| Message ID | 1537311828-4547-1-git-send-email-penghao122@sina.com.cn (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [RESEND] kvm/x86: propagate fetch fault into guest | expand |
On Tue, 2018-09-18 at 19:03 -0400, Peng Hao wrote: > From: Peng Hao <peng.hao2@zte.com.cn> > > When handling ept misconfig exit, it will call emulate instruction > with insn_len = 0. The decode instruction function may return a fetch > fault and should propagate to guest. What fault is encountered, e.g. #PF, #UD, etc...? Something like a #UD at this point should result in an exit to userspace as it means KVM was unable to emulate an instruction that was successfully decoded by hardware. > The problem will result to emulation fail. > KVM internal error. Suberror: 1 > emulation failure > EAX=f81a0024 EBX=f6a07000 ECX=f6a0737c EDX=f8be0118 > ESI=f6a0737c EDI=00000021 EBP=f6929f98 ESP=f6929f98 > EIP=f8bdd141 EFL=00010086 [--S--P-] CPL=0 II=0 A20=1 SMM=0 HLT=0 > ES =007b 00000000 ffffffff 00c0f300 DPL=3 DS [-WA] > CS =0060 00000000 ffffffff 00c09b00 DPL=0 CS32 [-RA] > SS =0068 00000000 ffffffff 00c09300 DPL=0 DS [-WA] > DS =007b 00000000 ffffffff 00c0f300 DPL=3 DS [-WA] > FS =00d8 2c044000 ffffffff 00809300 DPL=0 DS16 [-WA] > GS =0033 081a44c8 01000fff 00d0f300 DPL=3 DS [-WA] > LDT=0000 00000000 ffffffff 00000000 > TR =0080 f6ea0c80 0000206b 00008b00 DPL=0 TSS32-busy > GDT= f6e99000 000000ff > IDT= fffbb000 000007ff > CR0=80050033 CR2=b757d000 CR3=35d31000 CR4=001406d0 > > Signed-off-by: Peng Hao <peng.hao2@zte.com.cn> > Reviewed-by: Jiang Biao <jiang.biao2@zte.com.cn> > --- > arch/x86/kvm/emulate.c | 5 ++++- > arch/x86/kvm/x86.c | 4 +++- > 2 files changed, 7 insertions(+), 2 deletions(-) > > diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c > index 106482d..aecf9a72 100644 > --- a/arch/x86/kvm/emulate.c > +++ b/arch/x86/kvm/emulate.c > @@ -5105,8 +5105,11 @@ int x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len) > memcpy(ctxt->fetch.data, insn, insn_len); > else { > rc = __do_insn_fetch_bytes(ctxt, 1); > - if (rc != X86EMUL_CONTINUE) > + if (rc != X86EMUL_CONTINUE) { > + if (rc == X86EMUL_PROPAGATE_FAULT) > + ctxt->have_exception = true; > return rc; > + } > } > > switch (mode) { > diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c > index 542f631..27072e1 100644 > --- a/arch/x86/kvm/x86.c > +++ b/arch/x86/kvm/x86.c > @@ -6188,8 +6188,10 @@ int x86_emulate_instruction(struct kvm_vcpu *vcpu, > if (reexecute_instruction(vcpu, cr2, write_fault_to_spt, > emulation_type)) > return EMULATE_DONE; > - if (ctxt->have_exception && inject_emulated_exception(vcpu)) > + if (ctxt->have_exception) { > + inject_emulated_exception(vcpu); > return EMULATE_DONE; > + } > if (emulation_type & EMULTYPE_SKIP) > return EMULATE_FAIL; > return handle_emulation_failure(vcpu, emulation_type);
On 19/09/2018 16:33, Sean Christopherson wrote: > On Tue, 2018-09-18 at 19:03 -0400, Peng Hao wrote: >> From: Peng Hao <peng.hao2@zte.com.cn> >> >> When handling ept misconfig exit, it will call emulate instruction >> with insn_len = 0. The decode instruction function may return a fetch >> fault and should propagate to guest. > > What fault is encountered, e.g. #PF, #UD, etc...? Something like a > #UD at this point should result in an exit to userspace as it means > KVM was unable to emulate an instruction that was successfully > decoded by hardware. If it's the same bug as https://lore.kernel.org/patchwork/patch/850077/ it would be a #PF. Paolo >> The problem will result to emulation fail. >> KVM internal error. Suberror: 1 >> emulation failure >> EAX=f81a0024 EBX=f6a07000 ECX=f6a0737c EDX=f8be0118 >> ESI=f6a0737c EDI=00000021 EBP=f6929f98 ESP=f6929f98 >> EIP=f8bdd141 EFL=00010086 [--S--P-] CPL=0 II=0 A20=1 SMM=0 HLT=0 >> ES =007b 00000000 ffffffff 00c0f300 DPL=3 DS [-WA] >> CS =0060 00000000 ffffffff 00c09b00 DPL=0 CS32 [-RA] >> SS =0068 00000000 ffffffff 00c09300 DPL=0 DS [-WA] >> DS =007b 00000000 ffffffff 00c0f300 DPL=3 DS [-WA] >> FS =00d8 2c044000 ffffffff 00809300 DPL=0 DS16 [-WA] >> GS =0033 081a44c8 01000fff 00d0f300 DPL=3 DS [-WA] >> LDT=0000 00000000 ffffffff 00000000 >> TR =0080 f6ea0c80 0000206b 00008b00 DPL=0 TSS32-busy >> GDT= f6e99000 000000ff >> IDT= fffbb000 000007ff >> CR0=80050033 CR2=b757d000 CR3=35d31000 CR4=001406d0 >> >> Signed-off-by: Peng Hao <peng.hao2@zte.com.cn> >> Reviewed-by: Jiang Biao <jiang.biao2@zte.com.cn> >> --- >> arch/x86/kvm/emulate.c | 5 ++++- >> arch/x86/kvm/x86.c | 4 +++- >> 2 files changed, 7 insertions(+), 2 deletions(-) >> >> diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c >> index 106482d..aecf9a72 100644 >> --- a/arch/x86/kvm/emulate.c >> +++ b/arch/x86/kvm/emulate.c >> @@ -5105,8 +5105,11 @@ int x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len) >> memcpy(ctxt->fetch.data, insn, insn_len); >> else { >> rc = __do_insn_fetch_bytes(ctxt, 1); >> - if (rc != X86EMUL_CONTINUE) >> + if (rc != X86EMUL_CONTINUE) { >> + if (rc == X86EMUL_PROPAGATE_FAULT) >> + ctxt->have_exception = true; >> return rc; >> + } >> } >> >> switch (mode) { >> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c >> index 542f631..27072e1 100644 >> --- a/arch/x86/kvm/x86.c >> +++ b/arch/x86/kvm/x86.c >> @@ -6188,8 +6188,10 @@ int x86_emulate_instruction(struct kvm_vcpu *vcpu, >> if (reexecute_instruction(vcpu, cr2, write_fault_to_spt, >> emulation_type)) >> return EMULATE_DONE; >> - if (ctxt->have_exception && inject_emulated_exception(vcpu)) >> + if (ctxt->have_exception) { >> + inject_emulated_exception(vcpu); >> return EMULATE_DONE; >> + } >> if (emulation_type & EMULTYPE_SKIP) >> return EMULATE_FAIL; >> return handle_emulation_failure(vcpu, emulation_type);
On Thu, Sep 20, 2018 at 06:32:10PM +0200, Paolo Bonzini wrote: > On 19/09/2018 16:33, Sean Christopherson wrote: > > On Tue, 2018-09-18 at 19:03 -0400, Peng Hao wrote: > >> From: Peng Hao <peng.hao2@zte.com.cn> > >> > >> When handling ept misconfig exit, it will call emulate instruction > >> with insn_len = 0. The decode instruction function may return a fetch > >> fault and should propagate to guest. > > > > What fault is encountered, e.g. #PF, #UD, etc...? Something like a > > #UD at this point should result in an exit to userspace as it means > > KVM was unable to emulate an instruction that was successfully > > decoded by hardware. > > If it's the same bug as https://lore.kernel.org/patchwork/patch/850077/ > it would be a #PF. Indeed it is, let's move this discussion to that thread.
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index 106482d..aecf9a72 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -5105,8 +5105,11 @@ int x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len) memcpy(ctxt->fetch.data, insn, insn_len); else { rc = __do_insn_fetch_bytes(ctxt, 1); - if (rc != X86EMUL_CONTINUE) + if (rc != X86EMUL_CONTINUE) { + if (rc == X86EMUL_PROPAGATE_FAULT) + ctxt->have_exception = true; return rc; + } } switch (mode) { diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 542f631..27072e1 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -6188,8 +6188,10 @@ int x86_emulate_instruction(struct kvm_vcpu *vcpu, if (reexecute_instruction(vcpu, cr2, write_fault_to_spt, emulation_type)) return EMULATE_DONE; - if (ctxt->have_exception && inject_emulated_exception(vcpu)) + if (ctxt->have_exception) { + inject_emulated_exception(vcpu); return EMULATE_DONE; + } if (emulation_type & EMULTYPE_SKIP) return EMULATE_FAIL; return handle_emulation_failure(vcpu, emulation_type);