arm64: stacktrace: restore terminal records
authorMark Rutland <mark.rutland@arm.com>
Thu, 29 Apr 2021 10:20:04 +0000 (11:20 +0100)
committerCatalin Marinas <catalin.marinas@arm.com>
Fri, 30 Apr 2021 17:30:45 +0000 (18:30 +0100)
commit8533d5bfad41e74b7dd80d292fd484913cdfb374
treef4b0427bf323eea7321e1c71d7cf5479eebe23dc
parent388708028e6937f3fc5fc19aeeb847f8970f489c
arm64: stacktrace: restore terminal records

We removed the terminal frame records in commit:

   6106e1112cc69a36 ("arm64: remove EL0 exception frame record")

... on the assumption that as we no longer used them to find the pt_regs
at exception boundaries, they were no longer necessary.

However, Leo reports that as an unintended side-effect, this causes
traces which cross secondary_start_kernel to terminate one entry too
late, with a spurious "0" entry.

There are a few ways we could sovle this, but as we're planning to use
terminal records for RELIABLE_STACKTRACE, let's revert the logic change
for now, keeping the update comments and accounting for the changes in
commit:

  3c02600144bdb0a1 ("arm64: stacktrace: Report when we reach the end of the stack")

This is effectively a partial revert of commit:

  6106e1112cc69a36 ("arm64: remove EL0 exception frame record")

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Fixes: 6106e1112cc6 ("arm64: remove EL0 exception frame record")
Reported-by: Leo Yan <leo.yan@linaro.org>
Tested-by: Leo Yan <leo.yan@linaro.org>
Cc: Will Deacon <will@kernel.org>
Cc: Mark Brown <broonie@kernel.org>
Cc: "Madhavan T. Venkataraman" <madvenka@linux.microsoft.com>
Link: https://lore.kernel.org/r/20210429104813.GA33550@C02TD0UTHF1T.local
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
arch/arm64/kernel/entry.S
arch/arm64/kernel/stacktrace.c