From: "Radim Krčmář" <rkrcmar@redhat.com>
To: Andrey Korolyov <andrey@xdel.ru>
Cc: "kvm@vger.kernel.org" <kvm@vger.kernel.org>,
"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
"Dr. David Alan Gilbert" <dgilbert@redhat.com>,
Bandan Das <bsd@redhat.com>, Kevin O'Connor <kevin@koconnor.net>,
Gerd Hoffmann <kraxel@redhat.com>,
Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [Qemu-devel] E5-2620v2 - emulation stop error
Date: Tue, 31 Mar 2015 18:45:40 +0200 [thread overview]
Message-ID: <20150331164539.GD14262@potion.brq.redhat.com> (raw)
In-Reply-To: <CABYiri8fK3zta+m0eVpStn9vRmU5-doDim9JG2sr1uGFhMkWfg@mail.gmail.com>
2015-03-31 17:56+0300, Andrey Korolyov:
> > Chasing the culprit this way could take a long time, so a new tracepoint
> > that shows if 0xef is set on entry would let us guess the bug faster ...
> >
> > Please provide a failing trace with the following patch:
>
> Thanks, please see below:
>
> http://xdel.ru/downloads/kvm-e5v2-issue/new-tracepoint-fail-with-apicv.dat.gz
qemu-system-x86-4022 [006] 255.915978:
kvm_entry: vcpu 0
kvm_emulate_insn: f0000:d275: ea 7a d2 00 f0
kvm_emulate_insn: f0000:d27a: 2e 0f 01 1e f0 6c
kvm_emulate_insn: f0000:d280: 31 c0
kvm_emulate_insn: f0000:d282: 8e e0
kvm_emulate_insn: f0000:d284: 8e e8
kvm_emulate_insn: f0000:d286: 8e c0
kvm_emulate_insn: f0000:d288: 8e d8
kvm_emulate_insn: f0000:d28a: 8e d0
kvm_entry: vcpu 0
kvm_0xef: irr clear, isr clear, vmcs 0x0
kvm_exit: reason EPT_VIOLATION rip 0x8dd0 info 184 0
kvm_page_fault: address f8dd0 error_code 184
kvm_entry: vcpu 0
kvm_0xef: irr clear, isr clear, vmcs 0x0
kvm_exit: reason EPT_VIOLATION rip 0x76d6 info 184 0
kvm_page_fault: address f76d6 error_code 184
kvm_entry: vcpu 0
kvm_0xef: irr clear, isr clear, vmcs 0x0
kvm_exit: reason EXCEPTION_NMI rip 0xd331 info 0 80000b0d
kvm_userspace_exit: reason KVM_EXIT_INTERNAL_ERROR (17)
Ok, nothing obvious here either ... I've desperately added all
information I know about. Please run it again, thanks.
(The patch has to be applied instead of the previous one.)
---
diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h
index 7c7bc8bef21f..f986636ad9d0 100644
--- a/arch/x86/kvm/trace.h
+++ b/arch/x86/kvm/trace.h
@@ -742,6 +742,41 @@ TRACE_EVENT(kvm_emulate_insn,
#define trace_kvm_emulate_insn_start(vcpu) trace_kvm_emulate_insn(vcpu, 0)
#define trace_kvm_emulate_insn_failed(vcpu) trace_kvm_emulate_insn(vcpu, 1)
+TRACE_EVENT(kvm_0xef,
+ TP_PROTO(bool irr, bool isr, u32 info, bool on, bool pir, u16 status),
+ TP_ARGS(irr, isr, info, on, pir, status),
+
+ TP_STRUCT__entry(
+ __field(bool, irr )
+ __field(bool, isr )
+ __field(u32, info)
+ __field(bool, on )
+ __field(bool, pir )
+ __field(u8, rvi )
+ __field(u8, svi )
+ ),
+
+ TP_fast_assign(
+ __entry->irr = irr;
+ __entry->isr = isr;
+ __entry->info = info;
+ __entry->on = on;
+ __entry->pir = pir;
+ __entry->rvi = status & 0xff;
+ __entry->svi = status >> 8;
+ ),
+
+ TP_printk("irr %s, isr %s, info 0x%x, on %s, pir %s, rvi 0x%x, svi 0x%x",
+ __entry->irr ? "set " : "clear",
+ __entry->isr ? "set " : "clear",
+ __entry->info,
+ __entry->on ? "set " : "clear",
+ __entry->pir ? "set " : "clear",
+ __entry->rvi,
+ __entry->svi
+ )
+ );
+
TRACE_EVENT(
vcpu_match_mmio,
TP_PROTO(gva_t gva, gpa_t gpa, bool write, bool gpa_match),
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index eee63dc33d89..b461edc93d53 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -5047,6 +5047,25 @@ static int handle_machine_check(struct kvm_vcpu *vcpu)
return 1;
}
+#define VEC_POS(v) ((v) & (32 - 1))
+#define REG_POS(v) (((v) >> 5) << 4)
+static inline int apic_test_vector(int vec, void *bitmap)
+{
+ return test_bit(VEC_POS(vec), (bitmap) + REG_POS(vec));
+}
+
+static inline void random_trace(struct kvm_vcpu *vcpu)
+{
+ struct vcpu_vmx *vmx = to_vmx(vcpu);
+
+ trace_kvm_0xef(apic_test_vector(0xef, vcpu->arch.apic->regs + APIC_IRR),
+ apic_test_vector(0xef, vcpu->arch.apic->regs + APIC_ISR),
+ vmcs_read32(VM_ENTRY_INTR_INFO_FIELD),
+ test_bit(POSTED_INTR_ON, (unsigned long *)&vmx->pi_desc.control),
+ test_bit(0xef, (unsigned long *)vmx->pi_desc.pir),
+ vmcs_read16(GUEST_INTR_STATUS));
+}
+
static int handle_exception(struct kvm_vcpu *vcpu)
{
struct vcpu_vmx *vmx = to_vmx(vcpu);
@@ -5077,6 +5096,8 @@ static int handle_exception(struct kvm_vcpu *vcpu)
return 1;
}
+ random_trace(vcpu);
+
error_code = 0;
if (intr_info & INTR_INFO_DELIVER_CODE_MASK)
error_code = vmcs_read32(VM_EXIT_INTR_ERROR_CODE);
@@ -8143,6 +8164,8 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
if (vmx->emulation_required)
return;
+ random_trace(vcpu);
+
if (vmx->ple_window_dirty) {
vmx->ple_window_dirty = false;
vmcs_write32(PLE_WINDOW, vmx->ple_window);
@@ -8312,6 +8335,8 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu)
vmx->exit_reason = vmcs_read32(VM_EXIT_REASON);
trace_kvm_exit(vmx->exit_reason, vcpu, KVM_ISA_VMX);
+ random_trace(vcpu);
+
/*
* the KVM_REQ_EVENT optimization bit is only on for one entry, and if
* we did not inject a still-pending event to L1 now because of
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 32bf19ef3115..a45fa01bd354 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -7881,3 +7881,4 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_intercepts);
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_write_tsc_offset);
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_ple_window);
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_pml_full);
+EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_0xef);
next prev parent reply other threads:[~2015-03-31 16:45 UTC|newest]
Thread overview: 83+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-03-05 22:14 [Qemu-devel] E5-2620v2 - emulation stop error Andrey Korolyov
2015-03-05 23:44 ` Andrey Korolyov
2015-03-06 16:57 ` Bandan Das
2015-03-07 0:00 ` Andrey Korolyov
2015-03-10 14:24 ` Andrey Korolyov
2015-03-10 16:57 ` Dr. David Alan Gilbert
2015-03-10 18:08 ` Andrey Korolyov
2015-03-10 18:16 ` Dr. David Alan Gilbert
2015-03-10 18:21 ` Andrey Korolyov
2015-03-10 19:30 ` Paolo Bonzini
2015-03-10 18:10 ` Paolo Bonzini
2015-03-10 18:21 ` Bandan Das
2015-03-10 19:25 ` Paolo Bonzini
2015-03-10 19:37 ` Dr. David Alan Gilbert
2015-03-10 20:29 ` Dr. David Alan Gilbert
2015-03-11 2:38 ` Bandan Das
2015-03-11 13:45 ` Dr. David Alan Gilbert
2015-03-11 15:42 ` Kevin O'Connor
2015-03-11 15:53 ` Dr. David Alan Gilbert
2015-03-11 16:37 ` Kevin O'Connor
2015-03-11 16:52 ` Dr. David Alan Gilbert
2015-03-11 17:37 ` Kevin O'Connor
2015-03-11 17:41 ` Paolo Bonzini
2015-03-11 17:59 ` Dr. David Alan Gilbert
2015-03-11 18:24 ` Bandan Das
2015-03-11 18:40 ` Kevin O'Connor
2015-03-11 18:45 ` Kevin O'Connor
2015-03-11 19:19 ` Kevin O'Connor
2015-03-11 19:33 ` Dr. David Alan Gilbert
2015-03-11 19:47 ` Bandan Das
2015-03-11 19:47 ` Andrey Korolyov
2015-03-11 19:59 ` Dr. David Alan Gilbert
2015-03-11 20:09 ` Andrey Korolyov
2015-03-12 9:59 ` Dr. David Alan Gilbert
2015-03-12 10:47 ` Andrey Korolyov
2015-03-16 19:17 ` Andrey Korolyov
2015-03-16 19:26 ` Dr. David Alan Gilbert
2015-03-25 20:43 ` Andrey Korolyov
2015-03-25 20:46 ` Andrey Korolyov
2015-03-25 20:54 ` Kevin O'Connor
2015-03-25 22:31 ` Andrey Korolyov
2015-03-25 23:02 ` Kevin O'Connor
2015-03-25 23:35 ` Andrey Korolyov
2015-03-26 0:05 ` Kevin O'Connor
2015-03-26 15:58 ` Radim Krčmář
2015-03-26 16:36 ` Kevin O'Connor
2015-03-26 16:48 ` Andrey Korolyov
2015-03-26 17:06 ` Kevin O'Connor
2015-03-26 17:08 ` Andrey Korolyov
2015-03-26 17:18 ` Kevin O'Connor
2015-03-26 17:33 ` Andrey Korolyov
2015-03-26 17:40 ` Radim Krčmář
2015-03-26 18:24 ` Andrey Korolyov
2015-03-26 20:40 ` Radim Krčmář
2015-03-26 21:03 ` Bandan Das
2015-03-27 10:16 ` Andrey Korolyov
2015-03-30 18:56 ` Radim Krčmář
2015-03-30 19:32 ` Andrey Korolyov
2015-03-31 13:45 ` Radim Krčmář
2015-03-31 14:56 ` Andrey Korolyov
2015-03-31 16:45 ` Radim Krčmář [this message]
2015-03-31 17:40 ` Andrey Korolyov
2015-03-31 18:01 ` Bandan Das
2015-03-31 18:04 ` Bandan Das
2015-03-31 18:23 ` Andrey Korolyov
2015-04-01 11:49 ` Radim Krčmář
2015-04-01 12:05 ` Paolo Bonzini
2015-04-01 12:26 ` Andrey Korolyov
2015-04-01 13:19 ` Paolo Bonzini
2015-04-01 15:37 ` Andrey Korolyov
2015-04-01 16:29 ` Andrey Korolyov
2015-04-01 22:58 ` Andrey Korolyov
2015-04-05 14:12 ` Andrey Korolyov
2015-03-27 11:54 ` Andrey Korolyov
2015-03-30 19:28 ` Radim Krčmář
2015-03-26 17:35 ` Radim Krčmář
2015-03-26 17:34 ` Radim Krčmář
2015-03-26 2:47 ` Bandan Das
2015-03-26 9:18 ` Andrey Korolyov
2015-03-26 15:05 ` Andrey Korolyov
2015-03-11 17:09 ` Bandan Das
2015-03-11 17:32 ` Kevin O'Connor
2015-03-11 18:01 ` Bandan Das
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20150331164539.GD14262@potion.brq.redhat.com \
--to=rkrcmar@redhat.com \
--cc=andrey@xdel.ru \
--cc=bsd@redhat.com \
--cc=dgilbert@redhat.com \
--cc=kevin@koconnor.net \
--cc=kraxel@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).