Most of this is the Neon decodetree patches, followed by Edgar's versal cleanups. thanks -- PMM The following changes since commit 2ef486e76d64436be90f7359a3071fb2a56ce835: Merge remote-tracking branch 'remotes/marcel/tags/rdma-pull-request' into staging (2020-05-03 14:12:56 +0100) are available in the Git repository at: https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20200504 for you to fetch changes up to 9aefc6cf9b73f66062d2f914a0136756e7a28211: target/arm: Move gen_ function typedefs to translate.h (2020-05-04 12:59:26 +0100) ---------------------------------------------------------------- target-arm queue: * Start of conversion of Neon insns to decodetree * versal board: support SD and RTC * Implement ARMv8.2-TTS2UXN * Make VQDMULL undefined when U=1 * Some minor code cleanups ---------------------------------------------------------------- Edgar E. Iglesias (11): hw/arm: versal: Remove inclusion of arm_gicv3_common.h hw/arm: versal: Move misplaced comment hw/arm: versal-virt: Fix typo xlnx-ve -> xlnx-versal hw/arm: versal: Embed the UARTs into the SoC type hw/arm: versal: Embed the GEMs into the SoC type hw/arm: versal: Embed the ADMAs into the SoC type hw/arm: versal: Embed the APUs into the SoC type hw/arm: versal: Add support for SD hw/arm: versal: Add support for the RTC hw/arm: versal-virt: Add support for SD hw/arm: versal-virt: Add support for the RTC Fredrik Strupe (1): target/arm: Make VQDMULL undefined when U=1 Peter Maydell (25): target/arm: Don't use a TLB for ARMMMUIdx_Stage2 target/arm: Use enum constant in get_phys_addr_lpae() call target/arm: Add new 's1_is_el0' argument to get_phys_addr_lpae() target/arm: Implement ARMv8.2-TTS2UXN target/arm: Use correct variable for setting 'max' cpu's ID_AA64DFR0 target/arm/translate-vfp.inc.c: Remove duplicate simd_r32 check target/arm: Don't allow Thumb Neon insns without FEATURE_NEON target/arm: Add stubs for AArch32 Neon decodetree target/arm: Convert VCMLA (vector) to decodetree target/arm: Convert VCADD (vector) to decodetree target/arm: Convert V[US]DOT (vector) to decodetree target/arm: Convert VFM[AS]L (vector) to decodetree target/arm: Convert VCMLA (scalar) to decodetree target/arm: Convert V[US]DOT (scalar) to decodetree target/arm: Convert VFM[AS]L (scalar) to decodetree target/arm: Convert Neon load/store multiple structures to decodetree target/arm: Convert Neon 'load single structure to all lanes' to decodetree target/arm: Convert Neon 'load/store single structure' to decodetree target/arm: Convert Neon 3-reg-same VADD/VSUB to decodetree target/arm: Convert Neon 3-reg-same logic ops to decodetree target/arm: Convert Neon 3-reg-same VMAX/VMIN to decodetree target/arm: Convert Neon 3-reg-same comparisons to decodetree target/arm: Convert Neon 3-reg-same VQADD/VQSUB to decodetree target/arm: Convert Neon 3-reg-same VMUL, VMLA, VMLS, VSHL to decodetree target/arm: Move gen_ function typedefs to translate.h Philippe Mathieu-Daudé (2): hw/arm/mps2-tz: Use TYPE_IOTKIT instead of hardcoded string target/arm: Use uint64_t for midr field in CPU state struct include/hw/arm/xlnx-versal.h | 31 +- target/arm/cpu-param.h | 2 +- target/arm/cpu.h | 38 ++- target/arm/translate-a64.h | 9 - target/arm/translate.h | 26 ++ target/arm/neon-dp.decode | 86 +++++ target/arm/neon-ls.decode | 52 +++ target/arm/neon-shared.decode | 66 ++++ hw/arm/mps2-tz.c | 2 +- hw/arm/xlnx-versal-virt.c | 74 ++++- hw/arm/xlnx-versal.c | 115 +++++-- target/arm/cpu.c | 3 +- target/arm/cpu64.c | 8 +- target/arm/helper.c | 183 ++++------ target/arm/translate-a64.c | 17 - target/arm/translate-neon.inc.c | 714 +++++++++++++++++++++++++++++++++++++++ target/arm/translate-vfp.inc.c | 6 - target/arm/translate.c | 716 +++------------------------------------- target/arm/Makefile.objs | 18 + 19 files changed, 1302 insertions(+), 864 deletions(-) create mode 100644 target/arm/neon-dp.decode create mode 100644 target/arm/neon-ls.decode create mode 100644 target/arm/neon-shared.decode create mode 100644 target/arm/translate-neon.inc.c From: Fredrik Strupe <fredrik@strupe.net> According to Arm ARM, VQDMULL is only valid when U=0, while having U=1 is unallocated. Signed-off-by: Fredrik Strupe <fredrik@strupe.net> Fixes: 695272dcb976 ("target-arm: Handle UNDEF cases for Neon 3-regs-different-widths") Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/translate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) 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) {0, 0, 0, 0}, /* VMLSL */ {0, 0, 0, 9}, /* VQDMLSL */ {0, 0, 0, 0}, /* Integer VMULL */ - {0, 0, 0, 1}, /* VQDMULL */ + {0, 0, 0, 9}, /* VQDMULL */ {0, 0, 0, 0xa}, /* Polynomial VMULL */ {0, 0, 0, 7}, /* Reserved: always UNDEF */ }; -- 2.20.1 From: Philippe Mathieu-Daudé <f4bug@amsat.org> By using the TYPE_* definitions for devices, we can: - quickly find where devices are used with 'git-grep' - easily rename a device (one-line change). Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 20200428154650.21991-1-f4bug@amsat.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/mps2-tz.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/hw/arm/mps2-tz.c b/hw/arm/mps2-tz.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/mps2-tz.c +++ b/hw/arm/mps2-tz.c @@ -XXX,XX +XXX,XX @@ static void mps2tz_common_init(MachineState *machine) exit(EXIT_FAILURE); } - sysbus_init_child_obj(OBJECT(machine), "iotkit", &mms->iotkit, + sysbus_init_child_obj(OBJECT(machine), TYPE_IOTKIT, &mms->iotkit, sizeof(mms->iotkit), mmc->armsse_type); iotkitdev = DEVICE(&mms->iotkit); object_property_set_link(OBJECT(&mms->iotkit), OBJECT(system_memory), -- 2.20.1 We define ARMMMUIdx_Stage2 as being an MMU index which uses a QEMU TLB. However we never actually use the TLB -- all stage 2 lookups are done by direct calls to get_phys_addr_lpae() followed by a physical address load via address_space_ld*(). Remove Stage2 from the list of ARM MMU indexes which correspond to real core MMU indexes, and instead put it in the set of "NOTLB" ARM MMU indexes. This allows us to drop NB_MMU_MODES to 11. It also means we can safely add support for the ARMv8.3-TTS2UXN extension, which adds permission bits to the stage 2 descriptors which define execute permission separatel for EL0 and EL1; supporting that while keeping Stage2 in a QEMU TLB would require us to use separate TLBs for "Stage2 for an EL0 access" and "Stage2 for an EL1 access", which is a lot of extra complication given we aren't even using the QEMU TLB. In the process of updating the comment on our MMU index use, fix a couple of other minor errors: * NS EL2 EL2&0 was missing from the list in the comment * some text hadn't been updated from when we bumped NB_MMU_MODES above 8 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200330210400.11724-2-peter.maydell@linaro.org --- target/arm/cpu-param.h | 2 +- target/arm/cpu.h | 21 +++++--- target/arm/helper.c | 112 ++++------------------------------------- 3 files changed, 27 insertions(+), 108 deletions(-) diff --git a/target/arm/cpu-param.h b/target/arm/cpu-param.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu-param.h +++ b/target/arm/cpu-param.h @@ -XXX,XX +XXX,XX @@ # define TARGET_PAGE_BITS_MIN 10 #endif -#define NB_MMU_MODES 12 +#define NB_MMU_MODES 11 #endif diff --git a/target/arm/cpu.h b/target/arm/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.h +++ b/target/arm/cpu.h @@ -XXX,XX +XXX,XX @@ bool write_cpustate_to_list(ARMCPU *cpu, bool kvm_sync); * handling via the TLB. The only way to do a stage 1 translation without * the immediate stage 2 translation is via the ATS or AT system insns, * which can be slow-pathed and always do a page table walk. + * The only use of stage 2 translations is either as part of an s1+2 + * lookup or when loading the descriptors during a stage 1 page table walk, + * and in both those cases we don't use the TLB. * 4. we can also safely fold together the "32 bit EL3" and "64 bit EL3" * translation regimes, because they map reasonably well to each other * and they can't both be active at the same time. @@ -XXX,XX +XXX,XX @@ bool write_cpustate_to_list(ARMCPU *cpu, bool kvm_sync); * NS EL1 EL1&0 stage 1+2 (aka NS PL1) * NS EL1 EL1&0 stage 1+2 +PAN * NS EL0 EL2&0 + * NS EL2 EL2&0 * NS EL2 EL2&0 +PAN * NS EL2 (aka NS PL2) * S EL0 EL1&0 (aka S PL0) * S EL1 EL1&0 (not used if EL3 is 32 bit) * S EL1 EL1&0 +PAN * S EL3 (aka S PL1) - * NS EL1&0 stage 2 * - * for a total of 12 different mmu_idx. + * for a total of 11 different mmu_idx. * * R profile CPUs have an MPU, but can use the same set of MMU indexes * as A profile. They only need to distinguish NS EL0 and NS EL1 (and @@ -XXX,XX +XXX,XX @@ bool write_cpustate_to_list(ARMCPU *cpu, bool kvm_sync); * are not quite the same -- different CPU types (most notably M profile * vs A/R profile) would like to use MMU indexes with different semantics, * but since we don't ever need to use all of those in a single CPU we - * can avoid setting NB_MMU_MODES to more than 8. The lower bits of + * can avoid having to set NB_MMU_MODES to "total number of A profile MMU + * modes + total number of M profile MMU modes". The lower bits of * ARMMMUIdx are the core TLB mmu index, and the higher bits are always * the same for any particular CPU. * Variables of type ARMMUIdx are always full values, and the core @@ -XXX,XX +XXX,XX @@ typedef enum ARMMMUIdx { ARMMMUIdx_SE10_1_PAN = 9 | ARM_MMU_IDX_A, ARMMMUIdx_SE3 = 10 | ARM_MMU_IDX_A, - ARMMMUIdx_Stage2 = 11 | ARM_MMU_IDX_A, - /* * These are not allocated TLBs and are used only for AT system * instructions or for the first stage of an S12 page table walk. @@ -XXX,XX +XXX,XX @@ typedef enum ARMMMUIdx { ARMMMUIdx_Stage1_E0 = 0 | ARM_MMU_IDX_NOTLB, ARMMMUIdx_Stage1_E1 = 1 | ARM_MMU_IDX_NOTLB, ARMMMUIdx_Stage1_E1_PAN = 2 | ARM_MMU_IDX_NOTLB, + /* + * Not allocated a TLB: used only for second stage of an S12 page + * table walk, or for descriptor loads during first stage of an S1 + * page table walk. Note that if we ever want to have a TLB for this + * then various TLB flush insns which currently are no-ops or flush + * only stage 1 MMU indexes will need to change to flush stage 2. + */ + ARMMMUIdx_Stage2 = 3 | ARM_MMU_IDX_NOTLB, /* * M-profile. @@ -XXX,XX +XXX,XX @@ typedef enum ARMMMUIdxBit { TO_CORE_BIT(SE10_1), TO_CORE_BIT(SE10_1_PAN), TO_CORE_BIT(SE3), - TO_CORE_BIT(Stage2), TO_CORE_BIT(MUser), TO_CORE_BIT(MPriv), diff --git a/target/arm/helper.c b/target/arm/helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.c +++ b/target/arm/helper.c @@ -XXX,XX +XXX,XX @@ static void tlbiall_nsnh_write(CPUARMState *env, const ARMCPRegInfo *ri, tlb_flush_by_mmuidx(cs, ARMMMUIdxBit_E10_1 | ARMMMUIdxBit_E10_1_PAN | - ARMMMUIdxBit_E10_0 | - ARMMMUIdxBit_Stage2); + ARMMMUIdxBit_E10_0); } static void tlbiall_nsnh_is_write(CPUARMState *env, const ARMCPRegInfo *ri, @@ -XXX,XX +XXX,XX @@ static void tlbiall_nsnh_is_write(CPUARMState *env, const ARMCPRegInfo *ri, tlb_flush_by_mmuidx_all_cpus_synced(cs, ARMMMUIdxBit_E10_1 | ARMMMUIdxBit_E10_1_PAN | - ARMMMUIdxBit_E10_0 | - ARMMMUIdxBit_Stage2); + ARMMMUIdxBit_E10_0); } -static void tlbiipas2_write(CPUARMState *env, const ARMCPRegInfo *ri, - uint64_t value) -{ - /* Invalidate by IPA. This has to invalidate any structures that - * contain only stage 2 translation information, but does not need - * to apply to structures that contain combined stage 1 and stage 2 - * translation information. - * This must NOP if EL2 isn't implemented or SCR_EL3.NS is zero. - */ - CPUState *cs = env_cpu(env); - uint64_t pageaddr; - - if (!arm_feature(env, ARM_FEATURE_EL2) || !(env->cp15.scr_el3 & SCR_NS)) { - return; - } - - pageaddr = sextract64(value << 12, 0, 40); - - tlb_flush_page_by_mmuidx(cs, pageaddr, ARMMMUIdxBit_Stage2); -} - -static void tlbiipas2_is_write(CPUARMState *env, const ARMCPRegInfo *ri, - uint64_t value) -{ - CPUState *cs = env_cpu(env); - uint64_t pageaddr; - - if (!arm_feature(env, ARM_FEATURE_EL2) || !(env->cp15.scr_el3 & SCR_NS)) { - return; - } - - pageaddr = sextract64(value << 12, 0, 40); - - tlb_flush_page_by_mmuidx_all_cpus_synced(cs, pageaddr, - ARMMMUIdxBit_Stage2); -} static void tlbiall_hyp_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value) @@ -XXX,XX +XXX,XX @@ static void vttbr_write(CPUARMState *env, const ARMCPRegInfo *ri, tlb_flush_by_mmuidx(cs, ARMMMUIdxBit_E10_1 | ARMMMUIdxBit_E10_1_PAN | - ARMMMUIdxBit_E10_0 | - ARMMMUIdxBit_Stage2); + ARMMMUIdxBit_E10_0); raw_write(env, ri, value); } } @@ -XXX,XX +XXX,XX @@ static int alle1_tlbmask(CPUARMState *env) return ARMMMUIdxBit_SE10_1 | ARMMMUIdxBit_SE10_1_PAN | ARMMMUIdxBit_SE10_0; - } else if (arm_feature(env, ARM_FEATURE_EL2)) { - return ARMMMUIdxBit_E10_1 | - ARMMMUIdxBit_E10_1_PAN | - ARMMMUIdxBit_E10_0 | - ARMMMUIdxBit_Stage2; } else { return ARMMMUIdxBit_E10_1 | ARMMMUIdxBit_E10_1_PAN | @@ -XXX,XX +XXX,XX @@ static void tlbi_aa64_vae3is_write(CPUARMState *env, const ARMCPRegInfo *ri, ARMMMUIdxBit_SE3); } -static void tlbi_aa64_ipas2e1_write(CPUARMState *env, const ARMCPRegInfo *ri, - uint64_t value) -{ - /* Invalidate by IPA. This has to invalidate any structures that - * contain only stage 2 translation information, but does not need - * to apply to structures that contain combined stage 1 and stage 2 - * translation information. - * This must NOP if EL2 isn't implemented or SCR_EL3.NS is zero. - */ - ARMCPU *cpu = env_archcpu(env); - CPUState *cs = CPU(cpu); - uint64_t pageaddr; - - if (!arm_feature(env, ARM_FEATURE_EL2) || !(env->cp15.scr_el3 & SCR_NS)) { - return; - } - - pageaddr = sextract64(value << 12, 0, 48); - - tlb_flush_page_by_mmuidx(cs, pageaddr, ARMMMUIdxBit_Stage2); -} - -static void tlbi_aa64_ipas2e1is_write(CPUARMState *env, const ARMCPRegInfo *ri, - uint64_t value) -{ - CPUState *cs = env_cpu(env); - uint64_t pageaddr; - - if (!arm_feature(env, ARM_FEATURE_EL2) || !(env->cp15.scr_el3 & SCR_NS)) { - return; - } - - pageaddr = sextract64(value << 12, 0, 48); - - tlb_flush_page_by_mmuidx_all_cpus_synced(cs, pageaddr, - ARMMMUIdxBit_Stage2); -} - static CPAccessResult aa64_zva_access(CPUARMState *env, const ARMCPRegInfo *ri, bool isread) { @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v8_cp_reginfo[] = { .writefn = tlbi_aa64_vae1_write }, { .name = "TLBI_IPAS2E1IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 0, .opc2 = 1, - .access = PL2_W, .type = ARM_CP_NO_RAW, - .writefn = tlbi_aa64_ipas2e1is_write }, + .access = PL2_W, .type = ARM_CP_NOP }, { .name = "TLBI_IPAS2LE1IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 0, .opc2 = 5, - .access = PL2_W, .type = ARM_CP_NO_RAW, - .writefn = tlbi_aa64_ipas2e1is_write }, + .access = PL2_W, .type = ARM_CP_NOP }, { .name = "TLBI_ALLE1IS", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 3, .opc2 = 4, .access = PL2_W, .type = ARM_CP_NO_RAW, @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v8_cp_reginfo[] = { .writefn = tlbi_aa64_alle1is_write }, { .name = "TLBI_IPAS2E1", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 4, .opc2 = 1, - .access = PL2_W, .type = ARM_CP_NO_RAW, - .writefn = tlbi_aa64_ipas2e1_write }, + .access = PL2_W, .type = ARM_CP_NOP }, { .name = "TLBI_IPAS2LE1", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 4, .opc2 = 5, - .access = PL2_W, .type = ARM_CP_NO_RAW, - .writefn = tlbi_aa64_ipas2e1_write }, + .access = PL2_W, .type = ARM_CP_NOP }, { .name = "TLBI_ALLE1", .state = ARM_CP_STATE_AA64, .opc0 = 1, .opc1 = 4, .crn = 8, .crm = 7, .opc2 = 4, .access = PL2_W, .type = ARM_CP_NO_RAW, @@ -XXX,XX +XXX,XX @@ static const ARMCPRegInfo v8_cp_reginfo[] = { .writefn = tlbimva_hyp_is_write }, { .name = "TLBIIPAS2", .cp = 15, .opc1 = 4, .crn = 8, .crm = 4, .opc2 = 1, - .type = ARM_CP_NO_RAW, .access = PL2_W, - .writefn = tlbiipas2_write }, + .type = ARM_CP_NOP, .access = PL2_W }, { .name = "TLBIIPAS2IS", .cp = 15, .opc1 = 4, .crn = 8, .crm = 0, .opc2 = 1, - .type = ARM_CP_NO_RAW, .access = PL2_W, - .writefn = tlbiipas2_is_write }, + .type = ARM_CP_NOP, .access = PL2_W }, { .name = "TLBIIPAS2L", .cp = 15, .opc1 = 4, .crn = 8, .crm = 4, .opc2 = 5, - .type = ARM_CP_NO_RAW, .access = PL2_W, - .writefn = tlbiipas2_write }, + .type = ARM_CP_NOP, .access = PL2_W }, { .name = "TLBIIPAS2LIS", .cp = 15, .opc1 = 4, .crn = 8, .crm = 0, .opc2 = 5, - .type = ARM_CP_NO_RAW, .access = PL2_W, - .writefn = tlbiipas2_is_write }, + .type = ARM_CP_NOP, .access = PL2_W }, /* 32 bit cache operations */ { .name = "ICIALLUIS", .cp = 15, .opc1 = 0, .crn = 7, .crm = 1, .opc2 = 0, .type = ARM_CP_NOP, .access = PL1_W, .accessfn = aa64_cacheop_pou_access }, -- 2.20.1 The access_type argument to get_phys_addr_lpae() is an MMUAccessType; use the enum constant MMU_DATA_LOAD rather than a literal 0 when we call it in S1_ptw_translate(). Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200330210400.11724-3-peter.maydell@linaro.org --- target/arm/helper.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/target/arm/helper.c b/target/arm/helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.c +++ b/target/arm/helper.c @@ -XXX,XX +XXX,XX @@ static hwaddr S1_ptw_translate(CPUARMState *env, ARMMMUIdx mmu_idx, pcacheattrs = &cacheattrs; } - ret = get_phys_addr_lpae(env, addr, 0, ARMMMUIdx_Stage2, &s2pa, - &txattrs, &s2prot, &s2size, fi, pcacheattrs); + ret = get_phys_addr_lpae(env, addr, MMU_DATA_LOAD, ARMMMUIdx_Stage2, + &s2pa, &txattrs, &s2prot, &s2size, fi, + pcacheattrs); if (ret) { assert(fi->type != ARMFault_None); fi->s2addr = addr; -- 2.20.1 For ARMv8.2-TTS2UXN, the stage 2 page table walk wants to know whether the stage 1 access is for EL0 or not, because whether exec permission is given can depend on whether this is an EL0 or EL1 access. Add a new argument to get_phys_addr_lpae() so the call sites can pass this information in. Since get_phys_addr_lpae() doesn't already have a doc comment, add one so we have a place to put the documentation of the semantics of the new s1_is_el0 argument. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200330210400.11724-4-peter.maydell@linaro.org --- target/arm/helper.c | 29 ++++++++++++++++++++++++++++- 1 file changed, 28 insertions(+), 1 deletion(-) diff --git a/target/arm/helper.c b/target/arm/helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.c +++ b/target/arm/helper.c @@ -XXX,XX +XXX,XX @@ static bool get_phys_addr_lpae(CPUARMState *env, target_ulong address, MMUAccessType access_type, ARMMMUIdx mmu_idx, + bool s1_is_el0, hwaddr *phys_ptr, MemTxAttrs *txattrs, int *prot, target_ulong *page_size_ptr, ARMMMUFaultInfo *fi, ARMCacheAttrs *cacheattrs); @@ -XXX,XX +XXX,XX @@ static hwaddr S1_ptw_translate(CPUARMState *env, ARMMMUIdx mmu_idx, } ret = get_phys_addr_lpae(env, addr, MMU_DATA_LOAD, ARMMMUIdx_Stage2, + false, &s2pa, &txattrs, &s2prot, &s2size, fi, pcacheattrs); if (ret) { @@ -XXX,XX +XXX,XX @@ static ARMVAParameters aa32_va_parameters(CPUARMState *env, uint32_t va, }; } +/** + * get_phys_addr_lpae: perform one stage of page table walk, LPAE format + * + * Returns false if the translation was successful. Otherwise, phys_ptr, attrs, + * prot and page_size may not be filled in, and the populated fsr value provides + * information on why the translation aborted, in the format of a long-format + * DFSR/IFSR fault register, with the following caveats: + * * the WnR bit is never set (the caller must do this). + * + * @env: CPUARMState + * @address: virtual address to get physical address for + * @access_type: MMU_DATA_LOAD, MMU_DATA_STORE or MMU_INST_FETCH + * @mmu_idx: MMU index indicating required translation regime + * @s1_is_el0: if @mmu_idx is ARMMMUIdx_Stage2 (so this is a stage 2 page table + * walk), must be true if this is stage 2 of a stage 1+2 walk for an + * EL0 access). If @mmu_idx is anything else, @s1_is_el0 is ignored. + * @phys_ptr: set to the physical address corresponding to the virtual address + * @attrs: set to the memory transaction attributes to use + * @prot: set to the permissions for the page containing phys_ptr + * @page_size_ptr: set to the size of the page containing phys_ptr + * @fi: set to fault info if the translation fails + * @cacheattrs: (if non-NULL) set to the cacheability/shareability attributes + */ static bool get_phys_addr_lpae(CPUARMState *env, target_ulong address, MMUAccessType access_type, ARMMMUIdx mmu_idx, + bool s1_is_el0, hwaddr *phys_ptr, MemTxAttrs *txattrs, int *prot, target_ulong *page_size_ptr, ARMMMUFaultInfo *fi, ARMCacheAttrs *cacheattrs) @@ -XXX,XX +XXX,XX @@ bool get_phys_addr(CPUARMState *env, target_ulong address, /* S1 is done. Now do S2 translation. */ ret = get_phys_addr_lpae(env, ipa, access_type, ARMMMUIdx_Stage2, + mmu_idx == ARMMMUIdx_E10_0, phys_ptr, attrs, &s2_prot, page_size, fi, cacheattrs != NULL ? &cacheattrs2 : NULL); @@ -XXX,XX +XXX,XX @@ bool get_phys_addr(CPUARMState *env, target_ulong address, } if (regime_using_lpae_format(env, mmu_idx)) { - return get_phys_addr_lpae(env, address, access_type, mmu_idx, + return get_phys_addr_lpae(env, address, access_type, mmu_idx, false, phys_ptr, attrs, prot, page_size, fi, cacheattrs); } else if (regime_sctlr(env, mmu_idx) & SCTLR_XP) { -- 2.20.1 The ARMv8.2-TTS2UXN feature extends the XN field in stage 2 translation table descriptors from just bit [54] to bits [54:53], allowing stage 2 to control execution permissions separately for EL0 and EL1. Implement the new semantics of the XN field and enable the feature for our 'max' CPU. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200330210400.11724-5-peter.maydell@linaro.org --- target/arm/cpu.h | 15 +++++++++++++++ target/arm/cpu.c | 1 + target/arm/cpu64.c | 2 ++ target/arm/helper.c | 37 +++++++++++++++++++++++++++++++------ 4 files changed, 49 insertions(+), 6 deletions(-) diff --git a/target/arm/cpu.h b/target/arm/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.h +++ b/target/arm/cpu.h @@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa32_ccidx(const ARMISARegisters *id) return FIELD_EX32(id->id_mmfr4, ID_MMFR4, CCIDX) != 0; } +static inline bool isar_feature_aa32_tts2uxn(const ARMISARegisters *id) +{ + return FIELD_EX32(id->id_mmfr4, ID_MMFR4, XNX) != 0; +} + /* * 64-bit feature tests via id registers. */ @@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_aa64_ccidx(const ARMISARegisters *id) return FIELD_EX64(id->id_aa64mmfr2, ID_AA64MMFR2, CCIDX) != 0; } +static inline bool isar_feature_aa64_tts2uxn(const ARMISARegisters *id) +{ + return FIELD_EX64(id->id_aa64mmfr1, ID_AA64MMFR1, XNX) != 0; +} + /* * Feature tests for "does this exist in either 32-bit or 64-bit?" */ @@ -XXX,XX +XXX,XX @@ static inline bool isar_feature_any_ccidx(const ARMISARegisters *id) return isar_feature_aa64_ccidx(id) || isar_feature_aa32_ccidx(id); } +static inline bool isar_feature_any_tts2uxn(const ARMISARegisters *id) +{ + return isar_feature_aa64_tts2uxn(id) || isar_feature_aa32_tts2uxn(id); +} + /* * Forward to the above feature tests given an ARMCPU pointer. */ diff --git a/target/arm/cpu.c b/target/arm/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.c +++ b/target/arm/cpu.c @@ -XXX,XX +XXX,XX @@ static void arm_max_initfn(Object *obj) t = FIELD_DP32(t, ID_MMFR4, HPDS, 1); /* AA32HPD */ t = FIELD_DP32(t, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ t = FIELD_DP32(t, ID_MMFR4, CNP, 1); /* TTCNP */ + t = FIELD_DP32(t, ID_MMFR4, XNX, 1); /* TTS2UXN */ cpu->isar.id_mmfr4 = t; } #endif diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) t = FIELD_DP64(t, ID_AA64MMFR1, VH, 1); t = FIELD_DP64(t, ID_AA64MMFR1, PAN, 2); /* ATS1E1 */ t = FIELD_DP64(t, ID_AA64MMFR1, VMIDBITS, 2); /* VMID16 */ + t = FIELD_DP64(t, ID_AA64MMFR1, XNX, 1); /* TTS2UXN */ cpu->isar.id_aa64mmfr1 = t; t = cpu->isar.id_aa64mmfr2; @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) u = FIELD_DP32(u, ID_MMFR4, HPDS, 1); /* AA32HPD */ u = FIELD_DP32(u, ID_MMFR4, AC2, 1); /* ACTLR2, HACTLR2 */ u = FIELD_DP32(u, ID_MMFR4, CNP, 1); /* TTCNP */ + u = FIELD_DP32(u, ID_MMFR4, XNX, 1); /* TTS2UXN */ cpu->isar.id_mmfr4 = u; u = cpu->isar.id_aa64dfr0; diff --git a/target/arm/helper.c b/target/arm/helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.c +++ b/target/arm/helper.c @@ -XXX,XX +XXX,XX @@ simple_ap_to_rw_prot(CPUARMState *env, ARMMMUIdx mmu_idx, int ap) * * @env: CPUARMState * @s2ap: The 2-bit stage2 access permissions (S2AP) - * @xn: XN (execute-never) bit + * @xn: XN (execute-never) bits + * @s1_is_el0: true if this is S2 of an S1+2 walk for EL0 */ -static int get_S2prot(CPUARMState *env, int s2ap, int xn) +static int get_S2prot(CPUARMState *env, int s2ap, int xn, bool s1_is_el0) { int prot = 0; @@ -XXX,XX +XXX,XX @@ static int get_S2prot(CPUARMState *env, int s2ap, int xn) if (s2ap & 2) { prot |= PAGE_WRITE; } - if (!xn) { - if (arm_el_is_aa64(env, 2) || prot & PAGE_READ) { + + if (cpu_isar_feature(any_tts2uxn, env_archcpu(env))) { + switch (xn) { + case 0: prot |= PAGE_EXEC; + break; + case 1: + if (s1_is_el0) { + prot |= PAGE_EXEC; + } + break; + case 2: + break; + case 3: + if (!s1_is_el0) { + prot |= PAGE_EXEC; + } + break; + default: + g_assert_not_reached(); + } + } else { + if (!extract32(xn, 1, 1)) { + if (arm_el_is_aa64(env, 2) || prot & PAGE_READ) { + prot |= PAGE_EXEC; + } } } return prot; @@ -XXX,XX +XXX,XX @@ static bool get_phys_addr_lpae(CPUARMState *env, target_ulong address, } ap = extract32(attrs, 4, 2); - xn = extract32(attrs, 12, 1); if (mmu_idx == ARMMMUIdx_Stage2) { ns = true; - *prot = get_S2prot(env, ap, xn); + xn = extract32(attrs, 11, 2); + *prot = get_S2prot(env, ap, xn, s1_is_el0); } else { ns = extract32(attrs, 3, 1); + xn = extract32(attrs, 12, 1); pxn = extract32(attrs, 11, 1); *prot = get_S1prot(env, mmu_idx, aarch64, ap, ns, xn, pxn); } -- 2.20.1 In aarch64_max_initfn() we update both 32-bit and 64-bit ID registers. The intended pattern is that for 64-bit ID registers we use FIELD_DP64 and the uint64_t 't' register, while 32-bit ID registers use FIELD_DP32 and the uint32_t 'u' register. For ID_AA64DFR0 we accidentally used 'u', meaning that the top 32 bits of this 64-bit ID register would end up always zero. Luckily at the moment that's what they should be anyway, so this bug has no visible effects. Use the right-sized variable. Fixes: 3bec78447a958d481991 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Laurent Desnogues <laurent.desnogues@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 20200423110915.10527-1-peter.maydell@linaro.org --- target/arm/cpu64.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/target/arm/cpu64.c b/target/arm/cpu64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu64.c +++ b/target/arm/cpu64.c @@ -XXX,XX +XXX,XX @@ static void aarch64_max_initfn(Object *obj) u = FIELD_DP32(u, ID_MMFR4, XNX, 1); /* TTS2UXN */ cpu->isar.id_mmfr4 = u; - u = cpu->isar.id_aa64dfr0; - u = FIELD_DP64(u, ID_AA64DFR0, PMUVER, 5); /* v8.4-PMU */ - cpu->isar.id_aa64dfr0 = u; + t = cpu->isar.id_aa64dfr0; + t = FIELD_DP64(t, ID_AA64DFR0, PMUVER, 5); /* v8.4-PMU */ + cpu->isar.id_aa64dfr0 = t; u = cpu->isar.id_dfr0; u = FIELD_DP32(u, ID_DFR0, PERFMON, 5); /* v8.4-PMU */ -- 2.20.1 From: Philippe Mathieu-Daudé <f4bug@amsat.org> MIDR_EL1 is a 64-bit system register with the top 32-bit being RES0. Represent it in QEMU's ARMCPU struct with a uint64_t, not a uint32_t. This fixes an error when compiling with -Werror=conversion because we were manipulating the register value using a local uint64_t variable: target/arm/cpu64.c: In function ‘aarch64_max_initfn’: target/arm/cpu64.c:628:21: error: conversion from ‘uint64_t’ {aka ‘long unsigned int’} to ‘uint32_t’ {aka ‘unsigned int’} may change value [-Werror=conversion] 628 | cpu->midr = t; | ^ and future-proofs us against a possible future architecture change using some of the top 32 bits. Suggested-by: Laurent Desnogues <laurent.desnogues@gmail.com> Suggested-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Laurent Desnogues <laurent.desnogues@gmail.com> Message-id: 20200428172634.29707-1-f4bug@amsat.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.h | 2 +- target/arm/cpu.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/target/arm/cpu.h b/target/arm/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.h +++ b/target/arm/cpu.h @@ -XXX,XX +XXX,XX @@ struct ARMCPU { uint64_t id_aa64dfr0; uint64_t id_aa64dfr1; } isar; - uint32_t midr; + uint64_t midr; uint32_t revidr; uint32_t reset_fpsid; uint32_t ctr; diff --git a/target/arm/cpu.c b/target/arm/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu.c +++ b/target/arm/cpu.c @@ -XXX,XX +XXX,XX @@ static const ARMCPUInfo arm_cpus[] = { static Property arm_cpu_properties[] = { DEFINE_PROP_BOOL("start-powered-off", ARMCPU, start_powered_off, false), DEFINE_PROP_UINT32("psci-conduit", ARMCPU, psci_conduit, 0), - DEFINE_PROP_UINT32("midr", ARMCPU, midr, 0), + DEFINE_PROP_UINT64("midr", ARMCPU, midr, 0), DEFINE_PROP_UINT64("mp-affinity", ARMCPU, mp_affinity, ARM64_AFFINITY_INVALID), DEFINE_PROP_INT32("node-id", ARMCPU, node_id, CPU_UNSET_NUMA_NODE_ID), -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Remove inclusion of arm_gicv3_common.h, this already gets included via xlnx-versal.h. Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Luc Michel <luc.michel@greensocs.com> Message-id: 20200427181649.26851-2-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/xlnx-versal.c | 1 - 1 file changed, 1 deletion(-) 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 "hw/arm/boot.h" #include "kvm_arm.h" #include "hw/misc/unimp.h" -#include "hw/intc/arm_gicv3_common.h" #include "hw/arm/xlnx-versal.h" #include "hw/char/pl011.h" -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Move misplaced comment. Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Luc Michel <luc.michel@greensocs.com> Message-id: 20200427181649.26851-3-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/xlnx-versal.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) 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 @@ static void versal_create_apu_cpus(Versal *s) obj = object_new(XLNX_VERSAL_ACPU_TYPE); if (!obj) { - /* Secondary CPUs start in PSCI powered-down state */ error_report("Unable to create apu.cpu[%d] of type %s", i, XLNX_VERSAL_ACPU_TYPE); exit(EXIT_FAILURE); @@ -XXX,XX +XXX,XX @@ static void versal_create_apu_cpus(Versal *s) object_property_set_int(obj, s->cfg.psci_conduit, "psci-conduit", &error_abort); if (i) { + /* Secondary CPUs start in PSCI powered-down state */ object_property_set_bool(obj, true, "start-powered-off", &error_abort); } -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Fix typo xlnx-ve -> xlnx-versal. Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Luc Michel <luc.michel@greensocs.com> Message-id: 20200427181649.26851-4-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/xlnx-versal-virt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/hw/arm/xlnx-versal-virt.c b/hw/arm/xlnx-versal-virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/xlnx-versal-virt.c +++ b/hw/arm/xlnx-versal-virt.c @@ -XXX,XX +XXX,XX @@ static void versal_virt_init(MachineState *machine) psci_conduit = QEMU_PSCI_CONDUIT_SMC; } - sysbus_init_child_obj(OBJECT(machine), "xlnx-ve", &s->soc, + sysbus_init_child_obj(OBJECT(machine), "xlnx-versal", &s->soc, sizeof(s->soc), TYPE_XLNX_VERSAL); object_property_set_link(OBJECT(&s->soc), OBJECT(machine->ram), "ddr", &error_abort); -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Embed the UARTs into the SoC type. Suggested-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Luc Michel <luc.michel@greensocs.com> Message-id: 20200427181649.26851-5-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/xlnx-versal.h | 3 ++- hw/arm/xlnx-versal.c | 12 ++++++------ 2 files changed, 8 insertions(+), 7 deletions(-) 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/intc/arm_gicv3.h" +#include "hw/char/pl011.h" #define TYPE_XLNX_VERSAL "xlnx-versal" #define XLNX_VERSAL(obj) OBJECT_CHECK(Versal, (obj), TYPE_XLNX_VERSAL) @@ -XXX,XX +XXX,XX @@ typedef struct Versal { MemoryRegion mr_ocm; struct { - SysBusDevice *uart[XLNX_VERSAL_NR_UARTS]; + PL011State uart[XLNX_VERSAL_NR_UARTS]; SysBusDevice *gem[XLNX_VERSAL_NR_GEMS]; SysBusDevice *adma[XLNX_VERSAL_NR_ADMAS]; } iou; 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 "kvm_arm.h" #include "hw/misc/unimp.h" #include "hw/arm/xlnx-versal.h" -#include "hw/char/pl011.h" #define XLNX_VERSAL_ACPU_TYPE ARM_CPU_TYPE_NAME("cortex-a72") #define GEM_REVISION 0x40070106 @@ -XXX,XX +XXX,XX @@ static void versal_create_uarts(Versal *s, qemu_irq *pic) DeviceState *dev; MemoryRegion *mr; - dev = qdev_create(NULL, TYPE_PL011); - s->lpd.iou.uart[i] = SYS_BUS_DEVICE(dev); + sysbus_init_child_obj(OBJECT(s), name, + &s->lpd.iou.uart[i], sizeof(s->lpd.iou.uart[i]), + TYPE_PL011); + dev = DEVICE(&s->lpd.iou.uart[i]); qdev_prop_set_chr(dev, "chardev", serial_hd(i)); - object_property_add_child(OBJECT(s), name, OBJECT(dev), &error_fatal); qdev_init_nofail(dev); - mr = sysbus_mmio_get_region(s->lpd.iou.uart[i], 0); + mr = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0); memory_region_add_subregion(&s->mr_ps, addrs[i], mr); - sysbus_connect_irq(s->lpd.iou.uart[i], 0, pic[irqs[i]]); + sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0, pic[irqs[i]]); g_free(name); } } -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Embed the GEMs into the SoC type. Suggested-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Luc Michel <luc.michel@greensocs.com> Message-id: 20200427181649.26851-6-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/xlnx-versal.h | 3 ++- hw/arm/xlnx-versal.c | 15 ++++++++------- 2 files changed, 10 insertions(+), 8 deletions(-) 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/arm/boot.h" #include "hw/intc/arm_gicv3.h" #include "hw/char/pl011.h" +#include "hw/net/cadence_gem.h" #define TYPE_XLNX_VERSAL "xlnx-versal" #define XLNX_VERSAL(obj) OBJECT_CHECK(Versal, (obj), TYPE_XLNX_VERSAL) @@ -XXX,XX +XXX,XX @@ typedef struct Versal { struct { PL011State uart[XLNX_VERSAL_NR_UARTS]; - SysBusDevice *gem[XLNX_VERSAL_NR_GEMS]; + CadenceGEMState gem[XLNX_VERSAL_NR_GEMS]; SysBusDevice *adma[XLNX_VERSAL_NR_ADMAS]; } iou; } lpd; 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 @@ static void versal_create_gems(Versal *s, qemu_irq *pic) DeviceState *dev; MemoryRegion *mr; - dev = qdev_create(NULL, "cadence_gem"); - s->lpd.iou.gem[i] = SYS_BUS_DEVICE(dev); - object_property_add_child(OBJECT(s), name, OBJECT(dev), &error_fatal); + sysbus_init_child_obj(OBJECT(s), name, + &s->lpd.iou.gem[i], sizeof(s->lpd.iou.gem[i]), + TYPE_CADENCE_GEM); + dev = DEVICE(&s->lpd.iou.gem[i]); if (nd->used) { qemu_check_nic_model(nd, "cadence_gem"); qdev_set_nic_properties(dev, nd); } - object_property_set_int(OBJECT(s->lpd.iou.gem[i]), + object_property_set_int(OBJECT(dev), 2, "num-priority-queues", &error_abort); - object_property_set_link(OBJECT(s->lpd.iou.gem[i]), + object_property_set_link(OBJECT(dev), OBJECT(&s->mr_ps), "dma", &error_abort); qdev_init_nofail(dev); - mr = sysbus_mmio_get_region(s->lpd.iou.gem[i], 0); + mr = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0); memory_region_add_subregion(&s->mr_ps, addrs[i], mr); - sysbus_connect_irq(s->lpd.iou.gem[i], 0, pic[irqs[i]]); + sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0, pic[irqs[i]]); g_free(name); } } -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Embed the ADMAs into the SoC type. Suggested-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Luc Michel <luc.michel@greensocs.com> Message-id: 20200427181649.26851-7-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/xlnx-versal.h | 3 ++- hw/arm/xlnx-versal.c | 14 +++++++------- 2 files changed, 9 insertions(+), 8 deletions(-) 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/arm/boot.h" #include "hw/intc/arm_gicv3.h" #include "hw/char/pl011.h" +#include "hw/dma/xlnx-zdma.h" #include "hw/net/cadence_gem.h" #define TYPE_XLNX_VERSAL "xlnx-versal" @@ -XXX,XX +XXX,XX @@ typedef struct Versal { struct { PL011State uart[XLNX_VERSAL_NR_UARTS]; CadenceGEMState gem[XLNX_VERSAL_NR_GEMS]; - SysBusDevice *adma[XLNX_VERSAL_NR_ADMAS]; + XlnxZDMA adma[XLNX_VERSAL_NR_ADMAS]; } iou; } lpd; 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 @@ static void versal_create_admas(Versal *s, qemu_irq *pic) DeviceState *dev; MemoryRegion *mr; - dev = qdev_create(NULL, "xlnx.zdma"); - s->lpd.iou.adma[i] = SYS_BUS_DEVICE(dev); - object_property_set_int(OBJECT(s->lpd.iou.adma[i]), 128, "bus-width", - &error_abort); - object_property_add_child(OBJECT(s), name, OBJECT(dev), &error_fatal); + sysbus_init_child_obj(OBJECT(s), name, + &s->lpd.iou.adma[i], sizeof(s->lpd.iou.adma[i]), + TYPE_XLNX_ZDMA); + dev = DEVICE(&s->lpd.iou.adma[i]); + object_property_set_int(OBJECT(dev), 128, "bus-width", &error_abort); qdev_init_nofail(dev); - mr = sysbus_mmio_get_region(s->lpd.iou.adma[i], 0); + mr = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0); memory_region_add_subregion(&s->mr_ps, MM_ADMA_CH0 + i * MM_ADMA_CH0_SIZE, mr); - sysbus_connect_irq(s->lpd.iou.adma[i], 0, pic[VERSAL_ADMA_IRQ_0 + i]); + sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0, pic[VERSAL_ADMA_IRQ_0 + i]); g_free(name); } } -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Embed the APUs into the SoC type. Suggested-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Luc Michel <luc.michel@greensocs.com> Message-id: 20200427181649.26851-8-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/xlnx-versal.h | 2 +- hw/arm/xlnx-versal-virt.c | 4 ++-- hw/arm/xlnx-versal.c | 19 +++++-------------- 3 files changed, 8 insertions(+), 17 deletions(-) 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 @@ typedef struct Versal { struct { struct { MemoryRegion mr; - ARMCPU *cpu[XLNX_VERSAL_NR_ACPUS]; + ARMCPU cpu[XLNX_VERSAL_NR_ACPUS]; GICv3State gic; } apu; } fpd; diff --git a/hw/arm/xlnx-versal-virt.c b/hw/arm/xlnx-versal-virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/xlnx-versal-virt.c +++ b/hw/arm/xlnx-versal-virt.c @@ -XXX,XX +XXX,XX @@ static void versal_virt_init(MachineState *machine) s->binfo.get_dtb = versal_virt_get_dtb; s->binfo.modify_dtb = versal_virt_modify_dtb; if (machine->kernel_filename) { - arm_load_kernel(s->soc.fpd.apu.cpu[0], machine, &s->binfo); + arm_load_kernel(&s->soc.fpd.apu.cpu[0], machine, &s->binfo); } else { - AddressSpace *as = arm_boot_address_space(s->soc.fpd.apu.cpu[0], + AddressSpace *as = arm_boot_address_space(&s->soc.fpd.apu.cpu[0], &s->binfo); /* Some boot-loaders (e.g u-boot) don't like blobs at address 0 (NULL). * Offset things by 4K. */ 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 @@ static void versal_create_apu_cpus(Versal *s) for (i = 0; i < ARRAY_SIZE(s->fpd.apu.cpu); i++) { Object *obj; - char *name; - - obj = object_new(XLNX_VERSAL_ACPU_TYPE); - if (!obj) { - error_report("Unable to create apu.cpu[%d] of type %s", - i, XLNX_VERSAL_ACPU_TYPE); - exit(EXIT_FAILURE); - } - - name = g_strdup_printf("apu-cpu[%d]", i); - object_property_add_child(OBJECT(s), name, obj, &error_fatal); - g_free(name); + object_initialize_child(OBJECT(s), "apu-cpu[*]", + &s->fpd.apu.cpu[i], sizeof(s->fpd.apu.cpu[i]), + XLNX_VERSAL_ACPU_TYPE, &error_abort, NULL); + obj = OBJECT(&s->fpd.apu.cpu[i]); object_property_set_int(obj, s->cfg.psci_conduit, "psci-conduit", &error_abort); if (i) { @@ -XXX,XX +XXX,XX @@ static void versal_create_apu_cpus(Versal *s) object_property_set_link(obj, OBJECT(&s->fpd.apu.mr), "memory", &error_abort); object_property_set_bool(obj, true, "realized", &error_fatal); - s->fpd.apu.cpu[i] = ARM_CPU(obj); } } @@ -XXX,XX +XXX,XX @@ static void versal_create_apu_gic(Versal *s, qemu_irq *pic) } for (i = 0; i < nr_apu_cpus; i++) { - DeviceState *cpudev = DEVICE(s->fpd.apu.cpu[i]); + DeviceState *cpudev = DEVICE(&s->fpd.apu.cpu[i]); int ppibase = XLNX_VERSAL_NR_IRQS + i * GIC_INTERNAL + GIC_NR_SGIS; qemu_irq maint_irq; int ti; -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Add support for SD. Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Luc Michel <luc.michel@greensocs.com> Message-id: 20200427181649.26851-9-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/xlnx-versal.h | 12 ++++++++++++ hw/arm/xlnx-versal.c | 31 +++++++++++++++++++++++++++++++ 2 files changed, 43 insertions(+) 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/sd/sdhci.h" #include "hw/intc/arm_gicv3.h" #include "hw/char/pl011.h" #include "hw/dma/xlnx-zdma.h" @@ -XXX,XX +XXX,XX @@ #define XLNX_VERSAL_NR_UARTS 2 #define XLNX_VERSAL_NR_GEMS 2 #define XLNX_VERSAL_NR_ADMAS 8 +#define XLNX_VERSAL_NR_SDS 2 #define XLNX_VERSAL_NR_IRQS 192 typedef struct Versal { @@ -XXX,XX +XXX,XX @@ typedef struct Versal { } iou; } lpd; + /* The Platform Management Controller subsystem. */ + struct { + struct { + SDHCIState sd[XLNX_VERSAL_NR_SDS]; + } iou; + } pmc; + struct { MemoryRegion *mr_ddr; uint32_t psci_conduit; @@ -XXX,XX +XXX,XX @@ typedef struct Versal { #define VERSAL_GEM1_IRQ_0 58 #define VERSAL_GEM1_WAKE_IRQ_0 59 #define VERSAL_ADMA_IRQ_0 60 +#define VERSAL_SD0_IRQ_0 126 /* Architecturally reserved IRQs suitable for virtualization. */ #define VERSAL_RSVD_IRQ_FIRST 111 @@ -XXX,XX +XXX,XX @@ typedef struct Versal { #define MM_FPD_CRF 0xfd1a0000U #define MM_FPD_CRF_SIZE 0x140000 +#define MM_PMC_SD0 0xf1040000U +#define MM_PMC_SD0_SIZE 0x10000 #define MM_PMC_CRP 0xf1260000U #define MM_PMC_CRP_SIZE 0x10000 #endif 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 @@ static void versal_create_admas(Versal *s, qemu_irq *pic) } } +#define SDHCI_CAPABILITIES 0x280737ec6481 /* Same as on ZynqMP. */ +static void versal_create_sds(Versal *s, qemu_irq *pic) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(s->pmc.iou.sd); i++) { + DeviceState *dev; + MemoryRegion *mr; + + sysbus_init_child_obj(OBJECT(s), "sd[*]", + &s->pmc.iou.sd[i], sizeof(s->pmc.iou.sd[i]), + TYPE_SYSBUS_SDHCI); + dev = DEVICE(&s->pmc.iou.sd[i]); + + object_property_set_uint(OBJECT(dev), + 3, "sd-spec-version", &error_fatal); + object_property_set_uint(OBJECT(dev), SDHCI_CAPABILITIES, "capareg", + &error_fatal); + object_property_set_uint(OBJECT(dev), UHS_I, "uhs", &error_fatal); + qdev_init_nofail(dev); + + mr = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0); + memory_region_add_subregion(&s->mr_ps, + MM_PMC_SD0 + i * MM_PMC_SD0_SIZE, mr); + + sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0, + pic[VERSAL_SD0_IRQ_0 + i * 2]); + } +} + /* 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_uarts(s, pic); versal_create_gems(s, pic); versal_create_admas(s, pic); + versal_create_sds(s, pic); versal_map_ddr(s); versal_unimp(s); -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> hw/arm: versal: Add support for the RTC. Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Luc Michel <luc.michel@greensocs.com> Message-id: 20200427181649.26851-10-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/xlnx-versal.h | 8 ++++++++ hw/arm/xlnx-versal.c | 21 +++++++++++++++++++++ 2 files changed, 29 insertions(+) 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/char/pl011.h" #include "hw/dma/xlnx-zdma.h" #include "hw/net/cadence_gem.h" +#include "hw/rtc/xlnx-zynqmp-rtc.h" #define TYPE_XLNX_VERSAL "xlnx-versal" #define XLNX_VERSAL(obj) OBJECT_CHECK(Versal, (obj), TYPE_XLNX_VERSAL) @@ -XXX,XX +XXX,XX @@ typedef struct Versal { struct { SDHCIState sd[XLNX_VERSAL_NR_SDS]; } iou; + + XlnxZynqMPRTC rtc; } pmc; struct { @@ -XXX,XX +XXX,XX @@ typedef struct Versal { #define VERSAL_GEM1_IRQ_0 58 #define VERSAL_GEM1_WAKE_IRQ_0 59 #define VERSAL_ADMA_IRQ_0 60 +#define VERSAL_RTC_APB_ERR_IRQ 121 #define VERSAL_SD0_IRQ_0 126 +#define VERSAL_RTC_ALARM_IRQ 142 +#define VERSAL_RTC_SECONDS_IRQ 143 /* Architecturally reserved IRQs suitable for virtualization. */ #define VERSAL_RSVD_IRQ_FIRST 111 @@ -XXX,XX +XXX,XX @@ typedef struct Versal { #define MM_PMC_SD0_SIZE 0x10000 #define MM_PMC_CRP 0xf1260000U #define MM_PMC_CRP_SIZE 0x10000 +#define MM_PMC_RTC 0xf12a0000 +#define MM_PMC_RTC_SIZE 0x10000 #endif 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 @@ static void versal_create_sds(Versal *s, qemu_irq *pic) } } +static void versal_create_rtc(Versal *s, qemu_irq *pic) +{ + SysBusDevice *sbd; + MemoryRegion *mr; + + sysbus_init_child_obj(OBJECT(s), "rtc", &s->pmc.rtc, sizeof(s->pmc.rtc), + TYPE_XLNX_ZYNQMP_RTC); + sbd = SYS_BUS_DEVICE(&s->pmc.rtc); + qdev_init_nofail(DEVICE(sbd)); + + mr = sysbus_mmio_get_region(sbd, 0); + memory_region_add_subregion(&s->mr_ps, MM_PMC_RTC, mr); + + /* + * TODO: Connect the ALARM and SECONDS interrupts once our RTC model + * supports them. + */ + sysbus_connect_irq(sbd, 1, pic[VERSAL_RTC_APB_ERR_IRQ]); +} + /* 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_gems(s, pic); versal_create_admas(s, pic); versal_create_sds(s, pic); + versal_create_rtc(s, pic); versal_map_ddr(s); versal_unimp(s); -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Add support for SD. Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Luc Michel <luc.michel@greensocs.com> Message-id: 20200427181649.26851-11-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/xlnx-versal-virt.c | 46 +++++++++++++++++++++++++++++++++++++++ 1 file changed, 46 insertions(+) diff --git a/hw/arm/xlnx-versal-virt.c b/hw/arm/xlnx-versal-virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/xlnx-versal-virt.c +++ b/hw/arm/xlnx-versal-virt.c @@ -XXX,XX +XXX,XX @@ #include "hw/arm/sysbus-fdt.h" #include "hw/arm/fdt.h" #include "cpu.h" +#include "hw/qdev-properties.h" #include "hw/arm/xlnx-versal.h" #define TYPE_XLNX_VERSAL_VIRT_MACHINE MACHINE_TYPE_NAME("xlnx-versal-virt") @@ -XXX,XX +XXX,XX @@ static void fdt_add_zdma_nodes(VersalVirt *s) } } +static void fdt_add_sd_nodes(VersalVirt *s) +{ + const char clocknames[] = "clk_xin\0clk_ahb"; + const char compat[] = "arasan,sdhci-8.9a"; + int i; + + for (i = ARRAY_SIZE(s->soc.pmc.iou.sd) - 1; i >= 0; i--) { + uint64_t addr = MM_PMC_SD0 + MM_PMC_SD0_SIZE * i; + char *name = g_strdup_printf("/sdhci@%" PRIx64, addr); + + qemu_fdt_add_subnode(s->fdt, name); + + qemu_fdt_setprop_cells(s->fdt, name, "clocks", + s->phandle.clk_25Mhz, s->phandle.clk_25Mhz); + qemu_fdt_setprop(s->fdt, name, "clock-names", + clocknames, sizeof(clocknames)); + qemu_fdt_setprop_cells(s->fdt, name, "interrupts", + GIC_FDT_IRQ_TYPE_SPI, VERSAL_SD0_IRQ_0 + i * 2, + GIC_FDT_IRQ_FLAGS_LEVEL_HI); + qemu_fdt_setprop_sized_cells(s->fdt, name, "reg", + 2, addr, 2, MM_PMC_SD0_SIZE); + qemu_fdt_setprop(s->fdt, name, "compatible", compat, sizeof(compat)); + g_free(name); + } +} + static void fdt_nop_memory_nodes(void *fdt, Error **errp) { Error *err = NULL; @@ -XXX,XX +XXX,XX @@ static void create_virtio_regions(VersalVirt *s) } } +static void sd_plugin_card(SDHCIState *sd, DriveInfo *di) +{ + BlockBackend *blk = di ? blk_by_legacy_dinfo(di) : NULL; + DeviceState *card; + + card = qdev_create(qdev_get_child_bus(DEVICE(sd), "sd-bus"), TYPE_SD_CARD); + object_property_add_child(OBJECT(sd), "card[*]", OBJECT(card), + &error_fatal); + qdev_prop_set_drive(card, "drive", blk, &error_fatal); + object_property_set_bool(OBJECT(card), true, "realized", &error_fatal); +} + static void versal_virt_init(MachineState *machine) { VersalVirt *s = XLNX_VERSAL_VIRT_MACHINE(machine); int psci_conduit = QEMU_PSCI_CONDUIT_DISABLED; + int i; /* * If the user provides an Operating System to be loaded, we expect them @@ -XXX,XX +XXX,XX @@ static void versal_virt_init(MachineState *machine) fdt_add_gic_nodes(s); fdt_add_timer_nodes(s); fdt_add_zdma_nodes(s); + fdt_add_sd_nodes(s); fdt_add_cpu_nodes(s, psci_conduit); fdt_add_clk_node(s, "/clk125", 125000000, s->phandle.clk_125Mhz); fdt_add_clk_node(s, "/clk25", 25000000, s->phandle.clk_25Mhz); @@ -XXX,XX +XXX,XX @@ static void versal_virt_init(MachineState *machine) memory_region_add_subregion_overlap(get_system_memory(), 0, &s->soc.fpd.apu.mr, 0); + /* Plugin SD cards. */ + for (i = 0; i < ARRAY_SIZE(s->soc.pmc.iou.sd); i++) { + sd_plugin_card(&s->soc.pmc.iou.sd[i], drive_get_next(IF_SD)); + } + s->binfo.ram_size = machine->ram_size; s->binfo.loader_start = 0x0; s->binfo.get_dtb = versal_virt_get_dtb; -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Add support for the RTC. Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Luc Michel <luc.michel@greensocs.com> Message-id: 20200427181649.26851-12-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/xlnx-versal-virt.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) diff --git a/hw/arm/xlnx-versal-virt.c b/hw/arm/xlnx-versal-virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/xlnx-versal-virt.c +++ b/hw/arm/xlnx-versal-virt.c @@ -XXX,XX +XXX,XX @@ static void fdt_add_sd_nodes(VersalVirt *s) } } +static void fdt_add_rtc_node(VersalVirt *s) +{ + const char compat[] = "xlnx,zynqmp-rtc"; + const char interrupt_names[] = "alarm\0sec"; + char *name = g_strdup_printf("/rtc@%x", MM_PMC_RTC); + + qemu_fdt_add_subnode(s->fdt, name); + + qemu_fdt_setprop_cells(s->fdt, name, "interrupts", + GIC_FDT_IRQ_TYPE_SPI, VERSAL_RTC_ALARM_IRQ, + GIC_FDT_IRQ_FLAGS_LEVEL_HI, + GIC_FDT_IRQ_TYPE_SPI, VERSAL_RTC_SECONDS_IRQ, + GIC_FDT_IRQ_FLAGS_LEVEL_HI); + qemu_fdt_setprop(s->fdt, name, "interrupt-names", + interrupt_names, sizeof(interrupt_names)); + qemu_fdt_setprop_sized_cells(s->fdt, name, "reg", + 2, MM_PMC_RTC, 2, MM_PMC_RTC_SIZE); + qemu_fdt_setprop(s->fdt, name, "compatible", compat, sizeof(compat)); + g_free(name); +} + static void fdt_nop_memory_nodes(void *fdt, Error **errp) { Error *err = NULL; @@ -XXX,XX +XXX,XX @@ static void versal_virt_init(MachineState *machine) fdt_add_timer_nodes(s); fdt_add_zdma_nodes(s); fdt_add_sd_nodes(s); + fdt_add_rtc_node(s); fdt_add_cpu_nodes(s, psci_conduit); fdt_add_clk_node(s, "/clk125", 125000000, s->phandle.clk_125Mhz); fdt_add_clk_node(s, "/clk25", 25000000, s->phandle.clk_25Mhz); -- 2.20.1 Somewhere along theline we accidentally added a duplicate "using D16-D31 when they don't exist" check to do_vfm_dp() (probably an artifact of a patchseries rebase). Remove it. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 20200430181003.21682-2-peter.maydell@linaro.org --- target/arm/translate-vfp.inc.c | 6 ------ 1 file changed, 6 deletions(-) diff --git a/target/arm/translate-vfp.inc.c b/target/arm/translate-vfp.inc.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-vfp.inc.c +++ b/target/arm/translate-vfp.inc.c @@ -XXX,XX +XXX,XX @@ static bool do_vfm_dp(DisasContext *s, arg_VFMA_dp *a, bool neg_n, bool neg_d) 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 (!vfp_access_check(s)) { return true; } -- 2.20.1 We were accidentally permitting decode of Thumb Neon insns even if the CPU didn't have the FEATURE_NEON bit set, because the feature check was being done before the call to disas_neon_data_insn() and disas_neon_ls_insn() in the Arm decoder but was omitted from the Thumb decoder. Push the feature bit check down into the called functions so it is done for both Arm and Thumb encodings. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 20200430181003.21682-3-peter.maydell@linaro.org --- target/arm/translate.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) 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_ls_insn(DisasContext *s, uint32_t insn) TCGv_i32 tmp2; TCGv_i64 tmp64; + if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { + return 1; + } + /* FIXME: this access check should not take precedence over UNDEF * for invalid encodings; we will generate incorrect syndrome information * for attempts to execute invalid vfp/neon encodings with FP disabled. @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) TCGv_ptr ptr1, ptr2, ptr3; TCGv_i64 tmp64; + if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { + return 1; + } + /* FIXME: this access check should not take precedence over UNDEF * for invalid encodings; we will generate incorrect syndrome information * for attempts to execute invalid vfp/neon encodings with FP disabled. @@ -XXX,XX +XXX,XX @@ static void disas_arm_insn(DisasContext *s, unsigned int insn) if (((insn >> 25) & 7) == 1) { /* NEON Data processing. */ - if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { - goto illegal_op; - } - if (disas_neon_data_insn(s, insn)) { goto illegal_op; } @@ -XXX,XX +XXX,XX @@ static void disas_arm_insn(DisasContext *s, unsigned int insn) } if ((insn & 0x0f100000) == 0x04000000) { /* NEON load/store. */ - if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { - goto illegal_op; - } - if (disas_neon_ls_insn(s, insn)) { goto illegal_op; } -- 2.20.1 Add the infrastructure for building and invoking a decodetree decoder for the AArch32 Neon encodings. At the moment the new decoder covers nothing, so we always fall back to the existing hand-written decode. We follow the same pattern we did for the VFP decodetree conversion (commit 78e138bc1f672c145ef6ace74617d and following): code that deals with Neon will be moving gradually out to translate-neon.vfp.inc, which we #include into translate.c. In order to share the decode files between A32 and T32, we split Neon into 3 parts: * data-processing * load-store * 'shared' encodings The first two groups of instructions have similar but not identical A32 and T32 encodings, so we need to manually transform the T32 encoding into the A32 one before calling the decoder; the third group covers the Neon instructions which are identical in A32 and T32. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-4-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 29 ++++++++++++++++++++++++++ target/arm/neon-ls.decode | 29 ++++++++++++++++++++++++++ target/arm/neon-shared.decode | 27 +++++++++++++++++++++++++ target/arm/translate-neon.inc.c | 32 +++++++++++++++++++++++++++++ target/arm/translate.c | 36 +++++++++++++++++++++++++++++++-- target/arm/Makefile.objs | 18 +++++++++++++++++ 6 files changed, 169 insertions(+), 2 deletions(-) create mode 100644 target/arm/neon-dp.decode create mode 100644 target/arm/neon-ls.decode create mode 100644 target/arm/neon-shared.decode create mode 100644 target/arm/translate-neon.inc.c diff --git a/target/arm/neon-dp.decode b/target/arm/neon-dp.decode new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/target/arm/neon-dp.decode @@ -XXX,XX +XXX,XX @@ +# AArch32 Neon data-processing instruction descriptions +# +# Copyright (c) 2020 Linaro, Ltd +# +# 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/>. + +# +# This file is processed by scripts/decodetree.py +# + +# Encodings for Neon data processing instructions where the T32 encoding +# is a simple transformation of the A32 encoding. +# More specifically, this file covers instructions where the A32 encoding is +# 0b1111_001p_qqqq_qqqq_qqqq_qqqq_qqqq_qqqq +# and the T32 encoding is +# 0b111p_1111_qqqq_qqqq_qqqq_qqqq_qqqq_qqqq +# This file works on the A32 encoding only; calling code for T32 has to +# transform the insn into the A32 version first. diff --git a/target/arm/neon-ls.decode b/target/arm/neon-ls.decode new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/target/arm/neon-ls.decode @@ -XXX,XX +XXX,XX @@ +# AArch32 Neon load/store instruction descriptions +# +# Copyright (c) 2020 Linaro, Ltd +# +# 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/>. + +# +# This file is processed by scripts/decodetree.py +# + +# Encodings for Neon load/store instructions where the T32 encoding +# is a simple transformation of the A32 encoding. +# More specifically, this file covers instructions where the A32 encoding is +# 0b1111_0100_xxx0_xxxx_xxxx_xxxx_xxxx_xxxx +# and the T32 encoding is +# 0b1111_1001_xxx0_xxxx_xxxx_xxxx_xxxx_xxxx +# This file works on the A32 encoding only; calling code for T32 has to +# transform the insn into the A32 version first. diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/target/arm/neon-shared.decode @@ -XXX,XX +XXX,XX @@ +# AArch32 Neon instruction descriptions +# +# Copyright (c) 2020 Linaro, Ltd +# +# 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/>. + +# +# This file is processed by scripts/decodetree.py +# + +# Encodings for Neon instructions whose encoding is the same for +# both A32 and T32. + +# More specifically, this covers: +# 2reg scalar ext: 0b1111_1110_xxxx_xxxx_xxxx_1x0x_xxxx_xxxx +# 3same ext: 0b1111_110x_xxxx_xxxx_xxxx_1x0x_xxxx_xxxx diff --git a/target/arm/translate-neon.inc.c b/target/arm/translate-neon.inc.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/target/arm/translate-neon.inc.c @@ -XXX,XX +XXX,XX @@ +/* + * ARM translation: AArch32 Neon instructions + * + * Copyright (c) 2003 Fabrice Bellard + * Copyright (c) 2005-2007 CodeSourcery + * Copyright (c) 2007 OpenedHand, Ltd. + * Copyright (c) 2020 Linaro, Ltd. + * + * 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/>. + */ + +/* + * This file is intended to be included from translate.c; it uses + * some macros and definitions provided by that file. + * It might be possible to convert it to a standalone .c file eventually. + */ + +/* Include the generated Neon decoder */ +#include "decode-neon-dp.inc.c" +#include "decode-neon-ls.inc.c" +#include "decode-neon-shared.inc.c" 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 TCGv_ptr vfp_reg_ptr(bool dp, int reg) #define ARM_CP_RW_BIT (1 << 20) -/* Include the VFP decoder */ +/* Include the VFP and Neon decoders */ #include "translate-vfp.inc.c" +#include "translate-neon.inc.c" static inline void iwmmxt_load_reg(TCGv_i64 var, int reg) { @@ -XXX,XX +XXX,XX @@ static void disas_arm_insn(DisasContext *s, unsigned int insn) /* Unconditional instructions. */ /* TODO: Perhaps merge these into one decodetree output file. */ if (disas_a32_uncond(s, insn) || - disas_vfp_uncond(s, insn)) { + disas_vfp_uncond(s, insn) || + disas_neon_dp(s, insn) || + disas_neon_ls(s, insn) || + disas_neon_shared(s, insn)) { return; } /* fall back to legacy decoder */ @@ -XXX,XX +XXX,XX @@ static void disas_thumb2_insn(DisasContext *s, uint32_t insn) ARCH(6T2); } + if ((insn & 0xef000000) == 0xef000000) { + /* + * T32 encodings 0b111p_1111_qqqq_qqqq_qqqq_qqqq_qqqq_qqqq + * transform into + * A32 encodings 0b1111_001p_qqqq_qqqq_qqqq_qqqq_qqqq_qqqq + */ + uint32_t a32_insn = (insn & 0xe2ffffff) | + ((insn & (1 << 28)) >> 4) | (1 << 28); + + if (disas_neon_dp(s, a32_insn)) { + return; + } + } + + if ((insn & 0xff100000) == 0xf9000000) { + /* + * T32 encodings 0b1111_1001_ppp0_qqqq_qqqq_qqqq_qqqq_qqqq + * transform into + * A32 encodings 0b1111_0100_ppp0_qqqq_qqqq_qqqq_qqqq_qqqq + */ + uint32_t a32_insn = (insn & 0x00ffffff) | 0xf4000000; + + if (disas_neon_ls(s, a32_insn)) { + return; + } + } + /* * TODO: Perhaps merge these into one decodetree output file. * Note disas_vfp is written for a32 with cond field in the @@ -XXX,XX +XXX,XX @@ static void disas_thumb2_insn(DisasContext *s, uint32_t insn) */ if (disas_t32(s, insn) || disas_vfp_uncond(s, insn) || + disas_neon_shared(s, insn) || ((insn >> 28) == 0xe && disas_vfp(s, insn))) { return; } diff --git a/target/arm/Makefile.objs b/target/arm/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/target/arm/Makefile.objs +++ b/target/arm/Makefile.objs @@ -XXX,XX +XXX,XX @@ target/arm/decode-sve.inc.c: $(SRC_PATH)/target/arm/sve.decode $(DECODETREE) $(PYTHON) $(DECODETREE) --decode disas_sve -o $@ $<,\ "GEN", $(TARGET_DIR)$@) +target/arm/decode-neon-shared.inc.c: $(SRC_PATH)/target/arm/neon-shared.decode $(DECODETREE) + $(call quiet-command,\ + $(PYTHON) $(DECODETREE) --static-decode disas_neon_shared -o $@ $<,\ + "GEN", $(TARGET_DIR)$@) + +target/arm/decode-neon-dp.inc.c: $(SRC_PATH)/target/arm/neon-dp.decode $(DECODETREE) + $(call quiet-command,\ + $(PYTHON) $(DECODETREE) --static-decode disas_neon_dp -o $@ $<,\ + "GEN", $(TARGET_DIR)$@) + +target/arm/decode-neon-ls.inc.c: $(SRC_PATH)/target/arm/neon-ls.decode $(DECODETREE) + $(call quiet-command,\ + $(PYTHON) $(DECODETREE) --static-decode disas_neon_ls -o $@ $<,\ + "GEN", $(TARGET_DIR)$@) + target/arm/decode-vfp.inc.c: $(SRC_PATH)/target/arm/vfp.decode $(DECODETREE) $(call quiet-command,\ $(PYTHON) $(DECODETREE) --static-decode disas_vfp -o $@ $<,\ @@ -XXX,XX +XXX,XX @@ target/arm/decode-t16.inc.c: $(SRC_PATH)/target/arm/t16.decode $(DECODETREE) "GEN", $(TARGET_DIR)$@) target/arm/translate-sve.o: target/arm/decode-sve.inc.c +target/arm/translate.o: target/arm/decode-neon-shared.inc.c +target/arm/translate.o: target/arm/decode-neon-dp.inc.c +target/arm/translate.o: target/arm/decode-neon-ls.inc.c target/arm/translate.o: target/arm/decode-vfp.inc.c target/arm/translate.o: target/arm/decode-vfp-uncond.inc.c target/arm/translate.o: target/arm/decode-a32.inc.c -- 2.20.1 Convert the VCMLA (vector) insns in the 3same extension group to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-5-peter.maydell@linaro.org --- target/arm/neon-shared.decode | 11 ++++++++++ target/arm/translate-neon.inc.c | 37 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 11 +--------- 3 files changed, 49 insertions(+), 10 deletions(-) diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-shared.decode +++ b/target/arm/neon-shared.decode @@ -XXX,XX +XXX,XX @@ # More specifically, this covers: # 2reg scalar ext: 0b1111_1110_xxxx_xxxx_xxxx_1x0x_xxxx_xxxx # 3same ext: 0b1111_110x_xxxx_xxxx_xxxx_1x0x_xxxx_xxxx + +# VFP/Neon register fields; same as vfp.decode +%vm_dp 5:1 0:4 +%vm_sp 0:4 5:1 +%vn_dp 7:1 16:4 +%vn_sp 16:4 7:1 +%vd_dp 22:1 12:4 +%vd_sp 12:4 22:1 + +VCMLA 1111 110 rot:2 . 1 size:1 .... .... 1000 . q:1 . 0 .... \ + vm=%vm_dp vn=%vn_dp vd=%vd_dp 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 @@ #include "decode-neon-dp.inc.c" #include "decode-neon-ls.inc.c" #include "decode-neon-shared.inc.c" + +static bool trans_VCMLA(DisasContext *s, arg_VCMLA *a) +{ + int opr_sz; + TCGv_ptr fpst; + gen_helper_gvec_3_ptr *fn_gvec_ptr; + + if (!dc_isar_feature(aa32_vcma, s) + || (!a->size && !dc_isar_feature(aa32_fp16_arith, 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) & a->q) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + opr_sz = (1 + a->q) * 8; + fpst = get_fpstatus_ptr(1); + fn_gvec_ptr = a->size ? gen_helper_gvec_fcmlas : gen_helper_gvec_fcmlah; + tcg_gen_gvec_3_ptr(vfp_reg_offset(1, a->vd), + vfp_reg_offset(1, a->vn), + vfp_reg_offset(1, a->vm), + fpst, opr_sz, opr_sz, a->rot, + fn_gvec_ptr); + tcg_temp_free_ptr(fpst); + return 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_insn_3same_ext(DisasContext *s, uint32_t insn) bool is_long = false, q = extract32(insn, 6, 1); bool ptr_is_env = false; - if ((insn & 0xfe200f10) == 0xfc200800) { - /* VCMLA -- 1111 110R R.1S .... .... 1000 ...0 .... */ - int size = extract32(insn, 20, 1); - data = extract32(insn, 23, 2); /* rot */ - if (!dc_isar_feature(aa32_vcma, s) - || (!size && !dc_isar_feature(aa32_fp16_arith, s))) { - return 1; - } - fn_gvec_ptr = size ? gen_helper_gvec_fcmlas : gen_helper_gvec_fcmlah; - } else if ((insn & 0xfea00f10) == 0xfc800800) { + if ((insn & 0xfea00f10) == 0xfc800800) { /* VCADD -- 1111 110R 1.0S .... .... 1000 ...0 .... */ int size = extract32(insn, 20, 1); data = extract32(insn, 24, 1); /* rot */ -- 2.20.1 Convert the VCADD (vector) insns to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-6-peter.maydell@linaro.org --- target/arm/neon-shared.decode | 3 +++ target/arm/translate-neon.inc.c | 37 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 11 +--------- 3 files changed, 41 insertions(+), 10 deletions(-) diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-shared.decode +++ b/target/arm/neon-shared.decode @@ -XXX,XX +XXX,XX @@ VCMLA 1111 110 rot:2 . 1 size:1 .... .... 1000 . q:1 . 0 .... \ vm=%vm_dp vn=%vn_dp vd=%vd_dp + +VCADD 1111 110 rot:1 1 . 0 size:1 .... .... 1000 . q:1 . 0 .... \ + vm=%vm_dp vn=%vn_dp vd=%vd_dp 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_VCMLA(DisasContext *s, arg_VCMLA *a) tcg_temp_free_ptr(fpst); return true; } + +static bool trans_VCADD(DisasContext *s, arg_VCADD *a) +{ + int opr_sz; + TCGv_ptr fpst; + gen_helper_gvec_3_ptr *fn_gvec_ptr; + + if (!dc_isar_feature(aa32_vcma, s) + || (!a->size && !dc_isar_feature(aa32_fp16_arith, 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) & a->q) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + opr_sz = (1 + a->q) * 8; + fpst = get_fpstatus_ptr(1); + fn_gvec_ptr = a->size ? gen_helper_gvec_fcadds : gen_helper_gvec_fcaddh; + tcg_gen_gvec_3_ptr(vfp_reg_offset(1, a->vd), + vfp_reg_offset(1, a->vn), + vfp_reg_offset(1, a->vm), + fpst, opr_sz, opr_sz, a->rot, + fn_gvec_ptr); + tcg_temp_free_ptr(fpst); + return 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_insn_3same_ext(DisasContext *s, uint32_t insn) bool is_long = false, q = extract32(insn, 6, 1); bool ptr_is_env = false; - if ((insn & 0xfea00f10) == 0xfc800800) { - /* VCADD -- 1111 110R 1.0S .... .... 1000 ...0 .... */ - int size = extract32(insn, 20, 1); - data = extract32(insn, 24, 1); /* rot */ - if (!dc_isar_feature(aa32_vcma, s) - || (!size && !dc_isar_feature(aa32_fp16_arith, s))) { - return 1; - } - fn_gvec_ptr = size ? gen_helper_gvec_fcadds : gen_helper_gvec_fcaddh; - } else if ((insn & 0xfeb00f00) == 0xfc200d00) { + if ((insn & 0xfeb00f00) == 0xfc200d00) { /* V[US]DOT -- 1111 1100 0.10 .... .... 1101 .Q.U .... */ bool u = extract32(insn, 4, 1); if (!dc_isar_feature(aa32_dp, s)) { -- 2.20.1 Convert the V[US]DOT (vector) insns to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-7-peter.maydell@linaro.org --- target/arm/neon-shared.decode | 4 ++++ target/arm/translate-neon.inc.c | 32 ++++++++++++++++++++++++++++++++ target/arm/translate.c | 9 +-------- 3 files changed, 37 insertions(+), 8 deletions(-) diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-shared.decode +++ b/target/arm/neon-shared.decode @@ -XXX,XX +XXX,XX @@ VCMLA 1111 110 rot:2 . 1 size:1 .... .... 1000 . q:1 . 0 .... \ VCADD 1111 110 rot:1 1 . 0 size:1 .... .... 1000 . q:1 . 0 .... \ vm=%vm_dp vn=%vn_dp vd=%vd_dp + +# VUDOT and VSDOT +VDOT 1111 110 00 . 10 .... .... 1101 . q:1 . u:1 .... \ + vm=%vm_dp vn=%vn_dp vd=%vd_dp 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_VCADD(DisasContext *s, arg_VCADD *a) tcg_temp_free_ptr(fpst); return true; } + +static bool trans_VDOT(DisasContext *s, arg_VDOT *a) +{ + int opr_sz; + gen_helper_gvec_3 *fn_gvec; + + if (!dc_isar_feature(aa32_dp, 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) & a->q) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + opr_sz = (1 + a->q) * 8; + fn_gvec = a->u ? gen_helper_gvec_udot_b : gen_helper_gvec_sdot_b; + tcg_gen_gvec_3_ool(vfp_reg_offset(1, a->vd), + vfp_reg_offset(1, a->vn), + vfp_reg_offset(1, a->vm), + opr_sz, opr_sz, 0, fn_gvec); + return 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_insn_3same_ext(DisasContext *s, uint32_t insn) bool is_long = false, q = extract32(insn, 6, 1); bool ptr_is_env = false; - if ((insn & 0xfeb00f00) == 0xfc200d00) { - /* V[US]DOT -- 1111 1100 0.10 .... .... 1101 .Q.U .... */ - bool u = extract32(insn, 4, 1); - if (!dc_isar_feature(aa32_dp, s)) { - return 1; - } - fn_gvec = u ? gen_helper_gvec_udot_b : gen_helper_gvec_sdot_b; - } else if ((insn & 0xff300f10) == 0xfc200810) { + if ((insn & 0xff300f10) == 0xfc200810) { /* VFM[AS]L -- 1111 1100 S.10 .... .... 1000 .Q.1 .... */ int is_s = extract32(insn, 23, 1); if (!dc_isar_feature(aa32_fhm, s)) { -- 2.20.1 Convert the VFM[AS]L (vector) insns to decodetree. This is the last insn in the legacy decoder for the 3same_ext group, so we can delete the legacy decoder function for the group entirely. Note that in disas_thumb2_insn() the parts of this encoding space where the decodetree decoder returns false will correctly be directed to illegal_op by the "(insn & (1 << 28))" check so they won't fall into disas_coproc_insn() by mistake. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-8-peter.maydell@linaro.org --- target/arm/neon-shared.decode | 6 +++ target/arm/translate-neon.inc.c | 31 +++++++++++ target/arm/translate.c | 92 +-------------------------------- 3 files changed, 38 insertions(+), 91 deletions(-) diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-shared.decode +++ b/target/arm/neon-shared.decode @@ -XXX,XX +XXX,XX @@ VCADD 1111 110 rot:1 1 . 0 size:1 .... .... 1000 . q:1 . 0 .... \ # VUDOT and VSDOT VDOT 1111 110 00 . 10 .... .... 1101 . q:1 . u:1 .... \ vm=%vm_dp vn=%vn_dp vd=%vd_dp + +# VFM[AS]L +VFML 1111 110 0 s:1 . 10 .... .... 1000 . 0 . 1 .... \ + vm=%vm_sp vn=%vn_sp vd=%vd_dp q=0 +VFML 1111 110 0 s:1 . 10 .... .... 1000 . 1 . 1 .... \ + vm=%vm_dp vn=%vn_dp vd=%vd_dp q=1 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_VDOT(DisasContext *s, arg_VDOT *a) opr_sz, opr_sz, 0, fn_gvec); return true; } + +static bool trans_VFML(DisasContext *s, arg_VFML *a) +{ + int opr_sz; + + if (!dc_isar_feature(aa32_fhm, s)) { + 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; + } + + opr_sz = (1 + a->q) * 8; + tcg_gen_gvec_3_ptr(vfp_reg_offset(1, a->vd), + vfp_reg_offset(a->q, a->vn), + vfp_reg_offset(a->q, a->vm), + cpu_env, opr_sz, opr_sz, a->s, /* is_2 == 0 */ + gen_helper_gvec_fmlal_a32); + return 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) return 0; } -/* Advanced SIMD three registers of the same length extension. - * 31 25 23 22 20 16 12 11 10 9 8 3 0 - * +---------------+-----+---+-----+----+----+---+----+---+----+---------+----+ - * | 1 1 1 1 1 1 0 | op1 | D | op2 | Vn | Vd | 1 | o3 | 0 | o4 | N Q M U | Vm | - * +---------------+-----+---+-----+----+----+---+----+---+----+---------+----+ - */ -static int disas_neon_insn_3same_ext(DisasContext *s, uint32_t insn) -{ - gen_helper_gvec_3 *fn_gvec = NULL; - gen_helper_gvec_3_ptr *fn_gvec_ptr = NULL; - int rd, rn, rm, opr_sz; - int data = 0; - int off_rn, off_rm; - bool is_long = false, q = extract32(insn, 6, 1); - bool ptr_is_env = false; - - if ((insn & 0xff300f10) == 0xfc200810) { - /* VFM[AS]L -- 1111 1100 S.10 .... .... 1000 .Q.1 .... */ - int is_s = extract32(insn, 23, 1); - if (!dc_isar_feature(aa32_fhm, s)) { - return 1; - } - is_long = true; - data = is_s; /* is_2 == 0 */ - fn_gvec_ptr = gen_helper_gvec_fmlal_a32; - ptr_is_env = true; - } else { - return 1; - } - - VFP_DREG_D(rd, insn); - if (rd & q) { - return 1; - } - if (q || !is_long) { - VFP_DREG_N(rn, insn); - VFP_DREG_M(rm, insn); - if ((rn | rm) & q & !is_long) { - return 1; - } - off_rn = vfp_reg_offset(1, rn); - off_rm = vfp_reg_offset(1, rm); - } else { - rn = VFP_SREG_N(insn); - rm = VFP_SREG_M(insn); - off_rn = vfp_reg_offset(0, rn); - off_rm = vfp_reg_offset(0, rm); - } - - if (s->fp_excp_el) { - gen_exception_insn(s, s->pc_curr, EXCP_UDEF, - syn_simd_access_trap(1, 0xe, false), s->fp_excp_el); - return 0; - } - if (!s->vfp_enabled) { - return 1; - } - - opr_sz = (1 + q) * 8; - if (fn_gvec_ptr) { - TCGv_ptr ptr; - if (ptr_is_env) { - ptr = cpu_env; - } else { - ptr = get_fpstatus_ptr(1); - } - tcg_gen_gvec_3_ptr(vfp_reg_offset(1, rd), off_rn, off_rm, ptr, - opr_sz, opr_sz, data, fn_gvec_ptr); - if (!ptr_is_env) { - tcg_temp_free_ptr(ptr); - } - } else { - tcg_gen_gvec_3_ool(vfp_reg_offset(1, rd), off_rn, off_rm, - opr_sz, opr_sz, data, fn_gvec); - } - return 0; -} - /* Advanced SIMD two registers and a scalar extension. * 31 24 23 22 20 16 12 11 10 9 8 3 0 * +-----------------+----+---+----+----+----+---+----+---+----+---------+----+ @@ -XXX,XX +XXX,XX @@ static void disas_arm_insn(DisasContext *s, unsigned int insn) } } } - } else if ((insn & 0x0e000a00) == 0x0c000800 - && arm_dc_feature(s, ARM_FEATURE_V8)) { - if (disas_neon_insn_3same_ext(s, insn)) { - goto illegal_op; - } - return; } else if ((insn & 0x0f000a00) == 0x0e000800 && arm_dc_feature(s, ARM_FEATURE_V8)) { if (disas_neon_insn_2reg_scalar_ext(s, insn)) { @@ -XXX,XX +XXX,XX @@ static void disas_thumb2_insn(DisasContext *s, uint32_t insn) } break; } - if ((insn & 0xfe000a00) == 0xfc000800 + if ((insn & 0xff000a00) == 0xfe000800 && arm_dc_feature(s, ARM_FEATURE_V8)) { /* The Thumb2 and ARM encodings are identical. */ - if (disas_neon_insn_3same_ext(s, insn)) { - goto illegal_op; - } - } else if ((insn & 0xff000a00) == 0xfe000800 - && arm_dc_feature(s, ARM_FEATURE_V8)) { - /* The Thumb2 and ARM encodings are identical. */ if (disas_neon_insn_2reg_scalar_ext(s, insn)) { goto illegal_op; } -- 2.20.1 Convert VCMLA (scalar) in the 2reg-scalar-ext group to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-9-peter.maydell@linaro.org --- target/arm/neon-shared.decode | 5 +++++ target/arm/translate-neon.inc.c | 40 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 26 +-------------------- 3 files changed, 46 insertions(+), 25 deletions(-) diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-shared.decode +++ b/target/arm/neon-shared.decode @@ -XXX,XX +XXX,XX @@ VFML 1111 110 0 s:1 . 10 .... .... 1000 . 0 . 1 .... \ vm=%vm_sp vn=%vn_sp vd=%vd_dp q=0 VFML 1111 110 0 s:1 . 10 .... .... 1000 . 1 . 1 .... \ vm=%vm_dp vn=%vn_dp vd=%vd_dp q=1 + +VCMLA_scalar 1111 1110 0 . rot:2 .... .... 1000 . q:1 index:1 0 vm:4 \ + vn=%vn_dp vd=%vd_dp size=0 +VCMLA_scalar 1111 1110 1 . rot:2 .... .... 1000 . q:1 . 0 .... \ + vm=%vm_dp vn=%vn_dp vd=%vd_dp size=1 index=0 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_VFML(DisasContext *s, arg_VFML *a) gen_helper_gvec_fmlal_a32); return true; } + +static bool trans_VCMLA_scalar(DisasContext *s, arg_VCMLA_scalar *a) +{ + gen_helper_gvec_3_ptr *fn_gvec_ptr; + int opr_sz; + TCGv_ptr fpst; + + if (!dc_isar_feature(aa32_vcma, s)) { + return false; + } + if (a->size == 0 && !dc_isar_feature(aa32_fp16_arith, 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->vd | a->vn) & a->q) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + fn_gvec_ptr = (a->size ? gen_helper_gvec_fcmlas_idx + : gen_helper_gvec_fcmlah_idx); + opr_sz = (1 + a->q) * 8; + fpst = get_fpstatus_ptr(1); + tcg_gen_gvec_3_ptr(vfp_reg_offset(1, a->vd), + vfp_reg_offset(1, a->vn), + vfp_reg_offset(1, a->vm), + fpst, opr_sz, opr_sz, + (a->index << 2) | a->rot, fn_gvec_ptr); + tcg_temp_free_ptr(fpst); + return 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_insn_2reg_scalar_ext(DisasContext *s, uint32_t insn) bool is_long = false, q = extract32(insn, 6, 1); bool ptr_is_env = false; - if ((insn & 0xff000f10) == 0xfe000800) { - /* VCMLA (indexed) -- 1111 1110 S.RR .... .... 1000 ...0 .... */ - int rot = extract32(insn, 20, 2); - int size = extract32(insn, 23, 1); - int index; - - if (!dc_isar_feature(aa32_vcma, s)) { - return 1; - } - if (size == 0) { - if (!dc_isar_feature(aa32_fp16_arith, s)) { - return 1; - } - /* For fp16, rm is just Vm, and index is M. */ - rm = extract32(insn, 0, 4); - index = extract32(insn, 5, 1); - } else { - /* For fp32, rm is the usual M:Vm, and index is 0. */ - VFP_DREG_M(rm, insn); - index = 0; - } - data = (index << 2) | rot; - fn_gvec_ptr = (size ? gen_helper_gvec_fcmlas_idx - : gen_helper_gvec_fcmlah_idx); - } else if ((insn & 0xffb00f00) == 0xfe200d00) { + if ((insn & 0xffb00f00) == 0xfe200d00) { /* V[US]DOT -- 1111 1110 0.10 .... .... 1101 .Q.U .... */ int u = extract32(insn, 4, 1); -- 2.20.1 Convert the V[US]DOT (scalar) insns in the 2reg-scalar-ext group to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-10-peter.maydell@linaro.org --- target/arm/neon-shared.decode | 3 +++ target/arm/translate-neon.inc.c | 35 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 13 +----------- 3 files changed, 39 insertions(+), 12 deletions(-) diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-shared.decode +++ b/target/arm/neon-shared.decode @@ -XXX,XX +XXX,XX @@ VCMLA_scalar 1111 1110 0 . rot:2 .... .... 1000 . q:1 index:1 0 vm:4 \ vn=%vn_dp vd=%vd_dp size=0 VCMLA_scalar 1111 1110 1 . rot:2 .... .... 1000 . q:1 . 0 .... \ vm=%vm_dp vn=%vn_dp vd=%vd_dp size=1 index=0 + +VDOT_scalar 1111 1110 0 . 10 .... .... 1101 . q:1 index:1 u:1 rm:4 \ + vm=%vm_dp vn=%vn_dp vd=%vd_dp 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_VCMLA_scalar(DisasContext *s, arg_VCMLA_scalar *a) tcg_temp_free_ptr(fpst); return true; } + +static bool trans_VDOT_scalar(DisasContext *s, arg_VDOT_scalar *a) +{ + gen_helper_gvec_3 *fn_gvec; + int opr_sz; + TCGv_ptr fpst; + + if (!dc_isar_feature(aa32_dp, 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) & 0x10)) { + return false; + } + + if ((a->vd | a->vn) & a->q) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + fn_gvec = a->u ? gen_helper_gvec_udot_idx_b : gen_helper_gvec_sdot_idx_b; + opr_sz = (1 + a->q) * 8; + fpst = get_fpstatus_ptr(1); + tcg_gen_gvec_3_ool(vfp_reg_offset(1, a->vd), + vfp_reg_offset(1, a->vn), + vfp_reg_offset(1, a->rm), + opr_sz, opr_sz, a->index, fn_gvec); + tcg_temp_free_ptr(fpst); + return 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_insn_2reg_scalar_ext(DisasContext *s, uint32_t insn) bool is_long = false, q = extract32(insn, 6, 1); bool ptr_is_env = false; - if ((insn & 0xffb00f00) == 0xfe200d00) { - /* V[US]DOT -- 1111 1110 0.10 .... .... 1101 .Q.U .... */ - int u = extract32(insn, 4, 1); - - if (!dc_isar_feature(aa32_dp, s)) { - return 1; - } - fn_gvec = u ? gen_helper_gvec_udot_idx_b : gen_helper_gvec_sdot_idx_b; - /* rm is just Vm, and index is M. */ - data = extract32(insn, 5, 1); /* index */ - rm = extract32(insn, 0, 4); - } else if ((insn & 0xffa00f10) == 0xfe000810) { + if ((insn & 0xffa00f10) == 0xfe000810) { /* VFM[AS]L -- 1111 1110 0.0S .... .... 1000 .Q.1 .... */ int is_s = extract32(insn, 20, 1); int vm20 = extract32(insn, 0, 3); -- 2.20.1 Convert the VFM[AS]L (scalar) insns in the 2reg-scalar-ext group to decodetree. These are the last ones in the group so we can remove all the legacy decode for the group. Note that in disas_thumb2_insn() the parts of this encoding space where the decodetree decoder returns false will correctly be directed to illegal_op by the "(insn & (1 << 28))" check so they won't fall into disas_coproc_insn() by mistake. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-11-peter.maydell@linaro.org --- target/arm/neon-shared.decode | 7 +++ target/arm/translate-neon.inc.c | 32 ++++++++++ target/arm/translate.c | 107 +------------------------------- 3 files changed, 40 insertions(+), 106 deletions(-) diff --git a/target/arm/neon-shared.decode b/target/arm/neon-shared.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-shared.decode +++ b/target/arm/neon-shared.decode @@ -XXX,XX +XXX,XX @@ VCMLA_scalar 1111 1110 1 . rot:2 .... .... 1000 . q:1 . 0 .... \ VDOT_scalar 1111 1110 0 . 10 .... .... 1101 . q:1 index:1 u:1 rm:4 \ vm=%vm_dp vn=%vn_dp vd=%vd_dp + +%vfml_scalar_q0_rm 0:3 5:1 +%vfml_scalar_q1_index 5:1 3:1 +VFML_scalar 1111 1110 0 . 0 s:1 .... .... 1000 . 0 . 1 index:1 ... \ + rm=%vfml_scalar_q0_rm vn=%vn_sp vd=%vd_dp q=0 +VFML_scalar 1111 1110 0 . 0 s:1 .... .... 1000 . 1 . 1 . rm:3 \ + index=%vfml_scalar_q1_index vn=%vn_dp vd=%vd_dp q=1 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_VDOT_scalar(DisasContext *s, arg_VDOT_scalar *a) tcg_temp_free_ptr(fpst); return true; } + +static bool trans_VFML_scalar(DisasContext *s, arg_VFML_scalar *a) +{ + int opr_sz; + + if (!dc_isar_feature(aa32_fhm, s)) { + return false; + } + + /* UNDEF accesses to D16-D31 if they don't exist. */ + if (!dc_isar_feature(aa32_simd_r32, s) && + ((a->vd & 0x10) || (a->q && (a->vn & 0x10)))) { + return false; + } + + if (a->vd & a->q) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + opr_sz = (1 + a->q) * 8; + tcg_gen_gvec_3_ptr(vfp_reg_offset(1, a->vd), + vfp_reg_offset(a->q, a->vn), + vfp_reg_offset(a->q, a->rm), + cpu_env, opr_sz, opr_sz, + (a->index << 2) | a->s, /* is_2 == 0 */ + gen_helper_gvec_fmlal_idx_a32); + return 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_dsp_insn(DisasContext *s, uint32_t insn) } #define VFP_REG_SHR(x, n) (((n) > 0) ? (x) >> (n) : (x) << -(n)) -#define VFP_SREG(insn, bigbit, smallbit) \ - ((VFP_REG_SHR(insn, bigbit - 1) & 0x1e) | (((insn) >> (smallbit)) & 1)) #define VFP_DREG(reg, insn, bigbit, smallbit) do { \ if (dc_isar_feature(aa32_simd_r32, s)) { \ reg = (((insn) >> (bigbit)) & 0x0f) \ @@ -XXX,XX +XXX,XX @@ static int disas_dsp_insn(DisasContext *s, uint32_t insn) reg = ((insn) >> (bigbit)) & 0x0f; \ }} while (0) -#define VFP_SREG_D(insn) VFP_SREG(insn, 12, 22) #define VFP_DREG_D(reg, insn) VFP_DREG(reg, insn, 12, 22) -#define VFP_SREG_N(insn) VFP_SREG(insn, 16, 7) #define VFP_DREG_N(reg, insn) VFP_DREG(reg, insn, 16, 7) -#define VFP_SREG_M(insn) VFP_SREG(insn, 0, 5) #define VFP_DREG_M(reg, insn) VFP_DREG(reg, insn, 0, 5) static void gen_neon_dup_low16(TCGv_i32 var) @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) return 0; } -/* Advanced SIMD two registers and a scalar extension. - * 31 24 23 22 20 16 12 11 10 9 8 3 0 - * +-----------------+----+---+----+----+----+---+----+---+----+---------+----+ - * | 1 1 1 1 1 1 1 0 | o1 | D | o2 | Vn | Vd | 1 | o3 | 0 | o4 | N Q M U | Vm | - * +-----------------+----+---+----+----+----+---+----+---+----+---------+----+ - * - */ - -static int disas_neon_insn_2reg_scalar_ext(DisasContext *s, uint32_t insn) -{ - gen_helper_gvec_3 *fn_gvec = NULL; - gen_helper_gvec_3_ptr *fn_gvec_ptr = NULL; - int rd, rn, rm, opr_sz, data; - int off_rn, off_rm; - bool is_long = false, q = extract32(insn, 6, 1); - bool ptr_is_env = false; - - if ((insn & 0xffa00f10) == 0xfe000810) { - /* VFM[AS]L -- 1111 1110 0.0S .... .... 1000 .Q.1 .... */ - int is_s = extract32(insn, 20, 1); - int vm20 = extract32(insn, 0, 3); - int vm3 = extract32(insn, 3, 1); - int m = extract32(insn, 5, 1); - int index; - - if (!dc_isar_feature(aa32_fhm, s)) { - return 1; - } - if (q) { - rm = vm20; - index = m * 2 + vm3; - } else { - rm = vm20 * 2 + m; - index = vm3; - } - is_long = true; - data = (index << 2) | is_s; /* is_2 == 0 */ - fn_gvec_ptr = gen_helper_gvec_fmlal_idx_a32; - ptr_is_env = true; - } else { - return 1; - } - - VFP_DREG_D(rd, insn); - if (rd & q) { - return 1; - } - if (q || !is_long) { - VFP_DREG_N(rn, insn); - if (rn & q & !is_long) { - return 1; - } - off_rn = vfp_reg_offset(1, rn); - off_rm = vfp_reg_offset(1, rm); - } else { - rn = VFP_SREG_N(insn); - off_rn = vfp_reg_offset(0, rn); - off_rm = vfp_reg_offset(0, rm); - } - if (s->fp_excp_el) { - gen_exception_insn(s, s->pc_curr, EXCP_UDEF, - syn_simd_access_trap(1, 0xe, false), s->fp_excp_el); - return 0; - } - if (!s->vfp_enabled) { - return 1; - } - - opr_sz = (1 + q) * 8; - if (fn_gvec_ptr) { - TCGv_ptr ptr; - if (ptr_is_env) { - ptr = cpu_env; - } else { - ptr = get_fpstatus_ptr(1); - } - tcg_gen_gvec_3_ptr(vfp_reg_offset(1, rd), off_rn, off_rm, ptr, - opr_sz, opr_sz, data, fn_gvec_ptr); - if (!ptr_is_env) { - tcg_temp_free_ptr(ptr); - } - } else { - tcg_gen_gvec_3_ool(vfp_reg_offset(1, rd), off_rn, off_rm, - opr_sz, opr_sz, data, fn_gvec); - } - return 0; -} - static int disas_coproc_insn(DisasContext *s, uint32_t insn) { int cpnum, is64, crn, crm, opc1, opc2, isread, rt, rt2; @@ -XXX,XX +XXX,XX @@ static void disas_arm_insn(DisasContext *s, unsigned int insn) } } } - } else if ((insn & 0x0f000a00) == 0x0e000800 - && arm_dc_feature(s, ARM_FEATURE_V8)) { - if (disas_neon_insn_2reg_scalar_ext(s, insn)) { - goto illegal_op; - } - return; } goto illegal_op; } @@ -XXX,XX +XXX,XX @@ static void disas_thumb2_insn(DisasContext *s, uint32_t insn) } break; } - if ((insn & 0xff000a00) == 0xfe000800 - && arm_dc_feature(s, ARM_FEATURE_V8)) { - /* The Thumb2 and ARM encodings are identical. */ - if (disas_neon_insn_2reg_scalar_ext(s, insn)) { - goto illegal_op; - } - } else if (((insn >> 24) & 3) == 3) { + if (((insn >> 24) & 3) == 3) { /* Translate into the equivalent ARM encoding. */ insn = (insn & 0xe2ffffff) | ((insn & (1 << 28)) >> 4) | (1 << 28); if (disas_neon_data_insn(s, insn)) { -- 2.20.1 Convert the Neon "load/store multiple structures" insns to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-12-peter.maydell@linaro.org --- target/arm/neon-ls.decode | 7 ++ target/arm/translate-neon.inc.c | 124 ++++++++++++++++++++++++++++++++ target/arm/translate.c | 91 +---------------------- 3 files changed, 133 insertions(+), 89 deletions(-) diff --git a/target/arm/neon-ls.decode b/target/arm/neon-ls.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-ls.decode +++ b/target/arm/neon-ls.decode @@ -XXX,XX +XXX,XX @@ # 0b1111_1001_xxx0_xxxx_xxxx_xxxx_xxxx_xxxx # This file works on the A32 encoding only; calling code for T32 has to # transform the insn into the A32 version first. + +%vd_dp 22:1 12:4 + +# Neon load/store multiple structures + +VLDST_multiple 1111 0100 0 . l:1 0 rn:4 .... itype:4 size:2 align:2 rm:4 \ + vd=%vd_dp 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_VFML_scalar(DisasContext *s, arg_VFML_scalar *a) gen_helper_gvec_fmlal_idx_a32); return true; } + +static struct { + int nregs; + int interleave; + int spacing; +} const neon_ls_element_type[11] = { + {1, 4, 1}, + {1, 4, 2}, + {4, 1, 1}, + {2, 2, 2}, + {1, 3, 1}, + {1, 3, 2}, + {3, 1, 1}, + {1, 1, 1}, + {1, 2, 1}, + {1, 2, 2}, + {2, 1, 1} +}; + +static void gen_neon_ldst_base_update(DisasContext *s, int rm, int rn, + int stride) +{ + if (rm != 15) { + TCGv_i32 base; + + base = load_reg(s, rn); + if (rm == 13) { + tcg_gen_addi_i32(base, base, stride); + } else { + TCGv_i32 index; + index = load_reg(s, rm); + tcg_gen_add_i32(base, base, index); + tcg_temp_free_i32(index); + } + store_reg(s, rn, base); + } +} + +static bool trans_VLDST_multiple(DisasContext *s, arg_VLDST_multiple *a) +{ + /* Neon load/store multiple structures */ + int nregs, interleave, spacing, reg, n; + MemOp endian = s->be_data; + int mmu_idx = get_mem_index(s); + int size = a->size; + TCGv_i64 tmp64; + TCGv_i32 addr, tmp; + + 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->itype > 10) { + return false; + } + /* Catch UNDEF cases for bad values of align field */ + switch (a->itype & 0xc) { + case 4: + if (a->align >= 2) { + return false; + } + break; + case 8: + if (a->align == 3) { + return false; + } + break; + default: + break; + } + nregs = neon_ls_element_type[a->itype].nregs; + interleave = neon_ls_element_type[a->itype].interleave; + spacing = neon_ls_element_type[a->itype].spacing; + if (size == 3 && (interleave | spacing) != 1) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + /* For our purposes, bytes are always little-endian. */ + if (size == 0) { + endian = MO_LE; + } + /* + * Consecutive little-endian elements from a single register + * can be promoted to a larger little-endian operation. + */ + if (interleave == 1 && endian == MO_LE) { + size = 3; + } + tmp64 = tcg_temp_new_i64(); + addr = tcg_temp_new_i32(); + tmp = tcg_const_i32(1 << size); + load_reg_var(s, addr, a->rn); + for (reg = 0; reg < nregs; reg++) { + for (n = 0; n < 8 >> size; n++) { + int xs; + for (xs = 0; xs < interleave; xs++) { + int tt = a->vd + reg + spacing * xs; + + if (a->l) { + gen_aa32_ld_i64(s, tmp64, addr, mmu_idx, endian | size); + neon_store_element64(tt, n, size, tmp64); + } else { + neon_load_element64(tmp64, tt, n, size); + gen_aa32_st_i64(s, tmp64, addr, mmu_idx, endian | size); + } + tcg_gen_add_i32(addr, addr, tmp); + } + } + } + tcg_temp_free_i32(addr); + tcg_temp_free_i32(tmp); + tcg_temp_free_i64(tmp64); + + gen_neon_ldst_base_update(s, a->rm, a->rn, nregs * interleave * 8); + return 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 void gen_neon_trn_u16(TCGv_i32 t0, TCGv_i32 t1) } -static struct { - int nregs; - int interleave; - int spacing; -} const neon_ls_element_type[11] = { - {1, 4, 1}, - {1, 4, 2}, - {4, 1, 1}, - {2, 2, 2}, - {1, 3, 1}, - {1, 3, 2}, - {3, 1, 1}, - {1, 1, 1}, - {1, 2, 1}, - {1, 2, 2}, - {2, 1, 1} -}; - /* Translate a NEON load/store element instruction. Return nonzero if the instruction is invalid. */ static int disas_neon_ls_insn(DisasContext *s, uint32_t insn) { int rd, rn, rm; - int op; int nregs; - int interleave; - int spacing; int stride; int size; int reg; int load; - int n; int vec_size; - int mmu_idx; - MemOp endian; TCGv_i32 addr; TCGv_i32 tmp; - TCGv_i32 tmp2; - TCGv_i64 tmp64; if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { return 1; @@ -XXX,XX +XXX,XX @@ static int disas_neon_ls_insn(DisasContext *s, uint32_t insn) rn = (insn >> 16) & 0xf; rm = insn & 0xf; load = (insn & (1 << 21)) != 0; - endian = s->be_data; - mmu_idx = get_mem_index(s); if ((insn & (1 << 23)) == 0) { - /* Load store all elements. */ - op = (insn >> 8) & 0xf; - size = (insn >> 6) & 3; - if (op > 10) - return 1; - /* Catch UNDEF cases for bad values of align field */ - switch (op & 0xc) { - case 4: - if (((insn >> 5) & 1) == 1) { - return 1; - } - break; - case 8: - if (((insn >> 4) & 3) == 3) { - return 1; - } - break; - default: - break; - } - nregs = neon_ls_element_type[op].nregs; - interleave = neon_ls_element_type[op].interleave; - spacing = neon_ls_element_type[op].spacing; - if (size == 3 && (interleave | spacing) != 1) { - return 1; - } - /* For our purposes, bytes are always little-endian. */ - if (size == 0) { - endian = MO_LE; - } - /* Consecutive little-endian elements from a single register - * can be promoted to a larger little-endian operation. - */ - if (interleave == 1 && endian == MO_LE) { - size = 3; - } - tmp64 = tcg_temp_new_i64(); - addr = tcg_temp_new_i32(); - tmp2 = tcg_const_i32(1 << size); - load_reg_var(s, addr, rn); - for (reg = 0; reg < nregs; reg++) { - for (n = 0; n < 8 >> size; n++) { - int xs; - for (xs = 0; xs < interleave; xs++) { - int tt = rd + reg + spacing * xs; - - if (load) { - gen_aa32_ld_i64(s, tmp64, addr, mmu_idx, endian | size); - neon_store_element64(tt, n, size, tmp64); - } else { - neon_load_element64(tmp64, tt, n, size); - gen_aa32_st_i64(s, tmp64, addr, mmu_idx, endian | size); - } - tcg_gen_add_i32(addr, addr, tmp2); - } - } - } - tcg_temp_free_i32(addr); - tcg_temp_free_i32(tmp2); - tcg_temp_free_i64(tmp64); - stride = nregs * interleave * 8; + /* Load store all elements -- handled already by decodetree */ + return 1; } else { size = (insn >> 10) & 3; if (size == 3) { -- 2.20.1 Convert the Neon "load single structure to all lanes" insns to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-13-peter.maydell@linaro.org --- target/arm/neon-ls.decode | 5 +++ target/arm/translate-neon.inc.c | 73 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 55 +------------------------ 3 files changed, 80 insertions(+), 53 deletions(-) diff --git a/target/arm/neon-ls.decode b/target/arm/neon-ls.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-ls.decode +++ b/target/arm/neon-ls.decode @@ -XXX,XX +XXX,XX @@ VLDST_multiple 1111 0100 0 . l:1 0 rn:4 .... itype:4 size:2 align:2 rm:4 \ vd=%vd_dp + +# Neon load single element to all lanes + +VLD_all_lanes 1111 0100 1 . 1 0 rn:4 .... 11 n:2 size:2 t:1 a:1 rm:4 \ + vd=%vd_dp 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_VLDST_multiple(DisasContext *s, arg_VLDST_multiple *a) gen_neon_ldst_base_update(s, a->rm, a->rn, nregs * interleave * 8); return true; } + +static bool trans_VLD_all_lanes(DisasContext *s, arg_VLD_all_lanes *a) +{ + /* Neon load single structure to all lanes */ + int reg, stride, vec_size; + int vd = a->vd; + int size = a->size; + int nregs = a->n + 1; + TCGv_i32 addr, tmp; + + 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 (size == 3) { + if (nregs != 4 || a->a == 0) { + return false; + } + /* For VLD4 size == 3 a == 1 means 32 bits at 16 byte alignment */ + size = 2; + } + if (nregs == 1 && a->a == 1 && size == 0) { + return false; + } + if (nregs == 3 && a->a == 1) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + /* + * VLD1 to all lanes: T bit indicates how many Dregs to write. + * VLD2/3/4 to all lanes: T bit indicates register stride. + */ + stride = a->t ? 2 : 1; + vec_size = nregs == 1 ? stride * 8 : 8; + + tmp = tcg_temp_new_i32(); + addr = tcg_temp_new_i32(); + load_reg_var(s, addr, a->rn); + for (reg = 0; reg < nregs; reg++) { + gen_aa32_ld_i32(s, tmp, addr, get_mem_index(s), + s->be_data | size); + if ((vd & 1) && vec_size == 16) { + /* + * We cannot write 16 bytes at once because the + * destination is unaligned. + */ + tcg_gen_gvec_dup_i32(size, neon_reg_offset(vd, 0), + 8, 8, tmp); + tcg_gen_gvec_mov(0, neon_reg_offset(vd + 1, 0), + neon_reg_offset(vd, 0), 8, 8); + } else { + tcg_gen_gvec_dup_i32(size, neon_reg_offset(vd, 0), + vec_size, vec_size, tmp); + } + tcg_gen_addi_i32(addr, addr, 1 << size); + vd += stride; + } + tcg_temp_free_i32(tmp); + tcg_temp_free_i32(addr); + + gen_neon_ldst_base_update(s, a->rm, a->rn, (1 << size) * nregs); + + return 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_ls_insn(DisasContext *s, uint32_t insn) int size; int reg; int load; - int vec_size; TCGv_i32 addr; TCGv_i32 tmp; @@ -XXX,XX +XXX,XX @@ static int disas_neon_ls_insn(DisasContext *s, uint32_t insn) } else { size = (insn >> 10) & 3; if (size == 3) { - /* Load single element to all lanes. */ - int a = (insn >> 4) & 1; - if (!load) { - return 1; - } - size = (insn >> 6) & 3; - nregs = ((insn >> 8) & 3) + 1; - - if (size == 3) { - if (nregs != 4 || a == 0) { - return 1; - } - /* For VLD4 size==3 a == 1 means 32 bits at 16 byte alignment */ - size = 2; - } - if (nregs == 1 && a == 1 && size == 0) { - return 1; - } - if (nregs == 3 && a == 1) { - return 1; - } - addr = tcg_temp_new_i32(); - load_reg_var(s, addr, rn); - - /* VLD1 to all lanes: bit 5 indicates how many Dregs to write. - * VLD2/3/4 to all lanes: bit 5 indicates register stride. - */ - stride = (insn & (1 << 5)) ? 2 : 1; - vec_size = nregs == 1 ? stride * 8 : 8; - - tmp = tcg_temp_new_i32(); - for (reg = 0; reg < nregs; reg++) { - gen_aa32_ld_i32(s, tmp, addr, get_mem_index(s), - s->be_data | size); - if ((rd & 1) && vec_size == 16) { - /* We cannot write 16 bytes at once because the - * destination is unaligned. - */ - tcg_gen_gvec_dup_i32(size, neon_reg_offset(rd, 0), - 8, 8, tmp); - tcg_gen_gvec_mov(0, neon_reg_offset(rd + 1, 0), - neon_reg_offset(rd, 0), 8, 8); - } else { - tcg_gen_gvec_dup_i32(size, neon_reg_offset(rd, 0), - vec_size, vec_size, tmp); - } - tcg_gen_addi_i32(addr, addr, 1 << size); - rd += stride; - } - tcg_temp_free_i32(tmp); - tcg_temp_free_i32(addr); - stride = (1 << size) * nregs; + /* Load single element to all lanes -- handled by decodetree */ + return 1; } else { /* Single element. */ int idx = (insn >> 4) & 0xf; -- 2.20.1 Convert the Neon "load/store single structure to one lane" insns to decodetree. As this is the last set of insns in the neon load/store group, we can remove the whole disas_neon_ls_insn() function. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-14-peter.maydell@linaro.org --- target/arm/neon-ls.decode | 11 +++ target/arm/translate-neon.inc.c | 89 +++++++++++++++++++ target/arm/translate.c | 147 -------------------------------- 3 files changed, 100 insertions(+), 147 deletions(-) diff --git a/target/arm/neon-ls.decode b/target/arm/neon-ls.decode index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon-ls.decode +++ b/target/arm/neon-ls.decode @@ -XXX,XX +XXX,XX @@ VLDST_multiple 1111 0100 0 . l:1 0 rn:4 .... itype:4 size:2 align:2 rm:4 \ VLD_all_lanes 1111 0100 1 . 1 0 rn:4 .... 11 n:2 size:2 t:1 a:1 rm:4 \ vd=%vd_dp + +# Neon load/store single structure to one lane +%imm1_5_p1 5:1 !function=plus1 +%imm1_6_p1 6:1 !function=plus1 + +VLDST_single 1111 0100 1 . l:1 0 rn:4 .... 00 n:2 reg_idx:3 align:1 rm:4 \ + vd=%vd_dp size=0 stride=1 +VLDST_single 1111 0100 1 . l:1 0 rn:4 .... 01 n:2 reg_idx:2 align:2 rm:4 \ + vd=%vd_dp size=1 stride=%imm1_5_p1 +VLDST_single 1111 0100 1 . l:1 0 rn:4 .... 10 n:2 reg_idx:1 align:3 rm:4 \ + vd=%vd_dp size=2 stride=%imm1_6_p1 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 @@ * It might be possible to convert it to a standalone .c file eventually. */ +static inline int plus1(DisasContext *s, int x) +{ + return x + 1; +} + /* Include the generated Neon decoder */ #include "decode-neon-dp.inc.c" #include "decode-neon-ls.inc.c" @@ -XXX,XX +XXX,XX @@ static bool trans_VLD_all_lanes(DisasContext *s, arg_VLD_all_lanes *a) return true; } + +static bool trans_VLDST_single(DisasContext *s, arg_VLDST_single *a) +{ + /* Neon load/store single structure to one lane */ + int reg; + int nregs = a->n + 1; + int vd = a->vd; + TCGv_i32 addr, tmp; + + 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; + } + + /* Catch the UNDEF cases. This is unavoidably a bit messy. */ + switch (nregs) { + case 1: + if (((a->align & (1 << a->size)) != 0) || + (a->size == 2 && ((a->align & 3) == 1 || (a->align & 3) == 2))) { + return false; + } + break; + case 3: + if ((a->align & 1) != 0) { + return false; + } + /* fall through */ + case 2: + if (a->size == 2 && (a->align & 2) != 0) { + return false; + } + break; + case 4: + if ((a->size == 2) && ((a->align & 3) == 3)) { + return false; + } + break; + default: + abort(); + } + if ((vd + a->stride * (nregs - 1)) > 31) { + /* + * Attempts to write off the end of the register file are + * UNPREDICTABLE; we choose to UNDEF because otherwise we would + * access off the end of the array that holds the register data. + */ + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + tmp = tcg_temp_new_i32(); + addr = tcg_temp_new_i32(); + load_reg_var(s, addr, a->rn); + /* + * TODO: if we implemented alignment exceptions, we should check + * addr against the alignment encoded in a->align here. + */ + for (reg = 0; reg < nregs; reg++) { + if (a->l) { + gen_aa32_ld_i32(s, tmp, addr, get_mem_index(s), + s->be_data | a->size); + neon_store_element(vd, a->reg_idx, a->size, tmp); + } else { /* Store */ + neon_load_element(tmp, vd, a->reg_idx, a->size); + gen_aa32_st_i32(s, tmp, addr, get_mem_index(s), + s->be_data | a->size); + } + vd += a->stride; + tcg_gen_addi_i32(addr, addr, 1 << a->size); + } + tcg_temp_free_i32(addr); + tcg_temp_free_i32(tmp); + + gen_neon_ldst_base_update(s, a->rm, a->rn, (1 << a->size) * nregs); + + return 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 void gen_neon_trn_u16(TCGv_i32 t0, TCGv_i32 t1) tcg_temp_free_i32(rd); } - -/* Translate a NEON load/store element instruction. Return nonzero if the - instruction is invalid. */ -static int disas_neon_ls_insn(DisasContext *s, uint32_t insn) -{ - int rd, rn, rm; - int nregs; - int stride; - int size; - int reg; - int load; - TCGv_i32 addr; - TCGv_i32 tmp; - - if (!arm_dc_feature(s, ARM_FEATURE_NEON)) { - return 1; - } - - /* FIXME: this access check should not take precedence over UNDEF - * for invalid encodings; we will generate incorrect syndrome information - * for attempts to execute invalid vfp/neon encodings with FP disabled. - */ - if (s->fp_excp_el) { - gen_exception_insn(s, s->pc_curr, EXCP_UDEF, - syn_simd_access_trap(1, 0xe, false), s->fp_excp_el); - return 0; - } - - if (!s->vfp_enabled) - return 1; - VFP_DREG_D(rd, insn); - rn = (insn >> 16) & 0xf; - rm = insn & 0xf; - load = (insn & (1 << 21)) != 0; - if ((insn & (1 << 23)) == 0) { - /* Load store all elements -- handled already by decodetree */ - return 1; - } else { - size = (insn >> 10) & 3; - if (size == 3) { - /* Load single element to all lanes -- handled by decodetree */ - return 1; - } else { - /* Single element. */ - int idx = (insn >> 4) & 0xf; - int reg_idx; - switch (size) { - case 0: - reg_idx = (insn >> 5) & 7; - stride = 1; - break; - case 1: - reg_idx = (insn >> 6) & 3; - stride = (insn & (1 << 5)) ? 2 : 1; - break; - case 2: - reg_idx = (insn >> 7) & 1; - stride = (insn & (1 << 6)) ? 2 : 1; - break; - default: - abort(); - } - nregs = ((insn >> 8) & 3) + 1; - /* Catch the UNDEF cases. This is unavoidably a bit messy. */ - switch (nregs) { - case 1: - if (((idx & (1 << size)) != 0) || - (size == 2 && ((idx & 3) == 1 || (idx & 3) == 2))) { - return 1; - } - break; - case 3: - if ((idx & 1) != 0) { - return 1; - } - /* fall through */ - case 2: - if (size == 2 && (idx & 2) != 0) { - return 1; - } - break; - case 4: - if ((size == 2) && ((idx & 3) == 3)) { - return 1; - } - break; - default: - abort(); - } - if ((rd + stride * (nregs - 1)) > 31) { - /* Attempts to write off the end of the register file - * are UNPREDICTABLE; we choose to UNDEF because otherwise - * the neon_load_reg() would write off the end of the array. - */ - return 1; - } - tmp = tcg_temp_new_i32(); - addr = tcg_temp_new_i32(); - load_reg_var(s, addr, rn); - for (reg = 0; reg < nregs; reg++) { - if (load) { - gen_aa32_ld_i32(s, tmp, addr, get_mem_index(s), - s->be_data | size); - neon_store_element(rd, reg_idx, size, tmp); - } else { /* Store */ - neon_load_element(tmp, rd, reg_idx, size); - gen_aa32_st_i32(s, tmp, addr, get_mem_index(s), - s->be_data | size); - } - rd += stride; - tcg_gen_addi_i32(addr, addr, 1 << size); - } - tcg_temp_free_i32(addr); - tcg_temp_free_i32(tmp); - stride = nregs * (1 << size); - } - } - if (rm != 15) { - TCGv_i32 base; - - base = load_reg(s, rn); - if (rm == 13) { - tcg_gen_addi_i32(base, base, stride); - } else { - TCGv_i32 index; - index = load_reg(s, rm); - tcg_gen_add_i32(base, base, index); - tcg_temp_free_i32(index); - } - store_reg(s, rn, base); - } - return 0; -} - static inline void gen_neon_narrow(int size, TCGv_i32 dest, TCGv_i64 src) { switch (size) { @@ -XXX,XX +XXX,XX @@ static void disas_arm_insn(DisasContext *s, unsigned int insn) } return; } - if ((insn & 0x0f100000) == 0x04000000) { - /* NEON load/store. */ - if (disas_neon_ls_insn(s, insn)) { - goto illegal_op; - } - return; - } if ((insn & 0x0e000f00) == 0x0c000100) { if (arm_dc_feature(s, ARM_FEATURE_IWMMXT)) { /* iWMMXt register transfer. */ @@ -XXX,XX +XXX,XX @@ static void disas_thumb2_insn(DisasContext *s, uint32_t insn) } break; case 12: - if ((insn & 0x01100000) == 0x01000000) { - if (disas_neon_ls_insn(s, insn)) { - goto illegal_op; - } - break; - } goto illegal_op; default: illegal_op: -- 2.20.1 Convert the Neon 3-reg-same VADD and VSUB insns to decodetree. Note that we don't need the neon_3r_sizes[op] check here because all size values are OK for VADD and VSUB; we'll add this when we convert the first insn that has size restrictions. For this we need one of the GVecGen*Fn typedefs currently in translate-a64.h; move them all to translate.h as a block so they are visible to the 32-bit decoder. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-15-peter.maydell@linaro.org --- target/arm/translate-a64.h | 9 -------- target/arm/translate.h | 9 ++++++++ target/arm/neon-dp.decode | 17 +++++++++++++++ target/arm/translate-neon.inc.c | 38 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 14 ++++-------- 5 files changed, 68 insertions(+), 19 deletions(-) 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); -/* Note that the gvec expanders operate on offsets + sizes. */ -typedef void GVecGen2Fn(unsigned, uint32_t, uint32_t, uint32_t, uint32_t); -typedef void GVecGen2iFn(unsigned, uint32_t, uint32_t, int64_t, - uint32_t, uint32_t); -typedef void GVecGen3Fn(unsigned, uint32_t, uint32_t, - uint32_t, uint32_t, uint32_t); -typedef void GVecGen4Fn(unsigned, uint32_t, uint32_t, uint32_t, - uint32_t, uint32_t, uint32_t); - #endif /* TARGET_ARM_TRANSLATE_A64_H */ diff --git a/target/arm/translate.h b/target/arm/translate.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.h +++ b/target/arm/translate.h @@ -XXX,XX +XXX,XX @@ void gen_sshl_i64(TCGv_i64 d, TCGv_i64 a, TCGv_i64 b); #define dc_isar_feature(name, ctx) \ ({ DisasContext *ctx_ = (ctx); isar_feature_##name(ctx_->isar); }) +/* Note that the gvec expanders operate on offsets + sizes. */ +typedef void GVecGen2Fn(unsigned, uint32_t, uint32_t, uint32_t, uint32_t); +typedef void GVecGen2iFn(unsigned, uint32_t, uint32_t, int64_t, + uint32_t, uint32_t); +typedef void GVecGen3Fn(unsigned, uint32_t, uint32_t, + uint32_t, uint32_t, uint32_t); +typedef void GVecGen4Fn(unsigned, uint32_t, uint32_t, uint32_t, + uint32_t, uint32_t, uint32_t); + #endif /* TARGET_ARM_TRANSLATE_H */ 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 @@ # # This file is processed by scripts/decodetree.py # +# VFP/Neon register fields; same as vfp.decode +%vm_dp 5:1 0:4 +%vn_dp 7:1 16:4 +%vd_dp 22:1 12:4 # Encodings for Neon data processing instructions where the T32 encoding # is a simple transformation of the A32 encoding. @@ -XXX,XX +XXX,XX @@ # 0b111p_1111_qqqq_qqqq_qqqq_qqqq_qqqq_qqqq # This file works on the A32 encoding only; calling code for T32 has to # transform the insn into the A32 version first. + +###################################################################### +# 3-reg-same grouping: +# 1111 001 U 0 D sz:2 Vn:4 Vd:4 opc:4 N Q M op Vm:4 +###################################################################### + +&3same vm vn vd q size + +@3same .... ... . . . size:2 .... .... .... . q:1 . . .... \ + &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp + +VADD_3s 1111 001 0 0 . .. .... .... 1000 . . . 0 .... @3same +VSUB_3s 1111 001 1 0 . .. .... .... 1000 . . . 0 .... @3same 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_VLDST_single(DisasContext *s, arg_VLDST_single *a) return true; } + +static bool do_3same(DisasContext *s, arg_3same *a, GVecGen3Fn fn) +{ + int vec_size = a->q ? 16 : 8; + int rd_ofs = neon_reg_offset(a->vd, 0); + int rn_ofs = neon_reg_offset(a->vn, 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->vn | a->vm) & 0x10)) { + return false; + } + + if ((a->vn | a->vm | a->vd) & a->q) { + return false; + } + + if (!vfp_access_check(s)) { + return true; + } + + fn(a->size, rd_ofs, rn_ofs, rm_ofs, vec_size, vec_size); + return true; +} + +#define DO_3SAME(INSN, FUNC) \ + static bool trans_##INSN##_3s(DisasContext *s, arg_3same *a) \ + { \ + return do_3same(s, a, FUNC); \ + } + +DO_3SAME(VADD, tcg_gen_gvec_add) +DO_3SAME(VSUB, tcg_gen_gvec_sub) 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) } return 0; - case NEON_3R_VADD_VSUB: - if (u) { - tcg_gen_gvec_sub(size, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - } else { - tcg_gen_gvec_add(size, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - } - return 0; - case NEON_3R_VQADD: tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc), rn_ofs, rm_ofs, vec_size, vec_size, @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) tcg_gen_gvec_3(rd_ofs, rm_ofs, rn_ofs, vec_size, vec_size, u ? &ushl_op[size] : &sshl_op[size]); return 0; + + case NEON_3R_VADD_VSUB: + /* Already handled by decodetree */ + return 1; } if (size == 3) { -- 2.20.1 Convert the Neon logic ops in the 3-reg-same grouping to decodetree. Note that for the logic ops the 'size' field forms part of their decode and the actual operations are always bitwise. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-16-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 12 +++++++++++ target/arm/translate-neon.inc.c | 19 +++++++++++++++++ target/arm/translate.c | 38 +-------------------------------- 3 files changed, 32 insertions(+), 37 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 @@ @3same .... ... . . . size:2 .... .... .... . q:1 . . .... \ &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp +@3same_logic .... ... . . . .. .... .... .... . q:1 .. .... \ + &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp size=0 + +VAND_3s 1111 001 0 0 . 00 .... .... 0001 ... 1 .... @3same_logic +VBIC_3s 1111 001 0 0 . 01 .... .... 0001 ... 1 .... @3same_logic +VORR_3s 1111 001 0 0 . 10 .... .... 0001 ... 1 .... @3same_logic +VORN_3s 1111 001 0 0 . 11 .... .... 0001 ... 1 .... @3same_logic +VEOR_3s 1111 001 1 0 . 00 .... .... 0001 ... 1 .... @3same_logic +VBSL_3s 1111 001 1 0 . 01 .... .... 0001 ... 1 .... @3same_logic +VBIT_3s 1111 001 1 0 . 10 .... .... 0001 ... 1 .... @3same_logic +VBIF_3s 1111 001 1 0 . 11 .... .... 0001 ... 1 .... @3same_logic + VADD_3s 1111 001 0 0 . .. .... .... 1000 . . . 0 .... @3same VSUB_3s 1111 001 1 0 . .. .... .... 1000 . . . 0 .... @3same 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(DisasContext *s, arg_3same *a, GVecGen3Fn fn) DO_3SAME(VADD, tcg_gen_gvec_add) DO_3SAME(VSUB, tcg_gen_gvec_sub) +DO_3SAME(VAND, tcg_gen_gvec_and) +DO_3SAME(VBIC, tcg_gen_gvec_andc) +DO_3SAME(VORR, tcg_gen_gvec_or) +DO_3SAME(VORN, tcg_gen_gvec_orc) +DO_3SAME(VEOR, tcg_gen_gvec_xor) + +/* These insns are all gvec_bitsel but with the inputs in various orders. */ +#define DO_3SAME_BITSEL(INSN, O1, O2, O3) \ + static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ + uint32_t rn_ofs, uint32_t rm_ofs, \ + uint32_t oprsz, uint32_t maxsz) \ + { \ + tcg_gen_gvec_bitsel(vece, rd_ofs, O1, O2, O3, oprsz, maxsz); \ + } \ + DO_3SAME(INSN, gen_##INSN##_3s) + +DO_3SAME_BITSEL(VBSL, rd_ofs, rn_ofs, rm_ofs) +DO_3SAME_BITSEL(VBIT, rm_ofs, rn_ofs, rd_ofs) +DO_3SAME_BITSEL(VBIF, rm_ofs, rd_ofs, rn_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) } return 1; - case NEON_3R_LOGIC: /* Logic ops. */ - switch ((u << 2) | size) { - case 0: /* VAND */ - tcg_gen_gvec_and(0, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - break; - case 1: /* VBIC */ - tcg_gen_gvec_andc(0, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - break; - case 2: /* VORR */ - tcg_gen_gvec_or(0, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - break; - case 3: /* VORN */ - tcg_gen_gvec_orc(0, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - break; - case 4: /* VEOR */ - tcg_gen_gvec_xor(0, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - break; - case 5: /* VBSL */ - tcg_gen_gvec_bitsel(MO_8, rd_ofs, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - break; - case 6: /* VBIT */ - tcg_gen_gvec_bitsel(MO_8, rd_ofs, rm_ofs, rn_ofs, rd_ofs, - vec_size, vec_size); - break; - case 7: /* VBIF */ - tcg_gen_gvec_bitsel(MO_8, rd_ofs, rm_ofs, rd_ofs, rn_ofs, - vec_size, vec_size); - break; - } - return 0; - case NEON_3R_VQADD: tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc), rn_ofs, rm_ofs, vec_size, vec_size, @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) return 0; case NEON_3R_VADD_VSUB: + case NEON_3R_LOGIC: /* Already handled by decodetree */ return 1; } -- 2.20.1 Convert the Neon 3-reg-same VMAX and VMIN insns to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-17-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 5 +++++ target/arm/translate-neon.inc.c | 14 ++++++++++++++ target/arm/translate.c | 21 ++------------------- 3 files changed, 21 insertions(+), 19 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 @@ VBSL_3s 1111 001 1 0 . 01 .... .... 0001 ... 1 .... @3same_logic VBIT_3s 1111 001 1 0 . 10 .... .... 0001 ... 1 .... @3same_logic VBIF_3s 1111 001 1 0 . 11 .... .... 0001 ... 1 .... @3same_logic +VMAX_S_3s 1111 001 0 0 . .. .... .... 0110 . . . 0 .... @3same +VMAX_U_3s 1111 001 1 0 . .. .... .... 0110 . . . 0 .... @3same +VMIN_S_3s 1111 001 0 0 . .. .... .... 0110 . . . 1 .... @3same +VMIN_U_3s 1111 001 1 0 . .. .... .... 0110 . . . 1 .... @3same + VADD_3s 1111 001 0 0 . .. .... .... 1000 . . . 0 .... @3same VSUB_3s 1111 001 1 0 . .. .... .... 1000 . . . 0 .... @3same 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(VEOR, tcg_gen_gvec_xor) DO_3SAME_BITSEL(VBSL, rd_ofs, rn_ofs, rm_ofs) DO_3SAME_BITSEL(VBIT, rm_ofs, rn_ofs, rd_ofs) DO_3SAME_BITSEL(VBIF, rm_ofs, rd_ofs, rn_ofs) + +#define DO_3SAME_NO_SZ_3(INSN, FUNC) \ + static bool trans_##INSN##_3s(DisasContext *s, arg_3same *a) \ + { \ + if (a->size == 3) { \ + return false; \ + } \ + return do_3same(s, a, FUNC); \ + } + +DO_3SAME_NO_SZ_3(VMAX_S, tcg_gen_gvec_smax) +DO_3SAME_NO_SZ_3(VMAX_U, tcg_gen_gvec_umax) +DO_3SAME_NO_SZ_3(VMIN_S, tcg_gen_gvec_smin) +DO_3SAME_NO_SZ_3(VMIN_U, tcg_gen_gvec_umin) 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) rd_ofs, rn_ofs, rm_ofs, vec_size, vec_size); return 0; - case NEON_3R_VMAX: - if (u) { - tcg_gen_gvec_umax(size, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - } else { - tcg_gen_gvec_smax(size, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - } - return 0; - case NEON_3R_VMIN: - if (u) { - tcg_gen_gvec_umin(size, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - } else { - tcg_gen_gvec_smin(size, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - } - return 0; - case NEON_3R_VSHL: /* Note the operation is vshl vd,vm,vn */ tcg_gen_gvec_3(rd_ofs, rm_ofs, rn_ofs, vec_size, vec_size, @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VADD_VSUB: case NEON_3R_LOGIC: + case NEON_3R_VMAX: + case NEON_3R_VMIN: /* Already handled by decodetree */ return 1; } -- 2.20.1 Convert the Neon comparison ops in the 3-reg-same grouping to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-18-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 8 ++++++++ target/arm/translate-neon.inc.c | 22 ++++++++++++++++++++++ target/arm/translate.c | 23 +++-------------------- 3 files changed, 33 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 @@ VBSL_3s 1111 001 1 0 . 01 .... .... 0001 ... 1 .... @3same_logic VBIT_3s 1111 001 1 0 . 10 .... .... 0001 ... 1 .... @3same_logic VBIF_3s 1111 001 1 0 . 11 .... .... 0001 ... 1 .... @3same_logic +VCGT_S_3s 1111 001 0 0 . .. .... .... 0011 . . . 0 .... @3same +VCGT_U_3s 1111 001 1 0 . .. .... .... 0011 . . . 0 .... @3same +VCGE_S_3s 1111 001 0 0 . .. .... .... 0011 . . . 1 .... @3same +VCGE_U_3s 1111 001 1 0 . .. .... .... 0011 . . . 1 .... @3same + VMAX_S_3s 1111 001 0 0 . .. .... .... 0110 . . . 0 .... @3same VMAX_U_3s 1111 001 1 0 . .. .... .... 0110 . . . 0 .... @3same VMIN_S_3s 1111 001 0 0 . .. .... .... 0110 . . . 1 .... @3same @@ -XXX,XX +XXX,XX @@ VMIN_U_3s 1111 001 1 0 . .. .... .... 0110 . . . 1 .... @3same VADD_3s 1111 001 0 0 . .. .... .... 1000 . . . 0 .... @3same VSUB_3s 1111 001 1 0 . .. .... .... 1000 . . . 0 .... @3same + +VTST_3s 1111 001 0 0 . .. .... .... 1000 . . . 1 .... @3same +VCEQ_3s 1111 001 1 0 . .. .... .... 1000 . . . 1 .... @3same 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_NO_SZ_3(VMAX_S, tcg_gen_gvec_smax) DO_3SAME_NO_SZ_3(VMAX_U, tcg_gen_gvec_umax) DO_3SAME_NO_SZ_3(VMIN_S, tcg_gen_gvec_smin) DO_3SAME_NO_SZ_3(VMIN_U, tcg_gen_gvec_umin) + +#define DO_3SAME_CMP(INSN, COND) \ + static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ + uint32_t rn_ofs, uint32_t rm_ofs, \ + uint32_t oprsz, uint32_t maxsz) \ + { \ + tcg_gen_gvec_cmp(COND, vece, rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz); \ + } \ + DO_3SAME_NO_SZ_3(INSN, gen_##INSN##_3s) + +DO_3SAME_CMP(VCGT_S, TCG_COND_GT) +DO_3SAME_CMP(VCGT_U, TCG_COND_GTU) +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_VTST_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(rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz, &cmtst_op[vece]); +} +DO_3SAME_NO_SZ_3(VTST, gen_VTST_3s) 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) u ? &mls_op[size] : &mla_op[size]); return 0; - case NEON_3R_VTST_VCEQ: - if (u) { /* VCEQ */ - tcg_gen_gvec_cmp(TCG_COND_EQ, size, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - } else { /* VTST */ - tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size, &cmtst_op[size]); - } - return 0; - - case NEON_3R_VCGT: - tcg_gen_gvec_cmp(u ? TCG_COND_GTU : TCG_COND_GT, size, - rd_ofs, rn_ofs, rm_ofs, vec_size, vec_size); - return 0; - - case NEON_3R_VCGE: - tcg_gen_gvec_cmp(u ? TCG_COND_GEU : TCG_COND_GE, size, - rd_ofs, rn_ofs, rm_ofs, vec_size, vec_size); - return 0; - case NEON_3R_VSHL: /* Note the operation is vshl vd,vm,vn */ tcg_gen_gvec_3(rd_ofs, rm_ofs, rn_ofs, vec_size, vec_size, @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_LOGIC: case NEON_3R_VMAX: case NEON_3R_VMIN: + case NEON_3R_VTST_VCEQ: + case NEON_3R_VCGT: + case NEON_3R_VCGE: /* Already handled by decodetree */ return 1; } -- 2.20.1 Convert the Neon VQADD/VQSUB insns in the 3-reg-same grouping to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-19-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 6 ++++++ target/arm/translate-neon.inc.c | 15 +++++++++++++++ target/arm/translate.c | 14 ++------------ 3 files changed, 23 insertions(+), 12 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 @@ @3same .... ... . . . size:2 .... .... .... . q:1 . . .... \ &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp +VQADD_S_3s 1111 001 0 0 . .. .... .... 0000 . . . 1 .... @3same +VQADD_U_3s 1111 001 1 0 . .. .... .... 0000 . . . 1 .... @3same + @3same_logic .... ... . . . .. .... .... .... . q:1 .. .... \ &3same vm=%vm_dp vn=%vn_dp vd=%vd_dp size=0 @@ -XXX,XX +XXX,XX @@ VBSL_3s 1111 001 1 0 . 01 .... .... 0001 ... 1 .... @3same_logic VBIT_3s 1111 001 1 0 . 10 .... .... 0001 ... 1 .... @3same_logic VBIF_3s 1111 001 1 0 . 11 .... .... 0001 ... 1 .... @3same_logic +VQSUB_S_3s 1111 001 0 0 . .. .... .... 0010 . . . 1 .... @3same +VQSUB_U_3s 1111 001 1 0 . .. .... .... 0010 . . . 1 .... @3same + VCGT_S_3s 1111 001 0 0 . .. .... .... 0011 . . . 0 .... @3same VCGT_U_3s 1111 001 1 0 . .. .... .... 0011 . . . 0 .... @3same VCGE_S_3s 1111 001 0 0 . .. .... .... 0011 . . . 1 .... @3same 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 void gen_VTST_3s(unsigned vece, uint32_t rd_ofs, uint32_t rn_ofs, tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, oprsz, maxsz, &cmtst_op[vece]); } DO_3SAME_NO_SZ_3(VTST, gen_VTST_3s) + +#define DO_3SAME_GVEC4(INSN, OPARRAY) \ + static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ + uint32_t rn_ofs, uint32_t rm_ofs, \ + uint32_t oprsz, uint32_t maxsz) \ + { \ + tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc), \ + rn_ofs, rm_ofs, oprsz, maxsz, &OPARRAY[vece]); \ + } \ + DO_3SAME(INSN, gen_##INSN##_3s) + +DO_3SAME_GVEC4(VQADD_S, sqadd_op) +DO_3SAME_GVEC4(VQADD_U, uqadd_op) +DO_3SAME_GVEC4(VQSUB_S, sqsub_op) +DO_3SAME_GVEC4(VQSUB_U, uqsub_op) 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) } return 1; - case NEON_3R_VQADD: - tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc), - rn_ofs, rm_ofs, vec_size, vec_size, - (u ? uqadd_op : sqadd_op) + size); - return 0; - - case NEON_3R_VQSUB: - tcg_gen_gvec_4(rd_ofs, offsetof(CPUARMState, vfp.qc), - rn_ofs, rm_ofs, vec_size, vec_size, - (u ? uqsub_op : sqsub_op) + size); - return 0; - case NEON_3R_VMUL: /* VMUL */ if (u) { /* Polynomial case allows only P8. */ @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VTST_VCEQ: case NEON_3R_VCGT: case NEON_3R_VCGE: + case NEON_3R_VQADD: + case NEON_3R_VQSUB: /* Already handled by decodetree */ return 1; } -- 2.20.1 Convert the Neon VMUL, VMLA, VMLS and VSHL insns in the 3-reg-same grouping to decodetree. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-20-peter.maydell@linaro.org --- target/arm/neon-dp.decode | 9 +++++++ target/arm/translate-neon.inc.c | 44 +++++++++++++++++++++++++++++++++ target/arm/translate.c | 28 +++------------------ 3 files changed, 56 insertions(+), 25 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 @@ VCGT_U_3s 1111 001 1 0 . .. .... .... 0011 . . . 0 .... @3same VCGE_S_3s 1111 001 0 0 . .. .... .... 0011 . . . 1 .... @3same VCGE_U_3s 1111 001 1 0 . .. .... .... 0011 . . . 1 .... @3same +VSHL_S_3s 1111 001 0 0 . .. .... .... 0100 . . . 0 .... @3same +VSHL_U_3s 1111 001 1 0 . .. .... .... 0100 . . . 0 .... @3same + VMAX_S_3s 1111 001 0 0 . .. .... .... 0110 . . . 0 .... @3same VMAX_U_3s 1111 001 1 0 . .. .... .... 0110 . . . 0 .... @3same VMIN_S_3s 1111 001 0 0 . .. .... .... 0110 . . . 1 .... @3same @@ -XXX,XX +XXX,XX @@ VSUB_3s 1111 001 1 0 . .. .... .... 1000 . . . 0 .... @3same VTST_3s 1111 001 0 0 . .. .... .... 1000 . . . 1 .... @3same VCEQ_3s 1111 001 1 0 . .. .... .... 1000 . . . 1 .... @3same + +VMLA_3s 1111 001 0 0 . .. .... .... 1001 . . . 0 .... @3same +VMLS_3s 1111 001 1 0 . .. .... .... 1001 . . . 0 .... @3same + +VMUL_3s 1111 001 0 0 . .. .... .... 1001 . . . 1 .... @3same +VMUL_p_3s 1111 001 1 0 . .. .... .... 1001 . . . 1 .... @3same 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_NO_SZ_3(VMAX_S, tcg_gen_gvec_smax) DO_3SAME_NO_SZ_3(VMAX_U, tcg_gen_gvec_umax) DO_3SAME_NO_SZ_3(VMIN_S, tcg_gen_gvec_smin) DO_3SAME_NO_SZ_3(VMIN_U, tcg_gen_gvec_umin) +DO_3SAME_NO_SZ_3(VMUL, tcg_gen_gvec_mul) #define DO_3SAME_CMP(INSN, COND) \ static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ @@ -XXX,XX +XXX,XX @@ DO_3SAME_GVEC4(VQADD_S, sqadd_op) DO_3SAME_GVEC4(VQADD_U, uqadd_op) DO_3SAME_GVEC4(VQSUB_S, sqsub_op) DO_3SAME_GVEC4(VQSUB_U, uqsub_op) + +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); +} + +static bool trans_VMUL_p_3s(DisasContext *s, arg_3same *a) +{ + if (a->size != 0) { + return false; + } + return do_3same(s, a, gen_VMUL_p_3s); +} + +#define DO_3SAME_GVEC3_NO_SZ_3(INSN, OPARRAY) \ + static void gen_##INSN##_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(rd_ofs, rn_ofs, rm_ofs, \ + oprsz, maxsz, &OPARRAY[vece]); \ + } \ + DO_3SAME_NO_SZ_3(INSN, gen_##INSN##_3s) + + +DO_3SAME_GVEC3_NO_SZ_3(VMLA, mla_op) +DO_3SAME_GVEC3_NO_SZ_3(VMLS, mls_op) + +#define DO_3SAME_GVEC3_SHIFT(INSN, OPARRAY) \ + static void gen_##INSN##_3s(unsigned vece, uint32_t rd_ofs, \ + uint32_t rn_ofs, uint32_t rm_ofs, \ + uint32_t oprsz, uint32_t maxsz) \ + { \ + /* Note the operation is vshl vd,vm,vn */ \ + tcg_gen_gvec_3(rd_ofs, rm_ofs, rn_ofs, \ + oprsz, maxsz, &OPARRAY[vece]); \ + } \ + DO_3SAME(INSN, gen_##INSN##_3s) + +DO_3SAME_GVEC3_SHIFT(VSHL_S, sshl_op) +DO_3SAME_GVEC3_SHIFT(VSHL_U, ushl_op) 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) } return 1; - case NEON_3R_VMUL: /* VMUL */ - if (u) { - /* Polynomial case allows only P8. */ - if (size != 0) { - return 1; - } - tcg_gen_gvec_3_ool(rd_ofs, rn_ofs, rm_ofs, vec_size, vec_size, - 0, gen_helper_gvec_pmul_b); - } else { - tcg_gen_gvec_mul(size, rd_ofs, rn_ofs, rm_ofs, - vec_size, vec_size); - } - return 0; - - case NEON_3R_VML: /* VMLA, VMLS */ - tcg_gen_gvec_3(rd_ofs, rn_ofs, rm_ofs, vec_size, vec_size, - u ? &mls_op[size] : &mla_op[size]); - return 0; - - case NEON_3R_VSHL: - /* Note the operation is vshl vd,vm,vn */ - tcg_gen_gvec_3(rd_ofs, rm_ofs, rn_ofs, vec_size, vec_size, - u ? &ushl_op[size] : &sshl_op[size]); - return 0; - case NEON_3R_VADD_VSUB: case NEON_3R_LOGIC: case NEON_3R_VMAX: @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) case NEON_3R_VCGE: case NEON_3R_VQADD: case NEON_3R_VQSUB: + case NEON_3R_VMUL: + case NEON_3R_VML: + case NEON_3R_VSHL: /* Already handled by decodetree */ return 1; } -- 2.20.1 We're going to want at least some of the NeonGen* typedefs for the refactored 32-bit Neon decoder, so move them all to translate.h since it makes more sense to keep them in one group. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200430181003.21682-23-peter.maydell@linaro.org --- target/arm/translate.h | 17 +++++++++++++++++ target/arm/translate-a64.c | 17 ----------------- 2 files changed, 17 insertions(+), 17 deletions(-) diff --git a/target/arm/translate.h b/target/arm/translate.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate.h +++ b/target/arm/translate.h @@ -XXX,XX +XXX,XX @@ typedef void GVecGen3Fn(unsigned, uint32_t, uint32_t, typedef void GVecGen4Fn(unsigned, uint32_t, uint32_t, uint32_t, uint32_t, uint32_t, uint32_t); +/* Function prototype for gen_ functions for calling Neon helpers */ +typedef void NeonGenOneOpEnvFn(TCGv_i32, TCGv_ptr, TCGv_i32); +typedef void NeonGenTwoOpFn(TCGv_i32, TCGv_i32, TCGv_i32); +typedef void NeonGenTwoOpEnvFn(TCGv_i32, TCGv_ptr, TCGv_i32, TCGv_i32); +typedef void NeonGenTwo64OpFn(TCGv_i64, TCGv_i64, TCGv_i64); +typedef void NeonGenTwo64OpEnvFn(TCGv_i64, TCGv_ptr, TCGv_i64, TCGv_i64); +typedef void NeonGenNarrowFn(TCGv_i32, TCGv_i64); +typedef void NeonGenNarrowEnvFn(TCGv_i32, TCGv_ptr, TCGv_i64); +typedef void NeonGenWidenFn(TCGv_i64, TCGv_i32); +typedef void NeonGenTwoSingleOPFn(TCGv_i32, TCGv_i32, TCGv_i32, TCGv_ptr); +typedef void NeonGenTwoDoubleOPFn(TCGv_i64, TCGv_i64, TCGv_i64, TCGv_ptr); +typedef void NeonGenOneOpFn(TCGv_i64, TCGv_i64); +typedef void CryptoTwoOpFn(TCGv_ptr, TCGv_ptr); +typedef void CryptoThreeOpIntFn(TCGv_ptr, TCGv_ptr, TCGv_i32); +typedef void CryptoThreeOpFn(TCGv_ptr, TCGv_ptr, TCGv_ptr); +typedef void AtomicThreeOpFn(TCGv_i64, TCGv_i64, TCGv_i64, TCGArg, MemOp); + #endif /* TARGET_ARM_TRANSLATE_H */ 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 @@ typedef struct AArch64DecodeTable { AArch64DecodeFn *disas_fn; } AArch64DecodeTable; -/* Function prototype for gen_ functions for calling Neon helpers */ -typedef void NeonGenOneOpEnvFn(TCGv_i32, TCGv_ptr, TCGv_i32); -typedef void NeonGenTwoOpFn(TCGv_i32, TCGv_i32, TCGv_i32); -typedef void NeonGenTwoOpEnvFn(TCGv_i32, TCGv_ptr, TCGv_i32, TCGv_i32); -typedef void NeonGenTwo64OpFn(TCGv_i64, TCGv_i64, TCGv_i64); -typedef void NeonGenTwo64OpEnvFn(TCGv_i64, TCGv_ptr, TCGv_i64, TCGv_i64); -typedef void NeonGenNarrowFn(TCGv_i32, TCGv_i64); -typedef void NeonGenNarrowEnvFn(TCGv_i32, TCGv_ptr, TCGv_i64); -typedef void NeonGenWidenFn(TCGv_i64, TCGv_i32); -typedef void NeonGenTwoSingleOPFn(TCGv_i32, TCGv_i32, TCGv_i32, TCGv_ptr); -typedef void NeonGenTwoDoubleOPFn(TCGv_i64, TCGv_i64, TCGv_i64, TCGv_ptr); -typedef void NeonGenOneOpFn(TCGv_i64, TCGv_i64); -typedef void CryptoTwoOpFn(TCGv_ptr, TCGv_ptr); -typedef void CryptoThreeOpIntFn(TCGv_ptr, TCGv_ptr, TCGv_i32); -typedef void CryptoThreeOpFn(TCGv_ptr, TCGv_ptr, TCGv_ptr); -typedef void AtomicThreeOpFn(TCGv_i64, TCGv_i64, TCGv_i64, TCGArg, MemOp); - /* initialize TCG globals. */ void a64_translate_init(void) { -- 2.20.1 A grab-bag of minor stuff for the end of the year. My to-review queue is not empty, but it it at least in single figures... -- PMM The following changes since commit 5bfbd8170ce7acb98a1834ff49ed7340b0837144: Merge remote-tracking branch 'remotes/vivier2/tags/trivial-branch-for-6.0-pull-request' into staging (2020-12-14 20:32:38 +0000) are available in the Git repository at: https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20201215 for you to fetch changes up to 23af268566069183285bebbdf95b1b37cb7c0942: hw/block/m25p80: Fix Numonyx fast read dummy cycle count (2020-12-15 13:39:30 +0000) ---------------------------------------------------------------- target-arm queue: * gdbstub: Correct misparsing of vCont C/S requests * openrisc: Move pic_cpu code into CPU object proper * nios2: Move IIC code into CPU object proper * Improve reporting of ROM overlap errors * xlnx-versal: Add USB support * hw/misc/zynq_slcr: Avoid #DIV/0! error * Numonyx: Fix dummy cycles and check for SPI mode on cmds ---------------------------------------------------------------- Joe Komlodi (4): hw/block/m25p80: Make Numonyx config field names more accurate hw/block/m25p80: Fix when VCFG XIP bit is set for Numonyx hw/block/m25p80: Check SPI mode before running some Numonyx commands hw/block/m25p80: Fix Numonyx fast read dummy cycle count Peter Maydell (11): gdbstub: Correct misparsing of vCont C/S requests hw/openrisc/openrisc_sim: Use IRQ splitter when connecting IRQ to multiple CPUs hw/openrisc/openrisc_sim: Abstract out "get IRQ x of CPU y" target/openrisc: Move pic_cpu code into CPU object proper target/nios2: Move IIC code into CPU object proper target/nios2: Move nios2_check_interrupts() into target/nios2 target/nios2: Use deposit32() to update ipending register hw/core/loader.c: Track last-seen ROM in rom_check_and_register_reset() hw/core/loader.c: Improve reporting of ROM overlap errors elf_ops.h: Don't truncate name of the ROM blobs we create elf_ops.h: Be more verbose with ROM blob names Philippe Mathieu-Daudé (1): hw/misc/zynq_slcr: Avoid #DIV/0! error Sai Pavan Boddu (2): usb: Add versal-usb2-ctrl-regs module usb: xlnx-usb-subsystem: Add xilinx usb subsystem Vikram Garhwal (2): usb: Add DWC3 model arm: xlnx-versal: Connect usb to virt-versal include/hw/arm/xlnx-versal.h | 9 + include/hw/elf_ops.h | 5 +- include/hw/usb/hcd-dwc3.h | 55 +++ include/hw/usb/xlnx-usb-subsystem.h | 45 ++ include/hw/usb/xlnx-versal-usb2-ctrl-regs.h | 45 ++ target/nios2/cpu.h | 3 - target/openrisc/cpu.h | 1 - gdbstub.c | 2 +- hw/arm/xlnx-versal-virt.c | 55 +++ hw/arm/xlnx-versal.c | 26 ++ hw/block/m25p80.c | 158 +++++-- hw/core/loader.c | 67 ++- hw/intc/nios2_iic.c | 95 ---- hw/misc/zynq_slcr.c | 5 + hw/nios2/10m50_devboard.c | 13 +- hw/nios2/cpu_pic.c | 67 --- hw/openrisc/openrisc_sim.c | 46 +- hw/openrisc/pic_cpu.c | 61 --- hw/usb/hcd-dwc3.c | 689 ++++++++++++++++++++++++++++ hw/usb/xlnx-usb-subsystem.c | 94 ++++ hw/usb/xlnx-versal-usb2-ctrl-regs.c | 229 +++++++++ softmmu/vl.c | 1 - target/nios2/cpu.c | 29 ++ target/nios2/op_helper.c | 9 + target/openrisc/cpu.c | 32 ++ MAINTAINERS | 1 - hw/intc/meson.build | 1 - hw/nios2/meson.build | 2 +- hw/openrisc/Kconfig | 1 + hw/openrisc/meson.build | 2 +- hw/usb/Kconfig | 10 + hw/usb/meson.build | 3 + 32 files changed, 1557 insertions(+), 304 deletions(-) create mode 100644 include/hw/usb/hcd-dwc3.h create mode 100644 include/hw/usb/xlnx-usb-subsystem.h create mode 100644 include/hw/usb/xlnx-versal-usb2-ctrl-regs.h delete mode 100644 hw/intc/nios2_iic.c delete mode 100644 hw/nios2/cpu_pic.c delete mode 100644 hw/openrisc/pic_cpu.c create mode 100644 hw/usb/hcd-dwc3.c create mode 100644 hw/usb/xlnx-usb-subsystem.c create mode 100644 hw/usb/xlnx-versal-usb2-ctrl-regs.c In the vCont packet, two of the command actions (C and S) take an argument specifying the signal to be sent to the process/thread, which is sent as an ASCII string of two hex digits which immediately follow the 'C' or 'S' character. Our code for parsing this packet accidentally skipped the first of the two bytes of the signal value, because it started parsing the hex string at 'p + 1' when the preceding code had already moved past the 'C' or 'S' with "cur_action = *p++". This meant that we would only do the right thing for signals below 10, and would misinterpret the rest. For instance, when the debugger wants to send the process a SIGPROF (27 on x86-64) we mangle this into a SIGSEGV (11). Remove the accidental double increment. Fixes: https://bugs.launchpad.net/qemu/+bug/1773743 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org> Message-id: 20201121210342.10089-1-peter.maydell@linaro.org --- gdbstub.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/gdbstub.c b/gdbstub.c index XXXXXXX..XXXXXXX 100644 --- a/gdbstub.c +++ b/gdbstub.c @@ -XXX,XX +XXX,XX @@ static int gdb_handle_vcont(const char *p) cur_action = *p++; if (cur_action == 'C' || cur_action == 'S') { cur_action = qemu_tolower(cur_action); - res = qemu_strtoul(p + 1, &p, 16, &tmp); + res = qemu_strtoul(p, &p, 16, &tmp); if (res) { goto out; } -- 2.20.1 openrisc_sim_net_init() attempts to connect the IRQ line from the ethernet device to both CPUs in an SMP configuration by simply caling sysbus_connect_irq() for it twice. This doesn't work, because the second connection simply overrides the first. Fix this by creating a TYPE_SPLIT_IRQ to split the IRQ in the SMP case. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Stafford Horne <shorne@gmail.com> Message-id: 20201127225127.14770-2-peter.maydell@linaro.org --- hw/openrisc/openrisc_sim.c | 13 +++++++++++-- hw/openrisc/Kconfig | 1 + 2 files changed, 12 insertions(+), 2 deletions(-) diff --git a/hw/openrisc/openrisc_sim.c b/hw/openrisc/openrisc_sim.c index XXXXXXX..XXXXXXX 100644 --- a/hw/openrisc/openrisc_sim.c +++ b/hw/openrisc/openrisc_sim.c @@ -XXX,XX +XXX,XX @@ #include "hw/sysbus.h" #include "sysemu/qtest.h" #include "sysemu/reset.h" +#include "hw/core/split-irq.h" #define KERNEL_LOAD_ADDR 0x100 @@ -XXX,XX +XXX,XX @@ static void openrisc_sim_net_init(hwaddr base, hwaddr descriptors, s = SYS_BUS_DEVICE(dev); sysbus_realize_and_unref(s, &error_fatal); - for (i = 0; i < num_cpus; i++) { - sysbus_connect_irq(s, 0, cpu_irqs[i][irq_pin]); + if (num_cpus > 1) { + DeviceState *splitter = qdev_new(TYPE_SPLIT_IRQ); + qdev_prop_set_uint32(splitter, "num-lines", num_cpus); + qdev_realize_and_unref(splitter, NULL, &error_fatal); + for (i = 0; i < num_cpus; i++) { + qdev_connect_gpio_out(splitter, i, cpu_irqs[i][irq_pin]); + } + sysbus_connect_irq(s, 0, qdev_get_gpio_in(splitter, 0)); + } else { + sysbus_connect_irq(s, 0, cpu_irqs[0][irq_pin]); } sysbus_mmio_map(s, 0, base); sysbus_mmio_map(s, 1, descriptors); diff --git a/hw/openrisc/Kconfig b/hw/openrisc/Kconfig index XXXXXXX..XXXXXXX 100644 --- a/hw/openrisc/Kconfig +++ b/hw/openrisc/Kconfig @@ -XXX,XX +XXX,XX @@ config OR1K_SIM select SERIAL select OPENCORES_ETH select OMPIC + select SPLIT_IRQ -- 2.20.1 We're about to refactor the OpenRISC pic_cpu code in a way that means that just grabbing the whole qemu_irq[] array of inbound IRQs for a CPU won't be possible any more. Abstract out a function for "return the qemu_irq for IRQ x input of CPU y" so we can more easily replace the implementation. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Stafford Horne <shorne@gmail.com> Message-id: 20201127225127.14770-3-peter.maydell@linaro.org --- hw/openrisc/openrisc_sim.c | 38 +++++++++++++++++++++----------------- 1 file changed, 21 insertions(+), 17 deletions(-) diff --git a/hw/openrisc/openrisc_sim.c b/hw/openrisc/openrisc_sim.c index XXXXXXX..XXXXXXX 100644 --- a/hw/openrisc/openrisc_sim.c +++ b/hw/openrisc/openrisc_sim.c @@ -XXX,XX +XXX,XX @@ static void main_cpu_reset(void *opaque) cpu_set_pc(cs, boot_info.bootstrap_pc); } +static qemu_irq get_cpu_irq(OpenRISCCPU *cpus[], int cpunum, int irq_pin) +{ + return cpus[cpunum]->env.irq[irq_pin]; +} + static void openrisc_sim_net_init(hwaddr base, hwaddr descriptors, - int num_cpus, qemu_irq **cpu_irqs, + int num_cpus, OpenRISCCPU *cpus[], int irq_pin, NICInfo *nd) { DeviceState *dev; @@ -XXX,XX +XXX,XX @@ static void openrisc_sim_net_init(hwaddr base, hwaddr descriptors, qdev_prop_set_uint32(splitter, "num-lines", num_cpus); qdev_realize_and_unref(splitter, NULL, &error_fatal); for (i = 0; i < num_cpus; i++) { - qdev_connect_gpio_out(splitter, i, cpu_irqs[i][irq_pin]); + qdev_connect_gpio_out(splitter, i, get_cpu_irq(cpus, i, irq_pin)); } sysbus_connect_irq(s, 0, qdev_get_gpio_in(splitter, 0)); } else { - sysbus_connect_irq(s, 0, cpu_irqs[0][irq_pin]); + sysbus_connect_irq(s, 0, get_cpu_irq(cpus, 0, irq_pin)); } sysbus_mmio_map(s, 0, base); sysbus_mmio_map(s, 1, descriptors); } static void openrisc_sim_ompic_init(hwaddr base, int num_cpus, - qemu_irq **cpu_irqs, int irq_pin) + OpenRISCCPU *cpus[], int irq_pin) { DeviceState *dev; SysBusDevice *s; @@ -XXX,XX +XXX,XX @@ static void openrisc_sim_ompic_init(hwaddr base, int num_cpus, s = SYS_BUS_DEVICE(dev); sysbus_realize_and_unref(s, &error_fatal); for (i = 0; i < num_cpus; i++) { - sysbus_connect_irq(s, i, cpu_irqs[i][irq_pin]); + sysbus_connect_irq(s, i, get_cpu_irq(cpus, i, irq_pin)); } sysbus_mmio_map(s, 0, base); } @@ -XXX,XX +XXX,XX @@ static void openrisc_sim_init(MachineState *machine) { ram_addr_t ram_size = machine->ram_size; const char *kernel_filename = machine->kernel_filename; - OpenRISCCPU *cpu = NULL; + OpenRISCCPU *cpus[2] = {}; MemoryRegion *ram; - qemu_irq *cpu_irqs[2]; qemu_irq serial_irq; int n; unsigned int smp_cpus = machine->smp.cpus; assert(smp_cpus >= 1 && smp_cpus <= 2); for (n = 0; n < smp_cpus; n++) { - cpu = OPENRISC_CPU(cpu_create(machine->cpu_type)); - if (cpu == NULL) { + cpus[n] = OPENRISC_CPU(cpu_create(machine->cpu_type)); + if (cpus[n] == NULL) { fprintf(stderr, "Unable to find CPU definition!\n"); exit(1); } - cpu_openrisc_pic_init(cpu); - cpu_irqs[n] = (qemu_irq *) cpu->env.irq; + cpu_openrisc_pic_init(cpus[n]); - cpu_openrisc_clock_init(cpu); + cpu_openrisc_clock_init(cpus[n]); - qemu_register_reset(main_cpu_reset, cpu); + qemu_register_reset(main_cpu_reset, cpus[n]); } ram = g_malloc(sizeof(*ram)); @@ -XXX,XX +XXX,XX @@ static void openrisc_sim_init(MachineState *machine) if (nd_table[0].used) { openrisc_sim_net_init(0x92000000, 0x92000400, smp_cpus, - cpu_irqs, 4, nd_table); + cpus, 4, nd_table); } if (smp_cpus > 1) { - openrisc_sim_ompic_init(0x98000000, smp_cpus, cpu_irqs, 1); + openrisc_sim_ompic_init(0x98000000, smp_cpus, cpus, 1); - serial_irq = qemu_irq_split(cpu_irqs[0][2], cpu_irqs[1][2]); + serial_irq = qemu_irq_split(get_cpu_irq(cpus, 0, 2), + get_cpu_irq(cpus, 1, 2)); } else { - serial_irq = cpu_irqs[0][2]; + serial_irq = get_cpu_irq(cpus, 0, 2); } serial_mm_init(get_system_memory(), 0x90000000, 0, serial_irq, -- 2.20.1 The openrisc code uses an old style of interrupt handling, where a separate standalone set of qemu_irqs invoke a function openrisc_pic_cpu_handler() which signals the interrupt to the CPU proper by directly calling cpu_interrupt() and cpu_reset_interrupt(). Because CPU objects now inherit (indirectly) from TYPE_DEVICE, they can have GPIO input lines themselves, and the neater modern way to implement this is to simply have the CPU object itself provide the input IRQ lines. Create GPIO inputs to the OpenRISC CPU object, and make the only user of cpu_openrisc_pic_init() wire up directly to those instead. This allows us to delete the hw/openrisc/pic_cpu.c file entirely. This fixes a trivial memory leak reported by Coverity of the IRQs allocated in cpu_openrisc_pic_init(). Fixes: Coverity CID 1421934 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Stafford Horne <shorne@gmail.com> Message-id: 20201127225127.14770-4-peter.maydell@linaro.org --- target/openrisc/cpu.h | 1 - hw/openrisc/openrisc_sim.c | 3 +- hw/openrisc/pic_cpu.c | 61 -------------------------------------- target/openrisc/cpu.c | 32 ++++++++++++++++++++ hw/openrisc/meson.build | 2 +- 5 files changed, 34 insertions(+), 65 deletions(-) delete mode 100644 hw/openrisc/pic_cpu.c diff --git a/target/openrisc/cpu.h b/target/openrisc/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/openrisc/cpu.h +++ b/target/openrisc/cpu.h @@ -XXX,XX +XXX,XX @@ typedef struct CPUOpenRISCState { uint32_t picmr; /* Interrupt mask register */ uint32_t picsr; /* Interrupt contrl register*/ #endif - void *irq[32]; /* Interrupt irq input */ } CPUOpenRISCState; /** diff --git a/hw/openrisc/openrisc_sim.c b/hw/openrisc/openrisc_sim.c index XXXXXXX..XXXXXXX 100644 --- a/hw/openrisc/openrisc_sim.c +++ b/hw/openrisc/openrisc_sim.c @@ -XXX,XX +XXX,XX @@ static void main_cpu_reset(void *opaque) static qemu_irq get_cpu_irq(OpenRISCCPU *cpus[], int cpunum, int irq_pin) { - return cpus[cpunum]->env.irq[irq_pin]; + return qdev_get_gpio_in_named(DEVICE(cpus[cpunum]), "IRQ", irq_pin); } static void openrisc_sim_net_init(hwaddr base, hwaddr descriptors, @@ -XXX,XX +XXX,XX @@ static void openrisc_sim_init(MachineState *machine) fprintf(stderr, "Unable to find CPU definition!\n"); exit(1); } - cpu_openrisc_pic_init(cpus[n]); cpu_openrisc_clock_init(cpus[n]); diff --git a/hw/openrisc/pic_cpu.c b/hw/openrisc/pic_cpu.c deleted file mode 100644 index XXXXXXX..XXXXXXX --- a/hw/openrisc/pic_cpu.c +++ /dev/null @@ -XXX,XX +XXX,XX @@ -/* - * OpenRISC Programmable Interrupt Controller support. - * - * Copyright (c) 2011-2012 Jia Liu <proljc@gmail.com> - * Feng Gao <gf91597@gmail.com> - * - * 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 2.1 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/>. - */ - -#include "qemu/osdep.h" -#include "hw/irq.h" -#include "cpu.h" - -/* OpenRISC pic handler */ -static void openrisc_pic_cpu_handler(void *opaque, int irq, int level) -{ - OpenRISCCPU *cpu = (OpenRISCCPU *)opaque; - CPUState *cs = CPU(cpu); - uint32_t irq_bit; - - if (irq > 31 || irq < 0) { - return; - } - - irq_bit = 1U << irq; - - if (level) { - cpu->env.picsr |= irq_bit; - } else { - cpu->env.picsr &= ~irq_bit; - } - - if (cpu->env.picsr & cpu->env.picmr) { - cpu_interrupt(cs, CPU_INTERRUPT_HARD); - } else { - cpu_reset_interrupt(cs, CPU_INTERRUPT_HARD); - cpu->env.picsr = 0; - } -} - -void cpu_openrisc_pic_init(OpenRISCCPU *cpu) -{ - int i; - qemu_irq *qi; - qi = qemu_allocate_irqs(openrisc_pic_cpu_handler, cpu, NR_IRQS); - - for (i = 0; i < NR_IRQS; i++) { - cpu->env.irq[i] = qi[i]; - } -} diff --git a/target/openrisc/cpu.c b/target/openrisc/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/openrisc/cpu.c +++ b/target/openrisc/cpu.c @@ -XXX,XX +XXX,XX @@ static void openrisc_cpu_reset(DeviceState *dev) #endif } +#ifndef CONFIG_USER_ONLY +static void openrisc_cpu_set_irq(void *opaque, int irq, int level) +{ + OpenRISCCPU *cpu = (OpenRISCCPU *)opaque; + CPUState *cs = CPU(cpu); + uint32_t irq_bit; + + if (irq > 31 || irq < 0) { + return; + } + + irq_bit = 1U << irq; + + if (level) { + cpu->env.picsr |= irq_bit; + } else { + cpu->env.picsr &= ~irq_bit; + } + + if (cpu->env.picsr & cpu->env.picmr) { + cpu_interrupt(cs, CPU_INTERRUPT_HARD); + } else { + cpu_reset_interrupt(cs, CPU_INTERRUPT_HARD); + cpu->env.picsr = 0; + } +} +#endif + static void openrisc_cpu_realizefn(DeviceState *dev, Error **errp) { CPUState *cs = CPU(dev); @@ -XXX,XX +XXX,XX @@ static void openrisc_cpu_initfn(Object *obj) OpenRISCCPU *cpu = OPENRISC_CPU(obj); cpu_set_cpustate_pointers(cpu); + +#ifndef CONFIG_USER_ONLY + qdev_init_gpio_in_named(DEVICE(cpu), openrisc_cpu_set_irq, "IRQ", NR_IRQS); +#endif } /* CPU models */ diff --git a/hw/openrisc/meson.build b/hw/openrisc/meson.build index XXXXXXX..XXXXXXX 100644 --- a/hw/openrisc/meson.build +++ b/hw/openrisc/meson.build @@ -XXX,XX +XXX,XX @@ openrisc_ss = ss.source_set() -openrisc_ss.add(files('pic_cpu.c', 'cputimer.c')) +openrisc_ss.add(files('cputimer.c')) openrisc_ss.add(when: 'CONFIG_OR1K_SIM', if_true: files('openrisc_sim.c')) hw_arch += {'openrisc': openrisc_ss} -- 2.20.1 The Nios2 architecture supports two different interrupt controller options: * The IIC (Internal Interrupt Controller) is part of the CPU itself; it has 32 IRQ input lines and no NMI support. Interrupt status is queried and controlled via the CPU's ipending and istatus registers. * The EIC (External Interrupt Controller) interface allows the CPU to connect to an external interrupt controller. The interface allows the interrupt controller to present a packet of information containing: - handler address - interrupt level - register set - NMI mode QEMU does not model an EIC currently. We do model the IIC, but its implementation is split across code in hw/nios2/cpu_pic.c and hw/intc/nios2_iic.c. The code in those two files has no state of its own -- the IIC state is in the Nios2CPU state struct. Because CPU objects now inherit (indirectly) from TYPE_DEVICE, they can have GPIO input lines themselves, so we can implement the IIC directly in the CPU object the same way that real hardware does. Create named "IRQ" GPIO inputs to the Nios2 CPU object, and make the only user of the IIC wire up directly to those instead. Note that the old code had an "NMI" concept which was entirely unused and also as far as I can see not architecturally correct, since only the EIC has a concept of an NMI. This fixes a Coverity-reported trivial memory leak of the IRQ array allocated in nios2_cpu_pic_init(). Fixes: Coverity CID 1421916 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 20201129174022.26530-2-peter.maydell@linaro.org Reviewed-by: Wentong Wu <wentong.wu@intel.com> Tested-by: Wentong Wu <wentong.wu@intel.com> --- target/nios2/cpu.h | 1 - hw/intc/nios2_iic.c | 95 --------------------------------------- hw/nios2/10m50_devboard.c | 13 +----- hw/nios2/cpu_pic.c | 31 ------------- target/nios2/cpu.c | 30 +++++++++++++ MAINTAINERS | 1 - hw/intc/meson.build | 1 - 7 files changed, 32 insertions(+), 140 deletions(-) delete mode 100644 hw/intc/nios2_iic.c diff --git a/target/nios2/cpu.h b/target/nios2/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/nios2/cpu.h +++ b/target/nios2/cpu.h @@ -XXX,XX +XXX,XX @@ void nios2_cpu_do_unaligned_access(CPUState *cpu, vaddr addr, MMUAccessType access_type, int mmu_idx, uintptr_t retaddr); -qemu_irq *nios2_cpu_pic_init(Nios2CPU *cpu); void nios2_check_interrupts(CPUNios2State *env); void do_nios2_semihosting(CPUNios2State *env); diff --git a/hw/intc/nios2_iic.c b/hw/intc/nios2_iic.c deleted file mode 100644 index XXXXXXX..XXXXXXX --- a/hw/intc/nios2_iic.c +++ /dev/null @@ -XXX,XX +XXX,XX @@ -/* - * QEMU Altera Internal Interrupt Controller. - * - * Copyright (c) 2012 Chris Wulff <crwulff@gmail.com> - * - * 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 2.1 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/lgpl-2.1.html> - */ - -#include "qemu/osdep.h" -#include "qemu/module.h" -#include "qapi/error.h" - -#include "hw/irq.h" -#include "hw/sysbus.h" -#include "cpu.h" -#include "qom/object.h" - -#define TYPE_ALTERA_IIC "altera,iic" -OBJECT_DECLARE_SIMPLE_TYPE(AlteraIIC, ALTERA_IIC) - -struct AlteraIIC { - SysBusDevice parent_obj; - void *cpu; - qemu_irq parent_irq; -}; - -static void update_irq(AlteraIIC *pv) -{ - CPUNios2State *env = &((Nios2CPU *)(pv->cpu))->env; - - qemu_set_irq(pv->parent_irq, - env->regs[CR_IPENDING] & env->regs[CR_IENABLE]); -} - -static void irq_handler(void *opaque, int irq, int level) -{ - AlteraIIC *pv = opaque; - CPUNios2State *env = &((Nios2CPU *)(pv->cpu))->env; - - env->regs[CR_IPENDING] &= ~(1 << irq); - env->regs[CR_IPENDING] |= !!level << irq; - - update_irq(pv); -} - -static void altera_iic_init(Object *obj) -{ - AlteraIIC *pv = ALTERA_IIC(obj); - - qdev_init_gpio_in(DEVICE(pv), irq_handler, 32); - sysbus_init_irq(SYS_BUS_DEVICE(obj), &pv->parent_irq); -} - -static void altera_iic_realize(DeviceState *dev, Error **errp) -{ - struct AlteraIIC *pv = ALTERA_IIC(dev); - - pv->cpu = object_property_get_link(OBJECT(dev), "cpu", &error_abort); -} - -static void altera_iic_class_init(ObjectClass *klass, void *data) -{ - DeviceClass *dc = DEVICE_CLASS(klass); - - /* Reason: needs to be wired up, e.g. by nios2_10m50_ghrd_init() */ - dc->user_creatable = false; - dc->realize = altera_iic_realize; -} - -static TypeInfo altera_iic_info = { - .name = TYPE_ALTERA_IIC, - .parent = TYPE_SYS_BUS_DEVICE, - .instance_size = sizeof(AlteraIIC), - .instance_init = altera_iic_init, - .class_init = altera_iic_class_init, -}; - -static void altera_iic_register(void) -{ - type_register_static(&altera_iic_info); -} - -type_init(altera_iic_register) diff --git a/hw/nios2/10m50_devboard.c b/hw/nios2/10m50_devboard.c index XXXXXXX..XXXXXXX 100644 --- a/hw/nios2/10m50_devboard.c +++ b/hw/nios2/10m50_devboard.c @@ -XXX,XX +XXX,XX @@ static void nios2_10m50_ghrd_init(MachineState *machine) ram_addr_t tcm_size = 0x1000; /* 1 kiB, but QEMU limit is 4 kiB */ ram_addr_t ram_base = 0x08000000; ram_addr_t ram_size = 0x08000000; - qemu_irq *cpu_irq, irq[32]; + qemu_irq irq[32]; int i; /* Physical TCM (tb_ram_1k) with alias at 0xc0000000 */ @@ -XXX,XX +XXX,XX @@ static void nios2_10m50_ghrd_init(MachineState *machine) /* Create CPU -- FIXME */ cpu = NIOS2_CPU(cpu_create(TYPE_NIOS2_CPU)); - - /* Register: CPU interrupt controller (PIC) */ - cpu_irq = nios2_cpu_pic_init(cpu); - - /* Register: Internal Interrupt Controller (IIC) */ - dev = qdev_new("altera,iic"); - object_property_add_const_link(OBJECT(dev), "cpu", OBJECT(cpu)); - sysbus_realize_and_unref(SYS_BUS_DEVICE(dev), &error_fatal); - sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0, cpu_irq[0]); for (i = 0; i < 32; i++) { - irq[i] = qdev_get_gpio_in(dev, i); + irq[i] = qdev_get_gpio_in_named(DEVICE(cpu), "IRQ", i); } /* Register: Altera 16550 UART */ diff --git a/hw/nios2/cpu_pic.c b/hw/nios2/cpu_pic.c index XXXXXXX..XXXXXXX 100644 --- a/hw/nios2/cpu_pic.c +++ b/hw/nios2/cpu_pic.c @@ -XXX,XX +XXX,XX @@ #include "boot.h" -static void nios2_pic_cpu_handler(void *opaque, int irq, int level) -{ - Nios2CPU *cpu = opaque; - CPUNios2State *env = &cpu->env; - CPUState *cs = CPU(cpu); - int type = irq ? CPU_INTERRUPT_NMI : CPU_INTERRUPT_HARD; - - if (type == CPU_INTERRUPT_HARD) { - env->irq_pending = level; - - if (level && (env->regs[CR_STATUS] & CR_STATUS_PIE)) { - env->irq_pending = 0; - cpu_interrupt(cs, type); - } else if (!level) { - env->irq_pending = 0; - cpu_reset_interrupt(cs, type); - } - } else { - if (level) { - cpu_interrupt(cs, type); - } else { - cpu_reset_interrupt(cs, type); - } - } -} - void nios2_check_interrupts(CPUNios2State *env) { if (env->irq_pending && @@ -XXX,XX +XXX,XX @@ void nios2_check_interrupts(CPUNios2State *env) cpu_interrupt(env_cpu(env), CPU_INTERRUPT_HARD); } } - -qemu_irq *nios2_cpu_pic_init(Nios2CPU *cpu) -{ - return qemu_allocate_irqs(nios2_pic_cpu_handler, cpu, 2); -} diff --git a/target/nios2/cpu.c b/target/nios2/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/nios2/cpu.c +++ b/target/nios2/cpu.c @@ -XXX,XX +XXX,XX @@ static void nios2_cpu_reset(DeviceState *dev) #endif } +#ifndef CONFIG_USER_ONLY +static void nios2_cpu_set_irq(void *opaque, int irq, int level) +{ + Nios2CPU *cpu = opaque; + CPUNios2State *env = &cpu->env; + CPUState *cs = CPU(cpu); + + env->regs[CR_IPENDING] &= ~(1 << irq); + env->regs[CR_IPENDING] |= !!level << irq; + + env->irq_pending = env->regs[CR_IPENDING] & env->regs[CR_IENABLE]; + + if (env->irq_pending && (env->regs[CR_STATUS] & CR_STATUS_PIE)) { + env->irq_pending = 0; + cpu_interrupt(cs, CPU_INTERRUPT_HARD); + } else if (!env->irq_pending) { + cpu_reset_interrupt(cs, CPU_INTERRUPT_HARD); + } +} +#endif + static void nios2_cpu_initfn(Object *obj) { Nios2CPU *cpu = NIOS2_CPU(obj); @@ -XXX,XX +XXX,XX @@ static void nios2_cpu_initfn(Object *obj) #if !defined(CONFIG_USER_ONLY) mmu_init(&cpu->env); + + /* + * These interrupt lines model the IIC (internal interrupt + * controller). QEMU does not currently support the EIC + * (external interrupt controller) -- if we did it would be + * a separate device in hw/intc with a custom interface to + * the CPU, and boards using it would not wire up these IRQ lines. + */ + qdev_init_gpio_in_named(DEVICE(cpu), nios2_cpu_set_irq, "IRQ", 32); #endif } diff --git a/MAINTAINERS b/MAINTAINERS index XXXXXXX..XXXXXXX 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -XXX,XX +XXX,XX @@ M: Marek Vasut <marex@denx.de> S: Maintained F: target/nios2/ F: hw/nios2/ -F: hw/intc/nios2_iic.c F: disas/nios2.c F: default-configs/nios2-softmmu.mak diff --git a/hw/intc/meson.build b/hw/intc/meson.build index XXXXXXX..XXXXXXX 100644 --- a/hw/intc/meson.build +++ b/hw/intc/meson.build @@ -XXX,XX +XXX,XX @@ specific_ss.add(when: 'CONFIG_IBEX', if_true: files('ibex_plic.c')) specific_ss.add(when: 'CONFIG_IOAPIC', if_true: files('ioapic.c')) specific_ss.add(when: 'CONFIG_LOONGSON_LIOINTC', if_true: files('loongson_liointc.c')) specific_ss.add(when: 'CONFIG_MIPS_CPS', if_true: files('mips_gic.c')) -specific_ss.add(when: 'CONFIG_NIOS2', if_true: files('nios2_iic.c')) specific_ss.add(when: 'CONFIG_OMAP', if_true: files('omap_intc.c')) specific_ss.add(when: 'CONFIG_OMPIC', if_true: files('ompic.c')) specific_ss.add(when: 'CONFIG_OPENPIC_KVM', if_true: files('openpic_kvm.c')) -- 2.20.1 The function nios2_check_interrupts)() looks only at CPU-internal state; it belongs in target/nios2, not hw/nios2. Move it into the same file as its only caller, so it can just be local to that file. This removes the only remaining code from cpu_pic.c, so we can delete that file entirely. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 20201129174022.26530-3-peter.maydell@linaro.org Reviewed-by: Wentong Wu <wentong.wu@intel.com> Tested-by: Wentong Wu <wentong.wu@intel.com> --- target/nios2/cpu.h | 2 -- hw/nios2/cpu_pic.c | 36 ------------------------------------ target/nios2/op_helper.c | 9 +++++++++ hw/nios2/meson.build | 2 +- 4 files changed, 10 insertions(+), 39 deletions(-) delete mode 100644 hw/nios2/cpu_pic.c diff --git a/target/nios2/cpu.h b/target/nios2/cpu.h index XXXXXXX..XXXXXXX 100644 --- a/target/nios2/cpu.h +++ b/target/nios2/cpu.h @@ -XXX,XX +XXX,XX @@ void nios2_cpu_do_unaligned_access(CPUState *cpu, vaddr addr, MMUAccessType access_type, int mmu_idx, uintptr_t retaddr); -void nios2_check_interrupts(CPUNios2State *env); - void do_nios2_semihosting(CPUNios2State *env); #define CPU_RESOLVING_TYPE TYPE_NIOS2_CPU diff --git a/hw/nios2/cpu_pic.c b/hw/nios2/cpu_pic.c deleted file mode 100644 index XXXXXXX..XXXXXXX --- a/hw/nios2/cpu_pic.c +++ /dev/null @@ -XXX,XX +XXX,XX @@ -/* - * Altera Nios2 CPU PIC - * - * Copyright (c) 2016 Marek Vasut <marek.vasut@gmail.com> - * - * 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 2.1 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/lgpl-2.1.html> - */ - -#include "qemu/osdep.h" -#include "cpu.h" -#include "hw/irq.h" - -#include "qemu/config-file.h" - -#include "boot.h" - -void nios2_check_interrupts(CPUNios2State *env) -{ - if (env->irq_pending && - (env->regs[CR_STATUS] & CR_STATUS_PIE)) { - env->irq_pending = 0; - cpu_interrupt(env_cpu(env), CPU_INTERRUPT_HARD); - } -} diff --git a/target/nios2/op_helper.c b/target/nios2/op_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/nios2/op_helper.c +++ b/target/nios2/op_helper.c @@ -XXX,XX +XXX,XX @@ void helper_mmu_write(CPUNios2State *env, uint32_t rn, uint32_t v) mmu_write(env, rn, v); } +static void nios2_check_interrupts(CPUNios2State *env) +{ + if (env->irq_pending && + (env->regs[CR_STATUS] & CR_STATUS_PIE)) { + env->irq_pending = 0; + cpu_interrupt(env_cpu(env), CPU_INTERRUPT_HARD); + } +} + void helper_check_interrupts(CPUNios2State *env) { qemu_mutex_lock_iothread(); diff --git a/hw/nios2/meson.build b/hw/nios2/meson.build index XXXXXXX..XXXXXXX 100644 --- a/hw/nios2/meson.build +++ b/hw/nios2/meson.build @@ -XXX,XX +XXX,XX @@ nios2_ss = ss.source_set() -nios2_ss.add(files('boot.c', 'cpu_pic.c')) +nios2_ss.add(files('boot.c')) nios2_ss.add(when: 'CONFIG_NIOS2_10M50', if_true: files('10m50_devboard.c')) nios2_ss.add(when: 'CONFIG_NIOS2_GENERIC_NOMMU', if_true: files('generic_nommu.c')) -- 2.20.1 In nios2_cpu_set_irq(), use deposit32() rather than raw shift-and-mask operations to set the appropriate bit in the ipending register. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 20201129174022.26530-4-peter.maydell@linaro.org --- target/nios2/cpu.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/target/nios2/cpu.c b/target/nios2/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/target/nios2/cpu.c +++ b/target/nios2/cpu.c @@ -XXX,XX +XXX,XX @@ static void nios2_cpu_set_irq(void *opaque, int irq, int level) CPUNios2State *env = &cpu->env; CPUState *cs = CPU(cpu); - env->regs[CR_IPENDING] &= ~(1 << irq); - env->regs[CR_IPENDING] |= !!level << irq; + env->regs[CR_IPENDING] = deposit32(env->regs[CR_IPENDING], irq, 1, !!level); env->irq_pending = env->regs[CR_IPENDING] & env->regs[CR_IENABLE]; -- 2.20.1 In rom_check_and_register_reset() we detect overlaps by looking at whether the ROM blob we're currently examining is in the same address space and starts before the previous ROM blob ends. (This works because the ROM list is kept sorted in order by AddressSpace and then by address.) Instead of keeping the AddressSpace and last address of the previous ROM blob in local variables, just keep a pointer to it. This will allow us to print more useful information when we do detect an overlap. Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20201129203923.10622-2-peter.maydell@linaro.org --- hw/core/loader.c | 23 +++++++++++++++-------- 1 file changed, 15 insertions(+), 8 deletions(-) diff --git a/hw/core/loader.c b/hw/core/loader.c index XXXXXXX..XXXXXXX 100644 --- a/hw/core/loader.c +++ b/hw/core/loader.c @@ -XXX,XX +XXX,XX @@ static void rom_reset(void *unused) } } +/* Return true if two consecutive ROMs in the ROM list overlap */ +static bool roms_overlap(Rom *last_rom, Rom *this_rom) +{ + if (!last_rom) { + return false; + } + return last_rom->as == this_rom->as && + last_rom->addr + last_rom->romsize > this_rom->addr; +} + int rom_check_and_register_reset(void) { - hwaddr addr = 0; MemoryRegionSection section; - Rom *rom; - AddressSpace *as = NULL; + Rom *rom, *last_rom = NULL; QTAILQ_FOREACH(rom, &roms, next) { if (rom->fw_file) { continue; } if (!rom->mr) { - if ((addr > rom->addr) && (as == rom->as)) { + if (roms_overlap(last_rom, rom)) { fprintf(stderr, "rom: requested regions overlap " "(rom %s. free=0x" TARGET_FMT_plx ", addr=0x" TARGET_FMT_plx ")\n", - rom->name, addr, rom->addr); + rom->name, last_rom->addr + last_rom->romsize, + rom->addr); return -1; } - addr = rom->addr; - addr += rom->romsize; - as = rom->as; + last_rom = rom; } section = memory_region_find(rom->mr ? rom->mr : get_system_memory(), rom->addr, 1); -- 2.20.1 In rom_check_and_register_reset() we report to the user if there is a "ROM region overlap". This has a couple of problems: * the reported information is not very easy to intepret * the function just prints the overlap to stderr (and relies on its single callsite in vl.c to do an error_report() and exit) * only the first overlap encountered is diagnosed Make this function use error_report() and error_printf() and report a more user-friendly report with all the overlaps diagnosed. Sample old output: rom: requested regions overlap (rom dtb. free=0x0000000000008000, addr=0x0000000000000000) qemu-system-aarch64: rom check and register reset failed Sample new output: qemu-system-aarch64: Some ROM regions are overlapping These ROM regions might have been loaded by direct user request or by default. They could be BIOS/firmware images, a guest kernel, initrd or some other file loaded into guest memory. Check whether you intended to load all this guest code, and whether it has been built to load to the correct addresses. The following two regions overlap (in the cpu-memory-0 address space): phdr #0: /home/petmay01/linaro/qemu-misc-tests/ldmia-fault.axf (addresses 0x0000000000000000 - 0x0000000000008000) dtb (addresses 0x0000000000000000 - 0x0000000000100000) The following two regions overlap (in the cpu-memory-0 address space): phdr #1: /home/petmay01/linaro/qemu-misc-tests/bad-psci-call.axf (addresses 0x0000000040000000 - 0x0000000040000010) phdr #0: /home/petmay01/linaro/qemu-misc-tests/bp-test.elf (addresses 0x0000000040000000 - 0x0000000040000020) Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20201129203923.10622-3-peter.maydell@linaro.org --- hw/core/loader.c | 48 ++++++++++++++++++++++++++++++++++++++++++------ softmmu/vl.c | 1 - 2 files changed, 42 insertions(+), 7 deletions(-) diff --git a/hw/core/loader.c b/hw/core/loader.c index XXXXXXX..XXXXXXX 100644 --- a/hw/core/loader.c +++ b/hw/core/loader.c @@ -XXX,XX +XXX,XX @@ static bool roms_overlap(Rom *last_rom, Rom *this_rom) last_rom->addr + last_rom->romsize > this_rom->addr; } +static const char *rom_as_name(Rom *rom) +{ + const char *name = rom->as ? rom->as->name : NULL; + return name ?: "anonymous"; +} + +static void rom_print_overlap_error_header(void) +{ + error_report("Some ROM regions are overlapping"); + error_printf( + "These ROM regions might have been loaded by " + "direct user request or by default.\n" + "They could be BIOS/firmware images, a guest kernel, " + "initrd or some other file loaded into guest memory.\n" + "Check whether you intended to load all this guest code, and " + "whether it has been built to load to the correct addresses.\n"); +} + +static void rom_print_one_overlap_error(Rom *last_rom, Rom *rom) +{ + error_printf( + "\nThe following two regions overlap (in the %s address space):\n", + rom_as_name(rom)); + error_printf( + " %s (addresses 0x" TARGET_FMT_plx " - 0x" TARGET_FMT_plx ")\n", + last_rom->name, last_rom->addr, last_rom->addr + last_rom->romsize); + error_printf( + " %s (addresses 0x" TARGET_FMT_plx " - 0x" TARGET_FMT_plx ")\n", + rom->name, rom->addr, rom->addr + rom->romsize); +} + int rom_check_and_register_reset(void) { MemoryRegionSection section; Rom *rom, *last_rom = NULL; + bool found_overlap = false; QTAILQ_FOREACH(rom, &roms, next) { if (rom->fw_file) { @@ -XXX,XX +XXX,XX @@ int rom_check_and_register_reset(void) } if (!rom->mr) { if (roms_overlap(last_rom, rom)) { - fprintf(stderr, "rom: requested regions overlap " - "(rom %s. free=0x" TARGET_FMT_plx - ", addr=0x" TARGET_FMT_plx ")\n", - rom->name, last_rom->addr + last_rom->romsize, - rom->addr); - return -1; + if (!found_overlap) { + found_overlap = true; + rom_print_overlap_error_header(); + } + rom_print_one_overlap_error(last_rom, rom); + /* Keep going through the list so we report all overlaps */ } last_rom = rom; } @@ -XXX,XX +XXX,XX @@ int rom_check_and_register_reset(void) rom->isrom = int128_nz(section.size) && memory_region_is_rom(section.mr); memory_region_unref(section.mr); } + if (found_overlap) { + return -1; + } + qemu_register_reset(rom_reset, NULL); roms_loaded = 1; return 0; diff --git a/softmmu/vl.c b/softmmu/vl.c index XXXXXXX..XXXXXXX 100644 --- a/softmmu/vl.c +++ b/softmmu/vl.c @@ -XXX,XX +XXX,XX @@ static void qemu_machine_creation_done(void) qemu_run_machine_init_done_notifiers(); if (rom_check_and_register_reset() != 0) { - error_report("rom check and register reset failed"); exit(1); } -- 2.20.1 Currently the load_elf code assembles the ROM blob name into a local 128 byte fixed-size array. Use g_strdup_printf() instead so that we don't truncate the pathname if it happens to be long. (This matters mostly for monitor 'info roms' output and for the error messages if ROM blobs overlap.) Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20201129203923.10622-4-peter.maydell@linaro.org --- include/hw/elf_ops.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/hw/elf_ops.h b/include/hw/elf_ops.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/elf_ops.h +++ b/include/hw/elf_ops.h @@ -XXX,XX +XXX,XX @@ static int glue(load_elf, SZ)(const char *name, int fd, uint64_t addr, low = (uint64_t)-1, high = 0; GMappedFile *mapped_file = NULL; uint8_t *data = NULL; - char label[128]; int ret = ELF_LOAD_FAILED; if (read(fd, &ehdr, sizeof(ehdr)) != sizeof(ehdr)) @@ -XXX,XX +XXX,XX @@ static int glue(load_elf, SZ)(const char *name, int fd, */ if (mem_size != 0) { if (load_rom) { - snprintf(label, sizeof(label), "phdr #%d: %s", i, name); + g_autofree char *label = + g_strdup_printf("phdr #%d: %s", i, name); /* * rom_add_elf_program() takes its own reference to -- 2.20.1 Instead of making the ROM blob name something like: phdr #0: /home/petmay01/linaro/qemu-misc-tests/ldmia-fault.axf make it a little more self-explanatory for people who don't know ELF format details: /home/petmay01/linaro/qemu-misc-tests/ldmia-fault.axf ELF program header segment 0 Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20201129203923.10622-5-peter.maydell@linaro.org --- include/hw/elf_ops.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/include/hw/elf_ops.h b/include/hw/elf_ops.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/elf_ops.h +++ b/include/hw/elf_ops.h @@ -XXX,XX +XXX,XX @@ static int glue(load_elf, SZ)(const char *name, int fd, if (mem_size != 0) { if (load_rom) { g_autofree char *label = - g_strdup_printf("phdr #%d: %s", i, name); + g_strdup_printf("%s ELF program header segment %d", + name, i); /* * rom_add_elf_program() takes its own reference to -- 2.20.1 From: Sai Pavan Boddu <sai.pavan.boddu@xilinx.com> This module emulates control registers of versal usb2 controller, this is added just to make guest happy. In general this module would control the phy-reset signal from usb controller, data coherency of the transactions, signals the host system errors received from controller. Signed-off-by: Sai Pavan Boddu <sai.pavan.boddu@xilinx.com> Signed-off-by: Vikram Garhwal <fnu.vikram@xilinx.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 1607023357-5096-2-git-send-email-sai.pavan.boddu@xilinx.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/usb/xlnx-versal-usb2-ctrl-regs.h | 45 ++++ hw/usb/xlnx-versal-usb2-ctrl-regs.c | 229 ++++++++++++++++++++ hw/usb/meson.build | 1 + 3 files changed, 275 insertions(+) create mode 100644 include/hw/usb/xlnx-versal-usb2-ctrl-regs.h create mode 100644 hw/usb/xlnx-versal-usb2-ctrl-regs.c diff --git a/include/hw/usb/xlnx-versal-usb2-ctrl-regs.h b/include/hw/usb/xlnx-versal-usb2-ctrl-regs.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/usb/xlnx-versal-usb2-ctrl-regs.h @@ -XXX,XX +XXX,XX @@ +/* + * QEMU model of the VersalUsb2CtrlRegs Register control/Status block for + * USB2.0 controller + * + * Copyright (c) 2020 Xilinx Inc. Vikram Garhwal <fnu.vikram@xilinx.com> + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ + +#ifndef _XLNX_USB2_REGS_H_ +#define _XLNX_USB2_REGS_H_ + +#define TYPE_XILINX_VERSAL_USB2_CTRL_REGS "xlnx.versal-usb2-ctrl-regs" + +#define XILINX_VERSAL_USB2_CTRL_REGS(obj) \ + OBJECT_CHECK(VersalUsb2CtrlRegs, (obj), TYPE_XILINX_VERSAL_USB2_CTRL_REGS) + +#define USB2_REGS_R_MAX ((0x78 / 4) + 1) + +typedef struct VersalUsb2CtrlRegs { + SysBusDevice parent_obj; + MemoryRegion iomem; + qemu_irq irq_ir; + + uint32_t regs[USB2_REGS_R_MAX]; + RegisterInfo regs_info[USB2_REGS_R_MAX]; +} VersalUsb2CtrlRegs; + +#endif diff --git a/hw/usb/xlnx-versal-usb2-ctrl-regs.c b/hw/usb/xlnx-versal-usb2-ctrl-regs.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/usb/xlnx-versal-usb2-ctrl-regs.c @@ -XXX,XX +XXX,XX @@ +/* + * QEMU model of the VersalUsb2CtrlRegs Register control/Status block for + * USB2.0 controller + * + * This module should control phy_reset, permanent device plugs, frame length + * time adjust & setting of coherency paths. None of which are emulated in + * present model. + * + * Copyright (c) 2020 Xilinx Inc. Vikram Garhwal <fnu.vikram@xilinx.com> + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ + +#include "qemu/osdep.h" +#include "hw/sysbus.h" +#include "hw/irq.h" +#include "hw/register.h" +#include "qemu/bitops.h" +#include "qemu/log.h" +#include "qom/object.h" +#include "migration/vmstate.h" +#include "hw/usb/xlnx-versal-usb2-ctrl-regs.h" + +#ifndef XILINX_VERSAL_USB2_CTRL_REGS_ERR_DEBUG +#define XILINX_VERSAL_USB2_CTRL_REGS_ERR_DEBUG 0 +#endif + +REG32(BUS_FILTER, 0x30) + FIELD(BUS_FILTER, BYPASS, 0, 4) +REG32(PORT, 0x34) + FIELD(PORT, HOST_SMI_BAR_WR, 4, 1) + FIELD(PORT, HOST_SMI_PCI_CMD_REG_WR, 3, 1) + FIELD(PORT, HOST_MSI_ENABLE, 2, 1) + FIELD(PORT, PWR_CTRL_PRSNT, 1, 1) + FIELD(PORT, HUB_PERM_ATTACH, 0, 1) +REG32(JITTER_ADJUST, 0x38) + FIELD(JITTER_ADJUST, FLADJ, 0, 6) +REG32(BIGENDIAN, 0x40) + FIELD(BIGENDIAN, ENDIAN_GS, 0, 1) +REG32(COHERENCY, 0x44) + FIELD(COHERENCY, USB_COHERENCY, 0, 1) +REG32(XHC_BME, 0x48) + FIELD(XHC_BME, XHC_BME, 0, 1) +REG32(REG_CTRL, 0x60) + FIELD(REG_CTRL, SLVERR_ENABLE, 0, 1) +REG32(IR_STATUS, 0x64) + FIELD(IR_STATUS, HOST_SYS_ERR, 1, 1) + FIELD(IR_STATUS, ADDR_DEC_ERR, 0, 1) +REG32(IR_MASK, 0x68) + FIELD(IR_MASK, HOST_SYS_ERR, 1, 1) + FIELD(IR_MASK, ADDR_DEC_ERR, 0, 1) +REG32(IR_ENABLE, 0x6c) + FIELD(IR_ENABLE, HOST_SYS_ERR, 1, 1) + FIELD(IR_ENABLE, ADDR_DEC_ERR, 0, 1) +REG32(IR_DISABLE, 0x70) + FIELD(IR_DISABLE, HOST_SYS_ERR, 1, 1) + FIELD(IR_DISABLE, ADDR_DEC_ERR, 0, 1) +REG32(USB3, 0x78) + +static void ir_update_irq(VersalUsb2CtrlRegs *s) +{ + bool pending = s->regs[R_IR_STATUS] & ~s->regs[R_IR_MASK]; + qemu_set_irq(s->irq_ir, pending); +} + +static void ir_status_postw(RegisterInfo *reg, uint64_t val64) +{ + VersalUsb2CtrlRegs *s = XILINX_VERSAL_USB2_CTRL_REGS(reg->opaque); + /* + * TODO: This should also clear USBSTS.HSE field in USB XHCI register. + * May be combine both the modules. + */ + ir_update_irq(s); +} + +static uint64_t ir_enable_prew(RegisterInfo *reg, uint64_t val64) +{ + VersalUsb2CtrlRegs *s = XILINX_VERSAL_USB2_CTRL_REGS(reg->opaque); + uint32_t val = val64; + + s->regs[R_IR_MASK] &= ~val; + ir_update_irq(s); + return 0; +} + +static uint64_t ir_disable_prew(RegisterInfo *reg, uint64_t val64) +{ + VersalUsb2CtrlRegs *s = XILINX_VERSAL_USB2_CTRL_REGS(reg->opaque); + uint32_t val = val64; + + s->regs[R_IR_MASK] |= val; + ir_update_irq(s); + return 0; +} + +static const RegisterAccessInfo usb2_ctrl_regs_regs_info[] = { + { .name = "BUS_FILTER", .addr = A_BUS_FILTER, + .rsvd = 0xfffffff0, + },{ .name = "PORT", .addr = A_PORT, + .rsvd = 0xffffffe0, + },{ .name = "JITTER_ADJUST", .addr = A_JITTER_ADJUST, + .reset = 0x20, + .rsvd = 0xffffffc0, + },{ .name = "BIGENDIAN", .addr = A_BIGENDIAN, + .rsvd = 0xfffffffe, + },{ .name = "COHERENCY", .addr = A_COHERENCY, + .rsvd = 0xfffffffe, + },{ .name = "XHC_BME", .addr = A_XHC_BME, + .reset = 0x1, + .rsvd = 0xfffffffe, + },{ .name = "REG_CTRL", .addr = A_REG_CTRL, + .rsvd = 0xfffffffe, + },{ .name = "IR_STATUS", .addr = A_IR_STATUS, + .rsvd = 0xfffffffc, + .w1c = 0x3, + .post_write = ir_status_postw, + },{ .name = "IR_MASK", .addr = A_IR_MASK, + .reset = 0x3, + .rsvd = 0xfffffffc, + .ro = 0x3, + },{ .name = "IR_ENABLE", .addr = A_IR_ENABLE, + .rsvd = 0xfffffffc, + .pre_write = ir_enable_prew, + },{ .name = "IR_DISABLE", .addr = A_IR_DISABLE, + .rsvd = 0xfffffffc, + .pre_write = ir_disable_prew, + },{ .name = "USB3", .addr = A_USB3, + } +}; + +static void usb2_ctrl_regs_reset_init(Object *obj, ResetType type) +{ + VersalUsb2CtrlRegs *s = XILINX_VERSAL_USB2_CTRL_REGS(obj); + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(s->regs_info); ++i) { + register_reset(&s->regs_info[i]); + } +} + +static void usb2_ctrl_regs_reset_hold(Object *obj) +{ + VersalUsb2CtrlRegs *s = XILINX_VERSAL_USB2_CTRL_REGS(obj); + + ir_update_irq(s); +} + +static const MemoryRegionOps usb2_ctrl_regs_ops = { + .read = register_read_memory, + .write = register_write_memory, + .endianness = DEVICE_LITTLE_ENDIAN, + .valid = { + .min_access_size = 4, + .max_access_size = 4, + }, +}; + +static void usb2_ctrl_regs_init(Object *obj) +{ + VersalUsb2CtrlRegs *s = XILINX_VERSAL_USB2_CTRL_REGS(obj); + SysBusDevice *sbd = SYS_BUS_DEVICE(obj); + RegisterInfoArray *reg_array; + + memory_region_init(&s->iomem, obj, TYPE_XILINX_VERSAL_USB2_CTRL_REGS, + USB2_REGS_R_MAX * 4); + reg_array = + register_init_block32(DEVICE(obj), usb2_ctrl_regs_regs_info, + ARRAY_SIZE(usb2_ctrl_regs_regs_info), + s->regs_info, s->regs, + &usb2_ctrl_regs_ops, + XILINX_VERSAL_USB2_CTRL_REGS_ERR_DEBUG, + USB2_REGS_R_MAX * 4); + memory_region_add_subregion(&s->iomem, + 0x0, + &reg_array->mem); + sysbus_init_mmio(sbd, &s->iomem); + sysbus_init_irq(sbd, &s->irq_ir); +} + +static const VMStateDescription vmstate_usb2_ctrl_regs = { + .name = TYPE_XILINX_VERSAL_USB2_CTRL_REGS, + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32_ARRAY(regs, VersalUsb2CtrlRegs, USB2_REGS_R_MAX), + VMSTATE_END_OF_LIST(), + } +}; + +static void usb2_ctrl_regs_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + ResettableClass *rc = RESETTABLE_CLASS(klass); + + rc->phases.enter = usb2_ctrl_regs_reset_init; + rc->phases.hold = usb2_ctrl_regs_reset_hold; + dc->vmsd = &vmstate_usb2_ctrl_regs; +} + +static const TypeInfo usb2_ctrl_regs_info = { + .name = TYPE_XILINX_VERSAL_USB2_CTRL_REGS, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_size = sizeof(VersalUsb2CtrlRegs), + .class_init = usb2_ctrl_regs_class_init, + .instance_init = usb2_ctrl_regs_init, +}; + +static void usb2_ctrl_regs_register_types(void) +{ + type_register_static(&usb2_ctrl_regs_info); +} + +type_init(usb2_ctrl_regs_register_types) diff --git a/hw/usb/meson.build b/hw/usb/meson.build index XXXXXXX..XXXXXXX 100644 --- a/hw/usb/meson.build +++ b/hw/usb/meson.build @@ -XXX,XX +XXX,XX @@ softmmu_ss.add(when: 'CONFIG_USB_DWC2', if_true: files('hcd-dwc2.c')) softmmu_ss.add(when: 'CONFIG_TUSB6010', if_true: files('tusb6010.c')) softmmu_ss.add(when: 'CONFIG_IMX', if_true: files('chipidea.c')) softmmu_ss.add(when: 'CONFIG_IMX_USBPHY', if_true: files('imx-usb-phy.c')) +specific_ss.add(when: 'CONFIG_XLNX_VERSAL', if_true: files('xlnx-versal-usb2-ctrl-regs.c')) # emulated usb devices softmmu_ss.add(when: 'CONFIG_USB', if_true: files('dev-hub.c')) -- 2.20.1 From: Vikram Garhwal <fnu.vikram@xilinx.com> This patch adds skeleton model of dwc3 usb controller attached to xhci-sysbus device. It defines global register space of DWC3 controller, global registers control the AXI/AHB interfaces properties, external FIFO support and event count support. All of which are unimplemented at present,we are only supporting core reset and read of ID register. Signed-off-by: Vikram Garhwal <fnu.vikram@xilinx.com> Signed-off-by: Sai Pavan Boddu <sai.pavan.boddu@xilinx.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 1607023357-5096-3-git-send-email-sai.pavan.boddu@xilinx.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/usb/hcd-dwc3.h | 55 +++ hw/usb/hcd-dwc3.c | 689 ++++++++++++++++++++++++++++++++++++++ hw/usb/Kconfig | 5 + hw/usb/meson.build | 1 + 4 files changed, 750 insertions(+) create mode 100644 include/hw/usb/hcd-dwc3.h create mode 100644 hw/usb/hcd-dwc3.c diff --git a/include/hw/usb/hcd-dwc3.h b/include/hw/usb/hcd-dwc3.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/usb/hcd-dwc3.h @@ -XXX,XX +XXX,XX @@ +/* + * QEMU model of the USB DWC3 host controller emulation. + * + * Copyright (c) 2020 Xilinx Inc. + * + * Written by Vikram Garhwal<fnu.vikram@xilinx.com> + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ +#ifndef HCD_DWC3_H +#define HCD_DWC3_H + +#include "hw/usb/hcd-xhci.h" +#include "hw/usb/hcd-xhci-sysbus.h" + +#define TYPE_USB_DWC3 "usb_dwc3" + +#define USB_DWC3(obj) \ + OBJECT_CHECK(USBDWC3, (obj), TYPE_USB_DWC3) + +#define USB_DWC3_R_MAX ((0x530 / 4) + 1) +#define DWC3_SIZE 0x10000 + +typedef struct USBDWC3 { + SysBusDevice parent_obj; + MemoryRegion iomem; + XHCISysbusState sysbus_xhci; + + uint32_t regs[USB_DWC3_R_MAX]; + RegisterInfo regs_info[USB_DWC3_R_MAX]; + + struct { + uint8_t mode; + uint32_t dwc_usb3_user; + } cfg; + +} USBDWC3; + +#endif diff --git a/hw/usb/hcd-dwc3.c b/hw/usb/hcd-dwc3.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/usb/hcd-dwc3.c @@ -XXX,XX +XXX,XX @@ +/* + * QEMU model of the USB DWC3 host controller emulation. + * + * This model defines global register space of DWC3 controller. Global + * registers control the AXI/AHB interfaces properties, external FIFO support + * and event count support. All of which are unimplemented at present. We are + * only supporting core reset and read of ID register. + * + * Copyright (c) 2020 Xilinx Inc. Vikram Garhwal<fnu.vikram@xilinx.com> + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ + +#include "qemu/osdep.h" +#include "hw/sysbus.h" +#include "hw/register.h" +#include "qemu/bitops.h" +#include "qemu/log.h" +#include "qom/object.h" +#include "migration/vmstate.h" +#include "hw/qdev-properties.h" +#include "hw/usb/hcd-dwc3.h" +#include "qapi/error.h" + +#ifndef USB_DWC3_ERR_DEBUG +#define USB_DWC3_ERR_DEBUG 0 +#endif + +#define HOST_MODE 1 +#define FIFO_LEN 0x1000 + +REG32(GSBUSCFG0, 0x00) + FIELD(GSBUSCFG0, DATRDREQINFO, 28, 4) + FIELD(GSBUSCFG0, DESRDREQINFO, 24, 4) + FIELD(GSBUSCFG0, DATWRREQINFO, 20, 4) + FIELD(GSBUSCFG0, DESWRREQINFO, 16, 4) + FIELD(GSBUSCFG0, RESERVED_15_12, 12, 4) + FIELD(GSBUSCFG0, DATBIGEND, 11, 1) + FIELD(GSBUSCFG0, DESBIGEND, 10, 1) + FIELD(GSBUSCFG0, RESERVED_9_8, 8, 2) + FIELD(GSBUSCFG0, INCR256BRSTENA, 7, 1) + FIELD(GSBUSCFG0, INCR128BRSTENA, 6, 1) + FIELD(GSBUSCFG0, INCR64BRSTENA, 5, 1) + FIELD(GSBUSCFG0, INCR32BRSTENA, 4, 1) + FIELD(GSBUSCFG0, INCR16BRSTENA, 3, 1) + FIELD(GSBUSCFG0, INCR8BRSTENA, 2, 1) + FIELD(GSBUSCFG0, INCR4BRSTENA, 1, 1) + FIELD(GSBUSCFG0, INCRBRSTENA, 0, 1) +REG32(GSBUSCFG1, 0x04) + FIELD(GSBUSCFG1, RESERVED_31_13, 13, 19) + FIELD(GSBUSCFG1, EN1KPAGE, 12, 1) + FIELD(GSBUSCFG1, PIPETRANSLIMIT, 8, 4) + FIELD(GSBUSCFG1, RESERVED_7_0, 0, 8) +REG32(GTXTHRCFG, 0x08) + FIELD(GTXTHRCFG, RESERVED_31, 31, 1) + FIELD(GTXTHRCFG, RESERVED_30, 30, 1) + FIELD(GTXTHRCFG, USBTXPKTCNTSEL, 29, 1) + FIELD(GTXTHRCFG, RESERVED_28, 28, 1) + FIELD(GTXTHRCFG, USBTXPKTCNT, 24, 4) + FIELD(GTXTHRCFG, USBMAXTXBURSTSIZE, 16, 8) + FIELD(GTXTHRCFG, RESERVED_15, 15, 1) + FIELD(GTXTHRCFG, RESERVED_14, 14, 1) + FIELD(GTXTHRCFG, RESERVED_13_11, 11, 3) + FIELD(GTXTHRCFG, RESERVED_10_0, 0, 11) +REG32(GRXTHRCFG, 0x0c) + FIELD(GRXTHRCFG, RESERVED_31_30, 30, 2) + FIELD(GRXTHRCFG, USBRXPKTCNTSEL, 29, 1) + FIELD(GRXTHRCFG, RESERVED_28, 28, 1) + FIELD(GRXTHRCFG, USBRXPKTCNT, 24, 4) + FIELD(GRXTHRCFG, USBMAXRXBURSTSIZE, 19, 5) + FIELD(GRXTHRCFG, RESERVED_18_16, 16, 3) + FIELD(GRXTHRCFG, RESERVED_15, 15, 1) + FIELD(GRXTHRCFG, RESERVED_14_13, 13, 2) + FIELD(GRXTHRCFG, RESVISOCOUTSPC, 0, 13) +REG32(GCTL, 0x10) + FIELD(GCTL, PWRDNSCALE, 19, 13) + FIELD(GCTL, MASTERFILTBYPASS, 18, 1) + FIELD(GCTL, BYPSSETADDR, 17, 1) + FIELD(GCTL, U2RSTECN, 16, 1) + FIELD(GCTL, FRMSCLDWN, 14, 2) + FIELD(GCTL, PRTCAPDIR, 12, 2) + FIELD(GCTL, CORESOFTRESET, 11, 1) + FIELD(GCTL, U1U2TIMERSCALE, 9, 1) + FIELD(GCTL, DEBUGATTACH, 8, 1) + FIELD(GCTL, RAMCLKSEL, 6, 2) + FIELD(GCTL, SCALEDOWN, 4, 2) + FIELD(GCTL, DISSCRAMBLE, 3, 1) + FIELD(GCTL, U2EXIT_LFPS, 2, 1) + FIELD(GCTL, GBLHIBERNATIONEN, 1, 1) + FIELD(GCTL, DSBLCLKGTNG, 0, 1) +REG32(GPMSTS, 0x14) +REG32(GSTS, 0x18) + FIELD(GSTS, CBELT, 20, 12) + FIELD(GSTS, RESERVED_19_12, 12, 8) + FIELD(GSTS, SSIC_IP, 11, 1) + FIELD(GSTS, OTG_IP, 10, 1) + FIELD(GSTS, BC_IP, 9, 1) + FIELD(GSTS, ADP_IP, 8, 1) + FIELD(GSTS, HOST_IP, 7, 1) + FIELD(GSTS, DEVICE_IP, 6, 1) + FIELD(GSTS, CSRTIMEOUT, 5, 1) + FIELD(GSTS, BUSERRADDRVLD, 4, 1) + FIELD(GSTS, RESERVED_3_2, 2, 2) + FIELD(GSTS, CURMOD, 0, 2) +REG32(GUCTL1, 0x1c) + FIELD(GUCTL1, RESUME_OPMODE_HS_HOST, 10, 1) +REG32(GSNPSID, 0x20) +REG32(GGPIO, 0x24) + FIELD(GGPIO, GPO, 16, 16) + FIELD(GGPIO, GPI, 0, 16) +REG32(GUID, 0x28) +REG32(GUCTL, 0x2c) + FIELD(GUCTL, REFCLKPER, 22, 10) + FIELD(GUCTL, NOEXTRDL, 21, 1) + FIELD(GUCTL, RESERVED_20_18, 18, 3) + FIELD(GUCTL, SPRSCTRLTRANSEN, 17, 1) + FIELD(GUCTL, RESBWHSEPS, 16, 1) + FIELD(GUCTL, RESERVED_15, 15, 1) + FIELD(GUCTL, USBHSTINAUTORETRYEN, 14, 1) + FIELD(GUCTL, ENOVERLAPCHK, 13, 1) + FIELD(GUCTL, EXTCAPSUPPTEN, 12, 1) + FIELD(GUCTL, INSRTEXTRFSBODI, 11, 1) + FIELD(GUCTL, DTCT, 9, 2) + FIELD(GUCTL, DTFT, 0, 9) +REG32(GBUSERRADDRLO, 0x30) +REG32(GBUSERRADDRHI, 0x34) +REG32(GHWPARAMS0, 0x40) + FIELD(GHWPARAMS0, GHWPARAMS0_31_24, 24, 8) + FIELD(GHWPARAMS0, GHWPARAMS0_23_16, 16, 8) + FIELD(GHWPARAMS0, GHWPARAMS0_15_8, 8, 8) + FIELD(GHWPARAMS0, GHWPARAMS0_7_6, 6, 2) + FIELD(GHWPARAMS0, GHWPARAMS0_5_3, 3, 3) + FIELD(GHWPARAMS0, GHWPARAMS0_2_0, 0, 3) +REG32(GHWPARAMS1, 0x44) + FIELD(GHWPARAMS1, GHWPARAMS1_31, 31, 1) + FIELD(GHWPARAMS1, GHWPARAMS1_30, 30, 1) + FIELD(GHWPARAMS1, GHWPARAMS1_29, 29, 1) + FIELD(GHWPARAMS1, GHWPARAMS1_28, 28, 1) + FIELD(GHWPARAMS1, GHWPARAMS1_27, 27, 1) + FIELD(GHWPARAMS1, GHWPARAMS1_26, 26, 1) + FIELD(GHWPARAMS1, GHWPARAMS1_25_24, 24, 2) + FIELD(GHWPARAMS1, GHWPARAMS1_23, 23, 1) + FIELD(GHWPARAMS1, GHWPARAMS1_22_21, 21, 2) + FIELD(GHWPARAMS1, GHWPARAMS1_20_15, 15, 6) + FIELD(GHWPARAMS1, GHWPARAMS1_14_12, 12, 3) + FIELD(GHWPARAMS1, GHWPARAMS1_11_9, 9, 3) + FIELD(GHWPARAMS1, GHWPARAMS1_8_6, 6, 3) + FIELD(GHWPARAMS1, GHWPARAMS1_5_3, 3, 3) + FIELD(GHWPARAMS1, GHWPARAMS1_2_0, 0, 3) +REG32(GHWPARAMS2, 0x48) +REG32(GHWPARAMS3, 0x4c) + FIELD(GHWPARAMS3, GHWPARAMS3_31, 31, 1) + FIELD(GHWPARAMS3, GHWPARAMS3_30_23, 23, 8) + FIELD(GHWPARAMS3, GHWPARAMS3_22_18, 18, 5) + FIELD(GHWPARAMS3, GHWPARAMS3_17_12, 12, 6) + FIELD(GHWPARAMS3, GHWPARAMS3_11, 11, 1) + FIELD(GHWPARAMS3, GHWPARAMS3_10, 10, 1) + FIELD(GHWPARAMS3, GHWPARAMS3_9_8, 8, 2) + FIELD(GHWPARAMS3, GHWPARAMS3_7_6, 6, 2) + FIELD(GHWPARAMS3, GHWPARAMS3_5_4, 4, 2) + FIELD(GHWPARAMS3, GHWPARAMS3_3_2, 2, 2) + FIELD(GHWPARAMS3, GHWPARAMS3_1_0, 0, 2) +REG32(GHWPARAMS4, 0x50) + FIELD(GHWPARAMS4, GHWPARAMS4_31_28, 28, 4) + FIELD(GHWPARAMS4, GHWPARAMS4_27_24, 24, 4) + FIELD(GHWPARAMS4, GHWPARAMS4_23, 23, 1) + FIELD(GHWPARAMS4, GHWPARAMS4_22, 22, 1) + FIELD(GHWPARAMS4, GHWPARAMS4_21, 21, 1) + FIELD(GHWPARAMS4, GHWPARAMS4_20_17, 17, 4) + FIELD(GHWPARAMS4, GHWPARAMS4_16_13, 13, 4) + FIELD(GHWPARAMS4, GHWPARAMS4_12, 12, 1) + FIELD(GHWPARAMS4, GHWPARAMS4_11, 11, 1) + FIELD(GHWPARAMS4, GHWPARAMS4_10_9, 9, 2) + FIELD(GHWPARAMS4, GHWPARAMS4_8_7, 7, 2) + FIELD(GHWPARAMS4, GHWPARAMS4_6, 6, 1) + FIELD(GHWPARAMS4, GHWPARAMS4_5_0, 0, 6) +REG32(GHWPARAMS5, 0x54) + FIELD(GHWPARAMS5, GHWPARAMS5_31_28, 28, 4) + FIELD(GHWPARAMS5, GHWPARAMS5_27_22, 22, 6) + FIELD(GHWPARAMS5, GHWPARAMS5_21_16, 16, 6) + FIELD(GHWPARAMS5, GHWPARAMS5_15_10, 10, 6) + FIELD(GHWPARAMS5, GHWPARAMS5_9_4, 4, 6) + FIELD(GHWPARAMS5, GHWPARAMS5_3_0, 0, 4) +REG32(GHWPARAMS6, 0x58) + FIELD(GHWPARAMS6, GHWPARAMS6_31_16, 16, 16) + FIELD(GHWPARAMS6, BUSFLTRSSUPPORT, 15, 1) + FIELD(GHWPARAMS6, BCSUPPORT, 14, 1) + FIELD(GHWPARAMS6, OTG_SS_SUPPORT, 13, 1) + FIELD(GHWPARAMS6, ADPSUPPORT, 12, 1) + FIELD(GHWPARAMS6, HNPSUPPORT, 11, 1) + FIELD(GHWPARAMS6, SRPSUPPORT, 10, 1) + FIELD(GHWPARAMS6, GHWPARAMS6_9_8, 8, 2) + FIELD(GHWPARAMS6, GHWPARAMS6_7, 7, 1) + FIELD(GHWPARAMS6, GHWPARAMS6_6, 6, 1) + FIELD(GHWPARAMS6, GHWPARAMS6_5_0, 0, 6) +REG32(GHWPARAMS7, 0x5c) + FIELD(GHWPARAMS7, GHWPARAMS7_31_16, 16, 16) + FIELD(GHWPARAMS7, GHWPARAMS7_15_0, 0, 16) +REG32(GDBGFIFOSPACE, 0x60) + FIELD(GDBGFIFOSPACE, SPACE_AVAILABLE, 16, 16) + FIELD(GDBGFIFOSPACE, RESERVED_15_9, 9, 7) + FIELD(GDBGFIFOSPACE, FIFO_QUEUE_SELECT, 0, 9) +REG32(GUCTL2, 0x9c) + FIELD(GUCTL2, RESERVED_31_26, 26, 6) + FIELD(GUCTL2, EN_HP_PM_TIMER, 19, 7) + FIELD(GUCTL2, NOLOWPWRDUR, 15, 4) + FIELD(GUCTL2, RST_ACTBITLATER, 14, 1) + FIELD(GUCTL2, RESERVED_13, 13, 1) + FIELD(GUCTL2, DISABLECFC, 11, 1) +REG32(GUSB2PHYCFG, 0x100) + FIELD(GUSB2PHYCFG, U2_FREECLK_EXISTS, 30, 1) + FIELD(GUSB2PHYCFG, ULPI_LPM_WITH_OPMODE_CHK, 29, 1) + FIELD(GUSB2PHYCFG, RESERVED_25, 25, 1) + FIELD(GUSB2PHYCFG, LSTRD, 22, 3) + FIELD(GUSB2PHYCFG, LSIPD, 19, 3) + FIELD(GUSB2PHYCFG, ULPIEXTVBUSINDIACTOR, 18, 1) + FIELD(GUSB2PHYCFG, ULPIEXTVBUSDRV, 17, 1) + FIELD(GUSB2PHYCFG, RESERVED_16, 16, 1) + FIELD(GUSB2PHYCFG, ULPIAUTORES, 15, 1) + FIELD(GUSB2PHYCFG, RESERVED_14, 14, 1) + FIELD(GUSB2PHYCFG, USBTRDTIM, 10, 4) + FIELD(GUSB2PHYCFG, XCVRDLY, 9, 1) + FIELD(GUSB2PHYCFG, ENBLSLPM, 8, 1) + FIELD(GUSB2PHYCFG, PHYSEL, 7, 1) + FIELD(GUSB2PHYCFG, SUSPENDUSB20, 6, 1) + FIELD(GUSB2PHYCFG, FSINTF, 5, 1) + FIELD(GUSB2PHYCFG, ULPI_UTMI_SEL, 4, 1) + FIELD(GUSB2PHYCFG, PHYIF, 3, 1) + FIELD(GUSB2PHYCFG, TOUTCAL, 0, 3) +REG32(GUSB2I2CCTL, 0x140) +REG32(GUSB2PHYACC_ULPI, 0x180) + FIELD(GUSB2PHYACC_ULPI, RESERVED_31_27, 27, 5) + FIELD(GUSB2PHYACC_ULPI, DISUIPIDRVR, 26, 1) + FIELD(GUSB2PHYACC_ULPI, NEWREGREQ, 25, 1) + FIELD(GUSB2PHYACC_ULPI, VSTSDONE, 24, 1) + FIELD(GUSB2PHYACC_ULPI, VSTSBSY, 23, 1) + FIELD(GUSB2PHYACC_ULPI, REGWR, 22, 1) + FIELD(GUSB2PHYACC_ULPI, REGADDR, 16, 6) + FIELD(GUSB2PHYACC_ULPI, EXTREGADDR, 8, 8) + FIELD(GUSB2PHYACC_ULPI, REGDATA, 0, 8) +REG32(GTXFIFOSIZ0, 0x200) + FIELD(GTXFIFOSIZ0, TXFSTADDR_N, 16, 16) + FIELD(GTXFIFOSIZ0, TXFDEP_N, 0, 16) +REG32(GTXFIFOSIZ1, 0x204) + FIELD(GTXFIFOSIZ1, TXFSTADDR_N, 16, 16) + FIELD(GTXFIFOSIZ1, TXFDEP_N, 0, 16) +REG32(GTXFIFOSIZ2, 0x208) + FIELD(GTXFIFOSIZ2, TXFSTADDR_N, 16, 16) + FIELD(GTXFIFOSIZ2, TXFDEP_N, 0, 16) +REG32(GTXFIFOSIZ3, 0x20c) + FIELD(GTXFIFOSIZ3, TXFSTADDR_N, 16, 16) + FIELD(GTXFIFOSIZ3, TXFDEP_N, 0, 16) +REG32(GTXFIFOSIZ4, 0x210) + FIELD(GTXFIFOSIZ4, TXFSTADDR_N, 16, 16) + FIELD(GTXFIFOSIZ4, TXFDEP_N, 0, 16) +REG32(GTXFIFOSIZ5, 0x214) + FIELD(GTXFIFOSIZ5, TXFSTADDR_N, 16, 16) + FIELD(GTXFIFOSIZ5, TXFDEP_N, 0, 16) +REG32(GRXFIFOSIZ0, 0x280) + FIELD(GRXFIFOSIZ0, RXFSTADDR_N, 16, 16) + FIELD(GRXFIFOSIZ0, RXFDEP_N, 0, 16) +REG32(GRXFIFOSIZ1, 0x284) + FIELD(GRXFIFOSIZ1, RXFSTADDR_N, 16, 16) + FIELD(GRXFIFOSIZ1, RXFDEP_N, 0, 16) +REG32(GRXFIFOSIZ2, 0x288) + FIELD(GRXFIFOSIZ2, RXFSTADDR_N, 16, 16) + FIELD(GRXFIFOSIZ2, RXFDEP_N, 0, 16) +REG32(GEVNTADRLO_0, 0x300) +REG32(GEVNTADRHI_0, 0x304) +REG32(GEVNTSIZ_0, 0x308) + FIELD(GEVNTSIZ_0, EVNTINTRPTMASK, 31, 1) + FIELD(GEVNTSIZ_0, RESERVED_30_16, 16, 15) + FIELD(GEVNTSIZ_0, EVENTSIZ, 0, 16) +REG32(GEVNTCOUNT_0, 0x30c) + FIELD(GEVNTCOUNT_0, EVNT_HANDLER_BUSY, 31, 1) + FIELD(GEVNTCOUNT_0, RESERVED_30_16, 16, 15) + FIELD(GEVNTCOUNT_0, EVNTCOUNT, 0, 16) +REG32(GEVNTADRLO_1, 0x310) +REG32(GEVNTADRHI_1, 0x314) +REG32(GEVNTSIZ_1, 0x318) + FIELD(GEVNTSIZ_1, EVNTINTRPTMASK, 31, 1) + FIELD(GEVNTSIZ_1, RESERVED_30_16, 16, 15) + FIELD(GEVNTSIZ_1, EVENTSIZ, 0, 16) +REG32(GEVNTCOUNT_1, 0x31c) + FIELD(GEVNTCOUNT_1, EVNT_HANDLER_BUSY, 31, 1) + FIELD(GEVNTCOUNT_1, RESERVED_30_16, 16, 15) + FIELD(GEVNTCOUNT_1, EVNTCOUNT, 0, 16) +REG32(GEVNTADRLO_2, 0x320) +REG32(GEVNTADRHI_2, 0x324) +REG32(GEVNTSIZ_2, 0x328) + FIELD(GEVNTSIZ_2, EVNTINTRPTMASK, 31, 1) + FIELD(GEVNTSIZ_2, RESERVED_30_16, 16, 15) + FIELD(GEVNTSIZ_2, EVENTSIZ, 0, 16) +REG32(GEVNTCOUNT_2, 0x32c) + FIELD(GEVNTCOUNT_2, EVNT_HANDLER_BUSY, 31, 1) + FIELD(GEVNTCOUNT_2, RESERVED_30_16, 16, 15) + FIELD(GEVNTCOUNT_2, EVNTCOUNT, 0, 16) +REG32(GEVNTADRLO_3, 0x330) +REG32(GEVNTADRHI_3, 0x334) +REG32(GEVNTSIZ_3, 0x338) + FIELD(GEVNTSIZ_3, EVNTINTRPTMASK, 31, 1) + FIELD(GEVNTSIZ_3, RESERVED_30_16, 16, 15) + FIELD(GEVNTSIZ_3, EVENTSIZ, 0, 16) +REG32(GEVNTCOUNT_3, 0x33c) + FIELD(GEVNTCOUNT_3, EVNT_HANDLER_BUSY, 31, 1) + FIELD(GEVNTCOUNT_3, RESERVED_30_16, 16, 15) + FIELD(GEVNTCOUNT_3, EVNTCOUNT, 0, 16) +REG32(GHWPARAMS8, 0x500) +REG32(GTXFIFOPRIDEV, 0x510) + FIELD(GTXFIFOPRIDEV, RESERVED_31_N, 6, 26) + FIELD(GTXFIFOPRIDEV, GTXFIFOPRIDEV, 0, 6) +REG32(GTXFIFOPRIHST, 0x518) + FIELD(GTXFIFOPRIHST, RESERVED_31_16, 3, 29) + FIELD(GTXFIFOPRIHST, GTXFIFOPRIHST, 0, 3) +REG32(GRXFIFOPRIHST, 0x51c) + FIELD(GRXFIFOPRIHST, RESERVED_31_16, 3, 29) + FIELD(GRXFIFOPRIHST, GRXFIFOPRIHST, 0, 3) +REG32(GDMAHLRATIO, 0x524) + FIELD(GDMAHLRATIO, RESERVED_31_13, 13, 19) + FIELD(GDMAHLRATIO, HSTRXFIFO, 8, 5) + FIELD(GDMAHLRATIO, RESERVED_7_5, 5, 3) + FIELD(GDMAHLRATIO, HSTTXFIFO, 0, 5) +REG32(GFLADJ, 0x530) + FIELD(GFLADJ, GFLADJ_REFCLK_240MHZDECR_PLS1, 31, 1) + FIELD(GFLADJ, GFLADJ_REFCLK_240MHZ_DECR, 24, 7) + FIELD(GFLADJ, GFLADJ_REFCLK_LPM_SEL, 23, 1) + FIELD(GFLADJ, RESERVED_22, 22, 1) + FIELD(GFLADJ, GFLADJ_REFCLK_FLADJ, 8, 14) + FIELD(GFLADJ, GFLADJ_30MHZ_SDBND_SEL, 7, 1) + FIELD(GFLADJ, GFLADJ_30MHZ, 0, 6) + +#define DWC3_GLOBAL_OFFSET 0xC100 +static void reset_csr(USBDWC3 * s) +{ + int i = 0; + /* + * We reset all CSR regs except GCTL, GUCTL, GSTS, GSNPSID, GGPIO, GUID, + * GUSB2PHYCFGn registers and GUSB3PIPECTLn registers. We will skip PHY + * register as we don't implement them. + */ + for (i = 0; i < USB_DWC3_R_MAX; i++) { + switch (i) { + case R_GCTL: + break; + case R_GSTS: + break; + case R_GSNPSID: + break; + case R_GGPIO: + break; + case R_GUID: + break; + case R_GUCTL: + break; + case R_GHWPARAMS0...R_GHWPARAMS7: + break; + case R_GHWPARAMS8: + break; + default: + register_reset(&s->regs_info[i]); + break; + } + } + + xhci_sysbus_reset(DEVICE(&s->sysbus_xhci)); +} + +static void usb_dwc3_gctl_postw(RegisterInfo *reg, uint64_t val64) +{ + USBDWC3 *s = USB_DWC3(reg->opaque); + + if (ARRAY_FIELD_EX32(s->regs, GCTL, CORESOFTRESET)) { + reset_csr(s); + } +} + +static void usb_dwc3_guid_postw(RegisterInfo *reg, uint64_t val64) +{ + USBDWC3 *s = USB_DWC3(reg->opaque); + + s->regs[R_GUID] = s->cfg.dwc_usb3_user; +} + +static const RegisterAccessInfo usb_dwc3_regs_info[] = { + { .name = "GSBUSCFG0", .addr = A_GSBUSCFG0, + .ro = 0xf300, + .unimp = 0xffffffff, + },{ .name = "GSBUSCFG1", .addr = A_GSBUSCFG1, + .reset = 0x300, + .ro = 0xffffe0ff, + .unimp = 0xffffffff, + },{ .name = "GTXTHRCFG", .addr = A_GTXTHRCFG, + .ro = 0xd000ffff, + .unimp = 0xffffffff, + },{ .name = "GRXTHRCFG", .addr = A_GRXTHRCFG, + .ro = 0xd007e000, + .unimp = 0xffffffff, + },{ .name = "GCTL", .addr = A_GCTL, + .reset = 0x30c13004, .post_write = usb_dwc3_gctl_postw, + },{ .name = "GPMSTS", .addr = A_GPMSTS, + .ro = 0xfffffff, + .unimp = 0xffffffff, + },{ .name = "GSTS", .addr = A_GSTS, + .reset = 0x7e800000, + .ro = 0xffffffcf, + .w1c = 0x30, + .unimp = 0xffffffff, + },{ .name = "GUCTL1", .addr = A_GUCTL1, + .reset = 0x198a, + .ro = 0x7800, + .unimp = 0xffffffff, + },{ .name = "GSNPSID", .addr = A_GSNPSID, + .reset = 0x5533330a, + .ro = 0xffffffff, + },{ .name = "GGPIO", .addr = A_GGPIO, + .ro = 0xffff, + .unimp = 0xffffffff, + },{ .name = "GUID", .addr = A_GUID, + .reset = 0x12345678, .post_write = usb_dwc3_guid_postw, + },{ .name = "GUCTL", .addr = A_GUCTL, + .reset = 0x0c808010, + .ro = 0x1c8000, + .unimp = 0xffffffff, + },{ .name = "GBUSERRADDRLO", .addr = A_GBUSERRADDRLO, + .ro = 0xffffffff, + },{ .name = "GBUSERRADDRHI", .addr = A_GBUSERRADDRHI, + .ro = 0xffffffff, + },{ .name = "GHWPARAMS0", .addr = A_GHWPARAMS0, + .ro = 0xffffffff, + },{ .name = "GHWPARAMS1", .addr = A_GHWPARAMS1, + .ro = 0xffffffff, + },{ .name = "GHWPARAMS2", .addr = A_GHWPARAMS2, + .ro = 0xffffffff, + },{ .name = "GHWPARAMS3", .addr = A_GHWPARAMS3, + .ro = 0xffffffff, + },{ .name = "GHWPARAMS4", .addr = A_GHWPARAMS4, + .ro = 0xffffffff, + },{ .name = "GHWPARAMS5", .addr = A_GHWPARAMS5, + .ro = 0xffffffff, + },{ .name = "GHWPARAMS6", .addr = A_GHWPARAMS6, + .ro = 0xffffffff, + },{ .name = "GHWPARAMS7", .addr = A_GHWPARAMS7, + .ro = 0xffffffff, + },{ .name = "GDBGFIFOSPACE", .addr = A_GDBGFIFOSPACE, + .reset = 0xa0000, + .ro = 0xfffffe00, + .unimp = 0xffffffff, + },{ .name = "GUCTL2", .addr = A_GUCTL2, + .reset = 0x40d, + .ro = 0x2000, + .unimp = 0xffffffff, + },{ .name = "GUSB2PHYCFG", .addr = A_GUSB2PHYCFG, + .reset = 0x40102410, + .ro = 0x1e014030, + .unimp = 0xffffffff, + },{ .name = "GUSB2I2CCTL", .addr = A_GUSB2I2CCTL, + .ro = 0xffffffff, + .unimp = 0xffffffff, + },{ .name = "GUSB2PHYACC_ULPI", .addr = A_GUSB2PHYACC_ULPI, + .ro = 0xfd000000, + .unimp = 0xffffffff, + },{ .name = "GTXFIFOSIZ0", .addr = A_GTXFIFOSIZ0, + .reset = 0x2c7000a, + .unimp = 0xffffffff, + },{ .name = "GTXFIFOSIZ1", .addr = A_GTXFIFOSIZ1, + .reset = 0x2d10103, + .unimp = 0xffffffff, + },{ .name = "GTXFIFOSIZ2", .addr = A_GTXFIFOSIZ2, + .reset = 0x3d40103, + .unimp = 0xffffffff, + },{ .name = "GTXFIFOSIZ3", .addr = A_GTXFIFOSIZ3, + .reset = 0x4d70083, + .unimp = 0xffffffff, + },{ .name = "GTXFIFOSIZ4", .addr = A_GTXFIFOSIZ4, + .reset = 0x55a0083, + .unimp = 0xffffffff, + },{ .name = "GTXFIFOSIZ5", .addr = A_GTXFIFOSIZ5, + .reset = 0x5dd0083, + .unimp = 0xffffffff, + },{ .name = "GRXFIFOSIZ0", .addr = A_GRXFIFOSIZ0, + .reset = 0x1c20105, + .unimp = 0xffffffff, + },{ .name = "GRXFIFOSIZ1", .addr = A_GRXFIFOSIZ1, + .reset = 0x2c70000, + .unimp = 0xffffffff, + },{ .name = "GRXFIFOSIZ2", .addr = A_GRXFIFOSIZ2, + .reset = 0x2c70000, + .unimp = 0xffffffff, + },{ .name = "GEVNTADRLO_0", .addr = A_GEVNTADRLO_0, + .unimp = 0xffffffff, + },{ .name = "GEVNTADRHI_0", .addr = A_GEVNTADRHI_0, + .unimp = 0xffffffff, + },{ .name = "GEVNTSIZ_0", .addr = A_GEVNTSIZ_0, + .ro = 0x7fff0000, + .unimp = 0xffffffff, + },{ .name = "GEVNTCOUNT_0", .addr = A_GEVNTCOUNT_0, + .ro = 0x7fff0000, + .unimp = 0xffffffff, + },{ .name = "GEVNTADRLO_1", .addr = A_GEVNTADRLO_1, + .unimp = 0xffffffff, + },{ .name = "GEVNTADRHI_1", .addr = A_GEVNTADRHI_1, + .unimp = 0xffffffff, + },{ .name = "GEVNTSIZ_1", .addr = A_GEVNTSIZ_1, + .ro = 0x7fff0000, + .unimp = 0xffffffff, + },{ .name = "GEVNTCOUNT_1", .addr = A_GEVNTCOUNT_1, + .ro = 0x7fff0000, + .unimp = 0xffffffff, + },{ .name = "GEVNTADRLO_2", .addr = A_GEVNTADRLO_2, + .unimp = 0xffffffff, + },{ .name = "GEVNTADRHI_2", .addr = A_GEVNTADRHI_2, + .unimp = 0xffffffff, + },{ .name = "GEVNTSIZ_2", .addr = A_GEVNTSIZ_2, + .ro = 0x7fff0000, + .unimp = 0xffffffff, + },{ .name = "GEVNTCOUNT_2", .addr = A_GEVNTCOUNT_2, + .ro = 0x7fff0000, + .unimp = 0xffffffff, + },{ .name = "GEVNTADRLO_3", .addr = A_GEVNTADRLO_3, + .unimp = 0xffffffff, + },{ .name = "GEVNTADRHI_3", .addr = A_GEVNTADRHI_3, + .unimp = 0xffffffff, + },{ .name = "GEVNTSIZ_3", .addr = A_GEVNTSIZ_3, + .ro = 0x7fff0000, + .unimp = 0xffffffff, + },{ .name = "GEVNTCOUNT_3", .addr = A_GEVNTCOUNT_3, + .ro = 0x7fff0000, + .unimp = 0xffffffff, + },{ .name = "GHWPARAMS8", .addr = A_GHWPARAMS8, + .ro = 0xffffffff, + },{ .name = "GTXFIFOPRIDEV", .addr = A_GTXFIFOPRIDEV, + .ro = 0xffffffc0, + .unimp = 0xffffffff, + },{ .name = "GTXFIFOPRIHST", .addr = A_GTXFIFOPRIHST, + .ro = 0xfffffff8, + .unimp = 0xffffffff, + },{ .name = "GRXFIFOPRIHST", .addr = A_GRXFIFOPRIHST, + .ro = 0xfffffff8, + .unimp = 0xffffffff, + },{ .name = "GDMAHLRATIO", .addr = A_GDMAHLRATIO, + .ro = 0xffffe0e0, + .unimp = 0xffffffff, + },{ .name = "GFLADJ", .addr = A_GFLADJ, + .reset = 0xc83f020, + .rsvd = 0x40, + .ro = 0x400040, + .unimp = 0xffffffff, + } +}; + +static void usb_dwc3_reset(DeviceState *dev) +{ + USBDWC3 *s = USB_DWC3(dev); + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(s->regs_info); ++i) { + switch (i) { + case R_GHWPARAMS0...R_GHWPARAMS7: + break; + case R_GHWPARAMS8: + break; + default: + register_reset(&s->regs_info[i]); + }; + } + + xhci_sysbus_reset(DEVICE(&s->sysbus_xhci)); +} + +static const MemoryRegionOps usb_dwc3_ops = { + .read = register_read_memory, + .write = register_write_memory, + .endianness = DEVICE_LITTLE_ENDIAN, + .valid = { + .min_access_size = 4, + .max_access_size = 4, + }, +}; + +static void usb_dwc3_realize(DeviceState *dev, Error **errp) +{ + USBDWC3 *s = USB_DWC3(dev); + SysBusDevice *sbd = SYS_BUS_DEVICE(dev); + Error *err = NULL; + + sysbus_realize(SYS_BUS_DEVICE(&s->sysbus_xhci), &err); + if (err) { + error_propagate(errp, err); + return; + } + + memory_region_add_subregion(&s->iomem, 0, + sysbus_mmio_get_region(SYS_BUS_DEVICE(&s->sysbus_xhci), 0)); + sysbus_init_mmio(sbd, &s->iomem); + + /* + * Device Configuration + */ + s->regs[R_GHWPARAMS0] = 0x40204048 | s->cfg.mode; + s->regs[R_GHWPARAMS1] = 0x222493b; + s->regs[R_GHWPARAMS2] = 0x12345678; + s->regs[R_GHWPARAMS3] = 0x618c088; + s->regs[R_GHWPARAMS4] = 0x47822004; + s->regs[R_GHWPARAMS5] = 0x4202088; + s->regs[R_GHWPARAMS6] = 0x7850c20; + s->regs[R_GHWPARAMS7] = 0x0; + s->regs[R_GHWPARAMS8] = 0x478; +} + +static void usb_dwc3_init(Object *obj) +{ + USBDWC3 *s = USB_DWC3(obj); + RegisterInfoArray *reg_array; + + memory_region_init(&s->iomem, obj, TYPE_USB_DWC3, DWC3_SIZE); + reg_array = + register_init_block32(DEVICE(obj), usb_dwc3_regs_info, + ARRAY_SIZE(usb_dwc3_regs_info), + s->regs_info, s->regs, + &usb_dwc3_ops, + USB_DWC3_ERR_DEBUG, + USB_DWC3_R_MAX * 4); + memory_region_add_subregion(&s->iomem, + DWC3_GLOBAL_OFFSET, + &reg_array->mem); + object_initialize_child(obj, "dwc3-xhci", &s->sysbus_xhci, + TYPE_XHCI_SYSBUS); + qdev_alias_all_properties(DEVICE(&s->sysbus_xhci), obj); + + s->cfg.mode = HOST_MODE; +} + +static const VMStateDescription vmstate_usb_dwc3 = { + .name = "usb-dwc3", + .version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT32_ARRAY(regs, USBDWC3, USB_DWC3_R_MAX), + VMSTATE_UINT8(cfg.mode, USBDWC3), + VMSTATE_UINT32(cfg.dwc_usb3_user, USBDWC3), + VMSTATE_END_OF_LIST() + } +}; + +static Property usb_dwc3_properties[] = { + DEFINE_PROP_UINT32("DWC_USB3_USERID", USBDWC3, cfg.dwc_usb3_user, + 0x12345678), + DEFINE_PROP_END_OF_LIST(), +}; + +static void usb_dwc3_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->reset = usb_dwc3_reset; + dc->realize = usb_dwc3_realize; + dc->vmsd = &vmstate_usb_dwc3; + device_class_set_props(dc, usb_dwc3_properties); +} + +static const TypeInfo usb_dwc3_info = { + .name = TYPE_USB_DWC3, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_size = sizeof(USBDWC3), + .class_init = usb_dwc3_class_init, + .instance_init = usb_dwc3_init, +}; + +static void usb_dwc3_register_types(void) +{ + type_register_static(&usb_dwc3_info); +} + +type_init(usb_dwc3_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 IMX_USBPHY bool default y depends on USB + +config USB_DWC3 + bool + select USB_XHCI_SYSBUS + select REGISTER diff --git a/hw/usb/meson.build b/hw/usb/meson.build index XXXXXXX..XXXXXXX 100644 --- a/hw/usb/meson.build +++ b/hw/usb/meson.build @@ -XXX,XX +XXX,XX @@ softmmu_ss.add(when: 'CONFIG_USB_XHCI_SYSBUS', if_true: files('hcd-xhci-sysbus.c softmmu_ss.add(when: 'CONFIG_USB_XHCI_NEC', if_true: files('hcd-xhci-nec.c')) softmmu_ss.add(when: 'CONFIG_USB_MUSB', if_true: files('hcd-musb.c')) softmmu_ss.add(when: 'CONFIG_USB_DWC2', if_true: files('hcd-dwc2.c')) +softmmu_ss.add(when: 'CONFIG_USB_DWC3', if_true: files('hcd-dwc3.c')) softmmu_ss.add(when: 'CONFIG_TUSB6010', if_true: files('tusb6010.c')) softmmu_ss.add(when: 'CONFIG_IMX', if_true: files('chipidea.c')) -- 2.20.1 From: Sai Pavan Boddu <sai.pavan.boddu@xilinx.com> This model is a top level integration wrapper for hcd-dwc3 and versal-usb2-ctrl-regs modules, this is used by xilinx versal soc's and future xilinx usb subsystems would also be part of it. Signed-off-by: Sai Pavan Boddu <sai.pavan.boddu@xilinx.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Message-id: 1607023357-5096-4-git-send-email-sai.pavan.boddu@xilinx.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/usb/xlnx-usb-subsystem.h | 45 ++++++++++++++ hw/usb/xlnx-usb-subsystem.c | 94 +++++++++++++++++++++++++++++ hw/usb/Kconfig | 5 ++ hw/usb/meson.build | 1 + 4 files changed, 145 insertions(+) create mode 100644 include/hw/usb/xlnx-usb-subsystem.h create mode 100644 hw/usb/xlnx-usb-subsystem.c diff --git a/include/hw/usb/xlnx-usb-subsystem.h b/include/hw/usb/xlnx-usb-subsystem.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/usb/xlnx-usb-subsystem.h @@ -XXX,XX +XXX,XX @@ +/* + * QEMU model of the Xilinx usb subsystem + * + * Copyright (c) 2020 Xilinx Inc. Sai Pavan Boddu <sai.pavan.boddu@xilinx.com> + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ + +#ifndef _XLNX_VERSAL_USB_SUBSYSTEM_H_ +#define _XLNX_VERSAL_USB_SUBSYSTEM_H_ + +#include "hw/usb/xlnx-versal-usb2-ctrl-regs.h" +#include "hw/usb/hcd-dwc3.h" + +#define TYPE_XILINX_VERSAL_USB2 "xlnx.versal-usb2" + +#define VERSAL_USB2(obj) \ + OBJECT_CHECK(VersalUsb2, (obj), TYPE_XILINX_VERSAL_USB2) + +typedef struct VersalUsb2 { + SysBusDevice parent_obj; + MemoryRegion dwc3_mr; + MemoryRegion usb2Ctrl_mr; + + VersalUsb2CtrlRegs usb2Ctrl; + USBDWC3 dwc3; +} VersalUsb2; + +#endif diff --git a/hw/usb/xlnx-usb-subsystem.c b/hw/usb/xlnx-usb-subsystem.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/usb/xlnx-usb-subsystem.c @@ -XXX,XX +XXX,XX @@ +/* + * QEMU model of the Xilinx usb subsystem + * + * Copyright (c) 2020 Xilinx Inc. Sai Pavan Boddu <sai.pava.boddu@xilinx.com> + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this software and associated documentation files (the "Software"), to deal + * in the Software without restriction, including without limitation the rights + * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell + * copies of the Software, and to permit persons to whom the Software is + * furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN + * THE SOFTWARE. + */ + +#include "qemu/osdep.h" +#include "hw/sysbus.h" +#include "hw/irq.h" +#include "hw/register.h" +#include "qemu/bitops.h" +#include "qemu/log.h" +#include "qom/object.h" +#include "qapi/error.h" +#include "hw/qdev-properties.h" +#include "hw/usb/xlnx-usb-subsystem.h" + +static void versal_usb2_realize(DeviceState *dev, Error **errp) +{ + VersalUsb2 *s = VERSAL_USB2(dev); + SysBusDevice *sbd = SYS_BUS_DEVICE(dev); + Error *err = NULL; + + sysbus_realize(SYS_BUS_DEVICE(&s->dwc3), &err); + if (err) { + error_propagate(errp, err); + return; + } + sysbus_realize(SYS_BUS_DEVICE(&s->usb2Ctrl), &err); + if (err) { + error_propagate(errp, err); + return; + } + sysbus_init_mmio(sbd, &s->dwc3_mr); + sysbus_init_mmio(sbd, &s->usb2Ctrl_mr); + qdev_pass_gpios(DEVICE(&s->dwc3.sysbus_xhci), dev, SYSBUS_DEVICE_GPIO_IRQ); +} + +static void versal_usb2_init(Object *obj) +{ + VersalUsb2 *s = VERSAL_USB2(obj); + + object_initialize_child(obj, "versal.dwc3", &s->dwc3, + TYPE_USB_DWC3); + object_initialize_child(obj, "versal.usb2-ctrl", &s->usb2Ctrl, + TYPE_XILINX_VERSAL_USB2_CTRL_REGS); + memory_region_init_alias(&s->dwc3_mr, obj, "versal.dwc3_alias", + &s->dwc3.iomem, 0, DWC3_SIZE); + memory_region_init_alias(&s->usb2Ctrl_mr, obj, "versal.usb2Ctrl_alias", + &s->usb2Ctrl.iomem, 0, USB2_REGS_R_MAX * 4); + qdev_alias_all_properties(DEVICE(&s->dwc3), obj); + qdev_alias_all_properties(DEVICE(&s->dwc3.sysbus_xhci), obj); + object_property_add_alias(obj, "dma", OBJECT(&s->dwc3.sysbus_xhci), "dma"); +} + +static void versal_usb2_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->realize = versal_usb2_realize; +} + +static const TypeInfo versal_usb2_info = { + .name = TYPE_XILINX_VERSAL_USB2, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_size = sizeof(VersalUsb2), + .class_init = versal_usb2_class_init, + .instance_init = versal_usb2_init, +}; + +static void versal_usb_types(void) +{ + type_register_static(&versal_usb2_info); +} + +type_init(versal_usb_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_DWC3 bool select USB_XHCI_SYSBUS select REGISTER + +config XLNX_USB_SUBSYS + bool + default y if XLNX_VERSAL + select USB_DWC3 diff --git a/hw/usb/meson.build b/hw/usb/meson.build index XXXXXXX..XXXXXXX 100644 --- a/hw/usb/meson.build +++ b/hw/usb/meson.build @@ -XXX,XX +XXX,XX @@ softmmu_ss.add(when: 'CONFIG_TUSB6010', if_true: files('tusb6010.c')) softmmu_ss.add(when: 'CONFIG_IMX', if_true: files('chipidea.c')) softmmu_ss.add(when: 'CONFIG_IMX_USBPHY', if_true: files('imx-usb-phy.c')) specific_ss.add(when: 'CONFIG_XLNX_VERSAL', if_true: files('xlnx-versal-usb2-ctrl-regs.c')) +specific_ss.add(when: 'CONFIG_XLNX_USB_SUBSYS', if_true: files('xlnx-usb-subsystem.c')) # emulated usb devices softmmu_ss.add(when: 'CONFIG_USB', if_true: files('dev-hub.c')) -- 2.20.1 From: Vikram Garhwal <fnu.vikram@xilinx.com> Connect VersalUsb2 subsystem to xlnx-versal SOC, its placed in iou of lpd domain and configure it as dual port host controller. Add the respective guest dts nodes for "xlnx-versal-virt" machine. Signed-off-by: Vikram Garhwal <fnu.vikram@xilinx.com> Signed-off-by: Sai Pavan Boddu <sai.pavan.boddu@xilinx.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 1607023357-5096-5-git-send-email-sai.pavan.boddu@xilinx.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/xlnx-versal.h | 9 ++++++ hw/arm/xlnx-versal-virt.c | 55 ++++++++++++++++++++++++++++++++++++ hw/arm/xlnx-versal.c | 26 +++++++++++++++++ 3 files changed, 90 insertions(+) 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/net/cadence_gem.h" #include "hw/rtc/xlnx-zynqmp-rtc.h" #include "qom/object.h" +#include "hw/usb/xlnx-usb-subsystem.h" #define TYPE_XLNX_VERSAL "xlnx-versal" OBJECT_DECLARE_SIMPLE_TYPE(Versal, XLNX_VERSAL) @@ -XXX,XX +XXX,XX @@ struct Versal { PL011State uart[XLNX_VERSAL_NR_UARTS]; CadenceGEMState gem[XLNX_VERSAL_NR_GEMS]; XlnxZDMA adma[XLNX_VERSAL_NR_ADMAS]; + VersalUsb2 usb; } iou; } lpd; @@ -XXX,XX +XXX,XX @@ struct Versal { #define VERSAL_UART0_IRQ_0 18 #define VERSAL_UART1_IRQ_0 19 +#define VERSAL_USB0_IRQ_0 22 #define VERSAL_GEM0_IRQ_0 56 #define VERSAL_GEM0_WAKE_IRQ_0 57 #define VERSAL_GEM1_IRQ_0 58 @@ -XXX,XX +XXX,XX @@ struct Versal { #define MM_OCM 0xfffc0000U #define MM_OCM_SIZE 0x40000 +#define MM_USB2_CTRL_REGS 0xFF9D0000 +#define MM_USB2_CTRL_REGS_SIZE 0x10000 + +#define MM_USB_0 0xFE200000 +#define MM_USB_0_SIZE 0x10000 + #define MM_TOP_DDR 0x0 #define MM_TOP_DDR_SIZE 0x80000000U #define MM_TOP_DDR_2 0x800000000ULL diff --git a/hw/arm/xlnx-versal-virt.c b/hw/arm/xlnx-versal-virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/xlnx-versal-virt.c +++ b/hw/arm/xlnx-versal-virt.c @@ -XXX,XX +XXX,XX @@ struct VersalVirt { uint32_t ethernet_phy[2]; uint32_t clk_125Mhz; uint32_t clk_25Mhz; + uint32_t usb; + uint32_t dwc; } phandle; struct arm_boot_info binfo; @@ -XXX,XX +XXX,XX @@ static void fdt_create(VersalVirt *s) s->phandle.clk_25Mhz = qemu_fdt_alloc_phandle(s->fdt); s->phandle.clk_125Mhz = qemu_fdt_alloc_phandle(s->fdt); + s->phandle.usb = qemu_fdt_alloc_phandle(s->fdt); + s->phandle.dwc = qemu_fdt_alloc_phandle(s->fdt); /* Create /chosen node for load_dtb. */ qemu_fdt_add_subnode(s->fdt, "/chosen"); @@ -XXX,XX +XXX,XX @@ static void fdt_add_timer_nodes(VersalVirt *s) compat, sizeof(compat)); } +static void fdt_add_usb_xhci_nodes(VersalVirt *s) +{ + const char clocknames[] = "bus_clk\0ref_clk"; + const char irq_name[] = "dwc_usb3"; + const char compatVersalDWC3[] = "xlnx,versal-dwc3"; + const char compatDWC3[] = "snps,dwc3"; + char *name = g_strdup_printf("/usb@%" PRIx32, MM_USB2_CTRL_REGS); + + qemu_fdt_add_subnode(s->fdt, name); + qemu_fdt_setprop(s->fdt, name, "compatible", + compatVersalDWC3, sizeof(compatVersalDWC3)); + qemu_fdt_setprop_sized_cells(s->fdt, name, "reg", + 2, MM_USB2_CTRL_REGS, + 2, MM_USB2_CTRL_REGS_SIZE); + qemu_fdt_setprop(s->fdt, name, "clock-names", + clocknames, sizeof(clocknames)); + qemu_fdt_setprop_cells(s->fdt, name, "clocks", + s->phandle.clk_25Mhz, s->phandle.clk_125Mhz); + qemu_fdt_setprop(s->fdt, name, "ranges", NULL, 0); + qemu_fdt_setprop_cell(s->fdt, name, "#address-cells", 2); + qemu_fdt_setprop_cell(s->fdt, name, "#size-cells", 2); + qemu_fdt_setprop_cell(s->fdt, name, "phandle", s->phandle.usb); + g_free(name); + + name = g_strdup_printf("/usb@%" PRIx32 "/dwc3@%" PRIx32, + MM_USB2_CTRL_REGS, MM_USB_0); + qemu_fdt_add_subnode(s->fdt, name); + qemu_fdt_setprop(s->fdt, name, "compatible", + compatDWC3, sizeof(compatDWC3)); + qemu_fdt_setprop_sized_cells(s->fdt, name, "reg", + 2, MM_USB_0, 2, MM_USB_0_SIZE); + qemu_fdt_setprop(s->fdt, name, "interrupt-names", + irq_name, sizeof(irq_name)); + qemu_fdt_setprop_cells(s->fdt, name, "interrupts", + GIC_FDT_IRQ_TYPE_SPI, VERSAL_USB0_IRQ_0, + GIC_FDT_IRQ_FLAGS_LEVEL_HI); + qemu_fdt_setprop_cell(s->fdt, name, + "snps,quirk-frame-length-adjustment", 0x20); + qemu_fdt_setprop_cells(s->fdt, name, "#stream-id-cells", 1); + qemu_fdt_setprop_string(s->fdt, name, "dr_mode", "host"); + qemu_fdt_setprop_string(s->fdt, name, "phy-names", "usb3-phy"); + qemu_fdt_setprop(s->fdt, name, "snps,dis_u2_susphy_quirk", NULL, 0); + qemu_fdt_setprop(s->fdt, name, "snps,dis_u3_susphy_quirk", NULL, 0); + qemu_fdt_setprop(s->fdt, name, "snps,refclk_fladj", NULL, 0); + qemu_fdt_setprop(s->fdt, name, "snps,mask_phy_reset", NULL, 0); + qemu_fdt_setprop_cell(s->fdt, name, "phandle", s->phandle.dwc); + qemu_fdt_setprop_string(s->fdt, name, "maximum-speed", "high-speed"); + g_free(name); +} + static void fdt_add_uart_nodes(VersalVirt *s) { uint64_t addrs[] = { MM_UART1, MM_UART0 }; @@ -XXX,XX +XXX,XX @@ static void versal_virt_init(MachineState *machine) fdt_add_gic_nodes(s); fdt_add_timer_nodes(s); fdt_add_zdma_nodes(s); + fdt_add_usb_xhci_nodes(s); fdt_add_sd_nodes(s); fdt_add_rtc_node(s); fdt_add_cpu_nodes(s, psci_conduit); 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 @@ static void versal_create_uarts(Versal *s, qemu_irq *pic) } } +static void versal_create_usbs(Versal *s, qemu_irq *pic) +{ + DeviceState *dev; + MemoryRegion *mr; + + object_initialize_child(OBJECT(s), "usb2", &s->lpd.iou.usb, + TYPE_XILINX_VERSAL_USB2); + dev = DEVICE(&s->lpd.iou.usb); + + object_property_set_link(OBJECT(dev), "dma", OBJECT(&s->mr_ps), + &error_abort); + qdev_prop_set_uint32(dev, "intrs", 1); + qdev_prop_set_uint32(dev, "slots", 2); + + sysbus_realize(SYS_BUS_DEVICE(dev), &error_fatal); + + mr = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 0); + memory_region_add_subregion(&s->mr_ps, MM_USB_0, mr); + + sysbus_connect_irq(SYS_BUS_DEVICE(dev), 0, pic[VERSAL_USB0_IRQ_0]); + + mr = sysbus_mmio_get_region(SYS_BUS_DEVICE(dev), 1); + memory_region_add_subregion(&s->mr_ps, MM_USB2_CTRL_REGS, mr); +} + static void versal_create_gems(Versal *s, qemu_irq *pic) { int i; @@ -XXX,XX +XXX,XX @@ static void versal_realize(DeviceState *dev, Error **errp) versal_create_apu_cpus(s); versal_create_apu_gic(s, pic); versal_create_uarts(s, pic); + versal_create_usbs(s, pic); versal_create_gems(s, pic); versal_create_admas(s, pic); versal_create_sds(s, pic); -- 2.20.1 From: Philippe Mathieu-Daudé <f4bug@amsat.org> Malicious user can set the feedback divisor for the PLLs to zero, triggering a floating-point exception (SIGFPE). As the datasheet [*] is not clear how hardware behaves when these bits are zeroes, use the maximum divisor possible (128) to avoid the software FPE. [*] Zynq-7000 TRM, UG585 (v1.12.2) B.28 System Level Control Registers (slcr) -> "Register (slcr) ARM_PLL_CTRL" 25.10.4 PLLs -> "Software-Controlled PLL Update" Fixes: 38867cb7ec9 ("hw/misc/zynq_slcr: add clock generation for uarts") Reported-by: Gaoning Pan <pgn@zju.edu.cn> Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Damien Hedde <damien.hedde@greensocs.com> Message-id: 20201210141610.884600-1-f4bug@amsat.org Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/misc/zynq_slcr.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/hw/misc/zynq_slcr.c b/hw/misc/zynq_slcr.c index XXXXXXX..XXXXXXX 100644 --- a/hw/misc/zynq_slcr.c +++ b/hw/misc/zynq_slcr.c @@ -XXX,XX +XXX,XX @@ static uint64_t zynq_slcr_compute_pll(uint64_t input, uint32_t ctrl_reg) return 0; } + /* Consider zero feedback as maximum divide ratio possible */ + if (!mult) { + mult = 1 << R_xxx_PLL_CTRL_PLL_FPDIV_LENGTH; + } + /* frequency multiplier -> period division */ return input / mult; } -- 2.20.1 From: Joe Komlodi <joe.komlodi@xilinx.com> The previous naming of the configuration registers made it sound like that if the bits were set the settings would be enabled, while the opposite is true. Signed-off-by: Joe Komlodi <komlodi@xilinx.com> Reviewed-by: Francisco Iglesias <francisco.iglesias@xilinx.com> Message-id: 1605568264-26376-2-git-send-email-komlodi@xilinx.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/block/m25p80.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/hw/block/m25p80.c b/hw/block/m25p80.c index XXXXXXX..XXXXXXX 100644 --- a/hw/block/m25p80.c +++ b/hw/block/m25p80.c @@ -XXX,XX +XXX,XX @@ typedef struct FlashPartInfo { #define VCFG_WRAP_SEQUENTIAL 0x2 #define NVCFG_XIP_MODE_DISABLED (7 << 9) #define NVCFG_XIP_MODE_MASK (7 << 9) -#define VCFG_XIP_MODE_ENABLED (1 << 3) +#define VCFG_XIP_MODE_DISABLED (1 << 3) #define CFG_DUMMY_CLK_LEN 4 #define NVCFG_DUMMY_CLK_POS 12 #define VCFG_DUMMY_CLK_POS 4 @@ -XXX,XX +XXX,XX @@ typedef struct FlashPartInfo { #define EVCFG_VPP_ACCELERATOR (1 << 3) #define EVCFG_RESET_HOLD_ENABLED (1 << 4) #define NVCFG_DUAL_IO_MASK (1 << 2) -#define EVCFG_DUAL_IO_ENABLED (1 << 6) +#define EVCFG_DUAL_IO_DISABLED (1 << 6) #define NVCFG_QUAD_IO_MASK (1 << 3) -#define EVCFG_QUAD_IO_ENABLED (1 << 7) +#define EVCFG_QUAD_IO_DISABLED (1 << 7) #define NVCFG_4BYTE_ADDR_MASK (1 << 0) #define NVCFG_LOWER_SEGMENT_MASK (1 << 1) @@ -XXX,XX +XXX,XX @@ static void reset_memory(Flash *s) s->volatile_cfg |= VCFG_WRAP_SEQUENTIAL; if ((s->nonvolatile_cfg & NVCFG_XIP_MODE_MASK) != NVCFG_XIP_MODE_DISABLED) { - s->volatile_cfg |= VCFG_XIP_MODE_ENABLED; + s->volatile_cfg |= VCFG_XIP_MODE_DISABLED; } s->volatile_cfg |= deposit32(s->volatile_cfg, VCFG_DUMMY_CLK_POS, @@ -XXX,XX +XXX,XX @@ static void reset_memory(Flash *s) s->enh_volatile_cfg |= EVCFG_VPP_ACCELERATOR; s->enh_volatile_cfg |= EVCFG_RESET_HOLD_ENABLED; if (s->nonvolatile_cfg & NVCFG_DUAL_IO_MASK) { - s->enh_volatile_cfg |= EVCFG_DUAL_IO_ENABLED; + s->enh_volatile_cfg |= EVCFG_DUAL_IO_DISABLED; } if (s->nonvolatile_cfg & NVCFG_QUAD_IO_MASK) { - s->enh_volatile_cfg |= EVCFG_QUAD_IO_ENABLED; + s->enh_volatile_cfg |= EVCFG_QUAD_IO_DISABLED; } if (!(s->nonvolatile_cfg & NVCFG_4BYTE_ADDR_MASK)) { s->four_bytes_address_mode = true; -- 2.20.1 From: Joe Komlodi <joe.komlodi@xilinx.com> VCFG XIP is set (disabled) when the NVCFG XIP bits are all set (disabled). Signed-off-by: Joe Komlodi <komlodi@xilinx.com> Reviewed-by: Francisco Iglesias <francisco.iglesias@xilinx.com> Message-id: 1605568264-26376-3-git-send-email-komlodi@xilinx.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/block/m25p80.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/hw/block/m25p80.c b/hw/block/m25p80.c index XXXXXXX..XXXXXXX 100644 --- a/hw/block/m25p80.c +++ b/hw/block/m25p80.c @@ -XXX,XX +XXX,XX @@ static void reset_memory(Flash *s) s->volatile_cfg |= VCFG_DUMMY; s->volatile_cfg |= VCFG_WRAP_SEQUENTIAL; if ((s->nonvolatile_cfg & NVCFG_XIP_MODE_MASK) - != NVCFG_XIP_MODE_DISABLED) { + == NVCFG_XIP_MODE_DISABLED) { s->volatile_cfg |= VCFG_XIP_MODE_DISABLED; } s->volatile_cfg |= deposit32(s->volatile_cfg, -- 2.20.1 From: Joe Komlodi <joe.komlodi@xilinx.com> Some Numonyx flash commands cannot be executed in DIO and QIO mode, such as trying to do DPP or DOR when in QIO mode. Signed-off-by: Joe Komlodi <komlodi@xilinx.com> Reviewed-by: Francisco Iglesias <francisco.iglesias@xilinx.com> Message-id: 1605568264-26376-4-git-send-email-komlodi@xilinx.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/block/m25p80.c | 114 ++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 95 insertions(+), 19 deletions(-) diff --git a/hw/block/m25p80.c b/hw/block/m25p80.c index XXXXXXX..XXXXXXX 100644 --- a/hw/block/m25p80.c +++ b/hw/block/m25p80.c @@ -XXX,XX +XXX,XX @@ typedef enum { MAN_GENERIC, } Manufacturer; +typedef enum { + MODE_STD = 0, + MODE_DIO = 1, + MODE_QIO = 2 +} SPIMode; + #define M25P80_INTERNAL_DATA_BUFFER_SZ 16 struct Flash { @@ -XXX,XX +XXX,XX @@ static void reset_memory(Flash *s) trace_m25p80_reset_done(s); } +static uint8_t numonyx_mode(Flash *s) +{ + if (!(s->enh_volatile_cfg & EVCFG_QUAD_IO_DISABLED)) { + return MODE_QIO; + } else if (!(s->enh_volatile_cfg & EVCFG_DUAL_IO_DISABLED)) { + return MODE_DIO; + } else { + return MODE_STD; + } +} + static void decode_fast_read_cmd(Flash *s) { s->needed_bytes = get_addr_length(s); @@ -XXX,XX +XXX,XX @@ static void decode_new_cmd(Flash *s, uint32_t value) case ERASE4_32K: case ERASE_SECTOR: case ERASE4_SECTOR: - case READ: - case READ4: - case DPP: - case QPP: - case QPP_4: case PP: case PP4: - case PP4_4: case DIE_ERASE: case RDID_90: case RDID_AB: @@ -XXX,XX +XXX,XX @@ static void decode_new_cmd(Flash *s, uint32_t value) s->len = 0; s->state = STATE_COLLECTING_DATA; break; + case READ: + case READ4: + if (get_man(s) != MAN_NUMONYX || numonyx_mode(s) == MODE_STD) { + s->needed_bytes = get_addr_length(s); + s->pos = 0; + s->len = 0; + s->state = STATE_COLLECTING_DATA; + } else { + qemu_log_mask(LOG_GUEST_ERROR, "M25P80: Cannot execute cmd %x in " + "DIO or QIO mode\n", s->cmd_in_progress); + } + break; + case DPP: + if (get_man(s) != MAN_NUMONYX || numonyx_mode(s) != MODE_QIO) { + s->needed_bytes = get_addr_length(s); + s->pos = 0; + s->len = 0; + s->state = STATE_COLLECTING_DATA; + } else { + qemu_log_mask(LOG_GUEST_ERROR, "M25P80: Cannot execute cmd %x in " + "QIO mode\n", s->cmd_in_progress); + } + break; + case QPP: + case QPP_4: + case PP4_4: + if (get_man(s) != MAN_NUMONYX || numonyx_mode(s) != MODE_DIO) { + s->needed_bytes = get_addr_length(s); + s->pos = 0; + s->len = 0; + s->state = STATE_COLLECTING_DATA; + } else { + qemu_log_mask(LOG_GUEST_ERROR, "M25P80: Cannot execute cmd %x in " + "DIO mode\n", s->cmd_in_progress); + } + break; case FAST_READ: case FAST_READ4: + decode_fast_read_cmd(s); + break; case DOR: case DOR4: + if (get_man(s) != MAN_NUMONYX || numonyx_mode(s) != MODE_QIO) { + decode_fast_read_cmd(s); + } else { + qemu_log_mask(LOG_GUEST_ERROR, "M25P80: Cannot execute cmd %x in " + "QIO mode\n", s->cmd_in_progress); + } + break; case QOR: case QOR4: - decode_fast_read_cmd(s); + if (get_man(s) != MAN_NUMONYX || numonyx_mode(s) != MODE_DIO) { + decode_fast_read_cmd(s); + } else { + qemu_log_mask(LOG_GUEST_ERROR, "M25P80: Cannot execute cmd %x in " + "DIO mode\n", s->cmd_in_progress); + } break; case DIOR: case DIOR4: - decode_dio_read_cmd(s); + if (get_man(s) != MAN_NUMONYX || numonyx_mode(s) != MODE_QIO) { + decode_dio_read_cmd(s); + } else { + qemu_log_mask(LOG_GUEST_ERROR, "M25P80: Cannot execute cmd %x in " + "QIO mode\n", s->cmd_in_progress); + } break; case QIOR: case QIOR4: - decode_qio_read_cmd(s); + if (get_man(s) != MAN_NUMONYX || numonyx_mode(s) != MODE_DIO) { + decode_qio_read_cmd(s); + } else { + qemu_log_mask(LOG_GUEST_ERROR, "M25P80: Cannot execute cmd %x in " + "DIO mode\n", s->cmd_in_progress); + } break; case WRSR: @@ -XXX,XX +XXX,XX @@ static void decode_new_cmd(Flash *s, uint32_t value) break; case JEDEC_READ: - trace_m25p80_populated_jedec(s); - for (i = 0; i < s->pi->id_len; i++) { - s->data[i] = s->pi->id[i]; - } - for (; i < SPI_NOR_MAX_ID_LEN; i++) { - s->data[i] = 0; - } + if (get_man(s) != MAN_NUMONYX || numonyx_mode(s) == MODE_STD) { + trace_m25p80_populated_jedec(s); + for (i = 0; i < s->pi->id_len; i++) { + s->data[i] = s->pi->id[i]; + } + for (; i < SPI_NOR_MAX_ID_LEN; i++) { + s->data[i] = 0; + } - s->len = SPI_NOR_MAX_ID_LEN; - s->pos = 0; - s->state = STATE_READING_DATA; + s->len = SPI_NOR_MAX_ID_LEN; + s->pos = 0; + s->state = STATE_READING_DATA; + } else { + qemu_log_mask(LOG_GUEST_ERROR, "M25P80: Cannot execute JEDEC read " + "in DIO or QIO mode\n"); + } break; case RDCR: -- 2.20.1 From: Joe Komlodi <joe.komlodi@xilinx.com> Numonyx chips determine the number of cycles to wait based on bits 7:4 in the volatile configuration register. However, if these bits are 0x0 or 0xF, the number of dummy cycles to wait is 10 for QIOR and QIOR4 commands or when in QIO mode, and otherwise 8 for the currently supported fast read commands. [1] [1] https://www.micron.com/-/media/client/global/documents/products/data-sheet/nor-flash/serial-nor/mt25q/die-rev-b/mt25q_qlkt_u_02g_cbb_0.pdf?rev=9b167fbf2b3645efba6385949a72e453 Signed-off-by: Joe Komlodi <komlodi@xilinx.com> Reviewed-by: Francisco Iglesias <francisco.iglesias@xilinx.com> Message-id: 1605568264-26376-5-git-send-email-komlodi@xilinx.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/block/m25p80.c | 30 +++++++++++++++++++++++++++--- 1 file changed, 27 insertions(+), 3 deletions(-) diff --git a/hw/block/m25p80.c b/hw/block/m25p80.c index XXXXXXX..XXXXXXX 100644 --- a/hw/block/m25p80.c +++ b/hw/block/m25p80.c @@ -XXX,XX +XXX,XX @@ static uint8_t numonyx_mode(Flash *s) } } +static uint8_t numonyx_extract_cfg_num_dummies(Flash *s) +{ + uint8_t num_dummies; + uint8_t mode; + assert(get_man(s) == MAN_NUMONYX); + + mode = numonyx_mode(s); + num_dummies = extract32(s->volatile_cfg, 4, 4); + + if (num_dummies == 0x0 || num_dummies == 0xf) { + switch (s->cmd_in_progress) { + case QIOR: + case QIOR4: + num_dummies = 10; + break; + default: + num_dummies = (mode == MODE_QIO) ? 10 : 8; + break; + } + } + + return num_dummies; +} + static void decode_fast_read_cmd(Flash *s) { s->needed_bytes = get_addr_length(s); @@ -XXX,XX +XXX,XX @@ static void decode_fast_read_cmd(Flash *s) s->needed_bytes += 8; break; case MAN_NUMONYX: - s->needed_bytes += extract32(s->volatile_cfg, 4, 4); + s->needed_bytes += numonyx_extract_cfg_num_dummies(s); break; case MAN_MACRONIX: if (extract32(s->volatile_cfg, 6, 2) == 1) { @@ -XXX,XX +XXX,XX @@ static void decode_dio_read_cmd(Flash *s) ); break; case MAN_NUMONYX: - s->needed_bytes += extract32(s->volatile_cfg, 4, 4); + s->needed_bytes += numonyx_extract_cfg_num_dummies(s); break; case MAN_MACRONIX: switch (extract32(s->volatile_cfg, 6, 2)) { @@ -XXX,XX +XXX,XX @@ static void decode_qio_read_cmd(Flash *s) ); break; case MAN_NUMONYX: - s->needed_bytes += extract32(s->volatile_cfg, 4, 4); + s->needed_bytes += numonyx_extract_cfg_num_dummies(s); break; case MAN_MACRONIX: switch (extract32(s->volatile_cfg, 6, 2)) { -- 2.20.1