Skip to contentSkip to conversation

pstore fix for v4.10-rc8

GIT PULL Applied Kees Cook 3 participants 3 messages Feb 9, 2017 – Feb 16, 2017 Arm (32-bit)

lore
Share & tools
lore

Kees Cook

author
#0 lore raw

Hi,

This pstore regression was just noticed. If I'm too late, this can land in v4.11 and -stable, but if the fix window is still open for v4.10, I'd love to see this fix pulled for v4.10.

Thanks!

-Kees

The following changes since commit d966564fcdc19e13eb6ba1fbe6b8101070339c3d:

Revert "x86/ioapic: Restore IO-APIC irq_chip retrigger callback" (2017-02-08 18:08:29 -0800)

are available in the git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux.git tags/pstore-v4.10-rc8

for you to fetch changes up to 8672aed7bd865774257efd40929702759a869329:

pstore: don't OOPS when there are no ftrace zones (2017-02-09 11:49:49 -0800)

----------------------------------------------------------------
Fix pstore regression (boot Oops) when ftrace disabled, from Brian Norris.

----------------------------------------------------------------
Brian Norris (1):
pstore: don't OOPS when there are no ftrace zones

 fs/pstore/ram.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Signature · 2 lines
Kees Cook
Pixel Security
#1 lore raw

PATCH 1/1 ARM: multi_v7_defconfig: enable pstore configs

It's useful to get testing on these features on various boards (e.g., on KernelCI), and (when they're working) the ramoops feature is extremely helpful for debugging live systems.

---
+ ARM folks
- Linus

Quoted 1 line

From Kees's recent pull request:

Quoted 9 lines · Kees Cook

Hi,

This pstore regression was just noticed. If I'm too late, this can land in v4.11 and -stable, but if the fix window is still open for v4.10, I'd love to see this fix pulled for v4.10.

Thanks!

-Kees

[...]

Quoted 4 lines

----------------------------------------------------------------
Fix pstore regression (boot Oops) when ftrace disabled, from Brian Norris.

----------------------------------------------------------------

I expect we aren't getting regular testing coverage (e.g., from KernelCI, which has a few ARM-based Chromebooks in the lab [1], which would have trivially caught the bug) because pstore is not enabled on the multiplatform config that gets used there. Seems like a good idea, so we can catch regressions like this automatically?

[1] https://kernelci.org/boot/rk3288-veyron-jerry/

 arch/arm/configs/multi_v7_defconfig | 4 ++++
 1 file changed, 4 insertions(+)
arch/arm/configs/multi_v7_defconfig +4 -0
index 028d2b70e3b5..74bdb89faff6 100644
@@ -906,6 +906,10 @@ CONFIG_TMPFS=y
 CONFIG_SQUASHFS=y
 CONFIG_SQUASHFS_LZO=y
 CONFIG_SQUASHFS_XZ=y
+CONFIG_PSTORE=y
+CONFIG_PSTORE_CONSOLE=y
+CONFIG_PSTORE_PMSG=y
+CONFIG_PSTORE_RAM=y
 CONFIG_NFS_FS=y
 CONFIG_NFS_V3_ACL=y
 CONFIG_NFS_V4=y
Signature · 1 line
2.11.0.483
#2 lore raw

Re: [PATCH] ARM: multi_v7_defconfig: enable pstore configs

Quoted 8 lines · CET Brian Norris

It's useful to get testing on these features on various boards (e.g., on KernelCI), and (when they're working) the ramoops feature is extremely helpful for debugging live systems.

---
+ ARM folks
- Linus

Applied to next/defconfig, thanks!

Arnd