On powerpc the perf event interrupt is not masked when interrupts are
disabled, allowing it to function as an NMI.

This causes problems if perf is using vmalloc. If we take a page fault
on the vmalloc region the fault handler will fail the page fault because
it detects we are coming in from an NMI (see do_hash_page()).

We don't actually need or want vmalloc backed perf so just disable it on
powerpc.

Signed-off-by: Michael Ellerman <m...@ellerman.id.au>
---
 init/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/init/Kconfig b/init/Kconfig
index dc24dec60232..81050e4d9941 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -1637,7 +1637,7 @@ config PERF_EVENTS
 config DEBUG_PERF_USE_VMALLOC
        default n
        bool "Debug: use vmalloc to back perf mmap() buffers"
-       depends on PERF_EVENTS && DEBUG_KERNEL
+       depends on PERF_EVENTS && DEBUG_KERNEL && !PPC
        select PERF_USE_VMALLOC
        help
         Use vmalloc memory to back perf mmap() buffers.
-- 
2.1.0

_______________________________________________
Linuxppc-dev mailing list
Linuxppc-dev@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/linuxppc-dev

Reply via email to