diff options
Diffstat (limited to 'target')
33 files changed, 1393 insertions, 984 deletions
diff --git a/target/i386/cpu.c b/target/i386/cpu.c index 1e6a7d0a75..b0b87c3d81 100644 --- a/target/i386/cpu.c +++ b/target/i386/cpu.c @@ -18,6 +18,7 @@ */ #include "qemu/osdep.h" +#include "qemu/units.h" #include "qemu/cutils.h" #include "qemu/bitops.h" @@ -65,9 +66,6 @@ struct CPUID2CacheDescriptorInfo { int associativity; }; -#define KiB 1024 -#define MiB (1024 * 1024) - /* * Known CPUID 2 cache descriptors. * From Intel SDM Volume 2A, CPUID instruction @@ -751,7 +749,7 @@ static void x86_cpu_vendor_words2str(char *dst, uint32_t vendor1, #define TCG_EXT3_FEATURES (CPUID_EXT3_LAHF_LM | CPUID_EXT3_SVM | \ CPUID_EXT3_CR8LEG | CPUID_EXT3_ABM | CPUID_EXT3_SSE4A) #define TCG_EXT4_FEATURES 0 -#define TCG_SVM_FEATURES 0 +#define TCG_SVM_FEATURES CPUID_SVM_NPT #define TCG_KVM_FEATURES 0 #define TCG_7_0_EBX_FEATURES (CPUID_7_0_EBX_SMEP | CPUID_7_0_EBX_SMAP | \ CPUID_7_0_EBX_BMI1 | CPUID_7_0_EBX_BMI2 | CPUID_7_0_EBX_ADX | \ @@ -5415,6 +5413,7 @@ static Property x86_cpu_properties[] = { DEFINE_PROP_BOOL("hv-stimer", X86CPU, hyperv_stimer, false), DEFINE_PROP_BOOL("hv-frequencies", X86CPU, hyperv_frequencies, false), DEFINE_PROP_BOOL("hv-reenlightenment", X86CPU, hyperv_reenlightenment, false), + DEFINE_PROP_BOOL("hv-tlbflush", X86CPU, hyperv_tlbflush, false), DEFINE_PROP_BOOL("check", X86CPU, check_cpuid, true), DEFINE_PROP_BOOL("enforce", X86CPU, enforce_cpuid, false), DEFINE_PROP_BOOL("kvm", X86CPU, expose_kvm, true), diff --git a/target/i386/cpu.h b/target/i386/cpu.h index 8eaefeee3e..2c5a0d90a6 100644 --- a/target/i386/cpu.h +++ b/target/i386/cpu.h @@ -211,6 +211,7 @@ typedef enum X86Seg { #define HF2_VINTR_SHIFT 3 /* value of V_INTR_MASKING bit */ #define HF2_SMM_INSIDE_NMI_SHIFT 4 /* CPU serving SMI nested inside NMI */ #define HF2_MPX_PR_SHIFT 5 /* BNDCFGx.BNDPRESERVE */ +#define HF2_NPT_SHIFT 6 /* Nested Paging enabled */ #define HF2_GIF_MASK (1 << HF2_GIF_SHIFT) #define HF2_HIF_MASK (1 << HF2_HIF_SHIFT) @@ -218,6 +219,7 @@ typedef enum X86Seg { #define HF2_VINTR_MASK (1 << HF2_VINTR_SHIFT) #define HF2_SMM_INSIDE_NMI_MASK (1 << HF2_SMM_INSIDE_NMI_SHIFT) #define HF2_MPX_PR_MASK (1 << HF2_MPX_PR_SHIFT) +#define HF2_NPT_MASK (1 << HF2_NPT_SHIFT) #define CR0_PE_SHIFT 0 #define CR0_MP_SHIFT 1 @@ -1265,12 +1267,16 @@ typedef struct CPUX86State { uint16_t intercept_dr_read; uint16_t intercept_dr_write; uint32_t intercept_exceptions; + uint64_t nested_cr3; + uint32_t nested_pg_mode; uint8_t v_tpr; /* KVM states, automatically cleared on reset */ uint8_t nmi_injected; uint8_t nmi_pending; + uintptr_t retaddr; + /* Fields up to this point are cleared by a CPU reset */ struct {} end_reset_fields; @@ -1367,6 +1373,7 @@ struct X86CPU { bool hyperv_stimer; bool hyperv_frequencies; bool hyperv_reenlightenment; + bool hyperv_tlbflush; bool check_cpuid; bool enforce_cpuid; bool expose_kvm; diff --git a/target/i386/excp_helper.c b/target/i386/excp_helper.c index cb4d1b7d33..37a33d5ae0 100644 --- a/target/i386/excp_helper.c +++ b/target/i386/excp_helper.c @@ -157,6 +157,209 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr, int size, #else +static hwaddr get_hphys(CPUState *cs, hwaddr gphys, MMUAccessType access_type, + int *prot) +{ + CPUX86State *env = &X86_CPU(cs)->env; + uint64_t rsvd_mask = PG_HI_RSVD_MASK; + uint64_t ptep, pte; + uint64_t exit_info_1 = 0; + target_ulong pde_addr, pte_addr; + uint32_t page_offset; + int page_size; + + if (likely(!(env->hflags2 & HF2_NPT_MASK))) { + return gphys; + } + + if (!(env->nested_pg_mode & SVM_NPT_NXE)) { + rsvd_mask |= PG_NX_MASK; + } + + if (env->nested_pg_mode & SVM_NPT_PAE) { + uint64_t pde, pdpe; + target_ulong pdpe_addr; + +#ifdef TARGET_X86_64 + if (env->nested_pg_mode & SVM_NPT_LMA) { + uint64_t pml5e; + uint64_t pml4e_addr, pml4e; + + pml5e = env->nested_cr3; + ptep = PG_NX_MASK | PG_USER_MASK | PG_RW_MASK; + + pml4e_addr = (pml5e & PG_ADDRESS_MASK) + + (((gphys >> 39) & 0x1ff) << 3); + pml4e = x86_ldq_phys(cs, pml4e_addr); + if (!(pml4e & PG_PRESENT_MASK)) { + goto do_fault; + } + if (pml4e & (rsvd_mask | PG_PSE_MASK)) { + goto do_fault_rsvd; + } + if (!(pml4e & PG_ACCESSED_MASK)) { + pml4e |= PG_ACCESSED_MASK; + x86_stl_phys_notdirty(cs, pml4e_addr, pml4e); + } + ptep &= pml4e ^ PG_NX_MASK; + pdpe_addr = (pml4e & PG_ADDRESS_MASK) + + (((gphys >> 30) & 0x1ff) << 3); + pdpe = x86_ldq_phys(cs, pdpe_addr); + if (!(pdpe & PG_PRESENT_MASK)) { + goto do_fault; + } + if (pdpe & rsvd_mask) { + goto do_fault_rsvd; + } + ptep &= pdpe ^ PG_NX_MASK; + if (!(pdpe & PG_ACCESSED_MASK)) { + pdpe |= PG_ACCESSED_MASK; + x86_stl_phys_notdirty(cs, pdpe_addr, pdpe); + } + if (pdpe & PG_PSE_MASK) { + /* 1 GB page */ + page_size = 1024 * 1024 * 1024; + pte_addr = pdpe_addr; + pte = pdpe; + goto do_check_protect; + } + } else +#endif + { + pdpe_addr = (env->nested_cr3 & ~0x1f) + ((gphys >> 27) & 0x18); + pdpe = x86_ldq_phys(cs, pdpe_addr); + if (!(pdpe & PG_PRESENT_MASK)) { + goto do_fault; + } + rsvd_mask |= PG_HI_USER_MASK; + if (pdpe & (rsvd_mask | PG_NX_MASK)) { + goto do_fault_rsvd; + } + ptep = PG_NX_MASK | PG_USER_MASK | PG_RW_MASK; + } + + pde_addr = (pdpe & PG_ADDRESS_MASK) + (((gphys >> 21) & 0x1ff) << 3); + pde = x86_ldq_phys(cs, pde_addr); + if (!(pde & PG_PRESENT_MASK)) { + goto do_fault; + } + if (pde & rsvd_mask) { + goto do_fault_rsvd; + } + ptep &= pde ^ PG_NX_MASK; + if (pde & PG_PSE_MASK) { + /* 2 MB page */ + page_size = 2048 * 1024; + pte_addr = pde_addr; + pte = pde; + goto do_check_protect; + } + /* 4 KB page */ + if (!(pde & PG_ACCESSED_MASK)) { + pde |= PG_ACCESSED_MASK; + x86_stl_phys_notdirty(cs, pde_addr, pde); + } + pte_addr = (pde & PG_ADDRESS_MASK) + (((gphys >> 12) & 0x1ff) << 3); + pte = x86_ldq_phys(cs, pte_addr); + if (!(pte & PG_PRESENT_MASK)) { + goto do_fault; + } + if (pte & rsvd_mask) { + goto do_fault_rsvd; + } + /* combine pde and pte nx, user and rw protections */ + ptep &= pte ^ PG_NX_MASK; + page_size = 4096; + } else { + uint32_t pde; + + /* page directory entry */ + pde_addr = (env->nested_cr3 & ~0xfff) + ((gphys >> 20) & 0xffc); + pde = x86_ldl_phys(cs, pde_addr); + if (!(pde & PG_PRESENT_MASK)) { + goto do_fault; + } + ptep = pde | PG_NX_MASK; + + /* if PSE bit is set, then we use a 4MB page */ + if ((pde & PG_PSE_MASK) && (env->cr[4] & CR4_PSE_MASK)) { + page_size = 4096 * 1024; + pte_addr = pde_addr; + + /* Bits 20-13 provide bits 39-32 of the address, bit 21 is reserved. + * Leave bits 20-13 in place for setting accessed/dirty bits below. + */ + pte = pde | ((pde & 0x1fe000LL) << (32 - 13)); + rsvd_mask = 0x200000; + goto do_check_protect_pse36; + } + + if (!(pde & PG_ACCESSED_MASK)) { + pde |= PG_ACCESSED_MASK; + x86_stl_phys_notdirty(cs, pde_addr, pde); + } + + /* page directory entry */ + pte_addr = (pde & ~0xfff) + ((gphys >> 10) & 0xffc); + pte = x86_ldl_phys(cs, pte_addr); + if (!(pte & PG_PRESENT_MASK)) { + goto do_fault; + } + /* combine pde and pte user and rw protections */ + ptep &= pte | PG_NX_MASK; + page_size = 4096; + rsvd_mask = 0; + } + + do_check_protect: + rsvd_mask |= (page_size - 1) & PG_ADDRESS_MASK & ~PG_PSE_PAT_MASK; + do_check_protect_pse36: + if (pte & rsvd_mask) { + goto do_fault_rsvd; + } + ptep ^= PG_NX_MASK; + + if (!(ptep & PG_USER_MASK)) { + goto do_fault_protect; + } + if (ptep & PG_NX_MASK) { + if (access_type == MMU_INST_FETCH) { + goto do_fault_protect; + } + *prot &= ~PAGE_EXEC; + } + if (!(ptep & PG_RW_MASK)) { + if (access_type == MMU_DATA_STORE) { + goto do_fault_protect; + } + *prot &= ~PAGE_WRITE; + } + + pte &= PG_ADDRESS_MASK & ~(page_size - 1); + page_offset = gphys & (page_size - 1); + return pte + page_offset; + + do_fault_rsvd: + exit_info_1 |= SVM_NPTEXIT_RSVD; + do_fault_protect: + exit_info_1 |= SVM_NPTEXIT_P; + do_fault: + x86_stq_phys(cs, env->vm_vmcb + offsetof(struct vmcb, control.exit_info_2), + gphys); + exit_info_1 |= SVM_NPTEXIT_US; + if (access_type == MMU_DATA_STORE) { + exit_info_1 |= SVM_NPTEXIT_RW; + } else if (access_type == MMU_INST_FETCH) { + exit_info_1 |= SVM_NPTEXIT_ID; + } + if (prot) { + exit_info_1 |= SVM_NPTEXIT_GPA; + } else { /* page table access */ + exit_info_1 |= SVM_NPTEXIT_GPT; + } + cpu_vmexit(env, SVM_EXIT_NPF, exit_info_1, env->retaddr); +} + /* return value: * -1 = cannot handle fault * 0 = nothing more to do @@ -224,6 +427,7 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr, int size, if (la57) { pml5e_addr = ((env->cr[3] & ~0xfff) + (((addr >> 48) & 0x1ff) << 3)) & a20_mask; + pml5e_addr = get_hphys(cs, pml5e_addr, MMU_DATA_STORE, NULL); pml5e = x86_ldq_phys(cs, pml5e_addr); if (!(pml5e & PG_PRESENT_MASK)) { goto do_fault; @@ -243,6 +447,7 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr, int size, pml4e_addr = ((pml5e & PG_ADDRESS_MASK) + (((addr >> 39) & 0x1ff) << 3)) & a20_mask; + pml4e_addr = get_hphys(cs, pml4e_addr, MMU_DATA_STORE, false); pml4e = x86_ldq_phys(cs, pml4e_addr); if (!(pml4e & PG_PRESENT_MASK)) { goto do_fault; @@ -257,6 +462,7 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr, int size, ptep &= pml4e ^ PG_NX_MASK; pdpe_addr = ((pml4e & PG_ADDRESS_MASK) + (((addr >> 30) & 0x1ff) << 3)) & a20_mask; + pdpe_addr = get_hphys(cs, pdpe_addr, MMU_DATA_STORE, NULL); pdpe = x86_ldq_phys(cs, pdpe_addr); if (!(pdpe & PG_PRESENT_MASK)) { goto do_fault; @@ -282,6 +488,7 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr, int size, /* XXX: load them when cr3 is loaded ? */ pdpe_addr = ((env->cr[3] & ~0x1f) + ((addr >> 27) & 0x18)) & a20_mask; + pdpe_addr = get_hphys(cs, pdpe_addr, MMU_DATA_STORE, false); pdpe = x86_ldq_phys(cs, pdpe_addr); if (!(pdpe & PG_PRESENT_MASK)) { goto do_fault; @@ -295,6 +502,7 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr, int size, pde_addr = ((pdpe & PG_ADDRESS_MASK) + (((addr >> 21) & 0x1ff) << 3)) & a20_mask; + pde_addr = get_hphys(cs, pde_addr, MMU_DATA_STORE, NULL); pde = x86_ldq_phys(cs, pde_addr); if (!(pde & PG_PRESENT_MASK)) { goto do_fault; @@ -317,6 +525,7 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr, int size, } pte_addr = ((pde & PG_ADDRESS_MASK) + (((addr >> 12) & 0x1ff) << 3)) & a20_mask; + pte_addr = get_hphys(cs, pte_addr, MMU_DATA_STORE, NULL); pte = x86_ldq_phys(cs, pte_addr); if (!(pte & PG_PRESENT_MASK)) { goto do_fault; @@ -333,6 +542,7 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr, int size, /* page directory entry */ pde_addr = ((env->cr[3] & ~0xfff) + ((addr >> 20) & 0xffc)) & a20_mask; + pde_addr = get_hphys(cs, pde_addr, MMU_DATA_STORE, NULL); pde = x86_ldl_phys(cs, pde_addr); if (!(pde & PG_PRESENT_MASK)) { goto do_fault; @@ -360,6 +570,7 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr, int size, /* page directory entry */ pte_addr = ((pde & ~0xfff) + ((addr >> 10) & 0xffc)) & a20_mask; + pte_addr = get_hphys(cs, pte_addr, MMU_DATA_STORE, NULL); pte = x86_ldl_phys(cs, pte_addr); if (!(pte & PG_PRESENT_MASK)) { goto do_fault; @@ -442,12 +653,13 @@ do_check_protect_pse36: /* align to page_size */ pte &= PG_ADDRESS_MASK & ~(page_size - 1); + page_offset = addr & (page_size - 1); + paddr = get_hphys(cs, pte + page_offset, is_write1, &prot); /* Even if 4MB pages, we map only one 4KB page in the cache to avoid filling it too fast */ vaddr = addr & TARGET_PAGE_MASK; - page_offset = vaddr & (page_size - 1); - paddr = pte + page_offset; + paddr &= TARGET_PAGE_MASK; assert(prot & (1 << is_write1)); tlb_set_page_with_attrs(cs, vaddr, paddr, cpu_get_mem_attrs(env), diff --git a/target/i386/hyperv-proto.h b/target/i386/hyperv-proto.h index 93352ebd2a..d6d5a79293 100644 --- a/target/i386/hyperv-proto.h +++ b/target/i386/hyperv-proto.h @@ -58,6 +58,7 @@ #define HV_APIC_ACCESS_RECOMMENDED (1u << 3) #define HV_SYSTEM_RESET_RECOMMENDED (1u << 4) #define HV_RELAXED_TIMING_RECOMMENDED (1u << 5) +#define HV_EX_PROCESSOR_MASKS_RECOMMENDED (1u << 11) /* * Basic virtualized MSRs diff --git a/target/i386/kvm.c b/target/i386/kvm.c index 032f0ad2fc..ebb2d23aa4 100644 --- a/target/i386/kvm.c +++ b/target/i386/kvm.c @@ -601,7 +601,8 @@ static bool hyperv_enabled(X86CPU *cpu) cpu->hyperv_runtime || cpu->hyperv_synic || cpu->hyperv_stimer || - cpu->hyperv_reenlightenment); + cpu->hyperv_reenlightenment || + cpu->hyperv_tlbflush); } static int kvm_arch_set_tsc_khz(CPUState *cs) @@ -839,6 +840,18 @@ int kvm_arch_init_vcpu(CPUState *cs) if (cpu->hyperv_vapic) { c->eax |= HV_APIC_ACCESS_RECOMMENDED; } + if (cpu->hyperv_tlbflush) { + if (kvm_check_extension(cs->kvm_state, + KVM_CAP_HYPERV_TLBFLUSH) <= 0) { + fprintf(stderr, "Hyper-V TLB flush support " + "(requested by 'hv-tlbflush' cpu flag) " + " is not supported by kernel\n"); + return -ENOSYS; + } + c->eax |= HV_REMOTE_TLB_FLUSH_RECOMMENDED; + c->eax |= HV_EX_PROCESSOR_MASKS_RECOMMENDED; + } + c->ebx = cpu->hyperv_spinlock_attempts; c = &cpuid_data.entries[cpuid_i++]; diff --git a/target/i386/machine.c b/target/i386/machine.c index 4d98d367c1..8b64dff487 100644 --- a/target/i386/machine.c +++ b/target/i386/machine.c @@ -935,6 +935,26 @@ static const VMStateDescription vmstate_msr_virt_ssbd = { } }; +static bool svm_npt_needed(void *opaque) +{ + X86CPU *cpu = opaque; + CPUX86State *env = &cpu->env; + + return !!(env->hflags2 & HF2_NPT_MASK); +} + +static const VMStateDescription vmstate_svm_npt = { + .name = "cpu/svn_npt", + .version_id = 1, + .minimum_version_id = 1, + .needed = svm_npt_needed, + .fields = (VMStateField[]){ + VMSTATE_UINT64(env.nested_cr3, X86CPU), + VMSTATE_UINT32(env.nested_pg_mode, X86CPU), + VMSTATE_END_OF_LIST() + } +}; + VMStateDescription vmstate_x86_cpu = { .name = "cpu", .version_id = 12, @@ -1059,6 +1079,7 @@ VMStateDescription vmstate_x86_cpu = { &vmstate_mcg_ext_ctl, &vmstate_msr_intel_pt, &vmstate_msr_virt_ssbd, + &vmstate_svm_npt, NULL } }; diff --git a/target/i386/mem_helper.c b/target/i386/mem_helper.c index a8ae694a9c..30c26b9d9c 100644 --- a/target/i386/mem_helper.c +++ b/target/i386/mem_helper.c @@ -202,13 +202,13 @@ void helper_boundl(CPUX86State *env, target_ulong a0, int v) void tlb_fill(CPUState *cs, target_ulong addr, int size, MMUAccessType access_type, int mmu_idx, uintptr_t retaddr) { + X86CPU *cpu = X86_CPU(cs); + CPUX86State *env = &cpu->env; int ret; + env->retaddr = retaddr; ret = x86_cpu_handle_mmu_fault(cs, addr, size, access_type, mmu_idx); if (ret) { - X86CPU *cpu = X86_CPU(cs); - CPUX86State *env = &cpu->env; - raise_exception_err_ra(env, cs->exception_index, env->error_code, retaddr); } } diff --git a/target/i386/monitor.c b/target/i386/monitor.c index 6bbb3a96cd..74a13c571b 100644 --- a/target/i386/monitor.c +++ b/target/i386/monitor.c @@ -35,21 +35,28 @@ #include "sev_i386.h" #include "qapi/qapi-commands-misc.h" - -static void print_pte(Monitor *mon, CPUArchState *env, hwaddr addr, - hwaddr pte, hwaddr mask) +/* Perform linear address sign extension */ +static hwaddr addr_canonical(CPUArchState *env, hwaddr addr) { #ifdef TARGET_X86_64 if (env->cr[4] & CR4_LA57_MASK) { if (addr & (1ULL << 56)) { - addr |= -1LL << 57; + addr |= (hwaddr)-(1LL << 57); } } else { if (addr & (1ULL << 47)) { - addr |= -1LL << 48; + addr |= (hwaddr)-(1LL << 48); } } #endif + return addr; +} + +static void print_pte(Monitor *mon, CPUArchState *env, hwaddr addr, + hwaddr pte, hwaddr mask) +{ + addr = addr_canonical(env, addr); + monitor_printf(mon, TARGET_FMT_plx ": " TARGET_FMT_plx " %c%c%c%c%c%c%c%c%c\n", addr, @@ -243,8 +250,8 @@ void hmp_info_tlb(Monitor *mon, const QDict *qdict) } } -static void mem_print(Monitor *mon, hwaddr *pstart, - int *plast_prot, +static void mem_print(Monitor *mon, CPUArchState *env, + hwaddr *pstart, int *plast_prot, hwaddr end, int prot) { int prot1; @@ -253,7 +260,9 @@ static void mem_print(Monitor *mon, hwaddr *pstart, if (*pstart != -1) { monitor_printf(mon, TARGET_FMT_plx "-" TARGET_FMT_plx " " TARGET_FMT_plx " %c%c%c\n", - *pstart, end, end - *pstart, + addr_canonical(env, *pstart), + addr_canonical(env, end), + addr_canonical(env, end - *pstart), prot1 & PG_USER_MASK ? 'u' : '-', 'r', prot1 & PG_RW_MASK ? 'w' : '-'); @@ -283,7 +292,7 @@ static void mem_info_32(Monitor *mon, CPUArchState *env) if (pde & PG_PRESENT_MASK) { if ((pde & PG_PSE_MASK) && (env->cr[4] & CR4_PSE_MASK)) { prot = pde & (PG_USER_MASK | PG_RW_MASK | PG_PRESENT_MASK); - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); } else { for(l2 = 0; l2 < 1024; l2++) { cpu_physical_memory_read((pde & ~0xfff) + l2 * 4, &pte, 4); @@ -295,16 +304,16 @@ static void mem_info_32(Monitor *mon, CPUArchState *env) } else { prot = 0; } - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); } } } else { prot = 0; - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); } } /* Flush last range */ - mem_print(mon, &start, &last_prot, (hwaddr)1 << 32, 0); + mem_print(mon, env, &start, &last_prot, (hwaddr)1 << 32, 0); } static void mem_info_pae32(Monitor *mon, CPUArchState *env) @@ -332,7 +341,7 @@ static void mem_info_pae32(Monitor *mon, CPUArchState *env) if (pde & PG_PSE_MASK) { prot = pde & (PG_USER_MASK | PG_RW_MASK | PG_PRESENT_MASK); - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); } else { pt_addr = pde & 0x3fffffffff000ULL; for (l3 = 0; l3 < 512; l3++) { @@ -345,21 +354,21 @@ static void mem_info_pae32(Monitor *mon, CPUArchState *env) } else { prot = 0; } - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); } } } else { prot = 0; - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); } } } else { prot = 0; - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); } } /* Flush last range */ - mem_print(mon, &start, &last_prot, (hwaddr)1 << 32, 0); + mem_print(mon, env, &start, &last_prot, (hwaddr)1 << 32, 0); } @@ -389,7 +398,7 @@ static void mem_info_la48(Monitor *mon, CPUArchState *env) prot = pdpe & (PG_USER_MASK | PG_RW_MASK | PG_PRESENT_MASK); prot &= pml4e; - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); } else { pd_addr = pdpe & 0x3fffffffff000ULL; for (l3 = 0; l3 < 512; l3++) { @@ -401,7 +410,8 @@ static void mem_info_la48(Monitor *mon, CPUArchState *env) prot = pde & (PG_USER_MASK | PG_RW_MASK | PG_PRESENT_MASK); prot &= pml4e & pdpe; - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, + &last_prot, end, prot); } else { pt_addr = pde & 0x3fffffffff000ULL; for (l4 = 0; l4 < 512; l4++) { @@ -418,27 +428,29 @@ static void mem_info_la48(Monitor *mon, CPUArchState *env) } else { prot = 0; } - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, + &last_prot, end, prot); } } } else { prot = 0; - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, + &last_prot, end, prot); } } } } else { prot = 0; - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); } } } else { prot = 0; - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); } } /* Flush last range */ - mem_print(mon, &start, &last_prot, (hwaddr)1 << 48, 0); + mem_print(mon, env, &start, &last_prot, (hwaddr)1 << 48, 0); } static void mem_info_la57(Monitor *mon, CPUArchState *env) @@ -457,7 +469,7 @@ static void mem_info_la57(Monitor *mon, CPUArchState *env) end = l0 << 48; if (!(pml5e & PG_PRESENT_MASK)) { prot = 0; - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); continue; } @@ -468,7 +480,7 @@ static void mem_info_la57(Monitor *mon, CPUArchState *env) end = (l0 << 48) + (l1 << 39); if (!(pml4e & PG_PRESENT_MASK)) { prot = 0; - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); continue; } @@ -479,7 +491,7 @@ static void mem_info_la57(Monitor *mon, CPUArchState *env) end = (l0 << 48) + (l1 << 39) + (l2 << 30); if (pdpe & PG_PRESENT_MASK) { prot = 0; - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); continue; } @@ -487,7 +499,7 @@ static void mem_info_la57(Monitor *mon, CPUArchState *env) prot = pdpe & (PG_USER_MASK | PG_RW_MASK | PG_PRESENT_MASK); prot &= pml5e & pml4e; - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); continue; } @@ -498,7 +510,7 @@ static void mem_info_la57(Monitor *mon, CPUArchState *env) end = (l0 << 48) + (l1 << 39) + (l2 << 30) + (l3 << 21); if (pde & PG_PRESENT_MASK) { prot = 0; - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); continue; } @@ -506,7 +518,7 @@ static void mem_info_la57(Monitor *mon, CPUArchState *env) prot = pde & (PG_USER_MASK | PG_RW_MASK | PG_PRESENT_MASK); prot &= pml5e & pml4e & pdpe; - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); continue; } @@ -523,14 +535,14 @@ static void mem_info_la57(Monitor *mon, CPUArchState *env) } else { prot = 0; } - mem_print(mon, &start, &last_prot, end, prot); + mem_print(mon, env, &start, &last_prot, end, prot); } } } } } /* Flush last range */ - mem_print(mon, &start, &last_prot, (hwaddr)1 << 57, 0); + mem_print(mon, env, &start, &last_prot, (hwaddr)1 << 57, 0); } #endif /* TARGET_X86_64 */ diff --git a/target/i386/svm.h b/target/i386/svm.h index 922c8fd39c..23a3a040b8 100644 --- a/target/i386/svm.h +++ b/target/i386/svm.h @@ -130,6 +130,20 @@ #define SVM_CR0_SELECTIVE_MASK (1 << 3 | 1) /* TS and MP */ +#define SVM_NPT_ENABLED (1 << 0) + +#define SVM_NPT_PAE (1 << 0) +#define SVM_NPT_LMA (1 << 1) +#define SVM_NPT_NXE (1 << 2) + +#define SVM_NPTEXIT_P (1ULL << 0) +#define SVM_NPTEXIT_RW (1ULL << 1) +#define SVM_NPTEXIT_US (1ULL << 2) +#define SVM_NPTEXIT_RSVD (1ULL << 3) +#define SVM_NPTEXIT_ID (1ULL << 4) +#define SVM_NPTEXIT_GPA (1ULL << 32) +#define SVM_NPTEXIT_GPT (1ULL << 33) + struct QEMU_PACKED vmcb_control_area { uint16_t intercept_cr_read; uint16_t intercept_cr_write; diff --git a/target/i386/svm_helper.c b/target/i386/svm_helper.c index f245aec310..342ece082f 100644 --- a/target/i386/svm_helper.c +++ b/target/i386/svm_helper.c @@ -124,6 +124,7 @@ void helper_vmrun(CPUX86State *env, int aflag, int next_eip_addend) { CPUState *cs = CPU(x86_env_get_cpu(env)); target_ulong addr; + uint64_t nested_ctl; uint32_t event_inj; uint32_t int_ctl; @@ -206,6 +207,26 @@ void helper_vmrun(CPUX86State *env, int aflag, int next_eip_addend) control.intercept_exceptions )); + nested_ctl = x86_ldq_phys(cs, env->vm_vmcb + offsetof(struct vmcb, + control.nested_ctl)); + if (nested_ctl & SVM_NPT_ENABLED) { + env->nested_cr3 = x86_ldq_phys(cs, + env->vm_vmcb + offsetof(struct vmcb, + control.nested_cr3)); + env->hflags2 |= HF2_NPT_MASK; + + env->nested_pg_mode = 0; + if (env->cr[4] & CR4_PAE_MASK) { + env->nested_pg_mode |= SVM_NPT_PAE; + } + if (env->hflags & HF_LMA_MASK) { + env->nested_pg_mode |= SVM_NPT_LMA; + } + if (env->efer & MSR_EFER_NXE) { + env->nested_pg_mode |= SVM_NPT_NXE; + } + } + /* enable intercepts */ env->hflags |= HF_SVMI_MASK; @@ -616,6 +637,7 @@ void do_vmexit(CPUX86State *env, uint32_t exit_code, uint64_t exit_info_1) x86_stl_phys(cs, env->vm_vmcb + offsetof(struct vmcb, control.int_state), 0); } + env->hflags2 &= ~HF2_NPT_MASK; /* Save the VM state in the vmcb */ svm_save_seg(env, env->vm_vmcb + offsetof(struct vmcb, save.es), diff --git a/target/openrisc/Makefile.objs b/target/openrisc/Makefile.objs index 1b98a911ea..b5432f4684 100644 --- a/target/openrisc/Makefile.objs +++ b/target/openrisc/Makefile.objs @@ -1,7 +1,7 @@ obj-$(CONFIG_SOFTMMU) += machine.o -obj-y += cpu.o exception.o interrupt.o mmu.o translate.o +obj-y += cpu.o exception.o interrupt.o mmu.o translate.o disas.o obj-y += exception_helper.o fpu_helper.o \ - interrupt_helper.o mmu_helper.o sys_helper.o + interrupt_helper.o sys_helper.o obj-y += gdbstub.o DECODETREE = $(SRC_PATH)/scripts/decodetree.py @@ -12,3 +12,4 @@ target/openrisc/decode.inc.c: \ $(PYTHON) $(DECODETREE) -o $@ $<, "GEN", $(TARGET_DIR)$@) target/openrisc/translate.o: target/openrisc/decode.inc.c +target/openrisc/disas.o: target/openrisc/decode.inc.c diff --git a/target/openrisc/cpu.c b/target/openrisc/cpu.c index a692a98ec0..fb7cb5c507 100644 --- a/target/openrisc/cpu.c +++ b/target/openrisc/cpu.c @@ -27,6 +27,7 @@ static void openrisc_cpu_set_pc(CPUState *cs, vaddr value) OpenRISCCPU *cpu = OPENRISC_CPU(cs); cpu->env.pc = value; + cpu->env.dflag = 0; } static bool openrisc_cpu_has_work(CPUState *cs) @@ -35,6 +36,11 @@ static bool openrisc_cpu_has_work(CPUState *cs) CPU_INTERRUPT_TIMER); } +static void openrisc_disas_set_info(CPUState *cpu, disassemble_info *info) +{ + info->print_insn = print_insn_or1k; +} + /* CPUClass::reset() */ static void openrisc_cpu_reset(CPUState *s) { @@ -52,8 +58,10 @@ static void openrisc_cpu_reset(CPUState *s) cpu->env.upr = UPR_UP | UPR_DMP | UPR_IMP | UPR_PICP | UPR_TTP | UPR_PMP; - cpu->env.dmmucfgr = (DMMUCFGR_NTW & (0 << 2)) | (DMMUCFGR_NTS & (6 << 2)); - cpu->env.immucfgr = (IMMUCFGR_NTW & (0 << 2)) | (IMMUCFGR_NTS & (6 << 2)); + cpu->env.dmmucfgr = (DMMUCFGR_NTW & (0 << 2)) + | (DMMUCFGR_NTS & (ctz32(TLB_SIZE) << 2)); + cpu->env.immucfgr = (IMMUCFGR_NTW & (0 << 2)) + | (IMMUCFGR_NTS & (ctz32(TLB_SIZE) << 2)); #ifndef CONFIG_USER_ONLY cpu->env.picmr = 0x00000000; @@ -87,10 +95,6 @@ static void openrisc_cpu_initfn(Object *obj) OpenRISCCPU *cpu = OPENRISC_CPU(obj); cs->env_ptr = &cpu->env; - -#ifndef CONFIG_USER_ONLY - cpu_openrisc_mmu_init(cpu); -#endif } /* CPU models */ @@ -152,6 +156,7 @@ static void openrisc_cpu_class_init(ObjectClass *oc, void *data) #endif cc->gdb_num_core_regs = 32 + 3; cc->tcg_initialize = openrisc_translate_init; + cc->disas_set_info = openrisc_disas_set_info; } /* Sort alphabetically by type name, except for "any". */ diff --git a/target/openrisc/cpu.h b/target/openrisc/cpu.h index 35cab65f11..f1b31bc24a 100644 --- a/target/openrisc/cpu.h +++ b/target/openrisc/cpu.h @@ -222,12 +222,8 @@ enum { /* TLB size */ enum { - DTLB_WAYS = 1, - DTLB_SIZE = 64, - DTLB_MASK = (DTLB_SIZE-1), - ITLB_WAYS = 1, - ITLB_SIZE = 64, - ITLB_MASK = (ITLB_SIZE-1), + TLB_SIZE = 128, + TLB_MASK = TLB_SIZE - 1, }; /* TLB prot */ @@ -241,14 +237,6 @@ enum { UXE = (1 << 7), }; -/* check if tlb available */ -enum { - TLBRET_INVALID = -3, - TLBRET_NOMATCH = -2, - TLBRET_BADADDR = -1, - TLBRET_MATCH = 0 -}; - typedef struct OpenRISCTLBEntry { uint32_t mr; uint32_t tr; @@ -256,8 +244,8 @@ typedef struct OpenRISCTLBEntry { #ifndef CONFIG_USER_ONLY typedef struct CPUOpenRISCTLBContext { - OpenRISCTLBEntry itlb[ITLB_WAYS][ITLB_SIZE]; - OpenRISCTLBEntry dtlb[DTLB_WAYS][DTLB_SIZE]; + OpenRISCTLBEntry itlb[TLB_SIZE]; + OpenRISCTLBEntry dtlb[TLB_SIZE]; int (*cpu_openrisc_map_address_code)(struct OpenRISCCPU *cpu, hwaddr *physical, @@ -301,6 +289,10 @@ typedef struct CPUOpenRISCState { uint32_t dflag; /* In delay slot (boolean) */ +#ifndef CONFIG_USER_ONLY + CPUOpenRISCTLBContext tlb; +#endif + /* Fields up to this point are cleared by a CPU reset */ struct {} end_reset_fields; @@ -310,8 +302,6 @@ typedef struct CPUOpenRISCState { uint32_t cpucfgr; /* CPU configure register */ #ifndef CONFIG_USER_ONLY - CPUOpenRISCTLBContext * tlb; - QEMUTimer *timer; uint32_t ttmr; /* Timer tick mode register */ int is_counting; @@ -358,6 +348,7 @@ void openrisc_translate_init(void); int openrisc_cpu_handle_mmu_fault(CPUState *cpu, vaddr address, int size, int rw, int mmu_idx); int cpu_openrisc_signal_handler(int host_signum, void *pinfo, void *puc); +int print_insn_or1k(bfd_vma addr, disassemble_info *info); #define cpu_list cpu_openrisc_list #define cpu_signal_handler cpu_openrisc_signal_handler @@ -376,17 +367,6 @@ void cpu_openrisc_count_update(OpenRISCCPU *cpu); void cpu_openrisc_timer_update(OpenRISCCPU *cpu); void cpu_openrisc_count_start(OpenRISCCPU *cpu); void cpu_openrisc_count_stop(OpenRISCCPU *cpu); - -void cpu_openrisc_mmu_init(OpenRISCCPU *cpu); -int cpu_openrisc_get_phys_nommu(OpenRISCCPU *cpu, - hwaddr *physical, - int *prot, target_ulong address, int rw); -int cpu_openrisc_get_phys_code(OpenRISCCPU *cpu, - hwaddr *physical, - int *prot, target_ulong address, int rw); -int cpu_openrisc_get_phys_data(OpenRISCCPU *cpu, - hwaddr *physical, - int *prot, target_ulong address, int rw); #endif #define OPENRISC_CPU_TYPE_SUFFIX "-" TYPE_OPENRISC_CPU @@ -395,9 +375,12 @@ int cpu_openrisc_get_phys_data(OpenRISCCPU *cpu, #include "exec/cpu-all.h" -#define TB_FLAGS_DFLAG 1 -#define TB_FLAGS_R0_0 2 +#define TB_FLAGS_SM SR_SM +#define TB_FLAGS_DME SR_DME +#define TB_FLAGS_IME SR_IME #define TB_FLAGS_OVE SR_OVE +#define TB_FLAGS_DFLAG 2 /* reuse SR_TEE */ +#define TB_FLAGS_R0_0 4 /* reuse SR_IEE */ static inline uint32_t cpu_get_gpr(const CPUOpenRISCState *env, int i) { @@ -415,17 +398,21 @@ static inline void cpu_get_tb_cpu_state(CPUOpenRISCState *env, { *pc = env->pc; *cs_base = 0; - *flags = (env->dflag - | (cpu_get_gpr(env, 0) == 0 ? TB_FLAGS_R0_0 : 0) - | (env->sr & SR_OVE)); + *flags = (env->dflag ? TB_FLAGS_DFLAG : 0) + | (cpu_get_gpr(env, 0) ? 0 : TB_FLAGS_R0_0) + | (env->sr & (SR_SM | SR_DME | SR_IME | SR_OVE)); } static inline int cpu_mmu_index(CPUOpenRISCState *env, bool ifetch) { - if (!(env->sr & SR_IME)) { - return MMU_NOMMU_IDX; + int ret = MMU_NOMMU_IDX; /* mmu is disabled */ + + if (env->sr & (ifetch ? SR_IME : SR_DME)) { + /* The mmu is enabled; test supervisor state. */ + ret = env->sr & SR_SM ? MMU_SUPERVISOR_IDX : MMU_USER_IDX; } - return (env->sr & SR_SM) == 0 ? MMU_USER_IDX : MMU_SUPERVISOR_IDX; + + return ret; } static inline uint32_t cpu_get_sr(const CPUOpenRISCState *env) diff --git a/target/openrisc/disas.c b/target/openrisc/disas.c new file mode 100644 index 0000000000..4bfd2dd8a6 --- /dev/null +++ b/target/openrisc/disas.c @@ -0,0 +1,170 @@ +/* + * OpenRISC disassembler + * + * Copyright (c) 2018 Richard Henderson <rth@twiddle.net> + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#include "qemu/osdep.h" +#include "qemu-common.h" +#include "disas/bfd.h" +#include "qemu/bitops.h" +#include "cpu.h" + +typedef disassemble_info DisasContext; + +/* Include the auto-generated decoder. */ +#include "decode.inc.c" + +#define output(mnemonic, format, ...) \ + (info->fprintf_func(info->stream, "%-9s " format, \ + mnemonic, ##__VA_ARGS__)) + +int print_insn_or1k(bfd_vma addr, disassemble_info *info) +{ + bfd_byte buffer[4]; + uint32_t insn; + int status; + + status = info->read_memory_func(addr, buffer, 4, info); + if (status != 0) { + info->memory_error_func(status, addr, info); + return -1; + } + insn = bfd_getb32(buffer); + + if (!decode(info, insn)) { + output(".long", "%#08x", insn); + } + return 4; +} + +#define INSN(opcode, format, ...) \ +static bool trans_l_##opcode(disassemble_info *info, \ + arg_l_##opcode *a, uint32_t insn) \ +{ \ + output("l." #opcode, format, ##__VA_ARGS__); \ + return true; \ +} + +INSN(add, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(addc, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(sub, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(and, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(or, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(xor, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(sll, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(srl, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(sra, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(ror, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(exths, "r%d, r%d", a->d, a->a) +INSN(extbs, "r%d, r%d", a->d, a->a) +INSN(exthz, "r%d, r%d", a->d, a->a) +INSN(extbz, "r%d, r%d", a->d, a->a) +INSN(cmov, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(ff1, "r%d, r%d", a->d, a->a) +INSN(fl1, "r%d, r%d", a->d, a->a) +INSN(mul, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(mulu, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(div, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(divu, "r%d, r%d, r%d", a->d, a->a, a->b) +INSN(muld, "r%d, r%d", a->a, a->b) +INSN(muldu, "r%d, r%d", a->a, a->b) +INSN(j, "%d", a->n) +INSN(jal, "%d", a->n) +INSN(bf, "%d", a->n) +INSN(bnf, "%d", a->n) +INSN(jr, "r%d", a->b) +INSN(jalr, "r%d", a->b) +INSN(lwa, "r%d, %d(r%d)", a->d, a->i, a->a) +INSN(lwz, "r%d, %d(r%d)", a->d, a->i, a->a) +INSN(lws, "r%d, %d(r%d)", a->d, a->i, a->a) +INSN(lbz, "r%d, %d(r%d)", a->d, a->i, a->a) +INSN(lbs, "r%d, %d(r%d)", a->d, a->i, a->a) +INSN(lhz, "r%d, %d(r%d)", a->d, a->i, a->a) +INSN(lhs, "r%d, %d(r%d)", a->d, a->i, a->a) +INSN(swa, "%d(r%d), r%d", a->i, a->a, a->b) +INSN(sw, "%d(r%d), r%d", a->i, a->a, a->b) +INSN(sb, "%d(r%d), r%d", a->i, a->a, a->b) +INSN(sh, "%d(r%d), r%d", a->i, a->a, a->b) +INSN(nop, "") +INSN(addi, "r%d, r%d, %d", a->d, a->a, a->i) +INSN(addic, "r%d, r%d, %d", a->d, a->a, a->i) +INSN(muli, "r%d, r%d, %d", a->d, a->a, a->i) +INSN(maci, "r%d, %d", a->a, a->i) +INSN(andi, "r%d, r%d, %d", a->d, a->a, a->k) +INSN(ori, "r%d, r%d, %d", a->d, a->a, a->k) +INSN(xori, "r%d, r%d, %d", a->d, a->a, a->i) +INSN(mfspr, "r%d, r%d, %d", a->d, a->a, a->k) +INSN(mtspr, "r%d, r%d, %d", a->a, a->b, a->k) +INSN(mac, "r%d, r%d", a->a, a->b) +INSN(msb, "r%d, r%d", a->a, a->b) +INSN(macu, "r%d, r%d", a->a, a->b) +INSN(msbu, "r%d, r%d", a->a, a->b) +INSN(slli, "r%d, r%d, %d", a->d, a->a, a->l) +INSN(srli, "r%d, r%d, %d", a->d, a->a, a->l) +INSN(srai, "r%d, r%d, %d", a->d, a->a, a->l) +INSN(rori, "r%d, r%d, %d", a->d, a->a, a->l) +INSN(movhi, "r%d, %d", a->d, a->k) +INSN(macrc, "r%d", a->d) +INSN(sfeq, "r%d, r%d", a->a, a->b) +INSN(sfne, "r%d, r%d", a->a, a->b) +INSN(sfgtu, "r%d, r%d", a->a, a->b) +INSN(sfgeu, "r%d, r%d", a->a, a->b) +INSN(sfltu, "r%d, r%d", a->a, a->b) +INSN(sfleu, "r%d, r%d", a->a, a->b) +INSN(sfgts, "r%d, r%d", a->a, a->b) +INSN(sfges, "r%d, r%d", a->a, a->b) +INSN(sflts, "r%d, r%d", a->a, a->b) +INSN(sfles, "r%d, r%d", a->a, a->b) +INSN(sfeqi, "r%d, %d", a->a, a->i) +INSN(sfnei, "r%d, %d", a->a, a->i) +INSN(sfgtui, "r%d, %d", a->a, a->i) +INSN(sfgeui, "r%d, %d", a->a, a->i) +INSN(sfltui, "r%d, %d", a->a, a->i) +INSN(sfleui, "r%d, %d", a->a, a->i) +INSN(sfgtsi, "r%d, %d", a->a, a->i) +INSN(sfgesi, "r%d, %d", a->a, a->i) +INSN(sfltsi, "r%d, %d", a->a, a->i) +INSN(sflesi, "r%d, %d", a->a, a->i) +INSN(sys, "%d", a->k) +INSN(trap, "%d", a->k) +INSN(msync, "") +INSN(psync, "") +INSN(csync, "") +INSN(rfe, "") + +#define FP_INSN(opcode, suffix, format, ...) \ +static bool trans_lf_##opcode##_##suffix(disassemble_info *info, \ + arg_lf_##opcode##_##suffix *a, uint32_t insn) \ +{ \ + output("lf." #opcode "." #suffix, format, ##__VA_ARGS__); \ + return true; \ +} + +FP_INSN(add, s, "r%d, r%d, r%d", a->d, a->a, a->b) +FP_INSN(sub, s, "r%d, r%d, r%d", a->d, a->a, a->b) +FP_INSN(mul, s, "r%d, r%d, r%d", a->d, a->a, a->b) +FP_INSN(div, s, "r%d, r%d, r%d", a->d, a->a, a->b) +FP_INSN(rem, s, "r%d, r%d, r%d", a->d, a->a, a->b) +FP_INSN(itof, s, "r%d, r%d", a->d, a->a) +FP_INSN(ftoi, s, "r%d, r%d", a->d, a->a) +FP_INSN(madd, s, "r%d, r%d, r%d", a->d, a->a, a->b) +FP_INSN(sfeq, s, "r%d, r%d", a->a, a->b) +FP_INSN(sfne, s, "r%d, r%d", a->a, a->b) +FP_INSN(sfgt, s, "r%d, r%d", a->a, a->b) +FP_INSN(sfge, s, "r%d, r%d", a->a, a->b) +FP_INSN(sflt, s, "r%d, r%d", a->a, a->b) +FP_INSN(sfle, s, "r%d, r%d", a->a, a->b) diff --git a/target/openrisc/helper.h b/target/openrisc/helper.h index e37dabc77a..9db9bf3963 100644 --- a/target/openrisc/helper.h +++ b/target/openrisc/helper.h @@ -56,5 +56,5 @@ FOP_CMP(le) DEF_HELPER_FLAGS_1(rfe, 0, void, env) /* sys */ -DEF_HELPER_FLAGS_4(mtspr, 0, void, env, tl, tl, tl) -DEF_HELPER_FLAGS_4(mfspr, TCG_CALL_NO_WG, tl, env, tl, tl, tl) +DEF_HELPER_FLAGS_3(mtspr, 0, void, env, tl, tl) +DEF_HELPER_FLAGS_3(mfspr, TCG_CALL_NO_WG, tl, env, tl, tl) diff --git a/target/openrisc/interrupt.c b/target/openrisc/interrupt.c index 3959671c59..bbae956361 100644 --- a/target/openrisc/interrupt.c +++ b/target/openrisc/interrupt.c @@ -32,29 +32,22 @@ void openrisc_cpu_do_interrupt(CPUState *cs) #ifndef CONFIG_USER_ONLY OpenRISCCPU *cpu = OPENRISC_CPU(cs); CPUOpenRISCState *env = &cpu->env; + int exception = cs->exception_index; env->epcr = env->pc; - if (env->dflag) { - env->dflag = 0; - env->sr |= SR_DSX; - env->epcr -= 4; - } else { - env->sr &= ~SR_DSX; - } - if (cs->exception_index == EXCP_SYSCALL) { + if (exception == EXCP_SYSCALL) { env->epcr += 4; } /* When we have an illegal instruction the error effective address shall be set to the illegal instruction address. */ - if (cs->exception_index == EXCP_ILLEGAL) { + if (exception == EXCP_ILLEGAL) { env->eear = env->pc; } - /* For machine-state changed between user-mode and supervisor mode, - we need flush TLB when we enter&exit EXCP. */ - tlb_flush(cs); - + /* During exceptions esr is populared with the pre-exception sr. */ env->esr = cpu_get_sr(env); + /* In parallel sr is updated to disable mmu, interrupts, timers and + set the delay slot exception flag. */ env->sr &= ~SR_DME; env->sr &= ~SR_IME; env->sr |= SR_SM; @@ -62,12 +55,38 @@ void openrisc_cpu_do_interrupt(CPUState *cs) env->sr &= ~SR_TEE; env->pmr &= ~PMR_DME; env->pmr &= ~PMR_SME; - env->tlb->cpu_openrisc_map_address_data = &cpu_openrisc_get_phys_nommu; - env->tlb->cpu_openrisc_map_address_code = &cpu_openrisc_get_phys_nommu; env->lock_addr = -1; - if (cs->exception_index > 0 && cs->exception_index < EXCP_NR) { - hwaddr vect_pc = cs->exception_index << 8; + /* Set/clear dsx to indicate if we are in a delay slot exception. */ + if (env->dflag) { + env->dflag = 0; + env->sr |= SR_DSX; + env->epcr -= 4; + } else { + env->sr &= ~SR_DSX; + } + + if (exception > 0 && exception < EXCP_NR) { + static const char * const int_name[EXCP_NR] = { + [EXCP_RESET] = "RESET", + [EXCP_BUSERR] = "BUSERR (bus error)", + [EXCP_DPF] = "DFP (data protection fault)", + [EXCP_IPF] = "IPF (code protection fault)", + [EXCP_TICK] = "TICK (timer interrupt)", + [EXCP_ALIGN] = "ALIGN", + [EXCP_ILLEGAL] = "ILLEGAL", + [EXCP_INT] = "INT (device interrupt)", + [EXCP_DTLBMISS] = "DTLBMISS (data tlb miss)", + [EXCP_ITLBMISS] = "ITLBMISS (code tlb miss)", + [EXCP_RANGE] = "RANGE", + [EXCP_SYSCALL] = "SYSCALL", + [EXCP_FPE] = "FPE", + [EXCP_TRAP] = "TRAP", + }; + + qemu_log_mask(CPU_LOG_INT, "INT: %s\n", int_name[exception]); + + hwaddr vect_pc = exception << 8; if (env->cpucfgr & CPUCFGR_EVBARP) { vect_pc |= env->evbar; } @@ -76,7 +95,7 @@ void openrisc_cpu_do_interrupt(CPUState *cs) } env->pc = vect_pc; } else { - cpu_abort(cs, "Unhandled exception 0x%x\n", cs->exception_index); + cpu_abort(cs, "Unhandled exception 0x%x\n", exception); } #endif diff --git a/target/openrisc/interrupt_helper.c b/target/openrisc/interrupt_helper.c index 56620e0571..9c5489f5f7 100644 --- a/target/openrisc/interrupt_helper.c +++ b/target/openrisc/interrupt_helper.c @@ -25,36 +25,7 @@ void HELPER(rfe)(CPUOpenRISCState *env) { - OpenRISCCPU *cpu = openrisc_env_get_cpu(env); - CPUState *cs = CPU(cpu); -#ifndef CONFIG_USER_ONLY - int need_flush_tlb = (cpu->env.sr & (SR_SM | SR_IME | SR_DME)) ^ - (cpu->env.esr & (SR_SM | SR_IME | SR_DME)); -#endif - cpu->env.pc = cpu->env.epcr; - cpu_set_sr(&cpu->env, cpu->env.esr); - cpu->env.lock_addr = -1; - -#ifndef CONFIG_USER_ONLY - if (cpu->env.sr & SR_DME) { - cpu->env.tlb->cpu_openrisc_map_address_data = - &cpu_openrisc_get_phys_data; - } else { - cpu->env.tlb->cpu_openrisc_map_address_data = - &cpu_openrisc_get_phys_nommu; - } - - if (cpu->env.sr & SR_IME) { - cpu->env.tlb->cpu_openrisc_map_address_code = - &cpu_openrisc_get_phys_code; - } else { - cpu->env.tlb->cpu_openrisc_map_address_code = - &cpu_openrisc_get_phys_nommu; - } - - if (need_flush_tlb) { - tlb_flush(cs); - } -#endif - cs->interrupt_request |= CPU_INTERRUPT_EXITTB; + env->pc = env->epcr; + env->lock_addr = -1; + cpu_set_sr(env, env->esr); } diff --git a/target/openrisc/machine.c b/target/openrisc/machine.c index 0a793eb14d..1eedbf3dbe 100644 --- a/target/openrisc/machine.c +++ b/target/openrisc/machine.c @@ -24,31 +24,6 @@ #include "hw/boards.h" #include "migration/cpu.h" -static int env_post_load(void *opaque, int version_id) -{ - CPUOpenRISCState *env = opaque; - - /* Restore MMU handlers */ - if (env->sr & SR_DME) { - env->tlb->cpu_openrisc_map_address_data = - &cpu_openrisc_get_phys_data; - } else { - env->tlb->cpu_openrisc_map_address_data = - &cpu_openrisc_get_phys_nommu; - } - - if (env->sr & SR_IME) { - env->tlb->cpu_openrisc_map_address_code = - &cpu_openrisc_get_phys_code; - } else { - env->tlb->cpu_openrisc_map_address_code = - &cpu_openrisc_get_phys_nommu; - } - - - return 0; -} - static const VMStateDescription vmstate_tlb_entry = { .name = "tlb_entry", .version_id = 1, @@ -63,24 +38,17 @@ static const VMStateDescription vmstate_tlb_entry = { static const VMStateDescription vmstate_cpu_tlb = { .name = "cpu_tlb", - .version_id = 1, - .minimum_version_id = 1, - .minimum_version_id_old = 1, + .version_id = 2, + .minimum_version_id = 2, .fields = (VMStateField[]) { - VMSTATE_STRUCT_2DARRAY(itlb, CPUOpenRISCTLBContext, - ITLB_WAYS, ITLB_SIZE, 0, + VMSTATE_STRUCT_ARRAY(itlb, CPUOpenRISCTLBContext, TLB_SIZE, 0, vmstate_tlb_entry, OpenRISCTLBEntry), - VMSTATE_STRUCT_2DARRAY(dtlb, CPUOpenRISCTLBContext, - DTLB_WAYS, DTLB_SIZE, 0, + VMSTATE_STRUCT_ARRAY(dtlb, CPUOpenRISCTLBContext, TLB_SIZE, 0, vmstate_tlb_entry, OpenRISCTLBEntry), VMSTATE_END_OF_LIST() } }; -#define VMSTATE_CPU_TLB(_f, _s) \ - VMSTATE_STRUCT_POINTER(_f, _s, vmstate_cpu_tlb, CPUOpenRISCTLBContext) - - static int get_sr(QEMUFile *f, void *opaque, size_t size, VMStateField *field) { CPUOpenRISCState *env = opaque; @@ -106,7 +74,6 @@ static const VMStateDescription vmstate_env = { .name = "env", .version_id = 6, .minimum_version_id = 6, - .post_load = env_post_load, .fields = (VMStateField[]) { VMSTATE_UINTTL_2DARRAY(shadow_gpr, CPUOpenRISCState, 16, 32), VMSTATE_UINTTL(pc, CPUOpenRISCState), @@ -143,7 +110,8 @@ static const VMStateDescription vmstate_env = { VMSTATE_UINT32(fpcsr, CPUOpenRISCState), VMSTATE_UINT64(mac, CPUOpenRISCState), - VMSTATE_CPU_TLB(tlb, CPUOpenRISCState), + VMSTATE_STRUCT(tlb, CPUOpenRISCState, 1, + vmstate_cpu_tlb, CPUOpenRISCTLBContext), VMSTATE_TIMER_PTR(timer, CPUOpenRISCState), VMSTATE_UINT32(ttmr, CPUOpenRISCState), diff --git a/target/openrisc/mmu.c b/target/openrisc/mmu.c index 2bd782f89b..e7d5219e11 100644 --- a/target/openrisc/mmu.c +++ b/target/openrisc/mmu.c @@ -29,227 +29,156 @@ #endif #ifndef CONFIG_USER_ONLY -int cpu_openrisc_get_phys_nommu(OpenRISCCPU *cpu, - hwaddr *physical, - int *prot, target_ulong address, int rw) +static inline void get_phys_nommu(hwaddr *phys_addr, int *prot, + target_ulong address) { - *physical = address; + *phys_addr = address; *prot = PAGE_READ | PAGE_WRITE | PAGE_EXEC; - return TLBRET_MATCH; } -int cpu_openrisc_get_phys_code(OpenRISCCPU *cpu, - hwaddr *physical, - int *prot, target_ulong address, int rw) +static int get_phys_mmu(OpenRISCCPU *cpu, hwaddr *phys_addr, int *prot, + target_ulong addr, int need, bool super) { - int vpn = address >> TARGET_PAGE_BITS; - int idx = vpn & ITLB_MASK; - int right = 0; - - if ((cpu->env.tlb->itlb[0][idx].mr >> TARGET_PAGE_BITS) != vpn) { - return TLBRET_NOMATCH; - } - if (!(cpu->env.tlb->itlb[0][idx].mr & 1)) { - return TLBRET_INVALID; - } - - if (cpu->env.sr & SR_SM) { /* supervisor mode */ - if (cpu->env.tlb->itlb[0][idx].tr & SXE) { - right |= PAGE_EXEC; - } - } else { - if (cpu->env.tlb->itlb[0][idx].tr & UXE) { - right |= PAGE_EXEC; + int idx = (addr >> TARGET_PAGE_BITS) & TLB_MASK; + uint32_t imr = cpu->env.tlb.itlb[idx].mr; + uint32_t itr = cpu->env.tlb.itlb[idx].tr; + uint32_t dmr = cpu->env.tlb.dtlb[idx].mr; + uint32_t dtr = cpu->env.tlb.dtlb[idx].tr; + int right, match, valid; + + /* If the ITLB and DTLB indexes map to the same page, we want to + load all permissions all at once. If the destination pages do + not match, zap the one we don't need. */ + if (unlikely((itr ^ dtr) & TARGET_PAGE_MASK)) { + if (need & PAGE_EXEC) { + dmr = dtr = 0; + } else { + imr = itr = 0; } } - if ((rw & 2) && ((right & PAGE_EXEC) == 0)) { - return TLBRET_BADADDR; - } - - *physical = (cpu->env.tlb->itlb[0][idx].tr & TARGET_PAGE_MASK) | - (address & (TARGET_PAGE_SIZE-1)); + /* Check if either of the entries matches the source address. */ + match = (imr ^ addr) & TARGET_PAGE_MASK ? 0 : PAGE_EXEC; + match |= (dmr ^ addr) & TARGET_PAGE_MASK ? 0 : PAGE_READ | PAGE_WRITE; + + /* Check if either of the entries is valid. */ + valid = imr & 1 ? PAGE_EXEC : 0; + valid |= dmr & 1 ? PAGE_READ | PAGE_WRITE : 0; + valid &= match; + + /* Collect the permissions from the entries. */ + right = itr & (super ? SXE : UXE) ? PAGE_EXEC : 0; + right |= dtr & (super ? SRE : URE) ? PAGE_READ : 0; + right |= dtr & (super ? SWE : UWE) ? PAGE_WRITE : 0; + right &= valid; + + /* Note that above we validated that itr and dtr match on page. + So oring them together changes nothing without having to + check which one we needed. We also want to store to these + variables even on failure, as it avoids compiler warnings. */ + *phys_addr = ((itr | dtr) & TARGET_PAGE_MASK) | (addr & ~TARGET_PAGE_MASK); *prot = right; - return TLBRET_MATCH; -} -int cpu_openrisc_get_phys_data(OpenRISCCPU *cpu, - hwaddr *physical, - int *prot, target_ulong address, int rw) -{ - int vpn = address >> TARGET_PAGE_BITS; - int idx = vpn & DTLB_MASK; - int right = 0; + qemu_log_mask(CPU_LOG_MMU, + "MMU lookup: need %d match %d valid %d right %d -> %s\n", + need, match, valid, right, (need & right) ? "OK" : "FAIL"); - if ((cpu->env.tlb->dtlb[0][idx].mr >> TARGET_PAGE_BITS) != vpn) { - return TLBRET_NOMATCH; - } - if (!(cpu->env.tlb->dtlb[0][idx].mr & 1)) { - return TLBRET_INVALID; + /* Check the collective permissions are present. */ + if (likely(need & right)) { + return 0; /* success! */ } - if (cpu->env.sr & SR_SM) { /* supervisor mode */ - if (cpu->env.tlb->dtlb[0][idx].tr & SRE) { - right |= PAGE_READ; - } - if (cpu->env.tlb->dtlb[0][idx].tr & SWE) { - right |= PAGE_WRITE; - } + /* Determine what kind of failure we have. */ + if (need & valid) { + return need & PAGE_EXEC ? EXCP_IPF : EXCP_DPF; } else { - if (cpu->env.tlb->dtlb[0][idx].tr & URE) { - right |= PAGE_READ; - } - if (cpu->env.tlb->dtlb[0][idx].tr & UWE) { - right |= PAGE_WRITE; - } + return need & PAGE_EXEC ? EXCP_ITLBMISS : EXCP_DTLBMISS; } - - if (!(rw & 1) && ((right & PAGE_READ) == 0)) { - return TLBRET_BADADDR; - } - if ((rw & 1) && ((right & PAGE_WRITE) == 0)) { - return TLBRET_BADADDR; - } - - *physical = (cpu->env.tlb->dtlb[0][idx].tr & TARGET_PAGE_MASK) | - (address & (TARGET_PAGE_SIZE-1)); - *prot = right; - return TLBRET_MATCH; -} - -static int cpu_openrisc_get_phys_addr(OpenRISCCPU *cpu, - hwaddr *physical, - int *prot, target_ulong address, - int rw) -{ - int ret = TLBRET_MATCH; - - if (rw == MMU_INST_FETCH) { /* ITLB */ - *physical = 0; - ret = cpu->env.tlb->cpu_openrisc_map_address_code(cpu, physical, - prot, address, rw); - } else { /* DTLB */ - ret = cpu->env.tlb->cpu_openrisc_map_address_data(cpu, physical, - prot, address, rw); - } - - return ret; } #endif -static void cpu_openrisc_raise_mmu_exception(OpenRISCCPU *cpu, - target_ulong address, - int rw, int tlb_error) +static void raise_mmu_exception(OpenRISCCPU *cpu, target_ulong address, + int exception) { CPUState *cs = CPU(cpu); - int exception = 0; - - switch (tlb_error) { - default: - if (rw == 2) { - exception = EXCP_IPF; - } else { - exception = EXCP_DPF; - } - break; -#ifndef CONFIG_USER_ONLY - case TLBRET_BADADDR: - if (rw == 2) { - exception = EXCP_IPF; - } else { - exception = EXCP_DPF; - } - break; - case TLBRET_INVALID: - case TLBRET_NOMATCH: - /* No TLB match for a mapped address */ - if (rw == 2) { - exception = EXCP_ITLBMISS; - } else { - exception = EXCP_DTLBMISS; - } - break; -#endif - } cs->exception_index = exception; cpu->env.eear = address; cpu->env.lock_addr = -1; } -#ifndef CONFIG_USER_ONLY int openrisc_cpu_handle_mmu_fault(CPUState *cs, vaddr address, int size, int rw, int mmu_idx) { +#ifdef CONFIG_USER_ONLY OpenRISCCPU *cpu = OPENRISC_CPU(cs); - int ret = 0; - hwaddr physical = 0; - int prot = 0; - - ret = cpu_openrisc_get_phys_addr(cpu, &physical, &prot, - address, rw); - - if (ret == TLBRET_MATCH) { - tlb_set_page(cs, address & TARGET_PAGE_MASK, - physical & TARGET_PAGE_MASK, prot, - mmu_idx, TARGET_PAGE_SIZE); - ret = 0; - } else if (ret < 0) { - cpu_openrisc_raise_mmu_exception(cpu, address, rw, ret); - ret = 1; - } - - return ret; -} + raise_mmu_exception(cpu, address, EXCP_DPF); + return 1; #else -int openrisc_cpu_handle_mmu_fault(CPUState *cs, vaddr address, int size, - int rw, int mmu_idx) -{ - OpenRISCCPU *cpu = OPENRISC_CPU(cs); - int ret = 0; - - cpu_openrisc_raise_mmu_exception(cpu, address, rw, ret); - ret = 1; - - return ret; -} + g_assert_not_reached(); #endif +} #ifndef CONFIG_USER_ONLY hwaddr openrisc_cpu_get_phys_page_debug(CPUState *cs, vaddr addr) { OpenRISCCPU *cpu = OPENRISC_CPU(cs); + int prot, excp, sr = cpu->env.sr; hwaddr phys_addr; - int prot; - int miss; - /* Check memory for any kind of address, since during debug the - gdb can ask for anything, check data tlb for address */ - miss = cpu_openrisc_get_phys_addr(cpu, &phys_addr, &prot, addr, 0); - - /* Check instruction tlb */ - if (miss) { - miss = cpu_openrisc_get_phys_addr(cpu, &phys_addr, &prot, addr, - MMU_INST_FETCH); - } + switch (sr & (SR_DME | SR_IME)) { + case SR_DME | SR_IME: + /* The mmu is definitely enabled. */ + excp = get_phys_mmu(cpu, &phys_addr, &prot, addr, + PAGE_EXEC | PAGE_READ | PAGE_WRITE, + (sr & SR_SM) != 0); + return excp ? -1 : phys_addr; - /* Last, fall back to a plain address */ - if (miss) { - miss = cpu_openrisc_get_phys_nommu(cpu, &phys_addr, &prot, addr, 0); - } + default: + /* The mmu is partially enabled, and we don't really have + a "real" access type. Begin by trying the mmu, but if + that fails try again without. */ + excp = get_phys_mmu(cpu, &phys_addr, &prot, addr, + PAGE_EXEC | PAGE_READ | PAGE_WRITE, + (sr & SR_SM) != 0); + if (!excp) { + return phys_addr; + } + /* fallthru */ - if (miss) { - return -1; - } else { + case 0: + /* The mmu is definitely disabled; lookups never fail. */ + get_phys_nommu(&phys_addr, &prot, addr); return phys_addr; } } -void cpu_openrisc_mmu_init(OpenRISCCPU *cpu) +void tlb_fill(CPUState *cs, target_ulong addr, int size, + MMUAccessType access_type, int mmu_idx, uintptr_t retaddr) { - cpu->env.tlb = g_malloc0(sizeof(CPUOpenRISCTLBContext)); + OpenRISCCPU *cpu = OPENRISC_CPU(cs); + int prot, excp; + hwaddr phys_addr; + + if (mmu_idx == MMU_NOMMU_IDX) { + /* The mmu is disabled; lookups never fail. */ + get_phys_nommu(&phys_addr, &prot, addr); + excp = 0; + } else { + bool super = mmu_idx == MMU_SUPERVISOR_IDX; + int need = (access_type == MMU_INST_FETCH ? PAGE_EXEC + : access_type == MMU_DATA_STORE ? PAGE_WRITE + : PAGE_READ); + excp = get_phys_mmu(cpu, &phys_addr, &prot, addr, need, super); + } + + if (unlikely(excp)) { + raise_mmu_exception(cpu, addr, excp); + cpu_loop_exit_restore(cs, retaddr); + } - cpu->env.tlb->cpu_openrisc_map_address_code = &cpu_openrisc_get_phys_nommu; - cpu->env.tlb->cpu_openrisc_map_address_data = &cpu_openrisc_get_phys_nommu; + tlb_set_page(cs, addr & TARGET_PAGE_MASK, + phys_addr & TARGET_PAGE_MASK, prot, + mmu_idx, TARGET_PAGE_SIZE); } #endif diff --git a/target/openrisc/mmu_helper.c b/target/openrisc/mmu_helper.c deleted file mode 100644 index 97e1d17b5a..0000000000 --- a/target/openrisc/mmu_helper.c +++ /dev/null @@ -1,40 +0,0 @@ -/* - * OpenRISC MMU helper routines - * - * Copyright (c) 2011-2012 Jia Liu <proljc@gmail.com> - * Zhizhou Zhang <etouzh@gmail.com> - * - * This library is free software; you can redistribute it and/or - * modify it under the terms of the GNU Lesser General Public - * License as published by the Free Software Foundation; either - * version 2 of the License, or (at your option) any later version. - * - * This library is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * Lesser General Public License for more details. - * - * You should have received a copy of the GNU Lesser General Public - * License along with this library; if not, see <http://www.gnu.org/licenses/>. - */ - -#include "qemu/osdep.h" -#include "cpu.h" -#include "exec/exec-all.h" -#include "exec/cpu_ldst.h" - -#ifndef CONFIG_USER_ONLY - -void tlb_fill(CPUState *cs, target_ulong addr, int size, - MMUAccessType access_type, int mmu_idx, uintptr_t retaddr) -{ - int ret; - - ret = openrisc_cpu_handle_mmu_fault(cs, addr, size, access_type, mmu_idx); - - if (ret) { - /* Raise Exception. */ - cpu_loop_exit_restore(cs, retaddr); - } -} -#endif diff --git a/target/openrisc/sys_helper.c b/target/openrisc/sys_helper.c index b284064381..b66a45c1e0 100644 --- a/target/openrisc/sys_helper.c +++ b/target/openrisc/sys_helper.c @@ -27,13 +27,12 @@ #define TO_SPR(group, number) (((group) << 11) + (number)) -void HELPER(mtspr)(CPUOpenRISCState *env, - target_ulong ra, target_ulong rb, target_ulong offset) +void HELPER(mtspr)(CPUOpenRISCState *env, target_ulong spr, target_ulong rb) { #ifndef CONFIG_USER_ONLY OpenRISCCPU *cpu = openrisc_env_get_cpu(env); CPUState *cs = CPU(cpu); - int spr = (ra | offset); + target_ulong mr; int idx; switch (spr) { @@ -57,26 +56,7 @@ void HELPER(mtspr)(CPUOpenRISCState *env, break; case TO_SPR(0, 17): /* SR */ - if ((env->sr & (SR_IME | SR_DME | SR_SM)) ^ - (rb & (SR_IME | SR_DME | SR_SM))) { - tlb_flush(cs); - } cpu_set_sr(env, rb); - if (env->sr & SR_DME) { - env->tlb->cpu_openrisc_map_address_data = - &cpu_openrisc_get_phys_data; - } else { - env->tlb->cpu_openrisc_map_address_data = - &cpu_openrisc_get_phys_nommu; - } - - if (env->sr & SR_IME) { - env->tlb->cpu_openrisc_map_address_code = - &cpu_openrisc_get_phys_code; - } else { - env->tlb->cpu_openrisc_map_address_code = - &cpu_openrisc_get_phys_nommu; - } break; case TO_SPR(0, 18): /* PPC */ @@ -98,18 +78,22 @@ void HELPER(mtspr)(CPUOpenRISCState *env, case TO_SPR(0, 1024) ... TO_SPR(0, 1024 + (16 * 32)): /* Shadow GPRs */ idx = (spr - 1024); env->shadow_gpr[idx / 32][idx % 32] = rb; + break; - case TO_SPR(1, 512) ... TO_SPR(1, 512+DTLB_SIZE-1): /* DTLBW0MR 0-127 */ + case TO_SPR(1, 512) ... TO_SPR(1, 512 + TLB_SIZE - 1): /* DTLBW0MR 0-127 */ idx = spr - TO_SPR(1, 512); - if (!(rb & 1)) { - tlb_flush_page(cs, env->tlb->dtlb[0][idx].mr & TARGET_PAGE_MASK); + mr = env->tlb.dtlb[idx].mr; + if (mr & 1) { + tlb_flush_page(cs, mr & TARGET_PAGE_MASK); + } + if (rb & 1) { + tlb_flush_page(cs, rb & TARGET_PAGE_MASK); } - env->tlb->dtlb[0][idx].mr = rb; + env->tlb.dtlb[idx].mr = rb; break; - - case TO_SPR(1, 640) ... TO_SPR(1, 640+DTLB_SIZE-1): /* DTLBW0TR 0-127 */ + case TO_SPR(1, 640) ... TO_SPR(1, 640 + TLB_SIZE - 1): /* DTLBW0TR 0-127 */ idx = spr - TO_SPR(1, 640); - env->tlb->dtlb[0][idx].tr = rb; + env->tlb.dtlb[idx].tr = rb; break; case TO_SPR(1, 768) ... TO_SPR(1, 895): /* DTLBW1MR 0-127 */ case TO_SPR(1, 896) ... TO_SPR(1, 1023): /* DTLBW1TR 0-127 */ @@ -118,17 +102,21 @@ void HELPER(mtspr)(CPUOpenRISCState *env, case TO_SPR(1, 1280) ... TO_SPR(1, 1407): /* DTLBW3MR 0-127 */ case TO_SPR(1, 1408) ... TO_SPR(1, 1535): /* DTLBW3TR 0-127 */ break; - case TO_SPR(2, 512) ... TO_SPR(2, 512+ITLB_SIZE-1): /* ITLBW0MR 0-127 */ + + case TO_SPR(2, 512) ... TO_SPR(2, 512 + TLB_SIZE - 1): /* ITLBW0MR 0-127 */ idx = spr - TO_SPR(2, 512); - if (!(rb & 1)) { - tlb_flush_page(cs, env->tlb->itlb[0][idx].mr & TARGET_PAGE_MASK); + mr = env->tlb.itlb[idx].mr; + if (mr & 1) { + tlb_flush_page(cs, mr & TARGET_PAGE_MASK); + } + if (rb & 1) { + tlb_flush_page(cs, rb & TARGET_PAGE_MASK); } - env->tlb->itlb[0][idx].mr = rb; + env->tlb.itlb[idx].mr = rb; break; - - case TO_SPR(2, 640) ... TO_SPR(2, 640+ITLB_SIZE-1): /* ITLBW0TR 0-127 */ + case TO_SPR(2, 640) ... TO_SPR(2, 640 + TLB_SIZE - 1): /* ITLBW0TR 0-127 */ idx = spr - TO_SPR(2, 640); - env->tlb->itlb[0][idx].tr = rb; + env->tlb.itlb[idx].tr = rb; break; case TO_SPR(2, 768) ... TO_SPR(2, 895): /* ITLBW1MR 0-127 */ case TO_SPR(2, 896) ... TO_SPR(2, 1023): /* ITLBW1TR 0-127 */ @@ -137,6 +125,7 @@ void HELPER(mtspr)(CPUOpenRISCState *env, case TO_SPR(2, 1280) ... TO_SPR(2, 1407): /* ITLBW3MR 0-127 */ case TO_SPR(2, 1408) ... TO_SPR(2, 1535): /* ITLBW3TR 0-127 */ break; + case TO_SPR(5, 1): /* MACLO */ env->mac = deposit64(env->mac, 0, 32, rb); break; @@ -153,7 +142,7 @@ void HELPER(mtspr)(CPUOpenRISCState *env, } break; case TO_SPR(9, 0): /* PICMR */ - env->picmr |= rb; + env->picmr = rb; break; case TO_SPR(9, 2): /* PICSR */ env->picsr &= ~rb; @@ -201,13 +190,12 @@ void HELPER(mtspr)(CPUOpenRISCState *env, #endif } -target_ulong HELPER(mfspr)(CPUOpenRISCState *env, - target_ulong rd, target_ulong ra, uint32_t offset) +target_ulong HELPER(mfspr)(CPUOpenRISCState *env, target_ulong rd, + target_ulong spr) { #ifndef CONFIG_USER_ONLY OpenRISCCPU *cpu = openrisc_env_get_cpu(env); CPUState *cs = CPU(cpu); - int spr = (ra | offset); int idx; switch (spr) { @@ -259,13 +247,13 @@ target_ulong HELPER(mfspr)(CPUOpenRISCState *env, idx = (spr - 1024); return env->shadow_gpr[idx / 32][idx % 32]; - case TO_SPR(1, 512) ... TO_SPR(1, 512+DTLB_SIZE-1): /* DTLBW0MR 0-127 */ + case TO_SPR(1, 512) ... TO_SPR(1, 512 + TLB_SIZE - 1): /* DTLBW0MR 0-127 */ idx = spr - TO_SPR(1, 512); - return env->tlb->dtlb[0][idx].mr; + return env->tlb.dtlb[idx].mr; - case TO_SPR(1, 640) ... TO_SPR(1, 640+DTLB_SIZE-1): /* DTLBW0TR 0-127 */ + case TO_SPR(1, 640) ... TO_SPR(1, 640 + TLB_SIZE - 1): /* DTLBW0TR 0-127 */ idx = spr - TO_SPR(1, 640); - return env->tlb->dtlb[0][idx].tr; + return env->tlb.dtlb[idx].tr; case TO_SPR(1, 768) ... TO_SPR(1, 895): /* DTLBW1MR 0-127 */ case TO_SPR(1, 896) ... TO_SPR(1, 1023): /* DTLBW1TR 0-127 */ @@ -275,13 +263,13 @@ target_ulong HELPER(mfspr)(CPUOpenRISCState *env, case TO_SPR(1, 1408) ... TO_SPR(1, 1535): /* DTLBW3TR 0-127 */ break; - case TO_SPR(2, 512) ... TO_SPR(2, 512+ITLB_SIZE-1): /* ITLBW0MR 0-127 */ + case TO_SPR(2, 512) ... TO_SPR(2, 512 + TLB_SIZE - 1): /* ITLBW0MR 0-127 */ idx = spr - TO_SPR(2, 512); - return env->tlb->itlb[0][idx].mr; + return env->tlb.itlb[idx].mr; - case TO_SPR(2, 640) ... TO_SPR(2, 640+ITLB_SIZE-1): /* ITLBW0TR 0-127 */ + case TO_SPR(2, 640) ... TO_SPR(2, 640 + TLB_SIZE - 1): /* ITLBW0TR 0-127 */ idx = spr - TO_SPR(2, 640); - return env->tlb->itlb[0][idx].tr; + return env->tlb.itlb[idx].tr; case TO_SPR(2, 768) ... TO_SPR(2, 895): /* ITLBW1MR 0-127 */ case TO_SPR(2, 896) ... TO_SPR(2, 1023): /* ITLBW1TR 0-127 */ diff --git a/target/openrisc/translate.c b/target/openrisc/translate.c index d69f8d0422..a271cd3903 100644 --- a/target/openrisc/translate.c +++ b/target/openrisc/translate.c @@ -36,22 +36,29 @@ #include "trace-tcg.h" #include "exec/log.h" -#define LOG_DIS(str, ...) \ - qemu_log_mask(CPU_LOG_TB_IN_ASM, "%08x: " str, dc->base.pc_next, \ - ## __VA_ARGS__) - /* is_jmp field values */ -#define DISAS_JUMP DISAS_TARGET_0 /* only pc was modified dynamically */ -#define DISAS_UPDATE DISAS_TARGET_1 /* cpu state was modified dynamically */ -#define DISAS_TB_JUMP DISAS_TARGET_2 /* only pc was modified statically */ +#define DISAS_EXIT DISAS_TARGET_0 /* force exit to main loop */ +#define DISAS_JUMP DISAS_TARGET_1 /* exit via jmp_pc/jmp_pc_imm */ typedef struct DisasContext { DisasContextBase base; uint32_t mem_idx; uint32_t tb_flags; uint32_t delayed_branch; + + /* If not -1, jmp_pc contains this value and so is a direct jump. */ + target_ulong jmp_pc_imm; } DisasContext; +static inline bool is_user(DisasContext *dc) +{ +#ifdef CONFIG_USER_ONLY + return true; +#else + return !(dc->tb_flags & TB_FLAGS_SM); +#endif +} + /* Include the auto-generated decoder. */ #include "decode.inc.c" @@ -165,34 +172,6 @@ static void check_ov64s(DisasContext *dc) } \ } while (0) -static inline bool use_goto_tb(DisasContext *dc, target_ulong dest) -{ - if (unlikely(dc->base.singlestep_enabled)) { - return false; - } - -#ifndef CONFIG_USER_ONLY - return (dc->base.tb->pc & TARGET_PAGE_MASK) == (dest & TARGET_PAGE_MASK); -#else - return true; -#endif -} - -static void gen_goto_tb(DisasContext *dc, int n, target_ulong dest) -{ - if (use_goto_tb(dc, dest)) { - tcg_gen_movi_tl(cpu_pc, dest); - tcg_gen_goto_tb(n); - tcg_gen_exit_tb(dc->base.tb, n); - } else { - tcg_gen_movi_tl(cpu_pc, dest); - if (dc->base.singlestep_enabled) { - gen_exception(dc, EXCP_DEBUG); - } - tcg_gen_exit_tb(NULL, 0); - } -} - static void gen_ove_cy(DisasContext *dc) { if (dc->tb_flags & SR_OVE) { @@ -457,7 +436,6 @@ static void gen_msbu(DisasContext *dc, TCGv srca, TCGv srcb) static bool trans_l_add(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("l.add r%d, r%d, r%d\n", a->d, a->a, a->b); check_r0_write(a->d); gen_add(dc, cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); return true; @@ -465,7 +443,6 @@ static bool trans_l_add(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_addc(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("l.addc r%d, r%d, r%d\n", a->d, a->a, a->b); check_r0_write(a->d); gen_addc(dc, cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); return true; @@ -473,7 +450,6 @@ static bool trans_l_addc(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_sub(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("l.sub r%d, r%d, r%d\n", a->d, a->a, a->b); check_r0_write(a->d); gen_sub(dc, cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); return true; @@ -481,7 +457,6 @@ static bool trans_l_sub(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_and(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("l.and r%d, r%d, r%d\n", a->d, a->a, a->b); check_r0_write(a->d); tcg_gen_and_tl(cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); return true; @@ -489,7 +464,6 @@ static bool trans_l_and(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_or(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("l.or r%d, r%d, r%d\n", a->d, a->a, a->b); check_r0_write(a->d); tcg_gen_or_tl(cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); return true; @@ -497,7 +471,6 @@ static bool trans_l_or(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_xor(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("l.xor r%d, r%d, r%d\n", a->d, a->a, a->b); check_r0_write(a->d); tcg_gen_xor_tl(cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); return true; @@ -505,7 +478,6 @@ static bool trans_l_xor(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_sll(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("l.sll r%d, r%d, r%d\n", a->d, a->a, a->b); check_r0_write(a->d); tcg_gen_shl_tl(cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); return true; @@ -513,7 +485,6 @@ static bool trans_l_sll(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_srl(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("l.srl r%d, r%d, r%d\n", a->d, a->a, a->b); check_r0_write(a->d); tcg_gen_shr_tl(cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); return true; @@ -521,7 +492,6 @@ static bool trans_l_srl(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_sra(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("l.sra r%d, r%d, r%d\n", a->d, a->a, a->b); check_r0_write(a->d); tcg_gen_sar_tl(cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); return true; @@ -529,7 +499,6 @@ static bool trans_l_sra(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_ror(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("l.ror r%d, r%d, r%d\n", a->d, a->a, a->b); check_r0_write(a->d); tcg_gen_rotr_tl(cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); return true; @@ -537,7 +506,6 @@ static bool trans_l_ror(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_exths(DisasContext *dc, arg_da *a, uint32_t insn) { - LOG_DIS("l.exths r%d, r%d\n", a->d, a->a); check_r0_write(a->d); tcg_gen_ext16s_tl(cpu_R[a->d], cpu_R[a->a]); return true; @@ -545,7 +513,6 @@ static bool trans_l_exths(DisasContext *dc, arg_da *a, uint32_t insn) static bool trans_l_extbs(DisasContext *dc, arg_da *a, uint32_t insn) { - LOG_DIS("l.extbs r%d, r%d\n", a->d, a->a); check_r0_write(a->d); tcg_gen_ext8s_tl(cpu_R[a->d], cpu_R[a->a]); return true; @@ -553,7 +520,6 @@ static bool trans_l_extbs(DisasContext *dc, arg_da *a, uint32_t insn) static bool trans_l_exthz(DisasContext *dc, arg_da *a, uint32_t insn) { - LOG_DIS("l.exthz r%d, r%d\n", a->d, a->a); check_r0_write(a->d); tcg_gen_ext16u_tl(cpu_R[a->d], cpu_R[a->a]); return true; @@ -561,7 +527,6 @@ static bool trans_l_exthz(DisasContext *dc, arg_da *a, uint32_t insn) static bool trans_l_extbz(DisasContext *dc, arg_da *a, uint32_t insn) { - LOG_DIS("l.extbz r%d, r%d\n", a->d, a->a); check_r0_write(a->d); tcg_gen_ext8u_tl(cpu_R[a->d], cpu_R[a->a]); return true; @@ -570,7 +535,6 @@ static bool trans_l_extbz(DisasContext *dc, arg_da *a, uint32_t insn) static bool trans_l_cmov(DisasContext *dc, arg_dab *a, uint32_t insn) { TCGv zero; - LOG_DIS("l.cmov r%d, r%d, r%d\n", a->d, a->a, a->b); check_r0_write(a->d); zero = tcg_const_tl(0); @@ -582,8 +546,6 @@ static bool trans_l_cmov(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_ff1(DisasContext *dc, arg_da *a, uint32_t insn) { - LOG_DIS("l.ff1 r%d, r%d\n", a->d, a->a); - check_r0_write(a->d); tcg_gen_ctzi_tl(cpu_R[a->d], cpu_R[a->a], -1); tcg_gen_addi_tl(cpu_R[a->d], cpu_R[a->d], 1); @@ -592,8 +554,6 @@ static bool trans_l_ff1(DisasContext *dc, arg_da *a, uint32_t insn) static bool trans_l_fl1(DisasContext *dc, arg_da *a, uint32_t insn) { - LOG_DIS("l.fl1 r%d, r%d\n", a->d, a->a); - check_r0_write(a->d); tcg_gen_clzi_tl(cpu_R[a->d], cpu_R[a->a], TARGET_LONG_BITS); tcg_gen_subfi_tl(cpu_R[a->d], TARGET_LONG_BITS, cpu_R[a->d]); @@ -602,8 +562,6 @@ static bool trans_l_fl1(DisasContext *dc, arg_da *a, uint32_t insn) static bool trans_l_mul(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("l.mul r%d, r%d, r%d\n", a->d, a->a, a->b); - check_r0_write(a->d); gen_mul(dc, cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); return true; @@ -611,8 +569,6 @@ static bool trans_l_mul(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_mulu(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("l.mulu r%d, r%d, r%d\n", a->d, a->a, a->b); - check_r0_write(a->d); gen_mulu(dc, cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); return true; @@ -620,8 +576,6 @@ static bool trans_l_mulu(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_div(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("l.div r%d, r%d, r%d\n", a->d, a->a, a->b); - check_r0_write(a->d); gen_div(dc, cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); return true; @@ -629,8 +583,6 @@ static bool trans_l_div(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_divu(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("l.divu r%d, r%d, r%d\n", a->d, a->a, a->b); - check_r0_write(a->d); gen_divu(dc, cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); return true; @@ -638,14 +590,12 @@ static bool trans_l_divu(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_l_muld(DisasContext *dc, arg_ab *a, uint32_t insn) { - LOG_DIS("l.muld r%d, r%d\n", a->a, a->b); gen_muld(dc, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_muldu(DisasContext *dc, arg_ab *a, uint32_t insn) { - LOG_DIS("l.muldu r%d, r%d\n", a->a, a->b); gen_muldu(dc, cpu_R[a->a], cpu_R[a->b]); return true; } @@ -654,8 +604,8 @@ static bool trans_l_j(DisasContext *dc, arg_l_j *a, uint32_t insn) { target_ulong tmp_pc = dc->base.pc_next + a->n * 4; - LOG_DIS("l.j %d\n", a->n); tcg_gen_movi_tl(jmp_pc, tmp_pc); + dc->jmp_pc_imm = tmp_pc; dc->delayed_branch = 2; return true; } @@ -665,11 +615,11 @@ static bool trans_l_jal(DisasContext *dc, arg_l_jal *a, uint32_t insn) target_ulong tmp_pc = dc->base.pc_next + a->n * 4; target_ulong ret_pc = dc->base.pc_next + 8; - LOG_DIS("l.jal %d\n", a->n); tcg_gen_movi_tl(cpu_R[9], ret_pc); /* Optimize jal being used to load the PC for PIC. */ if (tmp_pc != ret_pc) { tcg_gen_movi_tl(jmp_pc, tmp_pc); + dc->jmp_pc_imm = tmp_pc; dc->delayed_branch = 2; } return true; @@ -692,21 +642,18 @@ static void do_bf(DisasContext *dc, arg_l_bf *a, TCGCond cond) static bool trans_l_bf(DisasContext *dc, arg_l_bf *a, uint32_t insn) { - LOG_DIS("l.bf %d\n", a->n); do_bf(dc, a, TCG_COND_NE); return true; } static bool trans_l_bnf(DisasContext *dc, arg_l_bf *a, uint32_t insn) { - LOG_DIS("l.bnf %d\n", a->n); do_bf(dc, a, TCG_COND_EQ); return true; } static bool trans_l_jr(DisasContext *dc, arg_l_jr *a, uint32_t insn) { - LOG_DIS("l.jr r%d\n", a->b); tcg_gen_mov_tl(jmp_pc, cpu_R[a->b]); dc->delayed_branch = 2; return true; @@ -714,7 +661,6 @@ static bool trans_l_jr(DisasContext *dc, arg_l_jr *a, uint32_t insn) static bool trans_l_jalr(DisasContext *dc, arg_l_jalr *a, uint32_t insn) { - LOG_DIS("l.jalr r%d\n", a->b); tcg_gen_mov_tl(jmp_pc, cpu_R[a->b]); tcg_gen_movi_tl(cpu_R[9], dc->base.pc_next + 8); dc->delayed_branch = 2; @@ -725,8 +671,6 @@ static bool trans_l_lwa(DisasContext *dc, arg_load *a, uint32_t insn) { TCGv ea; - LOG_DIS("l.lwa r%d, r%d, %d\n", a->d, a->a, a->i); - check_r0_write(a->d); ea = tcg_temp_new(); tcg_gen_addi_tl(ea, cpu_R[a->a], a->i); @@ -750,42 +694,36 @@ static void do_load(DisasContext *dc, arg_load *a, TCGMemOp mop) static bool trans_l_lwz(DisasContext *dc, arg_load *a, uint32_t insn) { - LOG_DIS("l.lwz r%d, r%d, %d\n", a->d, a->a, a->i); do_load(dc, a, MO_TEUL); return true; } static bool trans_l_lws(DisasContext *dc, arg_load *a, uint32_t insn) { - LOG_DIS("l.lws r%d, r%d, %d\n", a->d, a->a, a->i); do_load(dc, a, MO_TESL); return true; } static bool trans_l_lbz(DisasContext *dc, arg_load *a, uint32_t insn) { - LOG_DIS("l.lbz r%d, r%d, %d\n", a->d, a->a, a->i); do_load(dc, a, MO_UB); return true; } static bool trans_l_lbs(DisasContext *dc, arg_load *a, uint32_t insn) { - LOG_DIS("l.lbs r%d, r%d, %d\n", a->d, a->a, a->i); do_load(dc, a, MO_SB); return true; } static bool trans_l_lhz(DisasContext *dc, arg_load *a, uint32_t insn) { - LOG_DIS("l.lhz r%d, r%d, %d\n", a->d, a->a, a->i); do_load(dc, a, MO_TEUW); return true; } static bool trans_l_lhs(DisasContext *dc, arg_load *a, uint32_t insn) { - LOG_DIS("l.lhs r%d, r%d, %d\n", a->d, a->a, a->i); do_load(dc, a, MO_TESW); return true; } @@ -795,8 +733,6 @@ static bool trans_l_swa(DisasContext *dc, arg_store *a, uint32_t insn) TCGv ea, val; TCGLabel *lab_fail, *lab_done; - LOG_DIS("l.swa r%d, r%d, %d\n", a->a, a->b, a->i); - ea = tcg_temp_new(); tcg_gen_addi_tl(ea, cpu_R[a->a], a->i); @@ -837,28 +773,24 @@ static void do_store(DisasContext *dc, arg_store *a, TCGMemOp mop) static bool trans_l_sw(DisasContext *dc, arg_store *a, uint32_t insn) { - LOG_DIS("l.sw r%d, r%d, %d\n", a->a, a->b, a->i); do_store(dc, a, MO_TEUL); return true; } static bool trans_l_sb(DisasContext *dc, arg_store *a, uint32_t insn) { - LOG_DIS("l.sb r%d, r%d, %d\n", a->a, a->b, a->i); do_store(dc, a, MO_UB); return true; } static bool trans_l_sh(DisasContext *dc, arg_store *a, uint32_t insn) { - LOG_DIS("l.sh r%d, r%d, %d\n", a->a, a->b, a->i); do_store(dc, a, MO_TEUW); return true; } static bool trans_l_nop(DisasContext *dc, arg_l_nop *a, uint32_t insn) { - LOG_DIS("l.nop %d\n", a->k); return true; } @@ -866,7 +798,6 @@ static bool trans_l_addi(DisasContext *dc, arg_rri *a, uint32_t insn) { TCGv t0; - LOG_DIS("l.addi r%d, r%d, %d\n", a->d, a->a, a->i); check_r0_write(a->d); t0 = tcg_const_tl(a->i); gen_add(dc, cpu_R[a->d], cpu_R[a->a], t0); @@ -878,7 +809,6 @@ static bool trans_l_addic(DisasContext *dc, arg_rri *a, uint32_t insn) { TCGv t0; - LOG_DIS("l.addic r%d, r%d, %d\n", a->d, a->a, a->i); check_r0_write(a->d); t0 = tcg_const_tl(a->i); gen_addc(dc, cpu_R[a->d], cpu_R[a->a], t0); @@ -890,7 +820,6 @@ static bool trans_l_muli(DisasContext *dc, arg_rri *a, uint32_t insn) { TCGv t0; - LOG_DIS("l.muli r%d, r%d, %d\n", a->d, a->a, a->i); check_r0_write(a->d); t0 = tcg_const_tl(a->i); gen_mul(dc, cpu_R[a->d], cpu_R[a->a], t0); @@ -902,7 +831,6 @@ static bool trans_l_maci(DisasContext *dc, arg_l_maci *a, uint32_t insn) { TCGv t0; - LOG_DIS("l.maci r%d, %d\n", a->a, a->i); t0 = tcg_const_tl(a->i); gen_mac(dc, cpu_R[a->a], t0); tcg_temp_free(t0); @@ -911,7 +839,6 @@ static bool trans_l_maci(DisasContext *dc, arg_l_maci *a, uint32_t insn) static bool trans_l_andi(DisasContext *dc, arg_rrk *a, uint32_t insn) { - LOG_DIS("l.andi r%d, r%d, %d\n", a->d, a->a, a->k); check_r0_write(a->d); tcg_gen_andi_tl(cpu_R[a->d], cpu_R[a->a], a->k); return true; @@ -919,7 +846,6 @@ static bool trans_l_andi(DisasContext *dc, arg_rrk *a, uint32_t insn) static bool trans_l_ori(DisasContext *dc, arg_rrk *a, uint32_t insn) { - LOG_DIS("l.ori r%d, r%d, %d\n", a->d, a->a, a->k); check_r0_write(a->d); tcg_gen_ori_tl(cpu_R[a->d], cpu_R[a->a], a->k); return true; @@ -927,7 +853,6 @@ static bool trans_l_ori(DisasContext *dc, arg_rrk *a, uint32_t insn) static bool trans_l_xori(DisasContext *dc, arg_rri *a, uint32_t insn) { - LOG_DIS("l.xori r%d, r%d, %d\n", a->d, a->a, a->i); check_r0_write(a->d); tcg_gen_xori_tl(cpu_R[a->d], cpu_R[a->a], a->i); return true; @@ -935,72 +860,73 @@ static bool trans_l_xori(DisasContext *dc, arg_rri *a, uint32_t insn) static bool trans_l_mfspr(DisasContext *dc, arg_l_mfspr *a, uint32_t insn) { - LOG_DIS("l.mfspr r%d, r%d, %d\n", a->d, a->a, a->k); check_r0_write(a->d); -#ifdef CONFIG_USER_ONLY - gen_illegal_exception(dc); -#else - if (dc->mem_idx == MMU_USER_IDX) { + if (is_user(dc)) { gen_illegal_exception(dc); } else { - TCGv_i32 ti = tcg_const_i32(a->k); - gen_helper_mfspr(cpu_R[a->d], cpu_env, cpu_R[a->d], cpu_R[a->a], ti); - tcg_temp_free_i32(ti); + TCGv spr = tcg_temp_new(); + tcg_gen_ori_tl(spr, cpu_R[a->a], a->k); + gen_helper_mfspr(cpu_R[a->d], cpu_env, cpu_R[a->d], spr); + tcg_temp_free(spr); } -#endif return true; } static bool trans_l_mtspr(DisasContext *dc, arg_l_mtspr *a, uint32_t insn) { - LOG_DIS("l.mtspr r%d, r%d, %d\n", a->a, a->b, a->k); - -#ifdef CONFIG_USER_ONLY - gen_illegal_exception(dc); -#else - if (dc->mem_idx == MMU_USER_IDX) { + if (is_user(dc)) { gen_illegal_exception(dc); } else { - TCGv_i32 ti = tcg_const_i32(a->k); - gen_helper_mtspr(cpu_env, cpu_R[a->a], cpu_R[a->b], ti); - tcg_temp_free_i32(ti); + TCGv spr; + + /* For SR, we will need to exit the TB to recognize the new + * exception state. For NPC, in theory this counts as a branch + * (although the SPR only exists for use by an ICE). Save all + * of the cpu state first, allowing it to be overwritten. + */ + if (dc->delayed_branch) { + tcg_gen_mov_tl(cpu_pc, jmp_pc); + tcg_gen_discard_tl(jmp_pc); + } else { + tcg_gen_movi_tl(cpu_pc, dc->base.pc_next + 4); + } + dc->base.is_jmp = DISAS_EXIT; + + spr = tcg_temp_new(); + tcg_gen_ori_tl(spr, cpu_R[a->a], a->k); + gen_helper_mtspr(cpu_env, spr, cpu_R[a->b]); + tcg_temp_free(spr); } -#endif return true; } static bool trans_l_mac(DisasContext *dc, arg_ab *a, uint32_t insn) { - LOG_DIS("l.mac r%d, r%d\n", a->a, a->b); gen_mac(dc, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_msb(DisasContext *dc, arg_ab *a, uint32_t insn) { - LOG_DIS("l.msb r%d, r%d\n", a->a, a->b); gen_msb(dc, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_macu(DisasContext *dc, arg_ab *a, uint32_t insn) { - LOG_DIS("l.mac r%d, r%d\n", a->a, a->b); gen_macu(dc, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_msbu(DisasContext *dc, arg_ab *a, uint32_t insn) { - LOG_DIS("l.msb r%d, r%d\n", a->a, a->b); gen_msbu(dc, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_slli(DisasContext *dc, arg_dal *a, uint32_t insn) { - LOG_DIS("l.slli r%d, r%d, %d\n", a->d, a->a, a->l); check_r0_write(a->d); tcg_gen_shli_tl(cpu_R[a->d], cpu_R[a->a], a->l & (TARGET_LONG_BITS - 1)); return true; @@ -1008,7 +934,6 @@ static bool trans_l_slli(DisasContext *dc, arg_dal *a, uint32_t insn) static bool trans_l_srli(DisasContext *dc, arg_dal *a, uint32_t insn) { - LOG_DIS("l.srli r%d, r%d, %d\n", a->d, a->a, a->l); check_r0_write(a->d); tcg_gen_shri_tl(cpu_R[a->d], cpu_R[a->a], a->l & (TARGET_LONG_BITS - 1)); return true; @@ -1016,7 +941,6 @@ static bool trans_l_srli(DisasContext *dc, arg_dal *a, uint32_t insn) static bool trans_l_srai(DisasContext *dc, arg_dal *a, uint32_t insn) { - LOG_DIS("l.srai r%d, r%d, %d\n", a->d, a->a, a->l); check_r0_write(a->d); tcg_gen_sari_tl(cpu_R[a->d], cpu_R[a->a], a->l & (TARGET_LONG_BITS - 1)); return true; @@ -1024,7 +948,6 @@ static bool trans_l_srai(DisasContext *dc, arg_dal *a, uint32_t insn) static bool trans_l_rori(DisasContext *dc, arg_dal *a, uint32_t insn) { - LOG_DIS("l.rori r%d, r%d, %d\n", a->d, a->a, a->l); check_r0_write(a->d); tcg_gen_rotri_tl(cpu_R[a->d], cpu_R[a->a], a->l & (TARGET_LONG_BITS - 1)); return true; @@ -1032,7 +955,6 @@ static bool trans_l_rori(DisasContext *dc, arg_dal *a, uint32_t insn) static bool trans_l_movhi(DisasContext *dc, arg_l_movhi *a, uint32_t insn) { - LOG_DIS("l.movhi r%d, %d\n", a->d, a->k); check_r0_write(a->d); tcg_gen_movi_tl(cpu_R[a->d], a->k << 16); return true; @@ -1040,7 +962,6 @@ static bool trans_l_movhi(DisasContext *dc, arg_l_movhi *a, uint32_t insn) static bool trans_l_macrc(DisasContext *dc, arg_l_macrc *a, uint32_t insn) { - LOG_DIS("l.macrc r%d\n", a->d); check_r0_write(a->d); tcg_gen_trunc_i64_tl(cpu_R[a->d], cpu_mac); tcg_gen_movi_i64(cpu_mac, 0); @@ -1049,147 +970,126 @@ static bool trans_l_macrc(DisasContext *dc, arg_l_macrc *a, uint32_t insn) static bool trans_l_sfeq(DisasContext *dc, arg_ab *a, TCGCond cond) { - LOG_DIS("l.sfeq r%d, r%d\n", a->a, a->b); tcg_gen_setcond_tl(TCG_COND_EQ, cpu_sr_f, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_sfne(DisasContext *dc, arg_ab *a, TCGCond cond) { - LOG_DIS("l.sfne r%d, r%d\n", a->a, a->b); tcg_gen_setcond_tl(TCG_COND_NE, cpu_sr_f, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_sfgtu(DisasContext *dc, arg_ab *a, TCGCond cond) { - LOG_DIS("l.sfgtu r%d, r%d\n", a->a, a->b); tcg_gen_setcond_tl(TCG_COND_GTU, cpu_sr_f, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_sfgeu(DisasContext *dc, arg_ab *a, TCGCond cond) { - LOG_DIS("l.sfgeu r%d, r%d\n", a->a, a->b); tcg_gen_setcond_tl(TCG_COND_GEU, cpu_sr_f, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_sfltu(DisasContext *dc, arg_ab *a, TCGCond cond) { - LOG_DIS("l.sfltu r%d, r%d\n", a->a, a->b); tcg_gen_setcond_tl(TCG_COND_LTU, cpu_sr_f, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_sfleu(DisasContext *dc, arg_ab *a, TCGCond cond) { - LOG_DIS("l.sfleu r%d, r%d\n", a->a, a->b); tcg_gen_setcond_tl(TCG_COND_LEU, cpu_sr_f, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_sfgts(DisasContext *dc, arg_ab *a, TCGCond cond) { - LOG_DIS("l.sfgts r%d, r%d\n", a->a, a->b); tcg_gen_setcond_tl(TCG_COND_GT, cpu_sr_f, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_sfges(DisasContext *dc, arg_ab *a, TCGCond cond) { - LOG_DIS("l.sfges r%d, r%d\n", a->a, a->b); tcg_gen_setcond_tl(TCG_COND_GE, cpu_sr_f, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_sflts(DisasContext *dc, arg_ab *a, TCGCond cond) { - LOG_DIS("l.sflts r%d, r%d\n", a->a, a->b); tcg_gen_setcond_tl(TCG_COND_LT, cpu_sr_f, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_sfles(DisasContext *dc, arg_ab *a, TCGCond cond) { - LOG_DIS("l.sfles r%d, r%d\n", a->a, a->b); tcg_gen_setcond_tl(TCG_COND_LE, cpu_sr_f, cpu_R[a->a], cpu_R[a->b]); return true; } static bool trans_l_sfeqi(DisasContext *dc, arg_ai *a, TCGCond cond) { - LOG_DIS("l.sfeqi r%d, %d\n", a->a, a->i); tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_sr_f, cpu_R[a->a], a->i); return true; } static bool trans_l_sfnei(DisasContext *dc, arg_ai *a, TCGCond cond) { - LOG_DIS("l.sfnei r%d, %d\n", a->a, a->i); tcg_gen_setcondi_tl(TCG_COND_NE, cpu_sr_f, cpu_R[a->a], a->i); return true; } static bool trans_l_sfgtui(DisasContext *dc, arg_ai *a, TCGCond cond) { - LOG_DIS("l.sfgtui r%d, %d\n", a->a, a->i); tcg_gen_setcondi_tl(TCG_COND_GTU, cpu_sr_f, cpu_R[a->a], a->i); return true; } static bool trans_l_sfgeui(DisasContext *dc, arg_ai *a, TCGCond cond) { - LOG_DIS("l.sfgeui r%d, %d\n", a->a, a->i); tcg_gen_setcondi_tl(TCG_COND_GEU, cpu_sr_f, cpu_R[a->a], a->i); return true; } static bool trans_l_sfltui(DisasContext *dc, arg_ai *a, TCGCond cond) { - LOG_DIS("l.sfltui r%d, %d\n", a->a, a->i); tcg_gen_setcondi_tl(TCG_COND_LTU, cpu_sr_f, cpu_R[a->a], a->i); return true; } static bool trans_l_sfleui(DisasContext *dc, arg_ai *a, TCGCond cond) { - LOG_DIS("l.sfleui r%d, %d\n", a->a, a->i); tcg_gen_setcondi_tl(TCG_COND_LEU, cpu_sr_f, cpu_R[a->a], a->i); return true; } static bool trans_l_sfgtsi(DisasContext *dc, arg_ai *a, TCGCond cond) { - LOG_DIS("l.sfgtsi r%d, %d\n", a->a, a->i); tcg_gen_setcondi_tl(TCG_COND_GT, cpu_sr_f, cpu_R[a->a], a->i); return true; } static bool trans_l_sfgesi(DisasContext *dc, arg_ai *a, TCGCond cond) { - LOG_DIS("l.sfgesi r%d, %d\n", a->a, a->i); tcg_gen_setcondi_tl(TCG_COND_GE, cpu_sr_f, cpu_R[a->a], a->i); return true; } static bool trans_l_sfltsi(DisasContext *dc, arg_ai *a, TCGCond cond) { - LOG_DIS("l.sfltsi r%d, %d\n", a->a, a->i); tcg_gen_setcondi_tl(TCG_COND_LT, cpu_sr_f, cpu_R[a->a], a->i); return true; } static bool trans_l_sflesi(DisasContext *dc, arg_ai *a, TCGCond cond) { - LOG_DIS("l.sflesi r%d, %d\n", a->a, a->i); tcg_gen_setcondi_tl(TCG_COND_LE, cpu_sr_f, cpu_R[a->a], a->i); return true; } static bool trans_l_sys(DisasContext *dc, arg_l_sys *a, uint32_t insn) { - LOG_DIS("l.sys %d\n", a->k); tcg_gen_movi_tl(cpu_pc, dc->base.pc_next); gen_exception(dc, EXCP_SYSCALL); dc->base.is_jmp = DISAS_NORETURN; @@ -1198,7 +1098,6 @@ static bool trans_l_sys(DisasContext *dc, arg_l_sys *a, uint32_t insn) static bool trans_l_trap(DisasContext *dc, arg_l_trap *a, uint32_t insn) { - LOG_DIS("l.trap %d\n", a->k); tcg_gen_movi_tl(cpu_pc, dc->base.pc_next); gen_exception(dc, EXCP_TRAP); dc->base.is_jmp = DISAS_NORETURN; @@ -1207,37 +1106,28 @@ static bool trans_l_trap(DisasContext *dc, arg_l_trap *a, uint32_t insn) static bool trans_l_msync(DisasContext *dc, arg_l_msync *a, uint32_t insn) { - LOG_DIS("l.msync\n"); tcg_gen_mb(TCG_MO_ALL); return true; } static bool trans_l_psync(DisasContext *dc, arg_l_psync *a, uint32_t insn) { - LOG_DIS("l.psync\n"); return true; } static bool trans_l_csync(DisasContext *dc, arg_l_csync *a, uint32_t insn) { - LOG_DIS("l.csync\n"); return true; } static bool trans_l_rfe(DisasContext *dc, arg_l_rfe *a, uint32_t insn) { - LOG_DIS("l.rfe\n"); - -#ifdef CONFIG_USER_ONLY - gen_illegal_exception(dc); -#else - if (dc->mem_idx == MMU_USER_IDX) { + if (is_user(dc)) { gen_illegal_exception(dc); } else { gen_helper_rfe(cpu_env); - dc->base.is_jmp = DISAS_UPDATE; + dc->base.is_jmp = DISAS_EXIT; } -#endif return true; } @@ -1274,56 +1164,48 @@ static void do_fpcmp(DisasContext *dc, arg_ab *a, static bool trans_lf_add_s(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("lf.add.s r%d, r%d, r%d\n", a->d, a->a, a->b); do_fp3(dc, a, gen_helper_float_add_s); return true; } static bool trans_lf_sub_s(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("lf.sub.s r%d, r%d, r%d\n", a->d, a->a, a->b); do_fp3(dc, a, gen_helper_float_sub_s); return true; } static bool trans_lf_mul_s(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("lf.mul.s r%d, r%d, r%d\n", a->d, a->a, a->b); do_fp3(dc, a, gen_helper_float_mul_s); return true; } static bool trans_lf_div_s(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("lf.div.s r%d, r%d, r%d\n", a->d, a->a, a->b); do_fp3(dc, a, gen_helper_float_div_s); return true; } static bool trans_lf_rem_s(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("lf.rem.s r%d, r%d, r%d\n", a->d, a->a, a->b); do_fp3(dc, a, gen_helper_float_rem_s); return true; } static bool trans_lf_itof_s(DisasContext *dc, arg_da *a, uint32_t insn) { - LOG_DIS("lf.itof.s r%d, r%d\n", a->d, a->a); do_fp2(dc, a, gen_helper_itofs); return true; } static bool trans_lf_ftoi_s(DisasContext *dc, arg_da *a, uint32_t insn) { - LOG_DIS("lf.ftoi.s r%d, r%d\n", a->d, a->a); do_fp2(dc, a, gen_helper_ftois); return true; } static bool trans_lf_madd_s(DisasContext *dc, arg_dab *a, uint32_t insn) { - LOG_DIS("lf.madd.s r%d, r%d, r%d\n", a->d, a->a, a->b); check_r0_write(a->d); gen_helper_float_madd_s(cpu_R[a->d], cpu_env, cpu_R[a->d], cpu_R[a->a], cpu_R[a->b]); @@ -1333,42 +1215,36 @@ static bool trans_lf_madd_s(DisasContext *dc, arg_dab *a, uint32_t insn) static bool trans_lf_sfeq_s(DisasContext *dc, arg_ab *a, uint32_t insn) { - LOG_DIS("lf.sfeq.s r%d, r%d\n", a->a, a->b); do_fpcmp(dc, a, gen_helper_float_eq_s, false, false); return true; } static bool trans_lf_sfne_s(DisasContext *dc, arg_ab *a, uint32_t insn) { - LOG_DIS("lf.sfne.s r%d, r%d\n", a->a, a->b); do_fpcmp(dc, a, gen_helper_float_eq_s, true, false); return true; } static bool trans_lf_sfgt_s(DisasContext *dc, arg_ab *a, uint32_t insn) { - LOG_DIS("lf.sfgt.s r%d, r%d\n", a->a, a->b); do_fpcmp(dc, a, gen_helper_float_lt_s, false, true); return true; } static bool trans_lf_sfge_s(DisasContext *dc, arg_ab *a, uint32_t insn) { - LOG_DIS("lf.sfge.s r%d, r%d\n", a->a, a->b); do_fpcmp(dc, a, gen_helper_float_le_s, false, true); return true; } static bool trans_lf_sflt_s(DisasContext *dc, arg_ab *a, uint32_t insn) { - LOG_DIS("lf.sflt.s r%d, r%d\n", a->a, a->b); do_fpcmp(dc, a, gen_helper_float_lt_s, false, false); return true; } static bool trans_lf_sfle_s(DisasContext *dc, arg_ab *a, uint32_t insn) { - LOG_DIS("lf.sfle.s r%d, r%d\n", a->a, a->b); do_fpcmp(dc, a, gen_helper_float_le_s, false, false); return true; } @@ -1382,6 +1258,8 @@ static void openrisc_tr_init_disas_context(DisasContextBase *dcb, CPUState *cs) dc->mem_idx = cpu_mmu_index(env, false); dc->tb_flags = dc->base.tb->flags; dc->delayed_branch = (dc->tb_flags & TB_FLAGS_DFLAG) != 0; + dc->jmp_pc_imm = -1; + bound = -(dc->base.pc_first | TARGET_PAGE_MASK) / 4; dc->base.max_insns = MIN(dc->base.max_insns, bound); } @@ -1434,50 +1312,81 @@ static void openrisc_tr_translate_insn(DisasContextBase *dcbase, CPUState *cs) } dc->base.pc_next += 4; - /* delay slot */ - if (dc->delayed_branch) { - dc->delayed_branch--; - if (!dc->delayed_branch) { - tcg_gen_mov_tl(cpu_pc, jmp_pc); - tcg_gen_discard_tl(jmp_pc); - dc->base.is_jmp = DISAS_UPDATE; - return; - } + /* When exiting the delay slot normally, exit via jmp_pc. + * For DISAS_NORETURN, we have raised an exception and already exited. + * For DISAS_EXIT, we found l.rfe in a delay slot. There's nothing + * in the manual saying this is illegal, but it surely it should. + * At least or1ksim overrides pcnext and ignores the branch. + */ + if (dc->delayed_branch + && --dc->delayed_branch == 0 + && dc->base.is_jmp == DISAS_NEXT) { + dc->base.is_jmp = DISAS_JUMP; } } static void openrisc_tr_tb_stop(DisasContextBase *dcbase, CPUState *cs) { DisasContext *dc = container_of(dcbase, DisasContext, base); + target_ulong jmp_dest; + + /* If we have already exited the TB, nothing following has effect. */ + if (dc->base.is_jmp == DISAS_NORETURN) { + return; + } + /* Adjust the delayed branch state for the next TB. */ if ((dc->tb_flags & TB_FLAGS_DFLAG ? 1 : 0) != (dc->delayed_branch != 0)) { tcg_gen_movi_i32(cpu_dflag, dc->delayed_branch != 0); } - tcg_gen_movi_tl(cpu_ppc, dc->base.pc_next - 4); - if (dc->base.is_jmp == DISAS_NEXT) { - dc->base.is_jmp = DISAS_UPDATE; - tcg_gen_movi_tl(cpu_pc, dc->base.pc_next); - } - if (unlikely(dc->base.singlestep_enabled)) { - gen_exception(dc, EXCP_DEBUG); - } else { - switch (dc->base.is_jmp) { - case DISAS_TOO_MANY: - gen_goto_tb(dc, 0, dc->base.pc_next); - break; - case DISAS_NORETURN: - case DISAS_JUMP: - case DISAS_TB_JUMP: + /* For DISAS_TOO_MANY, jump to the next insn. */ + jmp_dest = dc->base.pc_next; + tcg_gen_movi_tl(cpu_ppc, jmp_dest - 4); + + switch (dc->base.is_jmp) { + case DISAS_JUMP: + jmp_dest = dc->jmp_pc_imm; + if (jmp_dest == -1) { + /* The jump destination is indirect/computed; use jmp_pc. */ + tcg_gen_mov_tl(cpu_pc, jmp_pc); + tcg_gen_discard_tl(jmp_pc); + if (unlikely(dc->base.singlestep_enabled)) { + gen_exception(dc, EXCP_DEBUG); + } else { + tcg_gen_lookup_and_goto_ptr(); + } break; - case DISAS_UPDATE: - /* indicate that the hash table must be used - to find the next TB */ + } + /* The jump destination is direct; use jmp_pc_imm. + However, we will have stored into jmp_pc as well; + we know now that it wasn't needed. */ + tcg_gen_discard_tl(jmp_pc); + /* fallthru */ + + case DISAS_TOO_MANY: + if (unlikely(dc->base.singlestep_enabled)) { + tcg_gen_movi_tl(cpu_pc, jmp_dest); + gen_exception(dc, EXCP_DEBUG); + } else if ((dc->base.pc_first ^ jmp_dest) & TARGET_PAGE_MASK) { + tcg_gen_movi_tl(cpu_pc, jmp_dest); + tcg_gen_lookup_and_goto_ptr(); + } else { + tcg_gen_goto_tb(0); + tcg_gen_movi_tl(cpu_pc, jmp_dest); + tcg_gen_exit_tb(dc->base.tb, 0); + } + break; + + case DISAS_EXIT: + if (unlikely(dc->base.singlestep_enabled)) { + gen_exception(dc, EXCP_DEBUG); + } else { tcg_gen_exit_tb(NULL, 0); - break; - default: - g_assert_not_reached(); } + break; + default: + g_assert_not_reached(); } } diff --git a/target/ppc/cpu.h b/target/ppc/cpu.h index c7f3fb6b73..4edcf62cf7 100644 --- a/target/ppc/cpu.h +++ b/target/ppc/cpu.h @@ -196,7 +196,6 @@ enum { /* QEMU exceptions: special cases we want to stop translation */ POWERPC_EXCP_SYNC = 0x202, /* context synchronizing instruction */ POWERPC_EXCP_SYSCALL_USER = 0x203, /* System call in user mode only */ - POWERPC_EXCP_STCX = 0x204 /* Conditional stores in user mode */ }; /* Exceptions error codes */ @@ -994,10 +993,6 @@ struct CPUPPCState { /* Reservation value */ target_ulong reserve_val; target_ulong reserve_val2; - /* Reservation store address */ - target_ulong reserve_ea; - /* Reserved store source register and size */ - target_ulong reserve_info; /* Those ones are used in supervisor mode only */ /* machine state register */ @@ -1015,6 +1010,9 @@ struct CPUPPCState { /* Next instruction pointer */ target_ulong nip; + /* High part of 128-bit helper return. */ + uint64_t retxh; + int access_type; /* when a memory exception occurs, the access type is stored here */ diff --git a/target/ppc/excp_helper.c b/target/ppc/excp_helper.c index c092fbead0..d6e97a90e0 100644 --- a/target/ppc/excp_helper.c +++ b/target/ppc/excp_helper.c @@ -22,7 +22,7 @@ #include "exec/helper-proto.h" #include "exec/exec-all.h" #include "exec/cpu_ldst.h" - +#include "internal.h" #include "helper_regs.h" //#define DEBUG_OP @@ -1198,3 +1198,19 @@ void helper_book3s_msgsnd(target_ulong rb) qemu_mutex_unlock_iothread(); } #endif + +void ppc_cpu_do_unaligned_access(CPUState *cs, vaddr vaddr, + MMUAccessType access_type, + int mmu_idx, uintptr_t retaddr) +{ + CPUPPCState *env = cs->env_ptr; + uint32_t insn; + + /* Restore state and reload the insn we executed, for filling in DSISR. */ + cpu_restore_state(cs, retaddr, true); + insn = cpu_ldl_code(env, env->nip); + + cs->exception_index = POWERPC_EXCP_ALIGN; + env->error_code = insn & 0x03FF0000; + cpu_loop_exit(cs); +} diff --git a/target/ppc/fpu_helper.c b/target/ppc/fpu_helper.c index 7714bfe0f9..8675d931b6 100644 --- a/target/ppc/fpu_helper.c +++ b/target/ppc/fpu_helper.c @@ -274,6 +274,7 @@ static inline void float_inexact_excp(CPUPPCState *env) { CPUState *cs = CPU(ppc_env_get_cpu(env)); + env->fpscr |= 1 << FPSCR_FI; env->fpscr |= 1 << FPSCR_XX; /* Update the floating-point exception summary */ env->fpscr |= FP_FX; @@ -533,6 +534,7 @@ static void do_float_check_status(CPUPPCState *env, uintptr_t raddr) { CPUState *cs = CPU(ppc_env_get_cpu(env)); int status = get_float_exception_flags(&env->fp_status); + bool inexact_happened = false; if (status & float_flag_divbyzero) { float_zero_divide_excp(env, raddr); @@ -542,6 +544,12 @@ static void do_float_check_status(CPUPPCState *env, uintptr_t raddr) float_underflow_excp(env); } else if (status & float_flag_inexact) { float_inexact_excp(env); + inexact_happened = true; + } + + /* if the inexact flag was not set */ + if (inexact_happened == false) { + env->fpscr &= ~(1 << FPSCR_FI); /* clear the FPSCR[FI] bit */ } if (cs->exception_index == POWERPC_EXCP_PROGRAM && diff --git a/target/ppc/helper.h b/target/ppc/helper.h index d751f0e219..5706c2497f 100644 --- a/target/ppc/helper.h +++ b/target/ppc/helper.h @@ -799,3 +799,14 @@ DEF_HELPER_4(dscliq, void, env, fprp, fprp, i32) DEF_HELPER_1(tbegin, void, env) DEF_HELPER_FLAGS_1(fixup_thrm, TCG_CALL_NO_RWG, void, env) + +#if defined(TARGET_PPC64) && defined(CONFIG_ATOMIC128) +DEF_HELPER_FLAGS_3(lq_le_parallel, TCG_CALL_NO_WG, i64, env, tl, i32) +DEF_HELPER_FLAGS_3(lq_be_parallel, TCG_CALL_NO_WG, i64, env, tl, i32) +DEF_HELPER_FLAGS_5(stq_le_parallel, TCG_CALL_NO_WG, + void, env, tl, i64, i64, i32) +DEF_HELPER_FLAGS_5(stq_be_parallel, TCG_CALL_NO_WG, + void, env, tl, i64, i64, i32) +DEF_HELPER_5(stqcx_le_parallel, i32, env, tl, i64, i64, i32) +DEF_HELPER_5(stqcx_be_parallel, i32, env, tl, i64, i64, i32) +#endif diff --git a/target/ppc/internal.h b/target/ppc/internal.h index 1f441c6483..a9bcadff42 100644 --- a/target/ppc/internal.h +++ b/target/ppc/internal.h @@ -252,4 +252,9 @@ static inline void putVSR(int n, ppc_vsr_t *vsr, CPUPPCState *env) void helper_compute_fprf_float16(CPUPPCState *env, float16 arg); void helper_compute_fprf_float32(CPUPPCState *env, float32 arg); void helper_compute_fprf_float128(CPUPPCState *env, float128 arg); + +/* Raise a data fault alignment exception for the specified virtual address */ +void ppc_cpu_do_unaligned_access(CPUState *cs, vaddr addr, + MMUAccessType access_type, + int mmu_idx, uintptr_t retaddr); #endif /* PPC_INTERNAL_H */ diff --git a/target/ppc/kvm.c b/target/ppc/kvm.c index 4df4ff6cbf..9211ee2ee1 100644 --- a/target/ppc/kvm.c +++ b/target/ppc/kvm.c @@ -248,107 +248,25 @@ static int kvm_booke206_tlb_init(PowerPCCPU *cpu) #if defined(TARGET_PPC64) -static void kvm_get_fallback_smmu_info(PowerPCCPU *cpu, - struct kvm_ppc_smmu_info *info) +static void kvm_get_smmu_info(struct kvm_ppc_smmu_info *info, Error **errp) { - CPUPPCState *env = &cpu->env; - CPUState *cs = CPU(cpu); - - memset(info, 0, sizeof(*info)); - - /* We don't have the new KVM_PPC_GET_SMMU_INFO ioctl, so - * need to "guess" what the supported page sizes are. - * - * For that to work we make a few assumptions: - * - * - Check whether we are running "PR" KVM which only supports 4K - * and 16M pages, but supports them regardless of the backing - * store characteritics. We also don't support 1T segments. - * - * This is safe as if HV KVM ever supports that capability or PR - * KVM grows supports for more page/segment sizes, those versions - * will have implemented KVM_CAP_PPC_GET_SMMU_INFO and thus we - * will not hit this fallback - * - * - Else we are running HV KVM. This means we only support page - * sizes that fit in the backing store. Additionally we only - * advertize 64K pages if the processor is ARCH 2.06 and we assume - * P7 encodings for the SLB and hash table. Here too, we assume - * support for any newer processor will mean a kernel that - * implements KVM_CAP_PPC_GET_SMMU_INFO and thus doesn't hit - * this fallback. - */ - if (kvmppc_is_pr(cs->kvm_state)) { - /* No flags */ - info->flags = 0; - info->slb_size = 64; - - /* Standard 4k base page size segment */ - info->sps[0].page_shift = 12; - info->sps[0].slb_enc = 0; - info->sps[0].enc[0].page_shift = 12; - info->sps[0].enc[0].pte_enc = 0; - - /* Standard 16M large page size segment */ - info->sps[1].page_shift = 24; - info->sps[1].slb_enc = SLB_VSID_L; - info->sps[1].enc[0].page_shift = 24; - info->sps[1].enc[0].pte_enc = 0; - } else { - int i = 0; - - /* HV KVM has backing store size restrictions */ - info->flags = KVM_PPC_PAGE_SIZES_REAL; - - if (ppc_hash64_has(cpu, PPC_HASH64_1TSEG)) { - info->flags |= KVM_PPC_1T_SEGMENTS; - } - - if (env->mmu_model == POWERPC_MMU_2_06 || - env->mmu_model == POWERPC_MMU_2_07) { - info->slb_size = 32; - } else { - info->slb_size = 64; - } + int ret; - /* Standard 4k base page size segment */ - info->sps[i].page_shift = 12; - info->sps[i].slb_enc = 0; - info->sps[i].enc[0].page_shift = 12; - info->sps[i].enc[0].pte_enc = 0; - i++; - - /* 64K on MMU 2.06 and later */ - if (env->mmu_model == POWERPC_MMU_2_06 || - env->mmu_model == POWERPC_MMU_2_07) { - info->sps[i].page_shift = 16; - info->sps[i].slb_enc = 0x110; - info->sps[i].enc[0].page_shift = 16; - info->sps[i].enc[0].pte_enc = 1; - i++; - } + assert(kvm_state != NULL); - /* Standard 16M large page size segment */ - info->sps[i].page_shift = 24; - info->sps[i].slb_enc = SLB_VSID_L; - info->sps[i].enc[0].page_shift = 24; - info->sps[i].enc[0].pte_enc = 0; + if (!kvm_check_extension(kvm_state, KVM_CAP_PPC_GET_SMMU_INFO)) { + error_setg(errp, "KVM doesn't expose the MMU features it supports"); + error_append_hint(errp, "Consider switching to a newer KVM\n"); + return; } -} - -static void kvm_get_smmu_info(PowerPCCPU *cpu, struct kvm_ppc_smmu_info *info) -{ - CPUState *cs = CPU(cpu); - int ret; - if (kvm_check_extension(cs->kvm_state, KVM_CAP_PPC_GET_SMMU_INFO)) { - ret = kvm_vm_ioctl(cs->kvm_state, KVM_PPC_GET_SMMU_INFO, info); - if (ret == 0) { - return; - } + ret = kvm_vm_ioctl(kvm_state, KVM_PPC_GET_SMMU_INFO, info); + if (ret == 0) { + return; } - kvm_get_fallback_smmu_info(cpu, info); + error_setg_errno(errp, -ret, + "KVM failed to provide the MMU features it supports"); } struct ppc_radix_page_info *kvm_get_radix_page_info(void) @@ -408,14 +326,13 @@ target_ulong kvmppc_configure_v3_mmu(PowerPCCPU *cpu, bool kvmppc_hpt_needs_host_contiguous_pages(void) { - PowerPCCPU *cpu = POWERPC_CPU(first_cpu); static struct kvm_ppc_smmu_info smmu_info; if (!kvm_enabled()) { return false; } - kvm_get_smmu_info(cpu, &smmu_info); + kvm_get_smmu_info(&smmu_info, &error_fatal); return !!(smmu_info.flags & KVM_PPC_PAGE_SIZES_REAL); } @@ -423,13 +340,18 @@ void kvm_check_mmu(PowerPCCPU *cpu, Error **errp) { struct kvm_ppc_smmu_info smmu_info; int iq, ik, jq, jk; + Error *local_err = NULL; /* For now, we only have anything to check on hash64 MMUs */ if (!cpu->hash64_opts || !kvm_enabled()) { return; } - kvm_get_smmu_info(cpu, &smmu_info); + kvm_get_smmu_info(&smmu_info, &local_err); + if (local_err) { + error_propagate(errp, local_err); + return; + } if (ppc_hash64_has(cpu, PPC_HASH64_1TSEG) && !(smmu_info.flags & KVM_PPC_1T_SEGMENTS)) { @@ -2168,7 +2090,7 @@ uint64_t kvmppc_rma_size(uint64_t current_size, unsigned int hash_shift) /* Find the largest hardware supported page size that's less than * or equal to the (logical) backing page size of guest RAM */ - kvm_get_smmu_info(POWERPC_CPU(first_cpu), &info); + kvm_get_smmu_info(&info, &error_fatal); rampagesize = qemu_getrampagesize(); best_page_shift = 0; diff --git a/target/ppc/mem_helper.c b/target/ppc/mem_helper.c index a34e604db3..8f0d86d104 100644 --- a/target/ppc/mem_helper.c +++ b/target/ppc/mem_helper.c @@ -21,9 +21,9 @@ #include "exec/exec-all.h" #include "qemu/host-utils.h" #include "exec/helper-proto.h" - #include "helper_regs.h" #include "exec/cpu_ldst.h" +#include "tcg.h" #include "internal.h" //#define DEBUG_OP @@ -215,6 +215,76 @@ target_ulong helper_lscbx(CPUPPCState *env, target_ulong addr, uint32_t reg, return i; } +#if defined(TARGET_PPC64) && defined(CONFIG_ATOMIC128) +uint64_t helper_lq_le_parallel(CPUPPCState *env, target_ulong addr, + uint32_t opidx) +{ + Int128 ret = helper_atomic_ldo_le_mmu(env, addr, opidx, GETPC()); + env->retxh = int128_gethi(ret); + return int128_getlo(ret); +} + +uint64_t helper_lq_be_parallel(CPUPPCState *env, target_ulong addr, + uint32_t opidx) +{ + Int128 ret = helper_atomic_ldo_be_mmu(env, addr, opidx, GETPC()); + env->retxh = int128_gethi(ret); + return int128_getlo(ret); +} + +void helper_stq_le_parallel(CPUPPCState *env, target_ulong addr, + uint64_t lo, uint64_t hi, uint32_t opidx) +{ + Int128 val = int128_make128(lo, hi); + helper_atomic_sto_le_mmu(env, addr, val, opidx, GETPC()); +} + +void helper_stq_be_parallel(CPUPPCState *env, target_ulong addr, + uint64_t lo, uint64_t hi, uint32_t opidx) +{ + Int128 val = int128_make128(lo, hi); + helper_atomic_sto_be_mmu(env, addr, val, opidx, GETPC()); +} + +uint32_t helper_stqcx_le_parallel(CPUPPCState *env, target_ulong addr, + uint64_t new_lo, uint64_t new_hi, + uint32_t opidx) +{ + bool success = false; + + if (likely(addr == env->reserve_addr)) { + Int128 oldv, cmpv, newv; + + cmpv = int128_make128(env->reserve_val2, env->reserve_val); + newv = int128_make128(new_lo, new_hi); + oldv = helper_atomic_cmpxchgo_le_mmu(env, addr, cmpv, newv, + opidx, GETPC()); + success = int128_eq(oldv, cmpv); + } + env->reserve_addr = -1; + return env->so + success * CRF_EQ_BIT; +} + +uint32_t helper_stqcx_be_parallel(CPUPPCState *env, target_ulong addr, + uint64_t new_lo, uint64_t new_hi, + uint32_t opidx) +{ + bool success = false; + + if (likely(addr == env->reserve_addr)) { + Int128 oldv, cmpv, newv; + + cmpv = int128_make128(env->reserve_val2, env->reserve_val); + newv = int128_make128(new_lo, new_hi); + oldv = helper_atomic_cmpxchgo_be_mmu(env, addr, cmpv, newv, + opidx, GETPC()); + success = int128_eq(oldv, cmpv); + } + env->reserve_addr = -1; + return env->so + success * CRF_EQ_BIT; +} +#endif + /*****************************************************************************/ /* Altivec extension helpers */ #if defined(HOST_WORDS_BIGENDIAN) diff --git a/target/ppc/mmu_helper.c b/target/ppc/mmu_helper.c index 98ce17985b..e6739e6c24 100644 --- a/target/ppc/mmu_helper.c +++ b/target/ppc/mmu_helper.c @@ -17,6 +17,7 @@ * License along with this library; if not, see <http://www.gnu.org/licenses/>. */ #include "qemu/osdep.h" +#include "qemu/units.h" #include "cpu.h" #include "exec/helper-proto.h" #include "sysemu/kvm.h" @@ -1090,11 +1091,10 @@ static void mmubooke_dump_mmu(FILE *f, fprintf_function cpu_fprintf, pa = entry->RPN & mask; /* Extend the physical address to 36 bits */ pa |= (hwaddr)(entry->RPN & 0xF) << 32; - size /= 1024; - if (size >= 1024) { - snprintf(size_buf, sizeof(size_buf), "%3" PRId64 "M", size / 1024); + if (size >= 1 * MiB) { + snprintf(size_buf, sizeof(size_buf), "%3" PRId64 "M", size / MiB); } else { - snprintf(size_buf, sizeof(size_buf), "%3" PRId64 "k", size); + snprintf(size_buf, sizeof(size_buf), "%3" PRId64 "k", size / KiB); } cpu_fprintf(f, "0x%016" PRIx64 " 0x%016" PRIx64 " %s %-5u %08x %08x\n", (uint64_t)ea, (uint64_t)pa, size_buf, (uint32_t)entry->PID, diff --git a/target/ppc/translate.c b/target/ppc/translate.c index 3a215a1dc6..9eaa10b421 100644 --- a/target/ppc/translate.c +++ b/target/ppc/translate.c @@ -2388,23 +2388,6 @@ static inline void gen_addr_add(DisasContext *ctx, TCGv ret, TCGv arg1, } } -static inline void gen_check_align(DisasContext *ctx, TCGv EA, int mask) -{ - TCGLabel *l1 = gen_new_label(); - TCGv t0 = tcg_temp_new(); - TCGv_i32 t1, t2; - tcg_gen_andi_tl(t0, EA, mask); - tcg_gen_brcondi_tl(TCG_COND_EQ, t0, 0, l1); - t1 = tcg_const_i32(POWERPC_EXCP_ALIGN); - t2 = tcg_const_i32(ctx->opcode & 0x03FF0000); - gen_update_nip(ctx, ctx->base.pc_next - 4); - gen_helper_raise_exception_err(cpu_env, t1, t2); - tcg_temp_free_i32(t1); - tcg_temp_free_i32(t2); - gen_set_label(l1); - tcg_temp_free(t0); -} - static inline void gen_align_no_le(DisasContext *ctx) { gen_exception_err(ctx, POWERPC_EXCP_ALIGN, @@ -2607,7 +2590,7 @@ static void gen_ld(DisasContext *ctx) static void gen_lq(DisasContext *ctx) { int ra, rd; - TCGv EA; + TCGv EA, hi, lo; /* lq is a legal user mode instruction starting in ISA 2.07 */ bool legal_in_user_mode = (ctx->insns_flags2 & PPC2_LSQ_ISA207) != 0; @@ -2633,16 +2616,35 @@ static void gen_lq(DisasContext *ctx) EA = tcg_temp_new(); gen_addr_imm_index(ctx, EA, 0x0F); - /* We only need to swap high and low halves. gen_qemu_ld64_i64 does - necessary 64-bit byteswap already. */ - if (unlikely(ctx->le_mode)) { - gen_qemu_ld64_i64(ctx, cpu_gpr[rd + 1], EA); + /* Note that the low part is always in RD+1, even in LE mode. */ + lo = cpu_gpr[rd + 1]; + hi = cpu_gpr[rd]; + + if (tb_cflags(ctx->base.tb) & CF_PARALLEL) { +#ifdef CONFIG_ATOMIC128 + TCGv_i32 oi = tcg_temp_new_i32(); + if (ctx->le_mode) { + tcg_gen_movi_i32(oi, make_memop_idx(MO_LEQ, ctx->mem_idx)); + gen_helper_lq_le_parallel(lo, cpu_env, EA, oi); + } else { + tcg_gen_movi_i32(oi, make_memop_idx(MO_BEQ, ctx->mem_idx)); + gen_helper_lq_be_parallel(lo, cpu_env, EA, oi); + } + tcg_temp_free_i32(oi); + tcg_gen_ld_i64(hi, cpu_env, offsetof(CPUPPCState, retxh)); +#else + /* Restart with exclusive lock. */ + gen_helper_exit_atomic(cpu_env); + ctx->base.is_jmp = DISAS_NORETURN; +#endif + } else if (ctx->le_mode) { + tcg_gen_qemu_ld_i64(lo, EA, ctx->mem_idx, MO_LEQ); gen_addr_add(ctx, EA, EA, 8); - gen_qemu_ld64_i64(ctx, cpu_gpr[rd], EA); + tcg_gen_qemu_ld_i64(hi, EA, ctx->mem_idx, MO_LEQ); } else { - gen_qemu_ld64_i64(ctx, cpu_gpr[rd], EA); + tcg_gen_qemu_ld_i64(hi, EA, ctx->mem_idx, MO_BEQ); gen_addr_add(ctx, EA, EA, 8); - gen_qemu_ld64_i64(ctx, cpu_gpr[rd + 1], EA); + tcg_gen_qemu_ld_i64(lo, EA, ctx->mem_idx, MO_BEQ); } tcg_temp_free(EA); } @@ -2741,6 +2743,7 @@ static void gen_std(DisasContext *ctx) if ((ctx->opcode & 0x3) == 0x2) { /* stq */ bool legal_in_user_mode = (ctx->insns_flags2 & PPC2_LSQ_ISA207) != 0; bool le_is_supported = (ctx->insns_flags2 & PPC2_LSQ_ISA207) != 0; + TCGv hi, lo; if (!(ctx->insns_flags & PPC_64BX)) { gen_inval_exception(ctx, POWERPC_EXCP_INVAL_INVAL); @@ -2764,20 +2767,38 @@ static void gen_std(DisasContext *ctx) EA = tcg_temp_new(); gen_addr_imm_index(ctx, EA, 0x03); - /* We only need to swap high and low halves. gen_qemu_st64_i64 does - necessary 64-bit byteswap already. */ - if (unlikely(ctx->le_mode)) { - gen_qemu_st64_i64(ctx, cpu_gpr[rs + 1], EA); + /* Note that the low part is always in RS+1, even in LE mode. */ + lo = cpu_gpr[rs + 1]; + hi = cpu_gpr[rs]; + + if (tb_cflags(ctx->base.tb) & CF_PARALLEL) { +#ifdef CONFIG_ATOMIC128 + TCGv_i32 oi = tcg_temp_new_i32(); + if (ctx->le_mode) { + tcg_gen_movi_i32(oi, make_memop_idx(MO_LEQ, ctx->mem_idx)); + gen_helper_stq_le_parallel(cpu_env, EA, lo, hi, oi); + } else { + tcg_gen_movi_i32(oi, make_memop_idx(MO_BEQ, ctx->mem_idx)); + gen_helper_stq_be_parallel(cpu_env, EA, lo, hi, oi); + } + tcg_temp_free_i32(oi); +#else + /* Restart with exclusive lock. */ + gen_helper_exit_atomic(cpu_env); + ctx->base.is_jmp = DISAS_NORETURN; +#endif + } else if (ctx->le_mode) { + tcg_gen_qemu_st_i64(lo, EA, ctx->mem_idx, MO_LEQ); gen_addr_add(ctx, EA, EA, 8); - gen_qemu_st64_i64(ctx, cpu_gpr[rs], EA); + tcg_gen_qemu_st_i64(hi, EA, ctx->mem_idx, MO_LEQ); } else { - gen_qemu_st64_i64(ctx, cpu_gpr[rs], EA); + tcg_gen_qemu_st_i64(hi, EA, ctx->mem_idx, MO_BEQ); gen_addr_add(ctx, EA, EA, 8); - gen_qemu_st64_i64(ctx, cpu_gpr[rs + 1], EA); + tcg_gen_qemu_st_i64(lo, EA, ctx->mem_idx, MO_BEQ); } tcg_temp_free(EA); } else { - /* std / stdu*/ + /* std / stdu */ if (Rc(ctx->opcode)) { if (unlikely(rA(ctx->opcode) == 0)) { gen_inval_exception(ctx, POWERPC_EXCP_INVAL_INVAL); @@ -3032,23 +3053,24 @@ static void gen_isync(DisasContext *ctx) #define MEMOP_GET_SIZE(x) (1 << ((x) & MO_SIZE)) -#define LARX(name, memop) \ -static void gen_##name(DisasContext *ctx) \ -{ \ - TCGv t0; \ - TCGv gpr = cpu_gpr[rD(ctx->opcode)]; \ - int len = MEMOP_GET_SIZE(memop); \ - gen_set_access_type(ctx, ACCESS_RES); \ - t0 = tcg_temp_local_new(); \ - gen_addr_reg_index(ctx, t0); \ - if ((len) > 1) { \ - gen_check_align(ctx, t0, (len)-1); \ - } \ - tcg_gen_qemu_ld_tl(gpr, t0, ctx->mem_idx, memop); \ - tcg_gen_mov_tl(cpu_reserve, t0); \ - tcg_gen_mov_tl(cpu_reserve_val, gpr); \ - tcg_gen_mb(TCG_MO_ALL | TCG_BAR_LDAQ); \ - tcg_temp_free(t0); \ +static void gen_load_locked(DisasContext *ctx, TCGMemOp memop) +{ + TCGv gpr = cpu_gpr[rD(ctx->opcode)]; + TCGv t0 = tcg_temp_new(); + + gen_set_access_type(ctx, ACCESS_RES); + gen_addr_reg_index(ctx, t0); + tcg_gen_qemu_ld_tl(gpr, t0, ctx->mem_idx, memop | MO_ALIGN); + tcg_gen_mov_tl(cpu_reserve, t0); + tcg_gen_mov_tl(cpu_reserve_val, gpr); + tcg_gen_mb(TCG_MO_ALL | TCG_BAR_LDAQ); + tcg_temp_free(t0); +} + +#define LARX(name, memop) \ +static void gen_##name(DisasContext *ctx) \ +{ \ + gen_load_locked(ctx, memop); \ } /* lwarx */ @@ -3056,134 +3078,239 @@ LARX(lbarx, DEF_MEMOP(MO_UB)) LARX(lharx, DEF_MEMOP(MO_UW)) LARX(lwarx, DEF_MEMOP(MO_UL)) -#define LD_ATOMIC(name, memop, tp, op, eop) \ -static void gen_##name(DisasContext *ctx) \ -{ \ - int len = MEMOP_GET_SIZE(memop); \ - uint32_t gpr_FC = FC(ctx->opcode); \ - TCGv EA = tcg_temp_local_new(); \ - TCGv_##tp t0, t1; \ - \ - gen_addr_register(ctx, EA); \ - if (len > 1) { \ - gen_check_align(ctx, EA, len - 1); \ - } \ - t0 = tcg_temp_new_##tp(); \ - t1 = tcg_temp_new_##tp(); \ - tcg_gen_##op(t0, cpu_gpr[rD(ctx->opcode) + 1]); \ - \ - switch (gpr_FC) { \ - case 0: /* Fetch and add */ \ - tcg_gen_atomic_fetch_add_##tp(t1, EA, t0, ctx->mem_idx, memop); \ - break; \ - case 1: /* Fetch and xor */ \ - tcg_gen_atomic_fetch_xor_##tp(t1, EA, t0, ctx->mem_idx, memop); \ - break; \ - case 2: /* Fetch and or */ \ - tcg_gen_atomic_fetch_or_##tp(t1, EA, t0, ctx->mem_idx, memop); \ - break; \ - case 3: /* Fetch and 'and' */ \ - tcg_gen_atomic_fetch_and_##tp(t1, EA, t0, ctx->mem_idx, memop); \ - break; \ - case 8: /* Swap */ \ - tcg_gen_atomic_xchg_##tp(t1, EA, t0, ctx->mem_idx, memop); \ - break; \ - case 4: /* Fetch and max unsigned */ \ - case 5: /* Fetch and max signed */ \ - case 6: /* Fetch and min unsigned */ \ - case 7: /* Fetch and min signed */ \ - case 16: /* compare and swap not equal */ \ - case 24: /* Fetch and increment bounded */ \ - case 25: /* Fetch and increment equal */ \ - case 28: /* Fetch and decrement bounded */ \ - gen_invalid(ctx); \ - break; \ - default: \ - /* invoke data storage error handler */ \ - gen_exception_err(ctx, POWERPC_EXCP_DSI, POWERPC_EXCP_INVAL); \ - } \ - tcg_gen_##eop(cpu_gpr[rD(ctx->opcode)], t1); \ - tcg_temp_free_##tp(t0); \ - tcg_temp_free_##tp(t1); \ - tcg_temp_free(EA); \ -} - -LD_ATOMIC(lwat, DEF_MEMOP(MO_UL), i32, trunc_tl_i32, extu_i32_tl) -#if defined(TARGET_PPC64) -LD_ATOMIC(ldat, DEF_MEMOP(MO_Q), i64, mov_i64, mov_i64) -#endif +static void gen_fetch_inc_conditional(DisasContext *ctx, TCGMemOp memop, + TCGv EA, TCGCond cond, int addend) +{ + TCGv t = tcg_temp_new(); + TCGv t2 = tcg_temp_new(); + TCGv u = tcg_temp_new(); -#define ST_ATOMIC(name, memop, tp, op) \ -static void gen_##name(DisasContext *ctx) \ -{ \ - int len = MEMOP_GET_SIZE(memop); \ - uint32_t gpr_FC = FC(ctx->opcode); \ - TCGv EA = tcg_temp_local_new(); \ - TCGv_##tp t0, t1; \ - \ - gen_addr_register(ctx, EA); \ - if (len > 1) { \ - gen_check_align(ctx, EA, len - 1); \ - } \ - t0 = tcg_temp_new_##tp(); \ - t1 = tcg_temp_new_##tp(); \ - tcg_gen_##op(t0, cpu_gpr[rD(ctx->opcode) + 1]); \ - \ - switch (gpr_FC) { \ - case 0: /* add and Store */ \ - tcg_gen_atomic_add_fetch_##tp(t1, EA, t0, ctx->mem_idx, memop); \ - break; \ - case 1: /* xor and Store */ \ - tcg_gen_atomic_xor_fetch_##tp(t1, EA, t0, ctx->mem_idx, memop); \ - break; \ - case 2: /* Or and Store */ \ - tcg_gen_atomic_or_fetch_##tp(t1, EA, t0, ctx->mem_idx, memop); \ - break; \ - case 3: /* 'and' and Store */ \ - tcg_gen_atomic_and_fetch_##tp(t1, EA, t0, ctx->mem_idx, memop); \ - break; \ - case 4: /* Store max unsigned */ \ - case 5: /* Store max signed */ \ - case 6: /* Store min unsigned */ \ - case 7: /* Store min signed */ \ - case 24: /* Store twin */ \ - gen_invalid(ctx); \ - break; \ - default: \ - /* invoke data storage error handler */ \ - gen_exception_err(ctx, POWERPC_EXCP_DSI, POWERPC_EXCP_INVAL); \ - } \ - tcg_temp_free_##tp(t0); \ - tcg_temp_free_##tp(t1); \ - tcg_temp_free(EA); \ -} - -ST_ATOMIC(stwat, DEF_MEMOP(MO_UL), i32, trunc_tl_i32) -#if defined(TARGET_PPC64) -ST_ATOMIC(stdat, DEF_MEMOP(MO_Q), i64, mov_i64) + tcg_gen_qemu_ld_tl(t, EA, ctx->mem_idx, memop); + tcg_gen_addi_tl(t2, EA, MEMOP_GET_SIZE(memop)); + tcg_gen_qemu_ld_tl(t2, t2, ctx->mem_idx, memop); + tcg_gen_addi_tl(u, t, addend); + + /* E.g. for fetch and increment bounded... */ + /* mem(EA,s) = (t != t2 ? u = t + 1 : t) */ + tcg_gen_movcond_tl(cond, u, t, t2, u, t); + tcg_gen_qemu_st_tl(u, EA, ctx->mem_idx, memop); + + /* RT = (t != t2 ? t : u = 1<<(s*8-1)) */ + tcg_gen_movi_tl(u, 1 << (MEMOP_GET_SIZE(memop) * 8 - 1)); + tcg_gen_movcond_tl(cond, cpu_gpr[rD(ctx->opcode)], t, t2, t, u); + + tcg_temp_free(t); + tcg_temp_free(t2); + tcg_temp_free(u); +} + +static void gen_ld_atomic(DisasContext *ctx, TCGMemOp memop) +{ + uint32_t gpr_FC = FC(ctx->opcode); + TCGv EA = tcg_temp_new(); + int rt = rD(ctx->opcode); + bool need_serial; + TCGv src, dst; + + gen_addr_register(ctx, EA); + dst = cpu_gpr[rt]; + src = cpu_gpr[(rt + 1) & 31]; + + need_serial = false; + memop |= MO_ALIGN; + switch (gpr_FC) { + case 0: /* Fetch and add */ + tcg_gen_atomic_fetch_add_tl(dst, EA, src, ctx->mem_idx, memop); + break; + case 1: /* Fetch and xor */ + tcg_gen_atomic_fetch_xor_tl(dst, EA, src, ctx->mem_idx, memop); + break; + case 2: /* Fetch and or */ + tcg_gen_atomic_fetch_or_tl(dst, EA, src, ctx->mem_idx, memop); + break; + case 3: /* Fetch and 'and' */ + tcg_gen_atomic_fetch_and_tl(dst, EA, src, ctx->mem_idx, memop); + break; + case 4: /* Fetch and max unsigned */ + tcg_gen_atomic_fetch_umax_tl(dst, EA, src, ctx->mem_idx, memop); + break; + case 5: /* Fetch and max signed */ + tcg_gen_atomic_fetch_smax_tl(dst, EA, src, ctx->mem_idx, memop); + break; + case 6: /* Fetch and min unsigned */ + tcg_gen_atomic_fetch_umin_tl(dst, EA, src, ctx->mem_idx, memop); + break; + case 7: /* Fetch and min signed */ + tcg_gen_atomic_fetch_smin_tl(dst, EA, src, ctx->mem_idx, memop); + break; + case 8: /* Swap */ + tcg_gen_atomic_xchg_tl(dst, EA, src, ctx->mem_idx, memop); + break; + + case 16: /* Compare and swap not equal */ + if (tb_cflags(ctx->base.tb) & CF_PARALLEL) { + need_serial = true; + } else { + TCGv t0 = tcg_temp_new(); + TCGv t1 = tcg_temp_new(); + + tcg_gen_qemu_ld_tl(t0, EA, ctx->mem_idx, memop); + if ((memop & MO_SIZE) == MO_64 || TARGET_LONG_BITS == 32) { + tcg_gen_mov_tl(t1, src); + } else { + tcg_gen_ext32u_tl(t1, src); + } + tcg_gen_movcond_tl(TCG_COND_NE, t1, t0, t1, + cpu_gpr[(rt + 2) & 31], t0); + tcg_gen_qemu_st_tl(t1, EA, ctx->mem_idx, memop); + tcg_gen_mov_tl(dst, t0); + + tcg_temp_free(t0); + tcg_temp_free(t1); + } + break; + + case 24: /* Fetch and increment bounded */ + if (tb_cflags(ctx->base.tb) & CF_PARALLEL) { + need_serial = true; + } else { + gen_fetch_inc_conditional(ctx, memop, EA, TCG_COND_NE, 1); + } + break; + case 25: /* Fetch and increment equal */ + if (tb_cflags(ctx->base.tb) & CF_PARALLEL) { + need_serial = true; + } else { + gen_fetch_inc_conditional(ctx, memop, EA, TCG_COND_EQ, 1); + } + break; + case 28: /* Fetch and decrement bounded */ + if (tb_cflags(ctx->base.tb) & CF_PARALLEL) { + need_serial = true; + } else { + gen_fetch_inc_conditional(ctx, memop, EA, TCG_COND_NE, -1); + } + break; + + default: + /* invoke data storage error handler */ + gen_exception_err(ctx, POWERPC_EXCP_DSI, POWERPC_EXCP_INVAL); + } + tcg_temp_free(EA); + + if (need_serial) { + /* Restart with exclusive lock. */ + gen_helper_exit_atomic(cpu_env); + ctx->base.is_jmp = DISAS_NORETURN; + } +} + +static void gen_lwat(DisasContext *ctx) +{ + gen_ld_atomic(ctx, DEF_MEMOP(MO_UL)); +} + +#ifdef TARGET_PPC64 +static void gen_ldat(DisasContext *ctx) +{ + gen_ld_atomic(ctx, DEF_MEMOP(MO_Q)); +} #endif -#if defined(CONFIG_USER_ONLY) -static void gen_conditional_store(DisasContext *ctx, TCGv EA, - int reg, int memop) +static void gen_st_atomic(DisasContext *ctx, TCGMemOp memop) { - TCGv t0 = tcg_temp_new(); + uint32_t gpr_FC = FC(ctx->opcode); + TCGv EA = tcg_temp_new(); + TCGv src, discard; - tcg_gen_st_tl(EA, cpu_env, offsetof(CPUPPCState, reserve_ea)); - tcg_gen_movi_tl(t0, (MEMOP_GET_SIZE(memop) << 5) | reg); - tcg_gen_st_tl(t0, cpu_env, offsetof(CPUPPCState, reserve_info)); - tcg_temp_free(t0); - gen_exception_err(ctx, POWERPC_EXCP_STCX, 0); + gen_addr_register(ctx, EA); + src = cpu_gpr[rD(ctx->opcode)]; + discard = tcg_temp_new(); + + memop |= MO_ALIGN; + switch (gpr_FC) { + case 0: /* add and Store */ + tcg_gen_atomic_add_fetch_tl(discard, EA, src, ctx->mem_idx, memop); + break; + case 1: /* xor and Store */ + tcg_gen_atomic_xor_fetch_tl(discard, EA, src, ctx->mem_idx, memop); + break; + case 2: /* Or and Store */ + tcg_gen_atomic_or_fetch_tl(discard, EA, src, ctx->mem_idx, memop); + break; + case 3: /* 'and' and Store */ + tcg_gen_atomic_and_fetch_tl(discard, EA, src, ctx->mem_idx, memop); + break; + case 4: /* Store max unsigned */ + tcg_gen_atomic_umax_fetch_tl(discard, EA, src, ctx->mem_idx, memop); + break; + case 5: /* Store max signed */ + tcg_gen_atomic_smax_fetch_tl(discard, EA, src, ctx->mem_idx, memop); + break; + case 6: /* Store min unsigned */ + tcg_gen_atomic_umin_fetch_tl(discard, EA, src, ctx->mem_idx, memop); + break; + case 7: /* Store min signed */ + tcg_gen_atomic_smin_fetch_tl(discard, EA, src, ctx->mem_idx, memop); + break; + case 24: /* Store twin */ + if (tb_cflags(ctx->base.tb) & CF_PARALLEL) { + /* Restart with exclusive lock. */ + gen_helper_exit_atomic(cpu_env); + ctx->base.is_jmp = DISAS_NORETURN; + } else { + TCGv t = tcg_temp_new(); + TCGv t2 = tcg_temp_new(); + TCGv s = tcg_temp_new(); + TCGv s2 = tcg_temp_new(); + TCGv ea_plus_s = tcg_temp_new(); + + tcg_gen_qemu_ld_tl(t, EA, ctx->mem_idx, memop); + tcg_gen_addi_tl(ea_plus_s, EA, MEMOP_GET_SIZE(memop)); + tcg_gen_qemu_ld_tl(t2, ea_plus_s, ctx->mem_idx, memop); + tcg_gen_movcond_tl(TCG_COND_EQ, s, t, t2, src, t); + tcg_gen_movcond_tl(TCG_COND_EQ, s2, t, t2, src, t2); + tcg_gen_qemu_st_tl(s, EA, ctx->mem_idx, memop); + tcg_gen_qemu_st_tl(s2, ea_plus_s, ctx->mem_idx, memop); + + tcg_temp_free(ea_plus_s); + tcg_temp_free(s2); + tcg_temp_free(s); + tcg_temp_free(t2); + tcg_temp_free(t); + } + break; + default: + /* invoke data storage error handler */ + gen_exception_err(ctx, POWERPC_EXCP_DSI, POWERPC_EXCP_INVAL); + } + tcg_temp_free(discard); + tcg_temp_free(EA); } -#else -static void gen_conditional_store(DisasContext *ctx, TCGv EA, - int reg, int memop) + +static void gen_stwat(DisasContext *ctx) +{ + gen_st_atomic(ctx, DEF_MEMOP(MO_UL)); +} + +#ifdef TARGET_PPC64 +static void gen_stdat(DisasContext *ctx) +{ + gen_st_atomic(ctx, DEF_MEMOP(MO_Q)); +} +#endif + +static void gen_conditional_store(DisasContext *ctx, TCGMemOp memop) { TCGLabel *l1 = gen_new_label(); TCGLabel *l2 = gen_new_label(); - TCGv t0; + TCGv t0 = tcg_temp_new(); + int reg = rS(ctx->opcode); - tcg_gen_brcond_tl(TCG_COND_NE, EA, cpu_reserve, l1); + gen_set_access_type(ctx, ACCESS_RES); + gen_addr_reg_index(ctx, t0); + tcg_gen_brcond_tl(TCG_COND_NE, t0, cpu_reserve, l1); + tcg_temp_free(t0); t0 = tcg_temp_new(); tcg_gen_atomic_cmpxchg_tl(t0, cpu_reserve, cpu_reserve_val, @@ -3206,21 +3333,11 @@ static void gen_conditional_store(DisasContext *ctx, TCGv EA, gen_set_label(l2); tcg_gen_movi_tl(cpu_reserve, -1); } -#endif -#define STCX(name, memop) \ -static void gen_##name(DisasContext *ctx) \ -{ \ - TCGv t0; \ - int len = MEMOP_GET_SIZE(memop); \ - gen_set_access_type(ctx, ACCESS_RES); \ - t0 = tcg_temp_local_new(); \ - gen_addr_reg_index(ctx, t0); \ - if (len > 1) { \ - gen_check_align(ctx, t0, (len) - 1); \ - } \ - gen_conditional_store(ctx, t0, rS(ctx->opcode), memop); \ - tcg_temp_free(t0); \ +#define STCX(name, memop) \ +static void gen_##name(DisasContext *ctx) \ +{ \ + gen_conditional_store(ctx, memop); \ } STCX(stbcx_, DEF_MEMOP(MO_UB)) @@ -3236,9 +3353,8 @@ STCX(stdcx_, DEF_MEMOP(MO_Q)) /* lqarx */ static void gen_lqarx(DisasContext *ctx) { - TCGv EA; int rd = rD(ctx->opcode); - TCGv gpr1, gpr2; + TCGv EA, hi, lo; if (unlikely((rd & 1) || (rd == rA(ctx->opcode)) || (rd == rB(ctx->opcode)))) { @@ -3247,73 +3363,125 @@ static void gen_lqarx(DisasContext *ctx) } gen_set_access_type(ctx, ACCESS_RES); - EA = tcg_temp_local_new(); + EA = tcg_temp_new(); gen_addr_reg_index(ctx, EA); - gen_check_align(ctx, EA, 15); - if (unlikely(ctx->le_mode)) { - gpr1 = cpu_gpr[rd+1]; - gpr2 = cpu_gpr[rd]; + + /* Note that the low part is always in RD+1, even in LE mode. */ + lo = cpu_gpr[rd + 1]; + hi = cpu_gpr[rd]; + + if (tb_cflags(ctx->base.tb) & CF_PARALLEL) { +#ifdef CONFIG_ATOMIC128 + TCGv_i32 oi = tcg_temp_new_i32(); + if (ctx->le_mode) { + tcg_gen_movi_i32(oi, make_memop_idx(MO_LEQ | MO_ALIGN_16, + ctx->mem_idx)); + gen_helper_lq_le_parallel(lo, cpu_env, EA, oi); + } else { + tcg_gen_movi_i32(oi, make_memop_idx(MO_BEQ | MO_ALIGN_16, + ctx->mem_idx)); + gen_helper_lq_be_parallel(lo, cpu_env, EA, oi); + } + tcg_temp_free_i32(oi); + tcg_gen_ld_i64(hi, cpu_env, offsetof(CPUPPCState, retxh)); +#else + /* Restart with exclusive lock. */ + gen_helper_exit_atomic(cpu_env); + ctx->base.is_jmp = DISAS_NORETURN; + tcg_temp_free(EA); + return; +#endif + } else if (ctx->le_mode) { + tcg_gen_qemu_ld_i64(lo, EA, ctx->mem_idx, MO_LEQ | MO_ALIGN_16); + tcg_gen_mov_tl(cpu_reserve, EA); + gen_addr_add(ctx, EA, EA, 8); + tcg_gen_qemu_ld_i64(hi, EA, ctx->mem_idx, MO_LEQ); } else { - gpr1 = cpu_gpr[rd]; - gpr2 = cpu_gpr[rd+1]; + tcg_gen_qemu_ld_i64(hi, EA, ctx->mem_idx, MO_BEQ | MO_ALIGN_16); + tcg_gen_mov_tl(cpu_reserve, EA); + gen_addr_add(ctx, EA, EA, 8); + tcg_gen_qemu_ld_i64(lo, EA, ctx->mem_idx, MO_BEQ); } - tcg_gen_qemu_ld_i64(gpr1, EA, ctx->mem_idx, DEF_MEMOP(MO_Q)); - tcg_gen_mov_tl(cpu_reserve, EA); - gen_addr_add(ctx, EA, EA, 8); - tcg_gen_qemu_ld_i64(gpr2, EA, ctx->mem_idx, DEF_MEMOP(MO_Q)); - - tcg_gen_st_tl(gpr1, cpu_env, offsetof(CPUPPCState, reserve_val)); - tcg_gen_st_tl(gpr2, cpu_env, offsetof(CPUPPCState, reserve_val2)); tcg_temp_free(EA); + + tcg_gen_st_tl(hi, cpu_env, offsetof(CPUPPCState, reserve_val)); + tcg_gen_st_tl(lo, cpu_env, offsetof(CPUPPCState, reserve_val2)); } /* stqcx. */ static void gen_stqcx_(DisasContext *ctx) { - TCGv EA; - int reg = rS(ctx->opcode); - int len = 16; -#if !defined(CONFIG_USER_ONLY) - TCGLabel *l1; - TCGv gpr1, gpr2; -#endif + int rs = rS(ctx->opcode); + TCGv EA, hi, lo; - if (unlikely((rD(ctx->opcode) & 1))) { + if (unlikely(rs & 1)) { gen_inval_exception(ctx, POWERPC_EXCP_INVAL_INVAL); return; } + gen_set_access_type(ctx, ACCESS_RES); - EA = tcg_temp_local_new(); + EA = tcg_temp_new(); gen_addr_reg_index(ctx, EA); - if (len > 1) { - gen_check_align(ctx, EA, (len) - 1); - } -#if defined(CONFIG_USER_ONLY) - gen_conditional_store(ctx, EA, reg, 16); -#else - tcg_gen_trunc_tl_i32(cpu_crf[0], cpu_so); - l1 = gen_new_label(); - tcg_gen_brcond_tl(TCG_COND_NE, EA, cpu_reserve, l1); - tcg_gen_ori_i32(cpu_crf[0], cpu_crf[0], CRF_EQ); + /* Note that the low part is always in RS+1, even in LE mode. */ + lo = cpu_gpr[rs + 1]; + hi = cpu_gpr[rs]; - if (unlikely(ctx->le_mode)) { - gpr1 = cpu_gpr[reg + 1]; - gpr2 = cpu_gpr[reg]; + if (tb_cflags(ctx->base.tb) & CF_PARALLEL) { + TCGv_i32 oi = tcg_const_i32(DEF_MEMOP(MO_Q) | MO_ALIGN_16); +#ifdef CONFIG_ATOMIC128 + if (ctx->le_mode) { + gen_helper_stqcx_le_parallel(cpu_crf[0], cpu_env, EA, lo, hi, oi); + } else { + gen_helper_stqcx_le_parallel(cpu_crf[0], cpu_env, EA, lo, hi, oi); + } +#else + /* Restart with exclusive lock. */ + gen_helper_exit_atomic(cpu_env); + ctx->base.is_jmp = DISAS_NORETURN; +#endif + tcg_temp_free(EA); + tcg_temp_free_i32(oi); } else { - gpr1 = cpu_gpr[reg]; - gpr2 = cpu_gpr[reg + 1]; - } - tcg_gen_qemu_st_tl(gpr1, EA, ctx->mem_idx, DEF_MEMOP(MO_Q)); - gen_addr_add(ctx, EA, EA, 8); - tcg_gen_qemu_st_tl(gpr2, EA, ctx->mem_idx, DEF_MEMOP(MO_Q)); + TCGLabel *lab_fail = gen_new_label(); + TCGLabel *lab_over = gen_new_label(); + TCGv_i64 t0 = tcg_temp_new_i64(); + TCGv_i64 t1 = tcg_temp_new_i64(); - gen_set_label(l1); - tcg_gen_movi_tl(cpu_reserve, -1); -#endif - tcg_temp_free(EA); -} + tcg_gen_brcond_tl(TCG_COND_NE, EA, cpu_reserve, lab_fail); + tcg_temp_free(EA); + gen_qemu_ld64_i64(ctx, t0, cpu_reserve); + tcg_gen_ld_i64(t1, cpu_env, (ctx->le_mode + ? offsetof(CPUPPCState, reserve_val2) + : offsetof(CPUPPCState, reserve_val))); + tcg_gen_brcond_i64(TCG_COND_NE, t0, t1, lab_fail); + + tcg_gen_addi_i64(t0, cpu_reserve, 8); + gen_qemu_ld64_i64(ctx, t0, t0); + tcg_gen_ld_i64(t1, cpu_env, (ctx->le_mode + ? offsetof(CPUPPCState, reserve_val) + : offsetof(CPUPPCState, reserve_val2))); + tcg_gen_brcond_i64(TCG_COND_NE, t0, t1, lab_fail); + + /* Success */ + gen_qemu_st64_i64(ctx, ctx->le_mode ? lo : hi, cpu_reserve); + tcg_gen_addi_i64(t0, cpu_reserve, 8); + gen_qemu_st64_i64(ctx, ctx->le_mode ? hi : lo, t0); + + tcg_gen_trunc_tl_i32(cpu_crf[0], cpu_so); + tcg_gen_ori_i32(cpu_crf[0], cpu_crf[0], CRF_EQ); + tcg_gen_br(lab_over); + + gen_set_label(lab_fail); + tcg_gen_trunc_tl_i32(cpu_crf[0], cpu_so); + + gen_set_label(lab_over); + tcg_gen_movi_tl(cpu_reserve, -1); + tcg_temp_free_i64(t0); + tcg_temp_free_i64(t1); + } +} #endif /* defined(TARGET_PPC64) */ /* sync */ @@ -4636,8 +4804,8 @@ static void gen_eciwx(DisasContext *ctx) gen_set_access_type(ctx, ACCESS_EXT); t0 = tcg_temp_new(); gen_addr_reg_index(ctx, t0); - gen_check_align(ctx, t0, 0x03); - gen_qemu_ld32u(ctx, cpu_gpr[rD(ctx->opcode)], t0); + tcg_gen_qemu_ld_tl(cpu_gpr[rD(ctx->opcode)], t0, ctx->mem_idx, + DEF_MEMOP(MO_UL | MO_ALIGN)); tcg_temp_free(t0); } @@ -4649,8 +4817,8 @@ static void gen_ecowx(DisasContext *ctx) gen_set_access_type(ctx, ACCESS_EXT); t0 = tcg_temp_new(); gen_addr_reg_index(ctx, t0); - gen_check_align(ctx, t0, 0x03); - gen_qemu_st32(ctx, cpu_gpr[rD(ctx->opcode)], t0); + tcg_gen_qemu_st_tl(cpu_gpr[rD(ctx->opcode)], t0, ctx->mem_idx, + DEF_MEMOP(MO_UL | MO_ALIGN)); tcg_temp_free(t0); } @@ -6886,7 +7054,7 @@ GEN_HANDLER(stop##u, opc, 0xFF, 0xFF, 0x00000000, type), #define GEN_STUX(name, stop, opc2, opc3, type) \ GEN_HANDLER(name##ux, 0x1F, opc2, opc3, 0x00000001, type), #define GEN_STX_E(name, stop, opc2, opc3, type, type2, chk) \ -GEN_HANDLER_E(name##x, 0x1F, opc2, opc3, 0x00000001, type, type2), +GEN_HANDLER_E(name##x, 0x1F, opc2, opc3, 0x00000000, type, type2), #define GEN_STS(name, stop, op, type) \ GEN_ST(name, stop, op | 0x20, type) \ GEN_STU(name, stop, op | 0x21, type) \ @@ -7314,6 +7482,7 @@ static bool ppc_tr_breakpoint_check(DisasContextBase *dcbase, CPUState *cs, DisasContext *ctx = container_of(dcbase, DisasContext, base); gen_debug_exception(ctx); + dcbase->is_jmp = DISAS_NORETURN; /* The address covered by the breakpoint must be included in [tb->pc, tb->pc + tb->size) in order to for it to be properly cleared -- thus we increment the PC here so that diff --git a/target/ppc/translate_init.inc.c b/target/ppc/translate_init.inc.c index 76d6f3fd5e..7813b1b004 100644 --- a/target/ppc/translate_init.inc.c +++ b/target/ppc/translate_init.inc.c @@ -10457,6 +10457,7 @@ static void ppc_cpu_class_init(ObjectClass *oc, void *data) cc->set_pc = ppc_cpu_set_pc; cc->gdb_read_register = ppc_cpu_gdb_read_register; cc->gdb_write_register = ppc_cpu_gdb_write_register; + cc->do_unaligned_access = ppc_cpu_do_unaligned_access; #ifdef CONFIG_USER_ONLY cc->handle_mmu_fault = ppc_cpu_handle_mmu_fault; #else diff --git a/target/xtensa/helper.c b/target/xtensa/helper.c index 34844eead3..c9a6132700 100644 --- a/target/xtensa/helper.c +++ b/target/xtensa/helper.c @@ -26,6 +26,7 @@ */ #include "qemu/osdep.h" +#include "qemu/units.h" #include "cpu.h" #include "exec/exec-all.h" #include "exec/gdbstub.h" @@ -726,10 +727,10 @@ static void dump_tlb(FILE *f, fprintf_function cpu_fprintf, bool print_header = true; if (sz >= 0x100000) { - sz >>= 20; + sz /= MiB; sz_text = "MB"; } else { - sz >>= 10; + sz /= KiB; sz_text = "KB"; } |