Re: [PATCH v3 05/14] perf cs-etm: Use the recorded instruction size for A32 and A64

[sashiko-bot@kernel.org]

> Use packet->last_instr_size in cs_etm__instr_size() instead of a
> hard-coded value of four. Ordinary A32/A64 instruction ranges are
> unchanged, since their instructions are always four bytes.
> 
> Prepare for synthesizing exception entries separately from the preceding
> instruction range. An exception packet may have last_instr_size = 0,
> meaning that no instruction completed at the exception source PC and the
> exception return address must not advance past that PC. The thread stack
> computes the return address as the source PC plus the instruction size,
> so returning four would incorrectly advance the return address by four
> bytes.
> 
> Keep per-address decoding for T32. A range can mix 16-bit and 32-bit
> instructions, so last_instr_size describes only the final instruction
> and cannot determine the size at an arbitrary address in the range.
> 
> Reviewed-by: James Clark <james.clark@linaro.org>
> Signed-off-by: Leo Yan <leo.yan@arm.com>

Sashiko has reviewed this patch and found no issues.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261030-perf_cs_etm_fix_non_taken-v3-0-7629b52ab047@arm.com?part=5