Re: [PATCH v4 01/16] perf/x86/intel: Add x86_pmu.pebs_vmx for Ice Lake Servers

From: Xu, Like
Date: Wed Apr 14 2021 - 21:38:45 EST


On 2021/4/14 22:49, Liuxiangdong wrote:
Hi Like,

On 2021/4/9 16:46, Like Xu wrote:
Hi Liuxiangdong,

On 2021/4/9 16:33, Liuxiangdong (Aven, Cloud Infrastructure Service Product Dept.) wrote:
Do you have any comments or ideas about it ?

https://lore.kernel.org/kvm/606E5EF6.2060402@xxxxxxxxxx/

My expectation is that there may be many fewer PEBS samples
on Skylake without any soft lockup.

You may need to confirm the statement

"All that matters is that the EPT pages don't get
unmapped ever while PEBS is active"

is true in the kernel level.

Try "-overcommit mem-lock=on" for your qemu.


Sorry, in fact, I don't quite understand
"My expectation is that there may be many fewer PEBS samples on Skylake without any soft lockup. "

For testcase: perf record -e instructions:pp ./workload

We can get 2242 samples on the ICX guest, but
only 17 samples or less on the Skylake guest.

In my testcase on Skylake, neither the host nor the guest triggered the soft lock.


And, I have used "-overcommit mem-lock=on"  when soft lockup happens.

I misunderstood the use of "mem-lock=on". It is not the same as the
guest mem pin and I believe more kernel patches are needed.



Now, I have tried to configure 1G-hugepages for 2G-mem vm. Each of guest numa nodes has 1G mem.
When I use pebs(perf record -e cycles:pp) in guest, there are successful pebs samples just for a while and
then I cannot get pebs samples. Host doesn't soft lockup in this process.

In the worst case, no samples are expected.


Are there something wrong on skylake for we can only get a few samples? IRQ?  Or using hugepage is not effecitve?

The few samples comes from hardware limitation.
The Skylake doesn't have this "EPT-Friendly PEBS" capabilityand
some PEBS records will be lost when used by guests.


Thanks!



On 2021/4/6 13:14, Xu, Like wrote:
Hi Xiangdong,

On 2021/4/6 11:24, Liuxiangdong (Aven, Cloud Infrastructure Service Product Dept.) wrote:
Hi,like.
Some questions about this new pebs patches set:
https://lore.kernel.org/kvm/20210329054137.120994-2-like.xu@xxxxxxxxxxxxxxx/

The new hardware facility supporting guest PEBS is only available
on Intel Ice Lake Server platforms for now.

Yes, we have documented this "EPT-friendly PEBS" capability in the SDM
18.3.10.1 Processor Event Based Sampling (PEBS) Facility

And again, this patch set doesn't officially support guest PEBS on the Skylake.



AFAIK, Icelake supports adaptive PEBS and extended PEBS which Skylake doesn't.
But we can still use IA32_PEBS_ENABLE MSR to indicate general-purpose counter in Skylake.

For Skylake, only the PMC0-PMC3 are valid for PEBS and you may
mask the other unsupported bits in the pmu->pebs_enable_mask.

Is there anything else that only Icelake supports in this patches set?

The PDIR counter on the Ice Lake is the fixed counter 0
while the PDIR counter on the Sky Lake is the gp counter 1.

You may also expose x86_pmu.pebs_vmx for Skylake in the 1st patch.



Besides, we have tried this patches set in Icelake.  We can use pebs(eg: "perf record -e cycles:pp")
when guest is kernel-5.11, but can't when kernel-4.18.  Is there a minimum guest kernel version requirement?

The Ice Lake CPU model has been added since v5.4.

You may double check whether the stable tree(s) code has
INTEL_FAM6_ICELAKE in the arch/x86/include/asm/intel-family.h.



Thanks,
Xiangdong Liu