Hi Amir,

On Wed, Jun 03, 2026 at 01:10:17PM -0700, Amir Ayupov wrote:

Hi James,

Thank you for picking it up.

I tested the v2 patch series and it looks good. There was a minor difference in 2/39 tested perf data files: the number of brstack samples differs by one, however, there was no loss of binary profile. The resulting BOLT profile converted from the perf script output was identical, so I'm OK with v2 patch as-is.

Sorry jumping in as brstack is mentioned.

Now branch stack is maintained per-CPU wise, it mixes up branch stack cross threads. The patch 03 in the series [1] refactors branch stack per-thread by using common code.

Hope this can benefit a bit the profiling data quality and in case you are interested in.

Thanks, Leo

[1] https://lore.kernel.org/linux-perf-users/20260526-b4-arm_cs_callchain_suppor...