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 First arm pullreq of the 5.1 cycle; mostly bugfixes and some cleanup patches. The new clock modelling framework is the big thing here. -- PMM The following changes since commit 648db19685b7030aa558a4ddbd3a8e53d8c9a062: Merge remote-tracking branch 'remotes/armbru/tags/pull-misc-2020-04-29' into staging (2020-04-29 15:07:33 +0100) are available in the Git repository at: https://git.linaro.org/people/pmaydell/qemu-arm.git tags/pull-target-arm-20200430 for you to fetch changes up to 1267437e593e85498f9105b3bdab796630d2e83f: hw/arm: xlnx-zcu102: Disable unsupported FDT firmware nodes (2020-04-30 11:52:29 +0100) ---------------------------------------------------------------- target-arm queue: * xlnx-zdma: Fix endianness handling of descriptor loading * nrf51: Fix last GPIO CNF address * gicv3: Use gicr_typer in arm_gicv3_icc_reset * msf2: Add EMAC block to SmartFusion2 SoC * New clock modelling framework * hw/arm: versal: Setup the ADMA with 128bit bus-width * Cadence: gem: fix wraparound in 64bit descriptors * cadence_gem: clear RX control descriptor * target/arm: Vectorize integer comparison vs zero * hw/arm/virt: dt: add kaslr-seed property * hw/arm: xlnx-zcu102: Disable unsupported FDT firmware nodes ---------------------------------------------------------------- Cameron Esfahani (1): nrf51: Fix last GPIO CNF address Damien Hedde (7): hw/core/clock-vmstate: define a vmstate entry for clock state qdev: add clock input&output support to devices. qdev-clock: introduce an init array to ease the device construction hw/misc/zynq_slcr: add clock generation for uarts hw/char/cadence_uart: add clock support hw/arm/xilinx_zynq: connect uart clocks to slcr qdev-monitor: print the device's clock with info qtree Edgar E. Iglesias (7): dma/xlnx-zdma: Fix descriptor loading (MEM) wrt endianness dma/xlnx-zdma: Fix descriptor loading (REG) wrt endianness hw/arm: versal: Setup the ADMA with 128bit bus-width device_tree: Allow name wildcards in qemu_fdt_node_path() device_tree: Constify compat in qemu_fdt_node_path() hw/arm: xlnx-zcu102: Move arm_boot_info into XlnxZCU102 hw/arm: xlnx-zcu102: Disable unsupported FDT firmware nodes Jerome Forissier (2): hw/arm/virt: dt: move creation of /secure-chosen to create_fdt() hw/arm/virt: dt: add kaslr-seed property Keqian Zhu (2): bugfix: Use gicr_typer in arm_gicv3_icc_reset Typo: Correct the name of CPU hotplug memory region Peter Maydell (2): hw/core/clock: introduce clock object docs/clocks: add device's clock documentation Philippe Mathieu-Daudé (3): target/arm: Restrict the Address Translate write operation to TCG accel target/arm/cpu: Use ARRAY_SIZE() to iterate over ARMCPUInfo[] target/arm/cpu: Update coding style to make checkpatch.pl happy Ramon Fried (2): Cadence: gem: fix wraparound in 64bit descriptors net: cadence_gem: clear RX control descriptor Richard Henderson (1): target/arm: Vectorize integer comparison vs zero Subbaraya Sundeep (3): hw/net: Add Smartfusion2 emac block msf2: Add EMAC block to SmartFusion2 SoC tests/boot_linux_console: Add ethernet test to SmartFusion2 Thomas Huth (1): target/arm: Make cpu_register() available for other files hw/core/Makefile.objs | 2 + hw/net/Makefile.objs | 1 + tests/Makefile.include | 1 + include/hw/arm/msf2-soc.h | 2 + include/hw/char/cadence_uart.h | 1 + include/hw/clock.h | 225 +++++++++++++ include/hw/gpio/nrf51_gpio.h | 2 +- include/hw/net/msf2-emac.h | 53 +++ include/hw/qdev-clock.h | 159 +++++++++ include/hw/qdev-core.h | 12 + include/sysemu/device_tree.h | 5 +- target/arm/cpu-qom.h | 9 +- target/arm/helper.h | 27 +- target/arm/translate.h | 5 + device_tree.c | 4 +- hw/acpi/cpu.c | 2 +- hw/arm/msf2-soc.c | 26 +- hw/arm/virt.c | 20 +- hw/arm/xilinx_zynq.c | 57 +++- hw/arm/xlnx-versal.c | 2 + hw/arm/xlnx-zcu102.c | 39 ++- hw/char/cadence_uart.c | 73 +++- hw/core/clock-vmstate.c | 25 ++ hw/core/clock.c | 130 ++++++++ hw/core/qdev-clock.c | 185 +++++++++++ hw/core/qdev.c | 12 + hw/dma/xlnx-zdma.c | 25 +- hw/intc/arm_gicv3_kvm.c | 4 +- hw/misc/zynq_slcr.c | 172 +++++++++- hw/net/cadence_gem.c | 16 +- hw/net/msf2-emac.c | 589 +++++++++++++++++++++++++++++++++ qdev-monitor.c | 9 + target/arm/cpu.c | 25 +- target/arm/cpu64.c | 16 +- target/arm/helper.c | 17 + target/arm/neon_helper.c | 24 -- target/arm/translate-a64.c | 64 +--- target/arm/translate.c | 256 ++++++++++++-- target/arm/vec_helper.c | 25 ++ MAINTAINERS | 2 + docs/devel/clocks.rst | 391 ++++++++++++++++++++++ docs/devel/index.rst | 1 + hw/char/trace-events | 3 + hw/core/trace-events | 7 + tests/acceptance/boot_linux_console.py | 15 +- 45 files changed, 2538 insertions(+), 202 deletions(-) create mode 100644 include/hw/clock.h create mode 100644 include/hw/net/msf2-emac.h create mode 100644 include/hw/qdev-clock.h create mode 100644 hw/core/clock-vmstate.c create mode 100644 hw/core/clock.c create mode 100644 hw/core/qdev-clock.c create mode 100644 hw/net/msf2-emac.c create mode 100644 docs/devel/clocks.rst From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Fix descriptor loading from memory wrt host endianness. Reported-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Francisco Iglesias <frasse.iglesias@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20200404122718.25111-2-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/dma/xlnx-zdma.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/hw/dma/xlnx-zdma.c b/hw/dma/xlnx-zdma.c index XXXXXXX..XXXXXXX 100644 --- a/hw/dma/xlnx-zdma.c +++ b/hw/dma/xlnx-zdma.c @@ -XXX,XX +XXX,XX @@ static void zdma_put_regaddr64(XlnxZDMA *s, unsigned int basereg, uint64_t addr) s->regs[basereg + 1] = addr >> 32; } -static bool zdma_load_descriptor(XlnxZDMA *s, uint64_t addr, void *buf) +static bool zdma_load_descriptor(XlnxZDMA *s, uint64_t addr, + XlnxZDMADescr *descr) { /* ZDMA descriptors must be aligned to their own size. */ if (addr % sizeof(XlnxZDMADescr)) { qemu_log_mask(LOG_GUEST_ERROR, "zdma: unaligned descriptor at %" PRIx64, addr); - memset(buf, 0x0, sizeof(XlnxZDMADescr)); + memset(descr, 0x0, sizeof(XlnxZDMADescr)); s->error = true; return false; } - address_space_read(s->dma_as, addr, s->attr, buf, sizeof(XlnxZDMADescr)); + descr->addr = address_space_ldq_le(s->dma_as, addr, s->attr, NULL); + descr->size = address_space_ldl_le(s->dma_as, addr + 8, s->attr, NULL); + descr->attr = address_space_ldl_le(s->dma_as, addr + 12, s->attr, NULL); return true; } @@ -XXX,XX +XXX,XX @@ static void zdma_update_descr_addr(XlnxZDMA *s, bool type, } else { addr = zdma_get_regaddr64(s, basereg); addr += sizeof(s->dsc_dst); - address_space_read(s->dma_as, addr, s->attr, (void *) &next, 8); + next = address_space_ldq_le(s->dma_as, addr, s->attr, NULL); } zdma_put_regaddr64(s, basereg, next); -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Fix descriptor loading from registers wrt host endianness. Reported-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Francisco Iglesias <frasse.iglesias@gmail.com> Message-id: 20200404122718.25111-3-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/dma/xlnx-zdma.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/hw/dma/xlnx-zdma.c b/hw/dma/xlnx-zdma.c index XXXXXXX..XXXXXXX 100644 --- a/hw/dma/xlnx-zdma.c +++ b/hw/dma/xlnx-zdma.c @@ -XXX,XX +XXX,XX @@ static void zdma_put_regaddr64(XlnxZDMA *s, unsigned int basereg, uint64_t addr) s->regs[basereg + 1] = addr >> 32; } +static void zdma_load_descriptor_reg(XlnxZDMA *s, unsigned int reg, + XlnxZDMADescr *descr) +{ + descr->addr = zdma_get_regaddr64(s, reg); + descr->size = s->regs[reg + 2]; + descr->attr = s->regs[reg + 3]; +} + static bool zdma_load_descriptor(XlnxZDMA *s, uint64_t addr, XlnxZDMADescr *descr) { @@ -XXX,XX +XXX,XX @@ static void zdma_load_src_descriptor(XlnxZDMA *s) unsigned int ptype = ARRAY_FIELD_EX32(s->regs, ZDMA_CH_CTRL0, POINT_TYPE); if (ptype == PT_REG) { - memcpy(&s->dsc_src, &s->regs[R_ZDMA_CH_SRC_DSCR_WORD0], - sizeof(s->dsc_src)); + zdma_load_descriptor_reg(s, R_ZDMA_CH_SRC_DSCR_WORD0, &s->dsc_src); return; } @@ -XXX,XX +XXX,XX @@ static void zdma_load_dst_descriptor(XlnxZDMA *s) bool dst_type; if (ptype == PT_REG) { - memcpy(&s->dsc_dst, &s->regs[R_ZDMA_CH_DST_DSCR_WORD0], - sizeof(s->dsc_dst)); + zdma_load_descriptor_reg(s, R_ZDMA_CH_DST_DSCR_WORD0, &s->dsc_dst); return; } -- 2.20.1 From: Cameron Esfahani <dirty@apple.com> NRF51_GPIO_REG_CNF_END doesn't actually refer to the start of the last valid CNF register: it's referring to the last byte of the last valid CNF register. This hasn't been a problem up to now, as current implementation in memory.c turns an unaligned 4-byte read from 0x77f to a single byte read and the qtest only looks at the least-significant byte of the register. But when running with patches which fix unaligned accesses in memory.c, the qtest breaks. Considering NRF51 doesn't support unaligned accesses, the simplest fix is to actually set NRF51_GPIO_REG_CNF_END to the start of the last valid CNF register: 0x77c. Now, qtests work with or without the unaligned access patches. Reviewed-by: Cédric Le Goater <clg@kaod.org> Tested-by: Cédric Le Goater <clg@kaod.org> Reviewed-by: Joel Stanley <joel@jms.id.au> Signed-off-by: Cameron Esfahani <dirty@apple.com> Message-id: 51b427f06838622da783d38ba56e3630d6d85c60.1586925392.git.dirty@apple.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/gpio/nrf51_gpio.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/hw/gpio/nrf51_gpio.h b/include/hw/gpio/nrf51_gpio.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/gpio/nrf51_gpio.h +++ b/include/hw/gpio/nrf51_gpio.h @@ -XXX,XX +XXX,XX @@ #define NRF51_GPIO_REG_DIRSET 0x518 #define NRF51_GPIO_REG_DIRCLR 0x51C #define NRF51_GPIO_REG_CNF_START 0x700 -#define NRF51_GPIO_REG_CNF_END 0x77F +#define NRF51_GPIO_REG_CNF_END 0x77C #define NRF51_GPIO_PULLDOWN 1 #define NRF51_GPIO_PULLUP 3 -- 2.20.1 From: Keqian Zhu <zhukeqian1@huawei.com> The KVM_VGIC_ATTR macro expect the second parameter as gicr_typer, of which high 32bit is constructed by mp_affinity. For most case, the high 32bit of mp_affinity is zero, so it will always access the ICC_CTLR_EL1 of CPU0. Signed-off-by: Keqian Zhu <zhukeqian1@huawei.com> Message-id: 20200413091552.62748-2-zhukeqian1@huawei.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/intc/arm_gicv3_kvm.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/hw/intc/arm_gicv3_kvm.c b/hw/intc/arm_gicv3_kvm.c index XXXXXXX..XXXXXXX 100644 --- a/hw/intc/arm_gicv3_kvm.c +++ b/hw/intc/arm_gicv3_kvm.c @@ -XXX,XX +XXX,XX @@ static void kvm_arm_gicv3_get(GICv3State *s) static void arm_gicv3_icc_reset(CPUARMState *env, const ARMCPRegInfo *ri) { - ARMCPU *cpu; GICv3State *s; GICv3CPUState *c; c = (GICv3CPUState *)env->gicv3state; s = c->gic; - cpu = ARM_CPU(c->cpu); c->icc_pmr_el1 = 0; c->icc_bpr[GICV3_G0] = GIC_MIN_BPR; @@ -XXX,XX +XXX,XX @@ static void arm_gicv3_icc_reset(CPUARMState *env, const ARMCPRegInfo *ri) /* Initialize to actual HW supported configuration */ kvm_device_access(s->dev_fd, KVM_DEV_ARM_VGIC_GRP_CPU_SYSREGS, - KVM_VGIC_ATTR(ICC_CTLR_EL1, cpu->mp_affinity), + KVM_VGIC_ATTR(ICC_CTLR_EL1, c->gicr_typer), &c->icc_ctlr_el1[GICV3_NS], false, &error_abort); c->icc_ctlr_el1[GICV3_S] = c->icc_ctlr_el1[GICV3_NS]; -- 2.20.1 From: Keqian Zhu <zhukeqian1@huawei.com> Replace "acpi-mem-hotplug" with "acpi-cpu-hotplug" Signed-off-by: Keqian Zhu <zhukeqian1@huawei.com> Message-id: 20200413091552.62748-4-zhukeqian1@huawei.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/acpi/cpu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/hw/acpi/cpu.c b/hw/acpi/cpu.c index XXXXXXX..XXXXXXX 100644 --- a/hw/acpi/cpu.c +++ b/hw/acpi/cpu.c @@ -XXX,XX +XXX,XX @@ void cpu_hotplug_hw_init(MemoryRegion *as, Object *owner, state->devs[i].arch_id = id_list->cpus[i].arch_id; } memory_region_init_io(&state->ctrl_reg, owner, &cpu_hotplug_ops, state, - "acpi-mem-hotplug", ACPI_CPU_HOTPLUG_REG_LEN); + "acpi-cpu-hotplug", ACPI_CPU_HOTPLUG_REG_LEN); memory_region_add_subregion(as, base_addr, &state->ctrl_reg); } -- 2.20.1 From: Subbaraya Sundeep <sundeep.lkml@gmail.com> Modelled Ethernet MAC of Smartfusion2 SoC. Micrel KSZ8051 PHY is present on Emcraft's SOM kit hence same PHY is emulated. Signed-off-by: Subbaraya Sundeep <sundeep.lkml@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Tested-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 1587048891-30493-2-git-send-email-sundeep.lkml@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/net/Makefile.objs | 1 + include/hw/net/msf2-emac.h | 53 ++++ hw/net/msf2-emac.c | 589 +++++++++++++++++++++++++++++++++++++ MAINTAINERS | 2 + 4 files changed, 645 insertions(+) create mode 100644 include/hw/net/msf2-emac.h create mode 100644 hw/net/msf2-emac.c diff --git a/hw/net/Makefile.objs b/hw/net/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/net/Makefile.objs +++ b/hw/net/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-$(CONFIG_ROCKER) += rocker/rocker.o rocker/rocker_fp.o \ obj-$(call lnot,$(CONFIG_ROCKER)) += rocker/qmp-norocker.o common-obj-$(CONFIG_CAN_BUS) += can/ +common-obj-$(CONFIG_MSF2) += msf2-emac.o diff --git a/include/hw/net/msf2-emac.h b/include/hw/net/msf2-emac.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/net/msf2-emac.h @@ -XXX,XX +XXX,XX @@ +/* + * QEMU model of the Smartfusion2 Ethernet MAC. + * + * Copyright (c) 2020 Subbaraya Sundeep <sundeep.lkml@gmail.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 "hw/sysbus.h" +#include "exec/memory.h" +#include "net/net.h" +#include "net/eth.h" + +#define TYPE_MSS_EMAC "msf2-emac" +#define MSS_EMAC(obj) \ + OBJECT_CHECK(MSF2EmacState, (obj), TYPE_MSS_EMAC) + +#define R_MAX (0x1a0 / 4) +#define PHY_MAX_REGS 32 + +typedef struct MSF2EmacState { + SysBusDevice parent; + + MemoryRegion mmio; + MemoryRegion *dma_mr; + AddressSpace dma_as; + + qemu_irq irq; + NICState *nic; + NICConf conf; + + uint8_t mac_addr[ETH_ALEN]; + uint32_t rx_desc; + uint16_t phy_regs[PHY_MAX_REGS]; + + uint32_t regs[R_MAX]; +} MSF2EmacState; diff --git a/hw/net/msf2-emac.c b/hw/net/msf2-emac.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/net/msf2-emac.c @@ -XXX,XX +XXX,XX @@ +/* + * QEMU model of the Smartfusion2 Ethernet MAC. + * + * Copyright (c) 2020 Subbaraya Sundeep <sundeep.lkml@gmail.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. + * + * Refer to section Ethernet MAC in the document: + * UG0331: SmartFusion2 Microcontroller Subsystem User Guide + * Datasheet URL: + * https://www.microsemi.com/document-portal/cat_view/56661-internal-documents/ + * 56758-soc?lang=en&limit=20&limitstart=220 + */ + +#include "qemu/osdep.h" +#include "qemu-common.h" +#include "qemu/log.h" +#include "qapi/error.h" +#include "exec/address-spaces.h" +#include "hw/registerfields.h" +#include "hw/net/msf2-emac.h" +#include "hw/net/mii.h" +#include "hw/irq.h" +#include "hw/qdev-properties.h" +#include "migration/vmstate.h" + +REG32(CFG1, 0x0) + FIELD(CFG1, RESET, 31, 1) + FIELD(CFG1, RX_EN, 2, 1) + FIELD(CFG1, TX_EN, 0, 1) + FIELD(CFG1, LB_EN, 8, 1) +REG32(CFG2, 0x4) +REG32(IFG, 0x8) +REG32(HALF_DUPLEX, 0xc) +REG32(MAX_FRAME_LENGTH, 0x10) +REG32(MII_CMD, 0x24) + FIELD(MII_CMD, READ, 0, 1) +REG32(MII_ADDR, 0x28) + FIELD(MII_ADDR, REGADDR, 0, 5) + FIELD(MII_ADDR, PHYADDR, 8, 5) +REG32(MII_CTL, 0x2c) +REG32(MII_STS, 0x30) +REG32(STA1, 0x40) +REG32(STA2, 0x44) +REG32(FIFO_CFG0, 0x48) +REG32(FIFO_CFG4, 0x58) + FIELD(FIFO_CFG4, BCAST, 9, 1) + FIELD(FIFO_CFG4, MCAST, 8, 1) +REG32(FIFO_CFG5, 0x5C) + FIELD(FIFO_CFG5, BCAST, 9, 1) + FIELD(FIFO_CFG5, MCAST, 8, 1) +REG32(DMA_TX_CTL, 0x180) + FIELD(DMA_TX_CTL, EN, 0, 1) +REG32(DMA_TX_DESC, 0x184) +REG32(DMA_TX_STATUS, 0x188) + FIELD(DMA_TX_STATUS, PKTCNT, 16, 8) + FIELD(DMA_TX_STATUS, UNDERRUN, 1, 1) + FIELD(DMA_TX_STATUS, PKT_SENT, 0, 1) +REG32(DMA_RX_CTL, 0x18c) + FIELD(DMA_RX_CTL, EN, 0, 1) +REG32(DMA_RX_DESC, 0x190) +REG32(DMA_RX_STATUS, 0x194) + FIELD(DMA_RX_STATUS, PKTCNT, 16, 8) + FIELD(DMA_RX_STATUS, OVERFLOW, 2, 1) + FIELD(DMA_RX_STATUS, PKT_RCVD, 0, 1) +REG32(DMA_IRQ_MASK, 0x198) +REG32(DMA_IRQ, 0x19c) + +#define EMPTY_MASK (1 << 31) +#define PKT_SIZE 0x7FF +#define PHYADDR 0x1 +#define MAX_PKT_SIZE 2048 + +typedef struct { + uint32_t pktaddr; + uint32_t pktsize; + uint32_t next; +} EmacDesc; + +static uint32_t emac_get_isr(MSF2EmacState *s) +{ + uint32_t ier = s->regs[R_DMA_IRQ_MASK]; + uint32_t tx = s->regs[R_DMA_TX_STATUS] & 0xF; + uint32_t rx = s->regs[R_DMA_RX_STATUS] & 0xF; + uint32_t isr = (rx << 4) | tx; + + s->regs[R_DMA_IRQ] = ier & isr; + return s->regs[R_DMA_IRQ]; +} + +static void emac_update_irq(MSF2EmacState *s) +{ + bool intr = emac_get_isr(s); + + qemu_set_irq(s->irq, intr); +} + +static void emac_load_desc(MSF2EmacState *s, EmacDesc *d, hwaddr desc) +{ + address_space_read(&s->dma_as, desc, MEMTXATTRS_UNSPECIFIED, d, sizeof *d); + /* Convert from LE into host endianness. */ + d->pktaddr = le32_to_cpu(d->pktaddr); + d->pktsize = le32_to_cpu(d->pktsize); + d->next = le32_to_cpu(d->next); +} + +static void emac_store_desc(MSF2EmacState *s, EmacDesc *d, hwaddr desc) +{ + /* Convert from host endianness into LE. */ + d->pktaddr = cpu_to_le32(d->pktaddr); + d->pktsize = cpu_to_le32(d->pktsize); + d->next = cpu_to_le32(d->next); + + address_space_write(&s->dma_as, desc, MEMTXATTRS_UNSPECIFIED, d, sizeof *d); +} + +static void msf2_dma_tx(MSF2EmacState *s) +{ + NetClientState *nc = qemu_get_queue(s->nic); + hwaddr desc = s->regs[R_DMA_TX_DESC]; + uint8_t buf[MAX_PKT_SIZE]; + EmacDesc d; + int size; + uint8_t pktcnt; + uint32_t status; + + if (!(s->regs[R_CFG1] & R_CFG1_TX_EN_MASK)) { + return; + } + + while (1) { + emac_load_desc(s, &d, desc); + if (d.pktsize & EMPTY_MASK) { + break; + } + size = d.pktsize & PKT_SIZE; + address_space_read(&s->dma_as, d.pktaddr, MEMTXATTRS_UNSPECIFIED, + buf, size); + /* + * This is very basic way to send packets. Ideally there should be + * a FIFO and packets should be sent out from FIFO only when + * R_CFG1 bit 0 is set. + */ + if (s->regs[R_CFG1] & R_CFG1_LB_EN_MASK) { + nc->info->receive(nc, buf, size); + } else { + qemu_send_packet(nc, buf, size); + } + d.pktsize |= EMPTY_MASK; + emac_store_desc(s, &d, desc); + /* update sent packets count */ + status = s->regs[R_DMA_TX_STATUS]; + pktcnt = FIELD_EX32(status, DMA_TX_STATUS, PKTCNT); + pktcnt++; + s->regs[R_DMA_TX_STATUS] = FIELD_DP32(status, DMA_TX_STATUS, + PKTCNT, pktcnt); + s->regs[R_DMA_TX_STATUS] |= R_DMA_TX_STATUS_PKT_SENT_MASK; + desc = d.next; + } + s->regs[R_DMA_TX_STATUS] |= R_DMA_TX_STATUS_UNDERRUN_MASK; + s->regs[R_DMA_TX_CTL] &= ~R_DMA_TX_CTL_EN_MASK; +} + +static void msf2_phy_update_link(MSF2EmacState *s) +{ + /* Autonegotiation status mirrors link status. */ + if (qemu_get_queue(s->nic)->link_down) { + s->phy_regs[MII_BMSR] &= ~(MII_BMSR_AN_COMP | + MII_BMSR_LINK_ST); + } else { + s->phy_regs[MII_BMSR] |= (MII_BMSR_AN_COMP | + MII_BMSR_LINK_ST); + } +} + +static void msf2_phy_reset(MSF2EmacState *s) +{ + memset(&s->phy_regs[0], 0, sizeof(s->phy_regs)); + s->phy_regs[MII_BMCR] = 0x1140; + s->phy_regs[MII_BMSR] = 0x7968; + s->phy_regs[MII_PHYID1] = 0x0022; + s->phy_regs[MII_PHYID2] = 0x1550; + s->phy_regs[MII_ANAR] = 0x01E1; + s->phy_regs[MII_ANLPAR] = 0xCDE1; + + msf2_phy_update_link(s); +} + +static void write_to_phy(MSF2EmacState *s) +{ + uint8_t reg_addr = s->regs[R_MII_ADDR] & R_MII_ADDR_REGADDR_MASK; + uint8_t phy_addr = (s->regs[R_MII_ADDR] >> R_MII_ADDR_PHYADDR_SHIFT) & + R_MII_ADDR_REGADDR_MASK; + uint16_t data = s->regs[R_MII_CTL] & 0xFFFF; + + if (phy_addr != PHYADDR) { + return; + } + + switch (reg_addr) { + case MII_BMCR: + if (data & MII_BMCR_RESET) { + /* Phy reset */ + msf2_phy_reset(s); + data &= ~MII_BMCR_RESET; + } + if (data & MII_BMCR_AUTOEN) { + /* Complete autonegotiation immediately */ + data &= ~MII_BMCR_AUTOEN; + s->phy_regs[MII_BMSR] |= MII_BMSR_AN_COMP; + } + break; + } + + s->phy_regs[reg_addr] = data; +} + +static uint16_t read_from_phy(MSF2EmacState *s) +{ + uint8_t reg_addr = s->regs[R_MII_ADDR] & R_MII_ADDR_REGADDR_MASK; + uint8_t phy_addr = (s->regs[R_MII_ADDR] >> R_MII_ADDR_PHYADDR_SHIFT) & + R_MII_ADDR_REGADDR_MASK; + + if (phy_addr == PHYADDR) { + return s->phy_regs[reg_addr]; + } else { + return 0xFFFF; + } +} + +static void msf2_emac_do_reset(MSF2EmacState *s) +{ + memset(&s->regs[0], 0, sizeof(s->regs)); + s->regs[R_CFG1] = 0x80000000; + s->regs[R_CFG2] = 0x00007000; + s->regs[R_IFG] = 0x40605060; + s->regs[R_HALF_DUPLEX] = 0x00A1F037; + s->regs[R_MAX_FRAME_LENGTH] = 0x00000600; + s->regs[R_FIFO_CFG5] = 0X3FFFF; + + msf2_phy_reset(s); +} + +static uint64_t emac_read(void *opaque, hwaddr addr, unsigned int size) +{ + MSF2EmacState *s = opaque; + uint32_t r = 0; + + addr >>= 2; + + switch (addr) { + case R_DMA_IRQ: + r = emac_get_isr(s); + break; + default: + if (addr >= ARRAY_SIZE(s->regs)) { + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad offset 0x%" HWADDR_PRIx "\n", __func__, + addr * 4); + return r; + } + r = s->regs[addr]; + break; + } + return r; +} + +static void emac_write(void *opaque, hwaddr addr, uint64_t val64, + unsigned int size) +{ + MSF2EmacState *s = opaque; + uint32_t value = val64; + uint32_t enreqbits; + uint8_t pktcnt; + + addr >>= 2; + switch (addr) { + case R_DMA_TX_CTL: + s->regs[addr] = value; + if (value & R_DMA_TX_CTL_EN_MASK) { + msf2_dma_tx(s); + } + break; + case R_DMA_RX_CTL: + s->regs[addr] = value; + if (value & R_DMA_RX_CTL_EN_MASK) { + s->rx_desc = s->regs[R_DMA_RX_DESC]; + qemu_flush_queued_packets(qemu_get_queue(s->nic)); + } + break; + case R_CFG1: + s->regs[addr] = value; + if (value & R_CFG1_RESET_MASK) { + msf2_emac_do_reset(s); + } + break; + case R_FIFO_CFG0: + /* + * For our implementation, turning on modules is instantaneous, + * so the states requested via the *ENREQ bits appear in the + * *ENRPLY bits immediately. Also the reset bits to reset PE-MCXMAC + * module are not emulated here since it deals with start of frames, + * inter-packet gap and control frames. + */ + enreqbits = extract32(value, 8, 5); + s->regs[addr] = deposit32(value, 16, 5, enreqbits); + break; + case R_DMA_TX_DESC: + if (value & 0x3) { + qemu_log_mask(LOG_GUEST_ERROR, "Tx Descriptor address should be" + " 32 bit aligned\n"); + } + /* Ignore [1:0] bits */ + s->regs[addr] = value & ~3; + break; + case R_DMA_RX_DESC: + if (value & 0x3) { + qemu_log_mask(LOG_GUEST_ERROR, "Rx Descriptor address should be" + " 32 bit aligned\n"); + } + /* Ignore [1:0] bits */ + s->regs[addr] = value & ~3; + break; + case R_DMA_TX_STATUS: + if (value & R_DMA_TX_STATUS_UNDERRUN_MASK) { + s->regs[addr] &= ~R_DMA_TX_STATUS_UNDERRUN_MASK; + } + if (value & R_DMA_TX_STATUS_PKT_SENT_MASK) { + pktcnt = FIELD_EX32(s->regs[addr], DMA_TX_STATUS, PKTCNT); + pktcnt--; + s->regs[addr] = FIELD_DP32(s->regs[addr], DMA_TX_STATUS, + PKTCNT, pktcnt); + if (pktcnt == 0) { + s->regs[addr] &= ~R_DMA_TX_STATUS_PKT_SENT_MASK; + } + } + break; + case R_DMA_RX_STATUS: + if (value & R_DMA_RX_STATUS_OVERFLOW_MASK) { + s->regs[addr] &= ~R_DMA_RX_STATUS_OVERFLOW_MASK; + } + if (value & R_DMA_RX_STATUS_PKT_RCVD_MASK) { + pktcnt = FIELD_EX32(s->regs[addr], DMA_RX_STATUS, PKTCNT); + pktcnt--; + s->regs[addr] = FIELD_DP32(s->regs[addr], DMA_RX_STATUS, + PKTCNT, pktcnt); + if (pktcnt == 0) { + s->regs[addr] &= ~R_DMA_RX_STATUS_PKT_RCVD_MASK; + } + } + break; + case R_DMA_IRQ: + break; + case R_MII_CMD: + if (value & R_MII_CMD_READ_MASK) { + s->regs[R_MII_STS] = read_from_phy(s); + } + break; + case R_MII_CTL: + s->regs[addr] = value; + write_to_phy(s); + break; + case R_STA1: + s->regs[addr] = value; + /* + * R_STA1 [31:24] : octet 1 of mac address + * R_STA1 [23:16] : octet 2 of mac address + * R_STA1 [15:8] : octet 3 of mac address + * R_STA1 [7:0] : octet 4 of mac address + */ + stl_be_p(s->mac_addr, value); + break; + case R_STA2: + s->regs[addr] = value; + /* + * R_STA2 [31:24] : octet 5 of mac address + * R_STA2 [23:16] : octet 6 of mac address + */ + stw_be_p(s->mac_addr + 4, value >> 16); + break; + default: + if (addr >= ARRAY_SIZE(s->regs)) { + qemu_log_mask(LOG_GUEST_ERROR, + "%s: Bad offset 0x%" HWADDR_PRIx "\n", __func__, + addr * 4); + return; + } + s->regs[addr] = value; + break; + } + emac_update_irq(s); +} + +static const MemoryRegionOps emac_ops = { + .read = emac_read, + .write = emac_write, + .endianness = DEVICE_NATIVE_ENDIAN, + .impl = { + .min_access_size = 4, + .max_access_size = 4 + } +}; + +static bool emac_can_rx(NetClientState *nc) +{ + MSF2EmacState *s = qemu_get_nic_opaque(nc); + + return (s->regs[R_CFG1] & R_CFG1_RX_EN_MASK) && + (s->regs[R_DMA_RX_CTL] & R_DMA_RX_CTL_EN_MASK); +} + +static bool addr_filter_ok(MSF2EmacState *s, const uint8_t *buf) +{ + /* The broadcast MAC address: FF:FF:FF:FF:FF:FF */ + const uint8_t broadcast_addr[] = { 0xFF, 0xFF, 0xFF, 0xFF, + 0xFF, 0xFF }; + bool bcast_en = true; + bool mcast_en = true; + + if (s->regs[R_FIFO_CFG5] & R_FIFO_CFG5_BCAST_MASK) { + bcast_en = true; /* Broadcast dont care for drop circuitry */ + } else if (s->regs[R_FIFO_CFG4] & R_FIFO_CFG4_BCAST_MASK) { + bcast_en = false; + } + + if (s->regs[R_FIFO_CFG5] & R_FIFO_CFG5_MCAST_MASK) { + mcast_en = true; /* Multicast dont care for drop circuitry */ + } else if (s->regs[R_FIFO_CFG4] & R_FIFO_CFG4_MCAST_MASK) { + mcast_en = false; + } + + if (!memcmp(buf, broadcast_addr, sizeof(broadcast_addr))) { + return bcast_en; + } + + if (buf[0] & 1) { + return mcast_en; + } + + return !memcmp(buf, s->mac_addr, sizeof(s->mac_addr)); +} + +static ssize_t emac_rx(NetClientState *nc, const uint8_t *buf, size_t size) +{ + MSF2EmacState *s = qemu_get_nic_opaque(nc); + EmacDesc d; + uint8_t pktcnt; + uint32_t status; + + if (size > (s->regs[R_MAX_FRAME_LENGTH] & 0xFFFF)) { + return size; + } + if (!addr_filter_ok(s, buf)) { + return size; + } + + emac_load_desc(s, &d, s->rx_desc); + + if (d.pktsize & EMPTY_MASK) { + address_space_write(&s->dma_as, d.pktaddr, MEMTXATTRS_UNSPECIFIED, + buf, size & PKT_SIZE); + d.pktsize = size & PKT_SIZE; + emac_store_desc(s, &d, s->rx_desc); + /* update received packets count */ + status = s->regs[R_DMA_RX_STATUS]; + pktcnt = FIELD_EX32(status, DMA_RX_STATUS, PKTCNT); + pktcnt++; + s->regs[R_DMA_RX_STATUS] = FIELD_DP32(status, DMA_RX_STATUS, + PKTCNT, pktcnt); + s->regs[R_DMA_RX_STATUS] |= R_DMA_RX_STATUS_PKT_RCVD_MASK; + s->rx_desc = d.next; + } else { + s->regs[R_DMA_RX_CTL] &= ~R_DMA_RX_CTL_EN_MASK; + s->regs[R_DMA_RX_STATUS] |= R_DMA_RX_STATUS_OVERFLOW_MASK; + } + emac_update_irq(s); + return size; +} + +static void msf2_emac_reset(DeviceState *dev) +{ + MSF2EmacState *s = MSS_EMAC(dev); + + msf2_emac_do_reset(s); +} + +static void emac_set_link(NetClientState *nc) +{ + MSF2EmacState *s = qemu_get_nic_opaque(nc); + + msf2_phy_update_link(s); +} + +static NetClientInfo net_msf2_emac_info = { + .type = NET_CLIENT_DRIVER_NIC, + .size = sizeof(NICState), + .can_receive = emac_can_rx, + .receive = emac_rx, + .link_status_changed = emac_set_link, +}; + +static void msf2_emac_realize(DeviceState *dev, Error **errp) +{ + MSF2EmacState *s = MSS_EMAC(dev); + + if (!s->dma_mr) { + error_setg(errp, "MSS_EMAC 'ahb-bus' link not set"); + return; + } + + address_space_init(&s->dma_as, s->dma_mr, "emac-ahb"); + + qemu_macaddr_default_if_unset(&s->conf.macaddr); + s->nic = qemu_new_nic(&net_msf2_emac_info, &s->conf, + object_get_typename(OBJECT(dev)), dev->id, s); + qemu_format_nic_info_str(qemu_get_queue(s->nic), s->conf.macaddr.a); +} + +static void msf2_emac_init(Object *obj) +{ + MSF2EmacState *s = MSS_EMAC(obj); + + sysbus_init_irq(SYS_BUS_DEVICE(obj), &s->irq); + + memory_region_init_io(&s->mmio, obj, &emac_ops, s, + "msf2-emac", R_MAX * 4); + sysbus_init_mmio(SYS_BUS_DEVICE(obj), &s->mmio); +} + +static Property msf2_emac_properties[] = { + DEFINE_PROP_LINK("ahb-bus", MSF2EmacState, dma_mr, + TYPE_MEMORY_REGION, MemoryRegion *), + DEFINE_NIC_PROPERTIES(MSF2EmacState, conf), + DEFINE_PROP_END_OF_LIST(), +}; + +static const VMStateDescription vmstate_msf2_emac = { + .name = TYPE_MSS_EMAC, + .version_id = 1, + .minimum_version_id = 1, + .fields = (VMStateField[]) { + VMSTATE_UINT8_ARRAY(mac_addr, MSF2EmacState, ETH_ALEN), + VMSTATE_UINT32(rx_desc, MSF2EmacState), + VMSTATE_UINT16_ARRAY(phy_regs, MSF2EmacState, PHY_MAX_REGS), + VMSTATE_UINT32_ARRAY(regs, MSF2EmacState, R_MAX), + VMSTATE_END_OF_LIST() + } +}; + +static void msf2_emac_class_init(ObjectClass *klass, void *data) +{ + DeviceClass *dc = DEVICE_CLASS(klass); + + dc->realize = msf2_emac_realize; + dc->reset = msf2_emac_reset; + dc->vmsd = &vmstate_msf2_emac; + device_class_set_props(dc, msf2_emac_properties); +} + +static const TypeInfo msf2_emac_info = { + .name = TYPE_MSS_EMAC, + .parent = TYPE_SYS_BUS_DEVICE, + .instance_size = sizeof(MSF2EmacState), + .instance_init = msf2_emac_init, + .class_init = msf2_emac_class_init, +}; + +static void msf2_emac_register_types(void) +{ + type_register_static(&msf2_emac_info); +} + +type_init(msf2_emac_register_types) diff --git a/MAINTAINERS b/MAINTAINERS index XXXXXXX..XXXXXXX 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -XXX,XX +XXX,XX @@ F: include/hw/arm/msf2-soc.h F: include/hw/misc/msf2-sysreg.h F: include/hw/timer/mss-timer.h F: include/hw/ssi/mss-spi.h +F: hw/net/msf2-emac.c +F: include/hw/net/msf2-emac.h Emcraft M2S-FG484 M: Subbaraya Sundeep <sundeep.lkml@gmail.com> -- 2.20.1 From: Subbaraya Sundeep <sundeep.lkml@gmail.com> With SmartFusion2 Ethernet MAC model in place this patch adds the same to SoC. Signed-off-by: Subbaraya Sundeep <sundeep.lkml@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Tested-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 1587048891-30493-3-git-send-email-sundeep.lkml@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/arm/msf2-soc.h | 2 ++ hw/arm/msf2-soc.c | 26 ++++++++++++++++++++++++-- 2 files changed, 26 insertions(+), 2 deletions(-) diff --git a/include/hw/arm/msf2-soc.h b/include/hw/arm/msf2-soc.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/arm/msf2-soc.h +++ b/include/hw/arm/msf2-soc.h @@ -XXX,XX +XXX,XX @@ #include "hw/timer/mss-timer.h" #include "hw/misc/msf2-sysreg.h" #include "hw/ssi/mss-spi.h" +#include "hw/net/msf2-emac.h" #define TYPE_MSF2_SOC "msf2-soc" #define MSF2_SOC(obj) OBJECT_CHECK(MSF2State, (obj), TYPE_MSF2_SOC) @@ -XXX,XX +XXX,XX @@ typedef struct MSF2State { MSF2SysregState sysreg; MSSTimerState timer; MSSSpiState spi[MSF2_NUM_SPIS]; + MSF2EmacState emac; } MSF2State; #endif diff --git a/hw/arm/msf2-soc.c b/hw/arm/msf2-soc.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/msf2-soc.c +++ b/hw/arm/msf2-soc.c @@ -XXX,XX +XXX,XX @@ /* * SmartFusion2 SoC emulation. * - * Copyright (c) 2017 Subbaraya Sundeep <sundeep.lkml@gmail.com> + * Copyright (c) 2017-2020 Subbaraya Sundeep <sundeep.lkml@gmail.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 @@ -XXX,XX +XXX,XX @@ #define MSF2_TIMER_BASE 0x40004000 #define MSF2_SYSREG_BASE 0x40038000 +#define MSF2_EMAC_BASE 0x40041000 #define ENVM_BASE_ADDRESS 0x60000000 #define SRAM_BASE_ADDRESS 0x20000000 +#define MSF2_EMAC_IRQ 12 + #define MSF2_ENVM_MAX_SIZE (512 * KiB) /* @@ -XXX,XX +XXX,XX @@ static void m2sxxx_soc_initfn(Object *obj) sysbus_init_child_obj(obj, "spi[*]", &s->spi[i], sizeof(s->spi[i]), TYPE_MSS_SPI); } + + sysbus_init_child_obj(obj, "emac", &s->emac, sizeof(s->emac), + TYPE_MSS_EMAC); + if (nd_table[0].used) { + qemu_check_nic_model(&nd_table[0], TYPE_MSS_EMAC); + qdev_set_nic_properties(DEVICE(&s->emac), &nd_table[0]); + } } static void m2sxxx_soc_realize(DeviceState *dev_soc, Error **errp) @@ -XXX,XX +XXX,XX @@ static void m2sxxx_soc_realize(DeviceState *dev_soc, Error **errp) g_free(bus_name); } + dev = DEVICE(&s->emac); + object_property_set_link(OBJECT(&s->emac), OBJECT(get_system_memory()), + "ahb-bus", &error_abort); + object_property_set_bool(OBJECT(&s->emac), true, "realized", &err); + if (err != NULL) { + error_propagate(errp, err); + return; + } + busdev = SYS_BUS_DEVICE(dev); + sysbus_mmio_map(busdev, 0, MSF2_EMAC_BASE); + sysbus_connect_irq(busdev, 0, + qdev_get_gpio_in(armv7m, MSF2_EMAC_IRQ)); + /* Below devices are not modelled yet. */ create_unimplemented_device("i2c_0", 0x40002000, 0x1000); create_unimplemented_device("dma", 0x40003000, 0x1000); @@ -XXX,XX +XXX,XX @@ static void m2sxxx_soc_realize(DeviceState *dev_soc, Error **errp) create_unimplemented_device("can", 0x40015000, 0x1000); create_unimplemented_device("rtc", 0x40017000, 0x1000); create_unimplemented_device("apb_config", 0x40020000, 0x10000); - create_unimplemented_device("emac", 0x40041000, 0x1000); create_unimplemented_device("usb", 0x40043000, 0x1000); } -- 2.20.1 From: Subbaraya Sundeep <sundeep.lkml@gmail.com> In addition to simple serial test this patch uses ping to test the ethernet block modelled in SmartFusion2 SoC. Signed-off-by: Subbaraya Sundeep <sundeep.lkml@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Tested-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Message-id: 1587048891-30493-4-git-send-email-sundeep.lkml@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- tests/acceptance/boot_linux_console.py | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) diff --git a/tests/acceptance/boot_linux_console.py b/tests/acceptance/boot_linux_console.py index XXXXXXX..XXXXXXX 100644 --- a/tests/acceptance/boot_linux_console.py +++ b/tests/acceptance/boot_linux_console.py @@ -XXX,XX +XXX,XX @@ class BootLinuxConsole(Test): """ uboot_url = ('https://raw.githubusercontent.com/' 'Subbaraya-Sundeep/qemu-test-binaries/' - 'fa030bd77a014a0b8e360d3b7011df89283a2f0b/u-boot') - uboot_hash = 'abba5d9c24cdd2d49cdc2a8aa92976cf20737eff' + 'fe371d32e50ca682391e1e70ab98c2942aeffb01/u-boot') + uboot_hash = 'cbb8cbab970f594bf6523b9855be209c08374ae2' uboot_path = self.fetch_asset(uboot_url, asset_hash=uboot_hash) spi_url = ('https://raw.githubusercontent.com/' 'Subbaraya-Sundeep/qemu-test-binaries/' - 'fa030bd77a014a0b8e360d3b7011df89283a2f0b/spi.bin') - spi_hash = '85f698329d38de63aea6e884a86fbde70890a78a' + 'fe371d32e50ca682391e1e70ab98c2942aeffb01/spi.bin') + spi_hash = '65523a1835949b6f4553be96dec1b6a38fb05501' spi_path = self.fetch_asset(spi_url, asset_hash=spi_hash) self.vm.set_console() @@ -XXX,XX +XXX,XX @@ class BootLinuxConsole(Test): '-drive', 'file=' + spi_path + ',if=mtd,format=raw', '-no-reboot') self.vm.launch() - self.wait_for_console_pattern('init started: BusyBox') + self.wait_for_console_pattern('Enter \'help\' for a list') + + exec_command_and_wait_for_pattern(self, 'ifconfig eth0 10.0.2.15', + 'eth0: link becomes ready') + exec_command_and_wait_for_pattern(self, 'ping -c 3 10.0.2.2', + '3 packets transmitted, 3 packets received, 0% packet loss') def do_test_arm_raspi2(self, uart_id): """ -- 2.20.1 This object may be used to represent a clock inside a clock tree. A clock may be connected to another clock so that it receives update, through a callback, whenever the source/parent clock is updated. Although only the root clock of a clock tree controls the values (represented as periods) of all clocks in tree, each clock holds a local state containing the current value so that it can be fetched independently. It will allows us to fullfill migration requirements by migrating each clock independently of others. This is based on the original work of Frederic Konrad. Signed-off-by: Damien Hedde <damien.hedde@greensocs.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20200406135251.157596-2-damien.hedde@greensocs.com [PMM: Use uint64_t rather than unsigned long long in trace events; the dtrace backend can't handle the latter] Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/core/Makefile.objs | 1 + include/hw/clock.h | 216 ++++++++++++++++++++++++++++++++++++++++++ hw/core/clock.c | 130 +++++++++++++++++++++++++ hw/core/trace-events | 7 ++ 4 files changed, 354 insertions(+) create mode 100644 include/hw/clock.h create mode 100644 hw/core/clock.c diff --git a/hw/core/Makefile.objs b/hw/core/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/core/Makefile.objs +++ b/hw/core/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-y += hotplug.o common-obj-y += vmstate-if.o # irq.o needed for qdev GPIO handling: common-obj-y += irq.o +common-obj-y += clock.o common-obj-$(CONFIG_SOFTMMU) += reset.o common-obj-$(CONFIG_SOFTMMU) += qdev-fw.o diff --git a/include/hw/clock.h b/include/hw/clock.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/clock.h @@ -XXX,XX +XXX,XX @@ +/* + * Hardware Clocks + * + * Copyright GreenSocs 2016-2020 + * + * Authors: + * Frederic Konrad + * Damien Hedde + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#ifndef QEMU_HW_CLOCK_H +#define QEMU_HW_CLOCK_H + +#include "qom/object.h" +#include "qemu/queue.h" + +#define TYPE_CLOCK "clock" +#define CLOCK(obj) OBJECT_CHECK(Clock, (obj), TYPE_CLOCK) + +typedef void ClockCallback(void *opaque); + +/* + * clock store a value representing the clock's period in 2^-32ns unit. + * It can represent: + * + periods from 2^-32ns up to 4seconds + * + frequency from ~0.25Hz 2e10Ghz + * Resolution of frequency representation decreases with frequency: + * + at 100MHz, resolution is ~2mHz + * + at 1Ghz, resolution is ~0.2Hz + * + at 10Ghz, resolution is ~20Hz + */ +#define CLOCK_SECOND (1000000000llu << 32) + +/* + * macro helpers to convert to hertz / nanosecond + */ +#define CLOCK_PERIOD_FROM_NS(ns) ((ns) * (CLOCK_SECOND / 1000000000llu)) +#define CLOCK_PERIOD_TO_NS(per) ((per) / (CLOCK_SECOND / 1000000000llu)) +#define CLOCK_PERIOD_FROM_HZ(hz) (((hz) != 0) ? CLOCK_SECOND / (hz) : 0u) +#define CLOCK_PERIOD_TO_HZ(per) (((per) != 0) ? CLOCK_SECOND / (per) : 0u) + +/** + * Clock: + * @parent_obj: parent class + * @period: unsigned integer representing the period of the clock + * @canonical_path: clock path string cache (used for trace purpose) + * @callback: called when clock changes + * @callback_opaque: argument for @callback + * @source: source (or parent in clock tree) of the clock + * @children: list of clocks connected to this one (it is their source) + * @sibling: structure used to form a clock list + */ + +typedef struct Clock Clock; + +struct Clock { + /*< private >*/ + Object parent_obj; + + /* all fields are private and should not be modified directly */ + + /* fields */ + uint64_t period; + char *canonical_path; + ClockCallback *callback; + void *callback_opaque; + + /* Clocks are organized in a clock tree */ + Clock *source; + QLIST_HEAD(, Clock) children; + QLIST_ENTRY(Clock) sibling; +}; + +/** + * clock_setup_canonical_path: + * @clk: clock + * + * compute the canonical path of the clock (used by log messages) + */ +void clock_setup_canonical_path(Clock *clk); + +/** + * clock_set_callback: + * @clk: the clock to register the callback into + * @cb: the callback function + * @opaque: the argument to the callback + * + * Register a callback called on every clock update. + */ +void clock_set_callback(Clock *clk, ClockCallback *cb, void *opaque); + +/** + * clock_clear_callback: + * @clk: the clock to delete the callback from + * + * Unregister the callback registered with clock_set_callback. + */ +void clock_clear_callback(Clock *clk); + +/** + * clock_set_source: + * @clk: the clock. + * @src: the source clock + * + * Setup @src as the clock source of @clk. The current @src period + * value is also copied to @clk and its subtree but no callback is + * called. + * Further @src update will be propagated to @clk and its subtree. + */ +void clock_set_source(Clock *clk, Clock *src); + +/** + * clock_set: + * @clk: the clock to initialize. + * @value: the clock's value, 0 means unclocked + * + * Set the local cached period value of @clk to @value. + */ +void clock_set(Clock *clk, uint64_t value); + +static inline void clock_set_hz(Clock *clk, unsigned hz) +{ + clock_set(clk, CLOCK_PERIOD_FROM_HZ(hz)); +} + +static inline void clock_set_ns(Clock *clk, unsigned ns) +{ + clock_set(clk, CLOCK_PERIOD_FROM_NS(ns)); +} + +/** + * clock_propagate: + * @clk: the clock + * + * Propagate the clock period that has been previously configured using + * @clock_set(). This will update recursively all connected clocks. + * It is an error to call this function on a clock which has a source. + * Note: this function must not be called during device inititialization + * or migration. + */ +void clock_propagate(Clock *clk); + +/** + * clock_update: + * @clk: the clock to update. + * @value: the new clock's value, 0 means unclocked + * + * Update the @clk to the new @value. All connected clocks will be informed + * of this update. This is equivalent to call @clock_set() then + * @clock_propagate(). + */ +static inline void clock_update(Clock *clk, uint64_t value) +{ + clock_set(clk, value); + clock_propagate(clk); +} + +static inline void clock_update_hz(Clock *clk, unsigned hz) +{ + clock_update(clk, CLOCK_PERIOD_FROM_HZ(hz)); +} + +static inline void clock_update_ns(Clock *clk, unsigned ns) +{ + clock_update(clk, CLOCK_PERIOD_FROM_NS(ns)); +} + +/** + * clock_get: + * @clk: the clk to fetch the clock + * + * @return: the current period. + */ +static inline uint64_t clock_get(const Clock *clk) +{ + return clk->period; +} + +static inline unsigned clock_get_hz(Clock *clk) +{ + return CLOCK_PERIOD_TO_HZ(clock_get(clk)); +} + +static inline unsigned clock_get_ns(Clock *clk) +{ + return CLOCK_PERIOD_TO_NS(clock_get(clk)); +} + +/** + * clock_is_enabled: + * @clk: a clock + * + * @return: true if the clock is running. + */ +static inline bool clock_is_enabled(const Clock *clk) +{ + return clock_get(clk) != 0; +} + +static inline void clock_init(Clock *clk, uint64_t value) +{ + clock_set(clk, value); +} +static inline void clock_init_hz(Clock *clk, uint64_t value) +{ + clock_set_hz(clk, value); +} +static inline void clock_init_ns(Clock *clk, uint64_t value) +{ + clock_set_ns(clk, value); +} + +#endif /* QEMU_HW_CLOCK_H */ diff --git a/hw/core/clock.c b/hw/core/clock.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/core/clock.c @@ -XXX,XX +XXX,XX @@ +/* + * Hardware Clocks + * + * Copyright GreenSocs 2016-2020 + * + * Authors: + * Frederic Konrad + * Damien Hedde + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#include "qemu/osdep.h" +#include "hw/clock.h" +#include "trace.h" + +#define CLOCK_PATH(_clk) (_clk->canonical_path) + +void clock_setup_canonical_path(Clock *clk) +{ + g_free(clk->canonical_path); + clk->canonical_path = object_get_canonical_path(OBJECT(clk)); +} + +void clock_set_callback(Clock *clk, ClockCallback *cb, void *opaque) +{ + clk->callback = cb; + clk->callback_opaque = opaque; +} + +void clock_clear_callback(Clock *clk) +{ + clock_set_callback(clk, NULL, NULL); +} + +void clock_set(Clock *clk, uint64_t period) +{ + trace_clock_set(CLOCK_PATH(clk), CLOCK_PERIOD_TO_NS(clk->period), + CLOCK_PERIOD_TO_NS(period)); + clk->period = period; +} + +static void clock_propagate_period(Clock *clk, bool call_callbacks) +{ + Clock *child; + + QLIST_FOREACH(child, &clk->children, sibling) { + if (child->period != clk->period) { + child->period = clk->period; + trace_clock_update(CLOCK_PATH(child), CLOCK_PATH(clk), + CLOCK_PERIOD_TO_NS(clk->period), + call_callbacks); + if (call_callbacks && child->callback) { + child->callback(child->callback_opaque); + } + clock_propagate_period(child, call_callbacks); + } + } +} + +void clock_propagate(Clock *clk) +{ + assert(clk->source == NULL); + trace_clock_propagate(CLOCK_PATH(clk)); + clock_propagate_period(clk, true); +} + +void clock_set_source(Clock *clk, Clock *src) +{ + /* changing clock source is not supported */ + assert(!clk->source); + + trace_clock_set_source(CLOCK_PATH(clk), CLOCK_PATH(src)); + + clk->period = src->period; + QLIST_INSERT_HEAD(&src->children, clk, sibling); + clk->source = src; + clock_propagate_period(clk, false); +} + +static void clock_disconnect(Clock *clk) +{ + if (clk->source == NULL) { + return; + } + + trace_clock_disconnect(CLOCK_PATH(clk)); + + clk->source = NULL; + QLIST_REMOVE(clk, sibling); +} + +static void clock_initfn(Object *obj) +{ + Clock *clk = CLOCK(obj); + + QLIST_INIT(&clk->children); +} + +static void clock_finalizefn(Object *obj) +{ + Clock *clk = CLOCK(obj); + Clock *child, *next; + + /* clear our list of children */ + QLIST_FOREACH_SAFE(child, &clk->children, sibling, next) { + clock_disconnect(child); + } + + /* remove us from source's children list */ + clock_disconnect(clk); + + g_free(clk->canonical_path); +} + +static const TypeInfo clock_info = { + .name = TYPE_CLOCK, + .parent = TYPE_OBJECT, + .instance_size = sizeof(Clock), + .instance_init = clock_initfn, + .instance_finalize = clock_finalizefn, +}; + +static void clock_register_types(void) +{ + type_register_static(&clock_info); +} + +type_init(clock_register_types) diff --git a/hw/core/trace-events b/hw/core/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/core/trace-events +++ b/hw/core/trace-events @@ -XXX,XX +XXX,XX @@ resettable_phase_exit_begin(void *obj, const char *objtype, unsigned count, int resettable_phase_exit_exec(void *obj, const char *objtype, int has_method) "obj=%p(%s) method=%d" resettable_phase_exit_end(void *obj, const char *objtype, unsigned count) "obj=%p(%s) count=%d" resettable_transitional_function(void *obj, const char *objtype) "obj=%p(%s)" + +# clock.c +clock_set_source(const char *clk, const char *src) "'%s', src='%s'" +clock_disconnect(const char *clk) "'%s'" +clock_set(const char *clk, uint64_t old, uint64_t new) "'%s', ns=%"PRIu64"->%"PRIu64 +clock_propagate(const char *clk) "'%s'" +clock_update(const char *clk, const char *src, uint64_t val, int cb) "'%s', src='%s', ns=%"PRIu64", cb=%d" -- 2.20.1 From: Damien Hedde <damien.hedde@greensocs.com> Signed-off-by: Damien Hedde <damien.hedde@greensocs.com> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20200406135251.157596-3-damien.hedde@greensocs.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/core/Makefile.objs | 1 + include/hw/clock.h | 9 +++++++++ hw/core/clock-vmstate.c | 25 +++++++++++++++++++++++++ 3 files changed, 35 insertions(+) create mode 100644 hw/core/clock-vmstate.c diff --git a/hw/core/Makefile.objs b/hw/core/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/core/Makefile.objs +++ b/hw/core/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-$(CONFIG_SOFTMMU) += null-machine.o common-obj-$(CONFIG_SOFTMMU) += loader.o common-obj-$(CONFIG_SOFTMMU) += machine-hmp-cmds.o common-obj-$(CONFIG_SOFTMMU) += numa.o +common-obj-$(CONFIG_SOFTMMU) += clock-vmstate.o obj-$(CONFIG_SOFTMMU) += machine-qmp-cmds.o common-obj-$(CONFIG_EMPTY_SLOT) += empty_slot.o diff --git a/include/hw/clock.h b/include/hw/clock.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/clock.h +++ b/include/hw/clock.h @@ -XXX,XX +XXX,XX @@ struct Clock { QLIST_ENTRY(Clock) sibling; }; +/* + * vmstate description entry to be added in device vmsd. + */ +extern const VMStateDescription vmstate_clock; +#define VMSTATE_CLOCK(field, state) \ + VMSTATE_CLOCK_V(field, state, 0) +#define VMSTATE_CLOCK_V(field, state, version) \ + VMSTATE_STRUCT_POINTER_V(field, state, version, vmstate_clock, Clock) + /** * clock_setup_canonical_path: * @clk: clock diff --git a/hw/core/clock-vmstate.c b/hw/core/clock-vmstate.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/core/clock-vmstate.c @@ -XXX,XX +XXX,XX @@ +/* + * Clock migration structure + * + * Copyright GreenSocs 2019-2020 + * + * Authors: + * Damien Hedde + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#include "qemu/osdep.h" +#include "migration/vmstate.h" +#include "hw/clock.h" + +const VMStateDescription vmstate_clock = { + .name = "clock", + .version_id = 0, + .minimum_version_id = 0, + .fields = (VMStateField[]) { + VMSTATE_UINT64(period, Clock), + VMSTATE_END_OF_LIST() + } +}; -- 2.20.1 From: Damien Hedde <damien.hedde@greensocs.com> Add functions to easily handle clocks with devices. Clock inputs and outputs should be used to handle clock propagation between devices. The API is very similar the GPIO API. This is based on the original work of Frederic Konrad. Signed-off-by: Damien Hedde <damien.hedde@greensocs.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20200406135251.157596-4-damien.hedde@greensocs.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/core/Makefile.objs | 2 +- tests/Makefile.include | 1 + include/hw/qdev-clock.h | 104 +++++++++++++++++++++++++ include/hw/qdev-core.h | 12 +++ hw/core/qdev-clock.c | 168 ++++++++++++++++++++++++++++++++++++++++ hw/core/qdev.c | 12 +++ 6 files changed, 298 insertions(+), 1 deletion(-) create mode 100644 include/hw/qdev-clock.h create mode 100644 hw/core/qdev-clock.c diff --git a/hw/core/Makefile.objs b/hw/core/Makefile.objs index XXXXXXX..XXXXXXX 100644 --- a/hw/core/Makefile.objs +++ b/hw/core/Makefile.objs @@ -XXX,XX +XXX,XX @@ common-obj-y += hotplug.o common-obj-y += vmstate-if.o # irq.o needed for qdev GPIO handling: common-obj-y += irq.o -common-obj-y += clock.o +common-obj-y += clock.o qdev-clock.o common-obj-$(CONFIG_SOFTMMU) += reset.o common-obj-$(CONFIG_SOFTMMU) += qdev-fw.o diff --git a/tests/Makefile.include b/tests/Makefile.include index XXXXXXX..XXXXXXX 100644 --- a/tests/Makefile.include +++ b/tests/Makefile.include @@ -XXX,XX +XXX,XX @@ tests/test-qdev-global-props$(EXESUF): tests/test-qdev-global-props.o \ hw/core/fw-path-provider.o \ hw/core/reset.o \ hw/core/vmstate-if.o \ + hw/core/clock.o hw/core/qdev-clock.o \ $(test-qapi-obj-y) tests/test-vmstate$(EXESUF): tests/test-vmstate.o \ migration/vmstate.o migration/vmstate-types.o migration/qemu-file.o \ diff --git a/include/hw/qdev-clock.h b/include/hw/qdev-clock.h new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/include/hw/qdev-clock.h @@ -XXX,XX +XXX,XX @@ +/* + * Device's clock input and output + * + * Copyright GreenSocs 2016-2020 + * + * Authors: + * Frederic Konrad + * Damien Hedde + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#ifndef QDEV_CLOCK_H +#define QDEV_CLOCK_H + +#include "hw/clock.h" + +/** + * qdev_init_clock_in: + * @dev: the device to add an input clock to + * @name: the name of the clock (can't be NULL). + * @callback: optional callback to be called on update or NULL. + * @opaque: argument for the callback + * @returns: a pointer to the newly added clock + * + * Add an input clock to device @dev as a clock named @name. + * This adds a child<> property. + * The callback will be called with @opaque as opaque parameter. + */ +Clock *qdev_init_clock_in(DeviceState *dev, const char *name, + ClockCallback *callback, void *opaque); + +/** + * qdev_init_clock_out: + * @dev: the device to add an output clock to + * @name: the name of the clock (can't be NULL). + * @returns: a pointer to the newly added clock + * + * Add an output clock to device @dev as a clock named @name. + * This adds a child<> property. + */ +Clock *qdev_init_clock_out(DeviceState *dev, const char *name); + +/** + * qdev_get_clock_in: + * @dev: the device which has the clock + * @name: the name of the clock (can't be NULL). + * @returns: a pointer to the clock + * + * Get the input clock @name from @dev or NULL if does not exist. + */ +Clock *qdev_get_clock_in(DeviceState *dev, const char *name); + +/** + * qdev_get_clock_out: + * @dev: the device which has the clock + * @name: the name of the clock (can't be NULL). + * @returns: a pointer to the clock + * + * Get the output clock @name from @dev or NULL if does not exist. + */ +Clock *qdev_get_clock_out(DeviceState *dev, const char *name); + +/** + * qdev_connect_clock_in: + * @dev: a device + * @name: the name of an input clock in @dev + * @source: the source clock (an output clock of another device for example) + * + * Set the source clock of input clock @name of device @dev to @source. + * @source period update will be propagated to @name clock. + */ +static inline void qdev_connect_clock_in(DeviceState *dev, const char *name, + Clock *source) +{ + clock_set_source(qdev_get_clock_in(dev, name), source); +} + +/** + * qdev_alias_clock: + * @dev: the device which has the clock + * @name: the name of the clock in @dev (can't be NULL) + * @alias_dev: the device to add the clock + * @alias_name: the name of the clock in @container + * @returns: a pointer to the clock + * + * Add a clock @alias_name in @alias_dev which is an alias of the clock @name + * in @dev. The direction _in_ or _out_ will the same as the original. + * An alias clock must not be modified or used by @alias_dev and should + * typically be only only for device composition purpose. + */ +Clock *qdev_alias_clock(DeviceState *dev, const char *name, + DeviceState *alias_dev, const char *alias_name); + +/** + * qdev_finalize_clocklist: + * @dev: the device being finalized + * + * Clear the clocklist from @dev. Only used internally in qdev. + */ +void qdev_finalize_clocklist(DeviceState *dev); + +#endif /* QDEV_CLOCK_H */ diff --git a/include/hw/qdev-core.h b/include/hw/qdev-core.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/qdev-core.h +++ b/include/hw/qdev-core.h @@ -XXX,XX +XXX,XX @@ struct NamedGPIOList { QLIST_ENTRY(NamedGPIOList) node; }; +typedef struct Clock Clock; +typedef struct NamedClockList NamedClockList; + +struct NamedClockList { + char *name; + Clock *clock; + bool output; + bool alias; + QLIST_ENTRY(NamedClockList) node; +}; + /** * DeviceState: * @realized: Indicates whether the device has been fully constructed. @@ -XXX,XX +XXX,XX @@ struct DeviceState { bool allow_unplug_during_migration; BusState *parent_bus; QLIST_HEAD(, NamedGPIOList) gpios; + QLIST_HEAD(, NamedClockList) clocks; QLIST_HEAD(, BusState) child_bus; int num_child_bus; int instance_id_alias; diff --git a/hw/core/qdev-clock.c b/hw/core/qdev-clock.c new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/hw/core/qdev-clock.c @@ -XXX,XX +XXX,XX @@ +/* + * Device's clock input and output + * + * Copyright GreenSocs 2016-2020 + * + * Authors: + * Frederic Konrad + * Damien Hedde + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + */ + +#include "qemu/osdep.h" +#include "hw/qdev-clock.h" +#include "hw/qdev-core.h" +#include "qapi/error.h" + +/* + * qdev_init_clocklist: + * Add a new clock in a device + */ +static NamedClockList *qdev_init_clocklist(DeviceState *dev, const char *name, + bool output, Clock *clk) +{ + NamedClockList *ncl; + + /* + * Clock must be added before realize() so that we can compute the + * clock's canonical path during device_realize(). + */ + assert(!dev->realized); + + /* + * The ncl structure is freed by qdev_finalize_clocklist() which will + * be called during @dev's device_finalize(). + */ + ncl = g_new0(NamedClockList, 1); + ncl->name = g_strdup(name); + ncl->output = output; + ncl->alias = (clk != NULL); + + /* + * Trying to create a clock whose name clashes with some other + * clock or property is a bug in the caller and we will abort(). + */ + if (clk == NULL) { + clk = CLOCK(object_new(TYPE_CLOCK)); + object_property_add_child(OBJECT(dev), name, OBJECT(clk), &error_abort); + if (output) { + /* + * Remove object_new()'s initial reference. + * Note that for inputs, the reference created by object_new() + * will be deleted in qdev_finalize_clocklist(). + */ + object_unref(OBJECT(clk)); + } + } else { + object_property_add_link(OBJECT(dev), name, + object_get_typename(OBJECT(clk)), + (Object **) &ncl->clock, + NULL, OBJ_PROP_LINK_STRONG, &error_abort); + } + + ncl->clock = clk; + + QLIST_INSERT_HEAD(&dev->clocks, ncl, node); + return ncl; +} + +void qdev_finalize_clocklist(DeviceState *dev) +{ + /* called by @dev's device_finalize() */ + NamedClockList *ncl, *ncl_next; + + QLIST_FOREACH_SAFE(ncl, &dev->clocks, node, ncl_next) { + QLIST_REMOVE(ncl, node); + if (!ncl->output && !ncl->alias) { + /* + * We kept a reference on the input clock to ensure it lives up to + * this point so we can safely remove the callback. + * It avoids having a callback to a deleted object if ncl->clock + * is still referenced somewhere else (eg: by a clock output). + */ + clock_clear_callback(ncl->clock); + object_unref(OBJECT(ncl->clock)); + } + g_free(ncl->name); + g_free(ncl); + } +} + +Clock *qdev_init_clock_out(DeviceState *dev, const char *name) +{ + NamedClockList *ncl; + + assert(name); + + ncl = qdev_init_clocklist(dev, name, true, NULL); + + return ncl->clock; +} + +Clock *qdev_init_clock_in(DeviceState *dev, const char *name, + ClockCallback *callback, void *opaque) +{ + NamedClockList *ncl; + + assert(name); + + ncl = qdev_init_clocklist(dev, name, false, NULL); + + if (callback) { + clock_set_callback(ncl->clock, callback, opaque); + } + return ncl->clock; +} + +static NamedClockList *qdev_get_clocklist(DeviceState *dev, const char *name) +{ + NamedClockList *ncl; + + QLIST_FOREACH(ncl, &dev->clocks, node) { + if (strcmp(name, ncl->name) == 0) { + return ncl; + } + } + + return NULL; +} + +Clock *qdev_get_clock_in(DeviceState *dev, const char *name) +{ + NamedClockList *ncl; + + assert(name); + + ncl = qdev_get_clocklist(dev, name); + assert(!ncl->output); + + return ncl->clock; +} + +Clock *qdev_get_clock_out(DeviceState *dev, const char *name) +{ + NamedClockList *ncl; + + assert(name); + + ncl = qdev_get_clocklist(dev, name); + assert(ncl->output); + + return ncl->clock; +} + +Clock *qdev_alias_clock(DeviceState *dev, const char *name, + DeviceState *alias_dev, const char *alias_name) +{ + NamedClockList *ncl; + + assert(name && alias_name); + + ncl = qdev_get_clocklist(dev, name); + + qdev_init_clocklist(alias_dev, alias_name, ncl->output, ncl->clock); + + return ncl->clock; +} diff --git a/hw/core/qdev.c b/hw/core/qdev.c index XXXXXXX..XXXXXXX 100644 --- a/hw/core/qdev.c +++ b/hw/core/qdev.c @@ -XXX,XX +XXX,XX @@ #include "hw/qdev-properties.h" #include "hw/boards.h" #include "hw/sysbus.h" +#include "hw/qdev-clock.h" #include "migration/vmstate.h" #include "trace.h" @@ -XXX,XX +XXX,XX @@ static void device_set_realized(Object *obj, bool value, Error **errp) DeviceClass *dc = DEVICE_GET_CLASS(dev); HotplugHandler *hotplug_ctrl; BusState *bus; + NamedClockList *ncl; Error *local_err = NULL; bool unattached_parent = false; static int unattached_count; @@ -XXX,XX +XXX,XX @@ static void device_set_realized(Object *obj, bool value, Error **errp) */ g_free(dev->canonical_path); dev->canonical_path = object_get_canonical_path(OBJECT(dev)); + QLIST_FOREACH(ncl, &dev->clocks, node) { + if (ncl->alias) { + continue; + } else { + clock_setup_canonical_path(ncl->clock); + } + } if (qdev_get_vmsd(dev)) { if (vmstate_register_with_alias_id(VMSTATE_IF(dev), @@ -XXX,XX +XXX,XX @@ static void device_initfn(Object *obj) dev->allow_unplug_during_migration = false; QLIST_INIT(&dev->gpios); + QLIST_INIT(&dev->clocks); } static void device_post_init(Object *obj) @@ -XXX,XX +XXX,XX @@ static void device_finalize(Object *obj) */ } + qdev_finalize_clocklist(dev); + /* Only send event if the device had been completely realized */ if (dev->pending_deleted_event) { g_assert(dev->canonical_path); -- 2.20.1 From: Damien Hedde <damien.hedde@greensocs.com> Introduce a function and macro helpers to setup several clocks in a device from a static array description. An element of the array describes the clock (name and direction) as well as the related callback and an optional offset to store the created object pointer in the device state structure. The array must be terminated by a special element QDEV_CLOCK_END. This is based on the original work of Frederic Konrad. Signed-off-by: Damien Hedde <damien.hedde@greensocs.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20200406135251.157596-5-damien.hedde@greensocs.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/qdev-clock.h | 55 +++++++++++++++++++++++++++++++++++++++++ hw/core/qdev-clock.c | 17 +++++++++++++ 2 files changed, 72 insertions(+) diff --git a/include/hw/qdev-clock.h b/include/hw/qdev-clock.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/qdev-clock.h +++ b/include/hw/qdev-clock.h @@ -XXX,XX +XXX,XX @@ Clock *qdev_alias_clock(DeviceState *dev, const char *name, */ void qdev_finalize_clocklist(DeviceState *dev); +/** + * ClockPortInitElem: + * @name: name of the clock (can't be NULL) + * @output: indicates whether the clock is input or output + * @callback: for inputs, optional callback to be called on clock's update + * with device as opaque + * @offset: optional offset to store the ClockIn or ClockOut pointer in device + * state structure (0 means unused) + */ +struct ClockPortInitElem { + const char *name; + bool is_output; + ClockCallback *callback; + size_t offset; +}; + +#define clock_offset_value(devstate, field) \ + (offsetof(devstate, field) + \ + type_check(Clock *, typeof_field(devstate, field))) + +#define QDEV_CLOCK(out_not_in, devstate, field, cb) { \ + .name = (stringify(field)), \ + .is_output = out_not_in, \ + .callback = cb, \ + .offset = clock_offset_value(devstate, field), \ +} + +/** + * QDEV_CLOCK_(IN|OUT): + * @devstate: structure type. @dev argument of qdev_init_clocks below must be + * a pointer to that same type. + * @field: a field in @_devstate (must be Clock*) + * @callback: (for input only) callback (or NULL) to be called with the device + * state as argument + * + * The name of the clock will be derived from @field + */ +#define QDEV_CLOCK_IN(devstate, field, callback) \ + QDEV_CLOCK(false, devstate, field, callback) + +#define QDEV_CLOCK_OUT(devstate, field) \ + QDEV_CLOCK(true, devstate, field, NULL) + +#define QDEV_CLOCK_END { .name = NULL } + +typedef struct ClockPortInitElem ClockPortInitArray[]; + +/** + * qdev_init_clocks: + * @dev: the device to add clocks to + * @clocks: a QDEV_CLOCK_END-terminated array which contains the + * clocks information. + */ +void qdev_init_clocks(DeviceState *dev, const ClockPortInitArray clocks); + #endif /* QDEV_CLOCK_H */ diff --git a/hw/core/qdev-clock.c b/hw/core/qdev-clock.c index XXXXXXX..XXXXXXX 100644 --- a/hw/core/qdev-clock.c +++ b/hw/core/qdev-clock.c @@ -XXX,XX +XXX,XX @@ Clock *qdev_init_clock_in(DeviceState *dev, const char *name, return ncl->clock; } +void qdev_init_clocks(DeviceState *dev, const ClockPortInitArray clocks) +{ + const struct ClockPortInitElem *elem; + + for (elem = &clocks[0]; elem->name != NULL; elem++) { + Clock **clkp; + /* offset cannot be inside the DeviceState part */ + assert(elem->offset > sizeof(DeviceState)); + clkp = (Clock **)(((void *) dev) + elem->offset); + if (elem->is_output) { + *clkp = qdev_init_clock_out(dev, elem->name); + } else { + *clkp = qdev_init_clock_in(dev, elem->name, elem->callback, dev); + } + } +} + static NamedClockList *qdev_get_clocklist(DeviceState *dev, const char *name) { NamedClockList *ncl; -- 2.20.1 Add the documentation about the clock inputs and outputs in devices. This is based on the original work of Frederic Konrad. Signed-off-by: Damien Hedde <damien.hedde@greensocs.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20200406135251.157596-6-damien.hedde@greensocs.com [PMM: Editing pass for minor grammar, style and Sphinx formatting fixes] Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- docs/devel/clocks.rst | 391 ++++++++++++++++++++++++++++++++++++++++++ docs/devel/index.rst | 1 + 2 files changed, 392 insertions(+) create mode 100644 docs/devel/clocks.rst diff --git a/docs/devel/clocks.rst b/docs/devel/clocks.rst new file mode 100644 index XXXXXXX..XXXXXXX --- /dev/null +++ b/docs/devel/clocks.rst @@ -XXX,XX +XXX,XX @@ +Modelling a clock tree in QEMU +============================== + +What are clocks? +---------------- + +Clocks are QOM objects developed for the purpose of modelling the +distribution of clocks in QEMU. + +They allow us to model the clock distribution of a platform and detect +configuration errors in the clock tree such as badly configured PLL, clock +source selection or disabled clock. + +The object is *Clock* and its QOM name is ``clock`` (in C code, the macro +``TYPE_CLOCK``). + +Clocks are typically used with devices where they are used to model inputs +and outputs. They are created in a similar way to GPIOs. Inputs and outputs +of different devices can be connected together. + +In these cases a Clock object is a child of a Device object, but this +is not a requirement. Clocks can be independent of devices. For +example it is possible to create a clock outside of any device to +model the main clock source of a machine. + +Here is an example of clocks:: + + +---------+ +----------------------+ +--------------+ + | Clock 1 | | Device B | | Device C | + | | | +-------+ +-------+ | | +-------+ | + | |>>-+-->>|Clock 2| |Clock 3|>>--->>|Clock 6| | + +---------+ | | | (in) | | (out) | | | | (in) | | + | | +-------+ +-------+ | | +-------+ | + | | +-------+ | +--------------+ + | | |Clock 4|>> + | | | (out) | | +--------------+ + | | +-------+ | | Device D | + | | +-------+ | | +-------+ | + | | |Clock 5|>>--->>|Clock 7| | + | | | (out) | | | | (in) | | + | | +-------+ | | +-------+ | + | +----------------------+ | | + | | +-------+ | + +----------------------------->>|Clock 8| | + | | (in) | | + | +-------+ | + +--------------+ + +Clocks are defined in the ``include/hw/clock.h`` header and device +related functions are defined in the ``include/hw/qdev-clock.h`` +header. + +The clock state +--------------- + +The state of a clock is its period; it is stored as an integer +representing it in units of 2 :sup:`-32` ns. The special value of 0 is used to +represent the clock being inactive or gated. The clocks do not model +the signal itself (pin toggling) or other properties such as the duty +cycle. + +All clocks contain this state: outputs as well as inputs. This allows +the current period of a clock to be fetched at any time. When a clock +is updated, the value is immediately propagated to all connected +clocks in the tree. + +To ease interaction with clocks, helpers with a unit suffix are defined for +every clock state setter or getter. The suffixes are: + +- ``_ns`` for handling periods in nanoseconds +- ``_hz`` for handling frequencies in hertz + +The 0 period value is converted to 0 in hertz and vice versa. 0 always means +that the clock is disabled. + +Adding a new clock +------------------ + +Adding clocks to a device must be done during the init method of the Device +instance. + +To add an input clock to a device, the function ``qdev_init_clock_in()`` +must be used. It takes the name, a callback and an opaque parameter +for the callback (this will be explained in a following section). +Output is simpler; only the name is required. Typically:: + + qdev_init_clock_in(DEVICE(dev), "clk_in", clk_in_callback, dev); + qdev_init_clock_out(DEVICE(dev), "clk_out"); + +Both functions return the created Clock pointer, which should be saved in the +device's state structure for further use. + +These objects will be automatically deleted by the QOM reference mechanism. + +Note that it is possible to create a static array describing clock inputs and +outputs. The function ``qdev_init_clocks()`` must be called with the array as +parameter to initialize the clocks: it has the same behaviour as calling the +``qdev_init_clock_in/out()`` for each clock in the array. To ease the array +construction, some macros are defined in ``include/hw/qdev-clock.h``. +As an example, the following creates 2 clocks to a device: one input and one +output. + +.. code-block:: c + + /* device structure containing pointers to the clock objects */ + typedef struct MyDeviceState { + DeviceState parent_obj; + Clock *clk_in; + Clock *clk_out; + } MyDeviceState; + + /* + * callback for the input clock (see "Callback on input clock + * change" section below for more information). + */ + static void clk_in_callback(void *opaque); + + /* + * static array describing clocks: + * + a clock input named "clk_in", whose pointer is stored in + * the clk_in field of a MyDeviceState structure with callback + * clk_in_callback. + * + a clock output named "clk_out" whose pointer is stored in + * the clk_out field of a MyDeviceState structure. + */ + static const ClockPortInitArray mydev_clocks = { + QDEV_CLOCK_IN(MyDeviceState, clk_in, clk_in_callback), + QDEV_CLOCK_OUT(MyDeviceState, clk_out), + QDEV_CLOCK_END + }; + + /* device initialization function */ + static void mydev_init(Object *obj) + { + /* cast to MyDeviceState */ + MyDeviceState *mydev = MYDEVICE(obj); + /* create and fill the pointer fields in the MyDeviceState */ + qdev_init_clocks(mydev, mydev_clocks); + [...] + } + +An alternative way to create a clock is to simply call +``object_new(TYPE_CLOCK)``. In that case the clock will neither be an +input nor an output of a device. After the whole QOM hierarchy of the +clock has been set ``clock_setup_canonical_path()`` should be called. + +At creation, the period of the clock is 0: the clock is disabled. You can +change it using ``clock_set_ns()`` or ``clock_set_hz()``. + +Note that if you are creating a clock with a fixed period which will never +change (for example the main clock source of a board), then you'll have +nothing else to do. This value will be propagated to other clocks when +connecting the clocks together and devices will fetch the right value during +the first reset. + +Retrieving clocks from a device +------------------------------- + +``qdev_get_clock_in()`` and ``dev_get_clock_out()`` are available to +get the clock inputs or outputs of a device. For example: + +.. code-block:: c + + Clock *clk = qdev_get_clock_in(DEVICE(mydev), "clk_in"); + +or: + +.. code-block:: c + + Clock *clk = qdev_get_clock_out(DEVICE(mydev), "clk_out"); + +Connecting two clocks together +------------------------------ + +To connect two clocks together, use the ``clock_set_source()`` function. +Given two clocks ``clk1``, and ``clk2``, ``clock_set_source(clk2, clk1);`` +configures ``clk2`` to follow the ``clk1`` period changes. Every time ``clk1`` +is updated, ``clk2`` will be updated too. + +When connecting clock between devices, prefer using the +``qdev_connect_clock_in()`` function to set the source of an input +device clock. For example, to connect the input clock ``clk2`` of +``devB`` to the output clock ``clk1`` of ``devA``, do: + +.. code-block:: c + + qdev_connect_clock_in(devB, "clk2", qdev_get_clock_out(devA, "clk1")) + +We used ``qdev_get_clock_out()`` above, but any clock can drive an +input clock, even another input clock. The following diagram shows +some examples of connections. Note also that a clock can drive several +other clocks. + +:: + + +------------+ +--------------------------------------------------+ + | Device A | | Device B | + | | | +---------------------+ | + | | | | Device C | | + | +-------+ | | +-------+ | +-------+ +-------+ | +-------+ | + | |Clock 1|>>-->>|Clock 2|>>+-->>|Clock 3| |Clock 5|>>>>|Clock 6|>> + | | (out) | | | | (in) | | | | (in) | | (out) | | | (out) | | + | +-------+ | | +-------+ | | +-------+ +-------+ | +-------+ | + +------------+ | | +---------------------+ | + | | | + | | +--------------+ | + | | | Device D | | + | | | +-------+ | | + | +-->>|Clock 4| | | + | | | (in) | | | + | | +-------+ | | + | +--------------+ | + +--------------------------------------------------+ + +In the above example, when *Clock 1* is updated by *Device A*, three +clocks get the new clock period value: *Clock 2*, *Clock 3* and *Clock 4*. + +It is not possible to disconnect a clock or to change the clock connection +after it is connected. + +Unconnected input clocks +------------------------ + +A newly created input clock is disabled (period of 0). This means the +clock will be considered as disabled until the period is updated. If +the clock remains unconnected it will always keep its initial value +of 0. If this is not the desired behaviour, ``clock_set()``, +``clock_set_ns()`` or ``clock_set_hz()`` should be called on the Clock +object during device instance init. For example: + +.. code-block:: c + + clk = qdev_init_clock_in(DEVICE(dev), "clk-in", clk_in_callback, + dev); + /* set initial value to 10ns / 100MHz */ + clock_set_ns(clk, 10); + +Fetching clock frequency/period +------------------------------- + +To get the current state of a clock, use the functions ``clock_get()``, +``clock_get_ns()`` or ``clock_get_hz()``. + +It is also possible to register a callback on clock frequency changes. +Here is an example: + +.. code-block:: c + + void clock_callback(void *opaque) { + MyDeviceState *s = (MyDeviceState *) opaque; + /* + * 'opaque' is the argument passed to qdev_init_clock_in(); + * usually this will be the device state pointer. + */ + + /* do something with the new period */ + fprintf(stdout, "device new period is %" PRIu64 "ns\n", + clock_get_ns(dev->my_clk_input)); + } + +Changing a clock period +----------------------- + +A device can change its outputs using the ``clock_update()``, +``clock_update_ns()`` or ``clock_update_hz()`` function. It will trigger +updates on every connected input. + +For example, let's say that we have an output clock *clkout* and we +have a pointer to it in the device state because we did the following +in init phase: + +.. code-block:: c + + dev->clkout = qdev_init_clock_out(DEVICE(dev), "clkout"); + +Then at any time (apart from the cases listed below), it is possible to +change the clock value by doing: + +.. code-block:: c + + clock_update_hz(dev->clkout, 1000 * 1000 * 1000); /* 1GHz */ + +Because updating a clock may trigger any side effects through +connected clocks and their callbacks, this operation must be done +while holding the qemu io lock. + +For the same reason, one can update clocks only when it is allowed to have +side effects on other objects. In consequence, it is forbidden: + +* during migration, +* and in the enter phase of reset. + +Note that calling ``clock_update[_ns|_hz]()`` is equivalent to calling +``clock_set[_ns|_hz]()`` (with the same arguments) then +``clock_propagate()`` on the clock. Thus, setting the clock value can +be separated from triggering the side-effects. This is often required +to factorize code to handle reset and migration in devices. + +Aliasing clocks +--------------- + +Sometimes, one needs to forward, or inherit, a clock from another +device. Typically, when doing device composition, a device might +expose a sub-device's clock without interfering with it. The function +``qdev_alias_clock()`` can be used to achieve this behaviour. Note +that it is possible to expose the clock under a different name. +``qdev_alias_clock()`` works for both input and output clocks. + +For example, if device B is a child of device A, +``device_a_instance_init()`` may do something like this: + +.. code-block:: c + + void device_a_instance_init(Object *obj) + { + AState *A = DEVICE_A(obj); + BState *B; + /* create object B as child of A */ + [...] + qdev_alias_clock(B, "clk", A, "b_clk"); + /* + * Now A has a clock "b_clk" which is an alias to + * the clock "clk" of its child B. + */ + } + +This function does not return any clock object. The new clock has the +same direction (input or output) as the original one. This function +only adds a link to the existing clock. In the above example, object B +remains the only object allowed to use the clock and device A must not +try to change the clock period or set a callback to the clock. This +diagram describes the example with an input clock:: + + +--------------------------+ + | Device A | + | +--------------+ | + | | Device B | | + | | +-------+ | | + >>"b_clk">>>| "clk" | | | + | (in) | | (in) | | | + | | +-------+ | | + | +--------------+ | + +--------------------------+ + +Migration +--------- + +Clock state is not migrated automatically. Every device must handle its +clock migration. Alias clocks must not be migrated. + +To ensure clock states are restored correctly during migration, there +are two solutions. + +Clock states can be migrated by adding an entry into the device +vmstate description. You should use the ``VMSTATE_CLOCK`` macro for this. +This is typically used to migrate an input clock state. For example: + +.. code-block:: c + + MyDeviceState { + DeviceState parent_obj; + [...] /* some fields */ + Clock *clk; + }; + + VMStateDescription my_device_vmstate = { + .name = "my_device", + .fields = (VMStateField[]) { + [...], /* other migrated fields */ + VMSTATE_CLOCK(clk, MyDeviceState), + VMSTATE_END_OF_LIST() + } + }; + +The second solution is to restore the clock state using information already +at our disposal. This can be used to restore output clock states using the +device state. The functions ``clock_set[_ns|_hz]()`` can be used during the +``post_load()`` migration callback. + +When adding clock support to an existing device, if you care about +migration compatibility you will need to be careful, as simply adding +a ``VMSTATE_CLOCK()`` line will break compatibility. Instead, you can +put the ``VMSTATE_CLOCK()`` line into a vmstate subsection with a +suitable ``needed`` function, and use ``clock_set()`` in a +``pre_load()`` function to set the default value that will be used if +the source virtual machine in the migration does not send the clock +state. + +Care should be taken not to use ``clock_update[_ns|_hz]()`` or +``clock_propagate()`` during the whole migration procedure because it +will trigger side effects to other devices in an unknown state. diff --git a/docs/devel/index.rst b/docs/devel/index.rst index XXXXXXX..XXXXXXX 100644 --- a/docs/devel/index.rst +++ b/docs/devel/index.rst @@ -XXX,XX +XXX,XX @@ Contents: bitops reset s390-dasd-ipl + clocks -- 2.20.1 From: Damien Hedde <damien.hedde@greensocs.com> Add some clocks to zynq_slcr + the main input clock (ps_clk) + the reference clock outputs for each uart (uart0 & 1) This commit also transitional the slcr to multi-phase reset as it is required to initialize the clocks correctly. The clock frequencies are computed using the internal pll & uart configuration registers and the input ps_clk frequency. Signed-off-by: Damien Hedde <damien.hedde@greensocs.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Acked-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20200406135251.157596-7-damien.hedde@greensocs.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/misc/zynq_slcr.c | 172 ++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 168 insertions(+), 4 deletions(-) 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 @@ #include "qemu/log.h" #include "qemu/module.h" #include "hw/registerfields.h" +#include "hw/qdev-clock.h" #ifndef ZYNQ_SLCR_ERR_DEBUG #define ZYNQ_SLCR_ERR_DEBUG 0 @@ -XXX,XX +XXX,XX @@ REG32(LOCKSTA, 0x00c) REG32(ARM_PLL_CTRL, 0x100) REG32(DDR_PLL_CTRL, 0x104) REG32(IO_PLL_CTRL, 0x108) +/* fields for [ARM|DDR|IO]_PLL_CTRL registers */ + FIELD(xxx_PLL_CTRL, PLL_RESET, 0, 1) + FIELD(xxx_PLL_CTRL, PLL_PWRDWN, 1, 1) + FIELD(xxx_PLL_CTRL, PLL_BYPASS_QUAL, 3, 1) + FIELD(xxx_PLL_CTRL, PLL_BYPASS_FORCE, 4, 1) + FIELD(xxx_PLL_CTRL, PLL_FPDIV, 12, 7) REG32(PLL_STATUS, 0x10c) REG32(ARM_PLL_CFG, 0x110) REG32(DDR_PLL_CFG, 0x114) @@ -XXX,XX +XXX,XX @@ REG32(SMC_CLK_CTRL, 0x148) REG32(LQSPI_CLK_CTRL, 0x14c) REG32(SDIO_CLK_CTRL, 0x150) REG32(UART_CLK_CTRL, 0x154) + FIELD(UART_CLK_CTRL, CLKACT0, 0, 1) + FIELD(UART_CLK_CTRL, CLKACT1, 1, 1) + FIELD(UART_CLK_CTRL, SRCSEL, 4, 2) + FIELD(UART_CLK_CTRL, DIVISOR, 8, 6) REG32(SPI_CLK_CTRL, 0x158) REG32(CAN_CLK_CTRL, 0x15c) REG32(CAN_MIOCLK_CTRL, 0x160) @@ -XXX,XX +XXX,XX @@ typedef struct ZynqSLCRState { MemoryRegion iomem; uint32_t regs[ZYNQ_SLCR_NUM_REGS]; + + Clock *ps_clk; + Clock *uart0_ref_clk; + Clock *uart1_ref_clk; } ZynqSLCRState; -static void zynq_slcr_reset(DeviceState *d) +/* + * return the output frequency of ARM/DDR/IO pll + * using input frequency and PLL_CTRL register + */ +static uint64_t zynq_slcr_compute_pll(uint64_t input, uint32_t ctrl_reg) { - ZynqSLCRState *s = ZYNQ_SLCR(d); + uint32_t mult = ((ctrl_reg & R_xxx_PLL_CTRL_PLL_FPDIV_MASK) >> + R_xxx_PLL_CTRL_PLL_FPDIV_SHIFT); + + /* first, check if pll is bypassed */ + if (ctrl_reg & R_xxx_PLL_CTRL_PLL_BYPASS_FORCE_MASK) { + return input; + } + + /* is pll disabled ? */ + if (ctrl_reg & (R_xxx_PLL_CTRL_PLL_RESET_MASK | + R_xxx_PLL_CTRL_PLL_PWRDWN_MASK)) { + return 0; + } + + /* frequency multiplier -> period division */ + return input / mult; +} + +/* + * return the output period of a clock given: + * + the periods in an array corresponding to input mux selector + * + the register xxx_CLK_CTRL value + * + enable bit index in ctrl register + * + * This function makes the assumption that the ctrl_reg value is organized as + * follows: + * + bits[13:8] clock frequency divisor + * + bits[5:4] clock mux selector (index in array) + * + bits[index] clock enable + */ +static uint64_t zynq_slcr_compute_clock(const uint64_t periods[], + uint32_t ctrl_reg, + unsigned index) +{ + uint32_t srcsel = extract32(ctrl_reg, 4, 2); /* bits [5:4] */ + uint32_t divisor = extract32(ctrl_reg, 8, 6); /* bits [13:8] */ + + /* first, check if clock is disabled */ + if (((ctrl_reg >> index) & 1u) == 0) { + return 0; + } + + /* + * according to the Zynq technical ref. manual UG585 v1.12.2 in + * Clocks chapter, section 25.10.1 page 705: + * "The 6-bit divider provides a divide range of 1 to 63" + * We follow here what is implemented in linux kernel and consider + * the 0 value as a bypass (no division). + */ + /* frequency divisor -> period multiplication */ + return periods[srcsel] * (divisor ? divisor : 1u); +} + +/* + * macro helper around zynq_slcr_compute_clock to avoid repeating + * the register name. + */ +#define ZYNQ_COMPUTE_CLK(state, plls, reg, enable_field) \ + zynq_slcr_compute_clock((plls), (state)->regs[reg], \ + reg ## _ ## enable_field ## _SHIFT) + +/** + * Compute and set the ouputs clocks periods. + * But do not propagate them further. Connected clocks + * will not receive any updates (See zynq_slcr_compute_clocks()) + */ +static void zynq_slcr_compute_clocks(ZynqSLCRState *s) +{ + uint64_t ps_clk = clock_get(s->ps_clk); + + /* consider outputs clocks are disabled while in reset */ + if (device_is_in_reset(DEVICE(s))) { + ps_clk = 0; + } + + uint64_t io_pll = zynq_slcr_compute_pll(ps_clk, s->regs[R_IO_PLL_CTRL]); + uint64_t arm_pll = zynq_slcr_compute_pll(ps_clk, s->regs[R_ARM_PLL_CTRL]); + uint64_t ddr_pll = zynq_slcr_compute_pll(ps_clk, s->regs[R_DDR_PLL_CTRL]); + + uint64_t uart_mux[4] = {io_pll, io_pll, arm_pll, ddr_pll}; + + /* compute uartX reference clocks */ + clock_set(s->uart0_ref_clk, + ZYNQ_COMPUTE_CLK(s, uart_mux, R_UART_CLK_CTRL, CLKACT0)); + clock_set(s->uart1_ref_clk, + ZYNQ_COMPUTE_CLK(s, uart_mux, R_UART_CLK_CTRL, CLKACT1)); +} + +/** + * Propagate the outputs clocks. + * zynq_slcr_compute_clocks() should have been called before + * to configure them. + */ +static void zynq_slcr_propagate_clocks(ZynqSLCRState *s) +{ + clock_propagate(s->uart0_ref_clk); + clock_propagate(s->uart1_ref_clk); +} + +static void zynq_slcr_ps_clk_callback(void *opaque) +{ + ZynqSLCRState *s = (ZynqSLCRState *) opaque; + zynq_slcr_compute_clocks(s); + zynq_slcr_propagate_clocks(s); +} + +static void zynq_slcr_reset_init(Object *obj, ResetType type) +{ + ZynqSLCRState *s = ZYNQ_SLCR(obj); int i; DB_PRINT("RESET\n"); @@ -XXX,XX +XXX,XX @@ static void zynq_slcr_reset(DeviceState *d) s->regs[R_DDRIOB + 12] = 0x00000021; } +static void zynq_slcr_reset_hold(Object *obj) +{ + ZynqSLCRState *s = ZYNQ_SLCR(obj); + + /* will disable all output clocks */ + zynq_slcr_compute_clocks(s); + zynq_slcr_propagate_clocks(s); +} + +static void zynq_slcr_reset_exit(Object *obj) +{ + ZynqSLCRState *s = ZYNQ_SLCR(obj); + + /* will compute output clocks according to ps_clk and registers */ + zynq_slcr_compute_clocks(s); + zynq_slcr_propagate_clocks(s); +} static bool zynq_slcr_check_offset(hwaddr offset, bool rnw) { @@ -XXX,XX +XXX,XX @@ static void zynq_slcr_write(void *opaque, hwaddr offset, qemu_system_reset_request(SHUTDOWN_CAUSE_GUEST_RESET); } break; + case R_IO_PLL_CTRL: + case R_ARM_PLL_CTRL: + case R_DDR_PLL_CTRL: + case R_UART_CLK_CTRL: + zynq_slcr_compute_clocks(s); + zynq_slcr_propagate_clocks(s); + break; } } @@ -XXX,XX +XXX,XX @@ static const MemoryRegionOps slcr_ops = { .endianness = DEVICE_NATIVE_ENDIAN, }; +static const ClockPortInitArray zynq_slcr_clocks = { + QDEV_CLOCK_IN(ZynqSLCRState, ps_clk, zynq_slcr_ps_clk_callback), + QDEV_CLOCK_OUT(ZynqSLCRState, uart0_ref_clk), + QDEV_CLOCK_OUT(ZynqSLCRState, uart1_ref_clk), + QDEV_CLOCK_END +}; + static void zynq_slcr_init(Object *obj) { ZynqSLCRState *s = ZYNQ_SLCR(obj); @@ -XXX,XX +XXX,XX @@ static void zynq_slcr_init(Object *obj) memory_region_init_io(&s->iomem, obj, &slcr_ops, s, "slcr", ZYNQ_SLCR_MMIO_SIZE); sysbus_init_mmio(SYS_BUS_DEVICE(obj), &s->iomem); + + qdev_init_clocks(DEVICE(obj), zynq_slcr_clocks); } static const VMStateDescription vmstate_zynq_slcr = { .name = "zynq_slcr", - .version_id = 2, + .version_id = 3, .minimum_version_id = 2, .fields = (VMStateField[]) { VMSTATE_UINT32_ARRAY(regs, ZynqSLCRState, ZYNQ_SLCR_NUM_REGS), + VMSTATE_CLOCK_V(ps_clk, ZynqSLCRState, 3), VMSTATE_END_OF_LIST() } }; @@ -XXX,XX +XXX,XX @@ static const VMStateDescription vmstate_zynq_slcr = { static void zynq_slcr_class_init(ObjectClass *klass, void *data) { DeviceClass *dc = DEVICE_CLASS(klass); + ResettableClass *rc = RESETTABLE_CLASS(klass); dc->vmsd = &vmstate_zynq_slcr; - dc->reset = zynq_slcr_reset; + rc->phases.enter = zynq_slcr_reset_init; + rc->phases.hold = zynq_slcr_reset_hold; + rc->phases.exit = zynq_slcr_reset_exit; } static const TypeInfo zynq_slcr_info = { -- 2.20.1 From: Damien Hedde <damien.hedde@greensocs.com> Switch the cadence uart to multi-phase reset and add the reference clock input. The input clock frequency is added to the migration structure. The reference clock controls the baudrate generation. If it disabled, any input characters and events are ignored. If this clock remains unconnected, the uart behaves as before (it default to a 50MHz ref clock). Signed-off-by: Damien Hedde <damien.hedde@greensocs.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Message-id: 20200406135251.157596-8-damien.hedde@greensocs.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/hw/char/cadence_uart.h | 1 + hw/char/cadence_uart.c | 73 +++++++++++++++++++++++++++++----- hw/char/trace-events | 3 ++ 3 files changed, 67 insertions(+), 10 deletions(-) diff --git a/include/hw/char/cadence_uart.h b/include/hw/char/cadence_uart.h index XXXXXXX..XXXXXXX 100644 --- a/include/hw/char/cadence_uart.h +++ b/include/hw/char/cadence_uart.h @@ -XXX,XX +XXX,XX @@ typedef struct { CharBackend chr; qemu_irq irq; QEMUTimer *fifo_trigger_handle; + Clock *refclk; } CadenceUARTState; static inline DeviceState *cadence_uart_create(hwaddr addr, diff --git a/hw/char/cadence_uart.c b/hw/char/cadence_uart.c index XXXXXXX..XXXXXXX 100644 --- a/hw/char/cadence_uart.c +++ b/hw/char/cadence_uart.c @@ -XXX,XX +XXX,XX @@ #include "qemu/module.h" #include "hw/char/cadence_uart.h" #include "hw/irq.h" +#include "hw/qdev-clock.h" +#include "trace.h" #ifdef CADENCE_UART_ERR_DEBUG #define DB_PRINT(...) do { \ @@ -XXX,XX +XXX,XX @@ #define LOCAL_LOOPBACK (0x2 << UART_MR_CHMODE_SH) #define REMOTE_LOOPBACK (0x3 << UART_MR_CHMODE_SH) -#define UART_INPUT_CLK 50000000 +#define UART_DEFAULT_REF_CLK (50 * 1000 * 1000) #define R_CR (0x00/4) #define R_MR (0x04/4) @@ -XXX,XX +XXX,XX @@ static void uart_send_breaks(CadenceUARTState *s) static void uart_parameters_setup(CadenceUARTState *s) { QEMUSerialSetParams ssp; - unsigned int baud_rate, packet_size; + unsigned int baud_rate, packet_size, input_clk; + input_clk = clock_get_hz(s->refclk); - baud_rate = (s->r[R_MR] & UART_MR_CLKS) ? - UART_INPUT_CLK / 8 : UART_INPUT_CLK; + baud_rate = (s->r[R_MR] & UART_MR_CLKS) ? input_clk / 8 : input_clk; + baud_rate /= (s->r[R_BRGR] * (s->r[R_BDIV] + 1)); + trace_cadence_uart_baudrate(baud_rate); + + ssp.speed = baud_rate; - ssp.speed = baud_rate / (s->r[R_BRGR] * (s->r[R_BDIV] + 1)); packet_size = 1; switch (s->r[R_MR] & UART_MR_PAR) { @@ -XXX,XX +XXX,XX @@ static void uart_parameters_setup(CadenceUARTState *s) } packet_size += ssp.data_bits + ssp.stop_bits; + if (ssp.speed == 0) { + /* + * Avoid division-by-zero below. + * TODO: find something better + */ + ssp.speed = 1; + } s->char_tx_time = (NANOSECONDS_PER_SECOND / ssp.speed) * packet_size; qemu_chr_fe_ioctl(&s->chr, CHR_IOCTL_SERIAL_SET_PARAMS, &ssp); } @@ -XXX,XX +XXX,XX @@ static void uart_receive(void *opaque, const uint8_t *buf, int size) CadenceUARTState *s = opaque; uint32_t ch_mode = s->r[R_MR] & UART_MR_CHMODE; + /* ignore characters when unclocked or in reset */ + if (!clock_is_enabled(s->refclk) || device_is_in_reset(DEVICE(s))) { + return; + } + if (ch_mode == NORMAL_MODE || ch_mode == ECHO_MODE) { uart_write_rx_fifo(opaque, buf, size); } @@ -XXX,XX +XXX,XX @@ static void uart_event(void *opaque, QEMUChrEvent event) CadenceUARTState *s = opaque; uint8_t buf = '\0'; + /* ignore characters when unclocked or in reset */ + if (!clock_is_enabled(s->refclk) || device_is_in_reset(DEVICE(s))) { + return; + } + if (event == CHR_EVENT_BREAK) { uart_write_rx_fifo(opaque, &buf, 1); } @@ -XXX,XX +XXX,XX @@ static const MemoryRegionOps uart_ops = { .endianness = DEVICE_NATIVE_ENDIAN, }; -static void cadence_uart_reset(DeviceState *dev) +static void cadence_uart_reset_init(Object *obj, ResetType type) { - CadenceUARTState *s = CADENCE_UART(dev); + CadenceUARTState *s = CADENCE_UART(obj); s->r[R_CR] = 0x00000128; s->r[R_IMR] = 0; @@ -XXX,XX +XXX,XX @@ static void cadence_uart_reset(DeviceState *dev) s->r[R_BRGR] = 0x0000028B; s->r[R_BDIV] = 0x0000000F; s->r[R_TTRIG] = 0x00000020; +} + +static void cadence_uart_reset_hold(Object *obj) +{ + CadenceUARTState *s = CADENCE_UART(obj); uart_rx_reset(s); uart_tx_reset(s); @@ -XXX,XX +XXX,XX @@ static void cadence_uart_realize(DeviceState *dev, Error **errp) uart_event, NULL, s, NULL, true); } +static void cadence_uart_refclk_update(void *opaque) +{ + CadenceUARTState *s = opaque; + + /* recompute uart's speed on clock change */ + uart_parameters_setup(s); +} + static void cadence_uart_init(Object *obj) { SysBusDevice *sbd = SYS_BUS_DEVICE(obj); @@ -XXX,XX +XXX,XX @@ static void cadence_uart_init(Object *obj) sysbus_init_mmio(sbd, &s->iomem); sysbus_init_irq(sbd, &s->irq); + s->refclk = qdev_init_clock_in(DEVICE(obj), "refclk", + cadence_uart_refclk_update, s); + /* initialize the frequency in case the clock remains unconnected */ + clock_set_hz(s->refclk, UART_DEFAULT_REF_CLK); + s->char_tx_time = (NANOSECONDS_PER_SECOND / 9600) * 10; } +static int cadence_uart_pre_load(void *opaque) +{ + CadenceUARTState *s = opaque; + + /* the frequency will be overriden if the refclk field is present */ + clock_set_hz(s->refclk, UART_DEFAULT_REF_CLK); + return 0; +} + static int cadence_uart_post_load(void *opaque, int version_id) { CadenceUARTState *s = opaque; @@ -XXX,XX +XXX,XX @@ static int cadence_uart_post_load(void *opaque, int version_id) static const VMStateDescription vmstate_cadence_uart = { .name = "cadence_uart", - .version_id = 2, + .version_id = 3, .minimum_version_id = 2, + .pre_load = cadence_uart_pre_load, .post_load = cadence_uart_post_load, .fields = (VMStateField[]) { VMSTATE_UINT32_ARRAY(r, CadenceUARTState, CADENCE_UART_R_MAX), @@ -XXX,XX +XXX,XX @@ static const VMStateDescription vmstate_cadence_uart = { VMSTATE_UINT32(tx_count, CadenceUARTState), VMSTATE_UINT32(rx_wpos, CadenceUARTState), VMSTATE_TIMER_PTR(fifo_trigger_handle, CadenceUARTState), + VMSTATE_CLOCK_V(refclk, CadenceUARTState, 3), VMSTATE_END_OF_LIST() - } + }, }; static Property cadence_uart_properties[] = { @@ -XXX,XX +XXX,XX @@ static Property cadence_uart_properties[] = { static void cadence_uart_class_init(ObjectClass *klass, void *data) { DeviceClass *dc = DEVICE_CLASS(klass); + ResettableClass *rc = RESETTABLE_CLASS(klass); dc->realize = cadence_uart_realize; dc->vmsd = &vmstate_cadence_uart; - dc->reset = cadence_uart_reset; + rc->phases.enter = cadence_uart_reset_init; + rc->phases.hold = cadence_uart_reset_hold; device_class_set_props(dc, cadence_uart_properties); } diff --git a/hw/char/trace-events b/hw/char/trace-events index XXXXXXX..XXXXXXX 100644 --- a/hw/char/trace-events +++ b/hw/char/trace-events @@ -XXX,XX +XXX,XX @@ exynos_uart_wo_read(uint32_t channel, const char *name, uint32_t reg) "UART%d: T exynos_uart_rxsize(uint32_t channel, uint32_t size) "UART%d: Rx FIFO size: %d" exynos_uart_channel_error(uint32_t channel) "Wrong UART channel number: %d" exynos_uart_rx_timeout(uint32_t channel, uint32_t stat, uint32_t intsp) "UART%d: Rx timeout stat=0x%x intsp=0x%x" + +# hw/char/cadence_uart.c +cadence_uart_baudrate(unsigned baudrate) "baudrate %u" -- 2.20.1 From: Damien Hedde <damien.hedde@greensocs.com> Add the connection between the slcr's output clocks and the uarts inputs. Also add the main board clock 'ps_clk', which is hard-coded to 33.33MHz (the default frequency). This clock is used to feed the slcr's input clock. Signed-off-by: Damien Hedde <damien.hedde@greensocs.com> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20200406135251.157596-9-damien.hedde@greensocs.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/xilinx_zynq.c | 57 +++++++++++++++++++++++++++++++++++++------- 1 file changed, 49 insertions(+), 8 deletions(-) diff --git a/hw/arm/xilinx_zynq.c b/hw/arm/xilinx_zynq.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/xilinx_zynq.c +++ b/hw/arm/xilinx_zynq.c @@ -XXX,XX +XXX,XX @@ #include "hw/char/cadence_uart.h" #include "hw/net/cadence_gem.h" #include "hw/cpu/a9mpcore.h" +#include "hw/qdev-clock.h" +#include "sysemu/reset.h" + +#define TYPE_ZYNQ_MACHINE MACHINE_TYPE_NAME("xilinx-zynq-a9") +#define ZYNQ_MACHINE(obj) \ + OBJECT_CHECK(ZynqMachineState, (obj), TYPE_ZYNQ_MACHINE) + +/* board base frequency: 33.333333 MHz */ +#define PS_CLK_FREQUENCY (100 * 1000 * 1000 / 3) #define NUM_SPI_FLASHES 4 #define NUM_QSPI_FLASHES 2 @@ -XXX,XX +XXX,XX @@ static const int dma_irqs[8] = { 0xe3401000 + ARMV7_IMM16(extract32((val), 16, 16)), /* movt r1 ... */ \ 0xe5801000 + (addr) +typedef struct ZynqMachineState { + MachineState parent; + Clock *ps_clk; +} ZynqMachineState; + static void zynq_write_board_setup(ARMCPU *cpu, const struct arm_boot_info *info) { @@ -XXX,XX +XXX,XX @@ static inline void zynq_init_spi_flashes(uint32_t base_addr, qemu_irq irq, static void zynq_init(MachineState *machine) { + ZynqMachineState *zynq_machine = ZYNQ_MACHINE(machine); ARMCPU *cpu; MemoryRegion *address_space_mem = get_system_memory(); MemoryRegion *ocm_ram = g_new(MemoryRegion, 1); - DeviceState *dev; + DeviceState *dev, *slcr; SysBusDevice *busdev; qemu_irq pic[64]; int n; @@ -XXX,XX +XXX,XX @@ static void zynq_init(MachineState *machine) 1, 0x0066, 0x0022, 0x0000, 0x0000, 0x0555, 0x2aa, 0); - dev = qdev_create(NULL, "xilinx,zynq_slcr"); - qdev_init_nofail(dev); - sysbus_mmio_map(SYS_BUS_DEVICE(dev), 0, 0xF8000000); + /* Create slcr, keep a pointer to connect clocks */ + slcr = qdev_create(NULL, "xilinx,zynq_slcr"); + qdev_init_nofail(slcr); + sysbus_mmio_map(SYS_BUS_DEVICE(slcr), 0, 0xF8000000); + + /* Create the main clock source, and feed slcr with it */ + zynq_machine->ps_clk = CLOCK(object_new(TYPE_CLOCK)); + object_property_add_child(OBJECT(zynq_machine), "ps_clk", + OBJECT(zynq_machine->ps_clk), &error_abort); + object_unref(OBJECT(zynq_machine->ps_clk)); + clock_set_hz(zynq_machine->ps_clk, PS_CLK_FREQUENCY); + qdev_connect_clock_in(slcr, "ps_clk", zynq_machine->ps_clk); dev = qdev_create(NULL, TYPE_A9MPCORE_PRIV); qdev_prop_set_uint32(dev, "num-cpu", 1); @@ -XXX,XX +XXX,XX @@ static void zynq_init(MachineState *machine) sysbus_create_simple(TYPE_CHIPIDEA, 0xE0002000, pic[53 - IRQ_OFFSET]); sysbus_create_simple(TYPE_CHIPIDEA, 0xE0003000, pic[76 - IRQ_OFFSET]); - cadence_uart_create(0xE0000000, pic[59 - IRQ_OFFSET], serial_hd(0)); - cadence_uart_create(0xE0001000, pic[82 - IRQ_OFFSET], serial_hd(1)); + dev = cadence_uart_create(0xE0000000, pic[59 - IRQ_OFFSET], serial_hd(0)); + qdev_connect_clock_in(dev, "refclk", + qdev_get_clock_out(slcr, "uart0_ref_clk")); + dev = cadence_uart_create(0xE0001000, pic[82 - IRQ_OFFSET], serial_hd(1)); + qdev_connect_clock_in(dev, "refclk", + qdev_get_clock_out(slcr, "uart1_ref_clk")); sysbus_create_varargs("cadence_ttc", 0xF8001000, pic[42-IRQ_OFFSET], pic[43-IRQ_OFFSET], pic[44-IRQ_OFFSET], NULL); @@ -XXX,XX +XXX,XX @@ static void zynq_init(MachineState *machine) arm_load_kernel(ARM_CPU(first_cpu), machine, &zynq_binfo); } -static void zynq_machine_init(MachineClass *mc) +static void zynq_machine_class_init(ObjectClass *oc, void *data) { + MachineClass *mc = MACHINE_CLASS(oc); mc->desc = "Xilinx Zynq Platform Baseboard for Cortex-A9"; mc->init = zynq_init; mc->max_cpus = 1; @@ -XXX,XX +XXX,XX @@ static void zynq_machine_init(MachineClass *mc) mc->default_ram_id = "zynq.ext_ram"; } -DEFINE_MACHINE("xilinx-zynq-a9", zynq_machine_init) +static const TypeInfo zynq_machine_type = { + .name = TYPE_ZYNQ_MACHINE, + .parent = TYPE_MACHINE, + .class_init = zynq_machine_class_init, + .instance_size = sizeof(ZynqMachineState), +}; + +static void zynq_machine_register_types(void) +{ + type_register_static(&zynq_machine_type); +} + +type_init(zynq_machine_register_types) -- 2.20.1 From: Damien Hedde <damien.hedde@greensocs.com> This prints the clocks attached to a DeviceState when using "info qtree" monitor command. For every clock, it displays the direction, the name and if the clock is forwarded. For input clock, it displays also the frequency. This is based on the original work of Frederic Konrad. Here follows a sample of `info qtree` output on xilinx_zynq machine after linux boot with only one uart clocked: > bus: main-system-bus > type System > [...] > dev: cadence_uart, id "" > gpio-out "sysbus-irq" 1 > clock-in "refclk" freq_hz=0.000000e+00 > chardev = "" > mmio 00000000e0001000/0000000000001000 > dev: cadence_uart, id "" > gpio-out "sysbus-irq" 1 > clock-in "refclk" freq_hz=1.375661e+07 > chardev = "serial0" > mmio 00000000e0000000/0000000000001000 > [...] > dev: xilinx,zynq_slcr, id "" > clock-out "uart1_ref_clk" freq_hz=0.000000e+00 > clock-out "uart0_ref_clk" freq_hz=1.375661e+07 > clock-in "ps_clk" freq_hz=3.333333e+07 > mmio 00000000f8000000/0000000000001000 Signed-off-by: Damien Hedde <damien.hedde@greensocs.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20200406135251.157596-10-damien.hedde@greensocs.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- qdev-monitor.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/qdev-monitor.c b/qdev-monitor.c index XXXXXXX..XXXXXXX 100644 --- a/qdev-monitor.c +++ b/qdev-monitor.c @@ -XXX,XX +XXX,XX @@ #include "migration/misc.h" #include "migration/migration.h" #include "qemu/cutils.h" +#include "hw/clock.h" /* * Aliases were a bad idea from the start. Let's keep them @@ -XXX,XX +XXX,XX @@ static void qdev_print(Monitor *mon, DeviceState *dev, int indent) ObjectClass *class; BusState *child; NamedGPIOList *ngl; + NamedClockList *ncl; qdev_printf("dev: %s, id \"%s\"\n", object_get_typename(OBJECT(dev)), dev->id ? dev->id : ""); @@ -XXX,XX +XXX,XX @@ static void qdev_print(Monitor *mon, DeviceState *dev, int indent) ngl->num_out); } } + QLIST_FOREACH(ncl, &dev->clocks, node) { + qdev_printf("clock-%s%s \"%s\" freq_hz=%e\n", + ncl->output ? "out" : "in", + ncl->alias ? " (alias)" : "", + ncl->name, + CLOCK_PERIOD_TO_HZ(1.0 * clock_get(ncl->clock))); + } class = object_get_class(OBJECT(dev)); do { qdev_print_props(mon, dev, DEVICE_CLASS(class)->props_, indent); -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Setup the ADMA with 128bit bus-width. This matters when FIXED BURST mode is used. 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: 20200417153800.27399-2-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/xlnx-versal.c | 2 ++ 1 file changed, 2 insertions(+) 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) 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); qdev_init_nofail(dev); -- 2.20.1 From: Ramon Fried <rfried.dev@gmail.com> Wraparound of TX descriptor cyclic buffer only updated the low 32 bits of the descriptor. Fix that by checking if we're working with 64bit descriptors. Signed-off-by: Ramon Fried <rfried.dev@gmail.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20200417171736.441607-1-rfried.dev@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/net/cadence_gem.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/hw/net/cadence_gem.c b/hw/net/cadence_gem.c index XXXXXXX..XXXXXXX 100644 --- a/hw/net/cadence_gem.c +++ b/hw/net/cadence_gem.c @@ -XXX,XX +XXX,XX @@ static void gem_transmit(CadenceGEMState *s) /* read next descriptor */ if (tx_desc_get_wrap(desc)) { tx_desc_set_last(desc); - packet_desc_addr = s->regs[GEM_TXQBASE]; + + if (s->regs[GEM_DMACFG] & GEM_DMACFG_ADDR_64B) { + packet_desc_addr = s->regs[GEM_TBQPH]; + packet_desc_addr <<= 32; + } else { + packet_desc_addr = 0; + } + packet_desc_addr |= s->regs[GEM_TXQBASE]; } else { packet_desc_addr += 4 * gem_get_desc_len(s, false); } -- 2.20.1 From: Ramon Fried <rfried.dev@gmail.com> The RX ring descriptors control field is used for setting SOF and EOF (start of frame and end of frame). The SOF and EOF weren't cleared from the previous descriptors, causing inconsistencies in ring buffer. Fix that by clearing the control field of every descriptors we're processing. Signed-off-by: Ramon Fried <rfried.dev@gmail.com> Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20200418085145.489726-1-rfried.dev@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/net/cadence_gem.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/hw/net/cadence_gem.c b/hw/net/cadence_gem.c index XXXXXXX..XXXXXXX 100644 --- a/hw/net/cadence_gem.c +++ b/hw/net/cadence_gem.c @@ -XXX,XX +XXX,XX @@ static inline void rx_desc_set_sof(uint32_t *desc) desc[1] |= DESC_1_RX_SOF; } +static inline void rx_desc_clear_control(uint32_t *desc) +{ + desc[1] = 0; +} + static inline void rx_desc_set_eof(uint32_t *desc) { desc[1] |= DESC_1_RX_EOF; @@ -XXX,XX +XXX,XX @@ static ssize_t gem_receive(NetClientState *nc, const uint8_t *buf, size_t size) rxbuf_ptr += MIN(bytes_to_copy, rxbufsize); bytes_to_copy -= MIN(bytes_to_copy, rxbufsize); + rx_desc_clear_control(s->rx_desc[q]); + /* Update the descriptor. */ if (first_desc) { rx_desc_set_sof(s->rx_desc[q]); -- 2.20.1 From: Richard Henderson <richard.henderson@linaro.org> These instructions are often used in glibc's string routines. They were the final uses of the 32-bit at a time neon helpers. Signed-off-by: Richard Henderson <richard.henderson@linaro.org> Message-id: 20200418162808.4680-1-richard.henderson@linaro.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.h | 27 ++-- target/arm/translate.h | 5 + target/arm/neon_helper.c | 24 ---- target/arm/translate-a64.c | 64 +++------- target/arm/translate.c | 256 +++++++++++++++++++++++++++++++------ target/arm/vec_helper.c | 25 ++++ 6 files changed, 278 insertions(+), 123 deletions(-) diff --git a/target/arm/helper.h b/target/arm/helper.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/helper.h +++ b/target/arm/helper.h @@ -XXX,XX +XXX,XX @@ DEF_HELPER_2(neon_hsub_u16, i32, i32, i32) DEF_HELPER_2(neon_hsub_s32, s32, s32, s32) DEF_HELPER_2(neon_hsub_u32, i32, i32, i32) -DEF_HELPER_2(neon_cgt_u8, i32, i32, i32) -DEF_HELPER_2(neon_cgt_s8, i32, i32, i32) -DEF_HELPER_2(neon_cgt_u16, i32, i32, i32) -DEF_HELPER_2(neon_cgt_s16, i32, i32, i32) -DEF_HELPER_2(neon_cgt_u32, i32, i32, i32) -DEF_HELPER_2(neon_cgt_s32, i32, i32, i32) -DEF_HELPER_2(neon_cge_u8, i32, i32, i32) -DEF_HELPER_2(neon_cge_s8, i32, i32, i32) -DEF_HELPER_2(neon_cge_u16, i32, i32, i32) -DEF_HELPER_2(neon_cge_s16, i32, i32, i32) -DEF_HELPER_2(neon_cge_u32, i32, i32, i32) -DEF_HELPER_2(neon_cge_s32, i32, i32, i32) - DEF_HELPER_2(neon_pmin_u8, i32, i32, i32) DEF_HELPER_2(neon_pmin_s8, i32, i32, i32) DEF_HELPER_2(neon_pmin_u16, i32, i32, i32) @@ -XXX,XX +XXX,XX @@ DEF_HELPER_2(neon_mul_u16, i32, i32, i32) DEF_HELPER_2(neon_tst_u8, i32, i32, i32) DEF_HELPER_2(neon_tst_u16, i32, i32, i32) DEF_HELPER_2(neon_tst_u32, i32, i32, i32) -DEF_HELPER_2(neon_ceq_u8, i32, i32, i32) -DEF_HELPER_2(neon_ceq_u16, i32, i32, i32) -DEF_HELPER_2(neon_ceq_u32, i32, i32, i32) DEF_HELPER_1(neon_clz_u8, i32, i32) DEF_HELPER_1(neon_clz_u16, i32, i32) @@ -XXX,XX +XXX,XX @@ DEF_HELPER_FLAGS_2(frint64_s, TCG_CALL_NO_RWG, f32, f32, ptr) DEF_HELPER_FLAGS_2(frint32_d, TCG_CALL_NO_RWG, f64, f64, ptr) DEF_HELPER_FLAGS_2(frint64_d, TCG_CALL_NO_RWG, f64, f64, ptr) +DEF_HELPER_FLAGS_3(gvec_ceq0_b, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_ceq0_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_clt0_b, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_clt0_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_cle0_b, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_cle0_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_cgt0_b, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_cgt0_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_cge0_b, TCG_CALL_NO_RWG, void, ptr, ptr, i32) +DEF_HELPER_FLAGS_3(gvec_cge0_h, TCG_CALL_NO_RWG, void, ptr, ptr, i32) + DEF_HELPER_FLAGS_4(gvec_sshl_b, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_4(gvec_sshl_h, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_4(gvec_ushl_b, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, i32) 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 @@ static inline void gen_swstep_exception(DisasContext *s, int isv, int ex) uint64_t vfp_expand_imm(int size, uint8_t imm8); /* Vector operations shared between ARM and AArch64. */ +extern const GVecGen2 ceq0_op[4]; +extern const GVecGen2 clt0_op[4]; +extern const GVecGen2 cgt0_op[4]; +extern const GVecGen2 cle0_op[4]; +extern const GVecGen2 cge0_op[4]; extern const GVecGen3 mla_op[4]; extern const GVecGen3 mls_op[4]; extern const GVecGen3 cmtst_op[4]; diff --git a/target/arm/neon_helper.c b/target/arm/neon_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/neon_helper.c +++ b/target/arm/neon_helper.c @@ -XXX,XX +XXX,XX @@ uint32_t HELPER(neon_hsub_u32)(uint32_t src1, uint32_t src2) return dest; } -#define NEON_FN(dest, src1, src2) dest = (src1 > src2) ? ~0 : 0 -NEON_VOP(cgt_s8, neon_s8, 4) -NEON_VOP(cgt_u8, neon_u8, 4) -NEON_VOP(cgt_s16, neon_s16, 2) -NEON_VOP(cgt_u16, neon_u16, 2) -NEON_VOP(cgt_s32, neon_s32, 1) -NEON_VOP(cgt_u32, neon_u32, 1) -#undef NEON_FN - -#define NEON_FN(dest, src1, src2) dest = (src1 >= src2) ? ~0 : 0 -NEON_VOP(cge_s8, neon_s8, 4) -NEON_VOP(cge_u8, neon_u8, 4) -NEON_VOP(cge_s16, neon_s16, 2) -NEON_VOP(cge_u16, neon_u16, 2) -NEON_VOP(cge_s32, neon_s32, 1) -NEON_VOP(cge_u32, neon_u32, 1) -#undef NEON_FN - #define NEON_FN(dest, src1, src2) dest = (src1 < src2) ? src1 : src2 NEON_POP(pmin_s8, neon_s8, 4) NEON_POP(pmin_u8, neon_u8, 4) @@ -XXX,XX +XXX,XX @@ NEON_VOP(tst_u16, neon_u16, 2) NEON_VOP(tst_u32, neon_u32, 1) #undef NEON_FN -#define NEON_FN(dest, src1, src2) dest = (src1 == src2) ? -1 : 0 -NEON_VOP(ceq_u8, neon_u8, 4) -NEON_VOP(ceq_u16, neon_u16, 2) -NEON_VOP(ceq_u32, neon_u32, 1) -#undef NEON_FN - /* Count Leading Sign/Zero Bits. */ static inline int do_clz8(uint8_t x) { diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -XXX,XX +XXX,XX @@ static void gen_gvec_fn4(DisasContext *s, bool is_q, int rd, int rn, int rm, is_q ? 16 : 8, vec_full_reg_size(s)); } +/* Expand a 2-operand AdvSIMD vector operation using an op descriptor. */ +static void gen_gvec_op2(DisasContext *s, bool is_q, int rd, + int rn, const GVecGen2 *gvec_op) +{ + tcg_gen_gvec_2(vec_full_reg_offset(s, rd), vec_full_reg_offset(s, rn), + is_q ? 16 : 8, vec_full_reg_size(s), gvec_op); +} + /* Expand a 2-operand + immediate AdvSIMD vector operation using * an op descriptor. */ @@ -XXX,XX +XXX,XX @@ static void disas_simd_two_reg_misc(DisasContext *s, uint32_t insn) return; } break; + case 0x8: /* CMGT, CMGE */ + gen_gvec_op2(s, is_q, rd, rn, u ? &cge0_op[size] : &cgt0_op[size]); + return; + case 0x9: /* CMEQ, CMLE */ + gen_gvec_op2(s, is_q, rd, rn, u ? &cle0_op[size] : &ceq0_op[size]); + return; + case 0xa: /* CMLT */ + gen_gvec_op2(s, is_q, rd, rn, &clt0_op[size]); + return; case 0xb: if (u) { /* ABS, NEG */ gen_gvec_fn2(s, is_q, rd, rn, tcg_gen_gvec_neg, size); @@ -XXX,XX +XXX,XX @@ static void disas_simd_two_reg_misc(DisasContext *s, uint32_t insn) for (pass = 0; pass < (is_q ? 4 : 2); pass++) { TCGv_i32 tcg_op = tcg_temp_new_i32(); TCGv_i32 tcg_res = tcg_temp_new_i32(); - TCGCond cond; read_vec_element_i32(s, tcg_op, rn, pass, MO_32); if (size == 2) { /* Special cases for 32 bit elements */ switch (opcode) { - case 0xa: /* CMLT */ - /* 32 bit integer comparison against zero, result is - * test ? (2^32 - 1) : 0. We implement via setcond(test) - * and inverting. - */ - cond = TCG_COND_LT; - do_cmop: - tcg_gen_setcondi_i32(cond, tcg_res, tcg_op, 0); - tcg_gen_neg_i32(tcg_res, tcg_res); - break; - case 0x8: /* CMGT, CMGE */ - cond = u ? TCG_COND_GE : TCG_COND_GT; - goto do_cmop; - case 0x9: /* CMEQ, CMLE */ - cond = u ? TCG_COND_LE : TCG_COND_EQ; - goto do_cmop; case 0x4: /* CLS */ if (u) { tcg_gen_clzi_i32(tcg_res, tcg_op, 32); @@ -XXX,XX +XXX,XX @@ static void disas_simd_two_reg_misc(DisasContext *s, uint32_t insn) genfn(tcg_res, cpu_env, tcg_op); break; } - case 0x8: /* CMGT, CMGE */ - case 0x9: /* CMEQ, CMLE */ - case 0xa: /* CMLT */ - { - static NeonGenTwoOpFn * const fns[3][2] = { - { gen_helper_neon_cgt_s8, gen_helper_neon_cgt_s16 }, - { gen_helper_neon_cge_s8, gen_helper_neon_cge_s16 }, - { gen_helper_neon_ceq_u8, gen_helper_neon_ceq_u16 }, - }; - NeonGenTwoOpFn *genfn; - int comp; - bool reverse; - TCGv_i32 tcg_zero = tcg_const_i32(0); - - /* comp = index into [CMGT, CMGE, CMEQ, CMLE, CMLT] */ - comp = (opcode - 0x8) * 2 + u; - /* ...but LE, LT are implemented as reverse GE, GT */ - reverse = (comp > 2); - if (reverse) { - comp = 4 - comp; - } - genfn = fns[comp][size]; - if (reverse) { - genfn(tcg_res, tcg_zero, tcg_op); - } else { - genfn(tcg_res, tcg_op, tcg_zero); - } - tcg_temp_free_i32(tcg_zero); - break; - } case 0x4: /* CLS, CLZ */ if (u) { if (size == 0) { 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 do_v81_helper(DisasContext *s, gen_helper_gvec_3_ptr *fn, return 1; } +static void gen_ceq0_i32(TCGv_i32 d, TCGv_i32 a) +{ + tcg_gen_setcondi_i32(TCG_COND_EQ, d, a, 0); + tcg_gen_neg_i32(d, d); +} + +static void gen_ceq0_i64(TCGv_i64 d, TCGv_i64 a) +{ + tcg_gen_setcondi_i64(TCG_COND_EQ, d, a, 0); + tcg_gen_neg_i64(d, d); +} + +static void gen_ceq0_vec(unsigned vece, TCGv_vec d, TCGv_vec a) +{ + TCGv_vec zero = tcg_const_zeros_vec_matching(d); + tcg_gen_cmp_vec(TCG_COND_EQ, vece, d, a, zero); + tcg_temp_free_vec(zero); +} + +static const TCGOpcode vecop_list_cmp[] = { + INDEX_op_cmp_vec, 0 +}; + +const GVecGen2 ceq0_op[4] = { + { .fno = gen_helper_gvec_ceq0_b, + .fniv = gen_ceq0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_8 }, + { .fno = gen_helper_gvec_ceq0_h, + .fniv = gen_ceq0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_16 }, + { .fni4 = gen_ceq0_i32, + .fniv = gen_ceq0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_32 }, + { .fni8 = gen_ceq0_i64, + .fniv = gen_ceq0_vec, + .opt_opc = vecop_list_cmp, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .vece = MO_64 }, +}; + +static void gen_cle0_i32(TCGv_i32 d, TCGv_i32 a) +{ + tcg_gen_setcondi_i32(TCG_COND_LE, d, a, 0); + tcg_gen_neg_i32(d, d); +} + +static void gen_cle0_i64(TCGv_i64 d, TCGv_i64 a) +{ + tcg_gen_setcondi_i64(TCG_COND_LE, d, a, 0); + tcg_gen_neg_i64(d, d); +} + +static void gen_cle0_vec(unsigned vece, TCGv_vec d, TCGv_vec a) +{ + TCGv_vec zero = tcg_const_zeros_vec_matching(d); + tcg_gen_cmp_vec(TCG_COND_LE, vece, d, a, zero); + tcg_temp_free_vec(zero); +} + +const GVecGen2 cle0_op[4] = { + { .fno = gen_helper_gvec_cle0_b, + .fniv = gen_cle0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_8 }, + { .fno = gen_helper_gvec_cle0_h, + .fniv = gen_cle0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_16 }, + { .fni4 = gen_cle0_i32, + .fniv = gen_cle0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_32 }, + { .fni8 = gen_cle0_i64, + .fniv = gen_cle0_vec, + .opt_opc = vecop_list_cmp, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .vece = MO_64 }, +}; + +static void gen_cge0_i32(TCGv_i32 d, TCGv_i32 a) +{ + tcg_gen_setcondi_i32(TCG_COND_GE, d, a, 0); + tcg_gen_neg_i32(d, d); +} + +static void gen_cge0_i64(TCGv_i64 d, TCGv_i64 a) +{ + tcg_gen_setcondi_i64(TCG_COND_GE, d, a, 0); + tcg_gen_neg_i64(d, d); +} + +static void gen_cge0_vec(unsigned vece, TCGv_vec d, TCGv_vec a) +{ + TCGv_vec zero = tcg_const_zeros_vec_matching(d); + tcg_gen_cmp_vec(TCG_COND_GE, vece, d, a, zero); + tcg_temp_free_vec(zero); +} + +const GVecGen2 cge0_op[4] = { + { .fno = gen_helper_gvec_cge0_b, + .fniv = gen_cge0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_8 }, + { .fno = gen_helper_gvec_cge0_h, + .fniv = gen_cge0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_16 }, + { .fni4 = gen_cge0_i32, + .fniv = gen_cge0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_32 }, + { .fni8 = gen_cge0_i64, + .fniv = gen_cge0_vec, + .opt_opc = vecop_list_cmp, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .vece = MO_64 }, +}; + +static void gen_clt0_i32(TCGv_i32 d, TCGv_i32 a) +{ + tcg_gen_setcondi_i32(TCG_COND_LT, d, a, 0); + tcg_gen_neg_i32(d, d); +} + +static void gen_clt0_i64(TCGv_i64 d, TCGv_i64 a) +{ + tcg_gen_setcondi_i64(TCG_COND_LT, d, a, 0); + tcg_gen_neg_i64(d, d); +} + +static void gen_clt0_vec(unsigned vece, TCGv_vec d, TCGv_vec a) +{ + TCGv_vec zero = tcg_const_zeros_vec_matching(d); + tcg_gen_cmp_vec(TCG_COND_LT, vece, d, a, zero); + tcg_temp_free_vec(zero); +} + +const GVecGen2 clt0_op[4] = { + { .fno = gen_helper_gvec_clt0_b, + .fniv = gen_clt0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_8 }, + { .fno = gen_helper_gvec_clt0_h, + .fniv = gen_clt0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_16 }, + { .fni4 = gen_clt0_i32, + .fniv = gen_clt0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_32 }, + { .fni8 = gen_clt0_i64, + .fniv = gen_clt0_vec, + .opt_opc = vecop_list_cmp, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .vece = MO_64 }, +}; + +static void gen_cgt0_i32(TCGv_i32 d, TCGv_i32 a) +{ + tcg_gen_setcondi_i32(TCG_COND_GT, d, a, 0); + tcg_gen_neg_i32(d, d); +} + +static void gen_cgt0_i64(TCGv_i64 d, TCGv_i64 a) +{ + tcg_gen_setcondi_i64(TCG_COND_GT, d, a, 0); + tcg_gen_neg_i64(d, d); +} + +static void gen_cgt0_vec(unsigned vece, TCGv_vec d, TCGv_vec a) +{ + TCGv_vec zero = tcg_const_zeros_vec_matching(d); + tcg_gen_cmp_vec(TCG_COND_GT, vece, d, a, zero); + tcg_temp_free_vec(zero); +} + +const GVecGen2 cgt0_op[4] = { + { .fno = gen_helper_gvec_cgt0_b, + .fniv = gen_cgt0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_8 }, + { .fno = gen_helper_gvec_cgt0_h, + .fniv = gen_cgt0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_16 }, + { .fni4 = gen_cgt0_i32, + .fniv = gen_cgt0_vec, + .opt_opc = vecop_list_cmp, + .vece = MO_32 }, + { .fni8 = gen_cgt0_i64, + .fniv = gen_cgt0_vec, + .opt_opc = vecop_list_cmp, + .prefer_i64 = TCG_TARGET_REG_BITS == 64, + .vece = MO_64 }, +}; + static void gen_ssra8_i64(TCGv_i64 d, TCGv_i64 a, int64_t shift) { tcg_gen_vec_sar8i_i64(a, a, shift); @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) tcg_gen_gvec_abs(size, rd_ofs, rm_ofs, vec_size, vec_size); break; + case NEON_2RM_VCEQ0: + tcg_gen_gvec_2(rd_ofs, rm_ofs, vec_size, + vec_size, &ceq0_op[size]); + break; + case NEON_2RM_VCGT0: + tcg_gen_gvec_2(rd_ofs, rm_ofs, vec_size, + vec_size, &cgt0_op[size]); + break; + case NEON_2RM_VCLE0: + tcg_gen_gvec_2(rd_ofs, rm_ofs, vec_size, + vec_size, &cle0_op[size]); + break; + case NEON_2RM_VCGE0: + tcg_gen_gvec_2(rd_ofs, rm_ofs, vec_size, + vec_size, &cge0_op[size]); + break; + case NEON_2RM_VCLT0: + tcg_gen_gvec_2(rd_ofs, rm_ofs, vec_size, + vec_size, &clt0_op[size]); + break; + default: elementwise: for (pass = 0; pass < (q ? 4 : 2); pass++) { @@ -XXX,XX +XXX,XX @@ static int disas_neon_data_insn(DisasContext *s, uint32_t insn) default: abort(); } break; - case NEON_2RM_VCGT0: case NEON_2RM_VCLE0: - tmp2 = tcg_const_i32(0); - switch(size) { - case 0: gen_helper_neon_cgt_s8(tmp, tmp, tmp2); break; - case 1: gen_helper_neon_cgt_s16(tmp, tmp, tmp2); break; - case 2: gen_helper_neon_cgt_s32(tmp, tmp, tmp2); break; - default: abort(); - } - tcg_temp_free_i32(tmp2); - if (op == NEON_2RM_VCLE0) { - tcg_gen_not_i32(tmp, tmp); - } - break; - case NEON_2RM_VCGE0: case NEON_2RM_VCLT0: - tmp2 = tcg_const_i32(0); - switch(size) { - case 0: gen_helper_neon_cge_s8(tmp, tmp, tmp2); break; - case 1: gen_helper_neon_cge_s16(tmp, tmp, tmp2); break; - case 2: gen_helper_neon_cge_s32(tmp, tmp, tmp2); break; - default: abort(); - } - tcg_temp_free_i32(tmp2); - if (op == NEON_2RM_VCLT0) { - tcg_gen_not_i32(tmp, tmp); - } - break; - case NEON_2RM_VCEQ0: - tmp2 = tcg_const_i32(0); - switch(size) { - case 0: gen_helper_neon_ceq_u8(tmp, tmp, tmp2); break; - case 1: gen_helper_neon_ceq_u16(tmp, tmp, tmp2); break; - case 2: gen_helper_neon_ceq_u32(tmp, tmp, tmp2); break; - default: abort(); - } - tcg_temp_free_i32(tmp2); - break; case NEON_2RM_VCGT0_F: { TCGv_ptr fpstatus = get_fpstatus_ptr(1); diff --git a/target/arm/vec_helper.c b/target/arm/vec_helper.c index XXXXXXX..XXXXXXX 100644 --- a/target/arm/vec_helper.c +++ b/target/arm/vec_helper.c @@ -XXX,XX +XXX,XX @@ void HELPER(sve2_pmull_h)(void *vd, void *vn, void *vm, uint32_t desc) } } #endif + +#define DO_CMP0(NAME, TYPE, OP) \ +void HELPER(NAME)(void *vd, void *vn, uint32_t desc) \ +{ \ + intptr_t i, opr_sz = simd_oprsz(desc); \ + for (i = 0; i < opr_sz; i += sizeof(TYPE)) { \ + TYPE nn = *(TYPE *)(vn + i); \ + *(TYPE *)(vd + i) = -(nn OP 0); \ + } \ + clear_tail(vd, opr_sz, simd_maxsz(desc)); \ +} + +DO_CMP0(gvec_ceq0_b, int8_t, ==) +DO_CMP0(gvec_clt0_b, int8_t, <) +DO_CMP0(gvec_cle0_b, int8_t, <=) +DO_CMP0(gvec_cgt0_b, int8_t, >) +DO_CMP0(gvec_cge0_b, int8_t, >=) + +DO_CMP0(gvec_ceq0_h, int16_t, ==) +DO_CMP0(gvec_clt0_h, int16_t, <) +DO_CMP0(gvec_cle0_h, int16_t, <=) +DO_CMP0(gvec_cgt0_h, int16_t, >) +DO_CMP0(gvec_cge0_h, int16_t, >=) + +#undef DO_CMP0 -- 2.20.1 From: Jerome Forissier <jerome@forissier.org> The /secure-chosen node is currently used only by create_uart(), but this will change. Therefore move the creation of this node to create_fdt(). Signed-off-by: Jerome Forissier <jerome@forissier.org> Message-id: 20200420121807.8204-2-jerome@forissier.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/virt.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/hw/arm/virt.c b/hw/arm/virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -XXX,XX +XXX,XX @@ static void create_fdt(VirtMachineState *vms) /* /chosen must exist for load_dtb to fill in necessary properties later */ qemu_fdt_add_subnode(fdt, "/chosen"); + if (vms->secure) { + qemu_fdt_add_subnode(fdt, "/secure-chosen"); + } + /* Clock node, for the benefit of the UART. The kernel device tree * binding documentation claims the PL011 node clock properties are * optional but in practice if you omit them the kernel refuses to @@ -XXX,XX +XXX,XX @@ static void create_uart(const VirtMachineState *vms, int uart, qemu_fdt_setprop_string(vms->fdt, nodename, "status", "disabled"); qemu_fdt_setprop_string(vms->fdt, nodename, "secure-status", "okay"); - qemu_fdt_add_subnode(vms->fdt, "/secure-chosen"); qemu_fdt_setprop_string(vms->fdt, "/secure-chosen", "stdout-path", nodename); } -- 2.20.1 From: Jerome Forissier <jerome@forissier.org> Generate random seeds to be used by the non-secure and/or secure OSes for ASLR. The seeds are 64-bit random values exported via the DT properties /chosen/kaslr-seed [1] and /secure-chosen/kaslr-seed, the latter being used by OP-TEE [2]. [1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=e5bc0c37c97e1 [2] https://github.com/OP-TEE/optee_os/commit/ef262691fe0e Signed-off-by: Jerome Forissier <jerome@forissier.org> Message-id: 20200420121807.8204-3-jerome@forissier.org Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/virt.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/hw/arm/virt.c b/hw/arm/virt.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/virt.c +++ b/hw/arm/virt.c @@ -XXX,XX +XXX,XX @@ #include "hw/acpi/generic_event_device.h" #include "hw/virtio/virtio-iommu.h" #include "hw/char/pl011.h" +#include "qemu/guest-random.h" #define DEFINE_VIRT_MACHINE_LATEST(major, minor, latest) \ static void virt_##major##_##minor##_class_init(ObjectClass *oc, \ @@ -XXX,XX +XXX,XX @@ static bool cpu_type_valid(const char *cpu) return false; } +static void create_kaslr_seed(VirtMachineState *vms, const char *node) +{ + Error *err = NULL; + uint64_t seed; + + if (qemu_guest_getrandom(&seed, sizeof(seed), &err)) { + error_free(err); + return; + } + qemu_fdt_setprop_u64(vms->fdt, node, "kaslr-seed", seed); +} + static void create_fdt(VirtMachineState *vms) { MachineState *ms = MACHINE(vms); @@ -XXX,XX +XXX,XX @@ static void create_fdt(VirtMachineState *vms) /* /chosen must exist for load_dtb to fill in necessary properties later */ qemu_fdt_add_subnode(fdt, "/chosen"); + create_kaslr_seed(vms, "/chosen"); if (vms->secure) { qemu_fdt_add_subnode(fdt, "/secure-chosen"); + create_kaslr_seed(vms, "/secure-chosen"); } /* Clock node, for the benefit of the UART. The kernel device tree -- 2.20.1 From: Philippe Mathieu-Daudé <philmd@redhat.com> Under KVM these registers are written by the hardware. Restrict the writefn handlers to TCG to avoid when building without TCG: LINK aarch64-softmmu/qemu-system-aarch64 target/arm/helper.o: In function `do_ats_write': target/arm/helper.c:3524: undefined reference to `raise_exception' Suggested-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20200423073358.27155-2-philmd@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/helper.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) 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 CPAccessResult ats_access(CPUARMState *env, const ARMCPRegInfo *ri, return CP_ACCESS_OK; } +#ifdef CONFIG_TCG static uint64_t do_ats_write(CPUARMState *env, uint64_t value, MMUAccessType access_type, ARMMMUIdx mmu_idx) { @@ -XXX,XX +XXX,XX @@ static uint64_t do_ats_write(CPUARMState *env, uint64_t value, } return par64; } +#endif /* CONFIG_TCG */ static void ats_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value) { +#ifdef CONFIG_TCG MMUAccessType access_type = ri->opc2 & 1 ? MMU_DATA_STORE : MMU_DATA_LOAD; uint64_t par64; ARMMMUIdx mmu_idx; @@ -XXX,XX +XXX,XX @@ static void ats_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value) par64 = do_ats_write(env, value, access_type, mmu_idx); A32_BANKED_CURRENT_REG_SET(env, par, par64); +#else + /* Handled by hardware accelerator. */ + g_assert_not_reached(); +#endif /* CONFIG_TCG */ } static void ats1h_write(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value) { +#ifdef CONFIG_TCG MMUAccessType access_type = ri->opc2 & 1 ? MMU_DATA_STORE : MMU_DATA_LOAD; uint64_t par64; par64 = do_ats_write(env, value, access_type, ARMMMUIdx_E2); A32_BANKED_CURRENT_REG_SET(env, par, par64); +#else + /* Handled by hardware accelerator. */ + g_assert_not_reached(); +#endif /* CONFIG_TCG */ } static CPAccessResult at_s1e2_access(CPUARMState *env, const ARMCPRegInfo *ri, @@ -XXX,XX +XXX,XX @@ static CPAccessResult at_s1e2_access(CPUARMState *env, const ARMCPRegInfo *ri, static void ats_write64(CPUARMState *env, const ARMCPRegInfo *ri, uint64_t value) { +#ifdef CONFIG_TCG MMUAccessType access_type = ri->opc2 & 1 ? MMU_DATA_STORE : MMU_DATA_LOAD; ARMMMUIdx mmu_idx; int secure = arm_is_secure_below_el3(env); @@ -XXX,XX +XXX,XX @@ static void ats_write64(CPUARMState *env, const ARMCPRegInfo *ri, } env->cp15.par_el[1] = do_ats_write(env, value, access_type, mmu_idx); +#else + /* Handled by hardware accelerator. */ + g_assert_not_reached(); +#endif /* CONFIG_TCG */ } #endif -- 2.20.1 From: Thomas Huth <thuth@redhat.com> Make cpu_register() (renamed to arm_cpu_register()) available from internals.h so we can register CPUs also from other files in the future. Signed-off-by: Thomas Huth <thuth@redhat.com> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Eric Auger <eric.auger@redhat.com> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20200423073358.27155-3-philmd@redhat.com Message-ID: <20190921150420.30743-2-thuth@redhat.com> [PMD: Only take cpu_register() from Thomas's patch] Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu-qom.h | 9 ++++++++- target/arm/cpu.c | 10 ++-------- target/arm/cpu64.c | 8 +------- 3 files changed, 11 insertions(+), 16 deletions(-) diff --git a/target/arm/cpu-qom.h b/target/arm/cpu-qom.h index XXXXXXX..XXXXXXX 100644 --- a/target/arm/cpu-qom.h +++ b/target/arm/cpu-qom.h @@ -XXX,XX +XXX,XX @@ struct arm_boot_info; #define TYPE_ARM_MAX_CPU "max-" TYPE_ARM_CPU -typedef struct ARMCPUInfo ARMCPUInfo; +typedef struct ARMCPUInfo { + const char *name; + void (*initfn)(Object *obj); + void (*class_init)(ObjectClass *oc, void *data); +} ARMCPUInfo; + +void arm_cpu_register(const ARMCPUInfo *info); +void aarch64_cpu_register(const ARMCPUInfo *info); /** * ARMCPUClass: 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) #endif /* !defined(CONFIG_USER_ONLY) || !defined(TARGET_AARCH64) */ -struct ARMCPUInfo { - const char *name; - void (*initfn)(Object *obj); - void (*class_init)(ObjectClass *oc, void *data); -}; - static const ARMCPUInfo arm_cpus[] = { #if !defined(CONFIG_USER_ONLY) || !defined(TARGET_AARCH64) { .name = "arm926", .initfn = arm926_initfn }, @@ -XXX,XX +XXX,XX @@ static void cpu_register_class_init(ObjectClass *oc, void *data) acc->info = data; } -static void cpu_register(const ARMCPUInfo *info) +void arm_cpu_register(const ARMCPUInfo *info) { TypeInfo type_info = { .parent = TYPE_ARM_CPU, @@ -XXX,XX +XXX,XX @@ static void arm_cpu_register_types(void) type_register_static(&idau_interface_type_info); while (info->name) { - cpu_register(info); + arm_cpu_register(info); info++; } 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) cpu_max_set_sve_max_vq, NULL, NULL, &error_fatal); } -struct ARMCPUInfo { - const char *name; - void (*initfn)(Object *obj); - void (*class_init)(ObjectClass *oc, void *data); -}; - static const ARMCPUInfo aarch64_cpus[] = { { .name = "cortex-a57", .initfn = aarch64_a57_initfn }, { .name = "cortex-a53", .initfn = aarch64_a53_initfn }, @@ -XXX,XX +XXX,XX @@ static void cpu_register_class_init(ObjectClass *oc, void *data) acc->info = data; } -static void aarch64_cpu_register(const ARMCPUInfo *info) +void aarch64_cpu_register(const ARMCPUInfo *info) { TypeInfo type_info = { .parent = TYPE_AARCH64_CPU, -- 2.20.1 From: Philippe Mathieu-Daudé <philmd@redhat.com> Suggested-by: Richard Henderson <richard.henderson@linaro.org> Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20200423073358.27155-4-philmd@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.c | 8 +++----- target/arm/cpu64.c | 8 +++----- 2 files changed, 6 insertions(+), 10 deletions(-) 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[] = { { .name = "any", .initfn = arm_max_initfn }, #endif #endif - { .name = NULL } }; static Property arm_cpu_properties[] = { @@ -XXX,XX +XXX,XX @@ static const TypeInfo idau_interface_type_info = { static void arm_cpu_register_types(void) { - const ARMCPUInfo *info = arm_cpus; + size_t i; type_register_static(&arm_cpu_type_info); type_register_static(&idau_interface_type_info); - while (info->name) { - arm_cpu_register(info); - info++; + for (i = 0; i < ARRAY_SIZE(arm_cpus); ++i) { + arm_cpu_register(&arm_cpus[i]); } #ifdef CONFIG_KVM 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 const ARMCPUInfo aarch64_cpus[] = { { .name = "cortex-a53", .initfn = aarch64_a53_initfn }, { .name = "cortex-a72", .initfn = aarch64_a72_initfn }, { .name = "max", .initfn = aarch64_max_initfn }, - { .name = NULL } }; static bool aarch64_cpu_get_aarch64(Object *obj, Error **errp) @@ -XXX,XX +XXX,XX @@ static const TypeInfo aarch64_cpu_type_info = { static void aarch64_cpu_register_types(void) { - const ARMCPUInfo *info = aarch64_cpus; + size_t i; type_register_static(&aarch64_cpu_type_info); - while (info->name) { - aarch64_cpu_register(info); - info++; + for (i = 0; i < ARRAY_SIZE(aarch64_cpus); ++i) { + aarch64_cpu_register(&aarch64_cpus[i]); } } -- 2.20.1 From: Philippe Mathieu-Daudé <philmd@redhat.com> We will move this code in the next commit. Clean it up first to avoid checkpatch.pl errors. Reviewed-by: Richard Henderson <richard.henderson@linaro.org> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> Message-id: 20200423073358.27155-5-philmd@redhat.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- target/arm/cpu.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) 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 bool arm_v7m_cpu_exec_interrupt(CPUState *cs, int interrupt_request) CPUARMState *env = &cpu->env; bool ret = false; - /* ARMv7-M interrupt masking works differently than -A or -R. + /* + * ARMv7-M interrupt masking works differently than -A or -R. * There is no FIQ/IRQ distinction. Instead of I and F bits * masking FIQ and IRQ interrupts, an exception is taken only * if it is higher priority than the current execution priority @@ -XXX,XX +XXX,XX @@ static void arm1026_initfn(Object *obj) static void arm1136_r2_initfn(Object *obj) { ARMCPU *cpu = ARM_CPU(obj); - /* What qemu calls "arm1136_r2" is actually the 1136 r0p2, ie an + /* + * What qemu calls "arm1136_r2" is actually the 1136 r0p2, ie an * older core than plain "arm1136". In particular this does not * have the v6K features. * These ID register values are correct for 1136 but may be wrong @@ -XXX,XX +XXX,XX @@ static const ARMCPUInfo arm_cpus[] = { { .name = "arm926", .initfn = arm926_initfn }, { .name = "arm946", .initfn = arm946_initfn }, { .name = "arm1026", .initfn = arm1026_initfn }, - /* What QEMU calls "arm1136-r2" is actually the 1136 r0p2, i.e. an + /* + * What QEMU calls "arm1136-r2" is actually the 1136 r0p2, i.e. an * older core than plain "arm1136". In particular this does not * have the v6K features. */ -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Allow name wildcards in qemu_fdt_node_path(). This is useful to find all nodes with a given compatibility string. Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20200423121114.4274-2-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/sysemu/device_tree.h | 3 +++ device_tree.c | 2 +- 2 files changed, 4 insertions(+), 1 deletion(-) diff --git a/include/sysemu/device_tree.h b/include/sysemu/device_tree.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/device_tree.h +++ b/include/sysemu/device_tree.h @@ -XXX,XX +XXX,XX @@ void *load_device_tree_from_sysfs(void); * NULL. If there is no error but no matching node was found, the * returned array contains a single element equal to NULL. If an error * was encountered when parsing the blob, the function returns NULL + * + * @name may be NULL to wildcard names and only match compatibility + * strings. */ char **qemu_fdt_node_path(void *fdt, const char *name, char *compat, Error **errp); diff --git a/device_tree.c b/device_tree.c index XXXXXXX..XXXXXXX 100644 --- a/device_tree.c +++ b/device_tree.c @@ -XXX,XX +XXX,XX @@ char **qemu_fdt_node_path(void *fdt, const char *name, char *compat, offset = len; break; } - if (!strcmp(iter_name, name)) { + if (!name || !strcmp(iter_name, name)) { char *path; path = g_malloc(path_len); -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Make compat in qemu_fdt_node_path() const char *. Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20200423121114.4274-3-edgar.iglesias@gmail.com Reviewed-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- include/sysemu/device_tree.h | 2 +- device_tree.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/include/sysemu/device_tree.h b/include/sysemu/device_tree.h index XXXXXXX..XXXXXXX 100644 --- a/include/sysemu/device_tree.h +++ b/include/sysemu/device_tree.h @@ -XXX,XX +XXX,XX @@ void *load_device_tree_from_sysfs(void); * @name may be NULL to wildcard names and only match compatibility * strings. */ -char **qemu_fdt_node_path(void *fdt, const char *name, char *compat, +char **qemu_fdt_node_path(void *fdt, const char *name, const char *compat, Error **errp); /** diff --git a/device_tree.c b/device_tree.c index XXXXXXX..XXXXXXX 100644 --- a/device_tree.c +++ b/device_tree.c @@ -XXX,XX +XXX,XX @@ char **qemu_fdt_node_unit_path(void *fdt, const char *name, Error **errp) return path_array; } -char **qemu_fdt_node_path(void *fdt, const char *name, char *compat, +char **qemu_fdt_node_path(void *fdt, const char *name, const char *compat, Error **errp) { int offset, len, ret; -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Move arm_boot_info into XlnxZCU102. Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20200423121114.4274-4-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/xlnx-zcu102.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/hw/arm/xlnx-zcu102.c b/hw/arm/xlnx-zcu102.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/xlnx-zcu102.c +++ b/hw/arm/xlnx-zcu102.c @@ -XXX,XX +XXX,XX @@ typedef struct XlnxZCU102 { bool secure; bool virt; + + struct arm_boot_info binfo; } XlnxZCU102; #define TYPE_ZCU102_MACHINE MACHINE_TYPE_NAME("xlnx-zcu102") #define ZCU102_MACHINE(obj) \ OBJECT_CHECK(XlnxZCU102, (obj), TYPE_ZCU102_MACHINE) -static struct arm_boot_info xlnx_zcu102_binfo; static bool zcu102_get_secure(Object *obj, Error **errp) { @@ -XXX,XX +XXX,XX @@ static void xlnx_zcu102_init(MachineState *machine) /* TODO create and connect IDE devices for ide_drive_get() */ - xlnx_zcu102_binfo.ram_size = ram_size; - xlnx_zcu102_binfo.loader_start = 0; - arm_load_kernel(s->soc.boot_cpu_ptr, machine, &xlnx_zcu102_binfo); + s->binfo.ram_size = ram_size; + s->binfo.loader_start = 0; + arm_load_kernel(s->soc.boot_cpu_ptr, machine, &s->binfo); } static void xlnx_zcu102_machine_instance_init(Object *obj) -- 2.20.1 From: "Edgar E. Iglesias" <edgar.iglesias@xilinx.com> Disable unsupported FDT firmware nodes if a user passes us a DTB with nodes enabled that the machine cannot support due to lack of EL3 or EL2 support. Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Message-id: 20200423121114.4274-5-edgar.iglesias@gmail.com Signed-off-by: Peter Maydell <peter.maydell@linaro.org> --- hw/arm/xlnx-zcu102.c | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) diff --git a/hw/arm/xlnx-zcu102.c b/hw/arm/xlnx-zcu102.c index XXXXXXX..XXXXXXX 100644 --- a/hw/arm/xlnx-zcu102.c +++ b/hw/arm/xlnx-zcu102.c @@ -XXX,XX +XXX,XX @@ #include "qemu/error-report.h" #include "qemu/log.h" #include "sysemu/qtest.h" +#include "sysemu/device_tree.h" typedef struct XlnxZCU102 { MachineState parent_obj; @@ -XXX,XX +XXX,XX @@ static void zcu102_set_virt(Object *obj, bool value, Error **errp) s->virt = value; } +static void zcu102_modify_dtb(const struct arm_boot_info *binfo, void *fdt) +{ + XlnxZCU102 *s = container_of(binfo, XlnxZCU102, binfo); + bool method_is_hvc; + char **node_path; + const char *r; + int prop_len; + int i; + + /* If EL3 is enabled, we keep all firmware nodes active. */ + if (!s->secure) { + node_path = qemu_fdt_node_path(fdt, NULL, "xlnx,zynqmp-firmware", + &error_fatal); + + for (i = 0; node_path && node_path[i]; i++) { + r = qemu_fdt_getprop(fdt, node_path[i], "method", &prop_len, NULL); + method_is_hvc = r && !strcmp("hvc", r); + + /* Allow HVC based firmware if EL2 is enabled. */ + if (method_is_hvc && s->virt) { + continue; + } + qemu_fdt_setprop_string(fdt, node_path[i], "status", "disabled"); + } + g_strfreev(node_path); + } +} + static void xlnx_zcu102_init(MachineState *machine) { XlnxZCU102 *s = ZCU102_MACHINE(machine); @@ -XXX,XX +XXX,XX @@ static void xlnx_zcu102_init(MachineState *machine) s->binfo.ram_size = ram_size; s->binfo.loader_start = 0; + s->binfo.modify_dtb = zcu102_modify_dtb; arm_load_kernel(s->soc.boot_cpu_ptr, machine, &s->binfo); } -- 2.20.1