]> asedeno.scripts.mit.edu Git - linux.git/commit
arm64: cpufeature: Fix handling of CTR_EL0.IDC field
authorSuzuki K Poulose <suzuki.poulose@arm.com>
Tue, 9 Oct 2018 13:47:06 +0000 (14:47 +0100)
committerCatalin Marinas <catalin.marinas@arm.com>
Tue, 16 Oct 2018 10:53:31 +0000 (11:53 +0100)
commit1602df02f33f61fe0de1bbfeba0d1c97c14bff19
tree97caef6e1301b3566fa5e33ab142f6912f623441
parent8ab66cbe63aeaf9e5970fb4aaef1c660fca59321
arm64: cpufeature: Fix handling of CTR_EL0.IDC field

CTR_EL0.IDC reports the data cache clean requirements for instruction
to data coherence. However, if the field is 0, we need to check the
CLIDR_EL1 fields to detect the status of the feature. Currently we
don't do this and generate a warning with tainting the kernel, when
there is a mismatch in the field among the CPUs. Also the userspace
doesn't have a reliable way to check the CLIDR_EL1 register to check
the status.

This patch fixes the problem by checking the CLIDR_EL1 fields, when
(CTR_EL0.IDC == 0) and updates the kernel's copy of the CTR_EL0 for
the CPU with the actual status of the feature. This would allow the
sanity check infrastructure to do the proper checking of the fields
and also allow the CTR_EL0 emulation code to supply the real status
of the feature.

Now, if a CPU has raw CTR_EL0.IDC == 0 and effective IDC == 1 (with
overall system wide IDC == 1), we need to expose the real value to
the user. So, we trap CTR_EL0 access on the CPU which reports incorrect
CTR_EL0.IDC.

Fixes: commit 6ae4b6e057888 ("arm64: Add support for new control bits CTR_EL0.DIC and CTR_EL0.IDC")
Cc: Shanker Donthineni <shankerd@codeaurora.org>
Cc: Philip Elcan <pelcan@codeaurora.org>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
arch/arm64/include/asm/cache.h
arch/arm64/kernel/cpu_errata.c
arch/arm64/kernel/cpufeature.c
arch/arm64/kernel/cpuinfo.c