From 54b8486f469475d6c8e8aec917b91239a54eb8c8 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Wed, 28 Apr 2010 19:15:44 +0300 Subject: [PATCH] KVM: x86 emulator: do not inject exception directly into vcpu Return exception as a result of instruction emulation and handle injection in KVM code. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_emulate.h | 6 ++ arch/x86/kvm/emulate.c | 124 ++++++++++++++++++----------- arch/x86/kvm/x86.c | 20 ++++- 3 files changed, 100 insertions(+), 50 deletions(-) diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index a87d95f09572..51cfd730ac5d 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -216,6 +216,12 @@ struct x86_emulate_ctxt { int interruptibility; bool restart; /* restart string instruction after writeback */ + + int exception; /* exception that happens during emulation or -1 */ + u32 error_code; /* error code for exception */ + bool error_code_valid; + unsigned long cr2; /* faulted address in case of #PF */ + /* decode cache */ struct decode_cache decode; }; diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index c40b40561dff..b43ac98ef790 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -653,6 +653,37 @@ static unsigned long ss_base(struct x86_emulate_ctxt *ctxt, return seg_base(ctxt, ops, VCPU_SREG_SS); } +static void emulate_exception(struct x86_emulate_ctxt *ctxt, int vec, + u32 error, bool valid) +{ + ctxt->exception = vec; + ctxt->error_code = error; + ctxt->error_code_valid = valid; + ctxt->restart = false; +} + +static void emulate_gp(struct x86_emulate_ctxt *ctxt, int err) +{ + emulate_exception(ctxt, GP_VECTOR, err, true); +} + +static void emulate_pf(struct x86_emulate_ctxt *ctxt, unsigned long addr, + int err) +{ + ctxt->cr2 = addr; + emulate_exception(ctxt, PF_VECTOR, err, true); +} + +static void emulate_ud(struct x86_emulate_ctxt *ctxt) +{ + emulate_exception(ctxt, UD_VECTOR, 0, false); +} + +static void emulate_ts(struct x86_emulate_ctxt *ctxt, int err) +{ + emulate_exception(ctxt, TS_VECTOR, err, true); +} + static int do_fetch_insn_byte(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops, unsigned long eip, u8 *dest) @@ -1285,7 +1316,7 @@ static int read_emulated(struct x86_emulate_ctxt *ctxt, rc = ops->read_emulated(addr, mc->data + mc->end, n, &err, ctxt->vcpu); if (rc == X86EMUL_PROPAGATE_FAULT) - kvm_inject_page_fault(ctxt->vcpu, addr, err); + emulate_pf(ctxt, addr, err); if (rc != X86EMUL_CONTINUE) return rc; mc->end += n; @@ -1366,13 +1397,13 @@ static int read_segment_descriptor(struct x86_emulate_ctxt *ctxt, get_descriptor_table_ptr(ctxt, ops, selector, &dt); if (dt.size < index * 8 + 7) { - kvm_inject_gp(ctxt->vcpu, selector & 0xfffc); + emulate_gp(ctxt, selector & 0xfffc); return X86EMUL_PROPAGATE_FAULT; } addr = dt.address + index * 8; ret = ops->read_std(addr, desc, sizeof *desc, ctxt->vcpu, &err); if (ret == X86EMUL_PROPAGATE_FAULT) - kvm_inject_page_fault(ctxt->vcpu, addr, err); + emulate_pf(ctxt, addr, err); return ret; } @@ -1391,14 +1422,14 @@ static int write_segment_descriptor(struct x86_emulate_ctxt *ctxt, get_descriptor_table_ptr(ctxt, ops, selector, &dt); if (dt.size < index * 8 + 7) { - kvm_inject_gp(ctxt->vcpu, selector & 0xfffc); + emulate_gp(ctxt, selector & 0xfffc); return X86EMUL_PROPAGATE_FAULT; } addr = dt.address + index * 8; ret = ops->write_std(addr, desc, sizeof *desc, ctxt->vcpu, &err); if (ret == X86EMUL_PROPAGATE_FAULT) - kvm_inject_page_fault(ctxt->vcpu, addr, err); + emulate_pf(ctxt, addr, err); return ret; } @@ -1517,7 +1548,7 @@ load: ops->set_cached_descriptor(&seg_desc, seg, ctxt->vcpu); return X86EMUL_CONTINUE; exception: - kvm_queue_exception_e(ctxt->vcpu, err_vec, err_code); + emulate_exception(ctxt, err_vec, err_code, true); return X86EMUL_PROPAGATE_FAULT; } @@ -1578,7 +1609,7 @@ static int emulate_popf(struct x86_emulate_ctxt *ctxt, break; case X86EMUL_MODE_VM86: if (iopl < 3) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } change_mask |= EFLG_IF; @@ -1829,7 +1860,7 @@ static inline int writeback(struct x86_emulate_ctxt *ctxt, &err, ctxt->vcpu); if (rc == X86EMUL_PROPAGATE_FAULT) - kvm_inject_page_fault(ctxt->vcpu, + emulate_pf(ctxt, (unsigned long)c->dst.ptr, err); if (rc != X86EMUL_CONTINUE) return rc; @@ -1883,7 +1914,7 @@ emulate_syscall(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) /* syscall is not available in real mode */ if (ctxt->mode == X86EMUL_MODE_REAL || ctxt->mode == X86EMUL_MODE_VM86) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); return X86EMUL_PROPAGATE_FAULT; } @@ -1937,7 +1968,7 @@ emulate_sysenter(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) /* inject #GP if in real mode */ if (ctxt->mode == X86EMUL_MODE_REAL) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } @@ -1945,7 +1976,7 @@ emulate_sysenter(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) * Therefore, we inject an #UD. */ if (ctxt->mode == X86EMUL_MODE_PROT64) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); return X86EMUL_PROPAGATE_FAULT; } @@ -1955,13 +1986,13 @@ emulate_sysenter(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) switch (ctxt->mode) { case X86EMUL_MODE_PROT32: if ((msr_data & 0xfffc) == 0x0) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } break; case X86EMUL_MODE_PROT64: if (msr_data == 0x0) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } break; @@ -2004,7 +2035,7 @@ emulate_sysexit(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) /* inject #GP if in real mode or Virtual 8086 mode */ if (ctxt->mode == X86EMUL_MODE_REAL || ctxt->mode == X86EMUL_MODE_VM86) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } @@ -2022,7 +2053,7 @@ emulate_sysexit(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) case X86EMUL_MODE_PROT32: cs_sel = (u16)(msr_data + 16); if ((msr_data & 0xfffc) == 0x0) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } ss_sel = (u16)(msr_data + 24); @@ -2030,7 +2061,7 @@ emulate_sysexit(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) case X86EMUL_MODE_PROT64: cs_sel = (u16)(msr_data + 32); if (msr_data == 0x0) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } ss_sel = cs_sel + 8; @@ -2192,7 +2223,7 @@ static int task_switch_16(struct x86_emulate_ctxt *ctxt, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, old_tss_base, err); + emulate_pf(ctxt, old_tss_base, err); return ret; } @@ -2202,7 +2233,7 @@ static int task_switch_16(struct x86_emulate_ctxt *ctxt, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, old_tss_base, err); + emulate_pf(ctxt, old_tss_base, err); return ret; } @@ -2210,7 +2241,7 @@ static int task_switch_16(struct x86_emulate_ctxt *ctxt, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, new_tss_base, err); + emulate_pf(ctxt, new_tss_base, err); return ret; } @@ -2223,7 +2254,7 @@ static int task_switch_16(struct x86_emulate_ctxt *ctxt, ctxt->vcpu, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, new_tss_base, err); + emulate_pf(ctxt, new_tss_base, err); return ret; } } @@ -2266,7 +2297,7 @@ static int load_state_from_tss32(struct x86_emulate_ctxt *ctxt, int ret; if (ops->set_cr(3, tss->cr3, ctxt->vcpu)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } c->eip = tss->eip; @@ -2334,7 +2365,7 @@ static int task_switch_32(struct x86_emulate_ctxt *ctxt, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, old_tss_base, err); + emulate_pf(ctxt, old_tss_base, err); return ret; } @@ -2344,7 +2375,7 @@ static int task_switch_32(struct x86_emulate_ctxt *ctxt, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, old_tss_base, err); + emulate_pf(ctxt, old_tss_base, err); return ret; } @@ -2352,7 +2383,7 @@ static int task_switch_32(struct x86_emulate_ctxt *ctxt, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, new_tss_base, err); + emulate_pf(ctxt, new_tss_base, err); return ret; } @@ -2365,7 +2396,7 @@ static int task_switch_32(struct x86_emulate_ctxt *ctxt, ctxt->vcpu, &err); if (ret == X86EMUL_PROPAGATE_FAULT) { /* FIXME: need to provide precise fault address */ - kvm_inject_page_fault(ctxt->vcpu, new_tss_base, err); + emulate_pf(ctxt, new_tss_base, err); return ret; } } @@ -2399,7 +2430,7 @@ static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt, if (reason != TASK_SWITCH_IRET) { if ((tss_selector & 3) > next_tss_desc.dpl || ops->cpl(ctxt->vcpu) > next_tss_desc.dpl) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); return X86EMUL_PROPAGATE_FAULT; } } @@ -2408,8 +2439,7 @@ static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt, if (!next_tss_desc.p || ((desc_limit < 0x67 && (next_tss_desc.type & 8)) || desc_limit < 0x2b)) { - kvm_queue_exception_e(ctxt->vcpu, TS_VECTOR, - tss_selector & 0xfffc); + emulate_ts(ctxt, tss_selector & 0xfffc); return X86EMUL_PROPAGATE_FAULT; } @@ -2505,19 +2535,19 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) ctxt->decode.mem_read.pos = 0; if (ctxt->mode == X86EMUL_MODE_PROT64 && (c->d & No64)) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; } /* LOCK prefix is allowed only with some instructions */ if (c->lock_prefix && (!(c->d & Lock) || c->dst.type != OP_MEM)) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; } /* Privileged instruction can be executed only in CPL=0 */ if ((c->d & Priv) && ops->cpl(ctxt->vcpu)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } @@ -2679,7 +2709,7 @@ special_insn: c->dst.bytes = min(c->dst.bytes, 4u); if (!emulator_io_permited(ctxt, ops, c->regs[VCPU_REGS_RDX], c->dst.bytes)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } if (!pio_in_emulated(ctxt, ops, c->dst.bytes, @@ -2691,7 +2721,7 @@ special_insn: c->src.bytes = min(c->src.bytes, 4u); if (!emulator_io_permited(ctxt, ops, c->regs[VCPU_REGS_RDX], c->src.bytes)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } ops->pio_out_emulated(c->src.bytes, c->regs[VCPU_REGS_RDX], @@ -2754,7 +2784,7 @@ special_insn: goto mov; case 0x8c: /* mov r/m, sreg */ if (c->modrm_reg > VCPU_SREG_GS) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; } c->dst.val = ops->get_segment_selector(c->modrm_reg, ctxt->vcpu); @@ -2769,7 +2799,7 @@ special_insn: if (c->modrm_reg == VCPU_SREG_CS || c->modrm_reg > VCPU_SREG_GS) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; } @@ -2895,7 +2925,7 @@ special_insn: do_io_in: c->dst.bytes = min(c->dst.bytes, 4u); if (!emulator_io_permited(ctxt, ops, c->src.val, c->dst.bytes)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } if (!pio_in_emulated(ctxt, ops, c->dst.bytes, c->src.val, @@ -2908,7 +2938,7 @@ special_insn: do_io_out: c->dst.bytes = min(c->dst.bytes, 4u); if (!emulator_io_permited(ctxt, ops, c->src.val, c->dst.bytes)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } ops->pio_out_emulated(c->dst.bytes, c->src.val, &c->dst.val, 1, @@ -2933,7 +2963,7 @@ special_insn: break; case 0xfa: /* cli */ if (emulator_bad_iopl(ctxt, ops)) - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); else { ctxt->eflags &= ~X86_EFLAGS_IF; c->dst.type = OP_NONE; /* Disable writeback. */ @@ -2941,7 +2971,7 @@ special_insn: break; case 0xfb: /* sti */ if (emulator_bad_iopl(ctxt, ops)) - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); else { ctxt->interruptibility = KVM_X86_SHADOW_INT_STI; ctxt->eflags |= X86_EFLAGS_IF; @@ -3069,7 +3099,7 @@ twobyte_insn: c->dst.type = OP_NONE; break; case 5: /* not defined */ - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; case 7: /* invlpg*/ emulate_invlpg(ctxt->vcpu, c->modrm_ea); @@ -3102,7 +3132,7 @@ twobyte_insn: case 1: case 5 ... 7: case 9 ... 15: - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; } c->regs[c->modrm_rm] = ops->get_cr(c->modrm_reg, ctxt->vcpu); @@ -3111,7 +3141,7 @@ twobyte_insn: case 0x21: /* mov from dr to reg */ if ((ops->get_cr(4, ctxt->vcpu) & X86_CR4_DE) && (c->modrm_reg == 4 || c->modrm_reg == 5)) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; } ops->get_dr(c->modrm_reg, &c->regs[c->modrm_rm], ctxt->vcpu); @@ -3119,7 +3149,7 @@ twobyte_insn: break; case 0x22: /* mov reg, cr */ if (ops->set_cr(c->modrm_reg, c->modrm_val, ctxt->vcpu)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } c->dst.type = OP_NONE; @@ -3127,7 +3157,7 @@ twobyte_insn: case 0x23: /* mov from reg to dr */ if ((ops->get_cr(4, ctxt->vcpu) & X86_CR4_DE) && (c->modrm_reg == 4 || c->modrm_reg == 5)) { - kvm_queue_exception(ctxt->vcpu, UD_VECTOR); + emulate_ud(ctxt); goto done; } @@ -3135,7 +3165,7 @@ twobyte_insn: ((ctxt->mode == X86EMUL_MODE_PROT64) ? ~0ULL : ~0U), ctxt->vcpu) < 0) { /* #UD condition is already handled by the code above */ - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } @@ -3146,7 +3176,7 @@ twobyte_insn: msr_data = (u32)c->regs[VCPU_REGS_RAX] | ((u64)c->regs[VCPU_REGS_RDX] << 32); if (ops->set_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], msr_data)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } rc = X86EMUL_CONTINUE; @@ -3155,7 +3185,7 @@ twobyte_insn: case 0x32: /* rdmsr */ if (ops->get_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], &msr_data)) { - kvm_inject_gp(ctxt->vcpu, 0); + emulate_gp(ctxt, 0); goto done; } else { c->regs[VCPU_REGS_RAX] = (u32)msr_data; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 91bfe7771f50..63c87adcec48 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3852,6 +3852,17 @@ static void toggle_interruptibility(struct kvm_vcpu *vcpu, u32 mask) kvm_x86_ops->set_interrupt_shadow(vcpu, mask); } +static void inject_emulated_exception(struct kvm_vcpu *vcpu) +{ + struct x86_emulate_ctxt *ctxt = &vcpu->arch.emulate_ctxt; + if (ctxt->exception == PF_VECTOR) + kvm_inject_page_fault(vcpu, ctxt->cr2, ctxt->error_code); + else if (ctxt->error_code_valid) + kvm_queue_exception_e(vcpu, ctxt->exception, ctxt->error_code); + else + kvm_queue_exception(vcpu, ctxt->exception); +} + int emulate_instruction(struct kvm_vcpu *vcpu, unsigned long cr2, u16 error_code, @@ -3886,6 +3897,7 @@ int emulate_instruction(struct kvm_vcpu *vcpu, memset(c, 0, sizeof(struct decode_cache)); memcpy(c->regs, vcpu->arch.regs, sizeof c->regs); vcpu->arch.emulate_ctxt.interruptibility = 0; + vcpu->arch.emulate_ctxt.exception = -1; r = x86_decode_insn(&vcpu->arch.emulate_ctxt, &emulate_ops); trace_kvm_emulate_insn_start(vcpu); @@ -3958,6 +3970,11 @@ restart: memcpy(vcpu->arch.regs, c->regs, sizeof c->regs); kvm_rip_write(vcpu, vcpu->arch.emulate_ctxt.eip); + if (vcpu->arch.emulate_ctxt.exception >= 0) { + inject_emulated_exception(vcpu); + return EMULATE_DONE; + } + if (vcpu->arch.pio.count) { if (!vcpu->arch.pio.in) vcpu->arch.pio.count = 0; @@ -3970,9 +3987,6 @@ restart: return EMULATE_DO_MMIO; } - if (vcpu->arch.exception.pending) - vcpu->arch.emulate_ctxt.restart = false; - if (vcpu->arch.emulate_ctxt.restart) goto restart;