* [PATCH v2 00/01] KVM: x86: never specify a sample period for virtualized in_tx_cp counters
@ 2017-02-01 4:06 Robert O'Callahan
2017-02-01 4:06 ` [PATCH v2 01/01] " Robert O'Callahan
2017-02-23 2:56 ` [PATCH v2 00/01] " Robert O'Callahan
0 siblings, 2 replies; 5+ messages in thread
From: Robert O'Callahan @ 2017-02-01 4:06 UTC (permalink / raw)
To: kvm
Cc: Paolo Bonzini, Radim Krčmář,
Thomas Gleixner, Ingo Molnar, H . Peter Anvin, x86, linux-kernel,
Robert O'Callahan
Changes in version 2:
- Fixed block comment formatting as requested by Ingo Molnar
- Changed commit summary to refer to a testcase that more easily shows the
problem.
For reference here's a complete testcase for an Intel TSX-enabled KVM guest:
#include <linux/perf_event.h>
#include <stdio.h>
#include <string.h>
#include <sys/ioctl.h>
#include <sys/syscall.h>
#include <unistd.h>
int main(void) {
struct perf_event_attr attr;
int fd;
long long count;
memset(&attr, 0, sizeof(attr));
attr.type = PERF_TYPE_RAW;
attr.size = sizeof(attr);
attr.config = 0x2005101c4; // conditional branches retired IN_TXCP
attr.sample_period = 0;
attr.exclude_kernel = 1;
attr.exclude_guest = 1;
fd = syscall(__NR_perf_event_open, &attr, 0, -1, -1, 0);
ioctl(fd, PERF_EVENT_IOC_DISABLE, 0);
ioctl(fd, PERF_EVENT_IOC_ENABLE, 0);
for (int i = 0; i < 500; ++i) {
putchar('.');
}
read(fd, &count, sizeof(count));
printf("\nConditional branches: %lld\n%s\n", count,
count < 500 ? "FAILED" : "PASSED");
return 0;
}
Robert O'Callahan (1):
KVM: x86: never specify a sample period for virtualized in_tx_cp
counters
arch/x86/kvm/pmu.c | 13 ++++++++++---
1 file changed, 10 insertions(+), 3 deletions(-)
--
2.9.3
^ permalink raw reply [flat|nested] 5+ messages in thread
* [PATCH v2 01/01] KVM: x86: never specify a sample period for virtualized in_tx_cp counters
2017-02-01 4:06 [PATCH v2 00/01] KVM: x86: never specify a sample period for virtualized in_tx_cp counters Robert O'Callahan
@ 2017-02-01 4:06 ` Robert O'Callahan
2017-02-23 2:56 ` [PATCH v2 00/01] " Robert O'Callahan
1 sibling, 0 replies; 5+ messages in thread
From: Robert O'Callahan @ 2017-02-01 4:06 UTC (permalink / raw)
To: kvm
Cc: Paolo Bonzini, Radim Krčmář,
Thomas Gleixner, Ingo Molnar, H . Peter Anvin, x86, linux-kernel,
Robert O'Callahan
pmc_reprogram_counter() always sets a sample period based on the value of
pmc->counter. However, hsw_hw_config() rejects sample periods less than
2^31 - 1. So for example, if a KVM guest does
struct perf_event_attr attr;
memset(&attr, 0, sizeof(attr));
attr.type = PERF_TYPE_RAW;
attr.size = sizeof(attr);
attr.config = 0x2005101c4; // conditional branches retired IN_TXCP
attr.sample_period = 0;
int fd = syscall(__NR_perf_event_open, &attr, 0, -1, -1, 0);
ioctl(fd, PERF_EVENT_IOC_DISABLE, 0);
ioctl(fd, PERF_EVENT_IOC_ENABLE, 0);
the guest kernel counts some conditional branch events, then updates the
virtual PMU register with a nonzero count. The host reaches
pmc_reprogram_counter() with nonzero pmc->counter, triggers EOPNOTSUPP
in hsw_hw_config(), prints "kvm_pmu: event creation failed" in
pmc_reprogram_counter(), and silently (from the guest's point of view) stops
counting events.
We fix event counting by forcing attr.sample_period to always be zero for
in_tx_cp counters. Sampling doesn't work, but it already didn't work and
can't be fixed without major changes to the approach in hsw_hw_config().
Signed-off-by: Robert O'Callahan <robert@ocallahan.org>
---
arch/x86/kvm/pmu.c | 13 ++++++++++---
1 file changed, 10 insertions(+), 3 deletions(-)
diff --git a/arch/x86/kvm/pmu.c b/arch/x86/kvm/pmu.c
index 06ce377..026db42 100644
--- a/arch/x86/kvm/pmu.c
+++ b/arch/x86/kvm/pmu.c
@@ -113,12 +113,19 @@ static void pmc_reprogram_counter(struct kvm_pmc *pmc, u32 type,
.config = config,
};
+ attr.sample_period = (-pmc->counter) & pmc_bitmask(pmc);
+
if (in_tx)
attr.config |= HSW_IN_TX;
- if (in_tx_cp)
+ if (in_tx_cp) {
+ /*
+ * HSW_IN_TX_CHECKPOINTED is not supported with nonzero
+ * period. Just clear the sample period so at least
+ * allocating the counter doesn't fail.
+ */
+ attr.sample_period = 0;
attr.config |= HSW_IN_TX_CHECKPOINTED;
-
- attr.sample_period = (-pmc->counter) & pmc_bitmask(pmc);
+ }
event = perf_event_create_kernel_counter(&attr, -1, current,
intr ? kvm_perf_overflow_intr :
--
2.9.3
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH v2 00/01] KVM: x86: never specify a sample period for virtualized in_tx_cp counters
2017-02-01 4:06 [PATCH v2 00/01] KVM: x86: never specify a sample period for virtualized in_tx_cp counters Robert O'Callahan
2017-02-01 4:06 ` [PATCH v2 01/01] " Robert O'Callahan
@ 2017-02-23 2:56 ` Robert O'Callahan
2017-02-23 8:31 ` Paolo Bonzini
1 sibling, 1 reply; 5+ messages in thread
From: Robert O'Callahan @ 2017-02-23 2:56 UTC (permalink / raw)
To: kvm
Cc: Paolo Bonzini, Radim Krčmář,
Thomas Gleixner, Ingo Molnar, H . Peter Anvin, x86, linux-kernel,
Robert O'Callahan
Ping? Is there something else I need to do to get someone to look at this?
Thanks,
Rob
--
lbir ye,ea yer.tnietoehr rdn rdsme,anea lurpr edna e hnysnenh hhe uresyf toD
selthor stor edna siewaoeodm or v sstvr esBa kbvted,t rdsme,aoreseoouoto
o l euetiuruewFa kbn e hnystoivateweh uresyf tulsa rehr rdm or rnea lurpr
.a war hsrer holsa rodvted,t nenh hneireseoouot.tniesiewaoeivatewt sstvr esn
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2 00/01] KVM: x86: never specify a sample period for virtualized in_tx_cp counters
2017-02-23 2:56 ` [PATCH v2 00/01] " Robert O'Callahan
@ 2017-02-23 8:31 ` Paolo Bonzini
2017-02-23 9:43 ` Robert O'Callahan
0 siblings, 1 reply; 5+ messages in thread
From: Paolo Bonzini @ 2017-02-23 8:31 UTC (permalink / raw)
To: robert, kvm
Cc: Radim Krčmář,
Thomas Gleixner, Ingo Molnar, H . Peter Anvin, x86, linux-kernel
On 23/02/2017 03:56, Robert O'Callahan wrote:
> Ping? Is there something else I need to do to get someone to look at this?
Nothing to do, we will pick it up for 4.11.
Paolo
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v2 00/01] KVM: x86: never specify a sample period for virtualized in_tx_cp counters
2017-02-23 8:31 ` Paolo Bonzini
@ 2017-02-23 9:43 ` Robert O'Callahan
0 siblings, 0 replies; 5+ messages in thread
From: Robert O'Callahan @ 2017-02-23 9:43 UTC (permalink / raw)
To: Paolo Bonzini
Cc: kvm, Radim Krčmář,
Thomas Gleixner, Ingo Molnar, H . Peter Anvin, x86, linux-kernel
Great, thanks!
Rob
--
lbir ye,ea yer.tnietoehr rdn rdsme,anea lurpr edna e hnysnenh hhe uresyf toD
selthor stor edna siewaoeodm or v sstvr esBa kbvted,t rdsme,aoreseoouoto
o l euetiuruewFa kbn e hnystoivateweh uresyf tulsa rehr rdm or rnea lurpr
.a war hsrer holsa rodvted,t nenh hneireseoouot.tniesiewaoeivatewt sstvr esn
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2017-02-23 9:43 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-02-01 4:06 [PATCH v2 00/01] KVM: x86: never specify a sample period for virtualized in_tx_cp counters Robert O'Callahan
2017-02-01 4:06 ` [PATCH v2 01/01] " Robert O'Callahan
2017-02-23 2:56 ` [PATCH v2 00/01] " Robert O'Callahan
2017-02-23 8:31 ` Paolo Bonzini
2017-02-23 9:43 ` Robert O'Callahan
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).