Arm queue; some of the simpler stuff, things other have reviewed (thanks!), etc. -- PMM The following changes since commit 5d2f557b47dfbf8f23277a5bdd8473d4607c681a: Merge remote-tracking branch 'remotes/kraxel/tags/vga-20200605-pull-request' into staging (2020-06-05 13:53:05 +0100) are available in the Git repository at: https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20200605 for you to fetch changes up to 2c35a39eda0b16c2ed85c94cec204bf5efb97812: target/arm: Convert Neon one-register-and-immediate insns to decodetree (2020-06-05 17:23:10 +0100) ---------------------------------------------------------------- target-arm queue: hw/ssi/imx_spi: Handle tx burst lengths other than 8 correctly hw/input/pxa2xx_keypad: Replace hw_error() by qemu_log_mask() hw/arm/pxa2xx: Replace printf() call by qemu_log_mask() target/arm: Convert crypto insns to gvec hw/adc/stm32f2xx_adc: Correct memory region size and access size tests/acceptance: Add a boot test for the xlnx-versal-virt machine docs/system: Document Aspeed boards raspi: Add model of the USB controller target/arm: Convert 2-reg-and-shift and 1-reg-imm Neon insns to decodetree ---------------------------------------------------------------- Cédric Le Goater (1): docs/system: Document Aspeed boards Eden Mikitas (2): hw/ssi/imx_spi: changed while statement to prevent underflow hw/ssi/imx_spi: Removed unnecessary cast of rx data received from slave Paul Zimmerman (7): raspi: add BCM2835 SOC MPHI emulation dwc-hsotg (dwc2) USB host controller register definitions dwc-hsotg (dwc2) USB host controller state definitions dwc-hsotg (dwc2) USB host controller emulation usb: add short-packet handling to usb-storage driver wire in the dwc-hsotg (dwc2) USB host controller emulation raspi2 acceptance test: add test for dwc-hsotg (dwc2) USB host Peter Maydell (9): target/arm: Convert Neon VSHL and VSLI 2-reg-shift insn to decodetree target/arm: Convert Neon VSHR 2-reg-shift insns to decodetree target/arm: Convert Neon VSRA, VSRI, VRSHR, VRSRA 2-reg-shift insns to decodetree target/arm: Convert VQSHLU, VQSHL 2-reg-shift insns to decodetree target/arm: Convert Neon narrowing shifts with op==8 to decodetree target/arm: Convert Neon narrowing shifts with op==9 to decodetree target/arm: Convert Neon VSHLL, VMOVL to decodetree target/arm: Convert VCVT fixed-point ops to decodetree target/arm: Convert Neon one-register-and-immediate insns to decodetree Philippe Mathieu-Daudé (3): hw/input/pxa2xx_keypad: Replace hw_error() by qemu_log_mask() hw/arm/pxa2xx: Replace printf() call by qemu_log_mask() hw/adc/stm32f2xx_adc: Correct memory region size and access size Richard Henderson (6): target/arm: Convert aes and sm4 to gvec helpers target/arm: Convert rax1 to gvec helpers target/arm: Convert sha512 and sm3 to gvec helpers target/arm: Convert sha1 and sha256 to gvec helpers target/arm: Split helper_crypto_sha1_3reg target/arm: Split helper_crypto_sm3tt Thomas Huth (1): tests/acceptance: Add a boot test for the xlnx-versal-virt machine docs/system/arm/aspeed.rst | 85 ++ docs/system/target-arm.rst | 1 + hw/usb/hcd-dwc2.h | 190 +++++ include/hw/arm/bcm2835_peripherals.h | 5 +- include/hw/misc/bcm2835_mphi.h | 44 + include/hw/usb/dwc2-regs.h | 899 ++++++++++++++++++++ target/arm/helper.h | 45 +- target/arm/translate-a64.h | 3 + target/arm/vec_internal.h | 33 + target/arm/neon-dp.decode | 214 ++++- hw/adc/stm32f2xx_adc.c | 4 +- hw/arm/bcm2835_peripherals.c | 38 +- hw/arm/pxa2xx.c | 66 +- hw/input/pxa2xx_keypad.c | 10 +- hw/misc/bcm2835_mphi.c | 191 +++++ hw/ssi/imx_spi.c | 4 +- hw/usb/dev-storage.c | 15 +- hw/usb/hcd-dwc2.c | 1417 ++++++++++++++++++++++++++++++++ target/arm/crypto_helper.c | 267 ++++-- target/arm/translate-a64.c | 198 ++--- target/arm/translate-neon.inc.c | 796 ++++++++++++++---- target/arm/translate.c | 539 +----------- target/arm/vec_helper.c | 12 +- hw/misc/Makefile.objs | 1 + hw/usb/Kconfig | 5 + hw/usb/Makefile.objs | 1 + hw/usb/trace-events | 50 ++ tests/acceptance/boot_linux_console.py | 35 +- 28 files changed, 4258 insertions(+), 910 deletions(-) create mode 100644 docs/system/arm/aspeed.rst create mode 100644 hw/usb/hcd-dwc2.h create mode 100644 include/hw/misc/bcm2835_mphi.h create mode 100644 include/hw/usb/dwc2-regs.h create mode 100644 target/arm/vec_internal.h create mode 100644 hw/misc/bcm2835_mphi.c create mode 100644 hw/usb/hcd-dwc2.c From: Eden Mikitas <e.mikitas@gmail.com> The while statement in question only checked if tx_burst is not 0. tx_burst is a signed int, which is assigned the value put by the guest driver in ECSPI_CONREG. The burst length can be anywhere between 1 and 4096, and since tx_burst is always decremented by 8 it could possibly underflow, causing an infinite loop. Signed-off-by: Eden Mikitas <e.mikitas@gmail.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/ssi/imx_spi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/hw/ssi/imx_spi.c b/hw/ssi/imx_spi.c index XXXXXXX..XXXXXXX 100644 --- a/hw/ssi/imx_spi.c +++ b/hw/ssi/imx_spi.c @@ -XXX,XX +XXX,XX @@ static void imx_spi_flush_txfifo(IMXSPIState *s) rx = 0; - while (tx_burst) { + while (tx_burst > 0) { uint8_t byte = tx & 0xff; DPRINTF("writing 0x%02x\n", (uint32_t)byte); -- 2.20.1 From: Eden Mikitas <e.mikitas@gmail.com> When inserting the value retrieved (rx) from the spi slave, rx is pushed to rx_fifo after being cast to uint8_t. rx_fifo is a fifo32, and the rx register the driver uses is also 32 bit. This zeroes the 24 most significant bits of rx. This proved problematic with devices that expect to use the whole 32 bits of the rx register. Signed-off-by: Eden Mikitas <e.mikitas@gmail.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/ssi/imx_spi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/hw/ssi/imx_spi.c b/hw/ssi/imx_spi.c index XXXXXXX..XXXXXXX 100644 --- a/hw/ssi/imx_spi.c +++ b/hw/ssi/imx_spi.c @@ -XXX,XX +XXX,XX @@ static void imx_spi_flush_txfifo(IMXSPIState *s) if (fifo32_is_full(&s->rx_fifo)) { s->regs[ECSPI_STATREG] |= ECSPI_STATREG_RO; } else { - fifo32_push(&s->rx_fifo, (uint8_t)rx); + fifo32_push(&s->rx_fifo, rx); } if (s->burst_length <= 0) { -- 2.20.1 From: Philippe Mathieu-Daudé <f4bug@amsat.org> hw_error() calls exit(). This a bit overkill when we can log the accesses as unimplemented or guest error. When fuzzing the devices, we don't want the whole process to exit. Replace some hw_error() calls by qemu_log_mask() (missed in commit 5a0001ec7e). Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 20200525114123.21317-2-f4bug@amsat.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/input/pxa2xx_keypad.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/hw/input/pxa2xx_keypad.c b/hw/input/pxa2xx_keypad.c index XXXXXXX..XXXXXXX 100644 --- a/hw/input/pxa2xx_keypad.c +++ b/hw/input/pxa2xx_keypad.c @@ -XXX,XX +XXX,XX @@ */ #include "qemu/osdep.h" -#include "hw/hw.h" +#include "qemu/log.h" #include "hw/irq.h" #include "migration/vmstate.h" #include "hw/arm/pxa.h" @@ -XXX,XX +XXX,XX @@ static uint64_t pxa2xx_keypad_read(void *opaque, hwaddr offset, return s->kpkdi; break; default: - hw_error("%s: Bad offset " REG_FMT "\n", __func__, offset); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad read offset 0x%"HWADDR_PRIx"\n", + __func__, offset); } return 0; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_keypad_write(void *opaque, hwaddr offset, break; default: - hw_error("%s: Bad offset " REG_FMT "\n", __func__, offset); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad write offset 0x%"HWADDR_PRIx"\n", + __func__, offset); } } -- 2.20.1 From: Philippe Mathieu-Daudé <f4bug@amsat.org> Replace printf() calls by qemu_log_mask(), which is disabled by default. This avoid flooding the terminal when fuzzing the device. Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 20200525114123.21317-3-f4bug@amsat.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/pxa2xx.c | 66 ++++++++++++++++++++++++++++++++++++------------- 1 file changed, 49 insertions(+), 17 deletions(-) diff --git a/hw/arm/pxa2xx.c b/hw/arm/pxa2xx.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/pxa2xx.c +++ b/hw/arm/pxa2xx.c @@ -XXX,XX +XXX,XX @@ #include "sysemu/blockdev.h" #include "sysemu/qtest.h" #include "qemu/cutils.h" +#include "qemu/log.h" static struct { hwaddr io_base; @@ -XXX,XX +XXX,XX @@ static uint64_t pxa2xx_pm_read(void *opaque, hwaddr addr, return s->pm_regs[addr >> 2]; default: fail: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad read offset 0x%"HWADDR_PRIx"\n", + __func__, addr); break; } return 0; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_pm_write(void *opaque, hwaddr addr, s->pm_regs[addr >> 2] = value; break; } - - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad write offset 0x%"HWADDR_PRIx"\n", + __func__, addr); break; } } @@ -XXX,XX +XXX,XX @@ static uint64_t pxa2xx_cm_read(void *opaque, hwaddr addr, return s->cm_regs[CCCR >> 2] | (3 << 28); default: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad read offset 0x%"HWADDR_PRIx"\n", + __func__, addr); break; } return 0; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_cm_write(void *opaque, hwaddr addr, break; default: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad write offset 0x%"HWADDR_PRIx"\n", + __func__, addr); break; } } @@ -XXX,XX +XXX,XX @@ static uint64_t pxa2xx_mm_read(void *opaque, hwaddr addr, return s->mm_regs[addr >> 2]; /* fall through */ default: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad read offset 0x%"HWADDR_PRIx"\n", + __func__, addr); break; } return 0; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_mm_write(void *opaque, hwaddr addr, } default: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad write offset 0x%"HWADDR_PRIx"\n", + __func__, addr); break; } } @@ -XXX,XX +XXX,XX @@ static uint64_t pxa2xx_ssp_read(void *opaque, hwaddr addr, case SSACD: return s->ssacd; default: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad read offset 0x%"HWADDR_PRIx"\n", + __func__, addr); break; } return 0; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_ssp_write(void *opaque, hwaddr addr, break; default: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad write offset 0x%"HWADDR_PRIx"\n", + __func__, addr); break; } } @@ -XXX,XX +XXX,XX @@ static uint64_t pxa2xx_rtc_read(void *opaque, hwaddr addr, else return s->last_swcr; default: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad read offset 0x%"HWADDR_PRIx"\n", + __func__, addr); break; } return 0; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_rtc_write(void *opaque, hwaddr addr, break; default: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad write offset 0x%"HWADDR_PRIx"\n", + __func__, addr); } } @@ -XXX,XX +XXX,XX @@ static uint64_t pxa2xx_i2c_read(void *opaque, hwaddr addr, s->ibmr = 0; return s->ibmr; default: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad read offset 0x%"HWADDR_PRIx"\n", + __func__, addr); break; } return 0; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_i2c_write(void *opaque, hwaddr addr, break; default: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad write offset 0x%"HWADDR_PRIx"\n", + __func__, addr); } } @@ -XXX,XX +XXX,XX @@ static uint64_t pxa2xx_i2s_read(void *opaque, hwaddr addr, } return 0; default: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad read offset 0x%"HWADDR_PRIx"\n", + __func__, addr); break; } return 0; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_i2s_write(void *opaque, hwaddr addr, } break; default: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad write offset 0x%"HWADDR_PRIx"\n", + __func__, addr); } } @@ -XXX,XX +XXX,XX @@ static uint64_t pxa2xx_fir_read(void *opaque, hwaddr addr, case ICFOR: return s->rx_len; default: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad read offset 0x%"HWADDR_PRIx"\n", + __func__, addr); break; } return 0; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_fir_write(void *opaque, hwaddr addr, case ICFOR: break; default: - printf("%s: Bad register " REG_FMT "\n", __func__, addr); + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad write offset 0x%"HWADDR_PRIx"\n", + __func__, addr); } } -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> With this conversion, we will be able to use the same helpers with sve. In particular, pass 3 vector parameters for the 3-operand operations; for advsimd the destination register is also an input. This also fixes a bug in which we failed to clear the high bits of the SVE register after an AdvSIMD operation. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200514212831.31248-2-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 6 ++-- target/arm/vec_internal.h | 33 +++++++++++++++++ target/arm/crypto_helper.c | 72 +++++++++++++++++++++++++++----------- target/arm/translate-a64.c | 55 ++++++++++++++++++----------- target/arm/translate.c | 27 +++++++------- target/arm/vec_helper.c | 12 +------ 6 files changed, 138 insertions(+), 67 deletions(-) create mode 100644 target/arm/vec_internal.h diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_2(neon_qzip8, TCG_CALL_NO_RWG, void, ptr, ptr) DEF_HELPER_FLAGS_2(neon_qzip16, TCG_CALL_NO_RWG, void, ptr, ptr) DEF_HELPER_FLAGS_2(neon_qzip32, TCG_CALL_NO_RWG, void, ptr, ptr) -DEF_HELPER_FLAGS_3(crypto_aese, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_aese, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_3(crypto_aesmc, TCG_CALL_NO_RWG, void, ptr, ptr, i32) DEF_HELPER_FLAGS_4(crypto_sha1_3reg, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_5(crypto_sm3tt, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32, i32) DEF_HELPER_FLAGS_3(crypto_sm3partw1, TCG_CALL_NO_RWG, void, ptr, ptr, ptr) DEF_HELPER_FLAGS_3(crypto_sm3partw2, TCG_CALL_NO_RWG, void, ptr, ptr, ptr) -DEF_HELPER_FLAGS_2(crypto_sm4e, TCG_CALL_NO_RWG, void, ptr, ptr) -DEF_HELPER_FLAGS_3(crypto_sm4ekey, TCG_CALL_NO_RWG, void, ptr, ptr, ptr) +DEF_HELPER_FLAGS_4(crypto_sm4e, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_sm4ekey, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_3(crc32, TCG_CALL_NO_RWG_SE, i32, i32, i32, i32) DEF_HELPER_FLAGS_3(crc32c, TCG_CALL_NO_RWG_SE, i32, i32, i32, i32) diff --git a/target/arm/vec_internal.h b/target/arm/vec_internal.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/target/arm/vec_internal.h @@ -XXX,XX +XXX,XX @@ +/* + * ARM AdvSIMD / SVE Vector Helpers + * + * Copyright (c) 2020 Linaro + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; if not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef TARGET_ARM_VEC_INTERNALS_H +#define TARGET_ARM_VEC_INTERNALS_H + +static inline void clear_tail(void *vd, uintptr_t opr_sz, uintptr_t max_sz) +{ + uint64_t *d = vd + opr_sz; + uintptr_t i; + + for (i = opr_sz; i < max_sz; i += 8) { + *d++ = 0; + } +} + +#endif /* TARGET_ARM_VEC_INTERNALS_H */ diff --git a/target/arm/crypto_helper.c b/target/arm/crypto_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/crypto_helper.c +++ b/target/arm/crypto_helper.c @@ -XXX,XX +XXX,XX @@ #include "cpu.h" #include "exec/helper-proto.h" +#include "tcg/tcg-gvec-desc.h" #include "crypto/aes.h" +#include "vec_internal.h" union CRYPTO_STATE { uint8_t bytes[16]; @@ -XXX,XX +XXX,XX @@ union CRYPTO_STATE { #define CR_ST_WORD(state, i) (state.words[i]) #endif -void HELPER(crypto_aese)(void *vd, void *vm, uint32_t decrypt) +static void do_crypto_aese(uint64_t *rd, uint64_t *rn, + uint64_t *rm, bool decrypt) { static uint8_t const * const sbox[2] = { AES_sbox, AES_isbox }; static uint8_t const * const shift[2] = { AES_shifts, AES_ishifts }; - uint64_t *rd = vd; - uint64_t *rm = vm; union CRYPTO_STATE rk = { .l = { rm[0], rm[1] } }; - union CRYPTO_STATE st = { .l = { rd[0], rd[1] } }; + union CRYPTO_STATE st = { .l = { rn[0], rn[1] } }; int i; - assert(decrypt < 2); - /* xor state vector with round key */ rk.l[0] ^= st.l[0]; rk.l[1] ^= st.l[1]; @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_aese)(void *vd, void *vm, uint32_t decrypt) rd[1] = st.l[1]; } -void HELPER(crypto_aesmc)(void *vd, void *vm, uint32_t decrypt) +void HELPER(crypto_aese)(void *vd, void *vn, void *vm, uint32_t desc) +{ + intptr_t i, opr_sz = simd_oprsz(desc); + bool decrypt = simd_data(desc); + + for (i = 0; i < opr_sz; i += 16) { + do_crypto_aese(vd + i, vn + i, vm + i, decrypt); + } + clear_tail(vd, opr_sz, simd_maxsz(desc)); +} + +static void do_crypto_aesmc(uint64_t *rd, uint64_t *rm, bool decrypt) { static uint32_t const mc[][256] = { { /* MixColumns lookup table */ @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_aesmc)(void *vd, void *vm, uint32_t decrypt) 0xbe805d9f, 0xb58d5491, 0xa89a4f83, 0xa397468d, } }; - uint64_t *rd = vd; - uint64_t *rm = vm; union CRYPTO_STATE st = { .l = { rm[0], rm[1] } }; int i; - assert(decrypt < 2); - for (i = 0; i < 16; i += 4) { CR_ST_WORD(st, i >> 2) = mc[decrypt][CR_ST_BYTE(st, i)] ^ @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_aesmc)(void *vd, void *vm, uint32_t decrypt) rd[1] = st.l[1]; } +void HELPER(crypto_aesmc)(void *vd, void *vm, uint32_t desc) +{ + intptr_t i, opr_sz = simd_oprsz(desc); + bool decrypt = simd_data(desc); + + for (i = 0; i < opr_sz; i += 16) { + do_crypto_aesmc(vd + i, vm + i, decrypt); + } + clear_tail(vd, opr_sz, simd_maxsz(desc)); +} + /* * SHA-1 logical functions */ @@ -XXX,XX +XXX,XX @@ static uint8_t const sm4_sbox[] = { 0x79, 0xee, 0x5f, 0x3e, 0xd7, 0xcb, 0x39, 0x48, }; -void HELPER(crypto_sm4e)(void *vd, void *vn) +static void do_crypto_sm4e(uint64_t *rd, uint64_t *rn, uint64_t *rm) { - uint64_t *rd = vd; - uint64_t *rn = vn; - union CRYPTO_STATE d = { .l = { rd[0], rd[1] } }; - union CRYPTO_STATE n = { .l = { rn[0], rn[1] } }; + union CRYPTO_STATE d = { .l = { rn[0], rn[1] } }; + union CRYPTO_STATE n = { .l = { rm[0], rm[1] } }; uint32_t t, i; for (i = 0; i < 4; i++) { @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sm4e)(void *vd, void *vn) rd[1] = d.l[1]; } -void HELPER(crypto_sm4ekey)(void *vd, void *vn, void* vm) +void HELPER(crypto_sm4e)(void *vd, void *vn, void *vm, uint32_t desc) +{ + intptr_t i, opr_sz = simd_oprsz(desc); + + for (i = 0; i < opr_sz; i += 16) { + do_crypto_sm4e(vd + i, vn + i, vm + i); + } + clear_tail(vd, opr_sz, simd_maxsz(desc)); +} + +static void do_crypto_sm4ekey(uint64_t *rd, uint64_t *rn, uint64_t *rm) { - uint64_t *rd = vd; - uint64_t *rn = vn; - uint64_t *rm = vm; union CRYPTO_STATE d; union CRYPTO_STATE n = { .l = { rn[0], rn[1] } }; union CRYPTO_STATE m = { .l = { rm[0], rm[1] } }; @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sm4ekey)(void *vd, void *vn, void* vm) rd[0] = d.l[0]; rd[1] = d.l[1]; } + +void HELPER(crypto_sm4ekey)(void *vd, void *vn, void* vm, uint32_t desc) +{ + intptr_t i, opr_sz = simd_oprsz(desc); + + for (i = 0; i < opr_sz; i += 16) { + do_crypto_sm4ekey(vd + i, vn + i, vm + i); + } + clear_tail(vd, opr_sz, simd_maxsz(desc)); +} diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void gen_gvec_fn4(DisasContext *s, bool is_q, int rd, int rn, int rm, is_q ? 16 : 8, vec_full_reg_size(s)); } +/* Expand a 2-operand operation using an out-of-line helper. */ +static void gen_gvec_op2_ool(DisasContext *s, bool is_q, int rd, + int rn, int data, gen_helper_gvec_2 *fn) +{ + tcg_gen_gvec_2_ool(vec_full_reg_offset(s, rd), + vec_full_reg_offset(s, rn), + is_q ? 16 : 8, vec_full_reg_size(s), data, fn); +} + /* Expand a 3-operand operation using an out-of-line helper. */ static void gen_gvec_op3_ool(DisasContext *s, bool is_q, int rd, int rn, int rm, int data, gen_helper_gvec_3 *fn) @@ -XXX,XX +XXX,XX @@ static void disas_crypto_aes(DisasContext *s, uint32_t insn) int rn = extract32(insn, 5, 5); int rd = extract32(insn, 0, 5); int decrypt; - TCGv_ptr tcg_rd_ptr, tcg_rn_ptr; - TCGv_i32 tcg_decrypt; - CryptoThreeOpIntFn *genfn; + gen_helper_gvec_2 *genfn2 = NULL; + gen_helper_gvec_3 *genfn3 = NULL; if (!dc_isar_feature(aa64_aes, s) || size != 0) { unallocated_encoding(s); @@ -XXX,XX +XXX,XX @@ static void disas_crypto_aes(DisasContext *s, uint32_t insn) switch (opcode) { case 0x4: /* AESE */ decrypt = 0; - genfn = gen_helper_crypto_aese; + genfn3 = gen_helper_crypto_aese; break; case 0x6: /* AESMC */ decrypt = 0; - genfn = gen_helper_crypto_aesmc; + genfn2 = gen_helper_crypto_aesmc; break; case 0x5: /* AESD */ decrypt = 1; - genfn = gen_helper_crypto_aese; + genfn3 = gen_helper_crypto_aese; break; case 0x7: /* AESIMC */ decrypt = 1; - genfn = gen_helper_crypto_aesmc; + genfn2 = gen_helper_crypto_aesmc; break; default: unallocated_encoding(s); @@ -XXX,XX +XXX,XX @@ static void disas_crypto_aes(DisasContext *s, uint32_t insn) if (!fp_access_check(s)) { return; } - - tcg_rd_ptr = vec_full_reg_ptr(s, rd); - tcg_rn_ptr = vec_full_reg_ptr(s, rn); - tcg_decrypt = tcg_const_i32(decrypt); - - genfn(tcg_rd_ptr, tcg_rn_ptr, tcg_decrypt); - - tcg_temp_free_ptr(tcg_rd_ptr); - tcg_temp_free_ptr(tcg_rn_ptr); - tcg_temp_free_i32(tcg_decrypt); + if (genfn2) { + gen_gvec_op2_ool(s, true, rd, rn, decrypt, genfn2); + } else { + gen_gvec_op3_ool(s, true, rd, rd, rn, decrypt, genfn3); + } } /* Crypto three-reg SHA @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha512(DisasContext *s, uint32_t insn) int rn = extract32(insn, 5, 5); int rd = extract32(insn, 0, 5); bool feature; - CryptoThreeOpFn *genfn; + CryptoThreeOpFn *genfn = NULL; + gen_helper_gvec_3 *oolfn = NULL; if (o == 0) { switch (opcode) { @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha512(DisasContext *s, uint32_t insn) break; case 2: /* SM4EKEY */ feature = dc_isar_feature(aa64_sm4, s); - genfn = gen_helper_crypto_sm4ekey; + oolfn = gen_helper_crypto_sm4ekey; break; default: unallocated_encoding(s); @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha512(DisasContext *s, uint32_t insn) return; } + if (oolfn) { + gen_gvec_op3_ool(s, true, rd, rn, rm, 0, oolfn); + return; + } + if (genfn) { TCGv_ptr tcg_rd_ptr, tcg_rn_ptr, tcg_rm_ptr; @@ -XXX,XX +XXX,XX @@ static void disas_crypto_two_reg_sha512(DisasContext *s, uint32_t insn) TCGv_ptr tcg_rd_ptr, tcg_rn_ptr; bool feature; CryptoTwoOpFn *genfn; + gen_helper_gvec_3 *oolfn = NULL; switch (opcode) { case 0: /* SHA512SU0 */ @@ -XXX,XX +XXX,XX @@ static void disas_crypto_two_reg_sha512(DisasContext *s, uint32_t insn) break; case 1: /* SM4E */ feature = dc_isar_feature(aa64_sm4, s); - genfn = gen_helper_crypto_sm4e; + oolfn = gen_helper_crypto_sm4e; break; default: unallocated_encoding(s); @@ -XXX,XX +XXX,XX @@ static void disas_crypto_two_reg_sha512(DisasContext *s, uint32_t insn) return; } + if (oolfn) { + gen_gvec_op3_ool(s, true, rd, rd, rn, 0, oolfn); + return; + } + tcg_rd_ptr = vec_full_reg_ptr(s, rd); tcg_rn_ptr = vec_full_reg_ptr(s, rn); diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) if (!dc_isar_feature(aa32_aes, s) || ((rm | rd) & 1)) { return 1; } - ptr1 = vfp_reg_ptr(true, rd); - ptr2 = vfp_reg_ptr(true, rm); - - /* Bit 6 is the lowest opcode bit; it distinguishes between - * encryption (AESE/AESMC) and decryption (AESD/AESIMC) - */ - tmp3 = tcg_const_i32(extract32(insn, 6, 1)); - + /* + * Bit 6 is the lowest opcode bit; it distinguishes + * between encryption (AESE/AESMC) and decryption + * (AESD/AESIMC). + */ if (op == NEON_2RM_AESE) { - gen_helper_crypto_aese(ptr1, ptr2, tmp3); + tcg_gen_gvec_3_ool(vfp_reg_offset(true, rd), + vfp_reg_offset(true, rd), + vfp_reg_offset(true, rm), + 16, 16, extract32(insn, 6, 1), + gen_helper_crypto_aese); } else { - gen_helper_crypto_aesmc(ptr1, ptr2, tmp3); + tcg_gen_gvec_2_ool(vfp_reg_offset(true, rd), + vfp_reg_offset(true, rm), + 16, 16, extract32(insn, 6, 1), + gen_helper_crypto_aesmc); } - tcg_temp_free_ptr(ptr1); - tcg_temp_free_ptr(ptr2); - tcg_temp_free_i32(tmp3); break; case NEON_2RM_SHA1H: if (!dc_isar_feature(aa32_sha1, s) || ((rm | rd) & 1)) { diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ #include "exec/helper-proto.h" #include "tcg/tcg-gvec-desc.h" #include "fpu/softfloat.h" - +#include "vec_internal.h" /* Note that vector data is stored in host-endian 64-bit chunks, so addressing units smaller than that needs a host-endian fixup. */ @@ -XXX,XX +XXX,XX @@ #define H4(x) (x) #endif -static void clear_tail(void *vd, uintptr_t opr_sz, uintptr_t max_sz) -{ - uint64_t *d = vd + opr_sz; - uintptr_t i; - - for (i = opr_sz; i < max_sz; i += 8) { - *d++ = 0; - } -} - /* Signed saturating rounding doubling multiply-accumulate high half, 16-bit */ static int16_t inl_qrdmlah_s16(int16_t src1, int16_t src2, int16_t src3, uint32_t *sat) -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> With this conversion, we will be able to use the same helpers with sve. This also fixes a bug in which we failed to clear the high bits of the SVE register after an AdvSIMD operation. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200514212831.31248-3-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 2 ++ target/arm/translate-a64.h | 3 ++ target/arm/crypto_helper.c | 11 +++++++ target/arm/translate-a64.c | 59 ++++++++++++++++++++------------------ 4 files changed, 47 insertions(+), 28 deletions(-) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_3(crypto_sm3partw2, TCG_CALL_NO_RWG, void, ptr, ptr, ptr) DEF_HELPER_FLAGS_4(crypto_sm4e, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_4(crypto_sm4ekey, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_rax1, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) + DEF_HELPER_FLAGS_3(crc32, TCG_CALL_NO_RWG_SE, i32, i32, i32, i32) DEF_HELPER_FLAGS_3(crc32c, TCG_CALL_NO_RWG_SE, i32, i32, i32, i32) diff --git a/target/arm/translate-a64.h b/target/arm/translate-a64.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.h +++ b/target/arm/translate-a64.h @@ -XXX,XX +XXX,XX @@ static inline int vec_full_reg_size(DisasContext *s) bool disas_sve(DisasContext *, uint32_t); +void gen_gvec_rax1(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz); + #endif /* TARGET_ARM_TRANSLATE_A64_H */ diff --git a/target/arm/crypto_helper.c b/target/arm/crypto_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/crypto_helper.c +++ b/target/arm/crypto_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sm4ekey)(void *vd, void *vn, void* vm, uint32_t desc) } clear_tail(vd, opr_sz, simd_maxsz(desc)); } + +void HELPER(crypto_rax1)(void *vd, void *vn, void *vm, uint32_t desc) +{ + intptr_t i, opr_sz = simd_oprsz(desc); + uint64_t *d = vd, *n = vn, *m = vm; + + for (i = 0; i < opr_sz / 8; ++i) { + d[i] = n[i] ^ rol64(m[i], 1); + } + clear_tail(vd, opr_sz, simd_maxsz(desc)); +} diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void disas_crypto_two_reg_sha(DisasContext *s, uint32_t insn) tcg_temp_free_ptr(tcg_rn_ptr); } +static void gen_rax1_i64(TCGv_i64 d, TCGv_i64 n, TCGv_i64 m) +{ + tcg_gen_rotli_i64(d, m, 1); + tcg_gen_xor_i64(d, d, n); +} + +static void gen_rax1_vec(unsigned vece, TCGv_vec d, TCGv_vec n, TCGv_vec m) +{ + tcg_gen_rotli_vec(vece, d, m, 1); + tcg_gen_xor_vec(vece, d, d, n); +} + +void gen_gvec_rax1(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, + uint32_t rm_ofs, uint32_t opr_sz, uint32_t max_sz) +{ + static const TCGOpcode vecop_list[] = { INDEX_op_rotli_vec, 0 }; + static const GVecGen3 op = { + .fni8 = gen_rax1_i64, + .fniv = gen_rax1_vec, + .opt_opc = vecop_list, + .fno = gen_helper_crypto_rax1, + .vece = MO_64, + }; + tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, opr_sz, max_sz, &op); +} + /* Crypto three-reg SHA512 * 31 21 20 16 15 14 13 12 11 10 9 5 4 0 * +-----------------------+------+---+---+-----+--------+------+------+ @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha512(DisasContext *s, uint32_t insn) bool feature; CryptoThreeOpFn *genfn = NULL; gen_helper_gvec_3 *oolfn = NULL; + GVecGen3Fn *gvecfn = NULL; if (o == 0) { switch (opcode) { @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha512(DisasContext *s, uint32_t insn) break; case 3: /* RAX1 */ feature = dc_isar_feature(aa64_sha3, s); - genfn = NULL; + gvecfn = gen_gvec_rax1; break; default: g_assert_not_reached(); @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha512(DisasContext *s, uint32_t insn) if (oolfn) { gen_gvec_op3_ool(s, true, rd, rn, rm, 0, oolfn); - return; - } - - if (genfn) { + } else if (gvecfn) { + gen_gvec_fn3(s, true, rd, rn, rm, gvecfn, MO_64); + } else { TCGv_ptr tcg_rd_ptr, tcg_rn_ptr, tcg_rm_ptr; tcg_rd_ptr = vec_full_reg_ptr(s, rd); @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha512(DisasContext *s, uint32_t insn) tcg_temp_free_ptr(tcg_rd_ptr); tcg_temp_free_ptr(tcg_rn_ptr); tcg_temp_free_ptr(tcg_rm_ptr); - } else { - TCGv_i64 tcg_op1, tcg_op2, tcg_res[2]; - int pass; - - tcg_op1 = tcg_temp_new_i64(); - tcg_op2 = tcg_temp_new_i64(); - tcg_res[0] = tcg_temp_new_i64(); - tcg_res[1] = tcg_temp_new_i64(); - - for (pass = 0; pass < 2; pass++) { - read_vec_element(s, tcg_op1, rn, pass, MO_64); - read_vec_element(s, tcg_op2, rm, pass, MO_64); - - tcg_gen_rotli_i64(tcg_res[pass], tcg_op2, 1); - tcg_gen_xor_i64(tcg_res[pass], tcg_res[pass], tcg_op1); - } - write_vec_element(s, tcg_res[0], rd, 0, MO_64); - write_vec_element(s, tcg_res[1], rd, 1, MO_64); - - tcg_temp_free_i64(tcg_op1); - tcg_temp_free_i64(tcg_op2); - tcg_temp_free_i64(tcg_res[0]); - tcg_temp_free_i64(tcg_res[1]); } } -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Do not yet convert the helpers to loop over opr_sz, but the descriptor allows the vector tail to be cleared. Which fixes an existing bug vs SVE. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200514212831.31248-4-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 15 +++++++----- target/arm/crypto_helper.c | 37 +++++++++++++++++++++++----- target/arm/translate-a64.c | 50 ++++++++++++-------------------------- 3 files changed, 55 insertions(+), 47 deletions(-) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_3(crypto_sha256h2, TCG_CALL_NO_RWG, void, ptr, ptr, ptr) DEF_HELPER_FLAGS_2(crypto_sha256su0, TCG_CALL_NO_RWG, void, ptr, ptr) DEF_HELPER_FLAGS_3(crypto_sha256su1, TCG_CALL_NO_RWG, void, ptr, ptr, ptr) -DEF_HELPER_FLAGS_3(crypto_sha512h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr) -DEF_HELPER_FLAGS_3(crypto_sha512h2, TCG_CALL_NO_RWG, void, ptr, ptr, ptr) -DEF_HELPER_FLAGS_2(crypto_sha512su0, TCG_CALL_NO_RWG, void, ptr, ptr) -DEF_HELPER_FLAGS_3(crypto_sha512su1, TCG_CALL_NO_RWG, void, ptr, ptr, ptr) +DEF_HELPER_FLAGS_4(crypto_sha512h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_sha512h2, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(crypto_sha512su0, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_sha512su1, TCG_CALL_NO_RWG, + void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_5(crypto_sm3tt, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32, i32) -DEF_HELPER_FLAGS_3(crypto_sm3partw1, TCG_CALL_NO_RWG, void, ptr, ptr, ptr) -DEF_HELPER_FLAGS_3(crypto_sm3partw2, TCG_CALL_NO_RWG, void, ptr, ptr, ptr) +DEF_HELPER_FLAGS_4(crypto_sm3partw1, TCG_CALL_NO_RWG, + void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_sm3partw2, TCG_CALL_NO_RWG, + void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_4(crypto_sm4e, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_4(crypto_sm4ekey, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) diff --git a/target/arm/crypto_helper.c b/target/arm/crypto_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/crypto_helper.c +++ b/target/arm/crypto_helper.c @@ -XXX,XX +XXX,XX @@ union CRYPTO_STATE { #define CR_ST_WORD(state, i) (state.words[i]) #endif +/* + * The caller has not been converted to full gvec, and so only + * modifies the low 16 bytes of the vector register. + */ +static void clear_tail_16(void *vd, uint32_t desc) +{ + int opr_sz = simd_oprsz(desc); + int max_sz = simd_maxsz(desc); + + assert(opr_sz == 16); + clear_tail(vd, opr_sz, max_sz); +} + static void do_crypto_aese(uint64_t *rd, uint64_t *rn, uint64_t *rm, bool decrypt) { @@ -XXX,XX +XXX,XX @@ static uint64_t s1_512(uint64_t x) return ror64(x, 19) ^ ror64(x, 61) ^ (x >> 6); } -void HELPER(crypto_sha512h)(void *vd, void *vn, void *vm) +void HELPER(crypto_sha512h)(void *vd, void *vn, void *vm, uint32_t desc) { uint64_t *rd = vd; uint64_t *rn = vn; @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sha512h)(void *vd, void *vn, void *vm) rd[0] = d0; rd[1] = d1; + + clear_tail_16(vd, desc); } -void HELPER(crypto_sha512h2)(void *vd, void *vn, void *vm) +void HELPER(crypto_sha512h2)(void *vd, void *vn, void *vm, uint32_t desc) { uint64_t *rd = vd; uint64_t *rn = vn; @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sha512h2)(void *vd, void *vn, void *vm) rd[0] = d0; rd[1] = d1; + + clear_tail_16(vd, desc); } -void HELPER(crypto_sha512su0)(void *vd, void *vn) +void HELPER(crypto_sha512su0)(void *vd, void *vn, uint32_t desc) { uint64_t *rd = vd; uint64_t *rn = vn; @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sha512su0)(void *vd, void *vn) rd[0] = d0; rd[1] = d1; + + clear_tail_16(vd, desc); } -void HELPER(crypto_sha512su1)(void *vd, void *vn, void *vm) +void HELPER(crypto_sha512su1)(void *vd, void *vn, void *vm, uint32_t desc) { uint64_t *rd = vd; uint64_t *rn = vn; @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sha512su1)(void *vd, void *vn, void *vm) rd[0] += s1_512(rn[0]) + rm[0]; rd[1] += s1_512(rn[1]) + rm[1]; + + clear_tail_16(vd, desc); } -void HELPER(crypto_sm3partw1)(void *vd, void *vn, void *vm) +void HELPER(crypto_sm3partw1)(void *vd, void *vn, void *vm, uint32_t desc) { uint64_t *rd = vd; uint64_t *rn = vn; @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sm3partw1)(void *vd, void *vn, void *vm) rd[0] = d.l[0]; rd[1] = d.l[1]; + + clear_tail_16(vd, desc); } -void HELPER(crypto_sm3partw2)(void *vd, void *vn, void *vm) +void HELPER(crypto_sm3partw2)(void *vd, void *vn, void *vm, uint32_t desc) { uint64_t *rd = vd; uint64_t *rn = vn; @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sm3partw2)(void *vd, void *vn, void *vm) rd[0] = d.l[0]; rd[1] = d.l[1]; + + clear_tail_16(vd, desc); } void HELPER(crypto_sm3tt)(void *vd, void *vn, void *vm, uint32_t imm2, diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha512(DisasContext *s, uint32_t insn) int rn = extract32(insn, 5, 5); int rd = extract32(insn, 0, 5); bool feature; - CryptoThreeOpFn *genfn = NULL; gen_helper_gvec_3 *oolfn = NULL; GVecGen3Fn *gvecfn = NULL; @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha512(DisasContext *s, uint32_t insn) switch (opcode) { case 0: /* SHA512H */ feature = dc_isar_feature(aa64_sha512, s); - genfn = gen_helper_crypto_sha512h; + oolfn = gen_helper_crypto_sha512h; break; case 1: /* SHA512H2 */ feature = dc_isar_feature(aa64_sha512, s); - genfn = gen_helper_crypto_sha512h2; + oolfn = gen_helper_crypto_sha512h2; break; case 2: /* SHA512SU1 */ feature = dc_isar_feature(aa64_sha512, s); - genfn = gen_helper_crypto_sha512su1; + oolfn = gen_helper_crypto_sha512su1; break; case 3: /* RAX1 */ feature = dc_isar_feature(aa64_sha3, s); @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha512(DisasContext *s, uint32_t insn) switch (opcode) { case 0: /* SM3PARTW1 */ feature = dc_isar_feature(aa64_sm3, s); - genfn = gen_helper_crypto_sm3partw1; + oolfn = gen_helper_crypto_sm3partw1; break; case 1: /* SM3PARTW2 */ feature = dc_isar_feature(aa64_sm3, s); - genfn = gen_helper_crypto_sm3partw2; + oolfn = gen_helper_crypto_sm3partw2; break; case 2: /* SM4EKEY */ feature = dc_isar_feature(aa64_sm4, s); @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha512(DisasContext *s, uint32_t insn) if (oolfn) { gen_gvec_op3_ool(s, true, rd, rn, rm, 0, oolfn); - } else if (gvecfn) { - gen_gvec_fn3(s, true, rd, rn, rm, gvecfn, MO_64); } else { - TCGv_ptr tcg_rd_ptr, tcg_rn_ptr, tcg_rm_ptr; - - tcg_rd_ptr = vec_full_reg_ptr(s, rd); - tcg_rn_ptr = vec_full_reg_ptr(s, rn); - tcg_rm_ptr = vec_full_reg_ptr(s, rm); - - genfn(tcg_rd_ptr, tcg_rn_ptr, tcg_rm_ptr); - - tcg_temp_free_ptr(tcg_rd_ptr); - tcg_temp_free_ptr(tcg_rn_ptr); - tcg_temp_free_ptr(tcg_rm_ptr); + gen_gvec_fn3(s, true, rd, rn, rm, gvecfn, MO_64); } } @@ -XXX,XX +XXX,XX @@ static void disas_crypto_two_reg_sha512(DisasContext *s, uint32_t insn) int opcode = extract32(insn, 10, 2); int rn = extract32(insn, 5, 5); int rd = extract32(insn, 0, 5); - TCGv_ptr tcg_rd_ptr, tcg_rn_ptr; bool feature; - CryptoTwoOpFn *genfn; - gen_helper_gvec_3 *oolfn = NULL; switch (opcode) { case 0: /* SHA512SU0 */ feature = dc_isar_feature(aa64_sha512, s); - genfn = gen_helper_crypto_sha512su0; break; case 1: /* SM4E */ feature = dc_isar_feature(aa64_sm4, s); - oolfn = gen_helper_crypto_sm4e; break; default: unallocated_encoding(s); @@ -XXX,XX +XXX,XX @@ static void disas_crypto_two_reg_sha512(DisasContext *s, uint32_t insn) return; } - if (oolfn) { - gen_gvec_op3_ool(s, true, rd, rd, rn, 0, oolfn); - return; + switch (opcode) { + case 0: /* SHA512SU0 */ + gen_gvec_op2_ool(s, true, rd, rn, 0, gen_helper_crypto_sha512su0); + break; + case 1: /* SM4E */ + gen_gvec_op3_ool(s, true, rd, rd, rn, 0, gen_helper_crypto_sm4e); + break; + default: + g_assert_not_reached(); } - - tcg_rd_ptr = vec_full_reg_ptr(s, rd); - tcg_rn_ptr = vec_full_reg_ptr(s, rn); - - genfn(tcg_rd_ptr, tcg_rn_ptr); - - tcg_temp_free_ptr(tcg_rd_ptr); - tcg_temp_free_ptr(tcg_rn_ptr); } /* Crypto four-register -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Do not yet convert the helpers to loop over opr_sz, but the descriptor allows the vector tail to be cleared. Which fixes an existing bug vs SVE. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200514212831.31248-5-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 12 ++-- target/arm/neon-dp.decode | 12 ++-- target/arm/crypto_helper.c | 24 +++++-- target/arm/translate-a64.c | 34 ++++----- target/arm/translate-neon.inc.c | 124 +++++--------------------------- target/arm/translate.c | 24 ++----- 6 files changed, 67 insertions(+), 163 deletions(-) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_4(crypto_aese, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_3(crypto_aesmc, TCG_CALL_NO_RWG, void, ptr, ptr, i32) DEF_HELPER_FLAGS_4(crypto_sha1_3reg, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) -DEF_HELPER_FLAGS_2(crypto_sha1h, TCG_CALL_NO_RWG, void, ptr, ptr) -DEF_HELPER_FLAGS_2(crypto_sha1su1, TCG_CALL_NO_RWG, void, ptr, ptr) +DEF_HELPER_FLAGS_3(crypto_sha1h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(crypto_sha1su1, TCG_CALL_NO_RWG, void, ptr, ptr, i32) -DEF_HELPER_FLAGS_3(crypto_sha256h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr) -DEF_HELPER_FLAGS_3(crypto_sha256h2, TCG_CALL_NO_RWG, void, ptr, ptr, ptr) -DEF_HELPER_FLAGS_2(crypto_sha256su0, TCG_CALL_NO_RWG, void, ptr, ptr) -DEF_HELPER_FLAGS_3(crypto_sha256su1, TCG_CALL_NO_RWG, void, ptr, ptr, ptr) +DEF_HELPER_FLAGS_4(crypto_sha256h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_sha256h2, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(crypto_sha256su0, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_sha256su1, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_4(crypto_sha512h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_4(crypto_sha512h2, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VPADD_3s 1111 001 0 0 . .. .... .... 1011 . . . 1 .... @3same_q0 VQRDMLAH_3s 1111 001 1 0 . .. .... .... 1011 ... 1 .... @3same +@3same_crypto .... .... .... .... .... .... .... .... \ + &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp size=0 q=1 + SHA1_3s 1111 001 0 0 . optype:2 .... .... 1100 . 1 . 0 .... \ vm=%vm_dp vn=%vn_dp vd=%vd_dp -SHA256H_3s 1111 001 1 0 . 00 .... .... 1100 . 1 . 0 .... \ - vm=%vm_dp vn=%vn_dp vd=%vd_dp -SHA256H2_3s 1111 001 1 0 . 01 .... .... 1100 . 1 . 0 .... \ - vm=%vm_dp vn=%vn_dp vd=%vd_dp -SHA256SU1_3s 1111 001 1 0 . 10 .... .... 1100 . 1 . 0 .... \ - vm=%vm_dp vn=%vn_dp vd=%vd_dp +SHA256H_3s 1111 001 1 0 . 00 .... .... 1100 . 1 . 0 .... @3same_crypto +SHA256H2_3s 1111 001 1 0 . 01 .... .... 1100 . 1 . 0 .... @3same_crypto +SHA256SU1_3s 1111 001 1 0 . 10 .... .... 1100 . 1 . 0 .... @3same_crypto VFMA_fp_3s 1111 001 0 0 . 0 . .... .... 1100 ... 1 .... @3same_fp VFMS_fp_3s 1111 001 0 0 . 1 . .... .... 1100 ... 1 .... @3same_fp diff --git a/target/arm/crypto_helper.c b/target/arm/crypto_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/crypto_helper.c +++ b/target/arm/crypto_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sha1_3reg)(void *vd, void *vn, void *vm, uint32_t op) rd[1] = d.l[1]; } -void HELPER(crypto_sha1h)(void *vd, void *vm) +void HELPER(crypto_sha1h)(void *vd, void *vm, uint32_t desc) { uint64_t *rd = vd; uint64_t *rm = vm; @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sha1h)(void *vd, void *vm) rd[0] = m.l[0]; rd[1] = m.l[1]; + + clear_tail_16(vd, desc); } -void HELPER(crypto_sha1su1)(void *vd, void *vm) +void HELPER(crypto_sha1su1)(void *vd, void *vm, uint32_t desc) { uint64_t *rd = vd; uint64_t *rm = vm; @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sha1su1)(void *vd, void *vm) rd[0] = d.l[0]; rd[1] = d.l[1]; + + clear_tail_16(vd, desc); } /* @@ -XXX,XX +XXX,XX @@ static uint32_t s1(uint32_t x) return ror32(x, 17) ^ ror32(x, 19) ^ (x >> 10); } -void HELPER(crypto_sha256h)(void *vd, void *vn, void *vm) +void HELPER(crypto_sha256h)(void *vd, void *vn, void *vm, uint32_t desc) { uint64_t *rd = vd; uint64_t *rn = vn; @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sha256h)(void *vd, void *vn, void *vm) rd[0] = d.l[0]; rd[1] = d.l[1]; + + clear_tail_16(vd, desc); } -void HELPER(crypto_sha256h2)(void *vd, void *vn, void *vm) +void HELPER(crypto_sha256h2)(void *vd, void *vn, void *vm, uint32_t desc) { uint64_t *rd = vd; uint64_t *rn = vn; @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sha256h2)(void *vd, void *vn, void *vm) rd[0] = d.l[0]; rd[1] = d.l[1]; + + clear_tail_16(vd, desc); } -void HELPER(crypto_sha256su0)(void *vd, void *vm) +void HELPER(crypto_sha256su0)(void *vd, void *vm, uint32_t desc) { uint64_t *rd = vd; uint64_t *rm = vm; @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sha256su0)(void *vd, void *vm) rd[0] = d.l[0]; rd[1] = d.l[1]; + + clear_tail_16(vd, desc); } -void HELPER(crypto_sha256su1)(void *vd, void *vn, void *vm) +void HELPER(crypto_sha256su1)(void *vd, void *vn, void *vm, uint32_t desc) { uint64_t *rd = vd; uint64_t *rn = vn; @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sha256su1)(void *vd, void *vn, void *vm) rd[0] = d.l[0]; rd[1] = d.l[1]; + + clear_tail_16(vd, desc); } /* diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha(DisasContext *s, uint32_t insn) int rm = extract32(insn, 16, 5); int rn = extract32(insn, 5, 5); int rd = extract32(insn, 0, 5); - CryptoThreeOpFn *genfn; - TCGv_ptr tcg_rd_ptr, tcg_rn_ptr, tcg_rm_ptr; + gen_helper_gvec_3 *genfn; bool feature; if (size != 0) { @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha(DisasContext *s, uint32_t insn) return; } - tcg_rd_ptr = vec_full_reg_ptr(s, rd); - tcg_rn_ptr = vec_full_reg_ptr(s, rn); - tcg_rm_ptr = vec_full_reg_ptr(s, rm); - if (genfn) { - genfn(tcg_rd_ptr, tcg_rn_ptr, tcg_rm_ptr); + gen_gvec_op3_ool(s, true, rd, rn, rm, 0, genfn); } else { TCGv_i32 tcg_opcode = tcg_const_i32(opcode); + TCGv_ptr tcg_rd_ptr = vec_full_reg_ptr(s, rd); + TCGv_ptr tcg_rn_ptr = vec_full_reg_ptr(s, rn); + TCGv_ptr tcg_rm_ptr = vec_full_reg_ptr(s, rm); gen_helper_crypto_sha1_3reg(tcg_rd_ptr, tcg_rn_ptr, tcg_rm_ptr, tcg_opcode); - tcg_temp_free_i32(tcg_opcode); - } - tcg_temp_free_ptr(tcg_rd_ptr); - tcg_temp_free_ptr(tcg_rn_ptr); - tcg_temp_free_ptr(tcg_rm_ptr); + tcg_temp_free_i32(tcg_opcode); + tcg_temp_free_ptr(tcg_rd_ptr); + tcg_temp_free_ptr(tcg_rn_ptr); + tcg_temp_free_ptr(tcg_rm_ptr); + } } /* Crypto two-reg SHA @@ -XXX,XX +XXX,XX @@ static void disas_crypto_two_reg_sha(DisasContext *s, uint32_t insn) int opcode = extract32(insn, 12, 5); int rn = extract32(insn, 5, 5); int rd = extract32(insn, 0, 5); - CryptoTwoOpFn *genfn; + gen_helper_gvec_2 *genfn; bool feature; - TCGv_ptr tcg_rd_ptr, tcg_rn_ptr; if (size != 0) { unallocated_encoding(s); @@ -XXX,XX +XXX,XX @@ static void disas_crypto_two_reg_sha(DisasContext *s, uint32_t insn) if (!fp_access_check(s)) { return; } - - tcg_rd_ptr = vec_full_reg_ptr(s, rd); - tcg_rn_ptr = vec_full_reg_ptr(s, rn); - - genfn(tcg_rd_ptr, tcg_rn_ptr); - - tcg_temp_free_ptr(tcg_rd_ptr); - tcg_temp_free_ptr(tcg_rn_ptr); + gen_gvec_op2_ool(s, true, rd, rn, 0, genfn); } static void gen_rax1_i64(TCGv_i64 d, TCGv_i64 n, TCGv_i64 m) diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_3SAME_CMP(VCGE_S, TCG_COND_GE) DO_3SAME_CMP(VCGE_U, TCG_COND_GEU) DO_3SAME_CMP(VCEQ, TCG_COND_EQ) -static void gen_VMUL_p_3s(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, - uint32_t rm_ofs, uint32_t oprsz, uint32_t maxsz) -{ - tcg_gen_gvec_3_ool(rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz, - 0, gen_helper_gvec_pmul_b); -} +#define WRAP_OOL_FN(WRAPNAME, FUNC) \ + static void WRAPNAME(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, \ + uint32_t rm_ofs, uint32_t oprsz, uint32_t maxsz) \ + { \ + tcg_gen_gvec_3_ool(rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz, 0, FUNC); \ + } + +WRAP_OOL_FN(gen_VMUL_p_3s, gen_helper_gvec_pmul_b) static bool trans_VMUL_p_3s(DisasContext *s, arg_3same *a) { @@ -XXX,XX +XXX,XX @@ static bool trans_SHA1_3s(DisasContext *s, arg_SHA1_3s *a) return true; } -static bool trans_SHA256H_3s(DisasContext *s, arg_SHA256H_3s *a) -{ - TCGv_ptr ptr1, ptr2, ptr3; - - if (!arm_dc_feature(s, ARM_FEATURE_NEON) || - !dc_isar_feature(aa32_sha2, s)) { - return false; +#define DO_SHA2(NAME, FUNC) \ + WRAP_OOL_FN(gen_##NAME##_3s, FUNC) \ + static bool trans_##NAME##_3s(DisasContext *s, arg_3same *a) \ + { \ + if (!dc_isar_feature(aa32_sha2, s)) { \ + return false; \ + } \ + return do_3same(s, a, gen_##NAME##_3s); \ } - /* UNDEF accesses to D16-D31 if they don't exist. */ - if (!dc_isar_feature(aa32_simd_r32, s) && - ((a->vd | a->vn | a->vm) & 0x10)) { - return false; - } - - if ((a->vn | a->vm | a->vd) & 1) { - return false; - } - - if (!vfp_access_check(s)) { - return true; - } - - ptr1 = vfp_reg_ptr(true, a->vd); - ptr2 = vfp_reg_ptr(true, a->vn); - ptr3 = vfp_reg_ptr(true, a->vm); - gen_helper_crypto_sha256h(ptr1, ptr2, ptr3); - tcg_temp_free_ptr(ptr1); - tcg_temp_free_ptr(ptr2); - tcg_temp_free_ptr(ptr3); - - return true; -} - -static bool trans_SHA256H2_3s(DisasContext *s, arg_SHA256H2_3s *a) -{ - TCGv_ptr ptr1, ptr2, ptr3; - - if (!arm_dc_feature(s, ARM_FEATURE_NEON) || - !dc_isar_feature(aa32_sha2, s)) { - return false; - } - - /* UNDEF accesses to D16-D31 if they don't exist. */ - if (!dc_isar_feature(aa32_simd_r32, s) && - ((a->vd | a->vn | a->vm) & 0x10)) { - return false; - } - - if ((a->vn | a->vm | a->vd) & 1) { - return false; - } - - if (!vfp_access_check(s)) { - return true; - } - - ptr1 = vfp_reg_ptr(true, a->vd); - ptr2 = vfp_reg_ptr(true, a->vn); - ptr3 = vfp_reg_ptr(true, a->vm); - gen_helper_crypto_sha256h2(ptr1, ptr2, ptr3); - tcg_temp_free_ptr(ptr1); - tcg_temp_free_ptr(ptr2); - tcg_temp_free_ptr(ptr3); - - return true; -} - -static bool trans_SHA256SU1_3s(DisasContext *s, arg_SHA256SU1_3s *a) -{ - TCGv_ptr ptr1, ptr2, ptr3; - - if (!arm_dc_feature(s, ARM_FEATURE_NEON) || - !dc_isar_feature(aa32_sha2, s)) { - return false; - } - - /* UNDEF accesses to D16-D31 if they don't exist. */ - if (!dc_isar_feature(aa32_simd_r32, s) && - ((a->vd | a->vn | a->vm) & 0x10)) { - return false; - } - - if ((a->vn | a->vm | a->vd) & 1) { - return false; - } - - if (!vfp_access_check(s)) { - return true; - } - - ptr1 = vfp_reg_ptr(true, a->vd); - ptr2 = vfp_reg_ptr(true, a->vn); - ptr3 = vfp_reg_ptr(true, a->vm); - gen_helper_crypto_sha256su1(ptr1, ptr2, ptr3); - tcg_temp_free_ptr(ptr1); - tcg_temp_free_ptr(ptr2); - tcg_temp_free_ptr(ptr3); - - return true; -} +DO_SHA2(SHA256H, gen_helper_crypto_sha256h) +DO_SHA2(SHA256H2, gen_helper_crypto_sha256h2) +DO_SHA2(SHA256SU1, gen_helper_crypto_sha256su1) #define DO_3SAME_64(INSN, FUNC) \ static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) int vec_size; uint32_t imm; TCGv_i32 tmp, tmp2, tmp3, tmp4, tmp5; - TCGv_ptr ptr1, ptr2; + TCGv_ptr ptr1; TCGv_i64 tmp64; if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) if (!dc_isar_feature(aa32_sha1, s) || ((rm | rd) & 1)) { return 1; } - ptr1 = vfp_reg_ptr(true, rd); - ptr2 = vfp_reg_ptr(true, rm); - - gen_helper_crypto_sha1h(ptr1, ptr2); - - tcg_temp_free_ptr(ptr1); - tcg_temp_free_ptr(ptr2); + tcg_gen_gvec_2_ool(rd_ofs, rm_ofs, 16, 16, 0, + gen_helper_crypto_sha1h); break; case NEON_2RM_SHA1SU1: if ((rm | rd) & 1) { @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) } else if (!dc_isar_feature(aa32_sha1, s)) { return 1; } - ptr1 = vfp_reg_ptr(true, rd); - ptr2 = vfp_reg_ptr(true, rm); - if (q) { - gen_helper_crypto_sha256su0(ptr1, ptr2); - } else { - gen_helper_crypto_sha1su1(ptr1, ptr2); - } - tcg_temp_free_ptr(ptr1); - tcg_temp_free_ptr(ptr2); + tcg_gen_gvec_2_ool(rd_ofs, rm_ofs, 16, 16, 0, + q ? gen_helper_crypto_sha256su0 + : gen_helper_crypto_sha1su1); break; - case NEON_2RM_VMVN: tcg_gen_gvec_not(0, rd_ofs, rm_ofs, vec_size, vec_size); break; -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Rather than passing an opcode to a helper, fully decode the operation at translate time. Use clear_tail_16 to zap the balance of the SVE register with the AdvSIMD write. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200514212831.31248-6-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 5 +- target/arm/neon-dp.decode | 6 +- target/arm/crypto_helper.c | 99 +++++++++++++++++++++------------ target/arm/translate-a64.c | 29 ++++------ target/arm/translate-neon.inc.c | 46 ++++----------- 5 files changed, 93 insertions(+), 92 deletions(-) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_2(neon_qzip32, TCG_CALL_NO_RWG, void, ptr, ptr) DEF_HELPER_FLAGS_4(crypto_aese, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_3(crypto_aesmc, TCG_CALL_NO_RWG, void, ptr, ptr, i32) -DEF_HELPER_FLAGS_4(crypto_sha1_3reg, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_sha1su0, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_sha1c, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_sha1p, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_sha1m, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_3(crypto_sha1h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) DEF_HELPER_FLAGS_3(crypto_sha1su1, TCG_CALL_NO_RWG, void, ptr, ptr, i32) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VQRDMLAH_3s 1111 001 1 0 . .. .... .... 1011 ... 1 .... @3same @3same_crypto .... .... .... .... .... .... .... .... \ &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp size=0 q=1 -SHA1_3s 1111 001 0 0 . optype:2 .... .... 1100 . 1 . 0 .... \ - vm=%vm_dp vn=%vn_dp vd=%vd_dp +SHA1C_3s 1111 001 0 0 . 00 .... .... 1100 . 1 . 0 .... @3same_crypto +SHA1P_3s 1111 001 0 0 . 01 .... .... 1100 . 1 . 0 .... @3same_crypto +SHA1M_3s 1111 001 0 0 . 10 .... .... 1100 . 1 . 0 .... @3same_crypto +SHA1SU0_3s 1111 001 0 0 . 11 .... .... 1100 . 1 . 0 .... @3same_crypto SHA256H_3s 1111 001 1 0 . 00 .... .... 1100 . 1 . 0 .... @3same_crypto SHA256H2_3s 1111 001 1 0 . 01 .... .... 1100 . 1 . 0 .... @3same_crypto SHA256SU1_3s 1111 001 1 0 . 10 .... .... 1100 . 1 . 0 .... @3same_crypto diff --git a/target/arm/crypto_helper.c b/target/arm/crypto_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/crypto_helper.c +++ b/target/arm/crypto_helper.c @@ -XXX,XX +XXX,XX @@ union CRYPTO_STATE { }; #ifdef HOST_WORDS_BIGENDIAN -#define CR_ST_BYTE(state, i) (state.bytes[(15 - (i)) ^ 8]) -#define CR_ST_WORD(state, i) (state.words[(3 - (i)) ^ 2]) +#define CR_ST_BYTE(state, i) ((state).bytes[(15 - (i)) ^ 8]) +#define CR_ST_WORD(state, i) ((state).words[(3 - (i)) ^ 2]) #else -#define CR_ST_BYTE(state, i) (state.bytes[i]) -#define CR_ST_WORD(state, i) (state.words[i]) +#define CR_ST_BYTE(state, i) ((state).bytes[i]) +#define CR_ST_WORD(state, i) ((state).words[i]) #endif /* @@ -XXX,XX +XXX,XX @@ static uint32_t maj(uint32_t x, uint32_t y, uint32_t z) return (x & y) | ((x | y) & z); } -void HELPER(crypto_sha1_3reg)(void *vd, void *vn, void *vm, uint32_t op) +void HELPER(crypto_sha1su0)(void *vd, void *vn, void *vm, uint32_t desc) +{ + uint64_t *d = vd, *n = vn, *m = vm; + uint64_t d0, d1; + + d0 = d[1] ^ d[0] ^ m[0]; + d1 = n[0] ^ d[1] ^ m[1]; + d[0] = d0; + d[1] = d1; + + clear_tail_16(vd, desc); +} + +static inline void crypto_sha1_3reg(uint64_t *rd, uint64_t *rn, + uint64_t *rm, uint32_t desc, + uint32_t (*fn)(union CRYPTO_STATE *d)) { - uint64_t *rd = vd; - uint64_t *rn = vn; - uint64_t *rm = vm; union CRYPTO_STATE d = { .l = { rd[0], rd[1] } }; union CRYPTO_STATE n = { .l = { rn[0], rn[1] } }; union CRYPTO_STATE m = { .l = { rm[0], rm[1] } }; + int i; - if (op == 3) { /* sha1su0 */ - d.l[0] ^= d.l[1] ^ m.l[0]; - d.l[1] ^= n.l[0] ^ m.l[1]; - } else { - int i; + for (i = 0; i < 4; i++) { + uint32_t t = fn(&d); - for (i = 0; i < 4; i++) { - uint32_t t; + t += rol32(CR_ST_WORD(d, 0), 5) + CR_ST_WORD(n, 0) + + CR_ST_WORD(m, i); - switch (op) { - case 0: /* sha1c */ - t = cho(CR_ST_WORD(d, 1), CR_ST_WORD(d, 2), CR_ST_WORD(d, 3)); - break; - case 1: /* sha1p */ - t = par(CR_ST_WORD(d, 1), CR_ST_WORD(d, 2), CR_ST_WORD(d, 3)); - break; - case 2: /* sha1m */ - t = maj(CR_ST_WORD(d, 1), CR_ST_WORD(d, 2), CR_ST_WORD(d, 3)); - break; - default: - g_assert_not_reached(); - } - t += rol32(CR_ST_WORD(d, 0), 5) + CR_ST_WORD(n, 0) - + CR_ST_WORD(m, i); - - CR_ST_WORD(n, 0) = CR_ST_WORD(d, 3); - CR_ST_WORD(d, 3) = CR_ST_WORD(d, 2); - CR_ST_WORD(d, 2) = ror32(CR_ST_WORD(d, 1), 2); - CR_ST_WORD(d, 1) = CR_ST_WORD(d, 0); - CR_ST_WORD(d, 0) = t; - } + CR_ST_WORD(n, 0) = CR_ST_WORD(d, 3); + CR_ST_WORD(d, 3) = CR_ST_WORD(d, 2); + CR_ST_WORD(d, 2) = ror32(CR_ST_WORD(d, 1), 2); + CR_ST_WORD(d, 1) = CR_ST_WORD(d, 0); + CR_ST_WORD(d, 0) = t; } rd[0] = d.l[0]; rd[1] = d.l[1]; + + clear_tail_16(rd, desc); +} + +static uint32_t do_sha1c(union CRYPTO_STATE *d) +{ + return cho(CR_ST_WORD(*d, 1), CR_ST_WORD(*d, 2), CR_ST_WORD(*d, 3)); +} + +void HELPER(crypto_sha1c)(void *vd, void *vn, void *vm, uint32_t desc) +{ + crypto_sha1_3reg(vd, vn, vm, desc, do_sha1c); +} + +static uint32_t do_sha1p(union CRYPTO_STATE *d) +{ + return par(CR_ST_WORD(*d, 1), CR_ST_WORD(*d, 2), CR_ST_WORD(*d, 3)); +} + +void HELPER(crypto_sha1p)(void *vd, void *vn, void *vm, uint32_t desc) +{ + crypto_sha1_3reg(vd, vn, vm, desc, do_sha1p); +} + +static uint32_t do_sha1m(union CRYPTO_STATE *d) +{ + return maj(CR_ST_WORD(*d, 1), CR_ST_WORD(*d, 2), CR_ST_WORD(*d, 3)); +} + +void HELPER(crypto_sha1m)(void *vd, void *vn, void *vm, uint32_t desc) +{ + crypto_sha1_3reg(vd, vn, vm, desc, do_sha1m); } void HELPER(crypto_sha1h)(void *vd, void *vm, uint32_t desc) diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha(DisasContext *s, uint32_t insn) switch (opcode) { case 0: /* SHA1C */ + genfn = gen_helper_crypto_sha1c; + feature = dc_isar_feature(aa64_sha1, s); + break; case 1: /* SHA1P */ + genfn = gen_helper_crypto_sha1p; + feature = dc_isar_feature(aa64_sha1, s); + break; case 2: /* SHA1M */ + genfn = gen_helper_crypto_sha1m; + feature = dc_isar_feature(aa64_sha1, s); + break; case 3: /* SHA1SU0 */ - genfn = NULL; + genfn = gen_helper_crypto_sha1su0; feature = dc_isar_feature(aa64_sha1, s); break; case 4: /* SHA256H */ @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_sha(DisasContext *s, uint32_t insn) if (!fp_access_check(s)) { return; } - - if (genfn) { - gen_gvec_op3_ool(s, true, rd, rn, rm, 0, genfn); - } else { - TCGv_i32 tcg_opcode = tcg_const_i32(opcode); - TCGv_ptr tcg_rd_ptr = vec_full_reg_ptr(s, rd); - TCGv_ptr tcg_rn_ptr = vec_full_reg_ptr(s, rn); - TCGv_ptr tcg_rm_ptr = vec_full_reg_ptr(s, rm); - - gen_helper_crypto_sha1_3reg(tcg_rd_ptr, tcg_rn_ptr, - tcg_rm_ptr, tcg_opcode); - - tcg_temp_free_i32(tcg_opcode); - tcg_temp_free_ptr(tcg_rd_ptr); - tcg_temp_free_ptr(tcg_rn_ptr); - tcg_temp_free_ptr(tcg_rm_ptr); - } + gen_gvec_op3_ool(s, true, rd, rn, rm, 0, genfn); } /* Crypto two-reg SHA diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ static bool trans_VMUL_p_3s(DisasContext *s, arg_3same *a) DO_VQRDMLAH(VQRDMLAH, gen_gvec_sqrdmlah_qc) DO_VQRDMLAH(VQRDMLSH, gen_gvec_sqrdmlsh_qc) -static bool trans_SHA1_3s(DisasContext *s, arg_SHA1_3s *a) -{ - TCGv_ptr ptr1, ptr2, ptr3; - TCGv_i32 tmp; - - if (!arm_dc_feature(s, ARM_FEATURE_NEON) || - !dc_isar_feature(aa32_sha1, s)) { - return false; +#define DO_SHA1(NAME, FUNC) \ + WRAP_OOL_FN(gen_##NAME##_3s, FUNC) \ + static bool trans_##NAME##_3s(DisasContext *s, arg_3same *a) \ + { \ + if (!dc_isar_feature(aa32_sha1, s)) { \ + return false; \ + } \ + return do_3same(s, a, gen_##NAME##_3s); \ } - /* UNDEF accesses to D16-D31 if they don't exist. */ - if (!dc_isar_feature(aa32_simd_r32, s) && - ((a->vd | a->vn | a->vm) & 0x10)) { - return false; - } - - if ((a->vn | a->vm | a->vd) & 1) { - return false; - } - - if (!vfp_access_check(s)) { - return true; - } - - ptr1 = vfp_reg_ptr(true, a->vd); - ptr2 = vfp_reg_ptr(true, a->vn); - ptr3 = vfp_reg_ptr(true, a->vm); - tmp = tcg_const_i32(a->optype); - gen_helper_crypto_sha1_3reg(ptr1, ptr2, ptr3, tmp); - tcg_temp_free_i32(tmp); - tcg_temp_free_ptr(ptr1); - tcg_temp_free_ptr(ptr2); - tcg_temp_free_ptr(ptr3); - - return true; -} +DO_SHA1(SHA1C, gen_helper_crypto_sha1c) +DO_SHA1(SHA1P, gen_helper_crypto_sha1p) +DO_SHA1(SHA1M, gen_helper_crypto_sha1m) +DO_SHA1(SHA1SU0, gen_helper_crypto_sha1su0) #define DO_SHA2(NAME, FUNC) \ WRAP_OOL_FN(gen_##NAME##_3s, FUNC) \ -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Rather than passing an opcode to a helper, fully decode the operation at translate time. Use clear_tail_16 to zap the balance of the SVE register with the AdvSIMD write. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200514212831.31248-7-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 5 ++++- target/arm/crypto_helper.c | 24 ++++++++++++++++++------ target/arm/translate-a64.c | 21 +++++---------------- 3 files changed, 27 insertions(+), 23 deletions(-) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_3(crypto_sha512su0, TCG_CALL_NO_RWG, void, ptr, ptr, i32) DEF_HELPER_FLAGS_4(crypto_sha512su1, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) -DEF_HELPER_FLAGS_5(crypto_sm3tt, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32, i32) +DEF_HELPER_FLAGS_4(crypto_sm3tt1a, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_sm3tt1b, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_sm3tt2a, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_4(crypto_sm3tt2b, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_4(crypto_sm3partw1, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_4(crypto_sm3partw2, TCG_CALL_NO_RWG, diff --git a/target/arm/crypto_helper.c b/target/arm/crypto_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/crypto_helper.c +++ b/target/arm/crypto_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sm3partw2)(void *vd, void *vn, void *vm, uint32_t desc) clear_tail_16(vd, desc); } -void HELPER(crypto_sm3tt)(void *vd, void *vn, void *vm, uint32_t imm2, - uint32_t opcode) +static inline void QEMU_ALWAYS_INLINE +crypto_sm3tt(uint64_t *rd, uint64_t *rn, uint64_t *rm, + uint32_t desc, uint32_t opcode) { - uint64_t *rd = vd; - uint64_t *rn = vn; - uint64_t *rm = vm; union CRYPTO_STATE d = { .l = { rd[0], rd[1] } }; union CRYPTO_STATE n = { .l = { rn[0], rn[1] } }; union CRYPTO_STATE m = { .l = { rm[0], rm[1] } }; + uint32_t imm2 = simd_data(desc); uint32_t t; assert(imm2 < 4); @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sm3tt)(void *vd, void *vn, void *vm, uint32_t imm2, /* SM3TT2B */ t = cho(CR_ST_WORD(d, 3), CR_ST_WORD(d, 2), CR_ST_WORD(d, 1)); } else { - g_assert_not_reached(); + qemu_build_not_reached(); } t += CR_ST_WORD(d, 0) + CR_ST_WORD(m, imm2); @@ -XXX,XX +XXX,XX @@ void HELPER(crypto_sm3tt)(void *vd, void *vn, void *vm, uint32_t imm2, rd[0] = d.l[0]; rd[1] = d.l[1]; + + clear_tail_16(rd, desc); } +#define DO_SM3TT(NAME, OPCODE) \ + void HELPER(NAME)(void *vd, void *vn, void *vm, uint32_t desc) \ + { crypto_sm3tt(vd, vn, vm, desc, OPCODE); } + +DO_SM3TT(crypto_sm3tt1a, 0) +DO_SM3TT(crypto_sm3tt1b, 1) +DO_SM3TT(crypto_sm3tt2a, 2) +DO_SM3TT(crypto_sm3tt2b, 3) + +#undef DO_SM3TT + static uint8_t const sm4_sbox[] = { 0xd6, 0x90, 0xe9, 0xfe, 0xcc, 0xe1, 0x3d, 0xb7, 0x16, 0xb6, 0x14, 0xc2, 0x28, 0xfb, 0x2c, 0x05, diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void disas_crypto_xar(DisasContext *s, uint32_t insn) */ static void disas_crypto_three_reg_imm2(DisasContext *s, uint32_t insn) { + static gen_helper_gvec_3 * const fns[4] = { + gen_helper_crypto_sm3tt1a, gen_helper_crypto_sm3tt1b, + gen_helper_crypto_sm3tt2a, gen_helper_crypto_sm3tt2b, + }; int opcode = extract32(insn, 10, 2); int imm2 = extract32(insn, 12, 2); int rm = extract32(insn, 16, 5); int rn = extract32(insn, 5, 5); int rd = extract32(insn, 0, 5); - TCGv_ptr tcg_rd_ptr, tcg_rn_ptr, tcg_rm_ptr; - TCGv_i32 tcg_imm2, tcg_opcode; if (!dc_isar_feature(aa64_sm3, s)) { unallocated_encoding(s); @@ -XXX,XX +XXX,XX @@ static void disas_crypto_three_reg_imm2(DisasContext *s, uint32_t insn) return; } - tcg_rd_ptr = vec_full_reg_ptr(s, rd); - tcg_rn_ptr = vec_full_reg_ptr(s, rn); - tcg_rm_ptr = vec_full_reg_ptr(s, rm); - tcg_imm2 = tcg_const_i32(imm2); - tcg_opcode = tcg_const_i32(opcode); - - gen_helper_crypto_sm3tt(tcg_rd_ptr, tcg_rn_ptr, tcg_rm_ptr, tcg_imm2, - tcg_opcode); - - tcg_temp_free_ptr(tcg_rd_ptr); - tcg_temp_free_ptr(tcg_rn_ptr); - tcg_temp_free_ptr(tcg_rm_ptr); - tcg_temp_free_i32(tcg_imm2); - tcg_temp_free_i32(tcg_opcode); + gen_gvec_op3_ool(s, true, rd, rn, rm, imm2, fns[opcode]); } /* C3.6 Data processing - SIMD, inc Crypto -- 2.20.1 From: Philippe Mathieu-Daudé <f4bug@amsat.org> The ADC region size is 256B, split as: - [0x00 - 0x4f] defined - [0x50 - 0xff] reserved All registers are 32-bit (thus when the datasheet mentions the last defined register is 0x4c, it means its address range is 0x4c .. 0x4f. This model implementation is also 32-bit. Set MemoryRegionOps 'impl' fields. See: 'RM0033 Reference manual Rev 8', Table 10.13.18 "ADC register map". Reported-by: Seth Kintigh <skintigh@gmail.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 20200603055915.17678-1-f4bug@amsat.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/adc/stm32f2xx_adc.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/hw/adc/stm32f2xx_adc.c b/hw/adc/stm32f2xx_adc.c index XXXXXXX..XXXXXXX 100644 --- a/hw/adc/stm32f2xx_adc.c +++ b/hw/adc/stm32f2xx_adc.c @@ -XXX,XX +XXX,XX @@ static const MemoryRegionOps stm32f2xx_adc_ops = { .read = stm32f2xx_adc_read, .write = stm32f2xx_adc_write, .endianness = DEVICE_NATIVE_ENDIAN, + .impl.min_access_size = 4, + .impl.max_access_size = 4, }; static const VMStateDescription vmstate_stm32f2xx_adc = { @@ -XXX,XX +XXX,XX @@ static void stm32f2xx_adc_init(Object *obj) sysbus_init_irq(SYS_BUS_DEVICE(obj), &s->irq); memory_region_init_io(&s->mmio, obj, &stm32f2xx_adc_ops, s, - TYPE_STM32F2XX_ADC, 0xFF); + TYPE_STM32F2XX_ADC, 0x100); sysbus_init_mmio(SYS_BUS_DEVICE(obj), &s->mmio); } -- 2.20.1 From: Thomas Huth <thuth@redhat.com> As described by Edgar here: https://www.mail-archive.com/qemu-devel@nongnu.org/msg605124.html we can use the Ubuntu kernel for testing the xlnx-versal-virt machine. So let's add a boot test for this now. Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com> Signed-off-by: Thomas Huth <thuth@redhat.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20200525141237.15243-1-thuth@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/acceptance/boot_linux_console.py | 26 ++++++++++++++++++++++++++ 1 file changed, 26 insertions(+) diff --git a/tests/acceptance/boot_linux_console.py b/tests/acceptance/boot_linux_console.py index XXXXXXX..XXXXXXX 100644 --- a/tests/acceptance/boot_linux_console.py +++ b/tests/acceptance/boot_linux_console.py @@ -XXX,XX +XXX,XX @@ class BootLinuxConsole(LinuxKernelTest): console_pattern = 'Kernel command line: %s' % kernel_command_line self.wait_for_console_pattern(console_pattern) + def test_aarch64_xlnx_versal_virt(self): + """ + :avocado: tags=arch:aarch64 + :avocado: tags=machine:xlnx-versal-virt + :avocado: tags=device:pl011 + :avocado: tags=device:arm_gicv3 + """ + kernel_url = ('http://ports.ubuntu.com/ubuntu-ports/dists/' + 'bionic-updates/main/installer-arm64/current/images/' + 'netboot/ubuntu-installer/arm64/linux') + kernel_hash = '5bfc54cf7ed8157d93f6e5b0241e727b6dc22c50' + kernel_path = self.fetch_asset(kernel_url, asset_hash=kernel_hash) + + initrd_url = ('http://ports.ubuntu.com/ubuntu-ports/dists/' + 'bionic-updates/main/installer-arm64/current/images/' + 'netboot/ubuntu-installer/arm64/initrd.gz') + initrd_hash = 'd385d3e88d53e2004c5d43cbe668b458a094f772' + initrd_path = self.fetch_asset(initrd_url, asset_hash=initrd_hash) + + self.vm.set_console() + self.vm.add_args('-m', '2G', + '-kernel', kernel_path, + '-initrd', initrd_path) + self.vm.launch() + self.wait_for_console_pattern('Checked W+X mappings: passed') + def test_arm_virt(self): """ :avocado: tags=arch:arm -- 2.20.1 From: Cédric Le Goater <clg@kaod.org> Signed-off-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 20200602135050.593692-1-clg@kaod.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/aspeed.rst | 85 ++++++++++++++++++++++++++++++++++++++ docs/system/target-arm.rst | 1 + 2 files changed, 86 insertions(+) create mode 100644 docs/system/arm/aspeed.rst diff --git a/docs/system/arm/aspeed.rst b/docs/system/arm/aspeed.rst new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/docs/system/arm/aspeed.rst @@ -XXX,XX +XXX,XX @@ +Aspeed family boards (``*-bmc``, ``ast2500-evb``, ``ast2600-evb``) +================================================================== + +The QEMU Aspeed machines model BMCs of various OpenPOWER systems and +Aspeed evaluation boards. They are based on different releases of the +Aspeed SoC : the AST2400 integrating an ARM926EJ-S CPU (400MHz), the +AST2500 with an ARM1176JZS CPU (800MHz) and more recently the AST2600 +with dual cores ARM Cortex A7 CPUs (1.2GHz). + +The SoC comes with RAM, Gigabit ethernet, USB, SD/MMC, USB, SPI, I2C, +etc. + +AST2400 SoC based machines : + +- ``palmetto-bmc`` OpenPOWER Palmetto POWER8 BMC + +AST2500 SoC based machines : + +- ``ast2500-evb`` Aspeed AST2500 Evaluation board +- ``romulus-bmc`` OpenPOWER Romulus POWER9 BMC +- ``witherspoon-bmc`` OpenPOWER Witherspoon POWER9 BMC +- ``sonorapass-bmc`` OCP SonoraPass BMC +- ``swift-bmc`` OpenPOWER Swift BMC POWER9 + +AST2600 SoC based machines : + +- ``ast2600-evb`` Aspeed AST2600 Evaluation board (Cortex A7) +- ``tacoma-bmc`` OpenPOWER Witherspoon POWER9 AST2600 BMC + +Supported devices +----------------- + + * SMP (for the AST2600 Cortex-A7) + * Interrupt Controller (VIC) + * Timer Controller + * RTC Controller + * I2C Controller + * System Control Unit (SCU) + * SRAM mapping + * X-DMA Controller (basic interface) + * Static Memory Controller (SMC or FMC) - Only SPI Flash support + * SPI Memory Controller + * USB 2.0 Controller + * SD/MMC storage controllers + * SDRAM controller (dummy interface for basic settings and training) + * Watchdog Controller + * GPIO Controller (Master only) + * UART + * Ethernet controllers + + +Missing devices +--------------- + + * Coprocessor support + * ADC (out of tree implementation) + * PWM and Fan Controller + * LPC Bus Controller + * Slave GPIO Controller + * Super I/O Controller + * Hash/Crypto Engine + * PCI-Express 1 Controller + * Graphic Display Controller + * PECI Controller + * MCTP Controller + * Mailbox Controller + * Virtual UART + * eSPI Controller + * I3C Controller + +Boot options +------------ + +The Aspeed machines can be started using the -kernel option to load a +Linux kernel or from a firmare image which can be downloaded from the +OpenPOWER jenkins : + + https://openpower.xyz/ + +The image should be attached as an MTD drive. Run : + +.. code-block:: bash + + $ qemu-system-arm -M romulus-bmc -nic user \ + -drive file=flash-romulus,format=raw,if=mtd -nographic diff --git a/docs/system/target-arm.rst b/docs/system/target-arm.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/target-arm.rst +++ b/docs/system/target-arm.rst @@ -XXX,XX +XXX,XX @@ undocumented; you can get a complete list by running arm/realview arm/versatile arm/vexpress + arm/aspeed arm/musicpal arm/nseries arm/orangepi -- 2.20.1 From: Paul Zimmerman <pauldzim@gmail.com> Add BCM2835 SOC MPHI (Message-based Parallel Host Interface) emulation. It is very basic, only providing the FIQ interrupt needed to allow the dwc-otg USB host controller driver in the Raspbian kernel to function. Signed-off-by: Paul Zimmerman <pauldzim@gmail.com> Acked-by: Philippe Mathieu-Daude <f4bug@amsat.org> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 20200520235349.21215-2-pauldzim@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/bcm2835_peripherals.h | 2 + include/hw/misc/bcm2835_mphi.h | 44 ++++++ hw/arm/bcm2835_peripherals.c | 17 +++ hw/misc/bcm2835_mphi.c | 191 +++++++++++++++++++++++++++ hw/misc/Makefile.objs | 1 + 5 files changed, 255 insertions(+) create mode 100644 include/hw/misc/bcm2835_mphi.h create mode 100644 hw/misc/bcm2835_mphi.c diff --git a/include/hw/arm/bcm2835_peripherals.h b/include/hw/arm/bcm2835_peripherals.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/bcm2835_peripherals.h +++ b/include/hw/arm/bcm2835_peripherals.h @@ -XXX,XX +XXX,XX @@ #include "hw/misc/bcm2835_property.h" #include "hw/misc/bcm2835_rng.h" #include "hw/misc/bcm2835_mbox.h" +#include "hw/misc/bcm2835_mphi.h" #include "hw/misc/bcm2835_thermal.h" #include "hw/sd/sdhci.h" #include "hw/sd/bcm2835_sdhost.h" @@ -XXX,XX +XXX,XX @@ typedef struct BCM2835PeripheralState { qemu_irq irq, fiq; BCM2835SystemTimerState systmr; + BCM2835MphiState mphi; UnimplementedDeviceState armtmr; UnimplementedDeviceState cprman; UnimplementedDeviceState a2w; diff --git a/include/hw/misc/bcm2835_mphi.h b/include/hw/misc/bcm2835_mphi.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/misc/bcm2835_mphi.h @@ -XXX,XX +XXX,XX @@ +/* + * BCM2835 SOC MPHI state definitions + * + * Copyright (c) 2020 Paul Zimmerman <pauldzim@gmail.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#ifndef HW_MISC_BCM2835_MPHI_H +#define HW_MISC_BCM2835_MPHI_H + +#include "hw/irq.h" +#include "hw/sysbus.h" + +#define MPHI_MMIO_SIZE 0x1000 + +typedef struct BCM2835MphiState BCM2835MphiState; + +struct BCM2835MphiState { + SysBusDevice parent_obj; + qemu_irq irq; + MemoryRegion iomem; + + uint32_t outdda; + uint32_t outddb; + uint32_t ctrl; + uint32_t intstat; + uint32_t swirq; +}; + +#define TYPE_BCM2835_MPHI "bcm2835-mphi" + +#define BCM2835_MPHI(obj) \ + OBJECT_CHECK(BCM2835MphiState, (obj), TYPE_BCM2835_MPHI) + +#endif diff --git a/hw/arm/bcm2835_peripherals.c b/hw/arm/bcm2835_peripherals.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/bcm2835_peripherals.c +++ b/hw/arm/bcm2835_peripherals.c @@ -XXX,XX +XXX,XX @@ static void bcm2835_peripherals_init(Object *obj) OBJECT(&s->sdhci.sdbus)); object_property_add_const_link(OBJECT(&s->gpio), "sdbus-sdhost", OBJECT(&s->sdhost.sdbus)); + + /* Mphi */ + sysbus_init_child_obj(obj, "mphi", &s->mphi, sizeof(s->mphi), + TYPE_BCM2835_MPHI); } static void bcm2835_peripherals_realize(DeviceState *dev, Error **errp) @@ -XXX,XX +XXX,XX @@ static void bcm2835_peripherals_realize(DeviceState *dev, Error **errp) object_property_add_alias(OBJECT(s), "sd-bus", OBJECT(&s->gpio), "sd-bus"); + /* Mphi */ + object_property_set_bool(OBJECT(&s->mphi), true, "realized", &err); + if (err) { + error_propagate(errp, err); + return; + } + + memory_region_add_subregion(&s->peri_mr, MPHI_OFFSET, + sysbus_mmio_get_region(SYS_BUS_DEVICE(&s->mphi), 0)); + sysbus_connect_irq(SYS_BUS_DEVICE(&s->mphi), 0, + qdev_get_gpio_in_named(DEVICE(&s->ic), BCM2835_IC_GPU_IRQ, + INTERRUPT_HOSTPORT)); + create_unimp(s, &s->armtmr, "bcm2835-sp804", ARMCTRL_TIMER0_1_OFFSET, 0x40); create_unimp(s, &s->cprman, "bcm2835-cprman", CPRMAN_OFFSET, 0x1000); create_unimp(s, &s->a2w, "bcm2835-a2w", A2W_OFFSET, 0x1000); diff --git a/hw/misc/bcm2835_mphi.c b/hw/misc/bcm2835_mphi.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/misc/bcm2835_mphi.c @@ -XXX,XX +XXX,XX @@ +/* + * BCM2835 SOC MPHI emulation + * + * Very basic emulation, only providing the FIQ interrupt needed to + * allow the dwc-otg USB host controller driver in the Raspbian kernel + * to function. + * + * Copyright (c) 2020 Paul Zimmerman <pauldzim@gmail.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include "qemu/osdep.h" +#include "qapi/error.h" +#include "hw/misc/bcm2835_mphi.h" +#include "migration/vmstate.h" +#include "qemu/error-report.h" +#include "qemu/log.h" +#include "qemu/main-loop.h" + +static inline void mphi_raise_irq(BCM2835MphiState *s) +{ + qemu_set_irq(s->irq, 1); +} + +static inline void mphi_lower_irq(BCM2835MphiState *s) +{ + qemu_set_irq(s->irq, 0); +} + +static uint64_t mphi_reg_read(void *ptr, hwaddr addr, unsigned size) +{ + BCM2835MphiState *s = ptr; + uint32_t val = 0; + + switch (addr) { + case 0x28: /* outdda */ + val = s->outdda; + break; + case 0x2c: /* outddb */ + val = s->outddb; + break; + case 0x4c: /* ctrl */ + val = s->ctrl; + val |= 1 << 17; + break; + case 0x50: /* intstat */ + val = s->intstat; + break; + case 0x1f0: /* swirq_set */ + val = s->swirq; + break; + case 0x1f4: /* swirq_clr */ + val = s->swirq; + break; + default: + qemu_log_mask(LOG_UNIMP, "read from unknown register"); + break; + } + + return val; +} + +static void mphi_reg_write(void *ptr, hwaddr addr, uint64_t val, unsigned size) +{ + BCM2835MphiState *s = ptr; + int do_irq = 0; + + switch (addr) { + case 0x28: /* outdda */ + s->outdda = val; + break; + case 0x2c: /* outddb */ + s->outddb = val; + if (val & (1 << 29)) { + do_irq = 1; + } + break; + case 0x4c: /* ctrl */ + s->ctrl = val; + if (val & (1 << 16)) { + do_irq = -1; + } + break; + case 0x50: /* intstat */ + s->intstat = val; + if (val & ((1 << 16) | (1 << 29))) { + do_irq = -1; + } + break; + case 0x1f0: /* swirq_set */ + s->swirq |= val; + do_irq = 1; + break; + case 0x1f4: /* swirq_clr */ + s->swirq &= ~val; + do_irq = -1; + break; + default: + qemu_log_mask(LOG_UNIMP, "write to unknown register"); + return; + } + + if (do_irq > 0) { + mphi_raise_irq(s); + } else if (do_irq < 0) { + mphi_lower_irq(s); + } +} + +static const MemoryRegionOps mphi_mmio_ops = { + .read = mphi_reg_read, + .write = mphi_reg_write, + .impl.min_access_size = 4, + .impl.max_access_size = 4, + .endianness = DEVICE_LITTLE_ENDIAN, +}; + +static void mphi_reset(DeviceState *dev) +{ + BCM2835MphiState *s = BCM2835_MPHI(dev); + + s->outdda = 0; + s->outddb = 0; + s->ctrl = 0; + s->intstat = 0; + s->swirq = 0; +} + +static void mphi_realize(DeviceState *dev, Error **errp) +{ + SysBusDevice *sbd = SYS_BUS_DEVICE(dev); + BCM2835MphiState *s = BCM2835_MPHI(dev); + + sysbus_init_irq(sbd, &s->irq); +} + +static void mphi_init(Object *obj) +{ + SysBusDevice *sbd = SYS_BUS_DEVICE(obj); + BCM2835MphiState *s = BCM2835_MPHI(obj); + + memory_region_init_io(&s->iomem, obj, &mphi_mmio_ops, s, "mphi", MPHI_MMIO_SIZE); + sysbus_init_mmio(sbd, &s->iomem); +} + +const VMStateDescription vmstate_mphi_state = { + .name = "mphi", + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32(outdda, BCM2835MphiState), + VMSTATE_UINT32(outddb, BCM2835MphiState), + VMSTATE_UINT32(ctrl, BCM2835MphiState), + VMSTATE_UINT32(intstat, BCM2835MphiState), + VMSTATE_UINT32(swirq, BCM2835MphiState), + VMSTATE_END_OF_LIST() + } +}; + +static void mphi_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->realize = mphi_realize; + dc->reset = mphi_reset; + dc->vmsd = &vmstate_mphi_state; +} + +static const TypeInfo bcm2835_mphi_type_info = { + .name = TYPE_BCM2835_MPHI, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_size = sizeof(BCM2835MphiState), + .instance_init = mphi_init, + .class_init = mphi_class_init, +}; + +static void bcm2835_mphi_register_types(void) +{ + type_register_static(&bcm2835_mphi_type_info); +} + +type_init(bcm2835_mphi_register_types) diff --git a/hw/misc/Makefile.objs b/hw/misc/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/Makefile.objs +++ b/hw/misc/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-$(CONFIG_OMAP) += omap_l4.o common-obj-$(CONFIG_OMAP) += omap_sdrc.o common-obj-$(CONFIG_OMAP) += omap_tap.o common-obj-$(CONFIG_RASPI) += bcm2835_mbox.o +common-obj-$(CONFIG_RASPI) += bcm2835_mphi.o common-obj-$(CONFIG_RASPI) += bcm2835_property.o common-obj-$(CONFIG_RASPI) += bcm2835_rng.o common-obj-$(CONFIG_RASPI) += bcm2835_thermal.o -- 2.20.1 From: Paul Zimmerman <pauldzim@gmail.com> Import the dwc-hsotg (dwc2) register definitions file from the Linux kernel. This is a copy of drivers/usb/dwc2/hw.h from the mainline Linux kernel, the only changes being to the header, and two instances of 'u32' changed to 'uint32_t' to allow it to compile. Checkpatch throws a boatload of errors due to the tab indentation, but I would rather import it as-is than reformat it. Signed-off-by: Paul Zimmerman <pauldzim@gmail.com> Message-id: 20200520235349.21215-3-pauldzim@gmail.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/usb/dwc2-regs.h | 899 +++++++++++++++++++++++++++++++++++++ 1 file changed, 899 insertions(+) create mode 100644 include/hw/usb/dwc2-regs.h diff --git a/include/hw/usb/dwc2-regs.h b/include/hw/usb/dwc2-regs.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/usb/dwc2-regs.h @@ -XXX,XX +XXX,XX @@ +/* SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause) */ +/* + * Imported from the Linux kernel file drivers/usb/dwc2/hw.h, commit + * a89bae709b3492b478480a2c9734e7e9393b279c ("usb: dwc2: Move + * UTMI_PHY_DATA defines closer") + * + * hw.h - DesignWare HS OTG Controller hardware definitions + * + * Copyright 2004-2013 Synopsys, Inc. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions, and the following disclaimer, + * without modification. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. The names of the above-listed copyright holders may not be used + * to endorse or promote products derived from this software without + * specific prior written permission. + * + * ALTERNATIVELY, this software may be distributed under the terms of the + * GNU General Public License ("GPL") as published by the Free Software + * Foundation; either version 2 of the License, or (at your option) any + * later version. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS + * IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR + * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR + * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, + * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, + * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR + * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF + * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING + * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + */ + +#ifndef __DWC2_HW_H__ +#define __DWC2_HW_H__ + +#define HSOTG_REG(x) (x) + +#define GOTGCTL HSOTG_REG(0x000) +#define GOTGCTL_CHIRPEN BIT(27) +#define GOTGCTL_MULT_VALID_BC_MASK (0x1f << 22) +#define GOTGCTL_MULT_VALID_BC_SHIFT 22 +#define GOTGCTL_OTGVER BIT(20) +#define GOTGCTL_BSESVLD BIT(19) +#define GOTGCTL_ASESVLD BIT(18) +#define GOTGCTL_DBNC_SHORT BIT(17) +#define GOTGCTL_CONID_B BIT(16) +#define GOTGCTL_DBNCE_FLTR_BYPASS BIT(15) +#define GOTGCTL_DEVHNPEN BIT(11) +#define GOTGCTL_HSTSETHNPEN BIT(10) +#define GOTGCTL_HNPREQ BIT(9) +#define GOTGCTL_HSTNEGSCS BIT(8) +#define GOTGCTL_SESREQ BIT(1) +#define GOTGCTL_SESREQSCS BIT(0) + +#define GOTGINT HSOTG_REG(0x004) +#define GOTGINT_DBNCE_DONE BIT(19) +#define GOTGINT_A_DEV_TOUT_CHG BIT(18) +#define GOTGINT_HST_NEG_DET BIT(17) +#define GOTGINT_HST_NEG_SUC_STS_CHNG BIT(9) +#define GOTGINT_SES_REQ_SUC_STS_CHNG BIT(8) +#define GOTGINT_SES_END_DET BIT(2) + +#define GAHBCFG HSOTG_REG(0x008) +#define GAHBCFG_AHB_SINGLE BIT(23) +#define GAHBCFG_NOTI_ALL_DMA_WRIT BIT(22) +#define GAHBCFG_REM_MEM_SUPP BIT(21) +#define GAHBCFG_P_TXF_EMP_LVL BIT(8) +#define GAHBCFG_NP_TXF_EMP_LVL BIT(7) +#define GAHBCFG_DMA_EN BIT(5) +#define GAHBCFG_HBSTLEN_MASK (0xf << 1) +#define GAHBCFG_HBSTLEN_SHIFT 1 +#define GAHBCFG_HBSTLEN_SINGLE 0 +#define GAHBCFG_HBSTLEN_INCR 1 +#define GAHBCFG_HBSTLEN_INCR4 3 +#define GAHBCFG_HBSTLEN_INCR8 5 +#define GAHBCFG_HBSTLEN_INCR16 7 +#define GAHBCFG_GLBL_INTR_EN BIT(0) +#define GAHBCFG_CTRL_MASK (GAHBCFG_P_TXF_EMP_LVL | \ + GAHBCFG_NP_TXF_EMP_LVL | \ + GAHBCFG_DMA_EN | \ + GAHBCFG_GLBL_INTR_EN) + +#define GUSBCFG HSOTG_REG(0x00C) +#define GUSBCFG_FORCEDEVMODE BIT(30) +#define GUSBCFG_FORCEHOSTMODE BIT(29) +#define GUSBCFG_TXENDDELAY BIT(28) +#define GUSBCFG_ICTRAFFICPULLREMOVE BIT(27) +#define GUSBCFG_ICUSBCAP BIT(26) +#define GUSBCFG_ULPI_INT_PROT_DIS BIT(25) +#define GUSBCFG_INDICATORPASSTHROUGH BIT(24) +#define GUSBCFG_INDICATORCOMPLEMENT BIT(23) +#define GUSBCFG_TERMSELDLPULSE BIT(22) +#define GUSBCFG_ULPI_INT_VBUS_IND BIT(21) +#define GUSBCFG_ULPI_EXT_VBUS_DRV BIT(20) +#define GUSBCFG_ULPI_CLK_SUSP_M BIT(19) +#define GUSBCFG_ULPI_AUTO_RES BIT(18) +#define GUSBCFG_ULPI_FS_LS BIT(17) +#define GUSBCFG_OTG_UTMI_FS_SEL BIT(16) +#define GUSBCFG_PHY_LP_CLK_SEL BIT(15) +#define GUSBCFG_USBTRDTIM_MASK (0xf << 10) +#define GUSBCFG_USBTRDTIM_SHIFT 10 +#define GUSBCFG_HNPCAP BIT(9) +#define GUSBCFG_SRPCAP BIT(8) +#define GUSBCFG_DDRSEL BIT(7) +#define GUSBCFG_PHYSEL BIT(6) +#define GUSBCFG_FSINTF BIT(5) +#define GUSBCFG_ULPI_UTMI_SEL BIT(4) +#define GUSBCFG_PHYIF16 BIT(3) +#define GUSBCFG_PHYIF8 (0 << 3) +#define GUSBCFG_TOUTCAL_MASK (0x7 << 0) +#define GUSBCFG_TOUTCAL_SHIFT 0 +#define GUSBCFG_TOUTCAL_LIMIT 0x7 +#define GUSBCFG_TOUTCAL(_x) ((_x) << 0) + +#define GRSTCTL HSOTG_REG(0x010) +#define GRSTCTL_AHBIDLE BIT(31) +#define GRSTCTL_DMAREQ BIT(30) +#define GRSTCTL_TXFNUM_MASK (0x1f << 6) +#define GRSTCTL_TXFNUM_SHIFT 6 +#define GRSTCTL_TXFNUM_LIMIT 0x1f +#define GRSTCTL_TXFNUM(_x) ((_x) << 6) +#define GRSTCTL_TXFFLSH BIT(5) +#define GRSTCTL_RXFFLSH BIT(4) +#define GRSTCTL_IN_TKNQ_FLSH BIT(3) +#define GRSTCTL_FRMCNTRRST BIT(2) +#define GRSTCTL_HSFTRST BIT(1) +#define GRSTCTL_CSFTRST BIT(0) + +#define GINTSTS HSOTG_REG(0x014) +#define GINTMSK HSOTG_REG(0x018) +#define GINTSTS_WKUPINT BIT(31) +#define GINTSTS_SESSREQINT BIT(30) +#define GINTSTS_DISCONNINT BIT(29) +#define GINTSTS_CONIDSTSCHNG BIT(28) +#define GINTSTS_LPMTRANRCVD BIT(27) +#define GINTSTS_PTXFEMP BIT(26) +#define GINTSTS_HCHINT BIT(25) +#define GINTSTS_PRTINT BIT(24) +#define GINTSTS_RESETDET BIT(23) +#define GINTSTS_FET_SUSP BIT(22) +#define GINTSTS_INCOMPL_IP BIT(21) +#define GINTSTS_INCOMPL_SOOUT BIT(21) +#define GINTSTS_INCOMPL_SOIN BIT(20) +#define GINTSTS_OEPINT BIT(19) +#define GINTSTS_IEPINT BIT(18) +#define GINTSTS_EPMIS BIT(17) +#define GINTSTS_RESTOREDONE BIT(16) +#define GINTSTS_EOPF BIT(15) +#define GINTSTS_ISOUTDROP BIT(14) +#define GINTSTS_ENUMDONE BIT(13) +#define GINTSTS_USBRST BIT(12) +#define GINTSTS_USBSUSP BIT(11) +#define GINTSTS_ERLYSUSP BIT(10) +#define GINTSTS_I2CINT BIT(9) +#define GINTSTS_ULPI_CK_INT BIT(8) +#define GINTSTS_GOUTNAKEFF BIT(7) +#define GINTSTS_GINNAKEFF BIT(6) +#define GINTSTS_NPTXFEMP BIT(5) +#define GINTSTS_RXFLVL BIT(4) +#define GINTSTS_SOF BIT(3) +#define GINTSTS_OTGINT BIT(2) +#define GINTSTS_MODEMIS BIT(1) +#define GINTSTS_CURMODE_HOST BIT(0) + +#define GRXSTSR HSOTG_REG(0x01C) +#define GRXSTSP HSOTG_REG(0x020) +#define GRXSTS_FN_MASK (0x7f << 25) +#define GRXSTS_FN_SHIFT 25 +#define GRXSTS_PKTSTS_MASK (0xf << 17) +#define GRXSTS_PKTSTS_SHIFT 17 +#define GRXSTS_PKTSTS_GLOBALOUTNAK 1 +#define GRXSTS_PKTSTS_OUTRX 2 +#define GRXSTS_PKTSTS_HCHIN 2 +#define GRXSTS_PKTSTS_OUTDONE 3 +#define GRXSTS_PKTSTS_HCHIN_XFER_COMP 3 +#define GRXSTS_PKTSTS_SETUPDONE 4 +#define GRXSTS_PKTSTS_DATATOGGLEERR 5 +#define GRXSTS_PKTSTS_SETUPRX 6 +#define GRXSTS_PKTSTS_HCHHALTED 7 +#define GRXSTS_HCHNUM_MASK (0xf << 0) +#define GRXSTS_HCHNUM_SHIFT 0 +#define GRXSTS_DPID_MASK (0x3 << 15) +#define GRXSTS_DPID_SHIFT 15 +#define GRXSTS_BYTECNT_MASK (0x7ff << 4) +#define GRXSTS_BYTECNT_SHIFT 4 +#define GRXSTS_EPNUM_MASK (0xf << 0) +#define GRXSTS_EPNUM_SHIFT 0 + +#define GRXFSIZ HSOTG_REG(0x024) +#define GRXFSIZ_DEPTH_MASK (0xffff << 0) +#define GRXFSIZ_DEPTH_SHIFT 0 + +#define GNPTXFSIZ HSOTG_REG(0x028) +/* Use FIFOSIZE_* constants to access this register */ + +#define GNPTXSTS HSOTG_REG(0x02C) +#define GNPTXSTS_NP_TXQ_TOP_MASK (0x7f << 24) +#define GNPTXSTS_NP_TXQ_TOP_SHIFT 24 +#define GNPTXSTS_NP_TXQ_SPC_AVAIL_MASK (0xff << 16) +#define GNPTXSTS_NP_TXQ_SPC_AVAIL_SHIFT 16 +#define GNPTXSTS_NP_TXQ_SPC_AVAIL_GET(_v) (((_v) >> 16) & 0xff) +#define GNPTXSTS_NP_TXF_SPC_AVAIL_MASK (0xffff << 0) +#define GNPTXSTS_NP_TXF_SPC_AVAIL_SHIFT 0 +#define GNPTXSTS_NP_TXF_SPC_AVAIL_GET(_v) (((_v) >> 0) & 0xffff) + +#define GI2CCTL HSOTG_REG(0x0030) +#define GI2CCTL_BSYDNE BIT(31) +#define GI2CCTL_RW BIT(30) +#define GI2CCTL_I2CDATSE0 BIT(28) +#define GI2CCTL_I2CDEVADDR_MASK (0x3 << 26) +#define GI2CCTL_I2CDEVADDR_SHIFT 26 +#define GI2CCTL_I2CSUSPCTL BIT(25) +#define GI2CCTL_ACK BIT(24) +#define GI2CCTL_I2CEN BIT(23) +#define GI2CCTL_ADDR_MASK (0x7f << 16) +#define GI2CCTL_ADDR_SHIFT 16 +#define GI2CCTL_REGADDR_MASK (0xff << 8) +#define GI2CCTL_REGADDR_SHIFT 8 +#define GI2CCTL_RWDATA_MASK (0xff << 0) +#define GI2CCTL_RWDATA_SHIFT 0 + +#define GPVNDCTL HSOTG_REG(0x0034) +#define GGPIO HSOTG_REG(0x0038) +#define GGPIO_STM32_OTG_GCCFG_PWRDWN BIT(16) + +#define GUID HSOTG_REG(0x003c) +#define GSNPSID HSOTG_REG(0x0040) +#define GHWCFG1 HSOTG_REG(0x0044) +#define GSNPSID_ID_MASK GENMASK(31, 16) + +#define GHWCFG2 HSOTG_REG(0x0048) +#define GHWCFG2_OTG_ENABLE_IC_USB BIT(31) +#define GHWCFG2_DEV_TOKEN_Q_DEPTH_MASK (0x1f << 26) +#define GHWCFG2_DEV_TOKEN_Q_DEPTH_SHIFT 26 +#define GHWCFG2_HOST_PERIO_TX_Q_DEPTH_MASK (0x3 << 24) +#define GHWCFG2_HOST_PERIO_TX_Q_DEPTH_SHIFT 24 +#define GHWCFG2_NONPERIO_TX_Q_DEPTH_MASK (0x3 << 22) +#define GHWCFG2_NONPERIO_TX_Q_DEPTH_SHIFT 22 +#define GHWCFG2_MULTI_PROC_INT BIT(20) +#define GHWCFG2_DYNAMIC_FIFO BIT(19) +#define GHWCFG2_PERIO_EP_SUPPORTED BIT(18) +#define GHWCFG2_NUM_HOST_CHAN_MASK (0xf << 14) +#define GHWCFG2_NUM_HOST_CHAN_SHIFT 14 +#define GHWCFG2_NUM_DEV_EP_MASK (0xf << 10) +#define GHWCFG2_NUM_DEV_EP_SHIFT 10 +#define GHWCFG2_FS_PHY_TYPE_MASK (0x3 << 8) +#define GHWCFG2_FS_PHY_TYPE_SHIFT 8 +#define GHWCFG2_FS_PHY_TYPE_NOT_SUPPORTED 0 +#define GHWCFG2_FS_PHY_TYPE_DEDICATED 1 +#define GHWCFG2_FS_PHY_TYPE_SHARED_UTMI 2 +#define GHWCFG2_FS_PHY_TYPE_SHARED_ULPI 3 +#define GHWCFG2_HS_PHY_TYPE_MASK (0x3 << 6) +#define GHWCFG2_HS_PHY_TYPE_SHIFT 6 +#define GHWCFG2_HS_PHY_TYPE_NOT_SUPPORTED 0 +#define GHWCFG2_HS_PHY_TYPE_UTMI 1 +#define GHWCFG2_HS_PHY_TYPE_ULPI 2 +#define GHWCFG2_HS_PHY_TYPE_UTMI_ULPI 3 +#define GHWCFG2_POINT2POINT BIT(5) +#define GHWCFG2_ARCHITECTURE_MASK (0x3 << 3) +#define GHWCFG2_ARCHITECTURE_SHIFT 3 +#define GHWCFG2_SLAVE_ONLY_ARCH 0 +#define GHWCFG2_EXT_DMA_ARCH 1 +#define GHWCFG2_INT_DMA_ARCH 2 +#define GHWCFG2_OP_MODE_MASK (0x7 << 0) +#define GHWCFG2_OP_MODE_SHIFT 0 +#define GHWCFG2_OP_MODE_HNP_SRP_CAPABLE 0 +#define GHWCFG2_OP_MODE_SRP_ONLY_CAPABLE 1 +#define GHWCFG2_OP_MODE_NO_HNP_SRP_CAPABLE 2 +#define GHWCFG2_OP_MODE_SRP_CAPABLE_DEVICE 3 +#define GHWCFG2_OP_MODE_NO_SRP_CAPABLE_DEVICE 4 +#define GHWCFG2_OP_MODE_SRP_CAPABLE_HOST 5 +#define GHWCFG2_OP_MODE_NO_SRP_CAPABLE_HOST 6 +#define GHWCFG2_OP_MODE_UNDEFINED 7 + +#define GHWCFG3 HSOTG_REG(0x004c) +#define GHWCFG3_DFIFO_DEPTH_MASK (0xffff << 16) +#define GHWCFG3_DFIFO_DEPTH_SHIFT 16 +#define GHWCFG3_OTG_LPM_EN BIT(15) +#define GHWCFG3_BC_SUPPORT BIT(14) +#define GHWCFG3_OTG_ENABLE_HSIC BIT(13) +#define GHWCFG3_ADP_SUPP BIT(12) +#define GHWCFG3_SYNCH_RESET_TYPE BIT(11) +#define GHWCFG3_OPTIONAL_FEATURES BIT(10) +#define GHWCFG3_VENDOR_CTRL_IF BIT(9) +#define GHWCFG3_I2C BIT(8) +#define GHWCFG3_OTG_FUNC BIT(7) +#define GHWCFG3_PACKET_SIZE_CNTR_WIDTH_MASK (0x7 << 4) +#define GHWCFG3_PACKET_SIZE_CNTR_WIDTH_SHIFT 4 +#define GHWCFG3_XFER_SIZE_CNTR_WIDTH_MASK (0xf << 0) +#define GHWCFG3_XFER_SIZE_CNTR_WIDTH_SHIFT 0 + +#define GHWCFG4 HSOTG_REG(0x0050) +#define GHWCFG4_DESC_DMA_DYN BIT(31) +#define GHWCFG4_DESC_DMA BIT(30) +#define GHWCFG4_NUM_IN_EPS_MASK (0xf << 26) +#define GHWCFG4_NUM_IN_EPS_SHIFT 26 +#define GHWCFG4_DED_FIFO_EN BIT(25) +#define GHWCFG4_DED_FIFO_SHIFT 25 +#define GHWCFG4_SESSION_END_FILT_EN BIT(24) +#define GHWCFG4_B_VALID_FILT_EN BIT(23) +#define GHWCFG4_A_VALID_FILT_EN BIT(22) +#define GHWCFG4_VBUS_VALID_FILT_EN BIT(21) +#define GHWCFG4_IDDIG_FILT_EN BIT(20) +#define GHWCFG4_NUM_DEV_MODE_CTRL_EP_MASK (0xf << 16) +#define GHWCFG4_NUM_DEV_MODE_CTRL_EP_SHIFT 16 +#define GHWCFG4_UTMI_PHY_DATA_WIDTH_MASK (0x3 << 14) +#define GHWCFG4_UTMI_PHY_DATA_WIDTH_SHIFT 14 +#define GHWCFG4_UTMI_PHY_DATA_WIDTH_8 0 +#define GHWCFG4_UTMI_PHY_DATA_WIDTH_16 1 +#define GHWCFG4_UTMI_PHY_DATA_WIDTH_8_OR_16 2 +#define GHWCFG4_ACG_SUPPORTED BIT(12) +#define GHWCFG4_IPG_ISOC_SUPPORTED BIT(11) +#define GHWCFG4_SERVICE_INTERVAL_SUPPORTED BIT(10) +#define GHWCFG4_XHIBER BIT(7) +#define GHWCFG4_HIBER BIT(6) +#define GHWCFG4_MIN_AHB_FREQ BIT(5) +#define GHWCFG4_POWER_OPTIMIZ BIT(4) +#define GHWCFG4_NUM_DEV_PERIO_IN_EP_MASK (0xf << 0) +#define GHWCFG4_NUM_DEV_PERIO_IN_EP_SHIFT 0 + +#define GLPMCFG HSOTG_REG(0x0054) +#define GLPMCFG_INVSELHSIC BIT(31) +#define GLPMCFG_HSICCON BIT(30) +#define GLPMCFG_RSTRSLPSTS BIT(29) +#define GLPMCFG_ENBESL BIT(28) +#define GLPMCFG_LPM_RETRYCNT_STS_MASK (0x7 << 25) +#define GLPMCFG_LPM_RETRYCNT_STS_SHIFT 25 +#define GLPMCFG_SNDLPM BIT(24) +#define GLPMCFG_RETRY_CNT_MASK (0x7 << 21) +#define GLPMCFG_RETRY_CNT_SHIFT 21 +#define GLPMCFG_LPM_REJECT_CTRL_CONTROL BIT(21) +#define GLPMCFG_LPM_ACCEPT_CTRL_ISOC BIT(22) +#define GLPMCFG_LPM_CHNL_INDX_MASK (0xf << 17) +#define GLPMCFG_LPM_CHNL_INDX_SHIFT 17 +#define GLPMCFG_L1RESUMEOK BIT(16) +#define GLPMCFG_SLPSTS BIT(15) +#define GLPMCFG_COREL1RES_MASK (0x3 << 13) +#define GLPMCFG_COREL1RES_SHIFT 13 +#define GLPMCFG_HIRD_THRES_MASK (0x1f << 8) +#define GLPMCFG_HIRD_THRES_SHIFT 8 +#define GLPMCFG_HIRD_THRES_EN (0x10 << 8) +#define GLPMCFG_ENBLSLPM BIT(7) +#define GLPMCFG_BREMOTEWAKE BIT(6) +#define GLPMCFG_HIRD_MASK (0xf << 2) +#define GLPMCFG_HIRD_SHIFT 2 +#define GLPMCFG_APPL1RES BIT(1) +#define GLPMCFG_LPMCAP BIT(0) + +#define GPWRDN HSOTG_REG(0x0058) +#define GPWRDN_MULT_VAL_ID_BC_MASK (0x1f << 24) +#define GPWRDN_MULT_VAL_ID_BC_SHIFT 24 +#define GPWRDN_ADP_INT BIT(23) +#define GPWRDN_BSESSVLD BIT(22) +#define GPWRDN_IDSTS BIT(21) +#define GPWRDN_LINESTATE_MASK (0x3 << 19) +#define GPWRDN_LINESTATE_SHIFT 19 +#define GPWRDN_STS_CHGINT_MSK BIT(18) +#define GPWRDN_STS_CHGINT BIT(17) +#define GPWRDN_SRP_DET_MSK BIT(16) +#define GPWRDN_SRP_DET BIT(15) +#define GPWRDN_CONNECT_DET_MSK BIT(14) +#define GPWRDN_CONNECT_DET BIT(13) +#define GPWRDN_DISCONN_DET_MSK BIT(12) +#define GPWRDN_DISCONN_DET BIT(11) +#define GPWRDN_RST_DET_MSK BIT(10) +#define GPWRDN_RST_DET BIT(9) +#define GPWRDN_LNSTSCHG_MSK BIT(8) +#define GPWRDN_LNSTSCHG BIT(7) +#define GPWRDN_DIS_VBUS BIT(6) +#define GPWRDN_PWRDNSWTCH BIT(5) +#define GPWRDN_PWRDNRSTN BIT(4) +#define GPWRDN_PWRDNCLMP BIT(3) +#define GPWRDN_RESTORE BIT(2) +#define GPWRDN_PMUACTV BIT(1) +#define GPWRDN_PMUINTSEL BIT(0) + +#define GDFIFOCFG HSOTG_REG(0x005c) +#define GDFIFOCFG_EPINFOBASE_MASK (0xffff << 16) +#define GDFIFOCFG_EPINFOBASE_SHIFT 16 +#define GDFIFOCFG_GDFIFOCFG_MASK (0xffff << 0) +#define GDFIFOCFG_GDFIFOCFG_SHIFT 0 + +#define ADPCTL HSOTG_REG(0x0060) +#define ADPCTL_AR_MASK (0x3 << 27) +#define ADPCTL_AR_SHIFT 27 +#define ADPCTL_ADP_TMOUT_INT_MSK BIT(26) +#define ADPCTL_ADP_SNS_INT_MSK BIT(25) +#define ADPCTL_ADP_PRB_INT_MSK BIT(24) +#define ADPCTL_ADP_TMOUT_INT BIT(23) +#define ADPCTL_ADP_SNS_INT BIT(22) +#define ADPCTL_ADP_PRB_INT BIT(21) +#define ADPCTL_ADPENA BIT(20) +#define ADPCTL_ADPRES BIT(19) +#define ADPCTL_ENASNS BIT(18) +#define ADPCTL_ENAPRB BIT(17) +#define ADPCTL_RTIM_MASK (0x7ff << 6) +#define ADPCTL_RTIM_SHIFT 6 +#define ADPCTL_PRB_PER_MASK (0x3 << 4) +#define ADPCTL_PRB_PER_SHIFT 4 +#define ADPCTL_PRB_DELTA_MASK (0x3 << 2) +#define ADPCTL_PRB_DELTA_SHIFT 2 +#define ADPCTL_PRB_DSCHRG_MASK (0x3 << 0) +#define ADPCTL_PRB_DSCHRG_SHIFT 0 + +#define GREFCLK HSOTG_REG(0x0064) +#define GREFCLK_REFCLKPER_MASK (0x1ffff << 15) +#define GREFCLK_REFCLKPER_SHIFT 15 +#define GREFCLK_REF_CLK_MODE BIT(14) +#define GREFCLK_SOF_CNT_WKUP_ALERT_MASK (0x3ff) +#define GREFCLK_SOF_CNT_WKUP_ALERT_SHIFT 0 + +#define GINTMSK2 HSOTG_REG(0x0068) +#define GINTMSK2_WKUP_ALERT_INT_MSK BIT(0) + +#define GINTSTS2 HSOTG_REG(0x006c) +#define GINTSTS2_WKUP_ALERT_INT BIT(0) + +#define HPTXFSIZ HSOTG_REG(0x100) +/* Use FIFOSIZE_* constants to access this register */ + +#define DPTXFSIZN(_a) HSOTG_REG(0x104 + (((_a) - 1) * 4)) +/* Use FIFOSIZE_* constants to access this register */ + +/* These apply to the GNPTXFSIZ, HPTXFSIZ and DPTXFSIZN registers */ +#define FIFOSIZE_DEPTH_MASK (0xffff << 16) +#define FIFOSIZE_DEPTH_SHIFT 16 +#define FIFOSIZE_STARTADDR_MASK (0xffff << 0) +#define FIFOSIZE_STARTADDR_SHIFT 0 +#define FIFOSIZE_DEPTH_GET(_x) (((_x) >> 16) & 0xffff) + +/* Device mode registers */ + +#define DCFG HSOTG_REG(0x800) +#define DCFG_DESCDMA_EN BIT(23) +#define DCFG_EPMISCNT_MASK (0x1f << 18) +#define DCFG_EPMISCNT_SHIFT 18 +#define DCFG_EPMISCNT_LIMIT 0x1f +#define DCFG_EPMISCNT(_x) ((_x) << 18) +#define DCFG_IPG_ISOC_SUPPORDED BIT(17) +#define DCFG_PERFRINT_MASK (0x3 << 11) +#define DCFG_PERFRINT_SHIFT 11 +#define DCFG_PERFRINT_LIMIT 0x3 +#define DCFG_PERFRINT(_x) ((_x) << 11) +#define DCFG_DEVADDR_MASK (0x7f << 4) +#define DCFG_DEVADDR_SHIFT 4 +#define DCFG_DEVADDR_LIMIT 0x7f +#define DCFG_DEVADDR(_x) ((_x) << 4) +#define DCFG_NZ_STS_OUT_HSHK BIT(2) +#define DCFG_DEVSPD_MASK (0x3 << 0) +#define DCFG_DEVSPD_SHIFT 0 +#define DCFG_DEVSPD_HS 0 +#define DCFG_DEVSPD_FS 1 +#define DCFG_DEVSPD_LS 2 +#define DCFG_DEVSPD_FS48 3 + +#define DCTL HSOTG_REG(0x804) +#define DCTL_SERVICE_INTERVAL_SUPPORTED BIT(19) +#define DCTL_PWRONPRGDONE BIT(11) +#define DCTL_CGOUTNAK BIT(10) +#define DCTL_SGOUTNAK BIT(9) +#define DCTL_CGNPINNAK BIT(8) +#define DCTL_SGNPINNAK BIT(7) +#define DCTL_TSTCTL_MASK (0x7 << 4) +#define DCTL_TSTCTL_SHIFT 4 +#define DCTL_GOUTNAKSTS BIT(3) +#define DCTL_GNPINNAKSTS BIT(2) +#define DCTL_SFTDISCON BIT(1) +#define DCTL_RMTWKUPSIG BIT(0) + +#define DSTS HSOTG_REG(0x808) +#define DSTS_SOFFN_MASK (0x3fff << 8) +#define DSTS_SOFFN_SHIFT 8 +#define DSTS_SOFFN_LIMIT 0x3fff +#define DSTS_SOFFN(_x) ((_x) << 8) +#define DSTS_ERRATICERR BIT(3) +#define DSTS_ENUMSPD_MASK (0x3 << 1) +#define DSTS_ENUMSPD_SHIFT 1 +#define DSTS_ENUMSPD_HS 0 +#define DSTS_ENUMSPD_FS 1 +#define DSTS_ENUMSPD_LS 2 +#define DSTS_ENUMSPD_FS48 3 +#define DSTS_SUSPSTS BIT(0) + +#define DIEPMSK HSOTG_REG(0x810) +#define DIEPMSK_NAKMSK BIT(13) +#define DIEPMSK_BNAININTRMSK BIT(9) +#define DIEPMSK_TXFIFOUNDRNMSK BIT(8) +#define DIEPMSK_TXFIFOEMPTY BIT(7) +#define DIEPMSK_INEPNAKEFFMSK BIT(6) +#define DIEPMSK_INTKNEPMISMSK BIT(5) +#define DIEPMSK_INTKNTXFEMPMSK BIT(4) +#define DIEPMSK_TIMEOUTMSK BIT(3) +#define DIEPMSK_AHBERRMSK BIT(2) +#define DIEPMSK_EPDISBLDMSK BIT(1) +#define DIEPMSK_XFERCOMPLMSK BIT(0) + +#define DOEPMSK HSOTG_REG(0x814) +#define DOEPMSK_BNAMSK BIT(9) +#define DOEPMSK_BACK2BACKSETUP BIT(6) +#define DOEPMSK_STSPHSERCVDMSK BIT(5) +#define DOEPMSK_OUTTKNEPDISMSK BIT(4) +#define DOEPMSK_SETUPMSK BIT(3) +#define DOEPMSK_AHBERRMSK BIT(2) +#define DOEPMSK_EPDISBLDMSK BIT(1) +#define DOEPMSK_XFERCOMPLMSK BIT(0) + +#define DAINT HSOTG_REG(0x818) +#define DAINTMSK HSOTG_REG(0x81C) +#define DAINT_OUTEP_SHIFT 16 +#define DAINT_OUTEP(_x) (1 << ((_x) + 16)) +#define DAINT_INEP(_x) (1 << (_x)) + +#define DTKNQR1 HSOTG_REG(0x820) +#define DTKNQR2 HSOTG_REG(0x824) +#define DTKNQR3 HSOTG_REG(0x830) +#define DTKNQR4 HSOTG_REG(0x834) +#define DIEPEMPMSK HSOTG_REG(0x834) + +#define DVBUSDIS HSOTG_REG(0x828) +#define DVBUSPULSE HSOTG_REG(0x82C) + +#define DIEPCTL0 HSOTG_REG(0x900) +#define DIEPCTL(_a) HSOTG_REG(0x900 + ((_a) * 0x20)) + +#define DOEPCTL0 HSOTG_REG(0xB00) +#define DOEPCTL(_a) HSOTG_REG(0xB00 + ((_a) * 0x20)) + +/* EP0 specialness: + * bits[29..28] - reserved (no SetD0PID, SetD1PID) + * bits[25..22] - should always be zero, this isn't a periodic endpoint + * bits[10..0] - MPS setting different for EP0 + */ +#define D0EPCTL_MPS_MASK (0x3 << 0) +#define D0EPCTL_MPS_SHIFT 0 +#define D0EPCTL_MPS_64 0 +#define D0EPCTL_MPS_32 1 +#define D0EPCTL_MPS_16 2 +#define D0EPCTL_MPS_8 3 + +#define DXEPCTL_EPENA BIT(31) +#define DXEPCTL_EPDIS BIT(30) +#define DXEPCTL_SETD1PID BIT(29) +#define DXEPCTL_SETODDFR BIT(29) +#define DXEPCTL_SETD0PID BIT(28) +#define DXEPCTL_SETEVENFR BIT(28) +#define DXEPCTL_SNAK BIT(27) +#define DXEPCTL_CNAK BIT(26) +#define DXEPCTL_TXFNUM_MASK (0xf << 22) +#define DXEPCTL_TXFNUM_SHIFT 22 +#define DXEPCTL_TXFNUM_LIMIT 0xf +#define DXEPCTL_TXFNUM(_x) ((_x) << 22) +#define DXEPCTL_STALL BIT(21) +#define DXEPCTL_SNP BIT(20) +#define DXEPCTL_EPTYPE_MASK (0x3 << 18) +#define DXEPCTL_EPTYPE_CONTROL (0x0 << 18) +#define DXEPCTL_EPTYPE_ISO (0x1 << 18) +#define DXEPCTL_EPTYPE_BULK (0x2 << 18) +#define DXEPCTL_EPTYPE_INTERRUPT (0x3 << 18) + +#define DXEPCTL_NAKSTS BIT(17) +#define DXEPCTL_DPID BIT(16) +#define DXEPCTL_EOFRNUM BIT(16) +#define DXEPCTL_USBACTEP BIT(15) +#define DXEPCTL_NEXTEP_MASK (0xf << 11) +#define DXEPCTL_NEXTEP_SHIFT 11 +#define DXEPCTL_NEXTEP_LIMIT 0xf +#define DXEPCTL_NEXTEP(_x) ((_x) << 11) +#define DXEPCTL_MPS_MASK (0x7ff << 0) +#define DXEPCTL_MPS_SHIFT 0 +#define DXEPCTL_MPS_LIMIT 0x7ff +#define DXEPCTL_MPS(_x) ((_x) << 0) + +#define DIEPINT(_a) HSOTG_REG(0x908 + ((_a) * 0x20)) +#define DOEPINT(_a) HSOTG_REG(0xB08 + ((_a) * 0x20)) +#define DXEPINT_SETUP_RCVD BIT(15) +#define DXEPINT_NYETINTRPT BIT(14) +#define DXEPINT_NAKINTRPT BIT(13) +#define DXEPINT_BBLEERRINTRPT BIT(12) +#define DXEPINT_PKTDRPSTS BIT(11) +#define DXEPINT_BNAINTR BIT(9) +#define DXEPINT_TXFIFOUNDRN BIT(8) +#define DXEPINT_OUTPKTERR BIT(8) +#define DXEPINT_TXFEMP BIT(7) +#define DXEPINT_INEPNAKEFF BIT(6) +#define DXEPINT_BACK2BACKSETUP BIT(6) +#define DXEPINT_INTKNEPMIS BIT(5) +#define DXEPINT_STSPHSERCVD BIT(5) +#define DXEPINT_INTKNTXFEMP BIT(4) +#define DXEPINT_OUTTKNEPDIS BIT(4) +#define DXEPINT_TIMEOUT BIT(3) +#define DXEPINT_SETUP BIT(3) +#define DXEPINT_AHBERR BIT(2) +#define DXEPINT_EPDISBLD BIT(1) +#define DXEPINT_XFERCOMPL BIT(0) + +#define DIEPTSIZ0 HSOTG_REG(0x910) +#define DIEPTSIZ0_PKTCNT_MASK (0x3 << 19) +#define DIEPTSIZ0_PKTCNT_SHIFT 19 +#define DIEPTSIZ0_PKTCNT_LIMIT 0x3 +#define DIEPTSIZ0_PKTCNT(_x) ((_x) << 19) +#define DIEPTSIZ0_XFERSIZE_MASK (0x7f << 0) +#define DIEPTSIZ0_XFERSIZE_SHIFT 0 +#define DIEPTSIZ0_XFERSIZE_LIMIT 0x7f +#define DIEPTSIZ0_XFERSIZE(_x) ((_x) << 0) + +#define DOEPTSIZ0 HSOTG_REG(0xB10) +#define DOEPTSIZ0_SUPCNT_MASK (0x3 << 29) +#define DOEPTSIZ0_SUPCNT_SHIFT 29 +#define DOEPTSIZ0_SUPCNT_LIMIT 0x3 +#define DOEPTSIZ0_SUPCNT(_x) ((_x) << 29) +#define DOEPTSIZ0_PKTCNT BIT(19) +#define DOEPTSIZ0_XFERSIZE_MASK (0x7f << 0) +#define DOEPTSIZ0_XFERSIZE_SHIFT 0 + +#define DIEPTSIZ(_a) HSOTG_REG(0x910 + ((_a) * 0x20)) +#define DOEPTSIZ(_a) HSOTG_REG(0xB10 + ((_a) * 0x20)) +#define DXEPTSIZ_MC_MASK (0x3 << 29) +#define DXEPTSIZ_MC_SHIFT 29 +#define DXEPTSIZ_MC_LIMIT 0x3 +#define DXEPTSIZ_MC(_x) ((_x) << 29) +#define DXEPTSIZ_PKTCNT_MASK (0x3ff << 19) +#define DXEPTSIZ_PKTCNT_SHIFT 19 +#define DXEPTSIZ_PKTCNT_LIMIT 0x3ff +#define DXEPTSIZ_PKTCNT_GET(_v) (((_v) >> 19) & 0x3ff) +#define DXEPTSIZ_PKTCNT(_x) ((_x) << 19) +#define DXEPTSIZ_XFERSIZE_MASK (0x7ffff << 0) +#define DXEPTSIZ_XFERSIZE_SHIFT 0 +#define DXEPTSIZ_XFERSIZE_LIMIT 0x7ffff +#define DXEPTSIZ_XFERSIZE_GET(_v) (((_v) >> 0) & 0x7ffff) +#define DXEPTSIZ_XFERSIZE(_x) ((_x) << 0) + +#define DIEPDMA(_a) HSOTG_REG(0x914 + ((_a) * 0x20)) +#define DOEPDMA(_a) HSOTG_REG(0xB14 + ((_a) * 0x20)) + +#define DTXFSTS(_a) HSOTG_REG(0x918 + ((_a) * 0x20)) + +#define PCGCTL HSOTG_REG(0x0e00) +#define PCGCTL_IF_DEV_MODE BIT(31) +#define PCGCTL_P2HD_PRT_SPD_MASK (0x3 << 29) +#define PCGCTL_P2HD_PRT_SPD_SHIFT 29 +#define PCGCTL_P2HD_DEV_ENUM_SPD_MASK (0x3 << 27) +#define PCGCTL_P2HD_DEV_ENUM_SPD_SHIFT 27 +#define PCGCTL_MAC_DEV_ADDR_MASK (0x7f << 20) +#define PCGCTL_MAC_DEV_ADDR_SHIFT 20 +#define PCGCTL_MAX_TERMSEL BIT(19) +#define PCGCTL_MAX_XCVRSELECT_MASK (0x3 << 17) +#define PCGCTL_MAX_XCVRSELECT_SHIFT 17 +#define PCGCTL_PORT_POWER BIT(16) +#define PCGCTL_PRT_CLK_SEL_MASK (0x3 << 14) +#define PCGCTL_PRT_CLK_SEL_SHIFT 14 +#define PCGCTL_ESS_REG_RESTORED BIT(13) +#define PCGCTL_EXTND_HIBER_SWITCH BIT(12) +#define PCGCTL_EXTND_HIBER_PWRCLMP BIT(11) +#define PCGCTL_ENBL_EXTND_HIBER BIT(10) +#define PCGCTL_RESTOREMODE BIT(9) +#define PCGCTL_RESETAFTSUSP BIT(8) +#define PCGCTL_DEEP_SLEEP BIT(7) +#define PCGCTL_PHY_IN_SLEEP BIT(6) +#define PCGCTL_ENBL_SLEEP_GATING BIT(5) +#define PCGCTL_RSTPDWNMODULE BIT(3) +#define PCGCTL_PWRCLMP BIT(2) +#define PCGCTL_GATEHCLK BIT(1) +#define PCGCTL_STOPPCLK BIT(0) + +#define PCGCCTL1 HSOTG_REG(0xe04) +#define PCGCCTL1_TIMER (0x3 << 1) +#define PCGCCTL1_GATEEN BIT(0) + +#define EPFIFO(_a) HSOTG_REG(0x1000 + ((_a) * 0x1000)) + +/* Host Mode Registers */ + +#define HCFG HSOTG_REG(0x0400) +#define HCFG_MODECHTIMEN BIT(31) +#define HCFG_PERSCHEDENA BIT(26) +#define HCFG_FRLISTEN_MASK (0x3 << 24) +#define HCFG_FRLISTEN_SHIFT 24 +#define HCFG_FRLISTEN_8 (0 << 24) +#define FRLISTEN_8_SIZE 8 +#define HCFG_FRLISTEN_16 BIT(24) +#define FRLISTEN_16_SIZE 16 +#define HCFG_FRLISTEN_32 (2 << 24) +#define FRLISTEN_32_SIZE 32 +#define HCFG_FRLISTEN_64 (3 << 24) +#define FRLISTEN_64_SIZE 64 +#define HCFG_DESCDMA BIT(23) +#define HCFG_RESVALID_MASK (0xff << 8) +#define HCFG_RESVALID_SHIFT 8 +#define HCFG_ENA32KHZ BIT(7) +#define HCFG_FSLSSUPP BIT(2) +#define HCFG_FSLSPCLKSEL_MASK (0x3 << 0) +#define HCFG_FSLSPCLKSEL_SHIFT 0 +#define HCFG_FSLSPCLKSEL_30_60_MHZ 0 +#define HCFG_FSLSPCLKSEL_48_MHZ 1 +#define HCFG_FSLSPCLKSEL_6_MHZ 2 + +#define HFIR HSOTG_REG(0x0404) +#define HFIR_FRINT_MASK (0xffff << 0) +#define HFIR_FRINT_SHIFT 0 +#define HFIR_RLDCTRL BIT(16) + +#define HFNUM HSOTG_REG(0x0408) +#define HFNUM_FRREM_MASK (0xffff << 16) +#define HFNUM_FRREM_SHIFT 16 +#define HFNUM_FRNUM_MASK (0xffff << 0) +#define HFNUM_FRNUM_SHIFT 0 +#define HFNUM_MAX_FRNUM 0x3fff + +#define HPTXSTS HSOTG_REG(0x0410) +#define TXSTS_QTOP_ODD BIT(31) +#define TXSTS_QTOP_CHNEP_MASK (0xf << 27) +#define TXSTS_QTOP_CHNEP_SHIFT 27 +#define TXSTS_QTOP_TOKEN_MASK (0x3 << 25) +#define TXSTS_QTOP_TOKEN_SHIFT 25 +#define TXSTS_QTOP_TERMINATE BIT(24) +#define TXSTS_QSPCAVAIL_MASK (0xff << 16) +#define TXSTS_QSPCAVAIL_SHIFT 16 +#define TXSTS_FSPCAVAIL_MASK (0xffff << 0) +#define TXSTS_FSPCAVAIL_SHIFT 0 + +#define HAINT HSOTG_REG(0x0414) +#define HAINTMSK HSOTG_REG(0x0418) +#define HFLBADDR HSOTG_REG(0x041c) + +#define HPRT0 HSOTG_REG(0x0440) +#define HPRT0_SPD_MASK (0x3 << 17) +#define HPRT0_SPD_SHIFT 17 +#define HPRT0_SPD_HIGH_SPEED 0 +#define HPRT0_SPD_FULL_SPEED 1 +#define HPRT0_SPD_LOW_SPEED 2 +#define HPRT0_TSTCTL_MASK (0xf << 13) +#define HPRT0_TSTCTL_SHIFT 13 +#define HPRT0_PWR BIT(12) +#define HPRT0_LNSTS_MASK (0x3 << 10) +#define HPRT0_LNSTS_SHIFT 10 +#define HPRT0_RST BIT(8) +#define HPRT0_SUSP BIT(7) +#define HPRT0_RES BIT(6) +#define HPRT0_OVRCURRCHG BIT(5) +#define HPRT0_OVRCURRACT BIT(4) +#define HPRT0_ENACHG BIT(3) +#define HPRT0_ENA BIT(2) +#define HPRT0_CONNDET BIT(1) +#define HPRT0_CONNSTS BIT(0) + +#define HCCHAR(_ch) HSOTG_REG(0x0500 + 0x20 * (_ch)) +#define HCCHAR_CHENA BIT(31) +#define HCCHAR_CHDIS BIT(30) +#define HCCHAR_ODDFRM BIT(29) +#define HCCHAR_DEVADDR_MASK (0x7f << 22) +#define HCCHAR_DEVADDR_SHIFT 22 +#define HCCHAR_MULTICNT_MASK (0x3 << 20) +#define HCCHAR_MULTICNT_SHIFT 20 +#define HCCHAR_EPTYPE_MASK (0x3 << 18) +#define HCCHAR_EPTYPE_SHIFT 18 +#define HCCHAR_LSPDDEV BIT(17) +#define HCCHAR_EPDIR BIT(15) +#define HCCHAR_EPNUM_MASK (0xf << 11) +#define HCCHAR_EPNUM_SHIFT 11 +#define HCCHAR_MPS_MASK (0x7ff << 0) +#define HCCHAR_MPS_SHIFT 0 + +#define HCSPLT(_ch) HSOTG_REG(0x0504 + 0x20 * (_ch)) +#define HCSPLT_SPLTENA BIT(31) +#define HCSPLT_COMPSPLT BIT(16) +#define HCSPLT_XACTPOS_MASK (0x3 << 14) +#define HCSPLT_XACTPOS_SHIFT 14 +#define HCSPLT_XACTPOS_MID 0 +#define HCSPLT_XACTPOS_END 1 +#define HCSPLT_XACTPOS_BEGIN 2 +#define HCSPLT_XACTPOS_ALL 3 +#define HCSPLT_HUBADDR_MASK (0x7f << 7) +#define HCSPLT_HUBADDR_SHIFT 7 +#define HCSPLT_PRTADDR_MASK (0x7f << 0) +#define HCSPLT_PRTADDR_SHIFT 0 + +#define HCINT(_ch) HSOTG_REG(0x0508 + 0x20 * (_ch)) +#define HCINTMSK(_ch) HSOTG_REG(0x050c + 0x20 * (_ch)) +#define HCINTMSK_RESERVED14_31 (0x3ffff << 14) +#define HCINTMSK_FRM_LIST_ROLL BIT(13) +#define HCINTMSK_XCS_XACT BIT(12) +#define HCINTMSK_BNA BIT(11) +#define HCINTMSK_DATATGLERR BIT(10) +#define HCINTMSK_FRMOVRUN BIT(9) +#define HCINTMSK_BBLERR BIT(8) +#define HCINTMSK_XACTERR BIT(7) +#define HCINTMSK_NYET BIT(6) +#define HCINTMSK_ACK BIT(5) +#define HCINTMSK_NAK BIT(4) +#define HCINTMSK_STALL BIT(3) +#define HCINTMSK_AHBERR BIT(2) +#define HCINTMSK_CHHLTD BIT(1) +#define HCINTMSK_XFERCOMPL BIT(0) + +#define HCTSIZ(_ch) HSOTG_REG(0x0510 + 0x20 * (_ch)) +#define TSIZ_DOPNG BIT(31) +#define TSIZ_SC_MC_PID_MASK (0x3 << 29) +#define TSIZ_SC_MC_PID_SHIFT 29 +#define TSIZ_SC_MC_PID_DATA0 0 +#define TSIZ_SC_MC_PID_DATA2 1 +#define TSIZ_SC_MC_PID_DATA1 2 +#define TSIZ_SC_MC_PID_MDATA 3 +#define TSIZ_SC_MC_PID_SETUP 3 +#define TSIZ_PKTCNT_MASK (0x3ff << 19) +#define TSIZ_PKTCNT_SHIFT 19 +#define TSIZ_NTD_MASK (0xff << 8) +#define TSIZ_NTD_SHIFT 8 +#define TSIZ_SCHINFO_MASK (0xff << 0) +#define TSIZ_SCHINFO_SHIFT 0 +#define TSIZ_XFERSIZE_MASK (0x7ffff << 0) +#define TSIZ_XFERSIZE_SHIFT 0 + +#define HCDMA(_ch) HSOTG_REG(0x0514 + 0x20 * (_ch)) + +#define HCDMAB(_ch) HSOTG_REG(0x051c + 0x20 * (_ch)) + +#define HCFIFO(_ch) HSOTG_REG(0x1000 + 0x1000 * (_ch)) + +/** + * struct dwc2_dma_desc - DMA descriptor structure, + * used for both host and gadget modes + * + * @status: DMA descriptor status quadlet + * @buf: DMA descriptor data buffer pointer + * + * DMA Descriptor structure contains two quadlets: + * Status quadlet and Data buffer pointer. + */ +struct dwc2_dma_desc { + uint32_t status; + uint32_t buf; +} __packed; + +/* Host Mode DMA descriptor status quadlet */ + +#define HOST_DMA_A BIT(31) +#define HOST_DMA_STS_MASK (0x3 << 28) +#define HOST_DMA_STS_SHIFT 28 +#define HOST_DMA_STS_PKTERR BIT(28) +#define HOST_DMA_EOL BIT(26) +#define HOST_DMA_IOC BIT(25) +#define HOST_DMA_SUP BIT(24) +#define HOST_DMA_ALT_QTD BIT(23) +#define HOST_DMA_QTD_OFFSET_MASK (0x3f << 17) +#define HOST_DMA_QTD_OFFSET_SHIFT 17 +#define HOST_DMA_ISOC_NBYTES_MASK (0xfff << 0) +#define HOST_DMA_ISOC_NBYTES_SHIFT 0 +#define HOST_DMA_NBYTES_MASK (0x1ffff << 0) +#define HOST_DMA_NBYTES_SHIFT 0 +#define HOST_DMA_NBYTES_LIMIT 131071 + +/* Device Mode DMA descriptor status quadlet */ + +#define DEV_DMA_BUFF_STS_MASK (0x3 << 30) +#define DEV_DMA_BUFF_STS_SHIFT 30 +#define DEV_DMA_BUFF_STS_HREADY 0 +#define DEV_DMA_BUFF_STS_DMABUSY 1 +#define DEV_DMA_BUFF_STS_DMADONE 2 +#define DEV_DMA_BUFF_STS_HBUSY 3 +#define DEV_DMA_STS_MASK (0x3 << 28) +#define DEV_DMA_STS_SHIFT 28 +#define DEV_DMA_STS_SUCC 0 +#define DEV_DMA_STS_BUFF_FLUSH 1 +#define DEV_DMA_STS_BUFF_ERR 3 +#define DEV_DMA_L BIT(27) +#define DEV_DMA_SHORT BIT(26) +#define DEV_DMA_IOC BIT(25) +#define DEV_DMA_SR BIT(24) +#define DEV_DMA_MTRF BIT(23) +#define DEV_DMA_ISOC_PID_MASK (0x3 << 23) +#define DEV_DMA_ISOC_PID_SHIFT 23 +#define DEV_DMA_ISOC_PID_DATA0 0 +#define DEV_DMA_ISOC_PID_DATA2 1 +#define DEV_DMA_ISOC_PID_DATA1 2 +#define DEV_DMA_ISOC_PID_MDATA 3 +#define DEV_DMA_ISOC_FRNUM_MASK (0x7ff << 12) +#define DEV_DMA_ISOC_FRNUM_SHIFT 12 +#define DEV_DMA_ISOC_TX_NBYTES_MASK (0xfff << 0) +#define DEV_DMA_ISOC_TX_NBYTES_LIMIT 0xfff +#define DEV_DMA_ISOC_RX_NBYTES_MASK (0x7ff << 0) +#define DEV_DMA_ISOC_RX_NBYTES_LIMIT 0x7ff +#define DEV_DMA_ISOC_NBYTES_SHIFT 0 +#define DEV_DMA_NBYTES_MASK (0xffff << 0) +#define DEV_DMA_NBYTES_SHIFT 0 +#define DEV_DMA_NBYTES_LIMIT 0xffff + +#define MAX_DMA_DESC_NUM_GENERIC 64 +#define MAX_DMA_DESC_NUM_HS_ISOC 256 + +#endif /* __DWC2_HW_H__ */ -- 2.20.1 From: Paul Zimmerman <pauldzim@gmail.com> Add the dwc-hsotg (dwc2) USB host controller state definitions. Mostly based on hw/usb/hcd-ehci.h. Signed-off-by: Paul Zimmerman <pauldzim@gmail.com> Message-id: 20200520235349.21215-4-pauldzim@gmail.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/usb/hcd-dwc2.h | 190 ++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 190 insertions(+) create mode 100644 hw/usb/hcd-dwc2.h diff --git a/hw/usb/hcd-dwc2.h b/hw/usb/hcd-dwc2.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/usb/hcd-dwc2.h @@ -XXX,XX +XXX,XX @@ +/* + * dwc-hsotg (dwc2) USB host controller state definitions + * + * Based on hw/usb/hcd-ehci.h + * + * Copyright (c) 2020 Paul Zimmerman <pauldzim@gmail.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#ifndef HW_USB_DWC2_H +#define HW_USB_DWC2_H + +#include "qemu/timer.h" +#include "hw/irq.h" +#include "hw/sysbus.h" +#include "hw/usb.h" +#include "sysemu/dma.h" + +#define DWC2_MMIO_SIZE 0x11000 + +#define DWC2_NB_CHAN 8 /* Number of host channels */ +#define DWC2_MAX_XFER_SIZE 65536 /* Max transfer size expected in HCTSIZ */ + +typedef struct DWC2Packet DWC2Packet; +typedef struct DWC2State DWC2State; +typedef struct DWC2Class DWC2Class; + +enum async_state { + DWC2_ASYNC_NONE = 0, + DWC2_ASYNC_INITIALIZED, + DWC2_ASYNC_INFLIGHT, + DWC2_ASYNC_FINISHED, +}; + +struct DWC2Packet { + USBPacket packet; + uint32_t devadr; + uint32_t epnum; + uint32_t epdir; + uint32_t mps; + uint32_t pid; + uint32_t index; + uint32_t pcnt; + uint32_t len; + int32_t async; + bool small; + bool needs_service; +}; + +struct DWC2State { + /*< private >*/ + SysBusDevice parent_obj; + + /*< public >*/ + USBBus bus; + qemu_irq irq; + MemoryRegion *dma_mr; + AddressSpace dma_as; + MemoryRegion container; + MemoryRegion hsotg; + MemoryRegion fifos; + + union { +#define DWC2_GLBREG_SIZE 0x70 + uint32_t glbreg[DWC2_GLBREG_SIZE / sizeof(uint32_t)]; + struct { + uint32_t gotgctl; /* 00 */ + uint32_t gotgint; /* 04 */ + uint32_t gahbcfg; /* 08 */ + uint32_t gusbcfg; /* 0c */ + uint32_t grstctl; /* 10 */ + uint32_t gintsts; /* 14 */ + uint32_t gintmsk; /* 18 */ + uint32_t grxstsr; /* 1c */ + uint32_t grxstsp; /* 20 */ + uint32_t grxfsiz; /* 24 */ + uint32_t gnptxfsiz; /* 28 */ + uint32_t gnptxsts; /* 2c */ + uint32_t gi2cctl; /* 30 */ + uint32_t gpvndctl; /* 34 */ + uint32_t ggpio; /* 38 */ + uint32_t guid; /* 3c */ + uint32_t gsnpsid; /* 40 */ + uint32_t ghwcfg1; /* 44 */ + uint32_t ghwcfg2; /* 48 */ + uint32_t ghwcfg3; /* 4c */ + uint32_t ghwcfg4; /* 50 */ + uint32_t glpmcfg; /* 54 */ + uint32_t gpwrdn; /* 58 */ + uint32_t gdfifocfg; /* 5c */ + uint32_t gadpctl; /* 60 */ + uint32_t grefclk; /* 64 */ + uint32_t gintmsk2; /* 68 */ + uint32_t gintsts2; /* 6c */ + }; + }; + + union { +#define DWC2_FSZREG_SIZE 0x04 + uint32_t fszreg[DWC2_FSZREG_SIZE / sizeof(uint32_t)]; + struct { + uint32_t hptxfsiz; /* 100 */ + }; + }; + + union { +#define DWC2_HREG0_SIZE 0x44 + uint32_t hreg0[DWC2_HREG0_SIZE / sizeof(uint32_t)]; + struct { + uint32_t hcfg; /* 400 */ + uint32_t hfir; /* 404 */ + uint32_t hfnum; /* 408 */ + uint32_t rsvd0; /* 40c */ + uint32_t hptxsts; /* 410 */ + uint32_t haint; /* 414 */ + uint32_t haintmsk; /* 418 */ + uint32_t hflbaddr; /* 41c */ + uint32_t rsvd1[8]; /* 420-43c */ + uint32_t hprt0; /* 440 */ + }; + }; + +#define DWC2_HREG1_SIZE (0x20 * DWC2_NB_CHAN) + uint32_t hreg1[DWC2_HREG1_SIZE / sizeof(uint32_t)]; + +#define hcchar(_ch) hreg1[((_ch) << 3) + 0] /* 500, 520, ... */ +#define hcsplt(_ch) hreg1[((_ch) << 3) + 1] /* 504, 524, ... */ +#define hcint(_ch) hreg1[((_ch) << 3) + 2] /* 508, 528, ... */ +#define hcintmsk(_ch) hreg1[((_ch) << 3) + 3] /* 50c, 52c, ... */ +#define hctsiz(_ch) hreg1[((_ch) << 3) + 4] /* 510, 530, ... */ +#define hcdma(_ch) hreg1[((_ch) << 3) + 5] /* 514, 534, ... */ +#define hcdmab(_ch) hreg1[((_ch) << 3) + 7] /* 51c, 53c, ... */ + + union { +#define DWC2_PCGREG_SIZE 0x08 + uint32_t pcgreg[DWC2_PCGREG_SIZE / sizeof(uint32_t)]; + struct { + uint32_t pcgctl; /* e00 */ + uint32_t pcgcctl1; /* e04 */ + }; + }; + + /* TODO - implement FIFO registers for slave mode */ +#define DWC2_HFIFO_SIZE (0x1000 * DWC2_NB_CHAN) + + /* + * Internal state + */ + QEMUTimer *eof_timer; + QEMUTimer *frame_timer; + QEMUBH *async_bh; + int64_t sof_time; + int64_t usb_frame_time; + int64_t usb_bit_time; + uint32_t usb_version; + uint16_t frame_number; + uint16_t fi; + uint16_t next_chan; + bool working; + USBPort uport; + DWC2Packet packet[DWC2_NB_CHAN]; /* one packet per chan */ + uint8_t usb_buf[DWC2_NB_CHAN][DWC2_MAX_XFER_SIZE]; /* one buffer per chan */ +}; + +struct DWC2Class { + /*< private >*/ + SysBusDeviceClass parent_class; + ResettablePhases parent_phases; + + /*< public >*/ +}; + +#define TYPE_DWC2_USB "dwc2-usb" +#define DWC2_USB(obj) \ + OBJECT_CHECK(DWC2State, (obj), TYPE_DWC2_USB) +#define DWC2_CLASS(klass) \ + OBJECT_CLASS_CHECK(DWC2Class, (klass), TYPE_DWC2_USB) +#define DWC2_GET_CLASS(obj) \ + OBJECT_GET_CLASS(DWC2Class, (obj), TYPE_DWC2_USB) + +#endif -- 2.20.1 From: Paul Zimmerman <pauldzim@gmail.com> Add the dwc-hsotg (dwc2) USB host controller emulation code. Based on hw/usb/hcd-ehci.c and hw/usb/hcd-ohci.c. Note that to use this with the dwc-otg driver in the Raspbian kernel, you must pass the option "dwc_otg.fiq_fsm_enable=0" on the kernel command line. Emulation of slave mode and of descriptor-DMA mode has not been implemented yet. These modes are seldom used. I have used some on-line sources of information while developing this emulation, including: http://www.capital-micro.com/PDF/CME-M7_Family_User_Guide_EN.pdf which has a pretty complete description of the controller starting on page 370. https://sourceforge.net/p/wive-ng/wive-ng-mt/ci/master/tree/docs/DataSheets/RT3050_5x_V2.0_081408_0902.pdf which has a description of the controller registers starting on page 130. Thanks to Felippe Mathieu-Daude for providing a cleaner method of implementing the memory regions for the controller registers. Signed-off-by: Paul Zimmerman <pauldzim@gmail.com> Message-id: 20200520235349.21215-5-pauldzim@gmail.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/usb/hcd-dwc2.c | 1417 ++++++++++++++++++++++++++++++++++++++++++ hw/usb/Kconfig | 5 + hw/usb/Makefile.objs | 1 + hw/usb/trace-events | 50 ++ 4 files changed, 1473 insertions(+) create mode 100644 hw/usb/hcd-dwc2.c diff --git a/hw/usb/hcd-dwc2.c b/hw/usb/hcd-dwc2.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/usb/hcd-dwc2.c @@ -XXX,XX +XXX,XX @@ +/* + * dwc-hsotg (dwc2) USB host controller emulation + * + * Based on hw/usb/hcd-ehci.c and hw/usb/hcd-ohci.c + * + * Note that to use this emulation with the dwc-otg driver in the + * Raspbian kernel, you must pass the option "dwc_otg.fiq_fsm_enable=0" + * on the kernel command line. + * + * Some useful documentation used to develop this emulation can be + * found online (as of April 2020) at: + * + * http://www.capital-micro.com/PDF/CME-M7_Family_User_Guide_EN.pdf + * which has a pretty complete description of the controller starting + * on page 370. + * + * https://sourceforge.net/p/wive-ng/wive-ng-mt/ci/master/tree/docs/DataSheets/RT3050_5x_V2.0_081408_0902.pdf + * which has a description of the controller registers starting on + * page 130. + * + * Copyright (c) 2020 Paul Zimmerman <pauldzim@gmail.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include "qemu/osdep.h" +#include "qemu/units.h" +#include "qapi/error.h" +#include "hw/usb/dwc2-regs.h" +#include "hw/usb/hcd-dwc2.h" +#include "migration/vmstate.h" +#include "trace.h" +#include "qemu/log.h" +#include "qemu/error-report.h" +#include "qemu/main-loop.h" +#include "hw/qdev-properties.h" + +#define USB_HZ_FS 12000000 +#define USB_HZ_HS 96000000 +#define USB_FRMINTVL 12000 + +/* nifty macros from Arnon's EHCI version */ +#define get_field(data, field) \ + (((data) & field##_MASK) >> field##_SHIFT) + +#define set_field(data, newval, field) do { \ + uint32_t val = *(data); \ + val &= ~field##_MASK; \ + val |= ((newval) << field##_SHIFT) & field##_MASK; \ + *(data) = val; \ +} while (0) + +#define get_bit(data, bitmask) \ + (!!((data) & (bitmask))) + +/* update irq line */ +static inline void dwc2_update_irq(DWC2State *s) +{ + static int oldlevel; + int level = 0; + + if ((s->gintsts & s->gintmsk) && (s->gahbcfg & GAHBCFG_GLBL_INTR_EN)) { + level = 1; + } + if (level != oldlevel) { + oldlevel = level; + trace_usb_dwc2_update_irq(level); + qemu_set_irq(s->irq, level); + } +} + +/* flag interrupt condition */ +static inline void dwc2_raise_global_irq(DWC2State *s, uint32_t intr) +{ + if (!(s->gintsts & intr)) { + s->gintsts |= intr; + trace_usb_dwc2_raise_global_irq(intr); + dwc2_update_irq(s); + } +} + +static inline void dwc2_lower_global_irq(DWC2State *s, uint32_t intr) +{ + if (s->gintsts & intr) { + s->gintsts &= ~intr; + trace_usb_dwc2_lower_global_irq(intr); + dwc2_update_irq(s); + } +} + +static inline void dwc2_raise_host_irq(DWC2State *s, uint32_t host_intr) +{ + if (!(s->haint & host_intr)) { + s->haint |= host_intr; + s->haint &= 0xffff; + trace_usb_dwc2_raise_host_irq(host_intr); + if (s->haint & s->haintmsk) { + dwc2_raise_global_irq(s, GINTSTS_HCHINT); + } + } +} + +static inline void dwc2_lower_host_irq(DWC2State *s, uint32_t host_intr) +{ + if (s->haint & host_intr) { + s->haint &= ~host_intr; + trace_usb_dwc2_lower_host_irq(host_intr); + if (!(s->haint & s->haintmsk)) { + dwc2_lower_global_irq(s, GINTSTS_HCHINT); + } + } +} + +static inline void dwc2_update_hc_irq(DWC2State *s, int index) +{ + uint32_t host_intr = 1 << (index >> 3); + + if (s->hreg1[index + 2] & s->hreg1[index + 3]) { + dwc2_raise_host_irq(s, host_intr); + } else { + dwc2_lower_host_irq(s, host_intr); + } +} + +/* set a timer for EOF */ +static void dwc2_eof_timer(DWC2State *s) +{ + timer_mod(s->eof_timer, s->sof_time + s->usb_frame_time); +} + +/* Set a timer for EOF and generate SOF event */ +static void dwc2_sof(DWC2State *s) +{ + s->sof_time += s->usb_frame_time; + trace_usb_dwc2_sof(s->sof_time); + dwc2_eof_timer(s); + dwc2_raise_global_irq(s, GINTSTS_SOF); +} + +/* Do frame processing on frame boundary */ +static void dwc2_frame_boundary(void *opaque) +{ + DWC2State *s = opaque; + int64_t now; + uint16_t frcnt; + + now = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL); + + /* Frame boundary, so do EOF stuff here */ + + /* Increment frame number */ + frcnt = (uint16_t)((now - s->sof_time) / s->fi); + s->frame_number = (s->frame_number + frcnt) & 0xffff; + s->hfnum = s->frame_number & HFNUM_MAX_FRNUM; + + /* Do SOF stuff here */ + dwc2_sof(s); +} + +/* Start sending SOF tokens on the USB bus */ +static void dwc2_bus_start(DWC2State *s) +{ + trace_usb_dwc2_bus_start(); + s->sof_time = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL); + dwc2_eof_timer(s); +} + +/* Stop sending SOF tokens on the USB bus */ +static void dwc2_bus_stop(DWC2State *s) +{ + trace_usb_dwc2_bus_stop(); + timer_del(s->eof_timer); +} + +static USBDevice *dwc2_find_device(DWC2State *s, uint8_t addr) +{ + USBDevice *dev; + + trace_usb_dwc2_find_device(addr); + + if (!(s->hprt0 & HPRT0_ENA)) { + trace_usb_dwc2_port_disabled(0); + } else { + dev = usb_find_device(&s->uport, addr); + if (dev != NULL) { + trace_usb_dwc2_device_found(0); + return dev; + } + } + + trace_usb_dwc2_device_not_found(); + return NULL; +} + +static const char *pstatus[] = { + "USB_RET_SUCCESS", "USB_RET_NODEV", "USB_RET_NAK", "USB_RET_STALL", + "USB_RET_BABBLE", "USB_RET_IOERROR", "USB_RET_ASYNC", + "USB_RET_ADD_TO_QUEUE", "USB_RET_REMOVE_FROM_QUEUE" +}; + +static uint32_t pintr[] = { + HCINTMSK_XFERCOMPL, HCINTMSK_XACTERR, HCINTMSK_NAK, HCINTMSK_STALL, + HCINTMSK_BBLERR, HCINTMSK_XACTERR, HCINTMSK_XACTERR, HCINTMSK_XACTERR, + HCINTMSK_XACTERR +}; + +static const char *types[] = { + "Ctrl", "Isoc", "Bulk", "Intr" +}; + +static const char *dirs[] = { + "Out", "In" +}; + +static void dwc2_handle_packet(DWC2State *s, uint32_t devadr, USBDevice *dev, + USBEndpoint *ep, uint32_t index, bool send) +{ + DWC2Packet *p; + uint32_t hcchar = s->hreg1[index]; + uint32_t hctsiz = s->hreg1[index + 4]; + uint32_t hcdma = s->hreg1[index + 5]; + uint32_t chan, epnum, epdir, eptype, mps, pid, pcnt, len, tlen, intr = 0; + uint32_t tpcnt, stsidx, actual = 0; + bool do_intr = false, done = false; + + epnum = get_field(hcchar, HCCHAR_EPNUM); + epdir = get_bit(hcchar, HCCHAR_EPDIR); + eptype = get_field(hcchar, HCCHAR_EPTYPE); + mps = get_field(hcchar, HCCHAR_MPS); + pid = get_field(hctsiz, TSIZ_SC_MC_PID); + pcnt = get_field(hctsiz, TSIZ_PKTCNT); + len = get_field(hctsiz, TSIZ_XFERSIZE); + assert(len <= DWC2_MAX_XFER_SIZE); + chan = index >> 3; + p = &s->packet[chan]; + + trace_usb_dwc2_handle_packet(chan, dev, &p->packet, epnum, types[eptype], + dirs[epdir], mps, len, pcnt); + + if (eptype == USB_ENDPOINT_XFER_CONTROL && pid == TSIZ_SC_MC_PID_SETUP) { + pid = USB_TOKEN_SETUP; + } else { + pid = epdir ? USB_TOKEN_IN : USB_TOKEN_OUT; + } + + if (send) { + tlen = len; + if (p->small) { + if (tlen > mps) { + tlen = mps; + } + } + + if (pid != USB_TOKEN_IN) { + trace_usb_dwc2_memory_read(hcdma, tlen); + if (dma_memory_read(&s->dma_as, hcdma, + s->usb_buf[chan], tlen) != MEMTX_OK) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: dma_memory_read failed\n", + __func__); + } + } + + usb_packet_init(&p->packet); + usb_packet_setup(&p->packet, pid, ep, 0, hcdma, + pid != USB_TOKEN_IN, true); + usb_packet_addbuf(&p->packet, s->usb_buf[chan], tlen); + p->async = DWC2_ASYNC_NONE; + usb_handle_packet(dev, &p->packet); + } else { + tlen = p->len; + } + + stsidx = -p->packet.status; + assert(stsidx < sizeof(pstatus) / sizeof(*pstatus)); + actual = p->packet.actual_length; + trace_usb_dwc2_packet_status(pstatus[stsidx], actual); + +babble: + if (p->packet.status != USB_RET_SUCCESS && + p->packet.status != USB_RET_NAK && + p->packet.status != USB_RET_STALL && + p->packet.status != USB_RET_ASYNC) { + trace_usb_dwc2_packet_error(pstatus[stsidx]); + } + + if (p->packet.status == USB_RET_ASYNC) { + trace_usb_dwc2_async_packet(&p->packet, chan, dev, epnum, + dirs[epdir], tlen); + usb_device_flush_ep_queue(dev, ep); + assert(p->async != DWC2_ASYNC_INFLIGHT); + p->devadr = devadr; + p->epnum = epnum; + p->epdir = epdir; + p->mps = mps; + p->pid = pid; + p->index = index; + p->pcnt = pcnt; + p->len = tlen; + p->async = DWC2_ASYNC_INFLIGHT; + p->needs_service = false; + return; + } + + if (p->packet.status == USB_RET_SUCCESS) { + if (actual > tlen) { + p->packet.status = USB_RET_BABBLE; + goto babble; + } + + if (pid == USB_TOKEN_IN) { + trace_usb_dwc2_memory_write(hcdma, actual); + if (dma_memory_write(&s->dma_as, hcdma, s->usb_buf[chan], + actual) != MEMTX_OK) { + qemu_log_mask(LOG_GUEST_ERROR, "%s: dma_memory_write failed\n", + __func__); + } + } + + tpcnt = actual / mps; + if (actual % mps) { + tpcnt++; + if (pid == USB_TOKEN_IN) { + done = true; + } + } + + pcnt -= tpcnt < pcnt ? tpcnt : pcnt; + set_field(&hctsiz, pcnt, TSIZ_PKTCNT); + len -= actual < len ? actual : len; + set_field(&hctsiz, len, TSIZ_XFERSIZE); + s->hreg1[index + 4] = hctsiz; + hcdma += actual; + s->hreg1[index + 5] = hcdma; + + if (!pcnt || len == 0 || actual == 0) { + done = true; + } + } else { + intr |= pintr[stsidx]; + if (p->packet.status == USB_RET_NAK && + (eptype == USB_ENDPOINT_XFER_CONTROL || + eptype == USB_ENDPOINT_XFER_BULK)) { + /* + * for ctrl/bulk, automatically retry on NAK, + * but send the interrupt anyway + */ + intr &= ~HCINTMSK_RESERVED14_31; + s->hreg1[index + 2] |= intr; + do_intr = true; + } else { + intr |= HCINTMSK_CHHLTD; + done = true; + } + } + + usb_packet_cleanup(&p->packet); + + if (done) { + hcchar &= ~HCCHAR_CHENA; + s->hreg1[index] = hcchar; + if (!(intr & HCINTMSK_CHHLTD)) { + intr |= HCINTMSK_CHHLTD | HCINTMSK_XFERCOMPL; + } + intr &= ~HCINTMSK_RESERVED14_31; + s->hreg1[index + 2] |= intr; + p->needs_service = false; + trace_usb_dwc2_packet_done(pstatus[stsidx], actual, len, pcnt); + dwc2_update_hc_irq(s, index); + return; + } + + p->devadr = devadr; + p->epnum = epnum; + p->epdir = epdir; + p->mps = mps; + p->pid = pid; + p->index = index; + p->pcnt = pcnt; + p->len = len; + p->needs_service = true; + trace_usb_dwc2_packet_next(pstatus[stsidx], len, pcnt); + if (do_intr) { + dwc2_update_hc_irq(s, index); + } +} + +/* Attach or detach a device on root hub */ + +static const char *speeds[] = { + "low", "full", "high" +}; + +static void dwc2_attach(USBPort *port) +{ + DWC2State *s = port->opaque; + int hispd = 0; + + trace_usb_dwc2_attach(port); + assert(port->index == 0); + + if (!port->dev || !port->dev->attached) { + return; + } + + assert(port->dev->speed <= USB_SPEED_HIGH); + trace_usb_dwc2_attach_speed(speeds[port->dev->speed]); + s->hprt0 &= ~HPRT0_SPD_MASK; + + switch (port->dev->speed) { + case USB_SPEED_LOW: + s->hprt0 |= HPRT0_SPD_LOW_SPEED << HPRT0_SPD_SHIFT; + break; + case USB_SPEED_FULL: + s->hprt0 |= HPRT0_SPD_FULL_SPEED << HPRT0_SPD_SHIFT; + break; + case USB_SPEED_HIGH: + s->hprt0 |= HPRT0_SPD_HIGH_SPEED << HPRT0_SPD_SHIFT; + hispd = 1; + break; + } + + if (hispd) { + s->usb_frame_time = NANOSECONDS_PER_SECOND / 8000; /* 125000 */ + if (NANOSECONDS_PER_SECOND >= USB_HZ_HS) { + s->usb_bit_time = NANOSECONDS_PER_SECOND / USB_HZ_HS; /* 10.4 */ + } else { + s->usb_bit_time = 1; + } + } else { + s->usb_frame_time = NANOSECONDS_PER_SECOND / 1000; /* 1000000 */ + if (NANOSECONDS_PER_SECOND >= USB_HZ_FS) { + s->usb_bit_time = NANOSECONDS_PER_SECOND / USB_HZ_FS; /* 83.3 */ + } else { + s->usb_bit_time = 1; + } + } + + s->fi = USB_FRMINTVL - 1; + s->hprt0 |= HPRT0_CONNDET | HPRT0_CONNSTS; + + dwc2_bus_start(s); + dwc2_raise_global_irq(s, GINTSTS_PRTINT); +} + +static void dwc2_detach(USBPort *port) +{ + DWC2State *s = port->opaque; + + trace_usb_dwc2_detach(port); + assert(port->index == 0); + + dwc2_bus_stop(s); + + s->hprt0 &= ~(HPRT0_SPD_MASK | HPRT0_SUSP | HPRT0_ENA | HPRT0_CONNSTS); + s->hprt0 |= HPRT0_CONNDET | HPRT0_ENACHG; + + dwc2_raise_global_irq(s, GINTSTS_PRTINT); +} + +static void dwc2_child_detach(USBPort *port, USBDevice *child) +{ + trace_usb_dwc2_child_detach(port, child); + assert(port->index == 0); +} + +static void dwc2_wakeup(USBPort *port) +{ + DWC2State *s = port->opaque; + + trace_usb_dwc2_wakeup(port); + assert(port->index == 0); + + if (s->hprt0 & HPRT0_SUSP) { + s->hprt0 |= HPRT0_RES; + dwc2_raise_global_irq(s, GINTSTS_PRTINT); + } + + qemu_bh_schedule(s->async_bh); +} + +static void dwc2_async_packet_complete(USBPort *port, USBPacket *packet) +{ + DWC2State *s = port->opaque; + DWC2Packet *p; + USBDevice *dev; + USBEndpoint *ep; + + assert(port->index == 0); + p = container_of(packet, DWC2Packet, packet); + dev = dwc2_find_device(s, p->devadr); + ep = usb_ep_get(dev, p->pid, p->epnum); + trace_usb_dwc2_async_packet_complete(port, packet, p->index >> 3, dev, + p->epnum, dirs[p->epdir], p->len); + assert(p->async == DWC2_ASYNC_INFLIGHT); + + if (packet->status == USB_RET_REMOVE_FROM_QUEUE) { + usb_cancel_packet(packet); + usb_packet_cleanup(packet); + return; + } + + dwc2_handle_packet(s, p->devadr, dev, ep, p->index, false); + + p->async = DWC2_ASYNC_FINISHED; + qemu_bh_schedule(s->async_bh); +} + +static USBPortOps dwc2_port_ops = { + .attach = dwc2_attach, + .detach = dwc2_detach, + .child_detach = dwc2_child_detach, + .wakeup = dwc2_wakeup, + .complete = dwc2_async_packet_complete, +}; + +static uint32_t dwc2_get_frame_remaining(DWC2State *s) +{ + uint32_t fr = 0; + int64_t tks; + + tks = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) - s->sof_time; + if (tks < 0) { + tks = 0; + } + + /* avoid muldiv if possible */ + if (tks >= s->usb_frame_time) { + goto out; + } + if (tks < s->usb_bit_time) { + fr = s->fi; + goto out; + } + + /* tks = number of ns since SOF, divided by 83 (fs) or 10 (hs) */ + tks = tks / s->usb_bit_time; + if (tks >= (int64_t)s->fi) { + goto out; + } + + /* remaining = frame interval minus tks */ + fr = (uint32_t)((int64_t)s->fi - tks); + +out: + return fr; +} + +static void dwc2_work_bh(void *opaque) +{ + DWC2State *s = opaque; + DWC2Packet *p; + USBDevice *dev; + USBEndpoint *ep; + int64_t t_now, expire_time; + int chan; + bool found = false; + + trace_usb_dwc2_work_bh(); + if (s->working) { + return; + } + s->working = true; + + t_now = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL); + chan = s->next_chan; + + do { + p = &s->packet[chan]; + if (p->needs_service) { + dev = dwc2_find_device(s, p->devadr); + ep = usb_ep_get(dev, p->pid, p->epnum); + trace_usb_dwc2_work_bh_service(s->next_chan, chan, dev, p->epnum); + dwc2_handle_packet(s, p->devadr, dev, ep, p->index, true); + found = true; + } + if (++chan == DWC2_NB_CHAN) { + chan = 0; + } + if (found) { + s->next_chan = chan; + trace_usb_dwc2_work_bh_next(chan); + } + } while (chan != s->next_chan); + + if (found) { + expire_time = t_now + NANOSECONDS_PER_SECOND / 4000; + timer_mod(s->frame_timer, expire_time); + } + s->working = false; +} + +static void dwc2_enable_chan(DWC2State *s, uint32_t index) +{ + USBDevice *dev; + USBEndpoint *ep; + uint32_t hcchar; + uint32_t hctsiz; + uint32_t devadr, epnum, epdir, eptype, pid, len; + DWC2Packet *p; + + assert((index >> 3) < DWC2_NB_CHAN); + p = &s->packet[index >> 3]; + hcchar = s->hreg1[index]; + hctsiz = s->hreg1[index + 4]; + devadr = get_field(hcchar, HCCHAR_DEVADDR); + epnum = get_field(hcchar, HCCHAR_EPNUM); + epdir = get_bit(hcchar, HCCHAR_EPDIR); + eptype = get_field(hcchar, HCCHAR_EPTYPE); + pid = get_field(hctsiz, TSIZ_SC_MC_PID); + len = get_field(hctsiz, TSIZ_XFERSIZE); + + dev = dwc2_find_device(s, devadr); + + trace_usb_dwc2_enable_chan(index >> 3, dev, &p->packet, epnum); + if (dev == NULL) { + return; + } + + if (eptype == USB_ENDPOINT_XFER_CONTROL && pid == TSIZ_SC_MC_PID_SETUP) { + pid = USB_TOKEN_SETUP; + } else { + pid = epdir ? USB_TOKEN_IN : USB_TOKEN_OUT; + } + + ep = usb_ep_get(dev, pid, epnum); + + /* + * Hack: Networking doesn't like us delivering large transfers, it kind + * of works but the latency is horrible. So if the transfer is <= the mtu + * size, we take that as a hint that this might be a network transfer, + * and do the transfer packet-by-packet. + */ + if (len > 1536) { + p->small = false; + } else { + p->small = true; + } + + dwc2_handle_packet(s, devadr, dev, ep, index, true); + qemu_bh_schedule(s->async_bh); +} + +static const char *glbregnm[] = { + "GOTGCTL ", "GOTGINT ", "GAHBCFG ", "GUSBCFG ", "GRSTCTL ", + "GINTSTS ", "GINTMSK ", "GRXSTSR ", "GRXSTSP ", "GRXFSIZ ", + "GNPTXFSIZ", "GNPTXSTS ", "GI2CCTL ", "GPVNDCTL ", "GGPIO ", + "GUID ", "GSNPSID ", "GHWCFG1 ", "GHWCFG2 ", "GHWCFG3 ", + "GHWCFG4 ", "GLPMCFG ", "GPWRDN ", "GDFIFOCFG", "GADPCTL ", + "GREFCLK ", "GINTMSK2 ", "GINTSTS2 " +}; + +static uint64_t dwc2_glbreg_read(void *ptr, hwaddr addr, int index, + unsigned size) +{ + DWC2State *s = ptr; + uint32_t val; + + assert(addr <= GINTSTS2); + val = s->glbreg[index]; + + switch (addr) { + case GRSTCTL: + /* clear any self-clearing bits that were set */ + val &= ~(GRSTCTL_TXFFLSH | GRSTCTL_RXFFLSH | GRSTCTL_IN_TKNQ_FLSH | + GRSTCTL_FRMCNTRRST | GRSTCTL_HSFTRST | GRSTCTL_CSFTRST); + s->glbreg[index] = val; + break; + default: + break; + } + + trace_usb_dwc2_glbreg_read(addr, glbregnm[index], val); + return val; +} + +static void dwc2_glbreg_write(void *ptr, hwaddr addr, int index, uint64_t val, + unsigned size) +{ + DWC2State *s = ptr; + uint64_t orig = val; + uint32_t *mmio; + uint32_t old; + int iflg = 0; + + assert(addr <= GINTSTS2); + mmio = &s->glbreg[index]; + old = *mmio; + + switch (addr) { + case GOTGCTL: + /* don't allow setting of read-only bits */ + val &= ~(GOTGCTL_MULT_VALID_BC_MASK | GOTGCTL_BSESVLD | + GOTGCTL_ASESVLD | GOTGCTL_DBNC_SHORT | GOTGCTL_CONID_B | + GOTGCTL_HSTNEGSCS | GOTGCTL_SESREQSCS); + /* don't allow clearing of read-only bits */ + val |= old & (GOTGCTL_MULT_VALID_BC_MASK | GOTGCTL_BSESVLD | + GOTGCTL_ASESVLD | GOTGCTL_DBNC_SHORT | GOTGCTL_CONID_B | + GOTGCTL_HSTNEGSCS | GOTGCTL_SESREQSCS); + break; + case GAHBCFG: + if ((val & GAHBCFG_GLBL_INTR_EN) && !(old & GAHBCFG_GLBL_INTR_EN)) { + iflg = 1; + } + break; + case GRSTCTL: + val |= GRSTCTL_AHBIDLE; + val &= ~GRSTCTL_DMAREQ; + if (!(old & GRSTCTL_TXFFLSH) && (val & GRSTCTL_TXFFLSH)) { + /* TODO - TX fifo flush */ + qemu_log_mask(LOG_UNIMP, "Tx FIFO flush not implemented\n"); + } + if (!(old & GRSTCTL_RXFFLSH) && (val & GRSTCTL_RXFFLSH)) { + /* TODO - RX fifo flush */ + qemu_log_mask(LOG_UNIMP, "Rx FIFO flush not implemented\n"); + } + if (!(old & GRSTCTL_IN_TKNQ_FLSH) && (val & GRSTCTL_IN_TKNQ_FLSH)) { + /* TODO - device IN token queue flush */ + qemu_log_mask(LOG_UNIMP, "Token queue flush not implemented\n"); + } + if (!(old & GRSTCTL_FRMCNTRRST) && (val & GRSTCTL_FRMCNTRRST)) { + /* TODO - host frame counter reset */ + qemu_log_mask(LOG_UNIMP, "Frame counter reset not implemented\n"); + } + if (!(old & GRSTCTL_HSFTRST) && (val & GRSTCTL_HSFTRST)) { + /* TODO - host soft reset */ + qemu_log_mask(LOG_UNIMP, "Host soft reset not implemented\n"); + } + if (!(old & GRSTCTL_CSFTRST) && (val & GRSTCTL_CSFTRST)) { + /* TODO - core soft reset */ + qemu_log_mask(LOG_UNIMP, "Core soft reset not implemented\n"); + } + /* don't allow clearing of self-clearing bits */ + val |= old & (GRSTCTL_TXFFLSH | GRSTCTL_RXFFLSH | + GRSTCTL_IN_TKNQ_FLSH | GRSTCTL_FRMCNTRRST | + GRSTCTL_HSFTRST | GRSTCTL_CSFTRST); + break; + case GINTSTS: + /* clear the write-1-to-clear bits */ + val |= ~old; + val = ~val; + /* don't allow clearing of read-only bits */ + val |= old & (GINTSTS_PTXFEMP | GINTSTS_HCHINT | GINTSTS_PRTINT | + GINTSTS_OEPINT | GINTSTS_IEPINT | GINTSTS_GOUTNAKEFF | + GINTSTS_GINNAKEFF | GINTSTS_NPTXFEMP | GINTSTS_RXFLVL | + GINTSTS_OTGINT | GINTSTS_CURMODE_HOST); + iflg = 1; + break; + case GINTMSK: + iflg = 1; + break; + default: + break; + } + + trace_usb_dwc2_glbreg_write(addr, glbregnm[index], orig, old, val); + *mmio = val; + + if (iflg) { + dwc2_update_irq(s); + } +} + +static uint64_t dwc2_fszreg_read(void *ptr, hwaddr addr, int index, + unsigned size) +{ + DWC2State *s = ptr; + uint32_t val; + + assert(addr == HPTXFSIZ); + val = s->fszreg[index]; + + trace_usb_dwc2_fszreg_read(addr, val); + return val; +} + +static void dwc2_fszreg_write(void *ptr, hwaddr addr, int index, uint64_t val, + unsigned size) +{ + DWC2State *s = ptr; + uint64_t orig = val; + uint32_t *mmio; + uint32_t old; + + assert(addr == HPTXFSIZ); + mmio = &s->fszreg[index]; + old = *mmio; + + trace_usb_dwc2_fszreg_write(addr, orig, old, val); + *mmio = val; +} + +static const char *hreg0nm[] = { + "HCFG ", "HFIR ", "HFNUM ", "<rsvd> ", "HPTXSTS ", + "HAINT ", "HAINTMSK ", "HFLBADDR ", "<rsvd> ", "<rsvd> ", + "<rsvd> ", "<rsvd> ", "<rsvd> ", "<rsvd> ", "<rsvd> ", + "<rsvd> ", "HPRT0 " +}; + +static uint64_t dwc2_hreg0_read(void *ptr, hwaddr addr, int index, + unsigned size) +{ + DWC2State *s = ptr; + uint32_t val; + + assert(addr >= HCFG && addr <= HPRT0); + val = s->hreg0[index]; + + switch (addr) { + case HFNUM: + val = (dwc2_get_frame_remaining(s) << HFNUM_FRREM_SHIFT) | + (s->hfnum << HFNUM_FRNUM_SHIFT); + break; + default: + break; + } + + trace_usb_dwc2_hreg0_read(addr, hreg0nm[index], val); + return val; +} + +static void dwc2_hreg0_write(void *ptr, hwaddr addr, int index, uint64_t val, + unsigned size) +{ + DWC2State *s = ptr; + USBDevice *dev = s->uport.dev; + uint64_t orig = val; + uint32_t *mmio; + uint32_t tval, told, old; + int prst = 0; + int iflg = 0; + + assert(addr >= HCFG && addr <= HPRT0); + mmio = &s->hreg0[index]; + old = *mmio; + + switch (addr) { + case HFIR: + break; + case HFNUM: + case HPTXSTS: + case HAINT: + qemu_log_mask(LOG_GUEST_ERROR, "%s: write to read-only register\n", + __func__); + return; + case HAINTMSK: + val &= 0xffff; + break; + case HPRT0: + /* don't allow clearing of read-only bits */ + val |= old & (HPRT0_SPD_MASK | HPRT0_LNSTS_MASK | HPRT0_OVRCURRACT | + HPRT0_CONNSTS); + /* don't allow clearing of self-clearing bits */ + val |= old & (HPRT0_SUSP | HPRT0_RES); + /* don't allow setting of self-setting bits */ + if (!(old & HPRT0_ENA) && (val & HPRT0_ENA)) { + val &= ~HPRT0_ENA; + } + /* clear the write-1-to-clear bits */ + tval = val & (HPRT0_OVRCURRCHG | HPRT0_ENACHG | HPRT0_ENA | + HPRT0_CONNDET); + told = old & (HPRT0_OVRCURRCHG | HPRT0_ENACHG | HPRT0_ENA | + HPRT0_CONNDET); + tval |= ~told; + tval = ~tval; + tval &= (HPRT0_OVRCURRCHG | HPRT0_ENACHG | HPRT0_ENA | + HPRT0_CONNDET); + val &= ~(HPRT0_OVRCURRCHG | HPRT0_ENACHG | HPRT0_ENA | + HPRT0_CONNDET); + val |= tval; + if (!(val & HPRT0_RST) && (old & HPRT0_RST)) { + if (dev && dev->attached) { + val |= HPRT0_ENA | HPRT0_ENACHG; + prst = 1; + } + } + if (val & (HPRT0_OVRCURRCHG | HPRT0_ENACHG | HPRT0_CONNDET)) { + iflg = 1; + } else { + iflg = -1; + } + break; + default: + break; + } + + if (prst) { + trace_usb_dwc2_hreg0_write(addr, hreg0nm[index], orig, old, + val & ~HPRT0_CONNDET); + trace_usb_dwc2_hreg0_action("call usb_port_reset"); + usb_port_reset(&s->uport); + val &= ~HPRT0_CONNDET; + } else { + trace_usb_dwc2_hreg0_write(addr, hreg0nm[index], orig, old, val); + } + + *mmio = val; + + if (iflg > 0) { + trace_usb_dwc2_hreg0_action("enable PRTINT"); + dwc2_raise_global_irq(s, GINTSTS_PRTINT); + } else if (iflg < 0) { + trace_usb_dwc2_hreg0_action("disable PRTINT"); + dwc2_lower_global_irq(s, GINTSTS_PRTINT); + } +} + +static const char *hreg1nm[] = { + "HCCHAR ", "HCSPLT ", "HCINT ", "HCINTMSK", "HCTSIZ ", "HCDMA ", + "<rsvd> ", "HCDMAB " +}; + +static uint64_t dwc2_hreg1_read(void *ptr, hwaddr addr, int index, + unsigned size) +{ + DWC2State *s = ptr; + uint32_t val; + + assert(addr >= HCCHAR(0) && addr <= HCDMAB(DWC2_NB_CHAN - 1)); + val = s->hreg1[index]; + + trace_usb_dwc2_hreg1_read(addr, hreg1nm[index & 7], addr >> 5, val); + return val; +} + +static void dwc2_hreg1_write(void *ptr, hwaddr addr, int index, uint64_t val, + unsigned size) +{ + DWC2State *s = ptr; + uint64_t orig = val; + uint32_t *mmio; + uint32_t old; + int iflg = 0; + int enflg = 0; + int disflg = 0; + + assert(addr >= HCCHAR(0) && addr <= HCDMAB(DWC2_NB_CHAN - 1)); + mmio = &s->hreg1[index]; + old = *mmio; + + switch (HSOTG_REG(0x500) + (addr & 0x1c)) { + case HCCHAR(0): + if ((val & HCCHAR_CHDIS) && !(old & HCCHAR_CHDIS)) { + val &= ~(HCCHAR_CHENA | HCCHAR_CHDIS); + disflg = 1; + } else { + val |= old & HCCHAR_CHDIS; + if ((val & HCCHAR_CHENA) && !(old & HCCHAR_CHENA)) { + val &= ~HCCHAR_CHDIS; + enflg = 1; + } else { + val |= old & HCCHAR_CHENA; + } + } + break; + case HCINT(0): + /* clear the write-1-to-clear bits */ + val |= ~old; + val = ~val; + val &= ~HCINTMSK_RESERVED14_31; + iflg = 1; + break; + case HCINTMSK(0): + val &= ~HCINTMSK_RESERVED14_31; + iflg = 1; + break; + case HCDMAB(0): + qemu_log_mask(LOG_GUEST_ERROR, "%s: write to read-only register\n", + __func__); + return; + default: + break; + } + + trace_usb_dwc2_hreg1_write(addr, hreg1nm[index & 7], index >> 3, orig, + old, val); + *mmio = val; + + if (disflg) { + /* set ChHltd in HCINT */ + s->hreg1[(index & ~7) + 2] |= HCINTMSK_CHHLTD; + iflg = 1; + } + + if (enflg) { + dwc2_enable_chan(s, index & ~7); + } + + if (iflg) { + dwc2_update_hc_irq(s, index & ~7); + } +} + +static const char *pcgregnm[] = { + "PCGCTL ", "PCGCCTL1 " +}; + +static uint64_t dwc2_pcgreg_read(void *ptr, hwaddr addr, int index, + unsigned size) +{ + DWC2State *s = ptr; + uint32_t val; + + assert(addr >= PCGCTL && addr <= PCGCCTL1); + val = s->pcgreg[index]; + + trace_usb_dwc2_pcgreg_read(addr, pcgregnm[index], val); + return val; +} + +static void dwc2_pcgreg_write(void *ptr, hwaddr addr, int index, + uint64_t val, unsigned size) +{ + DWC2State *s = ptr; + uint64_t orig = val; + uint32_t *mmio; + uint32_t old; + + assert(addr >= PCGCTL && addr <= PCGCCTL1); + mmio = &s->pcgreg[index]; + old = *mmio; + + trace_usb_dwc2_pcgreg_write(addr, pcgregnm[index], orig, old, val); + *mmio = val; +} + +static uint64_t dwc2_hsotg_read(void *ptr, hwaddr addr, unsigned size) +{ + uint64_t val; + + switch (addr) { + case HSOTG_REG(0x000) ... HSOTG_REG(0x0fc): + val = dwc2_glbreg_read(ptr, addr, (addr - HSOTG_REG(0x000)) >> 2, size); + break; + case HSOTG_REG(0x100): + val = dwc2_fszreg_read(ptr, addr, (addr - HSOTG_REG(0x100)) >> 2, size); + break; + case HSOTG_REG(0x104) ... HSOTG_REG(0x3fc): + /* Gadget-mode registers, just return 0 for now */ + val = 0; + break; + case HSOTG_REG(0x400) ... HSOTG_REG(0x4fc): + val = dwc2_hreg0_read(ptr, addr, (addr - HSOTG_REG(0x400)) >> 2, size); + break; + case HSOTG_REG(0x500) ... HSOTG_REG(0x7fc): + val = dwc2_hreg1_read(ptr, addr, (addr - HSOTG_REG(0x500)) >> 2, size); + break; + case HSOTG_REG(0x800) ... HSOTG_REG(0xdfc): + /* Gadget-mode registers, just return 0 for now */ + val = 0; + break; + case HSOTG_REG(0xe00) ... HSOTG_REG(0xffc): + val = dwc2_pcgreg_read(ptr, addr, (addr - HSOTG_REG(0xe00)) >> 2, size); + break; + default: + g_assert_not_reached(); + } + + return val; +} + +static void dwc2_hsotg_write(void *ptr, hwaddr addr, uint64_t val, + unsigned size) +{ + switch (addr) { + case HSOTG_REG(0x000) ... HSOTG_REG(0x0fc): + dwc2_glbreg_write(ptr, addr, (addr - HSOTG_REG(0x000)) >> 2, val, size); + break; + case HSOTG_REG(0x100): + dwc2_fszreg_write(ptr, addr, (addr - HSOTG_REG(0x100)) >> 2, val, size); + break; + case HSOTG_REG(0x104) ... HSOTG_REG(0x3fc): + /* Gadget-mode registers, do nothing for now */ + break; + case HSOTG_REG(0x400) ... HSOTG_REG(0x4fc): + dwc2_hreg0_write(ptr, addr, (addr - HSOTG_REG(0x400)) >> 2, val, size); + break; + case HSOTG_REG(0x500) ... HSOTG_REG(0x7fc): + dwc2_hreg1_write(ptr, addr, (addr - HSOTG_REG(0x500)) >> 2, val, size); + break; + case HSOTG_REG(0x800) ... HSOTG_REG(0xdfc): + /* Gadget-mode registers, do nothing for now */ + break; + case HSOTG_REG(0xe00) ... HSOTG_REG(0xffc): + dwc2_pcgreg_write(ptr, addr, (addr - HSOTG_REG(0xe00)) >> 2, val, size); + break; + default: + g_assert_not_reached(); + } +} + +static const MemoryRegionOps dwc2_mmio_hsotg_ops = { + .read = dwc2_hsotg_read, + .write = dwc2_hsotg_write, + .impl.min_access_size = 4, + .impl.max_access_size = 4, + .endianness = DEVICE_LITTLE_ENDIAN, +}; + +static uint64_t dwc2_hreg2_read(void *ptr, hwaddr addr, unsigned size) +{ + /* TODO - implement FIFOs to support slave mode */ + trace_usb_dwc2_hreg2_read(addr, addr >> 12, 0); + qemu_log_mask(LOG_UNIMP, "FIFO read not implemented\n"); + return 0; +} + +static void dwc2_hreg2_write(void *ptr, hwaddr addr, uint64_t val, + unsigned size) +{ + uint64_t orig = val; + + /* TODO - implement FIFOs to support slave mode */ + trace_usb_dwc2_hreg2_write(addr, addr >> 12, orig, 0, val); + qemu_log_mask(LOG_UNIMP, "FIFO write not implemented\n"); +} + +static const MemoryRegionOps dwc2_mmio_hreg2_ops = { + .read = dwc2_hreg2_read, + .write = dwc2_hreg2_write, + .impl.min_access_size = 4, + .impl.max_access_size = 4, + .endianness = DEVICE_LITTLE_ENDIAN, +}; + +static void dwc2_wakeup_endpoint(USBBus *bus, USBEndpoint *ep, + unsigned int stream) +{ + DWC2State *s = container_of(bus, DWC2State, bus); + + trace_usb_dwc2_wakeup_endpoint(ep, stream); + + /* TODO - do something here? */ + qemu_bh_schedule(s->async_bh); +} + +static USBBusOps dwc2_bus_ops = { + .wakeup_endpoint = dwc2_wakeup_endpoint, +}; + +static void dwc2_work_timer(void *opaque) +{ + DWC2State *s = opaque; + + trace_usb_dwc2_work_timer(); + qemu_bh_schedule(s->async_bh); +} + +static void dwc2_reset_enter(Object *obj, ResetType type) +{ + DWC2Class *c = DWC2_GET_CLASS(obj); + DWC2State *s = DWC2_USB(obj); + int i; + + trace_usb_dwc2_reset_enter(); + + if (c->parent_phases.enter) { + c->parent_phases.enter(obj, type); + } + + timer_del(s->frame_timer); + qemu_bh_cancel(s->async_bh); + + if (s->uport.dev && s->uport.dev->attached) { + usb_detach(&s->uport); + } + + dwc2_bus_stop(s); + + s->gotgctl = GOTGCTL_BSESVLD | GOTGCTL_ASESVLD | GOTGCTL_CONID_B; + s->gotgint = 0; + s->gahbcfg = 0; + s->gusbcfg = 5 << GUSBCFG_USBTRDTIM_SHIFT; + s->grstctl = GRSTCTL_AHBIDLE; + s->gintsts = GINTSTS_CONIDSTSCHNG | GINTSTS_PTXFEMP | GINTSTS_NPTXFEMP | + GINTSTS_CURMODE_HOST; + s->gintmsk = 0; + s->grxstsr = 0; + s->grxstsp = 0; + s->grxfsiz = 1024; + s->gnptxfsiz = 1024 << FIFOSIZE_DEPTH_SHIFT; + s->gnptxsts = (4 << FIFOSIZE_DEPTH_SHIFT) | 1024; + s->gi2cctl = GI2CCTL_I2CDATSE0 | GI2CCTL_ACK; + s->gpvndctl = 0; + s->ggpio = 0; + s->guid = 0; + s->gsnpsid = 0x4f54294a; + s->ghwcfg1 = 0; + s->ghwcfg2 = (8 << GHWCFG2_DEV_TOKEN_Q_DEPTH_SHIFT) | + (4 << GHWCFG2_HOST_PERIO_TX_Q_DEPTH_SHIFT) | + (4 << GHWCFG2_NONPERIO_TX_Q_DEPTH_SHIFT) | + GHWCFG2_DYNAMIC_FIFO | + GHWCFG2_PERIO_EP_SUPPORTED | + ((DWC2_NB_CHAN - 1) << GHWCFG2_NUM_HOST_CHAN_SHIFT) | + (GHWCFG2_INT_DMA_ARCH << GHWCFG2_ARCHITECTURE_SHIFT) | + (GHWCFG2_OP_MODE_NO_SRP_CAPABLE_HOST << GHWCFG2_OP_MODE_SHIFT); + s->ghwcfg3 = (4096 << GHWCFG3_DFIFO_DEPTH_SHIFT) | + (4 << GHWCFG3_PACKET_SIZE_CNTR_WIDTH_SHIFT) | + (4 << GHWCFG3_XFER_SIZE_CNTR_WIDTH_SHIFT); + s->ghwcfg4 = 0; + s->glpmcfg = 0; + s->gpwrdn = GPWRDN_PWRDNRSTN; + s->gdfifocfg = 0; + s->gadpctl = 0; + s->grefclk = 0; + s->gintmsk2 = 0; + s->gintsts2 = 0; + + s->hptxfsiz = 500 << FIFOSIZE_DEPTH_SHIFT; + + s->hcfg = 2 << HCFG_RESVALID_SHIFT; + s->hfir = 60000; + s->hfnum = 0x3fff; + s->hptxsts = (16 << TXSTS_QSPCAVAIL_SHIFT) | 32768; + s->haint = 0; + s->haintmsk = 0; + s->hprt0 = 0; + + memset(s->hreg1, 0, sizeof(s->hreg1)); + memset(s->pcgreg, 0, sizeof(s->pcgreg)); + + s->sof_time = 0; + s->frame_number = 0; + s->fi = USB_FRMINTVL - 1; + s->next_chan = 0; + s->working = false; + + for (i = 0; i < DWC2_NB_CHAN; i++) { + s->packet[i].needs_service = false; + } +} + +static void dwc2_reset_hold(Object *obj) +{ + DWC2Class *c = DWC2_GET_CLASS(obj); + DWC2State *s = DWC2_USB(obj); + + trace_usb_dwc2_reset_hold(); + + if (c->parent_phases.hold) { + c->parent_phases.hold(obj); + } + + dwc2_update_irq(s); +} + +static void dwc2_reset_exit(Object *obj) +{ + DWC2Class *c = DWC2_GET_CLASS(obj); + DWC2State *s = DWC2_USB(obj); + + trace_usb_dwc2_reset_exit(); + + if (c->parent_phases.exit) { + c->parent_phases.exit(obj); + } + + s->hprt0 = HPRT0_PWR; + if (s->uport.dev && s->uport.dev->attached) { + usb_attach(&s->uport); + usb_device_reset(s->uport.dev); + } +} + +static void dwc2_realize(DeviceState *dev, Error **errp) +{ + SysBusDevice *sbd = SYS_BUS_DEVICE(dev); + DWC2State *s = DWC2_USB(dev); + Object *obj; + Error *err = NULL; + + obj = object_property_get_link(OBJECT(dev), "dma-mr", &err); + if (err) { + error_setg(errp, "dwc2: required dma-mr link not found: %s", + error_get_pretty(err)); + return; + } + assert(obj != NULL); + + s->dma_mr = MEMORY_REGION(obj); + address_space_init(&s->dma_as, s->dma_mr, "dwc2"); + + usb_bus_new(&s->bus, sizeof(s->bus), &dwc2_bus_ops, dev); + usb_register_port(&s->bus, &s->uport, s, 0, &dwc2_port_ops, + USB_SPEED_MASK_LOW | USB_SPEED_MASK_FULL | + (s->usb_version == 2 ? USB_SPEED_MASK_HIGH : 0)); + s->uport.dev = 0; + + s->usb_frame_time = NANOSECONDS_PER_SECOND / 1000; /* 1000000 */ + if (NANOSECONDS_PER_SECOND >= USB_HZ_FS) { + s->usb_bit_time = NANOSECONDS_PER_SECOND / USB_HZ_FS; /* 83.3 */ + } else { + s->usb_bit_time = 1; + } + + s->fi = USB_FRMINTVL - 1; + s->eof_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, dwc2_frame_boundary, s); + s->frame_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, dwc2_work_timer, s); + s->async_bh = qemu_bh_new(dwc2_work_bh, s); + + sysbus_init_irq(sbd, &s->irq); +} + +static void dwc2_init(Object *obj) +{ + SysBusDevice *sbd = SYS_BUS_DEVICE(obj); + DWC2State *s = DWC2_USB(obj); + + memory_region_init(&s->container, obj, "dwc2", DWC2_MMIO_SIZE); + sysbus_init_mmio(sbd, &s->container); + + memory_region_init_io(&s->hsotg, obj, &dwc2_mmio_hsotg_ops, s, + "dwc2-io", 4 * KiB); + memory_region_add_subregion(&s->container, 0x0000, &s->hsotg); + + memory_region_init_io(&s->fifos, obj, &dwc2_mmio_hreg2_ops, s, + "dwc2-fifo", 64 * KiB); + memory_region_add_subregion(&s->container, 0x1000, &s->fifos); +} + +static const VMStateDescription vmstate_dwc2_state_packet = { + .name = "dwc2/packet", + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32(devadr, DWC2Packet), + VMSTATE_UINT32(epnum, DWC2Packet), + VMSTATE_UINT32(epdir, DWC2Packet), + VMSTATE_UINT32(mps, DWC2Packet), + VMSTATE_UINT32(pid, DWC2Packet), + VMSTATE_UINT32(index, DWC2Packet), + VMSTATE_UINT32(pcnt, DWC2Packet), + VMSTATE_UINT32(len, DWC2Packet), + VMSTATE_INT32(async, DWC2Packet), + VMSTATE_BOOL(small, DWC2Packet), + VMSTATE_BOOL(needs_service, DWC2Packet), + VMSTATE_END_OF_LIST() + }, +}; + +const VMStateDescription vmstate_dwc2_state = { + .name = "dwc2", + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32_ARRAY(glbreg, DWC2State, + DWC2_GLBREG_SIZE / sizeof(uint32_t)), + VMSTATE_UINT32_ARRAY(fszreg, DWC2State, + DWC2_FSZREG_SIZE / sizeof(uint32_t)), + VMSTATE_UINT32_ARRAY(hreg0, DWC2State, + DWC2_HREG0_SIZE / sizeof(uint32_t)), + VMSTATE_UINT32_ARRAY(hreg1, DWC2State, + DWC2_HREG1_SIZE / sizeof(uint32_t)), + VMSTATE_UINT32_ARRAY(pcgreg, DWC2State, + DWC2_PCGREG_SIZE / sizeof(uint32_t)), + + VMSTATE_TIMER_PTR(eof_timer, DWC2State), + VMSTATE_TIMER_PTR(frame_timer, DWC2State), + VMSTATE_INT64(sof_time, DWC2State), + VMSTATE_INT64(usb_frame_time, DWC2State), + VMSTATE_INT64(usb_bit_time, DWC2State), + VMSTATE_UINT32(usb_version, DWC2State), + VMSTATE_UINT16(frame_number, DWC2State), + VMSTATE_UINT16(fi, DWC2State), + VMSTATE_UINT16(next_chan, DWC2State), + VMSTATE_BOOL(working, DWC2State), + + VMSTATE_STRUCT_ARRAY(packet, DWC2State, DWC2_NB_CHAN, 1, + vmstate_dwc2_state_packet, DWC2Packet), + VMSTATE_UINT8_2DARRAY(usb_buf, DWC2State, DWC2_NB_CHAN, + DWC2_MAX_XFER_SIZE), + + VMSTATE_END_OF_LIST() + } +}; + +static Property dwc2_usb_properties[] = { + DEFINE_PROP_UINT32("usb_version", DWC2State, usb_version, 2), + DEFINE_PROP_END_OF_LIST(), +}; + +static void dwc2_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + DWC2Class *c = DWC2_CLASS(klass); + ResettableClass *rc = RESETTABLE_CLASS(klass); + + dc->realize = dwc2_realize; + dc->vmsd = &vmstate_dwc2_state; + set_bit(DEVICE_CATEGORY_USB, dc->categories); + device_class_set_props(dc, dwc2_usb_properties); + resettable_class_set_parent_phases(rc, dwc2_reset_enter, dwc2_reset_hold, + dwc2_reset_exit, &c->parent_phases); +} + +static const TypeInfo dwc2_usb_type_info = { + .name = TYPE_DWC2_USB, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_size = sizeof(DWC2State), + .instance_init = dwc2_init, + .class_size = sizeof(DWC2Class), + .class_init = dwc2_class_init, +}; + +static void dwc2_usb_register_types(void) +{ + type_register_static(&dwc2_usb_type_info); +} + +type_init(dwc2_usb_register_types) diff --git a/hw/usb/Kconfig b/hw/usb/Kconfig index XXXXXXX..XXXXXXX 100644 --- a/hw/usb/Kconfig +++ b/hw/usb/Kconfig @@ -XXX,XX +XXX,XX @@ config USB_MUSB bool select USB +config USB_DWC2 + bool + default y + select USB + config TUSB6010 bool select USB_MUSB diff --git a/hw/usb/Makefile.objs b/hw/usb/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/usb/Makefile.objs +++ b/hw/usb/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-$(CONFIG_USB_EHCI_SYSBUS) += hcd-ehci-sysbus.o common-obj-$(CONFIG_USB_XHCI) += hcd-xhci.o common-obj-$(CONFIG_USB_XHCI_NEC) += hcd-xhci-nec.o common-obj-$(CONFIG_USB_MUSB) += hcd-musb.o +common-obj-$(CONFIG_USB_DWC2) += hcd-dwc2.o common-obj-$(CONFIG_TUSB6010) += tusb6010.o common-obj-$(CONFIG_IMX) += chipidea.o diff --git a/hw/usb/trace-events b/hw/usb/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/usb/trace-events +++ b/hw/usb/trace-events @@ -XXX,XX +XXX,XX @@ usb_xhci_xfer_error(void *xfer, uint32_t ret) "%p: ret %d" usb_xhci_unimplemented(const char *item, int nr) "%s (0x%x)" usb_xhci_enforced_limit(const char *item) "%s" +# hcd-dwc2.c +usb_dwc2_update_irq(uint32_t level) "level=%d" +usb_dwc2_raise_global_irq(uint32_t intr) "0x%08x" +usb_dwc2_lower_global_irq(uint32_t intr) "0x%08x" +usb_dwc2_raise_host_irq(uint32_t intr) "0x%04x" +usb_dwc2_lower_host_irq(uint32_t intr) "0x%04x" +usb_dwc2_sof(int64_t next) "next SOF %" PRId64 +usb_dwc2_bus_start(void) "start SOFs" +usb_dwc2_bus_stop(void) "stop SOFs" +usb_dwc2_find_device(uint8_t addr) "%d" +usb_dwc2_port_disabled(uint32_t pnum) "port %d disabled" +usb_dwc2_device_found(uint32_t pnum) "device found on port %d" +usb_dwc2_device_not_found(void) "device not found" +usb_dwc2_handle_packet(uint32_t chan, void *dev, void *pkt, uint32_t ep, const char *type, const char *dir, uint32_t mps, uint32_t len, uint32_t pcnt) "ch %d dev %p pkt %p ep %d type %s dir %s mps %d len %d pcnt %d" +usb_dwc2_memory_read(uint32_t addr, uint32_t len) "addr %d len %d" +usb_dwc2_packet_status(const char *status, uint32_t len) "status %s len %d" +usb_dwc2_packet_error(const char *status) "ERROR %s" +usb_dwc2_async_packet(void *pkt, uint32_t chan, void *dev, uint32_t ep, const char *dir, uint32_t len) "pkt %p ch %d dev %p ep %d %s len %d" +usb_dwc2_memory_write(uint32_t addr, uint32_t len) "addr %d len %d" +usb_dwc2_packet_done(const char *status, uint32_t actual, uint32_t len, uint32_t pcnt) "status %s actual %d len %d pcnt %d" +usb_dwc2_packet_next(const char *status, uint32_t len, uint32_t pcnt) "status %s len %d pcnt %d" +usb_dwc2_attach(void *port) "port %p" +usb_dwc2_attach_speed(const char *speed) "%s-speed device attached" +usb_dwc2_detach(void *port) "port %p" +usb_dwc2_child_detach(void *port, void *child) "port %p child %p" +usb_dwc2_wakeup(void *port) "port %p" +usb_dwc2_async_packet_complete(void *port, void *pkt, uint32_t chan, void *dev, uint32_t ep, const char *dir, uint32_t len) "port %p packet %p ch %d dev %p ep %d %s len %d" +usb_dwc2_work_bh(void) "" +usb_dwc2_work_bh_service(uint32_t first, uint32_t current, void *dev, uint32_t ep) "first %d servicing %d dev %p ep %d" +usb_dwc2_work_bh_next(uint32_t chan) "next %d" +usb_dwc2_enable_chan(uint32_t chan, void *dev, void *pkt, uint32_t ep) "ch %d dev %p pkt %p ep %d" +usb_dwc2_glbreg_read(uint64_t addr, const char *reg, uint32_t val) " 0x%04" PRIx64 " %s val 0x%08x" +usb_dwc2_glbreg_write(uint64_t addr, const char *reg, uint64_t val, uint32_t old, uint64_t result) "0x%04" PRIx64 " %s val 0x%08" PRIx64 " old 0x%08x result 0x%08" PRIx64 +usb_dwc2_fszreg_read(uint64_t addr, uint32_t val) " 0x%04" PRIx64 " HPTXFSIZ val 0x%08x" +usb_dwc2_fszreg_write(uint64_t addr, uint64_t val, uint32_t old, uint64_t result) "0x%04" PRIx64 " HPTXFSIZ val 0x%08" PRIx64 " old 0x%08x result 0x%08" PRIx64 +usb_dwc2_hreg0_read(uint64_t addr, const char *reg, uint32_t val) " 0x%04" PRIx64 " %s val 0x%08x" +usb_dwc2_hreg0_write(uint64_t addr, const char *reg, uint64_t val, uint32_t old, uint64_t result) " 0x%04" PRIx64 " %s val 0x%08" PRIx64 " old 0x%08x result 0x%08" PRIx64 +usb_dwc2_hreg1_read(uint64_t addr, const char *reg, uint64_t chan, uint32_t val) " 0x%04" PRIx64 " %s%" PRId64 " val 0x%08x" +usb_dwc2_hreg1_write(uint64_t addr, const char *reg, uint64_t chan, uint64_t val, uint32_t old, uint64_t result) " 0x%04" PRIx64 " %s%" PRId64 " val 0x%08" PRIx64 " old 0x%08x result 0x%08" PRIx64 +usb_dwc2_pcgreg_read(uint64_t addr, const char *reg, uint32_t val) " 0x%04" PRIx64 " %s val 0x%08x" +usb_dwc2_pcgreg_write(uint64_t addr, const char *reg, uint64_t val, uint32_t old, uint64_t result) "0x%04" PRIx64 " %s val 0x%08" PRIx64 " old 0x%08x result 0x%08" PRIx64 +usb_dwc2_hreg2_read(uint64_t addr, uint64_t fifo, uint32_t val) " 0x%04" PRIx64 " FIFO%" PRId64 " val 0x%08x" +usb_dwc2_hreg2_write(uint64_t addr, uint64_t fifo, uint64_t val, uint32_t old, uint64_t result) " 0x%04" PRIx64 " FIFO%" PRId64 " val 0x%08" PRIx64 " old 0x%08x result 0x%08" PRIx64 +usb_dwc2_hreg0_action(const char *s) "%s" +usb_dwc2_wakeup_endpoint(void *ep, uint32_t stream) "endp %p stream %d" +usb_dwc2_work_timer(void) "" +usb_dwc2_reset_enter(void) "=== RESET enter ===" +usb_dwc2_reset_hold(void) "=== RESET hold ===" +usb_dwc2_reset_exit(void) "=== RESET exit ===" + # desc.c usb_desc_device(int addr, int len, int ret) "dev %d query device, len %d, ret %d" usb_desc_device_qualifier(int addr, int len, int ret) "dev %d query device qualifier, len %d, ret %d" -- 2.20.1 From: Paul Zimmerman <pauldzim@gmail.com> The dwc-hsotg (dwc2) USB host depends on a short packet to indicate the end of an IN transfer. The usb-storage driver currently doesn't provide this, so fix it. I have tested this change rather extensively using a PC emulation with xhci, ehci, and uhci controllers, and have not observed any regressions. Signed-off-by: Paul Zimmerman <pauldzim@gmail.com> Message-id: 20200520235349.21215-6-pauldzim@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/usb/dev-storage.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/hw/usb/dev-storage.c b/hw/usb/dev-storage.c index XXXXXXX..XXXXXXX 100644 --- a/hw/usb/dev-storage.c +++ b/hw/usb/dev-storage.c @@ -XXX,XX +XXX,XX @@ static void usb_msd_copy_data(MSDState *s, USBPacket *p) usb_packet_copy(p, scsi_req_get_buf(s->req) + s->scsi_off, len); s->scsi_len -= len; s->scsi_off += len; + if (len > s->data_len) { + len = s->data_len; + } s->data_len -= len; if (s->scsi_len == 0 || s->data_len == 0) { scsi_req_continue(s->req); @@ -XXX,XX +XXX,XX @@ static void usb_msd_command_complete(SCSIRequest *req, uint32_t status, size_t r if (s->data_len) { int len = (p->iov.size - p->actual_length); usb_packet_skip(p, len); + if (len > s->data_len) { + len = s->data_len; + } s->data_len -= len; } if (s->data_len == 0) { @@ -XXX,XX +XXX,XX @@ static void usb_msd_handle_data(USBDevice *dev, USBPacket *p) int len = p->iov.size - p->actual_length; if (len) { usb_packet_skip(p, len); + if (len > s->data_len) { + len = s->data_len; + } s->data_len -= len; if (s->data_len == 0) { s->mode = USB_MSDM_CSW; @@ -XXX,XX +XXX,XX @@ static void usb_msd_handle_data(USBDevice *dev, USBPacket *p) int len = p->iov.size - p->actual_length; if (len) { usb_packet_skip(p, len); + if (len > s->data_len) { + len = s->data_len; + } s->data_len -= len; if (s->data_len == 0) { s->mode = USB_MSDM_CSW; } } } - if (p->actual_length < p->iov.size) { + if (p->actual_length < p->iov.size && (p->short_not_ok || + s->scsi_len >= p->ep->max_packet_size)) { DPRINTF("Deferring packet %p [wait data-in]\n", p); s->packet = p; p->status = USB_RET_ASYNC; -- 2.20.1 From: Paul Zimmerman <pauldzim@gmail.com> Wire the dwc-hsotg (dwc2) emulation into Qemu Signed-off-by: Paul Zimmerman <pauldzim@gmail.com> Reviewed-by: Philippe Mathieu-Daude <f4bug@amsat.org> Message-id: 20200520235349.21215-7-pauldzim@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/bcm2835_peripherals.h | 3 ++- hw/arm/bcm2835_peripherals.c | 21 ++++++++++++++++++++- 2 files changed, 22 insertions(+), 2 deletions(-) diff --git a/include/hw/arm/bcm2835_peripherals.h b/include/hw/arm/bcm2835_peripherals.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/bcm2835_peripherals.h +++ b/include/hw/arm/bcm2835_peripherals.h @@ -XXX,XX +XXX,XX @@ #include "hw/sd/bcm2835_sdhost.h" #include "hw/gpio/bcm2835_gpio.h" #include "hw/timer/bcm2835_systmr.h" +#include "hw/usb/hcd-dwc2.h" #include "hw/misc/unimp.h" #define TYPE_BCM2835_PERIPHERALS "bcm2835-peripherals" @@ -XXX,XX +XXX,XX @@ typedef struct BCM2835PeripheralState { UnimplementedDeviceState ave0; UnimplementedDeviceState bscsl; UnimplementedDeviceState smi; - UnimplementedDeviceState dwc2; + DWC2State dwc2; UnimplementedDeviceState sdramc; } BCM2835PeripheralState; diff --git a/hw/arm/bcm2835_peripherals.c b/hw/arm/bcm2835_peripherals.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/bcm2835_peripherals.c +++ b/hw/arm/bcm2835_peripherals.c @@ -XXX,XX +XXX,XX @@ static void bcm2835_peripherals_init(Object *obj) /* Mphi */ sysbus_init_child_obj(obj, "mphi", &s->mphi, sizeof(s->mphi), TYPE_BCM2835_MPHI); + + /* DWC2 */ + sysbus_init_child_obj(obj, "dwc2", &s->dwc2, sizeof(s->dwc2), + TYPE_DWC2_USB); + + object_property_add_const_link(OBJECT(&s->dwc2), "dma-mr", + OBJECT(&s->gpu_bus_mr)); } static void bcm2835_peripherals_realize(DeviceState *dev, Error **errp) @@ -XXX,XX +XXX,XX @@ static void bcm2835_peripherals_realize(DeviceState *dev, Error **errp) qdev_get_gpio_in_named(DEVICE(&s->ic), BCM2835_IC_GPU_IRQ, INTERRUPT_HOSTPORT)); + /* DWC2 */ + object_property_set_bool(OBJECT(&s->dwc2), true, "realized", &err); + if (err) { + error_propagate(errp, err); + return; + } + + memory_region_add_subregion(&s->peri_mr, USB_OTG_OFFSET, + sysbus_mmio_get_region(SYS_BUS_DEVICE(&s->dwc2), 0)); + sysbus_connect_irq(SYS_BUS_DEVICE(&s->dwc2), 0, + qdev_get_gpio_in_named(DEVICE(&s->ic), BCM2835_IC_GPU_IRQ, + INTERRUPT_USB)); + create_unimp(s, &s->armtmr, "bcm2835-sp804", ARMCTRL_TIMER0_1_OFFSET, 0x40); create_unimp(s, &s->cprman, "bcm2835-cprman", CPRMAN_OFFSET, 0x1000); create_unimp(s, &s->a2w, "bcm2835-a2w", A2W_OFFSET, 0x1000); @@ -XXX,XX +XXX,XX @@ static void bcm2835_peripherals_realize(DeviceState *dev, Error **errp) create_unimp(s, &s->otp, "bcm2835-otp", OTP_OFFSET, 0x80); create_unimp(s, &s->dbus, "bcm2835-dbus", DBUS_OFFSET, 0x8000); create_unimp(s, &s->ave0, "bcm2835-ave0", AVE0_OFFSET, 0x8000); - create_unimp(s, &s->dwc2, "dwc-usb2", USB_OTG_OFFSET, 0x1000); create_unimp(s, &s->sdramc, "bcm2835-sdramc", SDRAMC_OFFSET, 0x100); } -- 2.20.1 From: Paul Zimmerman <pauldzim@gmail.com> Add a check for functional dwc-hsotg (dwc2) USB host emulation to the Raspi 2 acceptance test Signed-off-by: Paul Zimmerman <pauldzim@gmail.com> Reviewed-by: Philippe Mathieu-Daude <f4bug@amsat.org> Message-id: 20200520235349.21215-8-pauldzim@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/acceptance/boot_linux_console.py | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/tests/acceptance/boot_linux_console.py b/tests/acceptance/boot_linux_console.py index XXXXXXX..XXXXXXX 100644 --- a/tests/acceptance/boot_linux_console.py +++ b/tests/acceptance/boot_linux_console.py @@ -XXX,XX +XXX,XX @@ class BootLinuxConsole(LinuxKernelTest): self.vm.set_console() kernel_command_line = (self.KERNEL_COMMON_COMMAND_LINE + - serial_kernel_cmdline[uart_id]) + serial_kernel_cmdline[uart_id] + + ' root=/dev/mmcblk0p2 rootwait ' + + 'dwc_otg.fiq_fsm_enable=0') self.vm.add_args('-kernel', kernel_path, '-dtb', dtb_path, - '-append', kernel_command_line) + '-append', kernel_command_line, + '-device', 'usb-kbd') self.vm.launch() console_pattern = 'Kernel command line: %s' % kernel_command_line self.wait_for_console_pattern(console_pattern) + console_pattern = 'Product: QEMU USB Keyboard' + self.wait_for_console_pattern(console_pattern) def test_arm_raspi2_uart0(self): """ -- 2.20.1 Convert the VSHL and VSLI insns from the Neon 2-registers-and-a-shift group to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200522145520.6778-2-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 25 ++++++++++++++++++++++ target/arm/translate-neon.inc.c | 38 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 18 +++++++--------- 3 files changed, 71 insertions(+), 10 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VRECPS_fp_3s 1111 001 0 0 . 0 . .... .... 1111 ... 1 .... @3same_fp VRSQRTS_fp_3s 1111 001 0 0 . 1 . .... .... 1111 ... 1 .... @3same_fp VMAXNM_fp_3s 1111 001 1 0 . 0 . .... .... 1111 ... 1 .... @3same_fp VMINNM_fp_3s 1111 001 1 0 . 1 . .... .... 1111 ... 1 .... @3same_fp + +###################################################################### +# 2-reg-and-shift grouping: +# 1111 001 U 1 D immH:3 immL:3 Vd:4 opc:4 L Q M 1 Vm:4 +###################################################################### +&2reg_shift vm vd q shift size + +@2reg_shl_d .... ... . . . shift:6 .... .... 1 q:1 . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=3 +@2reg_shl_s .... ... . . . 1 shift:5 .... .... 0 q:1 . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=2 +@2reg_shl_h .... ... . . . 01 shift:4 .... .... 0 q:1 . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=1 +@2reg_shl_b .... ... . . . 001 shift:3 .... .... 0 q:1 . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=0 + +VSHL_2sh 1111 001 0 1 . ...... .... 0101 . . . 1 .... @2reg_shl_d +VSHL_2sh 1111 001 0 1 . ...... .... 0101 . . . 1 .... @2reg_shl_s +VSHL_2sh 1111 001 0 1 . ...... .... 0101 . . . 1 .... @2reg_shl_h +VSHL_2sh 1111 001 0 1 . ...... .... 0101 . . . 1 .... @2reg_shl_b + +VSLI_2sh 1111 001 1 1 . ...... .... 0101 . . . 1 .... @2reg_shl_d +VSLI_2sh 1111 001 1 1 . ...... .... 0101 . . . 1 .... @2reg_shl_s +VSLI_2sh 1111 001 1 1 . ...... .... 0101 . . . 1 .... @2reg_shl_h +VSLI_2sh 1111 001 1 1 . ...... .... 0101 . . . 1 .... @2reg_shl_b diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ static bool do_3same_fp_pair(DisasContext *s, arg_3same *a, VFPGen3OpSPFn *fn) DO_3S_FP_PAIR(VPADD, gen_helper_vfp_adds) DO_3S_FP_PAIR(VPMAX, gen_helper_vfp_maxs) DO_3S_FP_PAIR(VPMIN, gen_helper_vfp_mins) + +static bool do_vector_2sh(DisasContext *s, arg_2reg_shift *a, GVecGen2iFn *fn) +{ + /* Handle a 2-reg-shift insn which can be vectorized. */ + int vec_size = a->q ? 16 : 8; + int rd_ofs = neon_reg_offset(a->vd, 0); + int rm_ofs = neon_reg_offset(a->vm, 0); + + if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vm) & 0x10)) { + return false; + } + + if ((a->vm | a->vd) & a->q) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + fn(a->size, rd_ofs, rm_ofs, a->shift, vec_size, vec_size); + return true; +} + +#define DO_2SH(INSN, FUNC) \ + static bool trans_##INSN##_2sh(DisasContext *s, arg_2reg_shift *a) \ + { \ + return do_vector_2sh(s, a, FUNC); \ + } \ + +DO_2SH(VSHL, tcg_gen_gvec_shli) +DO_2SH(VSLI, gen_gvec_sli) diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) if ((insn & 0x00380080) != 0) { /* Two registers and shift. */ op = (insn >> 8) & 0xf; + + switch (op) { + case 5: /* VSHL, VSLI */ + return 1; /* handled by decodetree */ + default: + break; + } + if (insn & (1 << 7)) { /* 64-bit shift. */ if (op > 7) { @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) gen_gvec_sri(size, rd_ofs, rm_ofs, shift, vec_size, vec_size); return 0; - - case 5: /* VSHL, VSLI */ - if (u) { /* VSLI */ - gen_gvec_sli(size, rd_ofs, rm_ofs, shift, - vec_size, vec_size); - } else { /* VSHL */ - tcg_gen_gvec_shli(size, rd_ofs, rm_ofs, shift, - vec_size, vec_size); - } - return 0; } if (size == 3) { -- 2.20.1 Convert the VSHR 2-reg-shift insns to decodetree. Note that unlike the legacy decoder, we present the right shift amount to the trans_ function as a positive integer. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200522145520.6778-3-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 25 ++++++++++++++++++++ target/arm/translate-neon.inc.c | 41 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 21 +---------------- 3 files changed, 67 insertions(+), 20 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VMINNM_fp_3s 1111 001 1 0 . 1 . .... .... 1111 ... 1 .... @3same_fp ###################################################################### &2reg_shift vm vd q shift size +# Right shifts are encoded as N - shift, where N is the element size in bits. +%neon_rshift_i6 16:6 !function=rsub_64 +%neon_rshift_i5 16:5 !function=rsub_32 +%neon_rshift_i4 16:4 !function=rsub_16 +%neon_rshift_i3 16:3 !function=rsub_8 + +@2reg_shr_d .... ... . . . ...... .... .... 1 q:1 . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=3 shift=%neon_rshift_i6 +@2reg_shr_s .... ... . . . 1 ..... .... .... 0 q:1 . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=2 shift=%neon_rshift_i5 +@2reg_shr_h .... ... . . . 01 .... .... .... 0 q:1 . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=1 shift=%neon_rshift_i4 +@2reg_shr_b .... ... . . . 001 ... .... .... 0 q:1 . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=0 shift=%neon_rshift_i3 + @2reg_shl_d .... ... . . . shift:6 .... .... 1 q:1 . . .... \ &2reg_shift vm=%vm_dp vd=%vd_dp size=3 @2reg_shl_s .... ... . . . 1 shift:5 .... .... 0 q:1 . . .... \ @@ -XXX,XX +XXX,XX @@ VMINNM_fp_3s 1111 001 1 0 . 1 . .... .... 1111 ... 1 .... @3same_fp @2reg_shl_b .... ... . . . 001 shift:3 .... .... 0 q:1 . . .... \ &2reg_shift vm=%vm_dp vd=%vd_dp size=0 +VSHR_S_2sh 1111 001 0 1 . ...... .... 0000 . . . 1 .... @2reg_shr_d +VSHR_S_2sh 1111 001 0 1 . ...... .... 0000 . . . 1 .... @2reg_shr_s +VSHR_S_2sh 1111 001 0 1 . ...... .... 0000 . . . 1 .... @2reg_shr_h +VSHR_S_2sh 1111 001 0 1 . ...... .... 0000 . . . 1 .... @2reg_shr_b + +VSHR_U_2sh 1111 001 1 1 . ...... .... 0000 . . . 1 .... @2reg_shr_d +VSHR_U_2sh 1111 001 1 1 . ...... .... 0000 . . . 1 .... @2reg_shr_s +VSHR_U_2sh 1111 001 1 1 . ...... .... 0000 . . . 1 .... @2reg_shr_h +VSHR_U_2sh 1111 001 1 1 . ...... .... 0000 . . . 1 .... @2reg_shr_b + VSHL_2sh 1111 001 0 1 . ...... .... 0101 . . . 1 .... @2reg_shl_d VSHL_2sh 1111 001 0 1 . ...... .... 0101 . . . 1 .... @2reg_shl_s VSHL_2sh 1111 001 0 1 . ...... .... 0101 . . . 1 .... @2reg_shl_h diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ static inline int plus1(DisasContext *s, int x) return x + 1; } +static inline int rsub_64(DisasContext *s, int x) +{ + return 64 - x; +} + +static inline int rsub_32(DisasContext *s, int x) +{ + return 32 - x; +} +static inline int rsub_16(DisasContext *s, int x) +{ + return 16 - x; +} +static inline int rsub_8(DisasContext *s, int x) +{ + return 8 - x; +} + /* Include the generated Neon decoder */ #include "decode-neon-dp.inc.c" #include "decode-neon-ls.inc.c" @@ -XXX,XX +XXX,XX @@ static bool do_vector_2sh(DisasContext *s, arg_2reg_shift *a, GVecGen2iFn *fn) DO_2SH(VSHL, tcg_gen_gvec_shli) DO_2SH(VSLI, gen_gvec_sli) + +static bool trans_VSHR_S_2sh(DisasContext *s, arg_2reg_shift *a) +{ + /* Signed shift out of range results in all-sign-bits */ + a->shift = MIN(a->shift, (8 << a->size) - 1); + return do_vector_2sh(s, a, tcg_gen_gvec_sari); +} + +static void gen_zero_rd_2sh(unsigned vece, uint32_t rd_ofs, uint32_t rm_ofs, + int64_t shift, uint32_t oprsz, uint32_t maxsz) +{ + tcg_gen_gvec_dup_imm(vece, rd_ofs, oprsz, maxsz, 0); +} + +static bool trans_VSHR_U_2sh(DisasContext *s, arg_2reg_shift *a) +{ + /* Shift out of range is architecturally valid and results in zero. */ + if (a->shift >= (8 << a->size)) { + return do_vector_2sh(s, a, gen_zero_rd_2sh); + } else { + return do_vector_2sh(s, a, tcg_gen_gvec_shri); + } +} diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) op = (insn >> 8) & 0xf; switch (op) { + case 0: /* VSHR */ case 5: /* VSHL, VSLI */ return 1; /* handled by decodetree */ default: @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) } switch (op) { - case 0: /* VSHR */ - /* Right shift comes here negative. */ - shift = -shift; - /* Shifts larger than the element size are architecturally - * valid. Unsigned results in all zeros; signed results - * in all sign bits. - */ - if (!u) { - tcg_gen_gvec_sari(size, rd_ofs, rm_ofs, - MIN(shift, (8 << size) - 1), - vec_size, vec_size); - } else if (shift >= 8 << size) { - tcg_gen_gvec_dup_imm(MO_8, rd_ofs, vec_size, - vec_size, 0); - } else { - tcg_gen_gvec_shri(size, rd_ofs, rm_ofs, shift, - vec_size, vec_size); - } - return 0; - case 1: /* VSRA */ /* Right shift comes here negative. */ shift = -shift; -- 2.20.1 Convert the VSRA, VSRI, VRSHR, VRSRA 2-reg-shift insns to decodetree. (These are the last instructions in the group that are vectorized; the rest all require looping over each element.) Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200522145520.6778-4-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 35 ++++++++++++++++++++++ target/arm/translate-neon.inc.c | 7 +++++ target/arm/translate.c | 52 +++------------------------------ 3 files changed, 46 insertions(+), 48 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VSHR_U_2sh 1111 001 1 1 . ...... .... 0000 . . . 1 .... @2reg_shr_s VSHR_U_2sh 1111 001 1 1 . ...... .... 0000 . . . 1 .... @2reg_shr_h VSHR_U_2sh 1111 001 1 1 . ...... .... 0000 . . . 1 .... @2reg_shr_b +VSRA_S_2sh 1111 001 0 1 . ...... .... 0001 . . . 1 .... @2reg_shr_d +VSRA_S_2sh 1111 001 0 1 . ...... .... 0001 . . . 1 .... @2reg_shr_s +VSRA_S_2sh 1111 001 0 1 . ...... .... 0001 . . . 1 .... @2reg_shr_h +VSRA_S_2sh 1111 001 0 1 . ...... .... 0001 . . . 1 .... @2reg_shr_b + +VSRA_U_2sh 1111 001 1 1 . ...... .... 0001 . . . 1 .... @2reg_shr_d +VSRA_U_2sh 1111 001 1 1 . ...... .... 0001 . . . 1 .... @2reg_shr_s +VSRA_U_2sh 1111 001 1 1 . ...... .... 0001 . . . 1 .... @2reg_shr_h +VSRA_U_2sh 1111 001 1 1 . ...... .... 0001 . . . 1 .... @2reg_shr_b + +VRSHR_S_2sh 1111 001 0 1 . ...... .... 0010 . . . 1 .... @2reg_shr_d +VRSHR_S_2sh 1111 001 0 1 . ...... .... 0010 . . . 1 .... @2reg_shr_s +VRSHR_S_2sh 1111 001 0 1 . ...... .... 0010 . . . 1 .... @2reg_shr_h +VRSHR_S_2sh 1111 001 0 1 . ...... .... 0010 . . . 1 .... @2reg_shr_b + +VRSHR_U_2sh 1111 001 1 1 . ...... .... 0010 . . . 1 .... @2reg_shr_d +VRSHR_U_2sh 1111 001 1 1 . ...... .... 0010 . . . 1 .... @2reg_shr_s +VRSHR_U_2sh 1111 001 1 1 . ...... .... 0010 . . . 1 .... @2reg_shr_h +VRSHR_U_2sh 1111 001 1 1 . ...... .... 0010 . . . 1 .... @2reg_shr_b + +VRSRA_S_2sh 1111 001 0 1 . ...... .... 0011 . . . 1 .... @2reg_shr_d +VRSRA_S_2sh 1111 001 0 1 . ...... .... 0011 . . . 1 .... @2reg_shr_s +VRSRA_S_2sh 1111 001 0 1 . ...... .... 0011 . . . 1 .... @2reg_shr_h +VRSRA_S_2sh 1111 001 0 1 . ...... .... 0011 . . . 1 .... @2reg_shr_b + +VRSRA_U_2sh 1111 001 1 1 . ...... .... 0011 . . . 1 .... @2reg_shr_d +VRSRA_U_2sh 1111 001 1 1 . ...... .... 0011 . . . 1 .... @2reg_shr_s +VRSRA_U_2sh 1111 001 1 1 . ...... .... 0011 . . . 1 .... @2reg_shr_h +VRSRA_U_2sh 1111 001 1 1 . ...... .... 0011 . . . 1 .... @2reg_shr_b + +VSRI_2sh 1111 001 1 1 . ...... .... 0100 . . . 1 .... @2reg_shr_d +VSRI_2sh 1111 001 1 1 . ...... .... 0100 . . . 1 .... @2reg_shr_s +VSRI_2sh 1111 001 1 1 . ...... .... 0100 . . . 1 .... @2reg_shr_h +VSRI_2sh 1111 001 1 1 . ...... .... 0100 . . . 1 .... @2reg_shr_b + VSHL_2sh 1111 001 0 1 . ...... .... 0101 . . . 1 .... @2reg_shl_d VSHL_2sh 1111 001 0 1 . ...... .... 0101 . . . 1 .... @2reg_shl_s VSHL_2sh 1111 001 0 1 . ...... .... 0101 . . . 1 .... @2reg_shl_h diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ static bool do_vector_2sh(DisasContext *s, arg_2reg_shift *a, GVecGen2iFn *fn) DO_2SH(VSHL, tcg_gen_gvec_shli) DO_2SH(VSLI, gen_gvec_sli) +DO_2SH(VSRI, gen_gvec_sri) +DO_2SH(VSRA_S, gen_gvec_ssra) +DO_2SH(VSRA_U, gen_gvec_usra) +DO_2SH(VRSHR_S, gen_gvec_srshr) +DO_2SH(VRSHR_U, gen_gvec_urshr) +DO_2SH(VRSRA_S, gen_gvec_srsra) +DO_2SH(VRSRA_U, gen_gvec_ursra) static bool trans_VSHR_S_2sh(DisasContext *s, arg_2reg_shift *a) { diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) switch (op) { case 0: /* VSHR */ + case 1: /* VSRA */ + case 2: /* VRSHR */ + case 3: /* VRSRA */ + case 4: /* VSRI */ case 5: /* VSHL, VSLI */ return 1; /* handled by decodetree */ default: @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) shift = shift - (1 << (size + 3)); } - switch (op) { - case 1: /* VSRA */ - /* Right shift comes here negative. */ - shift = -shift; - if (u) { - gen_gvec_usra(size, rd_ofs, rm_ofs, shift, - vec_size, vec_size); - } else { - gen_gvec_ssra(size, rd_ofs, rm_ofs, shift, - vec_size, vec_size); - } - return 0; - - case 2: /* VRSHR */ - /* Right shift comes here negative. */ - shift = -shift; - if (u) { - gen_gvec_urshr(size, rd_ofs, rm_ofs, shift, - vec_size, vec_size); - } else { - gen_gvec_srshr(size, rd_ofs, rm_ofs, shift, - vec_size, vec_size); - } - return 0; - - case 3: /* VRSRA */ - /* Right shift comes here negative. */ - shift = -shift; - if (u) { - gen_gvec_ursra(size, rd_ofs, rm_ofs, shift, - vec_size, vec_size); - } else { - gen_gvec_srsra(size, rd_ofs, rm_ofs, shift, - vec_size, vec_size); - } - return 0; - - case 4: /* VSRI */ - if (!u) { - return 1; - } - /* Right shift comes here negative. */ - shift = -shift; - gen_gvec_sri(size, rd_ofs, rm_ofs, shift, - vec_size, vec_size); - return 0; - } - if (size == 3) { count = q + 1; } else { -- 2.20.1 Convert the VQSHLU and QVSHL 2-reg-shift insns to decodetree. These are the last of the simple shift-by-immediate insns. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200522145520.6778-5-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 15 +++++ target/arm/translate-neon.inc.c | 108 +++++++++++++++++++++++++++++++ target/arm/translate.c | 110 +------------------------------- 3 files changed, 126 insertions(+), 107 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VSLI_2sh 1111 001 1 1 . ...... .... 0101 . . . 1 .... @2reg_shl_d VSLI_2sh 1111 001 1 1 . ...... .... 0101 . . . 1 .... @2reg_shl_s VSLI_2sh 1111 001 1 1 . ...... .... 0101 . . . 1 .... @2reg_shl_h VSLI_2sh 1111 001 1 1 . ...... .... 0101 . . . 1 .... @2reg_shl_b + +VQSHLU_64_2sh 1111 001 1 1 . ...... .... 0110 . . . 1 .... @2reg_shl_d +VQSHLU_2sh 1111 001 1 1 . ...... .... 0110 . . . 1 .... @2reg_shl_s +VQSHLU_2sh 1111 001 1 1 . ...... .... 0110 . . . 1 .... @2reg_shl_h +VQSHLU_2sh 1111 001 1 1 . ...... .... 0110 . . . 1 .... @2reg_shl_b + +VQSHL_S_64_2sh 1111 001 0 1 . ...... .... 0111 . . . 1 .... @2reg_shl_d +VQSHL_S_2sh 1111 001 0 1 . ...... .... 0111 . . . 1 .... @2reg_shl_s +VQSHL_S_2sh 1111 001 0 1 . ...... .... 0111 . . . 1 .... @2reg_shl_h +VQSHL_S_2sh 1111 001 0 1 . ...... .... 0111 . . . 1 .... @2reg_shl_b + +VQSHL_U_64_2sh 1111 001 1 1 . ...... .... 0111 . . . 1 .... @2reg_shl_d +VQSHL_U_2sh 1111 001 1 1 . ...... .... 0111 . . . 1 .... @2reg_shl_s +VQSHL_U_2sh 1111 001 1 1 . ...... .... 0111 . . . 1 .... @2reg_shl_h +VQSHL_U_2sh 1111 001 1 1 . ...... .... 0111 . . . 1 .... @2reg_shl_b diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ static bool trans_VSHR_U_2sh(DisasContext *s, arg_2reg_shift *a) return do_vector_2sh(s, a, tcg_gen_gvec_shri); } } + +static bool do_2shift_env_64(DisasContext *s, arg_2reg_shift *a, + NeonGenTwo64OpEnvFn *fn) +{ + /* + * 2-reg-and-shift operations, size == 3 case, where the + * function needs to be passed cpu_env. + */ + TCGv_i64 constimm; + int pass; + + if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vm) & 0x10)) { + return false; + } + + if ((a->vm | a->vd) & a->q) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + /* + * To avoid excessive duplication of ops we implement shift + * by immediate using the variable shift operations. + */ + constimm = tcg_const_i64(dup_const(a->size, a->shift)); + + for (pass = 0; pass < a->q + 1; pass++) { + TCGv_i64 tmp = tcg_temp_new_i64(); + + neon_load_reg64(tmp, a->vm + pass); + fn(tmp, cpu_env, tmp, constimm); + neon_store_reg64(tmp, a->vd + pass); + } + tcg_temp_free_i64(constimm); + return true; +} + +static bool do_2shift_env_32(DisasContext *s, arg_2reg_shift *a, + NeonGenTwoOpEnvFn *fn) +{ + /* + * 2-reg-and-shift operations, size < 3 case, where the + * helper needs to be passed cpu_env. + */ + TCGv_i32 constimm; + int pass; + + if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vm) & 0x10)) { + return false; + } + + if ((a->vm | a->vd) & a->q) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + /* + * To avoid excessive duplication of ops we implement shift + * by immediate using the variable shift operations. + */ + constimm = tcg_const_i32(dup_const(a->size, a->shift)); + + for (pass = 0; pass < (a->q ? 4 : 2); pass++) { + TCGv_i32 tmp = neon_load_reg(a->vm, pass); + fn(tmp, cpu_env, tmp, constimm); + neon_store_reg(a->vd, pass, tmp); + } + tcg_temp_free_i32(constimm); + return true; +} + +#define DO_2SHIFT_ENV(INSN, FUNC) \ + static bool trans_##INSN##_64_2sh(DisasContext *s, arg_2reg_shift *a) \ + { \ + return do_2shift_env_64(s, a, gen_helper_neon_##FUNC##64); \ + } \ + static bool trans_##INSN##_2sh(DisasContext *s, arg_2reg_shift *a) \ + { \ + static NeonGenTwoOpEnvFn * const fns[] = { \ + gen_helper_neon_##FUNC##8, \ + gen_helper_neon_##FUNC##16, \ + gen_helper_neon_##FUNC##32, \ + }; \ + assert(a->size < ARRAY_SIZE(fns)); \ + return do_2shift_env_32(s, a, fns[a->size]); \ + } + +DO_2SHIFT_ENV(VQSHLU, qshlu_s) +DO_2SHIFT_ENV(VQSHL_U, qshl_u) +DO_2SHIFT_ENV(VQSHL_S, qshl_s) diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static inline void gen_neon_rsb(int size, TCGv_i32 t0, TCGv_i32 t1) } } -#define GEN_NEON_INTEGER_OP_ENV(name) do { \ - switch ((size << 1) | u) { \ - case 0: \ - gen_helper_neon_##name##_s8(tmp, cpu_env, tmp, tmp2); \ - break; \ - case 1: \ - gen_helper_neon_##name##_u8(tmp, cpu_env, tmp, tmp2); \ - break; \ - case 2: \ - gen_helper_neon_##name##_s16(tmp, cpu_env, tmp, tmp2); \ - break; \ - case 3: \ - gen_helper_neon_##name##_u16(tmp, cpu_env, tmp, tmp2); \ - break; \ - case 4: \ - gen_helper_neon_##name##_s32(tmp, cpu_env, tmp, tmp2); \ - break; \ - case 5: \ - gen_helper_neon_##name##_u32(tmp, cpu_env, tmp, tmp2); \ - break; \ - default: return 1; \ - }} while (0) - static TCGv_i32 neon_load_scratch(int scratch) { TCGv_i32 tmp = tcg_temp_new_i32(); @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) int size; int shift; int pass; - int count; int u; int vec_size; uint32_t imm; @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case 3: /* VRSRA */ case 4: /* VSRI */ case 5: /* VSHL, VSLI */ + case 6: /* VQSHLU */ + case 7: /* VQSHL */ return 1; /* handled by decodetree */ default: break; @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) size--; } shift = (insn >> 16) & ((1 << (3 + size)) - 1); - if (op < 8) { - /* Shift by immediate: - VSHR, VSRA, VRSHR, VRSRA, VSRI, VSHL, VQSHL, VQSHLU. */ - if (q && ((rd | rm) & 1)) { - return 1; - } - if (!u && (op == 4 || op == 6)) { - return 1; - } - /* Right shifts are encoded as N - shift, where N is the - element size in bits. */ - if (op <= 4) { - shift = shift - (1 << (size + 3)); - } - - if (size == 3) { - count = q + 1; - } else { - count = q ? 4: 2; - } - - /* To avoid excessive duplication of ops we implement shift - * by immediate using the variable shift operations. - */ - imm = dup_const(size, shift); - - for (pass = 0; pass < count; pass++) { - if (size == 3) { - neon_load_reg64(cpu_V0, rm + pass); - tcg_gen_movi_i64(cpu_V1, imm); - switch (op) { - case 6: /* VQSHLU */ - gen_helper_neon_qshlu_s64(cpu_V0, cpu_env, - cpu_V0, cpu_V1); - break; - case 7: /* VQSHL */ - if (u) { - gen_helper_neon_qshl_u64(cpu_V0, cpu_env, - cpu_V0, cpu_V1); - } else { - gen_helper_neon_qshl_s64(cpu_V0, cpu_env, - cpu_V0, cpu_V1); - } - break; - default: - g_assert_not_reached(); - } - neon_store_reg64(cpu_V0, rd + pass); - } else { /* size < 3 */ - /* Operands in T0 and T1. */ - tmp = neon_load_reg(rm, pass); - tmp2 = tcg_temp_new_i32(); - tcg_gen_movi_i32(tmp2, imm); - switch (op) { - case 6: /* VQSHLU */ - switch (size) { - case 0: - gen_helper_neon_qshlu_s8(tmp, cpu_env, - tmp, tmp2); - break; - case 1: - gen_helper_neon_qshlu_s16(tmp, cpu_env, - tmp, tmp2); - break; - case 2: - gen_helper_neon_qshlu_s32(tmp, cpu_env, - tmp, tmp2); - break; - default: - abort(); - } - break; - case 7: /* VQSHL */ - GEN_NEON_INTEGER_OP_ENV(qshl); - break; - default: - g_assert_not_reached(); - } - tcg_temp_free_i32(tmp2); - neon_store_reg(rd, pass, tmp); - } - } /* for pass */ - } else if (op < 10) { + if (op < 10) { /* Shift by immediate and narrow: VSHRN, VRSHRN, VQSHRN, VQRSHRN. */ int input_unsigned = (op == 8) ? !u : u; -- 2.20.1 Convert the Neon narrowing shifts where op==8 to decodetree: * VSHRN * VRSHRN * VQSHRUN * VQRSHRUN Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200522145520.6778-6-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 27 ++++++ target/arm/translate-neon.inc.c | 167 ++++++++++++++++++++++++++++++++ target/arm/translate.c | 1 + 3 files changed, 195 insertions(+) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VMINNM_fp_3s 1111 001 1 0 . 1 . .... .... 1111 ... 1 .... @3same_fp @2reg_shl_b .... ... . . . 001 shift:3 .... .... 0 q:1 . . .... \ &2reg_shift vm=%vm_dp vd=%vd_dp size=0 +# Narrowing right shifts: here the Q bit is part of the opcode decode +@2reg_shrn_d .... ... . . . 1 ..... .... .... 0 . . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=3 q=0 \ + shift=%neon_rshift_i5 +@2reg_shrn_s .... ... . . . 01 .... .... .... 0 . . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=2 q=0 \ + shift=%neon_rshift_i4 +@2reg_shrn_h .... ... . . . 001 ... .... .... 0 . . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=1 q=0 \ + shift=%neon_rshift_i3 + VSHR_S_2sh 1111 001 0 1 . ...... .... 0000 . . . 1 .... @2reg_shr_d VSHR_S_2sh 1111 001 0 1 . ...... .... 0000 . . . 1 .... @2reg_shr_s VSHR_S_2sh 1111 001 0 1 . ...... .... 0000 . . . 1 .... @2reg_shr_h @@ -XXX,XX +XXX,XX @@ VQSHL_U_64_2sh 1111 001 1 1 . ...... .... 0111 . . . 1 .... @2reg_shl_d VQSHL_U_2sh 1111 001 1 1 . ...... .... 0111 . . . 1 .... @2reg_shl_s VQSHL_U_2sh 1111 001 1 1 . ...... .... 0111 . . . 1 .... @2reg_shl_h VQSHL_U_2sh 1111 001 1 1 . ...... .... 0111 . . . 1 .... @2reg_shl_b + +VSHRN_64_2sh 1111 001 0 1 . ...... .... 1000 . 0 . 1 .... @2reg_shrn_d +VSHRN_32_2sh 1111 001 0 1 . ...... .... 1000 . 0 . 1 .... @2reg_shrn_s +VSHRN_16_2sh 1111 001 0 1 . ...... .... 1000 . 0 . 1 .... @2reg_shrn_h + +VRSHRN_64_2sh 1111 001 0 1 . ...... .... 1000 . 1 . 1 .... @2reg_shrn_d +VRSHRN_32_2sh 1111 001 0 1 . ...... .... 1000 . 1 . 1 .... @2reg_shrn_s +VRSHRN_16_2sh 1111 001 0 1 . ...... .... 1000 . 1 . 1 .... @2reg_shrn_h + +VQSHRUN_64_2sh 1111 001 1 1 . ...... .... 1000 . 0 . 1 .... @2reg_shrn_d +VQSHRUN_32_2sh 1111 001 1 1 . ...... .... 1000 . 0 . 1 .... @2reg_shrn_s +VQSHRUN_16_2sh 1111 001 1 1 . ...... .... 1000 . 0 . 1 .... @2reg_shrn_h + +VQRSHRUN_64_2sh 1111 001 1 1 . ...... .... 1000 . 1 . 1 .... @2reg_shrn_d +VQRSHRUN_32_2sh 1111 001 1 1 . ...... .... 1000 . 1 . 1 .... @2reg_shrn_s +VQRSHRUN_16_2sh 1111 001 1 1 . ...... .... 1000 . 1 . 1 .... @2reg_shrn_h diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ static bool do_2shift_env_32(DisasContext *s, arg_2reg_shift *a, DO_2SHIFT_ENV(VQSHLU, qshlu_s) DO_2SHIFT_ENV(VQSHL_U, qshl_u) DO_2SHIFT_ENV(VQSHL_S, qshl_s) + +static bool do_2shift_narrow_64(DisasContext *s, arg_2reg_shift *a, + NeonGenTwo64OpFn *shiftfn, + NeonGenNarrowEnvFn *narrowfn) +{ + /* 2-reg-and-shift narrowing-shift operations, size == 3 case */ + TCGv_i64 constimm, rm1, rm2; + TCGv_i32 rd; + + if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vm) & 0x10)) { + return false; + } + + if (a->vm & 1) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + /* + * This is always a right shift, and the shiftfn is always a + * left-shift helper, which thus needs the negated shift count. + */ + constimm = tcg_const_i64(-a->shift); + rm1 = tcg_temp_new_i64(); + rm2 = tcg_temp_new_i64(); + + /* Load both inputs first to avoid potential overwrite if rm == rd */ + neon_load_reg64(rm1, a->vm); + neon_load_reg64(rm2, a->vm + 1); + + shiftfn(rm1, rm1, constimm); + rd = tcg_temp_new_i32(); + narrowfn(rd, cpu_env, rm1); + neon_store_reg(a->vd, 0, rd); + + shiftfn(rm2, rm2, constimm); + rd = tcg_temp_new_i32(); + narrowfn(rd, cpu_env, rm2); + neon_store_reg(a->vd, 1, rd); + + tcg_temp_free_i64(rm1); + tcg_temp_free_i64(rm2); + tcg_temp_free_i64(constimm); + + return true; +} + +static bool do_2shift_narrow_32(DisasContext *s, arg_2reg_shift *a, + NeonGenTwoOpFn *shiftfn, + NeonGenNarrowEnvFn *narrowfn) +{ + /* 2-reg-and-shift narrowing-shift operations, size < 3 case */ + TCGv_i32 constimm, rm1, rm2, rm3, rm4; + TCGv_i64 rtmp; + uint32_t imm; + + if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vm) & 0x10)) { + return false; + } + + if (a->vm & 1) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + /* + * This is always a right shift, and the shiftfn is always a + * left-shift helper, which thus needs the negated shift count + * duplicated into each lane of the immediate value. + */ + if (a->size == 1) { + imm = (uint16_t)(-a->shift); + imm |= imm << 16; + } else { + /* size == 2 */ + imm = -a->shift; + } + constimm = tcg_const_i32(imm); + + /* Load all inputs first to avoid potential overwrite */ + rm1 = neon_load_reg(a->vm, 0); + rm2 = neon_load_reg(a->vm, 1); + rm3 = neon_load_reg(a->vm + 1, 0); + rm4 = neon_load_reg(a->vm + 1, 1); + rtmp = tcg_temp_new_i64(); + + shiftfn(rm1, rm1, constimm); + shiftfn(rm2, rm2, constimm); + + tcg_gen_concat_i32_i64(rtmp, rm1, rm2); + tcg_temp_free_i32(rm2); + + narrowfn(rm1, cpu_env, rtmp); + neon_store_reg(a->vd, 0, rm1); + + shiftfn(rm3, rm3, constimm); + shiftfn(rm4, rm4, constimm); + tcg_temp_free_i32(constimm); + + tcg_gen_concat_i32_i64(rtmp, rm3, rm4); + tcg_temp_free_i32(rm4); + + narrowfn(rm3, cpu_env, rtmp); + tcg_temp_free_i64(rtmp); + neon_store_reg(a->vd, 1, rm3); + return true; +} + +#define DO_2SN_64(INSN, FUNC, NARROWFUNC) \ + static bool trans_##INSN##_2sh(DisasContext *s, arg_2reg_shift *a) \ + { \ + return do_2shift_narrow_64(s, a, FUNC, NARROWFUNC); \ + } +#define DO_2SN_32(INSN, FUNC, NARROWFUNC) \ + static bool trans_##INSN##_2sh(DisasContext *s, arg_2reg_shift *a) \ + { \ + return do_2shift_narrow_32(s, a, FUNC, NARROWFUNC); \ + } + +static void gen_neon_narrow_u32(TCGv_i32 dest, TCGv_ptr env, TCGv_i64 src) +{ + tcg_gen_extrl_i64_i32(dest, src); +} + +static void gen_neon_narrow_u16(TCGv_i32 dest, TCGv_ptr env, TCGv_i64 src) +{ + gen_helper_neon_narrow_u16(dest, src); +} + +static void gen_neon_narrow_u8(TCGv_i32 dest, TCGv_ptr env, TCGv_i64 src) +{ + gen_helper_neon_narrow_u8(dest, src); +} + +DO_2SN_64(VSHRN_64, gen_ushl_i64, gen_neon_narrow_u32) +DO_2SN_32(VSHRN_32, gen_ushl_i32, gen_neon_narrow_u16) +DO_2SN_32(VSHRN_16, gen_helper_neon_shl_u16, gen_neon_narrow_u8) + +DO_2SN_64(VRSHRN_64, gen_helper_neon_rshl_u64, gen_neon_narrow_u32) +DO_2SN_32(VRSHRN_32, gen_helper_neon_rshl_u32, gen_neon_narrow_u16) +DO_2SN_32(VRSHRN_16, gen_helper_neon_rshl_u16, gen_neon_narrow_u8) + +DO_2SN_64(VQSHRUN_64, gen_sshl_i64, gen_helper_neon_unarrow_sat32) +DO_2SN_32(VQSHRUN_32, gen_sshl_i32, gen_helper_neon_unarrow_sat16) +DO_2SN_32(VQSHRUN_16, gen_helper_neon_shl_s16, gen_helper_neon_unarrow_sat8) + +DO_2SN_64(VQRSHRUN_64, gen_helper_neon_rshl_s64, gen_helper_neon_unarrow_sat32) +DO_2SN_32(VQRSHRUN_32, gen_helper_neon_rshl_s32, gen_helper_neon_unarrow_sat16) +DO_2SN_32(VQRSHRUN_16, gen_helper_neon_rshl_s16, gen_helper_neon_unarrow_sat8) diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case 5: /* VSHL, VSLI */ case 6: /* VQSHLU */ case 7: /* VQSHL */ + case 8: /* VSHRN, VRSHRN, VQSHRUN, VQRSHRUN */ return 1; /* handled by decodetree */ default: break; -- 2.20.1 Convert the remaining Neon narrowing shifts to decodetree: * VQSHRN * VQRSHRN Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200522145520.6778-7-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 20 ++++++ target/arm/translate-neon.inc.c | 15 +++++ target/arm/translate.c | 110 +------------------------------- 3 files changed, 37 insertions(+), 108 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VQSHRUN_16_2sh 1111 001 1 1 . ...... .... 1000 . 0 . 1 .... @2reg_shrn_h VQRSHRUN_64_2sh 1111 001 1 1 . ...... .... 1000 . 1 . 1 .... @2reg_shrn_d VQRSHRUN_32_2sh 1111 001 1 1 . ...... .... 1000 . 1 . 1 .... @2reg_shrn_s VQRSHRUN_16_2sh 1111 001 1 1 . ...... .... 1000 . 1 . 1 .... @2reg_shrn_h + +# VQSHRN with signed input +VQSHRN_S64_2sh 1111 001 0 1 . ...... .... 1001 . 0 . 1 .... @2reg_shrn_d +VQSHRN_S32_2sh 1111 001 0 1 . ...... .... 1001 . 0 . 1 .... @2reg_shrn_s +VQSHRN_S16_2sh 1111 001 0 1 . ...... .... 1001 . 0 . 1 .... @2reg_shrn_h + +# VQRSHRN with signed input +VQRSHRN_S64_2sh 1111 001 0 1 . ...... .... 1001 . 1 . 1 .... @2reg_shrn_d +VQRSHRN_S32_2sh 1111 001 0 1 . ...... .... 1001 . 1 . 1 .... @2reg_shrn_s +VQRSHRN_S16_2sh 1111 001 0 1 . ...... .... 1001 . 1 . 1 .... @2reg_shrn_h + +# VQSHRN with unsigned input +VQSHRN_U64_2sh 1111 001 1 1 . ...... .... 1001 . 0 . 1 .... @2reg_shrn_d +VQSHRN_U32_2sh 1111 001 1 1 . ...... .... 1001 . 0 . 1 .... @2reg_shrn_s +VQSHRN_U16_2sh 1111 001 1 1 . ...... .... 1001 . 0 . 1 .... @2reg_shrn_h + +# VQRSHRN with unsigned input +VQRSHRN_U64_2sh 1111 001 1 1 . ...... .... 1001 . 1 . 1 .... @2reg_shrn_d +VQRSHRN_U32_2sh 1111 001 1 1 . ...... .... 1001 . 1 . 1 .... @2reg_shrn_s +VQRSHRN_U16_2sh 1111 001 1 1 . ...... .... 1001 . 1 . 1 .... @2reg_shrn_h diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_2SN_32(VQSHRUN_16, gen_helper_neon_shl_s16, gen_helper_neon_unarrow_sat8) DO_2SN_64(VQRSHRUN_64, gen_helper_neon_rshl_s64, gen_helper_neon_unarrow_sat32) DO_2SN_32(VQRSHRUN_32, gen_helper_neon_rshl_s32, gen_helper_neon_unarrow_sat16) DO_2SN_32(VQRSHRUN_16, gen_helper_neon_rshl_s16, gen_helper_neon_unarrow_sat8) +DO_2SN_64(VQSHRN_S64, gen_sshl_i64, gen_helper_neon_narrow_sat_s32) +DO_2SN_32(VQSHRN_S32, gen_sshl_i32, gen_helper_neon_narrow_sat_s16) +DO_2SN_32(VQSHRN_S16, gen_helper_neon_shl_s16, gen_helper_neon_narrow_sat_s8) + +DO_2SN_64(VQRSHRN_S64, gen_helper_neon_rshl_s64, gen_helper_neon_narrow_sat_s32) +DO_2SN_32(VQRSHRN_S32, gen_helper_neon_rshl_s32, gen_helper_neon_narrow_sat_s16) +DO_2SN_32(VQRSHRN_S16, gen_helper_neon_rshl_s16, gen_helper_neon_narrow_sat_s8) + +DO_2SN_64(VQSHRN_U64, gen_ushl_i64, gen_helper_neon_narrow_sat_u32) +DO_2SN_32(VQSHRN_U32, gen_ushl_i32, gen_helper_neon_narrow_sat_u16) +DO_2SN_32(VQSHRN_U16, gen_helper_neon_shl_u16, gen_helper_neon_narrow_sat_u8) + +DO_2SN_64(VQRSHRN_U64, gen_helper_neon_rshl_u64, gen_helper_neon_narrow_sat_u32) +DO_2SN_32(VQRSHRN_U32, gen_helper_neon_rshl_u32, gen_helper_neon_narrow_sat_u16) +DO_2SN_32(VQRSHRN_U16, gen_helper_neon_rshl_u16, gen_helper_neon_narrow_sat_u8) diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static inline void gen_neon_unarrow_sats(int size, TCGv_i32 dest, TCGv_i64 src) } } -static inline void gen_neon_shift_narrow(int size, TCGv_i32 var, TCGv_i32 shift, - int q, int u) -{ - if (q) { - if (u) { - switch (size) { - case 1: gen_helper_neon_rshl_u16(var, var, shift); break; - case 2: gen_helper_neon_rshl_u32(var, var, shift); break; - default: abort(); - } - } else { - switch (size) { - case 1: gen_helper_neon_rshl_s16(var, var, shift); break; - case 2: gen_helper_neon_rshl_s32(var, var, shift); break; - default: abort(); - } - } - } else { - if (u) { - switch (size) { - case 1: gen_helper_neon_shl_u16(var, var, shift); break; - case 2: gen_ushl_i32(var, var, shift); break; - default: abort(); - } - } else { - switch (size) { - case 1: gen_helper_neon_shl_s16(var, var, shift); break; - case 2: gen_sshl_i32(var, var, shift); break; - default: abort(); - } - } - } -} - static inline void gen_neon_widen(TCGv_i64 dest, TCGv_i32 src, int size, int u) { if (u) { @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case 6: /* VQSHLU */ case 7: /* VQSHL */ case 8: /* VSHRN, VRSHRN, VQSHRUN, VQRSHRUN */ + case 9: /* VQSHRN, VQRSHRN */ return 1; /* handled by decodetree */ default: break; @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) size--; } shift = (insn >> 16) & ((1 << (3 + size)) - 1); - if (op < 10) { - /* Shift by immediate and narrow: - VSHRN, VRSHRN, VQSHRN, VQRSHRN. */ - int input_unsigned = (op == 8) ? !u : u; - if (rm & 1) { - return 1; - } - shift = shift - (1 << (size + 3)); - size++; - if (size == 3) { - tmp64 = tcg_const_i64(shift); - neon_load_reg64(cpu_V0, rm); - neon_load_reg64(cpu_V1, rm + 1); - for (pass = 0; pass < 2; pass++) { - TCGv_i64 in; - if (pass == 0) { - in = cpu_V0; - } else { - in = cpu_V1; - } - if (q) { - if (input_unsigned) { - gen_helper_neon_rshl_u64(cpu_V0, in, tmp64); - } else { - gen_helper_neon_rshl_s64(cpu_V0, in, tmp64); - } - } else { - if (input_unsigned) { - gen_ushl_i64(cpu_V0, in, tmp64); - } else { - gen_sshl_i64(cpu_V0, in, tmp64); - } - } - tmp = tcg_temp_new_i32(); - gen_neon_narrow_op(op == 8, u, size - 1, tmp, cpu_V0); - neon_store_reg(rd, pass, tmp); - } /* for pass */ - tcg_temp_free_i64(tmp64); - } else { - if (size == 1) { - imm = (uint16_t)shift; - imm |= imm << 16; - } else { - /* size == 2 */ - imm = (uint32_t)shift; - } - tmp2 = tcg_const_i32(imm); - tmp4 = neon_load_reg(rm + 1, 0); - tmp5 = neon_load_reg(rm + 1, 1); - for (pass = 0; pass < 2; pass++) { - if (pass == 0) { - tmp = neon_load_reg(rm, 0); - } else { - tmp = tmp4; - } - gen_neon_shift_narrow(size, tmp, tmp2, q, - input_unsigned); - if (pass == 0) { - tmp3 = neon_load_reg(rm, 1); - } else { - tmp3 = tmp5; - } - gen_neon_shift_narrow(size, tmp3, tmp2, q, - input_unsigned); - tcg_gen_concat_i32_i64(cpu_V0, tmp, tmp3); - tcg_temp_free_i32(tmp); - tcg_temp_free_i32(tmp3); - tmp = tcg_temp_new_i32(); - gen_neon_narrow_op(op == 8, u, size - 1, tmp, cpu_V0); - neon_store_reg(rd, pass, tmp); - } /* for pass */ - tcg_temp_free_i32(tmp2); - } - } else if (op == 10) { + if (op == 10) { /* VSHLL, VMOVL */ if (q || (rd & 1)) { return 1; -- 2.20.1 Convert the VSHLL and VMOVL insns from the 2-reg-shift group to decodetree. Since the loop always has two passes, we unroll it to avoid the awkward reassignment of one TCGv to another. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200522145520.6778-8-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 16 +++++++ target/arm/translate-neon.inc.c | 81 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 46 +------------------ 3 files changed, 99 insertions(+), 44 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VMINNM_fp_3s 1111 001 1 0 . 1 . .... .... 1111 ... 1 .... @3same_fp &2reg_shift vm=%vm_dp vd=%vd_dp size=1 q=0 \ shift=%neon_rshift_i3 +# Long left shifts: again Q is part of opcode decode +@2reg_shll_s .... ... . . . 1 shift:5 .... .... 0 . . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=2 q=0 +@2reg_shll_h .... ... . . . 01 shift:4 .... .... 0 . . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=1 q=0 +@2reg_shll_b .... ... . . . 001 shift:3 .... .... 0 . . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=0 q=0 + VSHR_S_2sh 1111 001 0 1 . ...... .... 0000 . . . 1 .... @2reg_shr_d VSHR_S_2sh 1111 001 0 1 . ...... .... 0000 . . . 1 .... @2reg_shr_s VSHR_S_2sh 1111 001 0 1 . ...... .... 0000 . . . 1 .... @2reg_shr_h @@ -XXX,XX +XXX,XX @@ VQSHRN_U16_2sh 1111 001 1 1 . ...... .... 1001 . 0 . 1 .... @2reg_shrn_h VQRSHRN_U64_2sh 1111 001 1 1 . ...... .... 1001 . 1 . 1 .... @2reg_shrn_d VQRSHRN_U32_2sh 1111 001 1 1 . ...... .... 1001 . 1 . 1 .... @2reg_shrn_s VQRSHRN_U16_2sh 1111 001 1 1 . ...... .... 1001 . 1 . 1 .... @2reg_shrn_h + +VSHLL_S_2sh 1111 001 0 1 . ...... .... 1010 . 0 . 1 .... @2reg_shll_s +VSHLL_S_2sh 1111 001 0 1 . ...... .... 1010 . 0 . 1 .... @2reg_shll_h +VSHLL_S_2sh 1111 001 0 1 . ...... .... 1010 . 0 . 1 .... @2reg_shll_b + +VSHLL_U_2sh 1111 001 1 1 . ...... .... 1010 . 0 . 1 .... @2reg_shll_s +VSHLL_U_2sh 1111 001 1 1 . ...... .... 1010 . 0 . 1 .... @2reg_shll_h +VSHLL_U_2sh 1111 001 1 1 . ...... .... 1010 . 0 . 1 .... @2reg_shll_b diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_2SN_32(VQSHRN_U16, gen_helper_neon_shl_u16, gen_helper_neon_narrow_sat_u8) DO_2SN_64(VQRSHRN_U64, gen_helper_neon_rshl_u64, gen_helper_neon_narrow_sat_u32) DO_2SN_32(VQRSHRN_U32, gen_helper_neon_rshl_u32, gen_helper_neon_narrow_sat_u16) DO_2SN_32(VQRSHRN_U16, gen_helper_neon_rshl_u16, gen_helper_neon_narrow_sat_u8) + +static bool do_vshll_2sh(DisasContext *s, arg_2reg_shift *a, + NeonGenWidenFn *widenfn, bool u) +{ + TCGv_i64 tmp; + TCGv_i32 rm0, rm1; + uint64_t widen_mask = 0; + + if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vm) & 0x10)) { + return false; + } + + if (a->vd & 1) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + /* + * This is a widen-and-shift operation. The shift is always less + * than the width of the source type, so after widening the input + * vector we can simply shift the whole 64-bit widened register, + * and then clear the potential overflow bits resulting from left + * bits of the narrow input appearing as right bits of the left + * neighbour narrow input. Calculate a mask of bits to clear. + */ + if ((a->shift != 0) && (a->size < 2 || u)) { + int esize = 8 << a->size; + widen_mask = MAKE_64BIT_MASK(0, esize); + widen_mask >>= esize - a->shift; + widen_mask = dup_const(a->size + 1, widen_mask); + } + + rm0 = neon_load_reg(a->vm, 0); + rm1 = neon_load_reg(a->vm, 1); + tmp = tcg_temp_new_i64(); + + widenfn(tmp, rm0); + if (a->shift != 0) { + tcg_gen_shli_i64(tmp, tmp, a->shift); + tcg_gen_andi_i64(tmp, tmp, ~widen_mask); + } + neon_store_reg64(tmp, a->vd); + + widenfn(tmp, rm1); + if (a->shift != 0) { + tcg_gen_shli_i64(tmp, tmp, a->shift); + tcg_gen_andi_i64(tmp, tmp, ~widen_mask); + } + neon_store_reg64(tmp, a->vd + 1); + tcg_temp_free_i64(tmp); + return true; +} + +static bool trans_VSHLL_S_2sh(DisasContext *s, arg_2reg_shift *a) +{ + NeonGenWidenFn *widenfn[] = { + gen_helper_neon_widen_s8, + gen_helper_neon_widen_s16, + tcg_gen_ext_i32_i64, + }; + return do_vshll_2sh(s, a, widenfn[a->size], false); +} + +static bool trans_VSHLL_U_2sh(DisasContext *s, arg_2reg_shift *a) +{ + NeonGenWidenFn *widenfn[] = { + gen_helper_neon_widen_u8, + gen_helper_neon_widen_u16, + tcg_gen_extu_i32_i64, + }; + return do_vshll_2sh(s, a, widenfn[a->size], true); +} diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case 7: /* VQSHL */ case 8: /* VSHRN, VRSHRN, VQSHRUN, VQRSHRUN */ case 9: /* VQSHRN, VQRSHRN */ + case 10: /* VSHLL, including VMOVL */ return 1; /* handled by decodetree */ default: break; @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) size--; } shift = (insn >> 16) & ((1 << (3 + size)) - 1); - if (op == 10) { - /* VSHLL, VMOVL */ - if (q || (rd & 1)) { - return 1; - } - tmp = neon_load_reg(rm, 0); - tmp2 = neon_load_reg(rm, 1); - for (pass = 0; pass < 2; pass++) { - if (pass == 1) - tmp = tmp2; - - gen_neon_widen(cpu_V0, tmp, size, u); - - if (shift != 0) { - /* The shift is less than the width of the source - type, so we can just shift the whole register. */ - tcg_gen_shli_i64(cpu_V0, cpu_V0, shift); - /* Widen the result of shift: we need to clear - * the potential overflow bits resulting from - * left bits of the narrow input appearing as - * right bits of left the neighbour narrow - * input. */ - if (size < 2 || !u) { - uint64_t imm64; - if (size == 0) { - imm = (0xffu >> (8 - shift)); - imm |= imm << 16; - } else if (size == 1) { - imm = 0xffff >> (16 - shift); - } else { - /* size == 2 */ - imm = 0xffffffff >> (32 - shift); - } - if (size < 2) { - imm64 = imm | (((uint64_t)imm) << 32); - } else { - imm64 = imm; - } - tcg_gen_andi_i64(cpu_V0, cpu_V0, ~imm64); - } - } - neon_store_reg64(cpu_V0, rd + pass); - } - } else if (op >= 14) { + if (op >= 14) { /* VCVT fixed-point. */ TCGv_ptr fpst; TCGv_i32 shiftv; -- 2.20.1 Convert the VCVT fixed-point conversion operations in the Neon 2-regs-and-shift group to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200522145520.6778-9-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 11 +++++ target/arm/translate-neon.inc.c | 49 +++++++++++++++++++++ target/arm/translate.c | 75 +-------------------------------- 3 files changed, 62 insertions(+), 73 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VMINNM_fp_3s 1111 001 1 0 . 1 . .... .... 1111 ... 1 .... @3same_fp @2reg_shll_b .... ... . . . 001 shift:3 .... .... 0 . . . .... \ &2reg_shift vm=%vm_dp vd=%vd_dp size=0 q=0 +# We use size=0 for fp32 and size=1 for fp16 to match the 3-same encodings. +@2reg_vcvt .... ... . . . 1 ..... .... .... . q:1 . . .... \ + &2reg_shift vm=%vm_dp vd=%vd_dp size=0 shift=%neon_rshift_i5 + VSHR_S_2sh 1111 001 0 1 . ...... .... 0000 . . . 1 .... @2reg_shr_d VSHR_S_2sh 1111 001 0 1 . ...... .... 0000 . . . 1 .... @2reg_shr_s VSHR_S_2sh 1111 001 0 1 . ...... .... 0000 . . . 1 .... @2reg_shr_h @@ -XXX,XX +XXX,XX @@ VSHLL_S_2sh 1111 001 0 1 . ...... .... 1010 . 0 . 1 .... @2reg_shll_b VSHLL_U_2sh 1111 001 1 1 . ...... .... 1010 . 0 . 1 .... @2reg_shll_s VSHLL_U_2sh 1111 001 1 1 . ...... .... 1010 . 0 . 1 .... @2reg_shll_h VSHLL_U_2sh 1111 001 1 1 . ...... .... 1010 . 0 . 1 .... @2reg_shll_b + +# VCVT fixed<->float conversions +# TODO: FP16 fixed<->float conversions are opc==0b1100 and 0b1101 +VCVT_SF_2sh 1111 001 0 1 . ...... .... 1110 0 . . 1 .... @2reg_vcvt +VCVT_UF_2sh 1111 001 1 1 . ...... .... 1110 0 . . 1 .... @2reg_vcvt +VCVT_FS_2sh 1111 001 0 1 . ...... .... 1111 0 . . 1 .... @2reg_vcvt +VCVT_FU_2sh 1111 001 1 1 . ...... .... 1111 0 . . 1 .... @2reg_vcvt diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ static bool trans_VSHLL_U_2sh(DisasContext *s, arg_2reg_shift *a) }; return do_vshll_2sh(s, a, widenfn[a->size], true); } + +static bool do_fp_2sh(DisasContext *s, arg_2reg_shift *a, + NeonGenTwoSingleOPFn *fn) +{ + /* FP operations in 2-reg-and-shift group */ + TCGv_i32 tmp, shiftv; + TCGv_ptr fpstatus; + int pass; + + if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd | a->vm) & 0x10)) { + return false; + } + + if ((a->vm | a->vd) & a->q) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + fpstatus = get_fpstatus_ptr(1); + shiftv = tcg_const_i32(a->shift); + for (pass = 0; pass < (a->q ? 4 : 2); pass++) { + tmp = neon_load_reg(a->vm, pass); + fn(tmp, tmp, shiftv, fpstatus); + neon_store_reg(a->vd, pass, tmp); + } + tcg_temp_free_ptr(fpstatus); + tcg_temp_free_i32(shiftv); + return true; +} + +#define DO_FP_2SH(INSN, FUNC) \ + static bool trans_##INSN##_2sh(DisasContext *s, arg_2reg_shift *a) \ + { \ + return do_fp_2sh(s, a, FUNC); \ + } + +DO_FP_2SH(VCVT_SF, gen_helper_vfp_sltos) +DO_FP_2SH(VCVT_UF, gen_helper_vfp_ultos) +DO_FP_2SH(VCVT_FS, gen_helper_vfp_tosls_round_to_zero) +DO_FP_2SH(VCVT_FU, gen_helper_vfp_touls_round_to_zero) diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) int q; int rd, rn, rm, rd_ofs, rn_ofs, rm_ofs; int size; - int shift; int pass; int u; int vec_size; @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) return 1; } else if (insn & (1 << 4)) { if ((insn & 0x00380080) != 0) { - /* Two registers and shift. */ - op = (insn >> 8) & 0xf; - - switch (op) { - case 0: /* VSHR */ - case 1: /* VSRA */ - case 2: /* VRSHR */ - case 3: /* VRSRA */ - case 4: /* VSRI */ - case 5: /* VSHL, VSLI */ - case 6: /* VQSHLU */ - case 7: /* VQSHL */ - case 8: /* VSHRN, VRSHRN, VQSHRUN, VQRSHRUN */ - case 9: /* VQSHRN, VQRSHRN */ - case 10: /* VSHLL, including VMOVL */ - return 1; /* handled by decodetree */ - default: - break; - } - - if (insn & (1 << 7)) { - /* 64-bit shift. */ - if (op > 7) { - return 1; - } - size = 3; - } else { - size = 2; - while ((insn & (1 << (size + 19))) == 0) - size--; - } - shift = (insn >> 16) & ((1 << (3 + size)) - 1); - if (op >= 14) { - /* VCVT fixed-point. */ - TCGv_ptr fpst; - TCGv_i32 shiftv; - VFPGenFixPointFn *fn; - - if (!(insn & (1 << 21)) || (q && ((rd | rm) & 1))) { - return 1; - } - - if (!(op & 1)) { - if (u) { - fn = gen_helper_vfp_ultos; - } else { - fn = gen_helper_vfp_sltos; - } - } else { - if (u) { - fn = gen_helper_vfp_touls_round_to_zero; - } else { - fn = gen_helper_vfp_tosls_round_to_zero; - } - } - - /* We have already masked out the must-be-1 top bit of imm6, - * hence this 32-shift where the ARM ARM has 64-imm6. - */ - shift = 32 - shift; - fpst = get_fpstatus_ptr(1); - shiftv = tcg_const_i32(shift); - for (pass = 0; pass < (q ? 4 : 2); pass++) { - TCGv_i32 tmpf = neon_load_reg(rm, pass); - fn(tmpf, tmpf, shiftv, fpst); - neon_store_reg(rd, pass, tmpf); - } - tcg_temp_free_ptr(fpst); - tcg_temp_free_i32(shiftv); - } else { - return 1; - } + /* Two registers and shift: handled by decodetree */ + return 1; } else { /* (insn & 0x00380080) == 0 */ int invert, reg_ofs, vec_size; -- 2.20.1 Convert the insns in the one-register-and-immediate group to decodetree. In the new decode, our asimd_imm_const() function returns a 64-bit value rather than a 32-bit one, which means we don't need to treat cmode=14 op=1 as a special case in the decoder (it is the only encoding where the two halves of the 64-bit value are different). Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200522145520.6778-10-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 22 ++++++ target/arm/translate-neon.inc.c | 118 ++++++++++++++++++++++++++++++++ target/arm/translate.c | 101 +-------------------------- 3 files changed, 142 insertions(+), 99 deletions(-) diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-dp.decode +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ VCVT_SF_2sh 1111 001 0 1 . ...... .... 1110 0 . . 1 .... @2reg_vcvt VCVT_UF_2sh 1111 001 1 1 . ...... .... 1110 0 . . 1 .... @2reg_vcvt VCVT_FS_2sh 1111 001 0 1 . ...... .... 1111 0 . . 1 .... @2reg_vcvt VCVT_FU_2sh 1111 001 1 1 . ...... .... 1111 0 . . 1 .... @2reg_vcvt + +###################################################################### +# 1-reg-and-modified-immediate grouping: +# 1111 001 i 1 D 000 imm:3 Vd:4 cmode:4 0 Q op 1 Vm:4 +###################################################################### + +&1reg_imm vd q imm cmode op + +%asimd_imm_value 24:1 16:3 0:4 + +@1reg_imm .... ... . . . ... ... .... .... . q:1 . . .... \ + &1reg_imm imm=%asimd_imm_value vd=%vd_dp + +# The cmode/op bits here decode VORR/VBIC/VMOV/VMNV, but +# not in a way we can conveniently represent in decodetree without +# a lot of repetition: +# VORR: op=0, (cmode & 1) && cmode < 12 +# VBIC: op=1, (cmode & 1) && cmode < 12 +# VMOV: everything else +# So we have a single decode line and check the cmode/op in the +# trans function. +Vimm_1r 1111 001 . 1 . 000 ... .... cmode:4 0 . op:1 1 .... @1reg_imm diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-neon.inc.c +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ DO_FP_2SH(VCVT_SF, gen_helper_vfp_sltos) DO_FP_2SH(VCVT_UF, gen_helper_vfp_ultos) DO_FP_2SH(VCVT_FS, gen_helper_vfp_tosls_round_to_zero) DO_FP_2SH(VCVT_FU, gen_helper_vfp_touls_round_to_zero) + +static uint64_t asimd_imm_const(uint32_t imm, int cmode, int op) +{ + /* + * Expand the encoded constant. + * Note that cmode = 2,3,4,5,6,7,10,11,12,13 imm=0 is UNPREDICTABLE. + * We choose to not special-case this and will behave as if a + * valid constant encoding of 0 had been given. + * cmode = 15 op = 1 must UNDEF; we assume decode has handled that. + */ + switch (cmode) { + case 0: case 1: + /* no-op */ + break; + case 2: case 3: + imm <<= 8; + break; + case 4: case 5: + imm <<= 16; + break; + case 6: case 7: + imm <<= 24; + break; + case 8: case 9: + imm |= imm << 16; + break; + case 10: case 11: + imm = (imm << 8) | (imm << 24); + break; + case 12: + imm = (imm << 8) | 0xff; + break; + case 13: + imm = (imm << 16) | 0xffff; + break; + case 14: + if (op) { + /* + * This is the only case where the top and bottom 32 bits + * of the encoded constant differ. + */ + uint64_t imm64 = 0; + int n; + + for (n = 0; n < 8; n++) { + if (imm & (1 << n)) { + imm64 |= (0xffULL << (n * 8)); + } + } + return imm64; + } + imm |= (imm << 8) | (imm << 16) | (imm << 24); + break; + case 15: + imm = ((imm & 0x80) << 24) | ((imm & 0x3f) << 19) + | ((imm & 0x40) ? (0x1f << 25) : (1 << 30)); + break; + } + if (op) { + imm = ~imm; + } + return dup_const(MO_32, imm); +} + +static bool do_1reg_imm(DisasContext *s, arg_1reg_imm *a, + GVecGen2iFn *fn) +{ + uint64_t imm; + int reg_ofs, vec_size; + + if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && (a->vd & 0x10)) { + return false; + } + + if (a->vd & a->q) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + reg_ofs = neon_reg_offset(a->vd, 0); + vec_size = a->q ? 16 : 8; + imm = asimd_imm_const(a->imm, a->cmode, a->op); + + fn(MO_64, reg_ofs, reg_ofs, imm, vec_size, vec_size); + return true; +} + +static void gen_VMOV_1r(unsigned vece, uint32_t dofs, uint32_t aofs, + int64_t c, uint32_t oprsz, uint32_t maxsz) +{ + tcg_gen_gvec_dup_imm(MO_64, dofs, oprsz, maxsz, c); +} + +static bool trans_Vimm_1r(DisasContext *s, arg_1reg_imm *a) +{ + /* Handle decode of cmode/op here between VORR/VBIC/VMOV */ + GVecGen2iFn *fn; + + if ((a->cmode & 1) && a->cmode < 12) { + /* for op=1, the imm will be inverted, so BIC becomes AND. */ + fn = a->op ? tcg_gen_gvec_andi : tcg_gen_gvec_ori; + } else { + /* There is one unallocated cmode/op combination in this space */ + if (a->cmode == 15 && a->op == 1) { + return false; + } + fn = gen_VMOV_1r; + } + return do_1reg_imm(s, a, fn); +} diff --git a/target/arm/translate.c b/target/arm/translate.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) /* Three register same length: handled by decodetree */ return 1; } else if (insn & (1 << 4)) { - if ((insn & 0x00380080) != 0) { - /* Two registers and shift: handled by decodetree */ - return 1; - } else { /* (insn & 0x00380080) == 0 */ - int invert, reg_ofs, vec_size; - - if (q && (rd & 1)) { - return 1; - } - - op = (insn >> 8) & 0xf; - /* One register and immediate. */ - imm = (u << 7) | ((insn >> 12) & 0x70) | (insn & 0xf); - invert = (insn & (1 << 5)) != 0; - /* Note that op = 2,3,4,5,6,7,10,11,12,13 imm=0 is UNPREDICTABLE. - * We choose to not special-case this and will behave as if a - * valid constant encoding of 0 had been given. - */ - switch (op) { - case 0: case 1: - /* no-op */ - break; - case 2: case 3: - imm <<= 8; - break; - case 4: case 5: - imm <<= 16; - break; - case 6: case 7: - imm <<= 24; - break; - case 8: case 9: - imm |= imm << 16; - break; - case 10: case 11: - imm = (imm << 8) | (imm << 24); - break; - case 12: - imm = (imm << 8) | 0xff; - break; - case 13: - imm = (imm << 16) | 0xffff; - break; - case 14: - imm |= (imm << 8) | (imm << 16) | (imm << 24); - if (invert) { - imm = ~imm; - } - break; - case 15: - if (invert) { - return 1; - } - imm = ((imm & 0x80) << 24) | ((imm & 0x3f) << 19) - | ((imm & 0x40) ? (0x1f << 25) : (1 << 30)); - break; - } - if (invert) { - imm = ~imm; - } - - reg_ofs = neon_reg_offset(rd, 0); - vec_size = q ? 16 : 8; - - if (op & 1 && op < 12) { - if (invert) { - /* The immediate value has already been inverted, - * so BIC becomes AND. - */ - tcg_gen_gvec_andi(MO_32, reg_ofs, reg_ofs, imm, - vec_size, vec_size); - } else { - tcg_gen_gvec_ori(MO_32, reg_ofs, reg_ofs, imm, - vec_size, vec_size); - } - } else { - /* VMOV, VMVN. */ - if (op == 14 && invert) { - TCGv_i64 t64 = tcg_temp_new_i64(); - - for (pass = 0; pass <= q; ++pass) { - uint64_t val = 0; - int n; - - for (n = 0; n < 8; n++) { - if (imm & (1 << (n + pass * 8))) { - val |= 0xffull << (n * 8); - } - } - tcg_gen_movi_i64(t64, val); - neon_store_reg64(t64, rd + pass); - } - tcg_temp_free_i64(t64); - } else { - tcg_gen_gvec_dup_imm(MO_32, reg_ofs, vec_size, - vec_size, imm); - } - } - } + /* Two registers and shift or reg and imm: handled by decodetree */ + return 1; } else { /* (insn & 0x00800010 == 0x00800000) */ if (size != 3) { op = (insn >> 8) & 0xf; -- 2.20.1 Last pullreq before 6.0 softfreeze: a few minor feature patches, some bugfixes, some cleanups. -- PMM The following changes since commit 6f34661b6c97a37a5efc27d31c037ddeda4547e2: Merge remote-tracking branch 'remotes/vivier2/tags/trivial-branch-for-6.0-pull-request' into staging (2021-03-11 18:55:27 +0000) are available in the Git repository at: https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20210312-1 for you to fetch changes up to 41f09f2e9f09e4dd386d84174a6dcb5136af17ca: hw/display/pxa2xx: Inline template header (2021-03-12 13:26:08 +0000) ---------------------------------------------------------------- target-arm queue: * versal: Support XRAMs and XRAM controller * smmu: Various minor bug fixes * SVE emulation: fix bugs handling odd vector lengths * allwinner-sun8i-emac: traverse transmit queue using TX_CUR_DESC register value * tests/acceptance: fix orangepi-pc acceptance tests * hw/timer/sse-timer: Propagate eventual error in sse_timer_realize() * hw/arm/virt: KVM: The IPA lower bound is 32 * npcm7xx: support MFT module * pl110, pxa2xx_lcd: tidy up template headers ---------------------------------------------------------------- Andrew Jones (2): accel: kvm: Fix kvm_type invocation hw/arm/virt: KVM: The IPA lower bound is 32 Edgar E. Iglesias (2): hw/misc: versal: Add a model of the XRAM controller hw/arm: versal: Add support for the XRAMs Eric Auger (7): intel_iommu: Fix mask may be uninitialized in vtd_context_device_invalidate dma: Introduce dma_aligned_pow2_mask() virtio-iommu: Handle non power of 2 range invalidations hw/arm/smmu-common: Fix smmu_iotlb_inv_iova when asid is not set hw/arm/smmuv3: Enforce invalidation on a power of two range hw/arm/smmuv3: Fix SMMU_CMD_CFGI_STE_RANGE handling hw/arm/smmuv3: Uniformize sid traces Hao Wu (5): hw/misc: Add GPIOs for duty in NPCM7xx PWM hw/misc: Add NPCM7XX MFT Module hw/arm: Add MFT device to NPCM7xx Soc hw/arm: Connect PWM fans in NPCM7XX boards tests/qtest: Test PWM fan RPM using MFT in PWM test Niek Linnenbank (5): hw/net/allwinner-sun8i-emac: traverse transmit queue using TX_CUR_DESC register value tests/acceptance/boot_linux_console: remove Armbian 19.11.3 bionic test for orangepi-pc machine tests/acceptance/boot_linux_console: change URL for test_arm_orangepi_bionic_20_08 tests/acceptance: update sunxi kernel from armbian to 5.10.16 tests/acceptance: drop ARMBIAN_ARTIFACTS_CACHED condition for orangepi-pc, cubieboard tests Peter Maydell (9): hw/display/pl110: Remove dead code for non-32-bpp surfaces hw/display/pl110: Pull included-once parts of template header into pl110.c hw/display/pl110: Remove use of BITS from pl110_template.h hw/display/pxa2xx_lcd: Remove dead code for non-32-bpp surfaces hw/display/pxa2xx_lcd: Remove dest_width state field hw/display/pxa2xx: Remove use of BITS in pxa2xx_template.h hw/display/pxa2xx: Apply brace-related coding style fixes to template header hw/display/pxa2xx: Apply whitespace-only coding style fixes to template header hw/display/pxa2xx: Inline template header Philippe Mathieu-Daudé (1): hw/timer/sse-timer: Propagate eventual error in sse_timer_realize() Richard Henderson (8): target/arm: Fix sve_uzp_p vs odd vector lengths target/arm: Fix sve_zip_p vs odd vector lengths target/arm: Fix sve_punpk_p vs odd vector lengths target/arm: Update find_last_active for PREDDESC target/arm: Update BRKA, BRKB, BRKN for PREDDESC target/arm: Update CNTP for PREDDESC target/arm: Update WHILE for PREDDESC target/arm: Update sve reduction vs simd_desc docs/system/arm/nuvoton.rst | 2 +- docs/system/arm/xlnx-versal-virt.rst | 1 + hw/arm/smmu-internal.h | 5 + hw/display/pl110_template.h | 120 +------- hw/display/pxa2xx_template.h | 447 --------------------------- include/hw/arm/npcm7xx.h | 13 +- include/hw/arm/xlnx-versal.h | 13 + include/hw/boards.h | 1 + include/hw/misc/npcm7xx_mft.h | 70 +++++ include/hw/misc/npcm7xx_pwm.h | 4 +- include/hw/misc/xlnx-versal-xramc.h | 97 ++++++ include/sysemu/dma.h | 12 + target/arm/kvm_arm.h | 6 +- accel/kvm/kvm-all.c | 2 + hw/arm/npcm7xx.c | 45 ++- hw/arm/npcm7xx_boards.c | 99 ++++++ hw/arm/smmu-common.c | 32 +- hw/arm/smmuv3.c | 58 ++-- hw/arm/virt.c | 23 +- hw/arm/xlnx-versal.c | 36 +++ hw/display/pl110.c | 123 +++++--- hw/display/pxa2xx_lcd.c | 520 ++++++++++++++++++++++++++----- hw/i386/intel_iommu.c | 32 +- hw/misc/npcm7xx_mft.c | 540 +++++++++++++++++++++++++++++++++ hw/misc/npcm7xx_pwm.c | 4 + hw/misc/xlnx-versal-xramc.c | 253 +++++++++++++++ hw/net/allwinner-sun8i-emac.c | 62 ++-- hw/timer/sse-timer.c | 1 + hw/virtio/virtio-iommu.c | 19 +- softmmu/dma-helpers.c | 26 ++ target/arm/kvm.c | 4 +- target/arm/sve_helper.c | 107 ++++--- target/arm/translate-sve.c | 26 +- tests/qtest/npcm7xx_pwm-test.c | 205 ++++++++++++- hw/arm/trace-events | 24 +- hw/misc/meson.build | 2 + hw/misc/trace-events | 8 + tests/acceptance/boot_linux_console.py | 120 +++----- tests/acceptance/replay_kernel.py | 10 +- 39 files changed, 2235 insertions(+), 937 deletions(-) delete mode 100644 hw/display/pxa2xx_template.h create mode 100644 include/hw/misc/npcm7xx_mft.h create mode 100644 include/hw/misc/xlnx-versal-xramc.h create mode 100644 hw/misc/npcm7xx_mft.c create mode 100644 hw/misc/xlnx-versal-xramc.c From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Add a model of the Xilinx Versal Accelerator RAM (XRAM). This is mainly a stub to make firmware happy. The size of the RAMs can be probed. The interrupt mask logic is modelled but none of the interrups will ever be raised unless injected. Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20210308224637.2949533-2-edgar.iglesias@gmail.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/misc/xlnx-versal-xramc.h | 97 +++++++++++ hw/misc/xlnx-versal-xramc.c | 253 ++++++++++++++++++++++++++++ hw/misc/meson.build | 1 + 3 files changed, 351 insertions(+) create mode 100644 include/hw/misc/xlnx-versal-xramc.h create mode 100644 hw/misc/xlnx-versal-xramc.c diff --git a/include/hw/misc/xlnx-versal-xramc.h b/include/hw/misc/xlnx-versal-xramc.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/misc/xlnx-versal-xramc.h @@ -XXX,XX +XXX,XX @@ +/* + * QEMU model of the Xilinx XRAM Controller. + * + * Copyright (c) 2021 Xilinx Inc. + * SPDX-License-Identifier: GPL-2.0-or-later + * Written by Edgar E. Iglesias <edgar.iglesias@xilinx.com> + */ + +#ifndef XLNX_VERSAL_XRAMC_H +#define XLNX_VERSAL_XRAMC_H + +#include "hw/sysbus.h" +#include "hw/register.h" + +#define TYPE_XLNX_XRAM_CTRL "xlnx.versal-xramc" + +#define XLNX_XRAM_CTRL(obj) \ + OBJECT_CHECK(XlnxXramCtrl, (obj), TYPE_XLNX_XRAM_CTRL) + +REG32(XRAM_ERR_CTRL, 0x0) + FIELD(XRAM_ERR_CTRL, UE_RES, 3, 1) + FIELD(XRAM_ERR_CTRL, PWR_ERR_RES, 2, 1) + FIELD(XRAM_ERR_CTRL, PZ_ERR_RES, 1, 1) + FIELD(XRAM_ERR_CTRL, APB_ERR_RES, 0, 1) +REG32(XRAM_ISR, 0x4) + FIELD(XRAM_ISR, INV_APB, 0, 1) +REG32(XRAM_IMR, 0x8) + FIELD(XRAM_IMR, INV_APB, 0, 1) +REG32(XRAM_IEN, 0xc) + FIELD(XRAM_IEN, INV_APB, 0, 1) +REG32(XRAM_IDS, 0x10) + FIELD(XRAM_IDS, INV_APB, 0, 1) +REG32(XRAM_ECC_CNTL, 0x14) + FIELD(XRAM_ECC_CNTL, FI_MODE, 2, 1) + FIELD(XRAM_ECC_CNTL, DET_ONLY, 1, 1) + FIELD(XRAM_ECC_CNTL, ECC_ON_OFF, 0, 1) +REG32(XRAM_CLR_EXE, 0x18) + FIELD(XRAM_CLR_EXE, MON_7, 7, 1) + FIELD(XRAM_CLR_EXE, MON_6, 6, 1) + FIELD(XRAM_CLR_EXE, MON_5, 5, 1) + FIELD(XRAM_CLR_EXE, MON_4, 4, 1) + FIELD(XRAM_CLR_EXE, MON_3, 3, 1) + FIELD(XRAM_CLR_EXE, MON_2, 2, 1) + FIELD(XRAM_CLR_EXE, MON_1, 1, 1) + FIELD(XRAM_CLR_EXE, MON_0, 0, 1) +REG32(XRAM_CE_FFA, 0x1c) + FIELD(XRAM_CE_FFA, ADDR, 0, 20) +REG32(XRAM_CE_FFD0, 0x20) +REG32(XRAM_CE_FFD1, 0x24) +REG32(XRAM_CE_FFD2, 0x28) +REG32(XRAM_CE_FFD3, 0x2c) +REG32(XRAM_CE_FFE, 0x30) + FIELD(XRAM_CE_FFE, SYNDROME, 0, 16) +REG32(XRAM_UE_FFA, 0x34) + FIELD(XRAM_UE_FFA, ADDR, 0, 20) +REG32(XRAM_UE_FFD0, 0x38) +REG32(XRAM_UE_FFD1, 0x3c) +REG32(XRAM_UE_FFD2, 0x40) +REG32(XRAM_UE_FFD3, 0x44) +REG32(XRAM_UE_FFE, 0x48) + FIELD(XRAM_UE_FFE, SYNDROME, 0, 16) +REG32(XRAM_FI_D0, 0x4c) +REG32(XRAM_FI_D1, 0x50) +REG32(XRAM_FI_D2, 0x54) +REG32(XRAM_FI_D3, 0x58) +REG32(XRAM_FI_SY, 0x5c) + FIELD(XRAM_FI_SY, DATA, 0, 16) +REG32(XRAM_RMW_UE_FFA, 0x70) + FIELD(XRAM_RMW_UE_FFA, ADDR, 0, 20) +REG32(XRAM_FI_CNTR, 0x74) + FIELD(XRAM_FI_CNTR, COUNT, 0, 24) +REG32(XRAM_IMP, 0x80) + FIELD(XRAM_IMP, SIZE, 0, 4) +REG32(XRAM_PRDY_DBG, 0x84) + FIELD(XRAM_PRDY_DBG, ISLAND3, 12, 4) + FIELD(XRAM_PRDY_DBG, ISLAND2, 8, 4) + FIELD(XRAM_PRDY_DBG, ISLAND1, 4, 4) + FIELD(XRAM_PRDY_DBG, ISLAND0, 0, 4) +REG32(XRAM_SAFETY_CHK, 0xff8) + +#define XRAM_CTRL_R_MAX (R_XRAM_SAFETY_CHK + 1) + +typedef struct XlnxXramCtrl { + SysBusDevice parent_obj; + MemoryRegion ram; + qemu_irq irq; + + struct { + uint64_t size; + unsigned int encoded_size; + } cfg; + + RegisterInfoArray *reg_array; + uint32_t regs[XRAM_CTRL_R_MAX]; + RegisterInfo regs_info[XRAM_CTRL_R_MAX]; +} XlnxXramCtrl; +#endif diff --git a/hw/misc/xlnx-versal-xramc.c b/hw/misc/xlnx-versal-xramc.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/misc/xlnx-versal-xramc.c @@ -XXX,XX +XXX,XX @@ +/* + * QEMU model of the Xilinx XRAM Controller. + * + * Copyright (c) 2021 Xilinx Inc. + * SPDX-License-Identifier: GPL-2.0-or-later + * Written by Edgar E. Iglesias <edgar.iglesias@xilinx.com> + */ + +#include "qemu/osdep.h" +#include "qemu/units.h" +#include "qapi/error.h" +#include "migration/vmstate.h" +#include "hw/sysbus.h" +#include "hw/register.h" +#include "hw/qdev-properties.h" +#include "hw/irq.h" +#include "hw/misc/xlnx-versal-xramc.h" + +#ifndef XLNX_XRAM_CTRL_ERR_DEBUG +#define XLNX_XRAM_CTRL_ERR_DEBUG 0 +#endif + +static void xram_update_irq(XlnxXramCtrl *s) +{ + bool pending = s->regs[R_XRAM_ISR] & ~s->regs[R_XRAM_IMR]; + qemu_set_irq(s->irq, pending); +} + +static void xram_isr_postw(RegisterInfo *reg, uint64_t val64) +{ + XlnxXramCtrl *s = XLNX_XRAM_CTRL(reg->opaque); + xram_update_irq(s); +} + +static uint64_t xram_ien_prew(RegisterInfo *reg, uint64_t val64) +{ + XlnxXramCtrl *s = XLNX_XRAM_CTRL(reg->opaque); + uint32_t val = val64; + + s->regs[R_XRAM_IMR] &= ~val; + xram_update_irq(s); + return 0; +} + +static uint64_t xram_ids_prew(RegisterInfo *reg, uint64_t val64) +{ + XlnxXramCtrl *s = XLNX_XRAM_CTRL(reg->opaque); + uint32_t val = val64; + + s->regs[R_XRAM_IMR] |= val; + xram_update_irq(s); + return 0; +} + +static const RegisterAccessInfo xram_ctrl_regs_info[] = { + { .name = "XRAM_ERR_CTRL", .addr = A_XRAM_ERR_CTRL, + .reset = 0xf, + .rsvd = 0xfffffff0, + },{ .name = "XRAM_ISR", .addr = A_XRAM_ISR, + .rsvd = 0xfffff800, + .w1c = 0x7ff, + .post_write = xram_isr_postw, + },{ .name = "XRAM_IMR", .addr = A_XRAM_IMR, + .reset = 0x7ff, + .rsvd = 0xfffff800, + .ro = 0x7ff, + },{ .name = "XRAM_IEN", .addr = A_XRAM_IEN, + .rsvd = 0xfffff800, + .pre_write = xram_ien_prew, + },{ .name = "XRAM_IDS", .addr = A_XRAM_IDS, + .rsvd = 0xfffff800, + .pre_write = xram_ids_prew, + },{ .name = "XRAM_ECC_CNTL", .addr = A_XRAM_ECC_CNTL, + .rsvd = 0xfffffff8, + },{ .name = "XRAM_CLR_EXE", .addr = A_XRAM_CLR_EXE, + .rsvd = 0xffffff00, + },{ .name = "XRAM_CE_FFA", .addr = A_XRAM_CE_FFA, + .rsvd = 0xfff00000, + .ro = 0xfffff, + },{ .name = "XRAM_CE_FFD0", .addr = A_XRAM_CE_FFD0, + .ro = 0xffffffff, + },{ .name = "XRAM_CE_FFD1", .addr = A_XRAM_CE_FFD1, + .ro = 0xffffffff, + },{ .name = "XRAM_CE_FFD2", .addr = A_XRAM_CE_FFD2, + .ro = 0xffffffff, + },{ .name = "XRAM_CE_FFD3", .addr = A_XRAM_CE_FFD3, + .ro = 0xffffffff, + },{ .name = "XRAM_CE_FFE", .addr = A_XRAM_CE_FFE, + .rsvd = 0xffff0000, + .ro = 0xffff, + },{ .name = "XRAM_UE_FFA", .addr = A_XRAM_UE_FFA, + .rsvd = 0xfff00000, + .ro = 0xfffff, + },{ .name = "XRAM_UE_FFD0", .addr = A_XRAM_UE_FFD0, + .ro = 0xffffffff, + },{ .name = "XRAM_UE_FFD1", .addr = A_XRAM_UE_FFD1, + .ro = 0xffffffff, + },{ .name = "XRAM_UE_FFD2", .addr = A_XRAM_UE_FFD2, + .ro = 0xffffffff, + },{ .name = "XRAM_UE_FFD3", .addr = A_XRAM_UE_FFD3, + .ro = 0xffffffff, + },{ .name = "XRAM_UE_FFE", .addr = A_XRAM_UE_FFE, + .rsvd = 0xffff0000, + .ro = 0xffff, + },{ .name = "XRAM_FI_D0", .addr = A_XRAM_FI_D0, + },{ .name = "XRAM_FI_D1", .addr = A_XRAM_FI_D1, + },{ .name = "XRAM_FI_D2", .addr = A_XRAM_FI_D2, + },{ .name = "XRAM_FI_D3", .addr = A_XRAM_FI_D3, + },{ .name = "XRAM_FI_SY", .addr = A_XRAM_FI_SY, + .rsvd = 0xffff0000, + },{ .name = "XRAM_RMW_UE_FFA", .addr = A_XRAM_RMW_UE_FFA, + .rsvd = 0xfff00000, + .ro = 0xfffff, + },{ .name = "XRAM_FI_CNTR", .addr = A_XRAM_FI_CNTR, + .rsvd = 0xff000000, + },{ .name = "XRAM_IMP", .addr = A_XRAM_IMP, + .reset = 0x4, + .rsvd = 0xfffffff0, + .ro = 0xf, + },{ .name = "XRAM_PRDY_DBG", .addr = A_XRAM_PRDY_DBG, + .reset = 0xffff, + .rsvd = 0xffff0000, + .ro = 0xffff, + },{ .name = "XRAM_SAFETY_CHK", .addr = A_XRAM_SAFETY_CHK, + } +}; + +static void xram_ctrl_reset_enter(Object *obj, ResetType type) +{ + XlnxXramCtrl *s = XLNX_XRAM_CTRL(obj); + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(s->regs_info); ++i) { + register_reset(&s->regs_info[i]); + } + + ARRAY_FIELD_DP32(s->regs, XRAM_IMP, SIZE, s->cfg.encoded_size); +} + +static void xram_ctrl_reset_hold(Object *obj) +{ + XlnxXramCtrl *s = XLNX_XRAM_CTRL(obj); + + xram_update_irq(s); +} + +static const MemoryRegionOps xram_ctrl_ops = { + .read = register_read_memory, + .write = register_write_memory, + .endianness = DEVICE_LITTLE_ENDIAN, + .valid = { + .min_access_size = 4, + .max_access_size = 4, + }, +}; + +static void xram_ctrl_realize(DeviceState *dev, Error **errp) +{ + SysBusDevice *sbd = SYS_BUS_DEVICE(dev); + XlnxXramCtrl *s = XLNX_XRAM_CTRL(dev); + + switch (s->cfg.size) { + case 64 * KiB: + s->cfg.encoded_size = 0; + break; + case 128 * KiB: + s->cfg.encoded_size = 1; + break; + case 256 * KiB: + s->cfg.encoded_size = 2; + break; + case 512 * KiB: + s->cfg.encoded_size = 3; + break; + case 1 * MiB: + s->cfg.encoded_size = 4; + break; + default: + error_setg(errp, "Unsupported XRAM size %" PRId64, s->cfg.size); + return; + } + + memory_region_init_ram(&s->ram, OBJECT(s), + object_get_canonical_path_component(OBJECT(s)), + s->cfg.size, &error_fatal); + sysbus_init_mmio(sbd, &s->ram); +} + +static void xram_ctrl_init(Object *obj) +{ + XlnxXramCtrl *s = XLNX_XRAM_CTRL(obj); + SysBusDevice *sbd = SYS_BUS_DEVICE(obj); + + s->reg_array = + register_init_block32(DEVICE(obj), xram_ctrl_regs_info, + ARRAY_SIZE(xram_ctrl_regs_info), + s->regs_info, s->regs, + &xram_ctrl_ops, + XLNX_XRAM_CTRL_ERR_DEBUG, + XRAM_CTRL_R_MAX * 4); + sysbus_init_mmio(sbd, &s->reg_array->mem); + sysbus_init_irq(sbd, &s->irq); +} + +static void xram_ctrl_finalize(Object *obj) +{ + XlnxXramCtrl *s = XLNX_XRAM_CTRL(obj); + register_finalize_block(s->reg_array); +} + +static const VMStateDescription vmstate_xram_ctrl = { + .name = TYPE_XLNX_XRAM_CTRL, + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32_ARRAY(regs, XlnxXramCtrl, XRAM_CTRL_R_MAX), + VMSTATE_END_OF_LIST(), + } +}; + +static Property xram_ctrl_properties[] = { + DEFINE_PROP_UINT64("size", XlnxXramCtrl, cfg.size, 1 * MiB), + DEFINE_PROP_END_OF_LIST(), +}; + +static void xram_ctrl_class_init(ObjectClass *klass, void *data) +{ + ResettableClass *rc = RESETTABLE_CLASS(klass); + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->realize = xram_ctrl_realize; + dc->vmsd = &vmstate_xram_ctrl; + device_class_set_props(dc, xram_ctrl_properties); + + rc->phases.enter = xram_ctrl_reset_enter; + rc->phases.hold = xram_ctrl_reset_hold; +} + +static const TypeInfo xram_ctrl_info = { + .name = TYPE_XLNX_XRAM_CTRL, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_size = sizeof(XlnxXramCtrl), + .class_init = xram_ctrl_class_init, + .instance_init = xram_ctrl_init, + .instance_finalize = xram_ctrl_finalize, +}; + +static void xram_ctrl_register_types(void) +{ + type_register_static(&xram_ctrl_info); +} + +type_init(xram_ctrl_register_types) diff --git a/hw/misc/meson.build b/hw/misc/meson.build index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/meson.build +++ b/hw/misc/meson.build @@ -XXX,XX +XXX,XX @@ softmmu_ss.add(when: 'CONFIG_RASPI', if_true: files( )) softmmu_ss.add(when: 'CONFIG_SLAVIO', if_true: files('slavio_misc.c')) softmmu_ss.add(when: 'CONFIG_ZYNQ', if_true: files('zynq_slcr.c', 'zynq-xadc.c')) +softmmu_ss.add(when: 'CONFIG_XLNX_VERSAL', if_true: files('xlnx-versal-xramc.c')) softmmu_ss.add(when: 'CONFIG_STM32F2XX_SYSCFG', if_true: files('stm32f2xx_syscfg.c')) softmmu_ss.add(when: 'CONFIG_STM32F4XX_SYSCFG', if_true: files('stm32f4xx_syscfg.c')) softmmu_ss.add(when: 'CONFIG_STM32F4XX_EXTI', if_true: files('stm32f4xx_exti.c')) -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Connect the support for the Versal Accelerator RAMs (XRAMs). Reviewed-by: Luc Michel <luc@lmichel.fr> Acked-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20210308224637.2949533-3-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/xlnx-versal-virt.rst | 1 + include/hw/arm/xlnx-versal.h | 13 ++++++++++ hw/arm/xlnx-versal.c | 36 ++++++++++++++++++++++++++++ 3 files changed, 50 insertions(+) diff --git a/docs/system/arm/xlnx-versal-virt.rst b/docs/system/arm/xlnx-versal-virt.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/arm/xlnx-versal-virt.rst +++ b/docs/system/arm/xlnx-versal-virt.rst @@ -XXX,XX +XXX,XX @@ Implemented devices: - 8 ADMA (Xilinx zDMA) channels - 2 SD Controllers - OCM (256KB of On Chip Memory) +- XRAM (4MB of on chip Accelerator RAM) - DDR memory QEMU does not yet model any other devices, including the PL and the AI Engine. diff --git a/include/hw/arm/xlnx-versal.h b/include/hw/arm/xlnx-versal.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/xlnx-versal.h +++ b/include/hw/arm/xlnx-versal.h @@ -XXX,XX +XXX,XX @@ #include "hw/sysbus.h" #include "hw/arm/boot.h" +#include "hw/or-irq.h" #include "hw/sd/sdhci.h" #include "hw/intc/arm_gicv3.h" #include "hw/char/pl011.h" @@ -XXX,XX +XXX,XX @@ #include "hw/rtc/xlnx-zynqmp-rtc.h" #include "qom/object.h" #include "hw/usb/xlnx-usb-subsystem.h" +#include "hw/misc/xlnx-versal-xramc.h" #define TYPE_XLNX_VERSAL "xlnx-versal" OBJECT_DECLARE_SIMPLE_TYPE(Versal, XLNX_VERSAL) @@ -XXX,XX +XXX,XX @@ OBJECT_DECLARE_SIMPLE_TYPE(Versal, XLNX_VERSAL) #define XLNX_VERSAL_NR_GEMS 2 #define XLNX_VERSAL_NR_ADMAS 8 #define XLNX_VERSAL_NR_SDS 2 +#define XLNX_VERSAL_NR_XRAM 4 #define XLNX_VERSAL_NR_IRQS 192 struct Versal { @@ -XXX,XX +XXX,XX @@ struct Versal { XlnxZDMA adma[XLNX_VERSAL_NR_ADMAS]; VersalUsb2 usb; } iou; + + struct { + qemu_or_irq irq_orgate; + XlnxXramCtrl ctrl[XLNX_VERSAL_NR_XRAM]; + } xram; } lpd; /* The Platform Management Controller subsystem. */ @@ -XXX,XX +XXX,XX @@ struct Versal { #define VERSAL_GEM1_IRQ_0 58 #define VERSAL_GEM1_WAKE_IRQ_0 59 #define VERSAL_ADMA_IRQ_0 60 +#define VERSAL_XRAM_IRQ_0 79 #define VERSAL_RTC_APB_ERR_IRQ 121 #define VERSAL_SD0_IRQ_0 126 #define VERSAL_RTC_ALARM_IRQ 142 @@ -XXX,XX +XXX,XX @@ struct Versal { #define MM_OCM 0xfffc0000U #define MM_OCM_SIZE 0x40000 +#define MM_XRAM 0xfe800000 +#define MM_XRAMC 0xff8e0000 +#define MM_XRAMC_SIZE 0x10000 + #define MM_USB2_CTRL_REGS 0xFF9D0000 #define MM_USB2_CTRL_REGS_SIZE 0x10000 diff --git a/hw/arm/xlnx-versal.c b/hw/arm/xlnx-versal.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/xlnx-versal.c +++ b/hw/arm/xlnx-versal.c @@ -XXX,XX +XXX,XX @@ */ #include "qemu/osdep.h" +#include "qemu/units.h" #include "qapi/error.h" #include "qemu/log.h" #include "qemu/module.h" @@ -XXX,XX +XXX,XX @@ static void versal_create_rtc(Versal *s, qemu_irq *pic) sysbus_connect_irq(sbd, 1, pic[VERSAL_RTC_APB_ERR_IRQ]); } +static void versal_create_xrams(Versal *s, qemu_irq *pic) +{ + int nr_xrams = ARRAY_SIZE(s->lpd.xram.ctrl); + DeviceState *orgate; + int i; + + /* XRAM IRQs get ORed into a single line. */ + object_initialize_child(OBJECT(s), "xram-irq-orgate", + &s->lpd.xram.irq_orgate, TYPE_OR_IRQ); + orgate = DEVICE(&s->lpd.xram.irq_orgate); + object_property_set_int(OBJECT(orgate), + "num-lines", nr_xrams, &error_fatal); + qdev_realize(orgate, NULL, &error_fatal); + qdev_connect_gpio_out(orgate, 0, pic[VERSAL_XRAM_IRQ_0]); + + for (i = 0; i < ARRAY_SIZE(s->lpd.xram.ctrl); i++) { + SysBusDevice *sbd; + MemoryRegion *mr; + + object_initialize_child(OBJECT(s), "xram[*]", &s->lpd.xram.ctrl[i], + TYPE_XLNX_XRAM_CTRL); + sbd = SYS_BUS_DEVICE(&s->lpd.xram.ctrl[i]); + sysbus_realize(sbd, &error_fatal); + + mr = sysbus_mmio_get_region(sbd, 0); + memory_region_add_subregion(&s->mr_ps, + MM_XRAMC + i * MM_XRAMC_SIZE, mr); + mr = sysbus_mmio_get_region(sbd, 1); + memory_region_add_subregion(&s->mr_ps, MM_XRAM + i * MiB, mr); + + sysbus_connect_irq(sbd, 0, qdev_get_gpio_in(orgate, i)); + } +} + /* This takes the board allocated linear DDR memory and creates aliases * for each split DDR range/aperture on the Versal address map. */ @@ -XXX,XX +XXX,XX @@ static void versal_realize(DeviceState *dev, Error **errp) versal_create_admas(s, pic); versal_create_sds(s, pic); versal_create_rtc(s, pic); + versal_create_xrams(s, pic); versal_map_ddr(s); versal_unimp(s); -- 2.20.1 From: Eric Auger <eric.auger@redhat.com> With -Werror=maybe-uninitialized configuration we get ../hw/i386/intel_iommu.c: In function ‘vtd_context_device_invalidate’: ../hw/i386/intel_iommu.c:1888:10: error: ‘mask’ may be used uninitialized in this function [-Werror=maybe-uninitialized] 1888 | mask = ~mask; | ~~~~~^~~~~~~ Add a g_assert_not_reached() to avoid the error. Signed-off-by: Eric Auger <eric.auger@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20210309102742.30442-2-eric.auger@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/i386/intel_iommu.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/hw/i386/intel_iommu.c b/hw/i386/intel_iommu.c index XXXXXXX..XXXXXXX 100644 --- a/hw/i386/intel_iommu.c +++ b/hw/i386/intel_iommu.c @@ -XXX,XX +XXX,XX @@ static void vtd_context_device_invalidate(IntelIOMMUState *s, case 3: mask = 7; /* Mask bit 2:0 in the SID field */ break; + default: + g_assert_not_reached(); } mask = ~mask; -- 2.20.1 From: Eric Auger <eric.auger@redhat.com> Currently get_naturally_aligned_size() is used by the intel iommu to compute the maximum invalidation range based on @size which is a power of 2 while being aligned with the @start address and less than the maximum range defined by @gaw. This helper is also useful for other iommu devices (virtio-iommu, SMMUv3) to make sure IOMMU UNMAP notifiers only are called with power of 2 range sizes. Let's move this latter into dma-helpers.c and rename it into dma_aligned_pow2_mask(). Also rewrite the helper so that it accomodates UINT64_MAX values for the size mask and max mask. It now returns a mask instead of a size. Change the caller. Signed-off-by: Eric Auger <eric.auger@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Message-id: 20210309102742.30442-3-eric.auger@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/sysemu/dma.h | 12 ++++++++++++ hw/i386/intel_iommu.c | 30 +++++++----------------------- softmmu/dma-helpers.c | 26 ++++++++++++++++++++++++++ 3 files changed, 45 insertions(+), 23 deletions(-) diff --git a/include/sysemu/dma.h b/include/sysemu/dma.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/dma.h +++ b/include/sysemu/dma.h @@ -XXX,XX +XXX,XX @@ uint64_t dma_buf_write(uint8_t *ptr, int32_t len, QEMUSGList *sg); void dma_acct_start(BlockBackend *blk, BlockAcctCookie *cookie, QEMUSGList *sg, enum BlockAcctType type); +/** + * dma_aligned_pow2_mask: Return the address bit mask of the largest + * power of 2 size less or equal than @end - @start + 1, aligned with @start, + * and bounded by 1 << @max_addr_bits bits. + * + * @start: range start address + * @end: range end address (greater than @start) + * @max_addr_bits: max address bits (<= 64) + */ +uint64_t dma_aligned_pow2_mask(uint64_t start, uint64_t end, + int max_addr_bits); + #endif diff --git a/hw/i386/intel_iommu.c b/hw/i386/intel_iommu.c index XXXXXXX..XXXXXXX 100644 --- a/hw/i386/intel_iommu.c +++ b/hw/i386/intel_iommu.c @@ -XXX,XX +XXX,XX @@ #include "hw/i386/x86-iommu.h" #include "hw/pci-host/q35.h" #include "sysemu/kvm.h" +#include "sysemu/dma.h" #include "sysemu/sysemu.h" #include "hw/i386/apic_internal.h" #include "kvm/kvm_i386.h" @@ -XXX,XX +XXX,XX @@ VTDAddressSpace *vtd_find_add_as(IntelIOMMUState *s, PCIBus *bus, int devfn) return vtd_dev_as; } -static uint64_t get_naturally_aligned_size(uint64_t start, - uint64_t size, int gaw) -{ - uint64_t max_mask = 1ULL << gaw; - uint64_t alignment = start ? start & -start : max_mask; - - alignment = MIN(alignment, max_mask); - size = MIN(size, max_mask); - - if (alignment <= size) { - /* Increase the alignment of start */ - return alignment; - } else { - /* Find the largest page mask from size */ - return 1ULL << (63 - clz64(size)); - } -} - /* Unmap the whole range in the notifier's scope. */ static void vtd_address_space_unmap(VTDAddressSpace *as, IOMMUNotifier *n) { @@ -XXX,XX +XXX,XX @@ static void vtd_address_space_unmap(VTDAddressSpace *as, IOMMUNotifier *n) while (remain >= VTD_PAGE_SIZE) { IOMMUTLBEvent event; - uint64_t mask = get_naturally_aligned_size(start, remain, s->aw_bits); + uint64_t mask = dma_aligned_pow2_mask(start, end, s->aw_bits); + uint64_t size = mask + 1; - assert(mask); + assert(size); event.type = IOMMU_NOTIFIER_UNMAP; event.entry.iova = start; - event.entry.addr_mask = mask - 1; + event.entry.addr_mask = mask; event.entry.target_as = &address_space_memory; event.entry.perm = IOMMU_NONE; /* This field is meaningless for unmap */ @@ -XXX,XX +XXX,XX @@ static void vtd_address_space_unmap(VTDAddressSpace *as, IOMMUNotifier *n) memory_region_notify_iommu_one(n, &event); - start += mask; - remain -= mask; + start += size; + remain -= size; } assert(!remain); diff --git a/softmmu/dma-helpers.c b/softmmu/dma-helpers.c index XXXXXXX..XXXXXXX 100644 --- a/softmmu/dma-helpers.c +++ b/softmmu/dma-helpers.c @@ -XXX,XX +XXX,XX @@ void dma_acct_start(BlockBackend *blk, BlockAcctCookie *cookie, { block_acct_start(blk_get_stats(blk), cookie, sg->size, type); } + +uint64_t dma_aligned_pow2_mask(uint64_t start, uint64_t end, int max_addr_bits) +{ + uint64_t max_mask = UINT64_MAX, addr_mask = end - start; + uint64_t alignment_mask, size_mask; + + if (max_addr_bits != 64) { + max_mask = (1ULL << max_addr_bits) - 1; + } + + alignment_mask = start ? (start & -start) - 1 : max_mask; + alignment_mask = MIN(alignment_mask, max_mask); + size_mask = MIN(addr_mask, max_mask); + + if (alignment_mask <= size_mask) { + /* Increase the alignment of start */ + return alignment_mask; + } else { + /* Find the largest page mask from size */ + if (addr_mask == UINT64_MAX) { + return UINT64_MAX; + } + return (1ULL << (63 - clz64(addr_mask + 1))) - 1; + } +} + -- 2.20.1 From: Eric Auger <eric.auger@redhat.com> Unmap notifiers work with an address mask assuming an invalidation range of a power of 2. Nothing mandates this in the VIRTIO-IOMMU spec. So in case the range is not a power of 2, split it into several invalidations. Signed-off-by: Eric Auger <eric.auger@redhat.com> Reviewed-by: Peter Xu <peterx@redhat.com> Message-id: 20210309102742.30442-4-eric.auger@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/virtio/virtio-iommu.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) diff --git a/hw/virtio/virtio-iommu.c b/hw/virtio/virtio-iommu.c index XXXXXXX..XXXXXXX 100644 --- a/hw/virtio/virtio-iommu.c +++ b/hw/virtio/virtio-iommu.c @@ -XXX,XX +XXX,XX @@ static void virtio_iommu_notify_unmap(IOMMUMemoryRegion *mr, hwaddr virt_start, hwaddr virt_end) { IOMMUTLBEvent event; + uint64_t delta = virt_end - virt_start; if (!(mr->iommu_notify_flags & IOMMU_NOTIFIER_UNMAP)) { return; @@ -XXX,XX +XXX,XX @@ static void virtio_iommu_notify_unmap(IOMMUMemoryRegion *mr, hwaddr virt_start, event.type = IOMMU_NOTIFIER_UNMAP; event.entry.target_as = &address_space_memory; - event.entry.addr_mask = virt_end - virt_start; - event.entry.iova = virt_start; event.entry.perm = IOMMU_NONE; event.entry.translated_addr = 0; + event.entry.addr_mask = delta; + event.entry.iova = virt_start; - memory_region_notify_iommu(mr, 0, event); + if (delta == UINT64_MAX) { + memory_region_notify_iommu(mr, 0, event); + } + + + while (virt_start != virt_end + 1) { + uint64_t mask = dma_aligned_pow2_mask(virt_start, virt_end, 64); + + event.entry.addr_mask = mask; + event.entry.iova = virt_start; + memory_region_notify_iommu(mr, 0, event); + virt_start += mask + 1; + } } static gboolean virtio_iommu_notify_unmap_cb(gpointer key, gpointer value, -- 2.20.1 From: Eric Auger <eric.auger@redhat.com> If the asid is not set, do not attempt to locate the key directly as all inserted keys have a valid asid. Use g_hash_table_foreach_remove instead. Signed-off-by: Eric Auger <eric.auger@redhat.com> Message-id: 20210309102742.30442-5-eric.auger@redhat.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/smmu-common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/hw/arm/smmu-common.c b/hw/arm/smmu-common.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/smmu-common.c +++ b/hw/arm/smmu-common.c @@ -XXX,XX +XXX,XX @@ inline void smmu_iotlb_inv_iova(SMMUState *s, int asid, dma_addr_t iova, uint8_t tg, uint64_t num_pages, uint8_t ttl) { - if (ttl && (num_pages == 1)) { + if (ttl && (num_pages == 1) && (asid >= 0)) { SMMUIOTLBKey key = smmu_get_iotlb_key(asid, iova, tg, ttl); g_hash_table_remove(s->iotlb, &key); -- 2.20.1 From: Eric Auger <eric.auger@redhat.com> As of today, the driver can invalidate a number of pages that is not a power of 2. However IOTLB unmap notifications and internal IOTLB invalidations work with masks leading to erroneous invalidations. In case the range is not a power of 2, split invalidations into power of 2 invalidations. When looking for a single page entry in the vSMMU internal IOTLB, let's make sure that if the entry is not found using a g_hash_table_remove() we iterate over all the entries to find a potential range that overlaps it. Signed-off-by: Eric Auger <eric.auger@redhat.com> Message-id: 20210309102742.30442-6-eric.auger@redhat.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/smmu-common.c | 30 ++++++++++++++++++------------ hw/arm/smmuv3.c | 24 ++++++++++++++++++++---- 2 files changed, 38 insertions(+), 16 deletions(-) diff --git a/hw/arm/smmu-common.c b/hw/arm/smmu-common.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/smmu-common.c +++ b/hw/arm/smmu-common.c @@ -XXX,XX +XXX,XX @@ inline void smmu_iotlb_inv_iova(SMMUState *s, int asid, dma_addr_t iova, uint8_t tg, uint64_t num_pages, uint8_t ttl) { + /* if tg is not set we use 4KB range invalidation */ + uint8_t granule = tg ? tg * 2 + 10 : 12; + if (ttl && (num_pages == 1) && (asid >= 0)) { SMMUIOTLBKey key = smmu_get_iotlb_key(asid, iova, tg, ttl); - g_hash_table_remove(s->iotlb, &key); - } else { - /* if tg is not set we use 4KB range invalidation */ - uint8_t granule = tg ? tg * 2 + 10 : 12; - - SMMUIOTLBPageInvInfo info = { - .asid = asid, .iova = iova, - .mask = (num_pages * 1 << granule) - 1}; - - g_hash_table_foreach_remove(s->iotlb, - smmu_hash_remove_by_asid_iova, - &info); + if (g_hash_table_remove(s->iotlb, &key)) { + return; + } + /* + * if the entry is not found, let's see if it does not + * belong to a larger IOTLB entry + */ } + + SMMUIOTLBPageInvInfo info = { + .asid = asid, .iova = iova, + .mask = (num_pages * 1 << granule) - 1}; + + g_hash_table_foreach_remove(s->iotlb, + smmu_hash_remove_by_asid_iova, + &info); } inline void smmu_iotlb_inv_asid(SMMUState *s, uint16_t asid) diff --git a/hw/arm/smmuv3.c b/hw/arm/smmuv3.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/smmuv3.c +++ b/hw/arm/smmuv3.c @@ -XXX,XX +XXX,XX @@ static void smmuv3_s1_range_inval(SMMUState *s, Cmd *cmd) uint16_t vmid = CMD_VMID(cmd); bool leaf = CMD_LEAF(cmd); uint8_t tg = CMD_TG(cmd); - hwaddr num_pages = 1; + uint64_t first_page = 0, last_page; + uint64_t num_pages = 1; int asid = -1; if (tg) { @@ -XXX,XX +XXX,XX @@ static void smmuv3_s1_range_inval(SMMUState *s, Cmd *cmd) if (type == SMMU_CMD_TLBI_NH_VA) { asid = CMD_ASID(cmd); } - trace_smmuv3_s1_range_inval(vmid, asid, addr, tg, num_pages, ttl, leaf); - smmuv3_inv_notifiers_iova(s, asid, addr, tg, num_pages); - smmu_iotlb_inv_iova(s, asid, addr, tg, num_pages, ttl); + + /* Split invalidations into ^2 range invalidations */ + last_page = num_pages - 1; + while (num_pages) { + uint8_t granule = tg * 2 + 10; + uint64_t mask, count; + + mask = dma_aligned_pow2_mask(first_page, last_page, 64 - granule); + count = mask + 1; + + trace_smmuv3_s1_range_inval(vmid, asid, addr, tg, count, ttl, leaf); + smmuv3_inv_notifiers_iova(s, asid, addr, tg, count); + smmu_iotlb_inv_iova(s, asid, addr, tg, count, ttl); + + num_pages -= count; + first_page += count; + addr += count * BIT_ULL(granule); + } } static int smmuv3_cmdq_consume(SMMUv3State *s) -- 2.20.1 From: Eric Auger <eric.auger@redhat.com> If the whole SID range (32b) is invalidated (SMMU_CMD_CFGI_ALL), @end overflows and we fail to handle the command properly. Once this gets fixed, the current code really is awkward in the sense it loops over the whole range instead of removing the currently cached configs through a hash table lookup. Fix both the overflow and the lookup. Signed-off-by: Eric Auger <eric.auger@redhat.com> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 20210309102742.30442-7-eric.auger@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/smmu-internal.h | 5 +++++ hw/arm/smmuv3.c | 34 ++++++++++++++++++++-------------- 2 files changed, 25 insertions(+), 14 deletions(-) diff --git a/hw/arm/smmu-internal.h b/hw/arm/smmu-internal.h index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/smmu-internal.h +++ b/hw/arm/smmu-internal.h @@ -XXX,XX +XXX,XX @@ typedef struct SMMUIOTLBPageInvInfo { uint64_t mask; } SMMUIOTLBPageInvInfo; +typedef struct SMMUSIDRange { + uint32_t start; + uint32_t end; +} SMMUSIDRange; + #endif diff --git a/hw/arm/smmuv3.c b/hw/arm/smmuv3.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/smmuv3.c +++ b/hw/arm/smmuv3.c @@ -XXX,XX +XXX,XX @@ #include "hw/arm/smmuv3.h" #include "smmuv3-internal.h" +#include "smmu-internal.h" /** * smmuv3_trigger_irq - pulse @irq if enabled and update @@ -XXX,XX +XXX,XX @@ static void smmuv3_s1_range_inval(SMMUState *s, Cmd *cmd) } } +static gboolean +smmuv3_invalidate_ste(gpointer key, gpointer value, gpointer user_data) +{ + SMMUDevice *sdev = (SMMUDevice *)key; + uint32_t sid = smmu_get_sid(sdev); + SMMUSIDRange *sid_range = (SMMUSIDRange *)user_data; + + if (sid < sid_range->start || sid > sid_range->end) { + return false; + } + trace_smmuv3_config_cache_inv(sid); + return true; +} + static int smmuv3_cmdq_consume(SMMUv3State *s) { SMMUState *bs = ARM_SMMU(s); @@ -XXX,XX +XXX,XX @@ static int smmuv3_cmdq_consume(SMMUv3State *s) } case SMMU_CMD_CFGI_STE_RANGE: /* same as SMMU_CMD_CFGI_ALL */ { - uint32_t start = CMD_SID(&cmd), end, i; + uint32_t start = CMD_SID(&cmd); uint8_t range = CMD_STE_RANGE(&cmd); + uint64_t end = start + (1ULL << (range + 1)) - 1; + SMMUSIDRange sid_range = {start, end}; if (CMD_SSEC(&cmd)) { cmd_error = SMMU_CERROR_ILL; break; } - - end = start + (1 << (range + 1)) - 1; trace_smmuv3_cmdq_cfgi_ste_range(start, end); - - for (i = start; i <= end; i++) { - IOMMUMemoryRegion *mr = smmu_iommu_mr(bs, i); - SMMUDevice *sdev; - - if (!mr) { - continue; - } - sdev = container_of(mr, SMMUDevice, iommu); - smmuv3_flush_config(sdev); - } + g_hash_table_foreach_remove(bs->configs, smmuv3_invalidate_ste, + &sid_range); break; } case SMMU_CMD_CFGI_CD: -- 2.20.1 From: Eric Auger <eric.auger@redhat.com> Convert all sid printouts to sid=0x%x. Signed-off-by: Eric Auger <eric.auger@redhat.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20210309102742.30442-8-eric.auger@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/trace-events | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/hw/arm/trace-events b/hw/arm/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/trace-events +++ b/hw/arm/trace-events @@ -XXX,XX +XXX,XX @@ smmuv3_cmdq_opcode(const char *opcode) "<--- %s" smmuv3_cmdq_consume_out(uint32_t prod, uint32_t cons, uint8_t prod_wrap, uint8_t cons_wrap) "prod:%d, cons:%d, prod_wrap:%d, cons_wrap:%d " smmuv3_cmdq_consume_error(const char *cmd_name, uint8_t cmd_error) "Error on %s command execution: %d" smmuv3_write_mmio(uint64_t addr, uint64_t val, unsigned size, uint32_t r) "addr: 0x%"PRIx64" val:0x%"PRIx64" size: 0x%x(%d)" -smmuv3_record_event(const char *type, uint32_t sid) "%s sid=%d" -smmuv3_find_ste(uint16_t sid, uint32_t features, uint16_t sid_split) "SID:0x%x features:0x%x, sid_split:0x%x" +smmuv3_record_event(const char *type, uint32_t sid) "%s sid=0x%x" +smmuv3_find_ste(uint16_t sid, uint32_t features, uint16_t sid_split) "sid=0x%x features:0x%x, sid_split:0x%x" smmuv3_find_ste_2lvl(uint64_t strtab_base, uint64_t l1ptr, int l1_ste_offset, uint64_t l2ptr, int l2_ste_offset, int max_l2_ste) "strtab_base:0x%"PRIx64" l1ptr:0x%"PRIx64" l1_off:0x%x, l2ptr:0x%"PRIx64" l2_off:0x%x max_l2_ste:%d" smmuv3_get_ste(uint64_t addr) "STE addr: 0x%"PRIx64 -smmuv3_translate_disable(const char *n, uint16_t sid, uint64_t addr, bool is_write) "%s sid=%d bypass (smmu disabled) iova:0x%"PRIx64" is_write=%d" -smmuv3_translate_bypass(const char *n, uint16_t sid, uint64_t addr, bool is_write) "%s sid=%d STE bypass iova:0x%"PRIx64" is_write=%d" -smmuv3_translate_abort(const char *n, uint16_t sid, uint64_t addr, bool is_write) "%s sid=%d abort on iova:0x%"PRIx64" is_write=%d" -smmuv3_translate_success(const char *n, uint16_t sid, uint64_t iova, uint64_t translated, int perm) "%s sid=%d iova=0x%"PRIx64" translated=0x%"PRIx64" perm=0x%x" +smmuv3_translate_disable(const char *n, uint16_t sid, uint64_t addr, bool is_write) "%s sid=0x%x bypass (smmu disabled) iova:0x%"PRIx64" is_write=%d" +smmuv3_translate_bypass(const char *n, uint16_t sid, uint64_t addr, bool is_write) "%s sid=0x%x STE bypass iova:0x%"PRIx64" is_write=%d" +smmuv3_translate_abort(const char *n, uint16_t sid, uint64_t addr, bool is_write) "%s sid=0x%x abort on iova:0x%"PRIx64" is_write=%d" +smmuv3_translate_success(const char *n, uint16_t sid, uint64_t iova, uint64_t translated, int perm) "%s sid=0x%x iova=0x%"PRIx64" translated=0x%"PRIx64" perm=0x%x" smmuv3_get_cd(uint64_t addr) "CD addr: 0x%"PRIx64 smmuv3_decode_cd(uint32_t oas) "oas=%d" smmuv3_decode_cd_tt(int i, uint32_t tsz, uint64_t ttb, uint32_t granule_sz, bool had) "TT[%d]:tsz:%d ttb:0x%"PRIx64" granule_sz:%d had:%d" -smmuv3_cmdq_cfgi_ste(int streamid) "streamid =%d" +smmuv3_cmdq_cfgi_ste(int streamid) "streamid= 0x%x" smmuv3_cmdq_cfgi_ste_range(int start, int end) "start=0x%x - end=0x%x" -smmuv3_cmdq_cfgi_cd(uint32_t sid) "streamid = %d" -smmuv3_config_cache_hit(uint32_t sid, uint32_t hits, uint32_t misses, uint32_t perc) "Config cache HIT for sid %d (hits=%d, misses=%d, hit rate=%d)" -smmuv3_config_cache_miss(uint32_t sid, uint32_t hits, uint32_t misses, uint32_t perc) "Config cache MISS for sid %d (hits=%d, misses=%d, hit rate=%d)" -smmuv3_s1_range_inval(int vmid, int asid, uint64_t addr, uint8_t tg, uint64_t num_pages, uint8_t ttl, bool leaf) "vmid =%d asid =%d addr=0x%"PRIx64" tg=%d num_pages=0x%"PRIx64" ttl=%d leaf=%d" +smmuv3_cmdq_cfgi_cd(uint32_t sid) "sid=0x%x" +smmuv3_config_cache_hit(uint32_t sid, uint32_t hits, uint32_t misses, uint32_t perc) "Config cache HIT for sid=0x%x (hits=%d, misses=%d, hit rate=%d)" +smmuv3_config_cache_miss(uint32_t sid, uint32_t hits, uint32_t misses, uint32_t perc) "Config cache MISS for sid=0x%x (hits=%d, misses=%d, hit rate=%d)" +smmuv3_s1_range_inval(int vmid, int asid, uint64_t addr, uint8_t tg, uint64_t num_pages, uint8_t ttl, bool leaf) "vmid=%d asid=%d addr=0x%"PRIx64" tg=%d num_pages=0x%"PRIx64" ttl=%d leaf=%d" smmuv3_cmdq_tlbi_nh(void) "" smmuv3_cmdq_tlbi_nh_asid(uint16_t asid) "asid=%d" -smmuv3_config_cache_inv(uint32_t sid) "Config cache INV for sid %d" +smmuv3_config_cache_inv(uint32_t sid) "Config cache INV for sid=0x%x" smmuv3_notify_flag_add(const char *iommu) "ADD SMMUNotifier node for iommu mr=%s" smmuv3_notify_flag_del(const char *iommu) "DEL SMMUNotifier node for iommu mr=%s" smmuv3_inv_notifiers_iova(const char *name, uint16_t asid, uint64_t iova, uint8_t tg, uint64_t num_pages) "iommu mr=%s asid=%d iova=0x%"PRIx64" tg=%d num_pages=0x%"PRIx64 -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Missed out on compressing the second half of a predicate with length vl % 512 > 256. Adjust all of the x + (y << s) to x | (y << s) as a general style fix. Drop the extract64 because the input uint64_t are known to be already zero-extended from the current size of the predicate. Reported-by: Laurent Desnogues <laurent.desnogues@gmail.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210309155305.11301-2-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/sve_helper.c | 30 +++++++++++++++++++++--------- 1 file changed, 21 insertions(+), 9 deletions(-) diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve_helper.c +++ b/target/arm/sve_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(sve_uzp_p)(void *vd, void *vn, void *vm, uint32_t pred_desc) if (oprsz <= 8) { l = compress_bits(n[0] >> odd, esz); h = compress_bits(m[0] >> odd, esz); - d[0] = extract64(l + (h << (4 * oprsz)), 0, 8 * oprsz); + d[0] = l | (h << (4 * oprsz)); } else { ARMPredicateReg tmp_m; intptr_t oprsz_16 = oprsz / 16; @@ -XXX,XX +XXX,XX @@ void HELPER(sve_uzp_p)(void *vd, void *vn, void *vm, uint32_t pred_desc) h = n[2 * i + 1]; l = compress_bits(l >> odd, esz); h = compress_bits(h >> odd, esz); - d[i] = l + (h << 32); + d[i] = l | (h << 32); } - /* For VL which is not a power of 2, the results from M do not - align nicely with the uint64_t for D. Put the aligned results - from M into TMP_M and then copy it into place afterward. */ + /* + * For VL which is not a multiple of 512, the results from M do not + * align nicely with the uint64_t for D. Put the aligned results + * from M into TMP_M and then copy it into place afterward. + */ if (oprsz & 15) { - d[i] = compress_bits(n[2 * i] >> odd, esz); + int final_shift = (oprsz & 15) * 2; + + l = n[2 * i + 0]; + h = n[2 * i + 1]; + l = compress_bits(l >> odd, esz); + h = compress_bits(h >> odd, esz); + d[i] = l | (h << final_shift); for (i = 0; i < oprsz_16; i++) { l = m[2 * i + 0]; h = m[2 * i + 1]; l = compress_bits(l >> odd, esz); h = compress_bits(h >> odd, esz); - tmp_m.p[i] = l + (h << 32); + tmp_m.p[i] = l | (h << 32); } - tmp_m.p[i] = compress_bits(m[2 * i] >> odd, esz); + l = m[2 * i + 0]; + h = m[2 * i + 1]; + l = compress_bits(l >> odd, esz); + h = compress_bits(h >> odd, esz); + tmp_m.p[i] = l | (h << final_shift); swap_memmove(vd + oprsz / 2, &tmp_m, oprsz / 2); } else { @@ -XXX,XX +XXX,XX @@ void HELPER(sve_uzp_p)(void *vd, void *vn, void *vm, uint32_t pred_desc) h = m[2 * i + 1]; l = compress_bits(l >> odd, esz); h = compress_bits(h >> odd, esz); - d[oprsz_16 + i] = l + (h << 32); + d[oprsz_16 + i] = l | (h << 32); } } } -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Wrote too much with low-half zip (zip1) with vl % 512 != 0. Adjust all of the x + (y << s) to x | (y << s) as a style fix. We only ever have exact overlap between D, M, and N. Therefore we only need a single temporary, and we do not need to check for partial overlap. Reported-by: Laurent Desnogues <laurent.desnogues@gmail.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210309155305.11301-3-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/sve_helper.c | 25 ++++++++++++++----------- 1 file changed, 14 insertions(+), 11 deletions(-) diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve_helper.c +++ b/target/arm/sve_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(sve_zip_p)(void *vd, void *vn, void *vm, uint32_t pred_desc) intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); int esz = FIELD_EX32(pred_desc, PREDDESC, ESZ); intptr_t high = FIELD_EX32(pred_desc, PREDDESC, DATA); + int esize = 1 << esz; uint64_t *d = vd; intptr_t i; @@ -XXX,XX +XXX,XX @@ void HELPER(sve_zip_p)(void *vd, void *vn, void *vm, uint32_t pred_desc) mm = extract64(mm, high * half, half); nn = expand_bits(nn, esz); mm = expand_bits(mm, esz); - d[0] = nn + (mm << (1 << esz)); + d[0] = nn | (mm << esize); } else { - ARMPredicateReg tmp_n, tmp_m; + ARMPredicateReg tmp; /* We produce output faster than we consume input. Therefore we must be mindful of possible overlap. */ - if ((vn - vd) < (uintptr_t)oprsz) { - vn = memcpy(&tmp_n, vn, oprsz); - } - if ((vm - vd) < (uintptr_t)oprsz) { - vm = memcpy(&tmp_m, vm, oprsz); + if (vd == vn) { + vn = memcpy(&tmp, vn, oprsz); + if (vd == vm) { + vm = vn; + } + } else if (vd == vm) { + vm = memcpy(&tmp, vm, oprsz); } if (high) { high = oprsz >> 1; } - if ((high & 3) == 0) { + if ((oprsz & 7) == 0) { uint32_t *n = vn, *m = vm; high >>= 2; - for (i = 0; i < DIV_ROUND_UP(oprsz, 8); i++) { + for (i = 0; i < oprsz / 8; i++) { uint64_t nn = n[H4(high + i)]; uint64_t mm = m[H4(high + i)]; nn = expand_bits(nn, esz); mm = expand_bits(mm, esz); - d[i] = nn + (mm << (1 << esz)); + d[i] = nn | (mm << esize); } } else { uint8_t *n = vn, *m = vm; @@ -XXX,XX +XXX,XX @@ void HELPER(sve_zip_p)(void *vd, void *vn, void *vm, uint32_t pred_desc) nn = expand_bits(nn, esz); mm = expand_bits(mm, esz); - d16[H2(i)] = nn + (mm << (1 << esz)); + d16[H2(i)] = nn | (mm << esize); } } } -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Wrote too much with punpk1 with vl % 512 != 0. Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Reported-by: Laurent Desnogues <laurent.desnogues@gmail.com> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210309155305.11301-4-richard.henderson@linaro.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/sve_helper.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve_helper.c +++ b/target/arm/sve_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(sve_punpk_p)(void *vd, void *vn, uint32_t pred_desc) high = oprsz >> 1; } - if ((high & 3) == 0) { + if ((oprsz & 7) == 0) { uint32_t *n = vn; high >>= 2; - for (i = 0; i < DIV_ROUND_UP(oprsz, 8); i++) { + for (i = 0; i < oprsz / 8; i++) { uint64_t nn = n[H4(high + i)]; d[i] = expand_bits(nn, 0); } -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Since b64ee454a4a0, all predicate operations should be using these field macros for predicates. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210309155305.11301-5-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/sve_helper.c | 6 +++--- target/arm/translate-sve.c | 7 +++---- 2 files changed, 6 insertions(+), 7 deletions(-) diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve_helper.c +++ b/target/arm/sve_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(sve_compact_d)(void *vd, void *vn, void *vg, uint32_t desc) */ int32_t HELPER(sve_last_active_element)(void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; - intptr_t esz = extract32(pred_desc, SIMD_DATA_SHIFT, 2); + intptr_t words = DIV_ROUND_UP(FIELD_EX32(pred_desc, PREDDESC, OPRSZ), 8); + intptr_t esz = FIELD_EX32(pred_desc, PREDDESC, ESZ); - return last_active_element(vg, DIV_ROUND_UP(oprsz, 8), esz); + return last_active_element(vg, words, esz); } void HELPER(sve_splice)(void *vd, void *vn, void *vm, void *vg, uint32_t desc) diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-sve.c +++ b/target/arm/translate-sve.c @@ -XXX,XX +XXX,XX @@ static void find_last_active(DisasContext *s, TCGv_i32 ret, int esz, int pg) */ TCGv_ptr t_p = tcg_temp_new_ptr(); TCGv_i32 t_desc; - unsigned vsz = pred_full_reg_size(s); - unsigned desc; + unsigned desc = 0; - desc = vsz - 2; - desc = deposit32(desc, SIMD_DATA_SHIFT, 2, esz); + desc = FIELD_DP32(desc, PREDDESC, OPRSZ, pred_full_reg_size(s)); + desc = FIELD_DP32(desc, PREDDESC, ESZ, esz); tcg_gen_addi_ptr(t_p, cpu_env, pred_full_reg_offset(s, pg)); t_desc = tcg_const_i32(desc); -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Since b64ee454a4a0, all predicate operations should be using these field macros for predicates. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210309155305.11301-6-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/sve_helper.c | 30 ++++++++++++++---------------- target/arm/translate-sve.c | 4 ++-- 2 files changed, 16 insertions(+), 18 deletions(-) diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve_helper.c +++ b/target/arm/sve_helper.c @@ -XXX,XX +XXX,XX @@ static uint32_t do_zero(ARMPredicateReg *d, intptr_t oprsz) void HELPER(sve_brkpa)(void *vd, void *vn, void *vm, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); if (last_active_pred(vn, vg, oprsz)) { compute_brk_z(vd, vm, vg, oprsz, true); } else { @@ -XXX,XX +XXX,XX @@ void HELPER(sve_brkpa)(void *vd, void *vn, void *vm, void *vg, uint32_t HELPER(sve_brkpas)(void *vd, void *vn, void *vm, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); if (last_active_pred(vn, vg, oprsz)) { return compute_brks_z(vd, vm, vg, oprsz, true); } else { @@ -XXX,XX +XXX,XX @@ uint32_t HELPER(sve_brkpas)(void *vd, void *vn, void *vm, void *vg, void HELPER(sve_brkpb)(void *vd, void *vn, void *vm, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); if (last_active_pred(vn, vg, oprsz)) { compute_brk_z(vd, vm, vg, oprsz, false); } else { @@ -XXX,XX +XXX,XX @@ void HELPER(sve_brkpb)(void *vd, void *vn, void *vm, void *vg, uint32_t HELPER(sve_brkpbs)(void *vd, void *vn, void *vm, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); if (last_active_pred(vn, vg, oprsz)) { return compute_brks_z(vd, vm, vg, oprsz, false); } else { @@ -XXX,XX +XXX,XX @@ uint32_t HELPER(sve_brkpbs)(void *vd, void *vn, void *vm, void *vg, void HELPER(sve_brka_z)(void *vd, void *vn, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); compute_brk_z(vd, vn, vg, oprsz, true); } uint32_t HELPER(sve_brkas_z)(void *vd, void *vn, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); return compute_brks_z(vd, vn, vg, oprsz, true); } void HELPER(sve_brkb_z)(void *vd, void *vn, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); compute_brk_z(vd, vn, vg, oprsz, false); } uint32_t HELPER(sve_brkbs_z)(void *vd, void *vn, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); return compute_brks_z(vd, vn, vg, oprsz, false); } void HELPER(sve_brka_m)(void *vd, void *vn, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); compute_brk_m(vd, vn, vg, oprsz, true); } uint32_t HELPER(sve_brkas_m)(void *vd, void *vn, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); return compute_brks_m(vd, vn, vg, oprsz, true); } void HELPER(sve_brkb_m)(void *vd, void *vn, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); compute_brk_m(vd, vn, vg, oprsz, false); } uint32_t HELPER(sve_brkbs_m)(void *vd, void *vn, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); return compute_brks_m(vd, vn, vg, oprsz, false); } void HELPER(sve_brkn)(void *vd, void *vn, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; - + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); if (!last_active_pred(vn, vg, oprsz)) { do_zero(vd, oprsz); } @@ -XXX,XX +XXX,XX @@ static uint32_t predtest_ones(ARMPredicateReg *d, intptr_t oprsz, uint32_t HELPER(sve_brkns)(void *vd, void *vn, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; - + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); if (last_active_pred(vn, vg, oprsz)) { return predtest_ones(vd, oprsz, -1); } else { diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-sve.c +++ b/target/arm/translate-sve.c @@ -XXX,XX +XXX,XX @@ static bool do_brk3(DisasContext *s, arg_rprr_s *a, TCGv_ptr n = tcg_temp_new_ptr(); TCGv_ptr m = tcg_temp_new_ptr(); TCGv_ptr g = tcg_temp_new_ptr(); - TCGv_i32 t = tcg_const_i32(vsz - 2); + TCGv_i32 t = tcg_const_i32(FIELD_DP32(0, PREDDESC, OPRSZ, vsz)); tcg_gen_addi_ptr(d, cpu_env, pred_full_reg_offset(s, a->rd)); tcg_gen_addi_ptr(n, cpu_env, pred_full_reg_offset(s, a->rn)); @@ -XXX,XX +XXX,XX @@ static bool do_brk2(DisasContext *s, arg_rpr_s *a, TCGv_ptr d = tcg_temp_new_ptr(); TCGv_ptr n = tcg_temp_new_ptr(); TCGv_ptr g = tcg_temp_new_ptr(); - TCGv_i32 t = tcg_const_i32(vsz - 2); + TCGv_i32 t = tcg_const_i32(FIELD_DP32(0, PREDDESC, OPRSZ, vsz)); tcg_gen_addi_ptr(d, cpu_env, pred_full_reg_offset(s, a->rd)); tcg_gen_addi_ptr(n, cpu_env, pred_full_reg_offset(s, a->rn)); -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Since b64ee454a4a0, all predicate operations should be using these field macros for predicates. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210309155305.11301-7-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/sve_helper.c | 6 +++--- target/arm/translate-sve.c | 6 +++--- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve_helper.c +++ b/target/arm/sve_helper.c @@ -XXX,XX +XXX,XX @@ uint32_t HELPER(sve_brkns)(void *vd, void *vn, void *vg, uint32_t pred_desc) uint64_t HELPER(sve_cntp)(void *vn, void *vg, uint32_t pred_desc) { - intptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; - intptr_t esz = extract32(pred_desc, SIMD_DATA_SHIFT, 2); + intptr_t words = DIV_ROUND_UP(FIELD_EX32(pred_desc, PREDDESC, OPRSZ), 8); + intptr_t esz = FIELD_EX32(pred_desc, PREDDESC, ESZ); uint64_t *n = vn, *g = vg, sum = 0, mask = pred_esz_masks[esz]; intptr_t i; - for (i = 0; i < DIV_ROUND_UP(oprsz, 8); ++i) { + for (i = 0; i < words; ++i) { uint64_t t = n[i] & g[i] & mask; sum += ctpop64(t); } diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-sve.c +++ b/target/arm/translate-sve.c @@ -XXX,XX +XXX,XX @@ static void do_cntp(DisasContext *s, TCGv_i64 val, int esz, int pn, int pg) } else { TCGv_ptr t_pn = tcg_temp_new_ptr(); TCGv_ptr t_pg = tcg_temp_new_ptr(); - unsigned desc; + unsigned desc = 0; TCGv_i32 t_desc; - desc = psz - 2; - desc = deposit32(desc, SIMD_DATA_SHIFT, 2, esz); + desc = FIELD_DP32(desc, PREDDESC, OPRSZ, psz); + desc = FIELD_DP32(desc, PREDDESC, ESZ, esz); tcg_gen_addi_ptr(t_pn, cpu_env, pred_full_reg_offset(s, pn)); tcg_gen_addi_ptr(t_pg, cpu_env, pred_full_reg_offset(s, pg)); -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> Since b64ee454a4a0, all predicate operations should be using these field macros for predicates. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210309155305.11301-8-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/sve_helper.c | 4 ++-- target/arm/translate-sve.c | 7 ++++--- 2 files changed, 6 insertions(+), 5 deletions(-) diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve_helper.c +++ b/target/arm/sve_helper.c @@ -XXX,XX +XXX,XX @@ uint64_t HELPER(sve_cntp)(void *vn, void *vg, uint32_t pred_desc) uint32_t HELPER(sve_while)(void *vd, uint32_t count, uint32_t pred_desc) { - uintptr_t oprsz = extract32(pred_desc, 0, SIMD_OPRSZ_BITS) + 2; - intptr_t esz = extract32(pred_desc, SIMD_DATA_SHIFT, 2); + intptr_t oprsz = FIELD_EX32(pred_desc, PREDDESC, OPRSZ); + intptr_t esz = FIELD_EX32(pred_desc, PREDDESC, ESZ); uint64_t esz_mask = pred_esz_masks[esz]; ARMPredicateReg *d = vd; uint32_t flags; diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-sve.c +++ b/target/arm/translate-sve.c @@ -XXX,XX +XXX,XX @@ static bool trans_WHILE(DisasContext *s, arg_WHILE *a) TCGv_i64 op0, op1, t0, t1, tmax; TCGv_i32 t2, t3; TCGv_ptr ptr; - unsigned desc, vsz = vec_full_reg_size(s); + unsigned vsz = vec_full_reg_size(s); + unsigned desc = 0; TCGCond cond; if (!sve_access_check(s)) { @@ -XXX,XX +XXX,XX @@ static bool trans_WHILE(DisasContext *s, arg_WHILE *a) /* Scale elements to bits. */ tcg_gen_shli_i32(t2, t2, a->esz); - desc = (vsz / 8) - 2; - desc = deposit32(desc, SIMD_DATA_SHIFT, 2, a->esz); + desc = FIELD_DP32(desc, PREDDESC, OPRSZ, vsz / 8); + desc = FIELD_DP32(desc, PREDDESC, ESZ, a->esz); t3 = tcg_const_i32(desc); ptr = tcg_temp_new_ptr(); -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> With the reduction operations, we intentionally increase maxsz to the next power of 2, so as to fill out the reduction tree correctly. Since e2e7168a214b, oprsz must equal maxsz, with exceptions for small vectors, so this triggers an assertion for vector sizes > 32 that are not themselves a power of 2. Pass the power-of-two value in the simd_data field instead. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20210309155305.11301-9-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/sve_helper.c | 2 +- target/arm/translate-sve.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/target/arm/sve_helper.c b/target/arm/sve_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/sve_helper.c +++ b/target/arm/sve_helper.c @@ -XXX,XX +XXX,XX @@ static TYPE NAME##_reduce(TYPE *data, float_status *status, uintptr_t n) \ } \ uint64_t HELPER(NAME)(void *vn, void *vg, void *vs, uint32_t desc) \ { \ - uintptr_t i, oprsz = simd_oprsz(desc), maxsz = simd_maxsz(desc); \ + uintptr_t i, oprsz = simd_oprsz(desc), maxsz = simd_data(desc); \ TYPE data[sizeof(ARMVectorReg) / sizeof(TYPE)]; \ for (i = 0; i < oprsz; ) { \ uint16_t pg = *(uint16_t *)(vg + H1_2(i >> 3)); \ diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-sve.c +++ b/target/arm/translate-sve.c @@ -XXX,XX +XXX,XX @@ static void do_reduce(DisasContext *s, arg_rpr_esz *a, { unsigned vsz = vec_full_reg_size(s); unsigned p2vsz = pow2ceil(vsz); - TCGv_i32 t_desc = tcg_const_i32(simd_desc(vsz, p2vsz, 0)); + TCGv_i32 t_desc = tcg_const_i32(simd_desc(vsz, vsz, p2vsz)); TCGv_ptr t_zn, t_pg, status; TCGv_i64 temp; -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> Currently the emulated EMAC for sun8i always traverses the transmit queue from the head when transferring packets. It searches for a list of consecutive descriptors whichs are flagged as ready for processing and transmits their payloads accordingly. The controller stops processing once it finds a descriptor that is not marked ready. While the above behaviour works in most situations, it is not the same as the actual EMAC in hardware. Actual hardware uses the TX_CUR_DESC register value to keep track of the last position in the transmit queue and continues processing from that position when software triggers the start of DMA processing. The currently emulated behaviour can lead to packet loss on transmit when software fills the transmit queue with ready descriptors that overlap the tail of the circular list. This commit modifies the emulated EMAC for sun8i such that it processes the transmit queue using the TX_CUR_DESC register in the same way as hardware. Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 20210310195820.21950-2-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/net/allwinner-sun8i-emac.c | 62 +++++++++++++++++++---------------- 1 file changed, 34 insertions(+), 28 deletions(-) diff --git a/hw/net/allwinner-sun8i-emac.c b/hw/net/allwinner-sun8i-emac.c index XXXXXXX..XXXXXXX 100644 --- a/hw/net/allwinner-sun8i-emac.c +++ b/hw/net/allwinner-sun8i-emac.c @@ -XXX,XX +XXX,XX @@ static void allwinner_sun8i_emac_update_irq(AwSun8iEmacState *s) qemu_set_irq(s->irq, (s->int_sta & s->int_en) != 0); } -static uint32_t allwinner_sun8i_emac_next_desc(AwSun8iEmacState *s, - FrameDescriptor *desc, - size_t min_size) +static bool allwinner_sun8i_emac_desc_owned(FrameDescriptor *desc, + size_t min_buf_size) { - uint32_t paddr = desc->next; - - dma_memory_read(&s->dma_as, paddr, desc, sizeof(*desc)); - - if ((desc->status & DESC_STATUS_CTL) && - (desc->status2 & DESC_STATUS2_BUF_SIZE_MASK) >= min_size) { - return paddr; - } else { - return 0; - } + return (desc->status & DESC_STATUS_CTL) && (min_buf_size == 0 || + (desc->status2 & DESC_STATUS2_BUF_SIZE_MASK) >= min_buf_size); } -static uint32_t allwinner_sun8i_emac_get_desc(AwSun8iEmacState *s, - FrameDescriptor *desc, - uint32_t start_addr, - size_t min_size) +static void allwinner_sun8i_emac_get_desc(AwSun8iEmacState *s, + FrameDescriptor *desc, + uint32_t phys_addr) +{ + dma_memory_read(&s->dma_as, phys_addr, desc, sizeof(*desc)); +} + +static uint32_t allwinner_sun8i_emac_next_desc(AwSun8iEmacState *s, + FrameDescriptor *desc) +{ + const uint32_t nxt = desc->next; + allwinner_sun8i_emac_get_desc(s, desc, nxt); + return nxt; +} + +static uint32_t allwinner_sun8i_emac_find_desc(AwSun8iEmacState *s, + FrameDescriptor *desc, + uint32_t start_addr, + size_t min_size) { uint32_t desc_addr = start_addr; /* Note that the list is a cycle. Last entry points back to the head. */ while (desc_addr != 0) { - dma_memory_read(&s->dma_as, desc_addr, desc, sizeof(*desc)); + allwinner_sun8i_emac_get_desc(s, desc, desc_addr); - if ((desc->status & DESC_STATUS_CTL) && - (desc->status2 & DESC_STATUS2_BUF_SIZE_MASK) >= min_size) { + if (allwinner_sun8i_emac_desc_owned(desc, min_size)) { return desc_addr; } else if (desc->next == start_addr) { break; @@ -XXX,XX +XXX,XX @@ static uint32_t allwinner_sun8i_emac_rx_desc(AwSun8iEmacState *s, FrameDescriptor *desc, size_t min_size) { - return allwinner_sun8i_emac_get_desc(s, desc, s->rx_desc_curr, min_size); + return allwinner_sun8i_emac_find_desc(s, desc, s->rx_desc_curr, min_size); } static uint32_t allwinner_sun8i_emac_tx_desc(AwSun8iEmacState *s, - FrameDescriptor *desc, - size_t min_size) + FrameDescriptor *desc) { - return allwinner_sun8i_emac_get_desc(s, desc, s->tx_desc_head, min_size); + allwinner_sun8i_emac_get_desc(s, desc, s->tx_desc_curr); + return s->tx_desc_curr; } static void allwinner_sun8i_emac_flush_desc(AwSun8iEmacState *s, @@ -XXX,XX +XXX,XX @@ static ssize_t allwinner_sun8i_emac_receive(NetClientState *nc, bytes_left -= desc_bytes; /* Move to the next descriptor */ - s->rx_desc_curr = allwinner_sun8i_emac_next_desc(s, &desc, 64); + s->rx_desc_curr = allwinner_sun8i_emac_find_desc(s, &desc, desc.next, + AW_SUN8I_EMAC_MIN_PKT_SZ); if (!s->rx_desc_curr) { /* Not enough buffer space available */ s->int_sta |= INT_STA_RX_BUF_UA; @@ -XXX,XX +XXX,XX @@ static void allwinner_sun8i_emac_transmit(AwSun8iEmacState *s) size_t transmitted = 0; static uint8_t packet_buf[2048]; - s->tx_desc_curr = allwinner_sun8i_emac_tx_desc(s, &desc, 0); + s->tx_desc_curr = allwinner_sun8i_emac_tx_desc(s, &desc); /* Read all transmit descriptors */ - while (s->tx_desc_curr != 0) { + while (allwinner_sun8i_emac_desc_owned(&desc, 0)) { /* Read from physical memory into packet buffer */ bytes = desc.status2 & DESC_STATUS2_BUF_SIZE_MASK; @@ -XXX,XX +XXX,XX @@ static void allwinner_sun8i_emac_transmit(AwSun8iEmacState *s) packet_bytes = 0; transmitted++; } - s->tx_desc_curr = allwinner_sun8i_emac_next_desc(s, &desc, 0); + s->tx_desc_curr = allwinner_sun8i_emac_next_desc(s, &desc); } /* Raise transmit completed interrupt */ -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> The image for Armbian 19.11.3 bionic has been removed from the armbian server. Without the image as input the test arm_orangepi_bionic_19_11 cannot run. This commit removes the test completely and merges the code of the generic function do_test_arm_orangepi_uboot_armbian back with the 20.08 test. Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Willian Rampazzo <willianr@redhat.com> Message-id: 20210310195820.21950-3-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/acceptance/boot_linux_console.py | 72 ++++++++------------------ 1 file changed, 23 insertions(+), 49 deletions(-) diff --git a/tests/acceptance/boot_linux_console.py b/tests/acceptance/boot_linux_console.py index XXXXXXX..XXXXXXX 100644 --- a/tests/acceptance/boot_linux_console.py +++ b/tests/acceptance/boot_linux_console.py @@ -XXX,XX +XXX,XX @@ def test_arm_orangepi_sd(self): # Wait for VM to shut down gracefully self.vm.wait() - def do_test_arm_orangepi_uboot_armbian(self, image_path): + @skipUnless(os.getenv('ARMBIAN_ARTIFACTS_CACHED'), + 'Test artifacts fetched from unreliable apt.armbian.com') + @skipUnless(os.getenv('AVOCADO_ALLOW_LARGE_STORAGE'), 'storage limited') + def test_arm_orangepi_bionic_20_08(self): + """ + :avocado: tags=arch:arm + :avocado: tags=machine:orangepi-pc + :avocado: tags=device:sd + """ + + # This test download a 275 MiB compressed image and expand it + # to 1036 MiB, but the underlying filesystem is 1552 MiB... + # As we expand it to 2 GiB we are safe. + + image_url = ('https://dl.armbian.com/orangepipc/archive/' + 'Armbian_20.08.1_Orangepipc_bionic_current_5.8.5.img.xz') + image_hash = ('b4d6775f5673486329e45a0586bf06b6' + 'dbe792199fd182ac6b9c7bb6c7d3e6dd') + image_path_xz = self.fetch_asset(image_url, asset_hash=image_hash, + algorithm='sha256') + image_path = archive.extract(image_path_xz, self.workdir) + image_pow2ceil_expand(image_path) + self.vm.set_console() self.vm.add_args('-drive', 'file=' + image_path + ',if=sd,format=raw', '-nic', 'user', @@ -XXX,XX +XXX,XX @@ def do_test_arm_orangepi_uboot_armbian(self, image_path): 'to <orangepipc>') self.wait_for_console_pattern('Starting Load Kernel Modules...') - @skipUnless(os.getenv('ARMBIAN_ARTIFACTS_CACHED'), - 'Test artifacts fetched from unreliable apt.armbian.com') - @skipUnless(os.getenv('AVOCADO_ALLOW_LARGE_STORAGE'), 'storage limited') - @skipUnless(P7ZIP_AVAILABLE, '7z not installed') - def test_arm_orangepi_bionic_19_11(self): - """ - :avocado: tags=arch:arm - :avocado: tags=machine:orangepi-pc - :avocado: tags=device:sd - """ - - # This test download a 196MB compressed image and expand it to 1GB - image_url = ('https://dl.armbian.com/orangepipc/archive/' - 'Armbian_19.11.3_Orangepipc_bionic_current_5.3.9.7z') - image_hash = '196a8ffb72b0123d92cea4a070894813d305c71e' - image_path_7z = self.fetch_asset(image_url, asset_hash=image_hash) - image_name = 'Armbian_19.11.3_Orangepipc_bionic_current_5.3.9.img' - image_path = os.path.join(self.workdir, image_name) - process.run("7z e -o%s %s" % (self.workdir, image_path_7z)) - image_pow2ceil_expand(image_path) - - self.do_test_arm_orangepi_uboot_armbian(image_path) - - @skipUnless(os.getenv('ARMBIAN_ARTIFACTS_CACHED'), - 'Test artifacts fetched from unreliable apt.armbian.com') - @skipUnless(os.getenv('AVOCADO_ALLOW_LARGE_STORAGE'), 'storage limited') - def test_arm_orangepi_bionic_20_08(self): - """ - :avocado: tags=arch:arm - :avocado: tags=machine:orangepi-pc - :avocado: tags=device:sd - """ - - # This test download a 275 MiB compressed image and expand it - # to 1036 MiB, but the underlying filesystem is 1552 MiB... - # As we expand it to 2 GiB we are safe. - - image_url = ('https://dl.armbian.com/orangepipc/archive/' - 'Armbian_20.08.1_Orangepipc_bionic_current_5.8.5.img.xz') - image_hash = ('b4d6775f5673486329e45a0586bf06b6' - 'dbe792199fd182ac6b9c7bb6c7d3e6dd') - image_path_xz = self.fetch_asset(image_url, asset_hash=image_hash, - algorithm='sha256') - image_path = archive.extract(image_path_xz, self.workdir) - image_pow2ceil_expand(image_path) - - self.do_test_arm_orangepi_uboot_armbian(image_path) - @skipUnless(os.getenv('AVOCADO_ALLOW_LARGE_STORAGE'), 'storage limited') def test_arm_orangepi_uboot_netbsd9(self): """ -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> Update the download URL of the Armbian 20.08 Bionic image for test_arm_orangepi_bionic_20_08 of the orangepi-pc machine. The archive.armbian.com URL contains more images and should keep stable for a longer period of time than dl.armbian.com. Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Willian Rampazzo <willianr@redhat.com> Message-id: 20210310195820.21950-4-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/acceptance/boot_linux_console.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/acceptance/boot_linux_console.py b/tests/acceptance/boot_linux_console.py index XXXXXXX..XXXXXXX 100644 --- a/tests/acceptance/boot_linux_console.py +++ b/tests/acceptance/boot_linux_console.py @@ -XXX,XX +XXX,XX @@ def test_arm_orangepi_bionic_20_08(self): # to 1036 MiB, but the underlying filesystem is 1552 MiB... # As we expand it to 2 GiB we are safe. - image_url = ('https://dl.armbian.com/orangepipc/archive/' + image_url = ('https://archive.armbian.com/orangepipc/archive/' 'Armbian_20.08.1_Orangepipc_bionic_current_5.8.5.img.xz') image_hash = ('b4d6775f5673486329e45a0586bf06b6' 'dbe792199fd182ac6b9c7bb6c7d3e6dd') -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> The linux kernel 4.20.7 binary for sunxi has been removed from apt.armbian.com: $ ARMBIAN_ARTIFACTS_CACHED=yes AVOCADO_ALLOW_LARGE_STORAGE=yes avocado --show=app,console run -t machine:orangepi-pc tests/acceptance/boot_linux_console.py Fetching asset from tests/acceptance/boot_linux_console.py:BootLinuxConsole.test_arm_orangepi ... (1/6) tests/acceptance/boot_linux_console.py:BootLinuxConsole.test_arm_orangepi: CANCEL: Missing asset https://apt.armbian.com/pool/main/l/linux-4.20.7-sunxi/linux-image-dev-sunxi_5.75_armhf.deb (0.55 s) This commit updates the sunxi kernel to 5.10.16 for the acceptance tests of the orangepi-pc and cubieboard machines. Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Willian Rampazzo <willianr@redhat.com> Message-id: 20210310195820.21950-5-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/acceptance/boot_linux_console.py | 40 +++++++++++++------------- tests/acceptance/replay_kernel.py | 8 +++--- 2 files changed, 24 insertions(+), 24 deletions(-) diff --git a/tests/acceptance/boot_linux_console.py b/tests/acceptance/boot_linux_console.py index XXXXXXX..XXXXXXX 100644 --- a/tests/acceptance/boot_linux_console.py +++ b/tests/acceptance/boot_linux_console.py @@ -XXX,XX +XXX,XX @@ def test_arm_cubieboard_initrd(self): :avocado: tags=machine:cubieboard """ deb_url = ('https://apt.armbian.com/pool/main/l/' - 'linux-4.20.7-sunxi/linux-image-dev-sunxi_5.75_armhf.deb') - deb_hash = '1334c29c44d984ffa05ed10de8c3361f33d78315' + 'linux-5.10.16-sunxi/linux-image-current-sunxi_21.02.2_armhf.deb') + deb_hash = '9fa84beda245cabf0b4fa84cf6eaa7738ead1da0' deb_path = self.fetch_asset(deb_url, asset_hash=deb_hash) kernel_path = self.extract_from_deb(deb_path, - '/boot/vmlinuz-4.20.7-sunxi') - dtb_path = '/usr/lib/linux-image-dev-sunxi/sun4i-a10-cubieboard.dtb' + '/boot/vmlinuz-5.10.16-sunxi') + dtb_path = '/usr/lib/linux-image-current-sunxi/sun4i-a10-cubieboard.dtb' dtb_path = self.extract_from_deb(deb_path, dtb_path) initrd_url = ('https://github.com/groeck/linux-build-test/raw/' '2eb0a73b5d5a28df3170c546ddaaa9757e1e0848/rootfs/' @@ -XXX,XX +XXX,XX @@ def test_arm_cubieboard_sata(self): :avocado: tags=machine:cubieboard """ deb_url = ('https://apt.armbian.com/pool/main/l/' - 'linux-4.20.7-sunxi/linux-image-dev-sunxi_5.75_armhf.deb') - deb_hash = '1334c29c44d984ffa05ed10de8c3361f33d78315' + 'linux-5.10.16-sunxi/linux-image-current-sunxi_21.02.2_armhf.deb') + deb_hash = '9fa84beda245cabf0b4fa84cf6eaa7738ead1da0' deb_path = self.fetch_asset(deb_url, asset_hash=deb_hash) kernel_path = self.extract_from_deb(deb_path, - '/boot/vmlinuz-4.20.7-sunxi') - dtb_path = '/usr/lib/linux-image-dev-sunxi/sun4i-a10-cubieboard.dtb' + '/boot/vmlinuz-5.10.16-sunxi') + dtb_path = '/usr/lib/linux-image-current-sunxi/sun4i-a10-cubieboard.dtb' dtb_path = self.extract_from_deb(deb_path, dtb_path) rootfs_url = ('https://github.com/groeck/linux-build-test/raw/' '2eb0a73b5d5a28df3170c546ddaaa9757e1e0848/rootfs/' @@ -XXX,XX +XXX,XX @@ def test_arm_orangepi(self): :avocado: tags=machine:orangepi-pc """ deb_url = ('https://apt.armbian.com/pool/main/l/' - 'linux-4.20.7-sunxi/linux-image-dev-sunxi_5.75_armhf.deb') - deb_hash = '1334c29c44d984ffa05ed10de8c3361f33d78315' + 'linux-5.10.16-sunxi/linux-image-current-sunxi_21.02.2_armhf.deb') + deb_hash = '9fa84beda245cabf0b4fa84cf6eaa7738ead1da0' deb_path = self.fetch_asset(deb_url, asset_hash=deb_hash) kernel_path = self.extract_from_deb(deb_path, - '/boot/vmlinuz-4.20.7-sunxi') - dtb_path = '/usr/lib/linux-image-dev-sunxi/sun8i-h3-orangepi-pc.dtb' + '/boot/vmlinuz-5.10.16-sunxi') + dtb_path = '/usr/lib/linux-image-current-sunxi/sun8i-h3-orangepi-pc.dtb' dtb_path = self.extract_from_deb(deb_path, dtb_path) self.vm.set_console() @@ -XXX,XX +XXX,XX @@ def test_arm_orangepi_initrd(self): :avocado: tags=machine:orangepi-pc """ deb_url = ('https://apt.armbian.com/pool/main/l/' - 'linux-4.20.7-sunxi/linux-image-dev-sunxi_5.75_armhf.deb') - deb_hash = '1334c29c44d984ffa05ed10de8c3361f33d78315' + 'linux-5.10.16-sunxi/linux-image-current-sunxi_21.02.2_armhf.deb') + deb_hash = '9fa84beda245cabf0b4fa84cf6eaa7738ead1da0' deb_path = self.fetch_asset(deb_url, asset_hash=deb_hash) kernel_path = self.extract_from_deb(deb_path, - '/boot/vmlinuz-4.20.7-sunxi') - dtb_path = '/usr/lib/linux-image-dev-sunxi/sun8i-h3-orangepi-pc.dtb' + '/boot/vmlinuz-5.10.16-sunxi') + dtb_path = '/usr/lib/linux-image-current-sunxi/sun8i-h3-orangepi-pc.dtb' dtb_path = self.extract_from_deb(deb_path, dtb_path) initrd_url = ('https://github.com/groeck/linux-build-test/raw/' '2eb0a73b5d5a28df3170c546ddaaa9757e1e0848/rootfs/' @@ -XXX,XX +XXX,XX @@ def test_arm_orangepi_sd(self): :avocado: tags=device:sd """ deb_url = ('https://apt.armbian.com/pool/main/l/' - 'linux-4.20.7-sunxi/linux-image-dev-sunxi_5.75_armhf.deb') - deb_hash = '1334c29c44d984ffa05ed10de8c3361f33d78315' + 'linux-5.10.16-sunxi/linux-image-current-sunxi_21.02.2_armhf.deb') + deb_hash = '9fa84beda245cabf0b4fa84cf6eaa7738ead1da0' deb_path = self.fetch_asset(deb_url, asset_hash=deb_hash) kernel_path = self.extract_from_deb(deb_path, - '/boot/vmlinuz-4.20.7-sunxi') - dtb_path = '/usr/lib/linux-image-dev-sunxi/sun8i-h3-orangepi-pc.dtb' + '/boot/vmlinuz-5.10.16-sunxi') + dtb_path = '/usr/lib/linux-image-current-sunxi/sun8i-h3-orangepi-pc.dtb' dtb_path = self.extract_from_deb(deb_path, dtb_path) rootfs_url = ('http://storage.kernelci.org/images/rootfs/buildroot/' 'kci-2019.02/armel/base/rootfs.ext2.xz') diff --git a/tests/acceptance/replay_kernel.py b/tests/acceptance/replay_kernel.py index XXXXXXX..XXXXXXX 100644 --- a/tests/acceptance/replay_kernel.py +++ b/tests/acceptance/replay_kernel.py @@ -XXX,XX +XXX,XX @@ def test_arm_cubieboard_initrd(self): :avocado: tags=machine:cubieboard """ deb_url = ('https://apt.armbian.com/pool/main/l/' - 'linux-4.20.7-sunxi/linux-image-dev-sunxi_5.75_armhf.deb') - deb_hash = '1334c29c44d984ffa05ed10de8c3361f33d78315' + 'linux-5.10.16-sunxi/linux-image-current-sunxi_21.02.2_armhf.deb') + deb_hash = '9fa84beda245cabf0b4fa84cf6eaa7738ead1da0' deb_path = self.fetch_asset(deb_url, asset_hash=deb_hash) kernel_path = self.extract_from_deb(deb_path, - '/boot/vmlinuz-4.20.7-sunxi') - dtb_path = '/usr/lib/linux-image-dev-sunxi/sun4i-a10-cubieboard.dtb' + '/boot/vmlinuz-5.10.16-sunxi') + dtb_path = '/usr/lib/linux-image-current-sunxi/sun4i-a10-cubieboard.dtb' dtb_path = self.extract_from_deb(deb_path, dtb_path) initrd_url = ('https://github.com/groeck/linux-build-test/raw/' '2eb0a73b5d5a28df3170c546ddaaa9757e1e0848/rootfs/' -- 2.20.1 From: Niek Linnenbank <nieklinnenbank@gmail.com> Previously the ARMBIAN_ARTIFACTS_CACHED pre-condition was added to allow running tests that have already existing armbian.com artifacts stored in the local avocado cache, but do not have working URLs to download a fresh copy. At this time of writing the URLs for artifacts on the armbian.com server are updated and working. Any future broken URLs will result in a skipped acceptance test, for example: (1/5) tests/acceptance/boot_linux_console.py:BootLinuxConsole.test_arm_orangepi: CANCEL: Missing asset https://apt.armbian.com/pool/main/l/linux-4.20.7-sunxi/linux-image-dev-sunxi_5.75_armhf.deb (0.53 s) This commits removes the ARMBIAN_ARTIFACTS_CACHED pre-condition such that the acceptance tests for the orangepi-pc and cubieboard machines can run. Signed-off-by: Niek Linnenbank <nieklinnenbank@gmail.com> Reviewed-by: Willian Rampazzo <willianr@redhat.com> Message-id: 20210310195820.21950-6-nieklinnenbank@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/acceptance/boot_linux_console.py | 12 ------------ tests/acceptance/replay_kernel.py | 2 -- 2 files changed, 14 deletions(-) diff --git a/tests/acceptance/boot_linux_console.py b/tests/acceptance/boot_linux_console.py index XXXXXXX..XXXXXXX 100644 --- a/tests/acceptance/boot_linux_console.py +++ b/tests/acceptance/boot_linux_console.py @@ -XXX,XX +XXX,XX @@ def test_arm_exynos4210_initrd(self): self.wait_for_console_pattern('Boot successful.') # TODO user command, for now the uart is stuck - @skipUnless(os.getenv('ARMBIAN_ARTIFACTS_CACHED'), - 'Test artifacts fetched from unreliable apt.armbian.com') def test_arm_cubieboard_initrd(self): """ :avocado: tags=arch:arm @@ -XXX,XX +XXX,XX @@ def test_arm_cubieboard_initrd(self): 'system-control@1c00000') # cubieboard's reboot is not functioning; omit reboot test. - @skipUnless(os.getenv('ARMBIAN_ARTIFACTS_CACHED'), - 'Test artifacts fetched from unreliable apt.armbian.com') def test_arm_cubieboard_sata(self): """ :avocado: tags=arch:arm @@ -XXX,XX +XXX,XX @@ def test_arm_quanta_gsj_initrd(self): self.wait_for_console_pattern( 'Give root password for system maintenance') - @skipUnless(os.getenv('ARMBIAN_ARTIFACTS_CACHED'), - 'Test artifacts fetched from unreliable apt.armbian.com') def test_arm_orangepi(self): """ :avocado: tags=arch:arm @@ -XXX,XX +XXX,XX @@ def test_arm_orangepi(self): console_pattern = 'Kernel command line: %s' % kernel_command_line self.wait_for_console_pattern(console_pattern) - @skipUnless(os.getenv('ARMBIAN_ARTIFACTS_CACHED'), - 'Test artifacts fetched from unreliable apt.armbian.com') def test_arm_orangepi_initrd(self): """ :avocado: tags=arch:arm @@ -XXX,XX +XXX,XX @@ def test_arm_orangepi_initrd(self): # Wait for VM to shut down gracefully self.vm.wait() - @skipUnless(os.getenv('ARMBIAN_ARTIFACTS_CACHED'), - 'Test artifacts fetched from unreliable apt.armbian.com') def test_arm_orangepi_sd(self): """ :avocado: tags=arch:arm @@ -XXX,XX +XXX,XX @@ def test_arm_orangepi_sd(self): # Wait for VM to shut down gracefully self.vm.wait() - @skipUnless(os.getenv('ARMBIAN_ARTIFACTS_CACHED'), - 'Test artifacts fetched from unreliable apt.armbian.com') @skipUnless(os.getenv('AVOCADO_ALLOW_LARGE_STORAGE'), 'storage limited') def test_arm_orangepi_bionic_20_08(self): """ diff --git a/tests/acceptance/replay_kernel.py b/tests/acceptance/replay_kernel.py index XXXXXXX..XXXXXXX 100644 --- a/tests/acceptance/replay_kernel.py +++ b/tests/acceptance/replay_kernel.py @@ -XXX,XX +XXX,XX @@ def test_arm_virt(self): self.run_rr(kernel_path, kernel_command_line, console_pattern, shift=1) @skipIf(os.getenv('GITLAB_CI'), 'Running on GitLab') - @skipUnless(os.getenv('ARMBIAN_ARTIFACTS_CACHED'), - 'Test artifacts fetched from unreliable apt.armbian.com') def test_arm_cubieboard_initrd(self): """ :avocado: tags=arch:arm -- 2.20.1 From: Philippe Mathieu-Daudé <f4bug@amsat.org> If the SSECounter link is absent, we set an error message in sse_timer_realize() but forgot to propagate this error. Add the missing 'return'. Fixes: CID 1450755 (Null pointer dereferences) Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 20210312001845.1562670-1-f4bug@amsat.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/timer/sse-timer.c | 1 + 1 file changed, 1 insertion(+) diff --git a/hw/timer/sse-timer.c b/hw/timer/sse-timer.c index XXXXXXX..XXXXXXX 100644 --- a/hw/timer/sse-timer.c +++ b/hw/timer/sse-timer.c @@ -XXX,XX +XXX,XX @@ static void sse_timer_realize(DeviceState *dev, Error **errp) if (!s->counter) { error_setg(errp, "counter property was not set"); + return; } s->counter_notifier.notify = sse_timer_counter_callback; -- 2.20.1 From: Andrew Jones <drjones@redhat.com> Prior to commit f2ce39b4f067 a MachineClass kvm_type method only needed to be registered to ensure it would be executed. With commit f2ce39b4f067 a kvm-type machine property must also be specified. hw/arm/virt relies on the kvm_type method to pass its selected IPA limit to KVM, but this is not exposed as a machine property. Restore the previous functionality of invoking kvm_type when it's present. Fixes: f2ce39b4f067 ("vl: make qemu_get_machine_opts static") Signed-off-by: Andrew Jones <drjones@redhat.com> Reviewed-by: Eric Auger <eric.auger@redhat.com> Message-id: 20210310135218.255205-2-drjones@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/boards.h | 1 + accel/kvm/kvm-all.c | 2 ++ 2 files changed, 3 insertions(+) diff --git a/include/hw/boards.h b/include/hw/boards.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/boards.h +++ b/include/hw/boards.h @@ -XXX,XX +XXX,XX @@ typedef struct { * @kvm_type: * Return the type of KVM corresponding to the kvm-type string option or * computed based on other criteria such as the host kernel capabilities. + * kvm-type may be NULL if it is not needed. * @numa_mem_supported: * true if '--numa node.mem' option is supported and false otherwise * @smp_parse: diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c index XXXXXXX..XXXXXXX 100644 --- a/accel/kvm/kvm-all.c +++ b/accel/kvm/kvm-all.c @@ -XXX,XX +XXX,XX @@ static int kvm_init(MachineState *ms) "kvm-type", &error_abort); type = mc->kvm_type(ms, kvm_type); + } else if (mc->kvm_type) { + type = mc->kvm_type(ms, NULL); } do { -- 2.20.1 From: Andrew Jones <drjones@redhat.com> The virt machine already checks KVM_CAP_ARM_VM_IPA_SIZE to get the upper bound of the IPA size. If that bound is lower than the highest possible GPA for the machine, then QEMU will error out. However, the IPA is set to 40 when the highest GPA is less than or equal to 40, even when KVM may support an IPA limit as low as 32. This means KVM may fail the VM creation unnecessarily. Additionally, 40 is selected with the value 0, which means use the default, and that gets around a check in some versions of KVM, causing a difficult to debug fail. Always use the IPA size that corresponds to the highest possible GPA, unless it's lower than 32, in which case use 32. Also, we must still use 0 when KVM only supports the legacy fixed 40 bit IPA. Suggested-by: Marc Zyngier <maz@kernel.org> Signed-off-by: Andrew Jones <drjones@redhat.com> Reviewed-by: Eric Auger <eric.auger@redhat.com> Reviewed-by: Marc Zyngier <maz@kernel.org> Message-id: 20210310135218.255205-3-drjones@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/kvm_arm.h | 6 ++++-- hw/arm/virt.c | 23 ++++++++++++++++------- target/arm/kvm.c | 4 +++- 3 files changed, 23 insertions(+), 10 deletions(-) diff --git a/target/arm/kvm_arm.h b/target/arm/kvm_arm.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/kvm_arm.h +++ b/target/arm/kvm_arm.h @@ -XXX,XX +XXX,XX @@ bool kvm_arm_sve_supported(void); /** * kvm_arm_get_max_vm_ipa_size: * @ms: Machine state handle + * @fixed_ipa: True when the IPA limit is fixed at 40. This is the case + * for legacy KVM. * * Returns the number of bits in the IPA address space supported by KVM */ -int kvm_arm_get_max_vm_ipa_size(MachineState *ms); +int kvm_arm_get_max_vm_ipa_size(MachineState *ms, bool *fixed_ipa); /** * kvm_arm_sync_mpstate_to_kvm: @@ -XXX,XX +XXX,XX @@ static inline void kvm_arm_add_vcpu_properties(Object *obj) g_assert_not_reached(); } -static inline int kvm_arm_get_max_vm_ipa_size(MachineState *ms) +static inline int kvm_arm_get_max_vm_ipa_size(MachineState *ms, bool *fixed_ipa) { g_assert_not_reached(); } diff --git a/hw/arm/virt.c b/hw/arm/virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -XXX,XX +XXX,XX @@ static HotplugHandler *virt_machine_get_hotplug_handler(MachineState *machine, static int virt_kvm_type(MachineState *ms, const char *type_str) { VirtMachineState *vms = VIRT_MACHINE(ms); - int max_vm_pa_size = kvm_arm_get_max_vm_ipa_size(ms); - int requested_pa_size; + int max_vm_pa_size, requested_pa_size; + bool fixed_ipa; + + max_vm_pa_size = kvm_arm_get_max_vm_ipa_size(ms, &fixed_ipa); /* we freeze the memory map to compute the highest gpa */ virt_set_memmap(vms); requested_pa_size = 64 - clz64(vms->highest_gpa); + /* + * KVM requires the IPA size to be at least 32 bits. + */ + if (requested_pa_size < 32) { + requested_pa_size = 32; + } + if (requested_pa_size > max_vm_pa_size) { error_report("-m and ,maxmem option values " "require an IPA range (%d bits) larger than " "the one supported by the host (%d bits)", requested_pa_size, max_vm_pa_size); - exit(1); + exit(1); } /* - * By default we return 0 which corresponds to an implicit legacy - * 40b IPA setting. Otherwise we return the actual requested PA - * logsize + * We return the requested PA log size, unless KVM only supports + * the implicit legacy 40b IPA setting, in which case the kvm_type + * must be 0. */ - return requested_pa_size > 40 ? requested_pa_size : 0; + return fixed_ipa ? 0 : requested_pa_size; } static void virt_machine_class_init(ObjectClass *oc, void *data) diff --git a/target/arm/kvm.c b/target/arm/kvm.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/kvm.c +++ b/target/arm/kvm.c @@ -XXX,XX +XXX,XX @@ bool kvm_arm_pmu_supported(void) return kvm_check_extension(kvm_state, KVM_CAP_ARM_PMU_V3); } -int kvm_arm_get_max_vm_ipa_size(MachineState *ms) +int kvm_arm_get_max_vm_ipa_size(MachineState *ms, bool *fixed_ipa) { KVMState *s = KVM_STATE(ms->accelerator); int ret; ret = kvm_check_extension(s, KVM_CAP_ARM_VM_IPA_SIZE); + *fixed_ipa = ret <= 0; + return ret > 0 ? ret : 40; } -- 2.20.1 From: Hao Wu <wuhaotsh@google.com> This patch adds GPIOs in NPCM7xx PWM module for its duty values. The purpose of this is to connect it to the MFT module to provide an input for measuring a PWM fan's RPM. Each PWM module has NPCM7XX_PWM_PER_MODULE of GPIOs, each one corresponds to one PWM instance and can connect to multiple fan instances in MFT. Reviewed-by: Doug Evans <dje@google.com> Reviewed-by: Tyrone Ting <kfting@nuvoton.com> Signed-off-by: Hao Wu <wuhaotsh@google.com> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 20210311180855.149764-2-wuhaotsh@google.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/misc/npcm7xx_pwm.h | 4 +++- hw/misc/npcm7xx_pwm.c | 4 ++++ 2 files changed, 7 insertions(+), 1 deletion(-) diff --git a/include/hw/misc/npcm7xx_pwm.h b/include/hw/misc/npcm7xx_pwm.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/misc/npcm7xx_pwm.h +++ b/include/hw/misc/npcm7xx_pwm.h @@ -XXX,XX +XXX,XX @@ typedef struct NPCM7xxPWM { * @iomem: Memory region through which registers are accessed. * @clock: The PWM clock. * @pwm: The PWM channels owned by this module. + * @duty_gpio_out: The duty cycle of each PWM channels as a output GPIO. * @ppr: The prescaler register. * @csr: The clock selector register. * @pcr: The control register. @@ -XXX,XX +XXX,XX @@ struct NPCM7xxPWMState { MemoryRegion iomem; Clock *clock; - NPCM7xxPWM pwm[NPCM7XX_PWM_PER_MODULE]; + NPCM7xxPWM pwm[NPCM7XX_PWM_PER_MODULE]; + qemu_irq duty_gpio_out[NPCM7XX_PWM_PER_MODULE]; uint32_t ppr; uint32_t csr; diff --git a/hw/misc/npcm7xx_pwm.c b/hw/misc/npcm7xx_pwm.c index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/npcm7xx_pwm.c +++ b/hw/misc/npcm7xx_pwm.c @@ -XXX,XX +XXX,XX @@ static void npcm7xx_pwm_update_duty(NPCM7xxPWM *p) trace_npcm7xx_pwm_update_duty(DEVICE(p->module)->canonical_path, p->index, p->duty, duty); p->duty = duty; + qemu_set_irq(p->module->duty_gpio_out[p->index], p->duty); } } @@ -XXX,XX +XXX,XX @@ static void npcm7xx_pwm_init(Object *obj) SysBusDevice *sbd = SYS_BUS_DEVICE(obj); int i; + QEMU_BUILD_BUG_ON(ARRAY_SIZE(s->pwm) != NPCM7XX_PWM_PER_MODULE); for (i = 0; i < NPCM7XX_PWM_PER_MODULE; i++) { NPCM7xxPWM *p = &s->pwm[i]; p->module = s; @@ -XXX,XX +XXX,XX @@ static void npcm7xx_pwm_init(Object *obj) object_property_add_uint32_ptr(obj, "duty[*]", &s->pwm[i].duty, OBJ_PROP_FLAG_READ); } + qdev_init_gpio_out_named(DEVICE(s), s->duty_gpio_out, + "duty-gpio-out", NPCM7XX_PWM_PER_MODULE); } static const VMStateDescription vmstate_npcm7xx_pwm = { -- 2.20.1 From: Hao Wu <wuhaotsh@google.com> This patch implements Multi Function Timer (MFT) module for NPCM7XX. This module is mainly used to configure PWM fans. It has just enough functionality to make the PWM fan kernel module work. The module takes two input, the max_rpm of a fan (modifiable via QMP) and duty cycle (a GPIO from the PWM module.) The actual measured RPM is equal to max_rpm * duty_cycle / NPCM7XX_PWM_MAX_DUTY. The RPM is measured as a counter compared to a prescaled input clock. The kernel driver reads this counter and report to user space. Refs: https://github.com/torvalds/linux/blob/master/drivers/hwmon/npcm750-pwm-fan.c Reviewed-by: Doug Evans <dje@google.com> Reviewed-by: Tyrone Ting <kfting@nuvoton.com> Signed-off-by: Hao Wu <wuhaotsh@google.com> Message-id: 20210311180855.149764-3-wuhaotsh@google.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/misc/npcm7xx_mft.h | 70 +++++ hw/misc/npcm7xx_mft.c | 540 ++++++++++++++++++++++++++++++++++ hw/misc/meson.build | 1 + hw/misc/trace-events | 8 + 4 files changed, 619 insertions(+) create mode 100644 include/hw/misc/npcm7xx_mft.h create mode 100644 hw/misc/npcm7xx_mft.c diff --git a/include/hw/misc/npcm7xx_mft.h b/include/hw/misc/npcm7xx_mft.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/misc/npcm7xx_mft.h @@ -XXX,XX +XXX,XX @@ +/* + * Nuvoton NPCM7xx MFT Module + * + * Copyright 2021 Google LLC + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ +#ifndef NPCM7XX_MFT_H +#define NPCM7XX_MFT_H + +#include "exec/memory.h" +#include "hw/clock.h" +#include "hw/irq.h" +#include "hw/sysbus.h" +#include "qom/object.h" + +/* Max Fan input number. */ +#define NPCM7XX_MFT_MAX_FAN_INPUT 19 + +/* + * Number of registers in one MFT module. Don't change this without increasing + * the version_id in vmstate. + */ +#define NPCM7XX_MFT_NR_REGS (0x20 / sizeof(uint16_t)) + +/* + * The MFT can take up to 4 inputs: A0, B0, A1, B1. It can measure one A and one + * B simultaneously. NPCM7XX_MFT_INASEL and NPCM7XX_MFT_INBSEL are used to + * select which A or B input are used. + */ +#define NPCM7XX_MFT_FANIN_COUNT 4 + +/** + * struct NPCM7xxMFTState - Multi Functional Tachometer device state. + * @parent: System bus device. + * @iomem: Memory region through which registers are accessed. + * @clock_in: The input clock for MFT from CLK module. + * @clock_{1,2}: The counter clocks for NPCM7XX_MFT_CNT{1,2} + * @irq: The IRQ for this MFT state. + * @regs: The MMIO registers. + * @max_rpm: The maximum rpm for fans. Order: A0, B0, A1, B1. + * @duty: The duty cycles for fans, relative to NPCM7XX_PWM_MAX_DUTY. + */ +typedef struct NPCM7xxMFTState { + SysBusDevice parent; + + MemoryRegion iomem; + + Clock *clock_in; + Clock *clock_1, *clock_2; + qemu_irq irq; + uint16_t regs[NPCM7XX_MFT_NR_REGS]; + + uint32_t max_rpm[NPCM7XX_MFT_FANIN_COUNT]; + uint32_t duty[NPCM7XX_MFT_FANIN_COUNT]; +} NPCM7xxMFTState; + +#define TYPE_NPCM7XX_MFT "npcm7xx-mft" +#define NPCM7XX_MFT(obj) \ + OBJECT_CHECK(NPCM7xxMFTState, (obj), TYPE_NPCM7XX_MFT) + +#endif /* NPCM7XX_MFT_H */ diff --git a/hw/misc/npcm7xx_mft.c b/hw/misc/npcm7xx_mft.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/misc/npcm7xx_mft.c @@ -XXX,XX +XXX,XX @@ +/* + * Nuvoton NPCM7xx MFT Module + * + * Copyright 2021 Google LLC + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License + * for more details. + */ + +#include "qemu/osdep.h" +#include "hw/irq.h" +#include "hw/qdev-clock.h" +#include "hw/qdev-properties.h" +#include "hw/misc/npcm7xx_mft.h" +#include "hw/misc/npcm7xx_pwm.h" +#include "hw/registerfields.h" +#include "migration/vmstate.h" +#include "qapi/error.h" +#include "qapi/visitor.h" +#include "qemu/bitops.h" +#include "qemu/error-report.h" +#include "qemu/log.h" +#include "qemu/module.h" +#include "qemu/timer.h" +#include "qemu/units.h" +#include "trace.h" + +/* + * Some of the registers can only accessed via 16-bit ops and some can only + * be accessed via 8-bit ops. However we mark all of them using REG16 to + * simplify implementation. npcm7xx_mft_check_mem_op checks the access length + * of memory operations. + */ +REG16(NPCM7XX_MFT_CNT1, 0x00); +REG16(NPCM7XX_MFT_CRA, 0x02); +REG16(NPCM7XX_MFT_CRB, 0x04); +REG16(NPCM7XX_MFT_CNT2, 0x06); +REG16(NPCM7XX_MFT_PRSC, 0x08); +REG16(NPCM7XX_MFT_CKC, 0x0a); +REG16(NPCM7XX_MFT_MCTRL, 0x0c); +REG16(NPCM7XX_MFT_ICTRL, 0x0e); +REG16(NPCM7XX_MFT_ICLR, 0x10); +REG16(NPCM7XX_MFT_IEN, 0x12); +REG16(NPCM7XX_MFT_CPA, 0x14); +REG16(NPCM7XX_MFT_CPB, 0x16); +REG16(NPCM7XX_MFT_CPCFG, 0x18); +REG16(NPCM7XX_MFT_INASEL, 0x1a); +REG16(NPCM7XX_MFT_INBSEL, 0x1c); + +/* Register Fields */ +#define NPCM7XX_MFT_CKC_C2CSEL BIT(3) +#define NPCM7XX_MFT_CKC_C1CSEL BIT(0) + +#define NPCM7XX_MFT_MCTRL_TBEN BIT(6) +#define NPCM7XX_MFT_MCTRL_TAEN BIT(5) +#define NPCM7XX_MFT_MCTRL_TBEDG BIT(4) +#define NPCM7XX_MFT_MCTRL_TAEDG BIT(3) +#define NPCM7XX_MFT_MCTRL_MODE5 BIT(2) + +#define NPCM7XX_MFT_ICTRL_TFPND BIT(5) +#define NPCM7XX_MFT_ICTRL_TEPND BIT(4) +#define NPCM7XX_MFT_ICTRL_TDPND BIT(3) +#define NPCM7XX_MFT_ICTRL_TCPND BIT(2) +#define NPCM7XX_MFT_ICTRL_TBPND BIT(1) +#define NPCM7XX_MFT_ICTRL_TAPND BIT(0) + +#define NPCM7XX_MFT_ICLR_TFCLR BIT(5) +#define NPCM7XX_MFT_ICLR_TECLR BIT(4) +#define NPCM7XX_MFT_ICLR_TDCLR BIT(3) +#define NPCM7XX_MFT_ICLR_TCCLR BIT(2) +#define NPCM7XX_MFT_ICLR_TBCLR BIT(1) +#define NPCM7XX_MFT_ICLR_TACLR BIT(0) + +#define NPCM7XX_MFT_IEN_TFIEN BIT(5) +#define NPCM7XX_MFT_IEN_TEIEN BIT(4) +#define NPCM7XX_MFT_IEN_TDIEN BIT(3) +#define NPCM7XX_MFT_IEN_TCIEN BIT(2) +#define NPCM7XX_MFT_IEN_TBIEN BIT(1) +#define NPCM7XX_MFT_IEN_TAIEN BIT(0) + +#define NPCM7XX_MFT_CPCFG_GET_B(rv) extract8((rv), 4, 4) +#define NPCM7XX_MFT_CPCFG_GET_A(rv) extract8((rv), 0, 4) +#define NPCM7XX_MFT_CPCFG_HIEN BIT(3) +#define NPCM7XX_MFT_CPCFG_EQEN BIT(2) +#define NPCM7XX_MFT_CPCFG_LOEN BIT(1) +#define NPCM7XX_MFT_CPCFG_CPSEL BIT(0) + +#define NPCM7XX_MFT_INASEL_SELA BIT(0) +#define NPCM7XX_MFT_INBSEL_SELB BIT(0) + +/* Max CNT values of the module. The CNT value is a countdown from it. */ +#define NPCM7XX_MFT_MAX_CNT 0xFFFF + +/* Each fan revolution should generated 2 pulses */ +#define NPCM7XX_MFT_PULSE_PER_REVOLUTION 2 + +typedef enum NPCM7xxMFTCaptureState { + /* capture succeeded with a valid CNT value. */ + NPCM7XX_CAPTURE_SUCCEED, + /* capture stopped prematurely due to reaching CPCFG condition. */ + NPCM7XX_CAPTURE_COMPARE_HIT, + /* capture fails since it reaches underflow condition for CNT. */ + NPCM7XX_CAPTURE_UNDERFLOW, +} NPCM7xxMFTCaptureState; + +static void npcm7xx_mft_reset(NPCM7xxMFTState *s) +{ + int i; + + /* Only registers PRSC ~ INBSEL need to be reset. */ + for (i = R_NPCM7XX_MFT_PRSC; i <= R_NPCM7XX_MFT_INBSEL; ++i) { + s->regs[i] = 0; + } +} + +static void npcm7xx_mft_clear_interrupt(NPCM7xxMFTState *s, uint8_t iclr) +{ + /* + * Clear bits in ICTRL where corresponding bits in iclr is 1. + * Both iclr and ictrl are 8-bit regs. (See npcm7xx_mft_check_mem_op) + */ + s->regs[R_NPCM7XX_MFT_ICTRL] &= ~iclr; +} + +/* + * If the CPCFG's condition should be triggered during count down from + * NPCM7XX_MFT_MAX_CNT to src if compared to tgt, return the count when + * the condition is triggered. + * Otherwise return -1. + * Since tgt is uint16_t it must always <= NPCM7XX_MFT_MAX_CNT. + */ +static int npcm7xx_mft_compare(int32_t src, uint16_t tgt, uint8_t cpcfg) +{ + if (cpcfg & NPCM7XX_MFT_CPCFG_HIEN) { + return NPCM7XX_MFT_MAX_CNT; + } + if ((cpcfg & NPCM7XX_MFT_CPCFG_EQEN) && (src <= tgt)) { + return tgt; + } + if ((cpcfg & NPCM7XX_MFT_CPCFG_LOEN) && (tgt > 0) && (src < tgt)) { + return tgt - 1; + } + + return -1; +} + +/* Compute CNT according to corresponding fan's RPM. */ +static NPCM7xxMFTCaptureState npcm7xx_mft_compute_cnt( + Clock *clock, uint32_t max_rpm, uint32_t duty, uint16_t tgt, + uint8_t cpcfg, uint16_t *cnt) +{ + uint32_t rpm = (uint64_t)max_rpm * (uint64_t)duty / NPCM7XX_PWM_MAX_DUTY; + int32_t count; + int stopped; + NPCM7xxMFTCaptureState state; + + if (rpm == 0) { + /* + * If RPM = 0, capture won't happen. CNT will continue count down. + * So it's effective equivalent to have a cnt > NPCM7XX_MFT_MAX_CNT + */ + count = NPCM7XX_MFT_MAX_CNT + 1; + } else { + /* + * RPM = revolution/min. The time for one revlution (in ns) is + * MINUTE_TO_NANOSECOND / RPM. + */ + count = clock_ns_to_ticks(clock, (60 * NANOSECONDS_PER_SECOND) / + (rpm * NPCM7XX_MFT_PULSE_PER_REVOLUTION)); + } + + if (count > NPCM7XX_MFT_MAX_CNT) { + count = -1; + } else { + /* The CNT is a countdown value from NPCM7XX_MFT_MAX_CNT. */ + count = NPCM7XX_MFT_MAX_CNT - count; + } + stopped = npcm7xx_mft_compare(count, tgt, cpcfg); + if (stopped == -1) { + if (count == -1) { + /* Underflow */ + state = NPCM7XX_CAPTURE_UNDERFLOW; + } else { + state = NPCM7XX_CAPTURE_SUCCEED; + } + } else { + count = stopped; + state = NPCM7XX_CAPTURE_COMPARE_HIT; + } + + if (count != -1) { + *cnt = count; + } + trace_npcm7xx_mft_rpm(clock->canonical_path, clock_get_hz(clock), + state, count, rpm, duty); + return state; +} + +/* + * Capture Fan RPM and update CNT and CR registers accordingly. + * Raise IRQ if certain contidions are met in IEN. + */ +static void npcm7xx_mft_capture(NPCM7xxMFTState *s) +{ + int irq_level = 0; + NPCM7xxMFTCaptureState state; + int sel; + uint8_t cpcfg; + + /* + * If not mode 5, the behavior is undefined. We just do nothing in this + * case. + */ + if (!(s->regs[R_NPCM7XX_MFT_MCTRL] & NPCM7XX_MFT_MCTRL_MODE5)) { + return; + } + + /* Capture input A. */ + if (s->regs[R_NPCM7XX_MFT_MCTRL] & NPCM7XX_MFT_MCTRL_TAEN && + s->regs[R_NPCM7XX_MFT_CKC] & NPCM7XX_MFT_CKC_C1CSEL) { + sel = s->regs[R_NPCM7XX_MFT_INASEL] & NPCM7XX_MFT_INASEL_SELA; + cpcfg = NPCM7XX_MFT_CPCFG_GET_A(s->regs[R_NPCM7XX_MFT_CPCFG]); + state = npcm7xx_mft_compute_cnt(s->clock_1, + sel ? s->max_rpm[2] : s->max_rpm[0], + sel ? s->duty[2] : s->duty[0], + s->regs[R_NPCM7XX_MFT_CPA], + cpcfg, + &s->regs[R_NPCM7XX_MFT_CNT1]); + switch (state) { + case NPCM7XX_CAPTURE_SUCCEED: + /* Interrupt on input capture on TAn transition - TAPND */ + s->regs[R_NPCM7XX_MFT_CRA] = s->regs[R_NPCM7XX_MFT_CNT1]; + s->regs[R_NPCM7XX_MFT_ICTRL] |= NPCM7XX_MFT_ICTRL_TAPND; + if (s->regs[R_NPCM7XX_MFT_IEN] & NPCM7XX_MFT_IEN_TAIEN) { + irq_level = 1; + } + break; + + case NPCM7XX_CAPTURE_COMPARE_HIT: + /* Compare Hit - TEPND */ + s->regs[R_NPCM7XX_MFT_ICTRL] |= NPCM7XX_MFT_ICTRL_TEPND; + if (s->regs[R_NPCM7XX_MFT_IEN] & NPCM7XX_MFT_IEN_TEIEN) { + irq_level = 1; + } + break; + + case NPCM7XX_CAPTURE_UNDERFLOW: + /* Underflow - TCPND */ + s->regs[R_NPCM7XX_MFT_ICTRL] |= NPCM7XX_MFT_ICTRL_TCPND; + if (s->regs[R_NPCM7XX_MFT_IEN] & NPCM7XX_MFT_IEN_TCIEN) { + irq_level = 1; + } + break; + + default: + g_assert_not_reached(); + } + } + + /* Capture input B. */ + if (s->regs[R_NPCM7XX_MFT_MCTRL] & NPCM7XX_MFT_MCTRL_TBEN && + s->regs[R_NPCM7XX_MFT_CKC] & NPCM7XX_MFT_CKC_C2CSEL) { + sel = s->regs[R_NPCM7XX_MFT_INBSEL] & NPCM7XX_MFT_INBSEL_SELB; + cpcfg = NPCM7XX_MFT_CPCFG_GET_B(s->regs[R_NPCM7XX_MFT_CPCFG]); + state = npcm7xx_mft_compute_cnt(s->clock_2, + sel ? s->max_rpm[3] : s->max_rpm[1], + sel ? s->duty[3] : s->duty[1], + s->regs[R_NPCM7XX_MFT_CPB], + cpcfg, + &s->regs[R_NPCM7XX_MFT_CNT2]); + switch (state) { + case NPCM7XX_CAPTURE_SUCCEED: + /* Interrupt on input capture on TBn transition - TBPND */ + s->regs[R_NPCM7XX_MFT_CRB] = s->regs[R_NPCM7XX_MFT_CNT2]; + s->regs[R_NPCM7XX_MFT_ICTRL] |= NPCM7XX_MFT_ICTRL_TBPND; + if (s->regs[R_NPCM7XX_MFT_IEN] & NPCM7XX_MFT_IEN_TBIEN) { + irq_level = 1; + } + break; + + case NPCM7XX_CAPTURE_COMPARE_HIT: + /* Compare Hit - TFPND */ + s->regs[R_NPCM7XX_MFT_ICTRL] |= NPCM7XX_MFT_ICTRL_TFPND; + if (s->regs[R_NPCM7XX_MFT_IEN] & NPCM7XX_MFT_IEN_TFIEN) { + irq_level = 1; + } + break; + + case NPCM7XX_CAPTURE_UNDERFLOW: + /* Underflow - TDPND */ + s->regs[R_NPCM7XX_MFT_ICTRL] |= NPCM7XX_MFT_ICTRL_TDPND; + if (s->regs[R_NPCM7XX_MFT_IEN] & NPCM7XX_MFT_IEN_TDIEN) { + irq_level = 1; + } + break; + + default: + g_assert_not_reached(); + } + } + + trace_npcm7xx_mft_capture(DEVICE(s)->canonical_path, irq_level); + qemu_set_irq(s->irq, irq_level); +} + +/* Update clock for counters. */ +static void npcm7xx_mft_update_clock(void *opaque, ClockEvent event) +{ + NPCM7xxMFTState *s = NPCM7XX_MFT(opaque); + uint64_t prescaled_clock_period; + + prescaled_clock_period = clock_get(s->clock_in) * + (s->regs[R_NPCM7XX_MFT_PRSC] + 1ULL); + trace_npcm7xx_mft_update_clock(s->clock_in->canonical_path, + s->regs[R_NPCM7XX_MFT_CKC], + clock_get(s->clock_in), + prescaled_clock_period); + /* Update clock 1 */ + if (s->regs[R_NPCM7XX_MFT_CKC] & NPCM7XX_MFT_CKC_C1CSEL) { + /* Clock is prescaled. */ + clock_update(s->clock_1, prescaled_clock_period); + } else { + /* Clock stopped. */ + clock_update(s->clock_1, 0); + } + /* Update clock 2 */ + if (s->regs[R_NPCM7XX_MFT_CKC] & NPCM7XX_MFT_CKC_C2CSEL) { + /* Clock is prescaled. */ + clock_update(s->clock_2, prescaled_clock_period); + } else { + /* Clock stopped. */ + clock_update(s->clock_2, 0); + } + + npcm7xx_mft_capture(s); +} + +static uint64_t npcm7xx_mft_read(void *opaque, hwaddr offset, unsigned size) +{ + NPCM7xxMFTState *s = NPCM7XX_MFT(opaque); + uint16_t value = 0; + + switch (offset) { + case A_NPCM7XX_MFT_ICLR: + qemu_log_mask(LOG_GUEST_ERROR, + "%s: register @ 0x%04" HWADDR_PRIx " is write-only\n", + __func__, offset); + break; + + default: + value = s->regs[offset / 2]; + } + + trace_npcm7xx_mft_read(DEVICE(s)->canonical_path, offset, value); + return value; +} + +static void npcm7xx_mft_write(void *opaque, hwaddr offset, + uint64_t v, unsigned size) +{ + NPCM7xxMFTState *s = NPCM7XX_MFT(opaque); + + trace_npcm7xx_mft_write(DEVICE(s)->canonical_path, offset, v); + switch (offset) { + case A_NPCM7XX_MFT_ICLR: + npcm7xx_mft_clear_interrupt(s, v); + break; + + case A_NPCM7XX_MFT_CKC: + case A_NPCM7XX_MFT_PRSC: + s->regs[offset / 2] = v; + npcm7xx_mft_update_clock(s, ClockUpdate); + break; + + default: + s->regs[offset / 2] = v; + npcm7xx_mft_capture(s); + break; + } +} + +static bool npcm7xx_mft_check_mem_op(void *opaque, hwaddr offset, + unsigned size, bool is_write, + MemTxAttrs attrs) +{ + switch (offset) { + /* 16-bit registers. Must be accessed with 16-bit read/write.*/ + case A_NPCM7XX_MFT_CNT1: + case A_NPCM7XX_MFT_CRA: + case A_NPCM7XX_MFT_CRB: + case A_NPCM7XX_MFT_CNT2: + case A_NPCM7XX_MFT_CPA: + case A_NPCM7XX_MFT_CPB: + return size == 2; + + /* 8-bit registers. Must be accessed with 8-bit read/write.*/ + case A_NPCM7XX_MFT_PRSC: + case A_NPCM7XX_MFT_CKC: + case A_NPCM7XX_MFT_MCTRL: + case A_NPCM7XX_MFT_ICTRL: + case A_NPCM7XX_MFT_ICLR: + case A_NPCM7XX_MFT_IEN: + case A_NPCM7XX_MFT_CPCFG: + case A_NPCM7XX_MFT_INASEL: + case A_NPCM7XX_MFT_INBSEL: + return size == 1; + + default: + /* Invalid registers. */ + return false; + } +} + +static void npcm7xx_mft_get_max_rpm(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + visit_type_uint32(v, name, (uint32_t *)opaque, errp); +} + +static void npcm7xx_mft_set_max_rpm(Object *obj, Visitor *v, const char *name, + void *opaque, Error **errp) +{ + NPCM7xxMFTState *s = NPCM7XX_MFT(obj); + uint32_t *max_rpm = opaque; + uint32_t value; + + if (!visit_type_uint32(v, name, &value, errp)) { + return; + } + + *max_rpm = value; + npcm7xx_mft_capture(s); +} + +static void npcm7xx_mft_duty_handler(void *opaque, int n, int value) +{ + NPCM7xxMFTState *s = NPCM7XX_MFT(opaque); + + trace_npcm7xx_mft_set_duty(DEVICE(s)->canonical_path, n, value); + s->duty[n] = value; + npcm7xx_mft_capture(s); +} + +static const struct MemoryRegionOps npcm7xx_mft_ops = { + .read = npcm7xx_mft_read, + .write = npcm7xx_mft_write, + .endianness = DEVICE_LITTLE_ENDIAN, + .valid = { + .min_access_size = 1, + .max_access_size = 2, + .unaligned = false, + .accepts = npcm7xx_mft_check_mem_op, + }, +}; + +static void npcm7xx_mft_enter_reset(Object *obj, ResetType type) +{ + NPCM7xxMFTState *s = NPCM7XX_MFT(obj); + + npcm7xx_mft_reset(s); +} + +static void npcm7xx_mft_hold_reset(Object *obj) +{ + NPCM7xxMFTState *s = NPCM7XX_MFT(obj); + + qemu_irq_lower(s->irq); +} + +static void npcm7xx_mft_init(Object *obj) +{ + NPCM7xxMFTState *s = NPCM7XX_MFT(obj); + SysBusDevice *sbd = SYS_BUS_DEVICE(obj); + DeviceState *dev = DEVICE(obj); + + memory_region_init_io(&s->iomem, obj, &npcm7xx_mft_ops, s, + TYPE_NPCM7XX_MFT, 4 * KiB); + sysbus_init_mmio(sbd, &s->iomem); + sysbus_init_irq(sbd, &s->irq); + s->clock_in = qdev_init_clock_in(dev, "clock-in", npcm7xx_mft_update_clock, + s, ClockUpdate); + s->clock_1 = qdev_init_clock_out(dev, "clock1"); + s->clock_2 = qdev_init_clock_out(dev, "clock2"); + + for (int i = 0; i < NPCM7XX_PWM_PER_MODULE; ++i) { + object_property_add(obj, "max_rpm[*]", "uint32", + npcm7xx_mft_get_max_rpm, + npcm7xx_mft_set_max_rpm, + NULL, &s->max_rpm[i]); + } + qdev_init_gpio_in_named(dev, npcm7xx_mft_duty_handler, "duty", + NPCM7XX_MFT_FANIN_COUNT); +} + +static const VMStateDescription vmstate_npcm7xx_mft = { + .name = "npcm7xx-mft-module", + .version_id = 0, + .minimum_version_id = 0, + .fields = (VMStateField[]) { + VMSTATE_CLOCK(clock_in, NPCM7xxMFTState), + VMSTATE_CLOCK(clock_1, NPCM7xxMFTState), + VMSTATE_CLOCK(clock_2, NPCM7xxMFTState), + VMSTATE_UINT16_ARRAY(regs, NPCM7xxMFTState, NPCM7XX_MFT_NR_REGS), + VMSTATE_UINT32_ARRAY(max_rpm, NPCM7xxMFTState, NPCM7XX_MFT_FANIN_COUNT), + VMSTATE_UINT32_ARRAY(duty, NPCM7xxMFTState, NPCM7XX_MFT_FANIN_COUNT), + VMSTATE_END_OF_LIST(), + }, +}; + +static void npcm7xx_mft_class_init(ObjectClass *klass, void *data) +{ + ResettableClass *rc = RESETTABLE_CLASS(klass); + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->desc = "NPCM7xx MFT Controller"; + dc->vmsd = &vmstate_npcm7xx_mft; + rc->phases.enter = npcm7xx_mft_enter_reset; + rc->phases.hold = npcm7xx_mft_hold_reset; +} + +static const TypeInfo npcm7xx_mft_info = { + .name = TYPE_NPCM7XX_MFT, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_size = sizeof(NPCM7xxMFTState), + .class_init = npcm7xx_mft_class_init, + .instance_init = npcm7xx_mft_init, +}; + +static void npcm7xx_mft_register_type(void) +{ + type_register_static(&npcm7xx_mft_info); +} +type_init(npcm7xx_mft_register_type); diff --git a/hw/misc/meson.build b/hw/misc/meson.build index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/meson.build +++ b/hw/misc/meson.build @@ -XXX,XX +XXX,XX @@ softmmu_ss.add(when: 'CONFIG_MAINSTONE', if_true: files('mst_fpga.c')) softmmu_ss.add(when: 'CONFIG_NPCM7XX', if_true: files( 'npcm7xx_clk.c', 'npcm7xx_gcr.c', + 'npcm7xx_mft.c', 'npcm7xx_pwm.c', 'npcm7xx_rng.c', )) diff --git a/hw/misc/trace-events b/hw/misc/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/trace-events +++ b/hw/misc/trace-events @@ -XXX,XX +XXX,XX @@ npcm7xx_clk_write(uint64_t offset, uint32_t value) "offset: 0x%04" PRIx64 " valu npcm7xx_gcr_read(uint64_t offset, uint32_t value) " offset: 0x%04" PRIx64 " value: 0x%08" PRIx32 npcm7xx_gcr_write(uint64_t offset, uint32_t value) "offset: 0x%04" PRIx64 " value: 0x%08" PRIx32 +# npcm7xx_mft.c +npcm7xx_mft_read(const char *name, uint64_t offset, uint16_t value) "%s: offset: 0x%04" PRIx64 " value: 0x%04" PRIx16 +npcm7xx_mft_write(const char *name, uint64_t offset, uint16_t value) "%s: offset: 0x%04" PRIx64 " value: 0x%04" PRIx16 +npcm7xx_mft_rpm(const char *clock, uint32_t clock_hz, int state, int32_t cnt, uint32_t rpm, uint32_t duty) " fan clk: %s clock_hz: %" PRIu32 ", state: %d, cnt: %" PRIi32 ", rpm: %" PRIu32 ", duty: %" PRIu32 +npcm7xx_mft_capture(const char *name, int irq_level) "%s: level: %d" +npcm7xx_mft_update_clock(const char *name, uint16_t sel, uint64_t clock_period, uint64_t prescaled_clock_period) "%s: sel: 0x%02" PRIx16 ", period: %" PRIu64 ", prescaled: %" PRIu64 +npcm7xx_mft_set_duty(const char *name, int n, int value) "%s[%d]: %d" + # npcm7xx_rng.c npcm7xx_rng_read(uint64_t offset, uint64_t value, unsigned size) "offset: 0x%04" PRIx64 " value: 0x%02" PRIx64 " size: %u" npcm7xx_rng_write(uint64_t offset, uint64_t value, unsigned size) "offset: 0x%04" PRIx64 " value: 0x%02" PRIx64 " size: %u" -- 2.20.1 From: Hao Wu <wuhaotsh@google.com> This patch adds the recently implemented MFT device to the NPCM7XX SoC file. Reviewed-by: Doug Evans <dje@google.com> Reviewed-by: Tyrone Ting <kfting@nuvoton.com> Signed-off-by: Hao Wu <wuhaotsh@google.com> Message-id: 20210311180855.149764-4-wuhaotsh@google.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/system/arm/nuvoton.rst | 2 +- include/hw/arm/npcm7xx.h | 2 ++ hw/arm/npcm7xx.c | 45 ++++++++++++++++++++++++++++++------- 3 files changed, 40 insertions(+), 9 deletions(-) diff --git a/docs/system/arm/nuvoton.rst b/docs/system/arm/nuvoton.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/system/arm/nuvoton.rst +++ b/docs/system/arm/nuvoton.rst @@ -XXX,XX +XXX,XX @@ Supported devices * Pulse Width Modulation (PWM) * SMBus controller (SMBF) * Ethernet controller (EMC) + * Tachometer Missing devices --------------- @@ -XXX,XX +XXX,XX @@ Missing devices * Peripheral SPI controller (PSPI) * SD/MMC host * PECI interface - * Tachometer * PCI and PCIe root complex and bridges * VDM and MCTP support * Serial I/O expansion diff --git a/include/hw/arm/npcm7xx.h b/include/hw/arm/npcm7xx.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/npcm7xx.h +++ b/include/hw/arm/npcm7xx.h @@ -XXX,XX +XXX,XX @@ #include "hw/mem/npcm7xx_mc.h" #include "hw/misc/npcm7xx_clk.h" #include "hw/misc/npcm7xx_gcr.h" +#include "hw/misc/npcm7xx_mft.h" #include "hw/misc/npcm7xx_pwm.h" #include "hw/misc/npcm7xx_rng.h" #include "hw/net/npcm7xx_emc.h" @@ -XXX,XX +XXX,XX @@ typedef struct NPCM7xxState { NPCM7xxTimerCtrlState tim[3]; NPCM7xxADCState adc; NPCM7xxPWMState pwm[2]; + NPCM7xxMFTState mft[8]; NPCM7xxOTPState key_storage; NPCM7xxOTPState fuse_array; NPCM7xxMCState mc; diff --git a/hw/arm/npcm7xx.c b/hw/arm/npcm7xx.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/npcm7xx.c +++ b/hw/arm/npcm7xx.c @@ -XXX,XX +XXX,XX @@ enum NPCM7xxInterrupt { NPCM7XX_SMBUS15_IRQ, NPCM7XX_PWM0_IRQ = 93, /* PWM module 0 */ NPCM7XX_PWM1_IRQ, /* PWM module 1 */ + NPCM7XX_MFT0_IRQ = 96, /* MFT module 0 */ + NPCM7XX_MFT1_IRQ, /* MFT module 1 */ + NPCM7XX_MFT2_IRQ, /* MFT module 2 */ + NPCM7XX_MFT3_IRQ, /* MFT module 3 */ + NPCM7XX_MFT4_IRQ, /* MFT module 4 */ + NPCM7XX_MFT5_IRQ, /* MFT module 5 */ + NPCM7XX_MFT6_IRQ, /* MFT module 6 */ + NPCM7XX_MFT7_IRQ, /* MFT module 7 */ NPCM7XX_EMC2RX_IRQ = 114, NPCM7XX_EMC2TX_IRQ, NPCM7XX_GPIO0_IRQ = 116, @@ -XXX,XX +XXX,XX @@ static const hwaddr npcm7xx_pwm_addr[] = { 0xf0104000, }; +/* Register base address for each MFT Module */ +static const hwaddr npcm7xx_mft_addr[] = { + 0xf0180000, + 0xf0181000, + 0xf0182000, + 0xf0183000, + 0xf0184000, + 0xf0185000, + 0xf0186000, + 0xf0187000, +}; + /* Direct memory-mapped access to each SMBus Module. */ static const hwaddr npcm7xx_smbus_addr[] = { 0xf0080000, @@ -XXX,XX +XXX,XX @@ static void npcm7xx_init(Object *obj) object_initialize_child(obj, "pwm[*]", &s->pwm[i], TYPE_NPCM7XX_PWM); } + for (i = 0; i < ARRAY_SIZE(s->mft); i++) { + object_initialize_child(obj, "mft[*]", &s->mft[i], TYPE_NPCM7XX_MFT); + } + for (i = 0; i < ARRAY_SIZE(s->emc); i++) { object_initialize_child(obj, "emc[*]", &s->emc[i], TYPE_NPCM7XX_EMC); } @@ -XXX,XX +XXX,XX @@ static void npcm7xx_realize(DeviceState *dev, Error **errp) sysbus_connect_irq(sbd, i, npcm7xx_irq(s, NPCM7XX_PWM0_IRQ + i)); } + /* MFT Modules. Cannot fail. */ + QEMU_BUILD_BUG_ON(ARRAY_SIZE(npcm7xx_mft_addr) != ARRAY_SIZE(s->mft)); + for (i = 0; i < ARRAY_SIZE(s->mft); i++) { + SysBusDevice *sbd = SYS_BUS_DEVICE(&s->mft[i]); + + qdev_connect_clock_in(DEVICE(&s->mft[i]), "clock-in", + qdev_get_clock_out(DEVICE(&s->clk), + "apb4-clock")); + sysbus_realize(sbd, &error_abort); + sysbus_mmio_map(sbd, 0, npcm7xx_mft_addr[i]); + sysbus_connect_irq(sbd, 0, npcm7xx_irq(s, NPCM7XX_MFT0_IRQ + i)); + } + /* * EMC Modules. Cannot fail. * The mapping of the device to its netdev backend works as follows: @@ -XXX,XX +XXX,XX @@ static void npcm7xx_realize(DeviceState *dev, Error **errp) create_unimplemented_device("npcm7xx.peci", 0xf0100000, 4 * KiB); create_unimplemented_device("npcm7xx.siox[1]", 0xf0101000, 4 * KiB); create_unimplemented_device("npcm7xx.siox[2]", 0xf0102000, 4 * KiB); - create_unimplemented_device("npcm7xx.mft[0]", 0xf0180000, 4 * KiB); - create_unimplemented_device("npcm7xx.mft[1]", 0xf0181000, 4 * KiB); - create_unimplemented_device("npcm7xx.mft[2]", 0xf0182000, 4 * KiB); - create_unimplemented_device("npcm7xx.mft[3]", 0xf0183000, 4 * KiB); - create_unimplemented_device("npcm7xx.mft[4]", 0xf0184000, 4 * KiB); - create_unimplemented_device("npcm7xx.mft[5]", 0xf0185000, 4 * KiB); - create_unimplemented_device("npcm7xx.mft[6]", 0xf0186000, 4 * KiB); - create_unimplemented_device("npcm7xx.mft[7]", 0xf0187000, 4 * KiB); create_unimplemented_device("npcm7xx.pspi1", 0xf0200000, 4 * KiB); create_unimplemented_device("npcm7xx.pspi2", 0xf0201000, 4 * KiB); create_unimplemented_device("npcm7xx.ahbpci", 0xf0400000, 1 * MiB); -- 2.20.1 From: Hao Wu <wuhaotsh@google.com> This patch adds fan_splitters (split IRQs) in NPCM7XX boards. Each fan splitter corresponds to 1 PWM output and can connect to multiple fan inputs (MFT devices). In NPCM7XX boards(NPCM750 EVB and Quanta GSJ boards), we initializes these splitters and connect them to their corresponding modules according their specific device trees. Reviewed-by: Doug Evans <dje@google.com> Reviewed-by: Tyrone Ting <kfting@nuvoton.com> Signed-off-by: Hao Wu <wuhaotsh@google.com> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 20210311180855.149764-5-wuhaotsh@google.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/npcm7xx.h | 11 ++++- hw/arm/npcm7xx_boards.c | 99 ++++++++++++++++++++++++++++++++++++++++ 2 files changed, 109 insertions(+), 1 deletion(-) diff --git a/include/hw/arm/npcm7xx.h b/include/hw/arm/npcm7xx.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/npcm7xx.h +++ b/include/hw/arm/npcm7xx.h @@ -XXX,XX +XXX,XX @@ #include "hw/boards.h" #include "hw/adc/npcm7xx_adc.h" +#include "hw/core/split-irq.h" #include "hw/cpu/a9mpcore.h" #include "hw/gpio/npcm7xx_gpio.h" #include "hw/i2c/npcm7xx_smbus.h" @@ -XXX,XX +XXX,XX @@ #define NPCM7XX_GIC_CPU_IF_ADDR (0xf03fe100) /* GIC within A9 */ #define NPCM7XX_BOARD_SETUP_ADDR (0xffff1000) /* Boot ROM */ +#define NPCM7XX_NR_PWM_MODULES 2 + typedef struct NPCM7xxMachine { MachineState parent; + /* + * PWM fan splitter. each splitter connects to one PWM output and + * multiple MFT inputs. + */ + SplitIRQ fan_splitter[NPCM7XX_NR_PWM_MODULES * + NPCM7XX_PWM_PER_MODULE]; } NPCM7xxMachine; #define TYPE_NPCM7XX_MACHINE MACHINE_TYPE_NAME("npcm7xx") @@ -XXX,XX +XXX,XX @@ typedef struct NPCM7xxState { NPCM7xxCLKState clk; NPCM7xxTimerCtrlState tim[3]; NPCM7xxADCState adc; - NPCM7xxPWMState pwm[2]; + NPCM7xxPWMState pwm[NPCM7XX_NR_PWM_MODULES]; NPCM7xxMFTState mft[8]; NPCM7xxOTPState key_storage; NPCM7xxOTPState fuse_array; diff --git a/hw/arm/npcm7xx_boards.c b/hw/arm/npcm7xx_boards.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/npcm7xx_boards.c +++ b/hw/arm/npcm7xx_boards.c @@ -XXX,XX +XXX,XX @@ #include "hw/core/cpu.h" #include "hw/i2c/smbus_eeprom.h" #include "hw/loader.h" +#include "hw/qdev-core.h" #include "hw/qdev-properties.h" #include "qapi/error.h" #include "qemu-common.h" @@ -XXX,XX +XXX,XX @@ static void at24c_eeprom_init(NPCM7xxState *soc, int bus, uint8_t addr, i2c_slave_realize_and_unref(i2c_dev, i2c_bus, &error_abort); } +static void npcm7xx_init_pwm_splitter(NPCM7xxMachine *machine, + NPCM7xxState *soc, const int *fan_counts) +{ + SplitIRQ *splitters = machine->fan_splitter; + + /* + * PWM 0~3 belong to module 0 output 0~3. + * PWM 4~7 belong to module 1 output 0~3. + */ + for (int i = 0; i < NPCM7XX_NR_PWM_MODULES; ++i) { + for (int j = 0; j < NPCM7XX_PWM_PER_MODULE; ++j) { + int splitter_no = i * NPCM7XX_PWM_PER_MODULE + j; + DeviceState *splitter; + + if (fan_counts[splitter_no] < 1) { + continue; + } + object_initialize_child(OBJECT(machine), "fan-splitter[*]", + &splitters[splitter_no], TYPE_SPLIT_IRQ); + splitter = DEVICE(&splitters[splitter_no]); + qdev_prop_set_uint16(splitter, "num-lines", + fan_counts[splitter_no]); + qdev_realize(splitter, NULL, &error_abort); + qdev_connect_gpio_out_named(DEVICE(&soc->pwm[i]), "duty-gpio-out", + j, qdev_get_gpio_in(splitter, 0)); + } + } +} + +static void npcm7xx_connect_pwm_fan(NPCM7xxState *soc, SplitIRQ *splitter, + int fan_no, int output_no) +{ + DeviceState *fan; + int fan_input; + qemu_irq fan_duty_gpio; + + g_assert(fan_no >= 0 && fan_no <= NPCM7XX_MFT_MAX_FAN_INPUT); + /* + * Fan 0~1 belong to module 0 input 0~1. + * Fan 2~3 belong to module 1 input 0~1. + * ... + * Fan 14~15 belong to module 7 input 0~1. + * Fan 16~17 belong to module 0 input 2~3. + * Fan 18~19 belong to module 1 input 2~3. + */ + if (fan_no < 16) { + fan = DEVICE(&soc->mft[fan_no / 2]); + fan_input = fan_no % 2; + } else { + fan = DEVICE(&soc->mft[(fan_no - 16) / 2]); + fan_input = fan_no % 2 + 2; + } + + /* Connect the Fan to PWM module */ + fan_duty_gpio = qdev_get_gpio_in_named(fan, "duty", fan_input); + qdev_connect_gpio_out(DEVICE(splitter), output_no, fan_duty_gpio); +} + static void npcm750_evb_i2c_init(NPCM7xxState *soc) { /* lm75 temperature sensor on SVB, tmp105 is compatible */ @@ -XXX,XX +XXX,XX @@ static void npcm750_evb_i2c_init(NPCM7xxState *soc) i2c_slave_create_simple(npcm7xx_i2c_get_bus(soc, 6), "tmp105", 0x48); } +static void npcm750_evb_fan_init(NPCM7xxMachine *machine, NPCM7xxState *soc) +{ + SplitIRQ *splitter = machine->fan_splitter; + static const int fan_counts[] = {2, 2, 2, 2, 2, 2, 2, 2}; + + npcm7xx_init_pwm_splitter(machine, soc, fan_counts); + npcm7xx_connect_pwm_fan(soc, &splitter[0], 0x00, 0); + npcm7xx_connect_pwm_fan(soc, &splitter[0], 0x01, 1); + npcm7xx_connect_pwm_fan(soc, &splitter[1], 0x02, 0); + npcm7xx_connect_pwm_fan(soc, &splitter[1], 0x03, 1); + npcm7xx_connect_pwm_fan(soc, &splitter[2], 0x04, 0); + npcm7xx_connect_pwm_fan(soc, &splitter[2], 0x05, 1); + npcm7xx_connect_pwm_fan(soc, &splitter[3], 0x06, 0); + npcm7xx_connect_pwm_fan(soc, &splitter[3], 0x07, 1); + npcm7xx_connect_pwm_fan(soc, &splitter[4], 0x08, 0); + npcm7xx_connect_pwm_fan(soc, &splitter[4], 0x09, 1); + npcm7xx_connect_pwm_fan(soc, &splitter[5], 0x0a, 0); + npcm7xx_connect_pwm_fan(soc, &splitter[5], 0x0b, 1); + npcm7xx_connect_pwm_fan(soc, &splitter[6], 0x0c, 0); + npcm7xx_connect_pwm_fan(soc, &splitter[6], 0x0d, 1); + npcm7xx_connect_pwm_fan(soc, &splitter[7], 0x0e, 0); + npcm7xx_connect_pwm_fan(soc, &splitter[7], 0x0f, 1); +} + static void quanta_gsj_i2c_init(NPCM7xxState *soc) { /* GSJ machine have 4 max31725 temperature sensors, tmp105 is compatible. */ @@ -XXX,XX +XXX,XX @@ static void quanta_gsj_i2c_init(NPCM7xxState *soc) /* TODO: Add additional i2c devices. */ } +static void quanta_gsj_fan_init(NPCM7xxMachine *machine, NPCM7xxState *soc) +{ + SplitIRQ *splitter = machine->fan_splitter; + static const int fan_counts[] = {2, 2, 2, 0, 0, 0, 0, 0}; + + npcm7xx_init_pwm_splitter(machine, soc, fan_counts); + npcm7xx_connect_pwm_fan(soc, &splitter[0], 0x00, 0); + npcm7xx_connect_pwm_fan(soc, &splitter[0], 0x01, 1); + npcm7xx_connect_pwm_fan(soc, &splitter[1], 0x02, 0); + npcm7xx_connect_pwm_fan(soc, &splitter[1], 0x03, 1); + npcm7xx_connect_pwm_fan(soc, &splitter[2], 0x04, 0); + npcm7xx_connect_pwm_fan(soc, &splitter[2], 0x05, 1); +} + static void npcm750_evb_init(MachineState *machine) { NPCM7xxState *soc; @@ -XXX,XX +XXX,XX @@ static void npcm750_evb_init(MachineState *machine) npcm7xx_load_bootrom(machine, soc); npcm7xx_connect_flash(&soc->fiu[0], 0, "w25q256", drive_get(IF_MTD, 0, 0)); npcm750_evb_i2c_init(soc); + npcm750_evb_fan_init(NPCM7XX_MACHINE(machine), soc); npcm7xx_load_kernel(machine, soc); } @@ -XXX,XX +XXX,XX @@ static void quanta_gsj_init(MachineState *machine) npcm7xx_connect_flash(&soc->fiu[0], 0, "mx25l25635e", drive_get(IF_MTD, 0, 0)); quanta_gsj_i2c_init(soc); + quanta_gsj_fan_init(NPCM7XX_MACHINE(machine), soc); npcm7xx_load_kernel(machine, soc); } -- 2.20.1 From: Hao Wu <wuhaotsh@google.com> This patch adds testing of PWM fan RPMs in the existing npcm7xx pwm test. It tests whether the MFT module can measure correct fan values for a PWM fan in NPCM7XX boards. Reviewed-by: Doug Evans <dje@google.com> Reviewed-by: Tyrone Ting <kfting@nuvoton.com> Signed-off-by: Hao Wu <wuhaotsh@google.com> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 20210311180855.149764-6-wuhaotsh@google.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/qtest/npcm7xx_pwm-test.c | 205 ++++++++++++++++++++++++++++++++- 1 file changed, 199 insertions(+), 6 deletions(-) diff --git a/tests/qtest/npcm7xx_pwm-test.c b/tests/qtest/npcm7xx_pwm-test.c index XXXXXXX..XXXXXXX 100644 --- a/tests/qtest/npcm7xx_pwm-test.c +++ b/tests/qtest/npcm7xx_pwm-test.c @@ -XXX,XX +XXX,XX @@ #define PLL_FBDV(rv) extract32((rv), 16, 12) #define PLL_OTDV1(rv) extract32((rv), 8, 3) #define PLL_OTDV2(rv) extract32((rv), 13, 3) +#define APB4CKDIV(rv) extract32((rv), 30, 2) #define APB3CKDIV(rv) extract32((rv), 28, 2) #define CLK2CKDIV(rv) extract32((rv), 0, 1) #define CLK4CKDIV(rv) extract32((rv), 26, 2) @@ -XXX,XX +XXX,XX @@ #define MAX_DUTY 1000000 +/* MFT (PWM fan) related */ +#define MFT_BA(n) (0xf0180000 + ((n) * 0x1000)) +#define MFT_IRQ(n) (96 + (n)) +#define MFT_CNT1 0x00 +#define MFT_CRA 0x02 +#define MFT_CRB 0x04 +#define MFT_CNT2 0x06 +#define MFT_PRSC 0x08 +#define MFT_CKC 0x0a +#define MFT_MCTRL 0x0c +#define MFT_ICTRL 0x0e +#define MFT_ICLR 0x10 +#define MFT_IEN 0x12 +#define MFT_CPA 0x14 +#define MFT_CPB 0x16 +#define MFT_CPCFG 0x18 +#define MFT_INASEL 0x1a +#define MFT_INBSEL 0x1c + +#define MFT_MCTRL_ALL 0x64 +#define MFT_ICLR_ALL 0x3f +#define MFT_IEN_ALL 0x3f +#define MFT_CPCFG_EQ_MODE 0x44 + +#define MFT_CKC_C2CSEL BIT(3) +#define MFT_CKC_C1CSEL BIT(0) + +#define MFT_ICTRL_TFPND BIT(5) +#define MFT_ICTRL_TEPND BIT(4) +#define MFT_ICTRL_TDPND BIT(3) +#define MFT_ICTRL_TCPND BIT(2) +#define MFT_ICTRL_TBPND BIT(1) +#define MFT_ICTRL_TAPND BIT(0) + +#define MFT_MAX_CNT 0xffff +#define MFT_TIMEOUT 0x5000 + +#define DEFAULT_RPM 19800 +#define DEFAULT_PRSC 255 +#define MFT_PULSE_PER_REVOLUTION 2 + +#define MAX_ERROR 1 + typedef struct PWMModule { int irq; uint64_t base_addr; @@ -XXX,XX +XXX,XX @@ static uint64_t pwm_get_duty(QTestState *qts, int module_index, int pwm_index) return pwm_qom_get(qts, path, name); } +static void mft_qom_set(QTestState *qts, int index, const char *name, + uint32_t value) +{ + QDict *response; + char *path = g_strdup_printf("/machine/soc/mft[%d]", index); + + g_test_message("Setting properties %s of mft[%d] with value %u", + name, index, value); + response = qtest_qmp(qts, "{ 'execute': 'qom-set'," + " 'arguments': { 'path': %s, " + " 'property': %s, 'value': %u}}", + path, name, value); + /* The qom set message returns successfully. */ + g_assert_true(qdict_haskey(response, "return")); +} + static uint32_t get_pll(uint32_t con) { return REF_HZ * PLL_FBDV(con) / (PLL_INDV(con) * PLL_OTDV1(con) * PLL_OTDV2(con)); } -static uint64_t read_pclk(QTestState *qts) +static uint64_t read_pclk(QTestState *qts, bool mft) { uint64_t freq = REF_HZ; uint32_t clksel = qtest_readl(qts, CLK_BA + CLKSEL); uint32_t pllcon; uint32_t clkdiv1 = qtest_readl(qts, CLK_BA + CLKDIV1); uint32_t clkdiv2 = qtest_readl(qts, CLK_BA + CLKDIV2); + uint32_t apbdiv = mft ? APB4CKDIV(clkdiv2) : APB3CKDIV(clkdiv2); switch (CPUCKSEL(clksel)) { case 0: @@ -XXX,XX +XXX,XX @@ static uint64_t read_pclk(QTestState *qts) g_assert_not_reached(); } - freq >>= (CLK2CKDIV(clkdiv1) + CLK4CKDIV(clkdiv1) + APB3CKDIV(clkdiv2)); + freq >>= (CLK2CKDIV(clkdiv1) + CLK4CKDIV(clkdiv1) + apbdiv); return freq; } @@ -XXX,XX +XXX,XX @@ static uint32_t pwm_selector(uint32_t csr) static uint64_t pwm_compute_freq(QTestState *qts, uint32_t ppr, uint32_t csr, uint32_t cnr) { - return read_pclk(qts) / ((ppr + 1) * pwm_selector(csr) * (cnr + 1)); + return read_pclk(qts, false) / ((ppr + 1) * pwm_selector(csr) * (cnr + 1)); } static uint64_t pwm_compute_duty(uint32_t cnr, uint32_t cmr, bool inverted) @@ -XXX,XX +XXX,XX @@ static void pwm_write(QTestState *qts, const TestData *td, unsigned offset, qtest_writel(qts, td->module->base_addr + offset, value); } +static uint8_t mft_readb(QTestState *qts, int index, unsigned offset) +{ + return qtest_readb(qts, MFT_BA(index) + offset); +} + +static uint16_t mft_readw(QTestState *qts, int index, unsigned offset) +{ + return qtest_readw(qts, MFT_BA(index) + offset); +} + +static void mft_writeb(QTestState *qts, int index, unsigned offset, + uint8_t value) +{ + qtest_writeb(qts, MFT_BA(index) + offset, value); +} + +static void mft_writew(QTestState *qts, int index, unsigned offset, + uint16_t value) +{ + return qtest_writew(qts, MFT_BA(index) + offset, value); +} + static uint32_t pwm_read_ppr(QTestState *qts, const TestData *td) { return extract32(pwm_read(qts, td, PPR), ppr_base[pwm_index(td->pwm)], 8); @@ -XXX,XX +XXX,XX @@ static void pwm_write_cmr(QTestState *qts, const TestData *td, uint32_t value) pwm_write(qts, td, td->pwm->cmr_offset, value); } +static int mft_compute_index(const TestData *td) +{ + int index = pwm_module_index(td->module) * ARRAY_SIZE(pwm_list) + + pwm_index(td->pwm); + + g_assert_cmpint(index, <, + ARRAY_SIZE(pwm_module_list) * ARRAY_SIZE(pwm_list)); + + return index; +} + +static void mft_reset_counters(QTestState *qts, int index) +{ + mft_writew(qts, index, MFT_CNT1, MFT_MAX_CNT); + mft_writew(qts, index, MFT_CNT2, MFT_MAX_CNT); + mft_writew(qts, index, MFT_CRA, MFT_MAX_CNT); + mft_writew(qts, index, MFT_CRB, MFT_MAX_CNT); + mft_writew(qts, index, MFT_CPA, MFT_MAX_CNT - MFT_TIMEOUT); + mft_writew(qts, index, MFT_CPB, MFT_MAX_CNT - MFT_TIMEOUT); +} + +static void mft_init(QTestState *qts, const TestData *td) +{ + int index = mft_compute_index(td); + + /* Enable everything */ + mft_writeb(qts, index, MFT_CKC, 0); + mft_writeb(qts, index, MFT_ICLR, MFT_ICLR_ALL); + mft_writeb(qts, index, MFT_MCTRL, MFT_MCTRL_ALL); + mft_writeb(qts, index, MFT_IEN, MFT_IEN_ALL); + mft_writeb(qts, index, MFT_INASEL, 0); + mft_writeb(qts, index, MFT_INBSEL, 0); + + /* Set cpcfg to use EQ mode, same as kernel driver */ + mft_writeb(qts, index, MFT_CPCFG, MFT_CPCFG_EQ_MODE); + + /* Write default counters, timeout and prescaler */ + mft_reset_counters(qts, index); + mft_writeb(qts, index, MFT_PRSC, DEFAULT_PRSC); + + /* Write default max rpm via QMP */ + mft_qom_set(qts, index, "max_rpm[0]", DEFAULT_RPM); + mft_qom_set(qts, index, "max_rpm[1]", DEFAULT_RPM); +} + +static int32_t mft_compute_cnt(uint32_t rpm, uint64_t clk) +{ + uint64_t cnt; + + if (rpm == 0) { + return -1; + } + + cnt = clk * 60 / ((DEFAULT_PRSC + 1) * rpm * MFT_PULSE_PER_REVOLUTION); + if (cnt >= MFT_TIMEOUT) { + return -1; + } + return MFT_MAX_CNT - cnt; +} + +static void mft_verify_rpm(QTestState *qts, const TestData *td, uint64_t duty) +{ + int index = mft_compute_index(td); + uint16_t cnt, cr; + uint32_t rpm = DEFAULT_RPM * duty / MAX_DUTY; + uint64_t clk = read_pclk(qts, true); + int32_t expected_cnt = mft_compute_cnt(rpm, clk); + + qtest_irq_intercept_in(qts, "/machine/soc/a9mpcore/gic"); + g_test_message( + "verifying rpm for mft[%d]: clk: %lu, duty: %lu, rpm: %u, cnt: %d", + index, clk, duty, rpm, expected_cnt); + + /* Verify rpm for fan A */ + /* Stop capture */ + mft_writeb(qts, index, MFT_CKC, 0); + mft_writeb(qts, index, MFT_ICLR, MFT_ICLR_ALL); + mft_reset_counters(qts, index); + g_assert_cmphex(mft_readw(qts, index, MFT_CNT1), ==, MFT_MAX_CNT); + g_assert_cmphex(mft_readw(qts, index, MFT_CRA), ==, MFT_MAX_CNT); + g_assert_cmphex(mft_readw(qts, index, MFT_CPA), ==, + MFT_MAX_CNT - MFT_TIMEOUT); + /* Start capture */ + mft_writeb(qts, index, MFT_CKC, MFT_CKC_C1CSEL); + g_assert_true(qtest_get_irq(qts, MFT_IRQ(index))); + if (expected_cnt == -1) { + g_assert_cmphex(mft_readb(qts, index, MFT_ICTRL), ==, MFT_ICTRL_TEPND); + } else { + g_assert_cmphex(mft_readb(qts, index, MFT_ICTRL), ==, MFT_ICTRL_TAPND); + cnt = mft_readw(qts, index, MFT_CNT1); + /* + * Due to error in clock measurement and rounding, we might have a small + * error in measuring RPM. + */ + g_assert_cmphex(cnt + MAX_ERROR, >=, expected_cnt); + g_assert_cmphex(cnt, <=, expected_cnt + MAX_ERROR); + cr = mft_readw(qts, index, MFT_CRA); + g_assert_cmphex(cnt, ==, cr); + } + + /* Verify rpm for fan B */ + + qtest_irq_intercept_out(qts, "/machine/soc/a9mpcore/gic"); +} + /* Check pwm registers can be reset to default value */ static void test_init(gconstpointer test_data) { const TestData *td = test_data; - QTestState *qts = qtest_init("-machine quanta-gsj"); + QTestState *qts = qtest_init("-machine npcm750-evb"); int module = pwm_module_index(td->module); int pwm = pwm_index(td->pwm); @@ -XXX,XX +XXX,XX @@ static void test_init(gconstpointer test_data) static void test_oneshot(gconstpointer test_data) { const TestData *td = test_data; - QTestState *qts = qtest_init("-machine quanta-gsj"); + QTestState *qts = qtest_init("-machine npcm750-evb"); int module = pwm_module_index(td->module); int pwm = pwm_index(td->pwm); uint32_t ppr, csr, pcr; @@ -XXX,XX +XXX,XX @@ static void test_oneshot(gconstpointer test_data) static void test_toggle(gconstpointer test_data) { const TestData *td = test_data; - QTestState *qts = qtest_init("-machine quanta-gsj"); + QTestState *qts = qtest_init("-machine npcm750-evb"); int module = pwm_module_index(td->module); int pwm = pwm_index(td->pwm); uint32_t ppr, csr, pcr, cnr, cmr; int i, j, k, l; uint64_t expected_freq, expected_duty; + mft_init(qts, td); + pcr = CH_EN | CH_MOD; for (i = 0; i < ARRAY_SIZE(ppr_list); ++i) { ppr = ppr_list[i]; @@ -XXX,XX +XXX,XX @@ static void test_toggle(gconstpointer test_data) ==, expected_freq); } + /* Test MFT's RPM is correct. */ + mft_verify_rpm(qts, td, expected_duty); + /* Test inverted mode */ expected_duty = pwm_compute_duty(cnr, cmr, true); pwm_write_pcr(qts, td, pcr | CH_INV); -- 2.20.1 For a long time now the UI layer has guaranteed that the console surface is always 32 bits per pixel. Remove the legacy dead code from the pl110 display device which was handling the possibility that the console surface was some other format. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Acked-by: Gerd Hoffmann <kraxel@redhat.com> Message-id: 20210211141515.8755-2-peter.maydell@linaro.org --- hw/display/pl110.c | 53 +++++++--------------------------------------- 1 file changed, 8 insertions(+), 45 deletions(-) diff --git a/hw/display/pl110.c b/hw/display/pl110.c index XXXXXXX..XXXXXXX 100644 --- a/hw/display/pl110.c +++ b/hw/display/pl110.c @@ -XXX,XX +XXX,XX @@ static const unsigned char *idregs[] = { pl111_id }; -#define BITS 8 -#include "pl110_template.h" -#define BITS 15 -#include "pl110_template.h" -#define BITS 16 -#include "pl110_template.h" -#define BITS 24 -#include "pl110_template.h" #define BITS 32 #include "pl110_template.h" @@ -XXX,XX +XXX,XX @@ static void pl110_update_display(void *opaque) PL110State *s = (PL110State *)opaque; SysBusDevice *sbd; DisplaySurface *surface = qemu_console_surface(s->con); - drawfn* fntable; drawfn fn; - int dest_width; int src_width; int bpp_offset; int first; @@ -XXX,XX +XXX,XX @@ static void pl110_update_display(void *opaque) sbd = SYS_BUS_DEVICE(s); - switch (surface_bits_per_pixel(surface)) { - case 0: - return; - case 8: - fntable = pl110_draw_fn_8; - dest_width = 1; - break; - case 15: - fntable = pl110_draw_fn_15; - dest_width = 2; - break; - case 16: - fntable = pl110_draw_fn_16; - dest_width = 2; - break; - case 24: - fntable = pl110_draw_fn_24; - dest_width = 3; - break; - case 32: - fntable = pl110_draw_fn_32; - dest_width = 4; - break; - default: - fprintf(stderr, "pl110: Bad color depth\n"); - exit(1); - } if (s->cr & PL110_CR_BGR) bpp_offset = 0; else @@ -XXX,XX +XXX,XX @@ static void pl110_update_display(void *opaque) } } - if (s->cr & PL110_CR_BEBO) - fn = fntable[s->bpp + 8 + bpp_offset]; - else if (s->cr & PL110_CR_BEPO) - fn = fntable[s->bpp + 16 + bpp_offset]; - else - fn = fntable[s->bpp + bpp_offset]; + if (s->cr & PL110_CR_BEBO) { + fn = pl110_draw_fn_32[s->bpp + 8 + bpp_offset]; + } else if (s->cr & PL110_CR_BEPO) { + fn = pl110_draw_fn_32[s->bpp + 16 + bpp_offset]; + } else { + fn = pl110_draw_fn_32[s->bpp + bpp_offset]; + } src_width = s->cols; switch (s->bpp) { @@ -XXX,XX +XXX,XX @@ static void pl110_update_display(void *opaque) src_width <<= 2; break; } - dest_width *= s->cols; first = 0; if (s->invalidate) { framebuffer_update_memory_section(&s->fbsection, @@ -XXX,XX +XXX,XX @@ static void pl110_update_display(void *opaque) framebuffer_update_display(surface, &s->fbsection, s->cols, s->rows, - src_width, dest_width, 0, + src_width, s->cols * 4, 0, s->invalidate, fn, s->palette, &first, &last); -- 2.20.1 The pl110_template.h header has a doubly-nested multiple-include pattern: * pl110.c includes it once for each host bit depth (now always 32) * every time it is included, it includes itself 6 times, to account for multiple guest device pixel and byte orders Now we only have to deal with 32-bit host bit depths, we can move the code corresponding to the outer layer of this double-nesting to be directly in pl110.c and reduce the template header to a single layer of nesting. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Acked-by: Gerd Hoffmann <kraxel@redhat.com> Message-id: 20210211141515.8755-3-peter.maydell@linaro.org --- hw/display/pl110_template.h | 100 +----------------------------------- hw/display/pl110.c | 79 ++++++++++++++++++++++++++++ 2 files changed, 80 insertions(+), 99 deletions(-) diff --git a/hw/display/pl110_template.h b/hw/display/pl110_template.h index XXXXXXX..XXXXXXX 100644 --- a/hw/display/pl110_template.h +++ b/hw/display/pl110_template.h @@ -XXX,XX +XXX,XX @@ */ #ifndef ORDER - -#if BITS == 8 -#define COPY_PIXEL(to, from) *(to++) = from -#elif BITS == 15 || BITS == 16 -#define COPY_PIXEL(to, from) do { *(uint16_t *)to = from; to += 2; } while (0) -#elif BITS == 24 -#define COPY_PIXEL(to, from) \ - do { \ - *(to++) = from; \ - *(to++) = (from) >> 8; \ - *(to++) = (from) >> 16; \ - } while (0) -#elif BITS == 32 -#define COPY_PIXEL(to, from) do { *(uint32_t *)to = from; to += 4; } while (0) -#else -#error unknown bit depth +#error "pl110_template.h is only for inclusion by pl110.c" #endif -#undef RGB -#define BORDER bgr -#define ORDER 0 -#include "pl110_template.h" -#define ORDER 1 -#include "pl110_template.h" -#define ORDER 2 -#include "pl110_template.h" -#undef BORDER -#define RGB -#define BORDER rgb -#define ORDER 0 -#include "pl110_template.h" -#define ORDER 1 -#include "pl110_template.h" -#define ORDER 2 -#include "pl110_template.h" -#undef BORDER - -static drawfn glue(pl110_draw_fn_,BITS)[48] = -{ - glue(pl110_draw_line1_lblp_bgr,BITS), - glue(pl110_draw_line2_lblp_bgr,BITS), - glue(pl110_draw_line4_lblp_bgr,BITS), - glue(pl110_draw_line8_lblp_bgr,BITS), - glue(pl110_draw_line16_555_lblp_bgr,BITS), - glue(pl110_draw_line32_lblp_bgr,BITS), - glue(pl110_draw_line16_lblp_bgr,BITS), - glue(pl110_draw_line12_lblp_bgr,BITS), - - glue(pl110_draw_line1_bbbp_bgr,BITS), - glue(pl110_draw_line2_bbbp_bgr,BITS), - glue(pl110_draw_line4_bbbp_bgr,BITS), - glue(pl110_draw_line8_bbbp_bgr,BITS), - glue(pl110_draw_line16_555_bbbp_bgr,BITS), - glue(pl110_draw_line32_bbbp_bgr,BITS), - glue(pl110_draw_line16_bbbp_bgr,BITS), - glue(pl110_draw_line12_bbbp_bgr,BITS), - - glue(pl110_draw_line1_lbbp_bgr,BITS), - glue(pl110_draw_line2_lbbp_bgr,BITS), - glue(pl110_draw_line4_lbbp_bgr,BITS), - glue(pl110_draw_line8_lbbp_bgr,BITS), - glue(pl110_draw_line16_555_lbbp_bgr,BITS), - glue(pl110_draw_line32_lbbp_bgr,BITS), - glue(pl110_draw_line16_lbbp_bgr,BITS), - glue(pl110_draw_line12_lbbp_bgr,BITS), - - glue(pl110_draw_line1_lblp_rgb,BITS), - glue(pl110_draw_line2_lblp_rgb,BITS), - glue(pl110_draw_line4_lblp_rgb,BITS), - glue(pl110_draw_line8_lblp_rgb,BITS), - glue(pl110_draw_line16_555_lblp_rgb,BITS), - glue(pl110_draw_line32_lblp_rgb,BITS), - glue(pl110_draw_line16_lblp_rgb,BITS), - glue(pl110_draw_line12_lblp_rgb,BITS), - - glue(pl110_draw_line1_bbbp_rgb,BITS), - glue(pl110_draw_line2_bbbp_rgb,BITS), - glue(pl110_draw_line4_bbbp_rgb,BITS), - glue(pl110_draw_line8_bbbp_rgb,BITS), - glue(pl110_draw_line16_555_bbbp_rgb,BITS), - glue(pl110_draw_line32_bbbp_rgb,BITS), - glue(pl110_draw_line16_bbbp_rgb,BITS), - glue(pl110_draw_line12_bbbp_rgb,BITS), - - glue(pl110_draw_line1_lbbp_rgb,BITS), - glue(pl110_draw_line2_lbbp_rgb,BITS), - glue(pl110_draw_line4_lbbp_rgb,BITS), - glue(pl110_draw_line8_lbbp_rgb,BITS), - glue(pl110_draw_line16_555_lbbp_rgb,BITS), - glue(pl110_draw_line32_lbbp_rgb,BITS), - glue(pl110_draw_line16_lbbp_rgb,BITS), - glue(pl110_draw_line12_lbbp_rgb,BITS), -}; - -#undef BITS -#undef COPY_PIXEL - -#else - #if ORDER == 0 #define NAME glue(glue(lblp_, BORDER), BITS) #ifdef HOST_WORDS_BIGENDIAN @@ -XXX,XX +XXX,XX @@ static void glue(pl110_draw_line12_,NAME)(void *opaque, uint8_t *d, const uint8_ #undef NAME #undef SWAP_WORDS #undef ORDER - -#endif diff --git a/hw/display/pl110.c b/hw/display/pl110.c index XXXXXXX..XXXXXXX 100644 --- a/hw/display/pl110.c +++ b/hw/display/pl110.c @@ -XXX,XX +XXX,XX @@ static const unsigned char *idregs[] = { }; #define BITS 32 +#define COPY_PIXEL(to, from) do { *(uint32_t *)to = from; to += 4; } while (0) + +#undef RGB +#define BORDER bgr +#define ORDER 0 #include "pl110_template.h" +#define ORDER 1 +#include "pl110_template.h" +#define ORDER 2 +#include "pl110_template.h" +#undef BORDER +#define RGB +#define BORDER rgb +#define ORDER 0 +#include "pl110_template.h" +#define ORDER 1 +#include "pl110_template.h" +#define ORDER 2 +#include "pl110_template.h" +#undef BORDER + +static drawfn pl110_draw_fn_32[48] = { + pl110_draw_line1_lblp_bgr32, + pl110_draw_line2_lblp_bgr32, + pl110_draw_line4_lblp_bgr32, + pl110_draw_line8_lblp_bgr32, + pl110_draw_line16_555_lblp_bgr32, + pl110_draw_line32_lblp_bgr32, + pl110_draw_line16_lblp_bgr32, + pl110_draw_line12_lblp_bgr32, + + pl110_draw_line1_bbbp_bgr32, + pl110_draw_line2_bbbp_bgr32, + pl110_draw_line4_bbbp_bgr32, + pl110_draw_line8_bbbp_bgr32, + pl110_draw_line16_555_bbbp_bgr32, + pl110_draw_line32_bbbp_bgr32, + pl110_draw_line16_bbbp_bgr32, + pl110_draw_line12_bbbp_bgr32, + + pl110_draw_line1_lbbp_bgr32, + pl110_draw_line2_lbbp_bgr32, + pl110_draw_line4_lbbp_bgr32, + pl110_draw_line8_lbbp_bgr32, + pl110_draw_line16_555_lbbp_bgr32, + pl110_draw_line32_lbbp_bgr32, + pl110_draw_line16_lbbp_bgr32, + pl110_draw_line12_lbbp_bgr32, + + pl110_draw_line1_lblp_rgb32, + pl110_draw_line2_lblp_rgb32, + pl110_draw_line4_lblp_rgb32, + pl110_draw_line8_lblp_rgb32, + pl110_draw_line16_555_lblp_rgb32, + pl110_draw_line32_lblp_rgb32, + pl110_draw_line16_lblp_rgb32, + pl110_draw_line12_lblp_rgb32, + + pl110_draw_line1_bbbp_rgb32, + pl110_draw_line2_bbbp_rgb32, + pl110_draw_line4_bbbp_rgb32, + pl110_draw_line8_bbbp_rgb32, + pl110_draw_line16_555_bbbp_rgb32, + pl110_draw_line32_bbbp_rgb32, + pl110_draw_line16_bbbp_rgb32, + pl110_draw_line12_bbbp_rgb32, + + pl110_draw_line1_lbbp_rgb32, + pl110_draw_line2_lbbp_rgb32, + pl110_draw_line4_lbbp_rgb32, + pl110_draw_line8_lbbp_rgb32, + pl110_draw_line16_555_lbbp_rgb32, + pl110_draw_line32_lbbp_rgb32, + pl110_draw_line16_lbbp_rgb32, + pl110_draw_line12_lbbp_rgb32, +}; + +#undef BITS +#undef COPY_PIXEL + static int pl110_enabled(PL110State *s) { -- 2.20.1 BITS is always 32, so remove all uses of it from the template header, by dropping the trailing '32' from the draw function names and not constructing the name of rgb_to_pixel32() via the glue() macro. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Acked-by: Gerd Hoffmann <kraxel@redhat.com> Message-id: 20210211141515.8755-4-peter.maydell@linaro.org --- hw/display/pl110_template.h | 20 +++---- hw/display/pl110.c | 113 ++++++++++++++++++------------------ 2 files changed, 65 insertions(+), 68 deletions(-) diff --git a/hw/display/pl110_template.h b/hw/display/pl110_template.h index XXXXXXX..XXXXXXX 100644 --- a/hw/display/pl110_template.h +++ b/hw/display/pl110_template.h @@ -XXX,XX +XXX,XX @@ #endif #if ORDER == 0 -#define NAME glue(glue(lblp_, BORDER), BITS) +#define NAME glue(lblp_, BORDER) #ifdef HOST_WORDS_BIGENDIAN #define SWAP_WORDS 1 #endif #elif ORDER == 1 -#define NAME glue(glue(bbbp_, BORDER), BITS) +#define NAME glue(bbbp_, BORDER) #ifndef HOST_WORDS_BIGENDIAN #define SWAP_WORDS 1 #endif #else #define SWAP_PIXELS 1 -#define NAME glue(glue(lbbp_, BORDER), BITS) +#define NAME glue(lbbp_, BORDER) #ifdef HOST_WORDS_BIGENDIAN #define SWAP_WORDS 1 #endif @@ -XXX,XX +XXX,XX @@ static void glue(pl110_draw_line16_,NAME)(void *opaque, uint8_t *d, const uint8_ MSB = (data & 0x1f) << 3; data >>= 5; #endif - COPY_PIXEL(d, glue(rgb_to_pixel,BITS)(r, g, b)); + COPY_PIXEL(d, rgb_to_pixel32(r, g, b)); LSB = (data & 0x1f) << 3; data >>= 5; g = (data & 0x3f) << 2; data >>= 6; MSB = (data & 0x1f) << 3; data >>= 5; - COPY_PIXEL(d, glue(rgb_to_pixel,BITS)(r, g, b)); + COPY_PIXEL(d, rgb_to_pixel32(r, g, b)); #undef MSB #undef LSB width -= 2; @@ -XXX,XX +XXX,XX @@ static void glue(pl110_draw_line32_,NAME)(void *opaque, uint8_t *d, const uint8_ g = (data >> 16) & 0xff; MSB = (data >> 8) & 0xff; #endif - COPY_PIXEL(d, glue(rgb_to_pixel,BITS)(r, g, b)); + COPY_PIXEL(d, rgb_to_pixel32(r, g, b)); #undef MSB #undef LSB width--; @@ -XXX,XX +XXX,XX @@ static void glue(pl110_draw_line16_555_,NAME)(void *opaque, uint8_t *d, const ui data >>= 5; MSB = (data & 0x1f) << 3; data >>= 5; - COPY_PIXEL(d, glue(rgb_to_pixel,BITS)(r, g, b)); + COPY_PIXEL(d, rgb_to_pixel32(r, g, b)); LSB = (data & 0x1f) << 3; data >>= 5; g = (data & 0x1f) << 3; data >>= 5; MSB = (data & 0x1f) << 3; data >>= 6; - COPY_PIXEL(d, glue(rgb_to_pixel,BITS)(r, g, b)); + COPY_PIXEL(d, rgb_to_pixel32(r, g, b)); #undef MSB #undef LSB width -= 2; @@ -XXX,XX +XXX,XX @@ static void glue(pl110_draw_line12_,NAME)(void *opaque, uint8_t *d, const uint8_ data >>= 4; MSB = (data & 0xf) << 4; data >>= 8; - COPY_PIXEL(d, glue(rgb_to_pixel,BITS)(r, g, b)); + COPY_PIXEL(d, rgb_to_pixel32(r, g, b)); LSB = (data & 0xf) << 4; data >>= 4; g = (data & 0xf) << 4; data >>= 4; MSB = (data & 0xf) << 4; data >>= 8; - COPY_PIXEL(d, glue(rgb_to_pixel,BITS)(r, g, b)); + COPY_PIXEL(d, rgb_to_pixel32(r, g, b)); #undef MSB #undef LSB width -= 2; diff --git a/hw/display/pl110.c b/hw/display/pl110.c index XXXXXXX..XXXXXXX 100644 --- a/hw/display/pl110.c +++ b/hw/display/pl110.c @@ -XXX,XX +XXX,XX @@ static const unsigned char *idregs[] = { pl111_id }; -#define BITS 32 #define COPY_PIXEL(to, from) do { *(uint32_t *)to = from; to += 4; } while (0) #undef RGB @@ -XXX,XX +XXX,XX @@ static const unsigned char *idregs[] = { #include "pl110_template.h" #undef BORDER -static drawfn pl110_draw_fn_32[48] = { - pl110_draw_line1_lblp_bgr32, - pl110_draw_line2_lblp_bgr32, - pl110_draw_line4_lblp_bgr32, - pl110_draw_line8_lblp_bgr32, - pl110_draw_line16_555_lblp_bgr32, - pl110_draw_line32_lblp_bgr32, - pl110_draw_line16_lblp_bgr32, - pl110_draw_line12_lblp_bgr32, - - pl110_draw_line1_bbbp_bgr32, - pl110_draw_line2_bbbp_bgr32, - pl110_draw_line4_bbbp_bgr32, - pl110_draw_line8_bbbp_bgr32, - pl110_draw_line16_555_bbbp_bgr32, - pl110_draw_line32_bbbp_bgr32, - pl110_draw_line16_bbbp_bgr32, - pl110_draw_line12_bbbp_bgr32, - - pl110_draw_line1_lbbp_bgr32, - pl110_draw_line2_lbbp_bgr32, - pl110_draw_line4_lbbp_bgr32, - pl110_draw_line8_lbbp_bgr32, - pl110_draw_line16_555_lbbp_bgr32, - pl110_draw_line32_lbbp_bgr32, - pl110_draw_line16_lbbp_bgr32, - pl110_draw_line12_lbbp_bgr32, - - pl110_draw_line1_lblp_rgb32, - pl110_draw_line2_lblp_rgb32, - pl110_draw_line4_lblp_rgb32, - pl110_draw_line8_lblp_rgb32, - pl110_draw_line16_555_lblp_rgb32, - pl110_draw_line32_lblp_rgb32, - pl110_draw_line16_lblp_rgb32, - pl110_draw_line12_lblp_rgb32, - - pl110_draw_line1_bbbp_rgb32, - pl110_draw_line2_bbbp_rgb32, - pl110_draw_line4_bbbp_rgb32, - pl110_draw_line8_bbbp_rgb32, - pl110_draw_line16_555_bbbp_rgb32, - pl110_draw_line32_bbbp_rgb32, - pl110_draw_line16_bbbp_rgb32, - pl110_draw_line12_bbbp_rgb32, - - pl110_draw_line1_lbbp_rgb32, - pl110_draw_line2_lbbp_rgb32, - pl110_draw_line4_lbbp_rgb32, - pl110_draw_line8_lbbp_rgb32, - pl110_draw_line16_555_lbbp_rgb32, - pl110_draw_line32_lbbp_rgb32, - pl110_draw_line16_lbbp_rgb32, - pl110_draw_line12_lbbp_rgb32, -}; - -#undef BITS #undef COPY_PIXEL +static drawfn pl110_draw_fn_32[48] = { + pl110_draw_line1_lblp_bgr, + pl110_draw_line2_lblp_bgr, + pl110_draw_line4_lblp_bgr, + pl110_draw_line8_lblp_bgr, + pl110_draw_line16_555_lblp_bgr, + pl110_draw_line32_lblp_bgr, + pl110_draw_line16_lblp_bgr, + pl110_draw_line12_lblp_bgr, + + pl110_draw_line1_bbbp_bgr, + pl110_draw_line2_bbbp_bgr, + pl110_draw_line4_bbbp_bgr, + pl110_draw_line8_bbbp_bgr, + pl110_draw_line16_555_bbbp_bgr, + pl110_draw_line32_bbbp_bgr, + pl110_draw_line16_bbbp_bgr, + pl110_draw_line12_bbbp_bgr, + + pl110_draw_line1_lbbp_bgr, + pl110_draw_line2_lbbp_bgr, + pl110_draw_line4_lbbp_bgr, + pl110_draw_line8_lbbp_bgr, + pl110_draw_line16_555_lbbp_bgr, + pl110_draw_line32_lbbp_bgr, + pl110_draw_line16_lbbp_bgr, + pl110_draw_line12_lbbp_bgr, + + pl110_draw_line1_lblp_rgb, + pl110_draw_line2_lblp_rgb, + pl110_draw_line4_lblp_rgb, + pl110_draw_line8_lblp_rgb, + pl110_draw_line16_555_lblp_rgb, + pl110_draw_line32_lblp_rgb, + pl110_draw_line16_lblp_rgb, + pl110_draw_line12_lblp_rgb, + + pl110_draw_line1_bbbp_rgb, + pl110_draw_line2_bbbp_rgb, + pl110_draw_line4_bbbp_rgb, + pl110_draw_line8_bbbp_rgb, + pl110_draw_line16_555_bbbp_rgb, + pl110_draw_line32_bbbp_rgb, + pl110_draw_line16_bbbp_rgb, + pl110_draw_line12_bbbp_rgb, + + pl110_draw_line1_lbbp_rgb, + pl110_draw_line2_lbbp_rgb, + pl110_draw_line4_lbbp_rgb, + pl110_draw_line8_lbbp_rgb, + pl110_draw_line16_555_lbbp_rgb, + pl110_draw_line32_lbbp_rgb, + pl110_draw_line16_lbbp_rgb, + pl110_draw_line12_lbbp_rgb, +}; static int pl110_enabled(PL110State *s) { -- 2.20.1 For a long time now the UI layer has guaranteed that the console surface is always 32 bits per pixel. Remove the legacy dead code from the pxa2xx_lcd display device which was handling the possibility that the console surface was some other format. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Acked-by: Gerd Hoffmann <kraxel@redhat.com> Message-id: 20210211141515.8755-5-peter.maydell@linaro.org --- hw/display/pxa2xx_lcd.c | 79 +++++++++-------------------------------- 1 file changed, 17 insertions(+), 62 deletions(-) diff --git a/hw/display/pxa2xx_lcd.c b/hw/display/pxa2xx_lcd.c index XXXXXXX..XXXXXXX 100644 --- a/hw/display/pxa2xx_lcd.c +++ b/hw/display/pxa2xx_lcd.c @@ -XXX,XX +XXX,XX @@ struct PXA2xxLCDState { int invalidated; QemuConsole *con; - drawfn *line_fn[2]; int dest_width; int xres, yres; int pal_for; @@ -XXX,XX +XXX,XX @@ typedef struct QEMU_PACKED { #define LDCMD_SOFINT (1 << 22) #define LDCMD_PAL (1 << 26) +#define BITS 32 +#include "pxa2xx_template.h" + /* Route internal interrupt lines to the global IC */ static void pxa2xx_lcdc_int_update(PXA2xxLCDState *s) { @@ -XXX,XX +XXX,XX @@ static void pxa2xx_palette_parse(PXA2xxLCDState *s, int ch, int bpp) } } +static inline drawfn pxa2xx_drawfn(PXA2xxLCDState *s) +{ + if (s->transp) { + return pxa2xx_draw_fn_32t[s->bpp]; + } else { + return pxa2xx_draw_fn_32[s->bpp]; + } +} + static void pxa2xx_lcdc_dma0_redraw_rot0(PXA2xxLCDState *s, hwaddr addr, int *miny, int *maxy) { DisplaySurface *surface = qemu_console_surface(s->con); int src_width, dest_width; - drawfn fn = NULL; - if (s->dest_width) - fn = s->line_fn[s->transp][s->bpp]; + drawfn fn = pxa2xx_drawfn(s); if (!fn) return; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_lcdc_dma0_redraw_rot90(PXA2xxLCDState *s, { DisplaySurface *surface = qemu_console_surface(s->con); int src_width, dest_width; - drawfn fn = NULL; - if (s->dest_width) - fn = s->line_fn[s->transp][s->bpp]; + drawfn fn = pxa2xx_drawfn(s); if (!fn) return; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_lcdc_dma0_redraw_rot180(PXA2xxLCDState *s, { DisplaySurface *surface = qemu_console_surface(s->con); int src_width, dest_width; - drawfn fn = NULL; - if (s->dest_width) { - fn = s->line_fn[s->transp][s->bpp]; - } + drawfn fn = pxa2xx_drawfn(s); if (!fn) { return; } @@ -XXX,XX +XXX,XX @@ static void pxa2xx_lcdc_dma0_redraw_rot270(PXA2xxLCDState *s, { DisplaySurface *surface = qemu_console_surface(s->con); int src_width, dest_width; - drawfn fn = NULL; - if (s->dest_width) { - fn = s->line_fn[s->transp][s->bpp]; - } + drawfn fn = pxa2xx_drawfn(s); if (!fn) { return; } @@ -XXX,XX +XXX,XX @@ static const VMStateDescription vmstate_pxa2xx_lcdc = { } }; -#define BITS 8 -#include "pxa2xx_template.h" -#define BITS 15 -#include "pxa2xx_template.h" -#define BITS 16 -#include "pxa2xx_template.h" -#define BITS 24 -#include "pxa2xx_template.h" -#define BITS 32 -#include "pxa2xx_template.h" - static const GraphicHwOps pxa2xx_ops = { .invalidate = pxa2xx_invalidate_display, .gfx_update = pxa2xx_update_display, @@ -XXX,XX +XXX,XX @@ PXA2xxLCDState *pxa2xx_lcdc_init(MemoryRegion *sysmem, hwaddr base, qemu_irq irq) { PXA2xxLCDState *s; - DisplaySurface *surface; s = (PXA2xxLCDState *) g_malloc0(sizeof(PXA2xxLCDState)); s->invalidated = 1; @@ -XXX,XX +XXX,XX @@ PXA2xxLCDState *pxa2xx_lcdc_init(MemoryRegion *sysmem, memory_region_add_subregion(sysmem, base, &s->iomem); s->con = graphic_console_init(NULL, 0, &pxa2xx_ops, s); - surface = qemu_console_surface(s->con); - - switch (surface_bits_per_pixel(surface)) { - case 0: - s->dest_width = 0; - break; - case 8: - s->line_fn[0] = pxa2xx_draw_fn_8; - s->line_fn[1] = pxa2xx_draw_fn_8t; - s->dest_width = 1; - break; - case 15: - s->line_fn[0] = pxa2xx_draw_fn_15; - s->line_fn[1] = pxa2xx_draw_fn_15t; - s->dest_width = 2; - break; - case 16: - s->line_fn[0] = pxa2xx_draw_fn_16; - s->line_fn[1] = pxa2xx_draw_fn_16t; - s->dest_width = 2; - break; - case 24: - s->line_fn[0] = pxa2xx_draw_fn_24; - s->line_fn[1] = pxa2xx_draw_fn_24t; - s->dest_width = 3; - break; - case 32: - s->line_fn[0] = pxa2xx_draw_fn_32; - s->line_fn[1] = pxa2xx_draw_fn_32t; - s->dest_width = 4; - break; - default: - fprintf(stderr, "%s: Bad color depth\n", __func__); - exit(1); - } + s->dest_width = 4; vmstate_register(NULL, 0, &vmstate_pxa2xx_lcdc, s); -- 2.20.1 Since the dest_width is now always 4 because the output surface is 32bpp, we can replace the dest_width state field with a constant. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Acked-by: Gerd Hoffmann <kraxel@redhat.com> Message-id: 20210211141515.8755-6-peter.maydell@linaro.org --- hw/display/pxa2xx_lcd.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) diff --git a/hw/display/pxa2xx_lcd.c b/hw/display/pxa2xx_lcd.c index XXXXXXX..XXXXXXX 100644 --- a/hw/display/pxa2xx_lcd.c +++ b/hw/display/pxa2xx_lcd.c @@ -XXX,XX +XXX,XX @@ typedef struct QEMU_PACKED { #define LDCMD_SOFINT (1 << 22) #define LDCMD_PAL (1 << 26) +/* Size of a pixel in the QEMU UI output surface, in bytes */ +#define DEST_PIXEL_WIDTH 4 + #define BITS 32 #include "pxa2xx_template.h" @@ -XXX,XX +XXX,XX @@ static void pxa2xx_lcdc_dma0_redraw_rot0(PXA2xxLCDState *s, else if (s->bpp > pxa_lcdc_8bpp) src_width *= 2; - dest_width = s->xres * s->dest_width; + dest_width = s->xres * DEST_PIXEL_WIDTH; *miny = 0; if (s->invalidated) { framebuffer_update_memory_section(&s->fbsection, s->sysmem, addr, s->yres, src_width); } framebuffer_update_display(surface, &s->fbsection, s->xres, s->yres, - src_width, dest_width, s->dest_width, + src_width, dest_width, DEST_PIXEL_WIDTH, s->invalidated, fn, s->dma_ch[0].palette, miny, maxy); } @@ -XXX,XX +XXX,XX @@ static void pxa2xx_lcdc_dma0_redraw_rot90(PXA2xxLCDState *s, else if (s->bpp > pxa_lcdc_8bpp) src_width *= 2; - dest_width = s->yres * s->dest_width; + dest_width = s->yres * DEST_PIXEL_WIDTH; *miny = 0; if (s->invalidated) { framebuffer_update_memory_section(&s->fbsection, s->sysmem, addr, s->yres, src_width); } framebuffer_update_display(surface, &s->fbsection, s->xres, s->yres, - src_width, s->dest_width, -dest_width, + src_width, DEST_PIXEL_WIDTH, -dest_width, s->invalidated, fn, s->dma_ch[0].palette, miny, maxy); @@ -XXX,XX +XXX,XX @@ static void pxa2xx_lcdc_dma0_redraw_rot180(PXA2xxLCDState *s, src_width *= 2; } - dest_width = s->xres * s->dest_width; + dest_width = s->xres * DEST_PIXEL_WIDTH; *miny = 0; if (s->invalidated) { framebuffer_update_memory_section(&s->fbsection, s->sysmem, addr, s->yres, src_width); } framebuffer_update_display(surface, &s->fbsection, s->xres, s->yres, - src_width, -dest_width, -s->dest_width, + src_width, -dest_width, -DEST_PIXEL_WIDTH, s->invalidated, fn, s->dma_ch[0].palette, miny, maxy); } @@ -XXX,XX +XXX,XX @@ static void pxa2xx_lcdc_dma0_redraw_rot270(PXA2xxLCDState *s, src_width *= 2; } - dest_width = s->yres * s->dest_width; + dest_width = s->yres * DEST_PIXEL_WIDTH; *miny = 0; if (s->invalidated) { framebuffer_update_memory_section(&s->fbsection, s->sysmem, addr, s->yres, src_width); } framebuffer_update_display(surface, &s->fbsection, s->xres, s->yres, - src_width, -s->dest_width, dest_width, + src_width, -DEST_PIXEL_WIDTH, dest_width, s->invalidated, fn, s->dma_ch[0].palette, miny, maxy); @@ -XXX,XX +XXX,XX @@ PXA2xxLCDState *pxa2xx_lcdc_init(MemoryRegion *sysmem, memory_region_add_subregion(sysmem, base, &s->iomem); s->con = graphic_console_init(NULL, 0, &pxa2xx_ops, s); - s->dest_width = 4; vmstate_register(NULL, 0, &vmstate_pxa2xx_lcdc, s); -- 2.20.1 Now that BITS is always 32, expand out all its uses in the template header, including removing now-useless uses of the glue() macro. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Acked-by: Gerd Hoffmann <kraxel@redhat.com> Message-id: 20210211141515.8755-7-peter.maydell@linaro.org --- hw/display/pxa2xx_template.h | 110 ++++++++++++++--------------------- 1 file changed, 45 insertions(+), 65 deletions(-) diff --git a/hw/display/pxa2xx_template.h b/hw/display/pxa2xx_template.h index XXXXXXX..XXXXXXX 100644 --- a/hw/display/pxa2xx_template.h +++ b/hw/display/pxa2xx_template.h @@ -XXX,XX +XXX,XX @@ */ # define SKIP_PIXEL(to) to += deststep -#if BITS == 8 -# define COPY_PIXEL(to, from) do { *to = from; SKIP_PIXEL(to); } while (0) -#elif BITS == 15 || BITS == 16 -# define COPY_PIXEL(to, from) \ - do { \ - *(uint16_t *) to = from; \ - SKIP_PIXEL(to); \ - } while (0) -#elif BITS == 24 -# define COPY_PIXEL(to, from) \ - do { \ - *(uint16_t *) to = from; \ - *(to + 2) = (from) >> 16; \ - SKIP_PIXEL(to); \ - } while (0) -#elif BITS == 32 # define COPY_PIXEL(to, from) \ do { \ *(uint32_t *) to = from; \ SKIP_PIXEL(to); \ } while (0) -#else -# error unknown bit depth -#endif #ifdef HOST_WORDS_BIGENDIAN # define SWAP_WORDS 1 @@ -XXX,XX +XXX,XX @@ #define FN_2(x) FN(x + 1) FN(x) #define FN_4(x) FN_2(x + 2) FN_2(x) -static void glue(pxa2xx_draw_line2_, BITS)(void *opaque, +static void pxa2xx_draw_line2(void *opaque, uint8_t *dest, const uint8_t *src, int width, int deststep) { uint32_t *palette = opaque; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line2_, BITS)(void *opaque, } } -static void glue(pxa2xx_draw_line4_, BITS)(void *opaque, +static void pxa2xx_draw_line4(void *opaque, uint8_t *dest, const uint8_t *src, int width, int deststep) { uint32_t *palette = opaque; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line4_, BITS)(void *opaque, } } -static void glue(pxa2xx_draw_line8_, BITS)(void *opaque, +static void pxa2xx_draw_line8(void *opaque, uint8_t *dest, const uint8_t *src, int width, int deststep) { uint32_t *palette = opaque; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line8_, BITS)(void *opaque, } } -static void glue(pxa2xx_draw_line16_, BITS)(void *opaque, +static void pxa2xx_draw_line16(void *opaque, uint8_t *dest, const uint8_t *src, int width, int deststep) { uint32_t data; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line16_, BITS)(void *opaque, data >>= 6; r = (data & 0x1f) << 3; data >>= 5; - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); b = (data & 0x1f) << 3; data >>= 5; g = (data & 0x3f) << 2; data >>= 6; r = (data & 0x1f) << 3; - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); width -= 2; src += 4; } } -static void glue(pxa2xx_draw_line16t_, BITS)(void *opaque, +static void pxa2xx_draw_line16t(void *opaque, uint8_t *dest, const uint8_t *src, int width, int deststep) { uint32_t data; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line16t_, BITS)(void *opaque, if (data & 1) SKIP_PIXEL(dest); else - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); data >>= 1; b = (data & 0x1f) << 3; data >>= 5; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line16t_, BITS)(void *opaque, if (data & 1) SKIP_PIXEL(dest); else - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); width -= 2; src += 4; } } -static void glue(pxa2xx_draw_line18_, BITS)(void *opaque, +static void pxa2xx_draw_line18(void *opaque, uint8_t *dest, const uint8_t *src, int width, int deststep) { uint32_t data; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line18_, BITS)(void *opaque, g = (data & 0x3f) << 2; data >>= 6; r = (data & 0x3f) << 2; - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); width -= 1; src += 4; } } /* The wicked packed format */ -static void glue(pxa2xx_draw_line18p_, BITS)(void *opaque, +static void pxa2xx_draw_line18p(void *opaque, uint8_t *dest, const uint8_t *src, int width, int deststep) { uint32_t data[3]; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line18p_, BITS)(void *opaque, data[0] >>= 6; r = (data[0] & 0x3f) << 2; data[0] >>= 12; - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); b = (data[0] & 0x3f) << 2; data[0] >>= 6; g = ((data[1] & 0xf) << 4) | (data[0] << 2); data[1] >>= 4; r = (data[1] & 0x3f) << 2; data[1] >>= 12; - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); b = (data[1] & 0x3f) << 2; data[1] >>= 6; g = (data[1] & 0x3f) << 2; data[1] >>= 6; r = ((data[2] & 0x3) << 6) | (data[1] << 2); data[2] >>= 8; - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); b = (data[2] & 0x3f) << 2; data[2] >>= 6; g = (data[2] & 0x3f) << 2; data[2] >>= 6; r = data[2] << 2; - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); width -= 4; } } -static void glue(pxa2xx_draw_line19_, BITS)(void *opaque, +static void pxa2xx_draw_line19(void *opaque, uint8_t *dest, const uint8_t *src, int width, int deststep) { uint32_t data; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line19_, BITS)(void *opaque, if (data & 1) SKIP_PIXEL(dest); else - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); width -= 1; src += 4; } } /* The wicked packed format */ -static void glue(pxa2xx_draw_line19p_, BITS)(void *opaque, +static void pxa2xx_draw_line19p(void *opaque, uint8_t *dest, const uint8_t *src, int width, int deststep) { uint32_t data[3]; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line19p_, BITS)(void *opaque, if (data[0] & 1) SKIP_PIXEL(dest); else - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); data[0] >>= 6; b = (data[0] & 0x3f) << 2; data[0] >>= 6; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line19p_, BITS)(void *opaque, if (data[1] & 1) SKIP_PIXEL(dest); else - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); data[1] >>= 6; b = (data[1] & 0x3f) << 2; data[1] >>= 6; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line19p_, BITS)(void *opaque, if (data[2] & 1) SKIP_PIXEL(dest); else - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); data[2] >>= 6; b = (data[2] & 0x3f) << 2; data[2] >>= 6; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line19p_, BITS)(void *opaque, if (data[2] & 1) SKIP_PIXEL(dest); else - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); width -= 4; } } -static void glue(pxa2xx_draw_line24_, BITS)(void *opaque, +static void pxa2xx_draw_line24(void *opaque, uint8_t *dest, const uint8_t *src, int width, int deststep) { uint32_t data; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line24_, BITS)(void *opaque, g = data & 0xff; data >>= 8; r = data & 0xff; - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); width -= 1; src += 4; } } -static void glue(pxa2xx_draw_line24t_, BITS)(void *opaque, +static void pxa2xx_draw_line24t(void *opaque, uint8_t *dest, const uint8_t *src, int width, int deststep) { uint32_t data; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line24t_, BITS)(void *opaque, if (data & 1) SKIP_PIXEL(dest); else - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); width -= 1; src += 4; } } -static void glue(pxa2xx_draw_line25_, BITS)(void *opaque, +static void pxa2xx_draw_line25(void *opaque, uint8_t *dest, const uint8_t *src, int width, int deststep) { uint32_t data; @@ -XXX,XX +XXX,XX @@ static void glue(pxa2xx_draw_line25_, BITS)(void *opaque, if (data & 1) SKIP_PIXEL(dest); else - COPY_PIXEL(dest, glue(rgb_to_pixel, BITS)(r, g, b)); + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); width -= 1; src += 4; } } /* Overlay planes disabled, no transparency */ -static drawfn glue(pxa2xx_draw_fn_, BITS)[16] = +static drawfn pxa2xx_draw_fn_32[16] = { [0 ... 0xf] = NULL, - [pxa_lcdc_2bpp] = glue(pxa2xx_draw_line2_, BITS), - [pxa_lcdc_4bpp] = glue(pxa2xx_draw_line4_, BITS), - [pxa_lcdc_8bpp] = glue(pxa2xx_draw_line8_, BITS), - [pxa_lcdc_16bpp] = glue(pxa2xx_draw_line16_, BITS), - [pxa_lcdc_18bpp] = glue(pxa2xx_draw_line18_, BITS), - [pxa_lcdc_18pbpp] = glue(pxa2xx_draw_line18p_, BITS), - [pxa_lcdc_24bpp] = glue(pxa2xx_draw_line24_, BITS), + [pxa_lcdc_2bpp] = pxa2xx_draw_line2, + [pxa_lcdc_4bpp] = pxa2xx_draw_line4, + [pxa_lcdc_8bpp] = pxa2xx_draw_line8, + [pxa_lcdc_16bpp] = pxa2xx_draw_line16, + [pxa_lcdc_18bpp] = pxa2xx_draw_line18, + [pxa_lcdc_18pbpp] = pxa2xx_draw_line18p, + [pxa_lcdc_24bpp] = pxa2xx_draw_line24, }; /* Overlay planes enabled, transparency used */ -static drawfn glue(glue(pxa2xx_draw_fn_, BITS), t)[16] = +static drawfn pxa2xx_draw_fn_32t[16] = { [0 ... 0xf] = NULL, - [pxa_lcdc_4bpp] = glue(pxa2xx_draw_line4_, BITS), - [pxa_lcdc_8bpp] = glue(pxa2xx_draw_line8_, BITS), - [pxa_lcdc_16bpp] = glue(pxa2xx_draw_line16t_, BITS), - [pxa_lcdc_19bpp] = glue(pxa2xx_draw_line19_, BITS), - [pxa_lcdc_19pbpp] = glue(pxa2xx_draw_line19p_, BITS), - [pxa_lcdc_24bpp] = glue(pxa2xx_draw_line24t_, BITS), - [pxa_lcdc_25bpp] = glue(pxa2xx_draw_line25_, BITS), + [pxa_lcdc_4bpp] = pxa2xx_draw_line4, + [pxa_lcdc_8bpp] = pxa2xx_draw_line8, + [pxa_lcdc_16bpp] = pxa2xx_draw_line16t, + [pxa_lcdc_19bpp] = pxa2xx_draw_line19, + [pxa_lcdc_19pbpp] = pxa2xx_draw_line19p, + [pxa_lcdc_24bpp] = pxa2xx_draw_line24t, + [pxa_lcdc_25bpp] = pxa2xx_draw_line25, }; -#undef BITS #undef COPY_PIXEL #undef SKIP_PIXEL -- 2.20.1 We're about to move code from the template header into pxa2xx_lcd.c. Before doing that, make coding style fixes so checkpatch doesn't complain about the patch which moves the code. This commit fixes missing braces in the SKIP_PIXEL() macro definition and in if() statements. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Acked-by: Gerd Hoffmann <kraxel@redhat.com> Message-id: 20210211141515.8755-8-peter.maydell@linaro.org --- hw/display/pxa2xx_template.h | 47 +++++++++++++++++++++--------------- 1 file changed, 28 insertions(+), 19 deletions(-) diff --git a/hw/display/pxa2xx_template.h b/hw/display/pxa2xx_template.h index XXXXXXX..XXXXXXX 100644 --- a/hw/display/pxa2xx_template.h +++ b/hw/display/pxa2xx_template.h @@ -XXX,XX +XXX,XX @@ * Framebuffer format conversion routines. */ -# define SKIP_PIXEL(to) to += deststep +# define SKIP_PIXEL(to) do { to += deststep; } while (0) # define COPY_PIXEL(to, from) \ do { \ *(uint32_t *) to = from; \ @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line16t(void *opaque, data >>= 5; r = (data & 0x1f) << 3; data >>= 5; - if (data & 1) + if (data & 1) { SKIP_PIXEL(dest); - else + } else { COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } data >>= 1; b = (data & 0x1f) << 3; data >>= 5; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line16t(void *opaque, data >>= 5; r = (data & 0x1f) << 3; data >>= 5; - if (data & 1) + if (data & 1) { SKIP_PIXEL(dest); - else + } else { COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } width -= 2; src += 4; } @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line19(void *opaque, data >>= 6; r = (data & 0x3f) << 2; data >>= 6; - if (data & 1) + if (data & 1) { SKIP_PIXEL(dest); - else + } else { COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } width -= 1; src += 4; } @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line19p(void *opaque, data[0] >>= 6; r = (data[0] & 0x3f) << 2; data[0] >>= 6; - if (data[0] & 1) + if (data[0] & 1) { SKIP_PIXEL(dest); - else + } else { COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } data[0] >>= 6; b = (data[0] & 0x3f) << 2; data[0] >>= 6; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line19p(void *opaque, data[1] >>= 4; r = (data[1] & 0x3f) << 2; data[1] >>= 6; - if (data[1] & 1) + if (data[1] & 1) { SKIP_PIXEL(dest); - else + } else { COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } data[1] >>= 6; b = (data[1] & 0x3f) << 2; data[1] >>= 6; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line19p(void *opaque, data[1] >>= 6; r = ((data[2] & 0x3) << 6) | (data[1] << 2); data[2] >>= 2; - if (data[2] & 1) + if (data[2] & 1) { SKIP_PIXEL(dest); - else + } else { COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } data[2] >>= 6; b = (data[2] & 0x3f) << 2; data[2] >>= 6; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line19p(void *opaque, data[2] >>= 6; r = data[2] << 2; data[2] >>= 6; - if (data[2] & 1) + if (data[2] & 1) { SKIP_PIXEL(dest); - else + } else { COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } width -= 4; } } @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line24t(void *opaque, data >>= 8; r = data & 0xff; data >>= 8; - if (data & 1) + if (data & 1) { SKIP_PIXEL(dest); - else + } else { COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } width -= 1; src += 4; } @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line25(void *opaque, data >>= 8; r = data & 0xff; data >>= 8; - if (data & 1) + if (data & 1) { SKIP_PIXEL(dest); - else + } else { COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } width -= 1; src += 4; } -- 2.20.1 We're about to move code from the template header into pxa2xx_lcd.c. Before doing that, make coding style fixes so checkpatch doesn't complain about the patch which moves the code. This commit is whitespace changes only: * avoid hard-coded tabs * fix ident on function prototypes * no newline before open brace on array definitions Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Acked-by: Gerd Hoffmann <kraxel@redhat.com> Message-id: 20210211141515.8755-9-peter.maydell@linaro.org --- hw/display/pxa2xx_template.h | 66 +++++++++++++++++------------------- 1 file changed, 32 insertions(+), 34 deletions(-) diff --git a/hw/display/pxa2xx_template.h b/hw/display/pxa2xx_template.h index XXXXXXX..XXXXXXX 100644 --- a/hw/display/pxa2xx_template.h +++ b/hw/display/pxa2xx_template.h @@ -XXX,XX +XXX,XX @@ } while (0) #ifdef HOST_WORDS_BIGENDIAN -# define SWAP_WORDS 1 +# define SWAP_WORDS 1 #endif -#define FN_2(x) FN(x + 1) FN(x) -#define FN_4(x) FN_2(x + 2) FN_2(x) +#define FN_2(x) FN(x + 1) FN(x) +#define FN_4(x) FN_2(x + 2) FN_2(x) -static void pxa2xx_draw_line2(void *opaque, - uint8_t *dest, const uint8_t *src, int width, int deststep) +static void pxa2xx_draw_line2(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) { uint32_t *palette = opaque; uint32_t data; while (width > 0) { data = *(uint32_t *) src; -#define FN(x) COPY_PIXEL(dest, palette[(data >> ((x) * 2)) & 3]); +#define FN(x) COPY_PIXEL(dest, palette[(data >> ((x) * 2)) & 3]); #ifdef SWAP_WORDS FN_4(12) FN_4(8) @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line2(void *opaque, } } -static void pxa2xx_draw_line4(void *opaque, - uint8_t *dest, const uint8_t *src, int width, int deststep) +static void pxa2xx_draw_line4(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) { uint32_t *palette = opaque; uint32_t data; while (width > 0) { data = *(uint32_t *) src; -#define FN(x) COPY_PIXEL(dest, palette[(data >> ((x) * 4)) & 0xf]); +#define FN(x) COPY_PIXEL(dest, palette[(data >> ((x) * 4)) & 0xf]); #ifdef SWAP_WORDS FN_2(6) FN_2(4) @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line4(void *opaque, } } -static void pxa2xx_draw_line8(void *opaque, - uint8_t *dest, const uint8_t *src, int width, int deststep) +static void pxa2xx_draw_line8(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) { uint32_t *palette = opaque; uint32_t data; while (width > 0) { data = *(uint32_t *) src; -#define FN(x) COPY_PIXEL(dest, palette[(data >> (x)) & 0xff]); +#define FN(x) COPY_PIXEL(dest, palette[(data >> (x)) & 0xff]); #ifdef SWAP_WORDS FN(24) FN(16) @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line8(void *opaque, } } -static void pxa2xx_draw_line16(void *opaque, - uint8_t *dest, const uint8_t *src, int width, int deststep) +static void pxa2xx_draw_line16(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) { uint32_t data; unsigned int r, g, b; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line16(void *opaque, } } -static void pxa2xx_draw_line16t(void *opaque, - uint8_t *dest, const uint8_t *src, int width, int deststep) +static void pxa2xx_draw_line16t(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) { uint32_t data; unsigned int r, g, b; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line16t(void *opaque, } } -static void pxa2xx_draw_line18(void *opaque, - uint8_t *dest, const uint8_t *src, int width, int deststep) +static void pxa2xx_draw_line18(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) { uint32_t data; unsigned int r, g, b; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line18(void *opaque, } /* The wicked packed format */ -static void pxa2xx_draw_line18p(void *opaque, - uint8_t *dest, const uint8_t *src, int width, int deststep) +static void pxa2xx_draw_line18p(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) { uint32_t data[3]; unsigned int r, g, b; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line18p(void *opaque, } } -static void pxa2xx_draw_line19(void *opaque, - uint8_t *dest, const uint8_t *src, int width, int deststep) +static void pxa2xx_draw_line19(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) { uint32_t data; unsigned int r, g, b; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line19(void *opaque, } /* The wicked packed format */ -static void pxa2xx_draw_line19p(void *opaque, - uint8_t *dest, const uint8_t *src, int width, int deststep) +static void pxa2xx_draw_line19p(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) { uint32_t data[3]; unsigned int r, g, b; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line19p(void *opaque, } } -static void pxa2xx_draw_line24(void *opaque, - uint8_t *dest, const uint8_t *src, int width, int deststep) +static void pxa2xx_draw_line24(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) { uint32_t data; unsigned int r, g, b; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line24(void *opaque, } } -static void pxa2xx_draw_line24t(void *opaque, - uint8_t *dest, const uint8_t *src, int width, int deststep) +static void pxa2xx_draw_line24t(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) { uint32_t data; unsigned int r, g, b; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line24t(void *opaque, } } -static void pxa2xx_draw_line25(void *opaque, - uint8_t *dest, const uint8_t *src, int width, int deststep) +static void pxa2xx_draw_line25(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) { uint32_t data; unsigned int r, g, b; @@ -XXX,XX +XXX,XX @@ static void pxa2xx_draw_line25(void *opaque, } /* Overlay planes disabled, no transparency */ -static drawfn pxa2xx_draw_fn_32[16] = -{ +static drawfn pxa2xx_draw_fn_32[16] = { [0 ... 0xf] = NULL, [pxa_lcdc_2bpp] = pxa2xx_draw_line2, [pxa_lcdc_4bpp] = pxa2xx_draw_line4, @@ -XXX,XX +XXX,XX @@ static drawfn pxa2xx_draw_fn_32[16] = }; /* Overlay planes enabled, transparency used */ -static drawfn pxa2xx_draw_fn_32t[16] = -{ +static drawfn pxa2xx_draw_fn_32t[16] = { [0 ... 0xf] = NULL, [pxa_lcdc_4bpp] = pxa2xx_draw_line4, [pxa_lcdc_8bpp] = pxa2xx_draw_line8, -- 2.20.1 The template header is now included only once; just inline its contents in hw/display/pxa2xx_lcd.c. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Acked-by: Gerd Hoffmann <kraxel@redhat.com> Message-id: 20210211141515.8755-10-peter.maydell@linaro.org --- hw/display/pxa2xx_template.h | 434 ----------------------------------- hw/display/pxa2xx_lcd.c | 427 +++++++++++++++++++++++++++++++++- 2 files changed, 425 insertions(+), 436 deletions(-) delete mode 100644 hw/display/pxa2xx_template.h diff --git a/hw/display/pxa2xx_template.h b/hw/display/pxa2xx_template.h deleted file mode 100644 index XXXXXXX..XXXXXXX --- a/hw/display/pxa2xx_template.h +++ /dev/null @@ -XXX,XX +XXX,XX @@ -/* - * Intel XScale PXA255/270 LCDC emulation. - * - * Copyright (c) 2006 Openedhand Ltd. - * Written by Andrzej Zaborowski <balrog@zabor.org> - * - * This code is licensed under the GPLv2. - * - * Framebuffer format conversion routines. - */ - -# define SKIP_PIXEL(to) do { to += deststep; } while (0) -# define COPY_PIXEL(to, from) \ - do { \ - *(uint32_t *) to = from; \ - SKIP_PIXEL(to); \ - } while (0) - -#ifdef HOST_WORDS_BIGENDIAN -# define SWAP_WORDS 1 -#endif - -#define FN_2(x) FN(x + 1) FN(x) -#define FN_4(x) FN_2(x + 2) FN_2(x) - -static void pxa2xx_draw_line2(void *opaque, uint8_t *dest, const uint8_t *src, - int width, int deststep) -{ - uint32_t *palette = opaque; - uint32_t data; - while (width > 0) { - data = *(uint32_t *) src; -#define FN(x) COPY_PIXEL(dest, palette[(data >> ((x) * 2)) & 3]); -#ifdef SWAP_WORDS - FN_4(12) - FN_4(8) - FN_4(4) - FN_4(0) -#else - FN_4(0) - FN_4(4) - FN_4(8) - FN_4(12) -#endif -#undef FN - width -= 16; - src += 4; - } -} - -static void pxa2xx_draw_line4(void *opaque, uint8_t *dest, const uint8_t *src, - int width, int deststep) -{ - uint32_t *palette = opaque; - uint32_t data; - while (width > 0) { - data = *(uint32_t *) src; -#define FN(x) COPY_PIXEL(dest, palette[(data >> ((x) * 4)) & 0xf]); -#ifdef SWAP_WORDS - FN_2(6) - FN_2(4) - FN_2(2) - FN_2(0) -#else - FN_2(0) - FN_2(2) - FN_2(4) - FN_2(6) -#endif -#undef FN - width -= 8; - src += 4; - } -} - -static void pxa2xx_draw_line8(void *opaque, uint8_t *dest, const uint8_t *src, - int width, int deststep) -{ - uint32_t *palette = opaque; - uint32_t data; - while (width > 0) { - data = *(uint32_t *) src; -#define FN(x) COPY_PIXEL(dest, palette[(data >> (x)) & 0xff]); -#ifdef SWAP_WORDS - FN(24) - FN(16) - FN(8) - FN(0) -#else - FN(0) - FN(8) - FN(16) - FN(24) -#endif -#undef FN - width -= 4; - src += 4; - } -} - -static void pxa2xx_draw_line16(void *opaque, uint8_t *dest, const uint8_t *src, - int width, int deststep) -{ - uint32_t data; - unsigned int r, g, b; - while (width > 0) { - data = *(uint32_t *) src; -#ifdef SWAP_WORDS - data = bswap32(data); -#endif - b = (data & 0x1f) << 3; - data >>= 5; - g = (data & 0x3f) << 2; - data >>= 6; - r = (data & 0x1f) << 3; - data >>= 5; - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - b = (data & 0x1f) << 3; - data >>= 5; - g = (data & 0x3f) << 2; - data >>= 6; - r = (data & 0x1f) << 3; - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - width -= 2; - src += 4; - } -} - -static void pxa2xx_draw_line16t(void *opaque, uint8_t *dest, const uint8_t *src, - int width, int deststep) -{ - uint32_t data; - unsigned int r, g, b; - while (width > 0) { - data = *(uint32_t *) src; -#ifdef SWAP_WORDS - data = bswap32(data); -#endif - b = (data & 0x1f) << 3; - data >>= 5; - g = (data & 0x1f) << 3; - data >>= 5; - r = (data & 0x1f) << 3; - data >>= 5; - if (data & 1) { - SKIP_PIXEL(dest); - } else { - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - } - data >>= 1; - b = (data & 0x1f) << 3; - data >>= 5; - g = (data & 0x1f) << 3; - data >>= 5; - r = (data & 0x1f) << 3; - data >>= 5; - if (data & 1) { - SKIP_PIXEL(dest); - } else { - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - } - width -= 2; - src += 4; - } -} - -static void pxa2xx_draw_line18(void *opaque, uint8_t *dest, const uint8_t *src, - int width, int deststep) -{ - uint32_t data; - unsigned int r, g, b; - while (width > 0) { - data = *(uint32_t *) src; -#ifdef SWAP_WORDS - data = bswap32(data); -#endif - b = (data & 0x3f) << 2; - data >>= 6; - g = (data & 0x3f) << 2; - data >>= 6; - r = (data & 0x3f) << 2; - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - width -= 1; - src += 4; - } -} - -/* The wicked packed format */ -static void pxa2xx_draw_line18p(void *opaque, uint8_t *dest, const uint8_t *src, - int width, int deststep) -{ - uint32_t data[3]; - unsigned int r, g, b; - while (width > 0) { - data[0] = *(uint32_t *) src; - src += 4; - data[1] = *(uint32_t *) src; - src += 4; - data[2] = *(uint32_t *) src; - src += 4; -#ifdef SWAP_WORDS - data[0] = bswap32(data[0]); - data[1] = bswap32(data[1]); - data[2] = bswap32(data[2]); -#endif - b = (data[0] & 0x3f) << 2; - data[0] >>= 6; - g = (data[0] & 0x3f) << 2; - data[0] >>= 6; - r = (data[0] & 0x3f) << 2; - data[0] >>= 12; - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - b = (data[0] & 0x3f) << 2; - data[0] >>= 6; - g = ((data[1] & 0xf) << 4) | (data[0] << 2); - data[1] >>= 4; - r = (data[1] & 0x3f) << 2; - data[1] >>= 12; - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - b = (data[1] & 0x3f) << 2; - data[1] >>= 6; - g = (data[1] & 0x3f) << 2; - data[1] >>= 6; - r = ((data[2] & 0x3) << 6) | (data[1] << 2); - data[2] >>= 8; - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - b = (data[2] & 0x3f) << 2; - data[2] >>= 6; - g = (data[2] & 0x3f) << 2; - data[2] >>= 6; - r = data[2] << 2; - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - width -= 4; - } -} - -static void pxa2xx_draw_line19(void *opaque, uint8_t *dest, const uint8_t *src, - int width, int deststep) -{ - uint32_t data; - unsigned int r, g, b; - while (width > 0) { - data = *(uint32_t *) src; -#ifdef SWAP_WORDS - data = bswap32(data); -#endif - b = (data & 0x3f) << 2; - data >>= 6; - g = (data & 0x3f) << 2; - data >>= 6; - r = (data & 0x3f) << 2; - data >>= 6; - if (data & 1) { - SKIP_PIXEL(dest); - } else { - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - } - width -= 1; - src += 4; - } -} - -/* The wicked packed format */ -static void pxa2xx_draw_line19p(void *opaque, uint8_t *dest, const uint8_t *src, - int width, int deststep) -{ - uint32_t data[3]; - unsigned int r, g, b; - while (width > 0) { - data[0] = *(uint32_t *) src; - src += 4; - data[1] = *(uint32_t *) src; - src += 4; - data[2] = *(uint32_t *) src; - src += 4; -# ifdef SWAP_WORDS - data[0] = bswap32(data[0]); - data[1] = bswap32(data[1]); - data[2] = bswap32(data[2]); -# endif - b = (data[0] & 0x3f) << 2; - data[0] >>= 6; - g = (data[0] & 0x3f) << 2; - data[0] >>= 6; - r = (data[0] & 0x3f) << 2; - data[0] >>= 6; - if (data[0] & 1) { - SKIP_PIXEL(dest); - } else { - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - } - data[0] >>= 6; - b = (data[0] & 0x3f) << 2; - data[0] >>= 6; - g = ((data[1] & 0xf) << 4) | (data[0] << 2); - data[1] >>= 4; - r = (data[1] & 0x3f) << 2; - data[1] >>= 6; - if (data[1] & 1) { - SKIP_PIXEL(dest); - } else { - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - } - data[1] >>= 6; - b = (data[1] & 0x3f) << 2; - data[1] >>= 6; - g = (data[1] & 0x3f) << 2; - data[1] >>= 6; - r = ((data[2] & 0x3) << 6) | (data[1] << 2); - data[2] >>= 2; - if (data[2] & 1) { - SKIP_PIXEL(dest); - } else { - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - } - data[2] >>= 6; - b = (data[2] & 0x3f) << 2; - data[2] >>= 6; - g = (data[2] & 0x3f) << 2; - data[2] >>= 6; - r = data[2] << 2; - data[2] >>= 6; - if (data[2] & 1) { - SKIP_PIXEL(dest); - } else { - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - } - width -= 4; - } -} - -static void pxa2xx_draw_line24(void *opaque, uint8_t *dest, const uint8_t *src, - int width, int deststep) -{ - uint32_t data; - unsigned int r, g, b; - while (width > 0) { - data = *(uint32_t *) src; -#ifdef SWAP_WORDS - data = bswap32(data); -#endif - b = data & 0xff; - data >>= 8; - g = data & 0xff; - data >>= 8; - r = data & 0xff; - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - width -= 1; - src += 4; - } -} - -static void pxa2xx_draw_line24t(void *opaque, uint8_t *dest, const uint8_t *src, - int width, int deststep) -{ - uint32_t data; - unsigned int r, g, b; - while (width > 0) { - data = *(uint32_t *) src; -#ifdef SWAP_WORDS - data = bswap32(data); -#endif - b = (data & 0x7f) << 1; - data >>= 7; - g = data & 0xff; - data >>= 8; - r = data & 0xff; - data >>= 8; - if (data & 1) { - SKIP_PIXEL(dest); - } else { - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - } - width -= 1; - src += 4; - } -} - -static void pxa2xx_draw_line25(void *opaque, uint8_t *dest, const uint8_t *src, - int width, int deststep) -{ - uint32_t data; - unsigned int r, g, b; - while (width > 0) { - data = *(uint32_t *) src; -#ifdef SWAP_WORDS - data = bswap32(data); -#endif - b = data & 0xff; - data >>= 8; - g = data & 0xff; - data >>= 8; - r = data & 0xff; - data >>= 8; - if (data & 1) { - SKIP_PIXEL(dest); - } else { - COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); - } - width -= 1; - src += 4; - } -} - -/* Overlay planes disabled, no transparency */ -static drawfn pxa2xx_draw_fn_32[16] = { - [0 ... 0xf] = NULL, - [pxa_lcdc_2bpp] = pxa2xx_draw_line2, - [pxa_lcdc_4bpp] = pxa2xx_draw_line4, - [pxa_lcdc_8bpp] = pxa2xx_draw_line8, - [pxa_lcdc_16bpp] = pxa2xx_draw_line16, - [pxa_lcdc_18bpp] = pxa2xx_draw_line18, - [pxa_lcdc_18pbpp] = pxa2xx_draw_line18p, - [pxa_lcdc_24bpp] = pxa2xx_draw_line24, -}; - -/* Overlay planes enabled, transparency used */ -static drawfn pxa2xx_draw_fn_32t[16] = { - [0 ... 0xf] = NULL, - [pxa_lcdc_4bpp] = pxa2xx_draw_line4, - [pxa_lcdc_8bpp] = pxa2xx_draw_line8, - [pxa_lcdc_16bpp] = pxa2xx_draw_line16t, - [pxa_lcdc_19bpp] = pxa2xx_draw_line19, - [pxa_lcdc_19pbpp] = pxa2xx_draw_line19p, - [pxa_lcdc_24bpp] = pxa2xx_draw_line24t, - [pxa_lcdc_25bpp] = pxa2xx_draw_line25, -}; - -#undef COPY_PIXEL -#undef SKIP_PIXEL - -#ifdef SWAP_WORDS -# undef SWAP_WORDS -#endif diff --git a/hw/display/pxa2xx_lcd.c b/hw/display/pxa2xx_lcd.c index XXXXXXX..XXXXXXX 100644 --- a/hw/display/pxa2xx_lcd.c +++ b/hw/display/pxa2xx_lcd.c @@ -XXX,XX +XXX,XX @@ typedef struct QEMU_PACKED { /* Size of a pixel in the QEMU UI output surface, in bytes */ #define DEST_PIXEL_WIDTH 4 -#define BITS 32 -#include "pxa2xx_template.h" +/* Line drawing code to handle the various possible guest pixel formats */ + +# define SKIP_PIXEL(to) do { to += deststep; } while (0) +# define COPY_PIXEL(to, from) \ + do { \ + *(uint32_t *) to = from; \ + SKIP_PIXEL(to); \ + } while (0) + +#ifdef HOST_WORDS_BIGENDIAN +# define SWAP_WORDS 1 +#endif + +#define FN_2(x) FN(x + 1) FN(x) +#define FN_4(x) FN_2(x + 2) FN_2(x) + +static void pxa2xx_draw_line2(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) +{ + uint32_t *palette = opaque; + uint32_t data; + while (width > 0) { + data = *(uint32_t *) src; +#define FN(x) COPY_PIXEL(dest, palette[(data >> ((x) * 2)) & 3]); +#ifdef SWAP_WORDS + FN_4(12) + FN_4(8) + FN_4(4) + FN_4(0) +#else + FN_4(0) + FN_4(4) + FN_4(8) + FN_4(12) +#endif +#undef FN + width -= 16; + src += 4; + } +} + +static void pxa2xx_draw_line4(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) +{ + uint32_t *palette = opaque; + uint32_t data; + while (width > 0) { + data = *(uint32_t *) src; +#define FN(x) COPY_PIXEL(dest, palette[(data >> ((x) * 4)) & 0xf]); +#ifdef SWAP_WORDS + FN_2(6) + FN_2(4) + FN_2(2) + FN_2(0) +#else + FN_2(0) + FN_2(2) + FN_2(4) + FN_2(6) +#endif +#undef FN + width -= 8; + src += 4; + } +} + +static void pxa2xx_draw_line8(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) +{ + uint32_t *palette = opaque; + uint32_t data; + while (width > 0) { + data = *(uint32_t *) src; +#define FN(x) COPY_PIXEL(dest, palette[(data >> (x)) & 0xff]); +#ifdef SWAP_WORDS + FN(24) + FN(16) + FN(8) + FN(0) +#else + FN(0) + FN(8) + FN(16) + FN(24) +#endif +#undef FN + width -= 4; + src += 4; + } +} + +static void pxa2xx_draw_line16(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) +{ + uint32_t data; + unsigned int r, g, b; + while (width > 0) { + data = *(uint32_t *) src; +#ifdef SWAP_WORDS + data = bswap32(data); +#endif + b = (data & 0x1f) << 3; + data >>= 5; + g = (data & 0x3f) << 2; + data >>= 6; + r = (data & 0x1f) << 3; + data >>= 5; + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + b = (data & 0x1f) << 3; + data >>= 5; + g = (data & 0x3f) << 2; + data >>= 6; + r = (data & 0x1f) << 3; + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + width -= 2; + src += 4; + } +} + +static void pxa2xx_draw_line16t(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) +{ + uint32_t data; + unsigned int r, g, b; + while (width > 0) { + data = *(uint32_t *) src; +#ifdef SWAP_WORDS + data = bswap32(data); +#endif + b = (data & 0x1f) << 3; + data >>= 5; + g = (data & 0x1f) << 3; + data >>= 5; + r = (data & 0x1f) << 3; + data >>= 5; + if (data & 1) { + SKIP_PIXEL(dest); + } else { + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } + data >>= 1; + b = (data & 0x1f) << 3; + data >>= 5; + g = (data & 0x1f) << 3; + data >>= 5; + r = (data & 0x1f) << 3; + data >>= 5; + if (data & 1) { + SKIP_PIXEL(dest); + } else { + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } + width -= 2; + src += 4; + } +} + +static void pxa2xx_draw_line18(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) +{ + uint32_t data; + unsigned int r, g, b; + while (width > 0) { + data = *(uint32_t *) src; +#ifdef SWAP_WORDS + data = bswap32(data); +#endif + b = (data & 0x3f) << 2; + data >>= 6; + g = (data & 0x3f) << 2; + data >>= 6; + r = (data & 0x3f) << 2; + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + width -= 1; + src += 4; + } +} + +/* The wicked packed format */ +static void pxa2xx_draw_line18p(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) +{ + uint32_t data[3]; + unsigned int r, g, b; + while (width > 0) { + data[0] = *(uint32_t *) src; + src += 4; + data[1] = *(uint32_t *) src; + src += 4; + data[2] = *(uint32_t *) src; + src += 4; +#ifdef SWAP_WORDS + data[0] = bswap32(data[0]); + data[1] = bswap32(data[1]); + data[2] = bswap32(data[2]); +#endif + b = (data[0] & 0x3f) << 2; + data[0] >>= 6; + g = (data[0] & 0x3f) << 2; + data[0] >>= 6; + r = (data[0] & 0x3f) << 2; + data[0] >>= 12; + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + b = (data[0] & 0x3f) << 2; + data[0] >>= 6; + g = ((data[1] & 0xf) << 4) | (data[0] << 2); + data[1] >>= 4; + r = (data[1] & 0x3f) << 2; + data[1] >>= 12; + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + b = (data[1] & 0x3f) << 2; + data[1] >>= 6; + g = (data[1] & 0x3f) << 2; + data[1] >>= 6; + r = ((data[2] & 0x3) << 6) | (data[1] << 2); + data[2] >>= 8; + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + b = (data[2] & 0x3f) << 2; + data[2] >>= 6; + g = (data[2] & 0x3f) << 2; + data[2] >>= 6; + r = data[2] << 2; + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + width -= 4; + } +} + +static void pxa2xx_draw_line19(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) +{ + uint32_t data; + unsigned int r, g, b; + while (width > 0) { + data = *(uint32_t *) src; +#ifdef SWAP_WORDS + data = bswap32(data); +#endif + b = (data & 0x3f) << 2; + data >>= 6; + g = (data & 0x3f) << 2; + data >>= 6; + r = (data & 0x3f) << 2; + data >>= 6; + if (data & 1) { + SKIP_PIXEL(dest); + } else { + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } + width -= 1; + src += 4; + } +} + +/* The wicked packed format */ +static void pxa2xx_draw_line19p(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) +{ + uint32_t data[3]; + unsigned int r, g, b; + while (width > 0) { + data[0] = *(uint32_t *) src; + src += 4; + data[1] = *(uint32_t *) src; + src += 4; + data[2] = *(uint32_t *) src; + src += 4; +# ifdef SWAP_WORDS + data[0] = bswap32(data[0]); + data[1] = bswap32(data[1]); + data[2] = bswap32(data[2]); +# endif + b = (data[0] & 0x3f) << 2; + data[0] >>= 6; + g = (data[0] & 0x3f) << 2; + data[0] >>= 6; + r = (data[0] & 0x3f) << 2; + data[0] >>= 6; + if (data[0] & 1) { + SKIP_PIXEL(dest); + } else { + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } + data[0] >>= 6; + b = (data[0] & 0x3f) << 2; + data[0] >>= 6; + g = ((data[1] & 0xf) << 4) | (data[0] << 2); + data[1] >>= 4; + r = (data[1] & 0x3f) << 2; + data[1] >>= 6; + if (data[1] & 1) { + SKIP_PIXEL(dest); + } else { + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } + data[1] >>= 6; + b = (data[1] & 0x3f) << 2; + data[1] >>= 6; + g = (data[1] & 0x3f) << 2; + data[1] >>= 6; + r = ((data[2] & 0x3) << 6) | (data[1] << 2); + data[2] >>= 2; + if (data[2] & 1) { + SKIP_PIXEL(dest); + } else { + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } + data[2] >>= 6; + b = (data[2] & 0x3f) << 2; + data[2] >>= 6; + g = (data[2] & 0x3f) << 2; + data[2] >>= 6; + r = data[2] << 2; + data[2] >>= 6; + if (data[2] & 1) { + SKIP_PIXEL(dest); + } else { + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } + width -= 4; + } +} + +static void pxa2xx_draw_line24(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) +{ + uint32_t data; + unsigned int r, g, b; + while (width > 0) { + data = *(uint32_t *) src; +#ifdef SWAP_WORDS + data = bswap32(data); +#endif + b = data & 0xff; + data >>= 8; + g = data & 0xff; + data >>= 8; + r = data & 0xff; + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + width -= 1; + src += 4; + } +} + +static void pxa2xx_draw_line24t(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) +{ + uint32_t data; + unsigned int r, g, b; + while (width > 0) { + data = *(uint32_t *) src; +#ifdef SWAP_WORDS + data = bswap32(data); +#endif + b = (data & 0x7f) << 1; + data >>= 7; + g = data & 0xff; + data >>= 8; + r = data & 0xff; + data >>= 8; + if (data & 1) { + SKIP_PIXEL(dest); + } else { + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } + width -= 1; + src += 4; + } +} + +static void pxa2xx_draw_line25(void *opaque, uint8_t *dest, const uint8_t *src, + int width, int deststep) +{ + uint32_t data; + unsigned int r, g, b; + while (width > 0) { + data = *(uint32_t *) src; +#ifdef SWAP_WORDS + data = bswap32(data); +#endif + b = data & 0xff; + data >>= 8; + g = data & 0xff; + data >>= 8; + r = data & 0xff; + data >>= 8; + if (data & 1) { + SKIP_PIXEL(dest); + } else { + COPY_PIXEL(dest, rgb_to_pixel32(r, g, b)); + } + width -= 1; + src += 4; + } +} + +/* Overlay planes disabled, no transparency */ +static drawfn pxa2xx_draw_fn_32[16] = { + [0 ... 0xf] = NULL, + [pxa_lcdc_2bpp] = pxa2xx_draw_line2, + [pxa_lcdc_4bpp] = pxa2xx_draw_line4, + [pxa_lcdc_8bpp] = pxa2xx_draw_line8, + [pxa_lcdc_16bpp] = pxa2xx_draw_line16, + [pxa_lcdc_18bpp] = pxa2xx_draw_line18, + [pxa_lcdc_18pbpp] = pxa2xx_draw_line18p, + [pxa_lcdc_24bpp] = pxa2xx_draw_line24, +}; + +/* Overlay planes enabled, transparency used */ +static drawfn pxa2xx_draw_fn_32t[16] = { + [0 ... 0xf] = NULL, + [pxa_lcdc_4bpp] = pxa2xx_draw_line4, + [pxa_lcdc_8bpp] = pxa2xx_draw_line8, + [pxa_lcdc_16bpp] = pxa2xx_draw_line16t, + [pxa_lcdc_19bpp] = pxa2xx_draw_line19, + [pxa_lcdc_19pbpp] = pxa2xx_draw_line19p, + [pxa_lcdc_24bpp] = pxa2xx_draw_line24t, + [pxa_lcdc_25bpp] = pxa2xx_draw_line25, +}; + +#undef COPY_PIXEL +#undef SKIP_PIXEL + +#ifdef SWAP_WORDS +# undef SWAP_WORDS +#endif /* Route internal interrupt lines to the global IC */ static void pxa2xx_lcdc_int_update(PXA2xxLCDState *s) -- 2.20.1