2020-08-21 07:16:48

by Christophe Leroy

Subject: [PATCH] powerpc/32s: Fix module loading failure when VMALLOC_END is over 0xf0000000

In is_module_segment(), when VMALLOC_END is over 0xf0000000,
ALIGN(VMALLOC_END, SZ_256M) has value 0.

In that case, addr >= ALIGN(VMALLOC_END, SZ_256M) is always
true then is_module_segment() always returns false.

Use (ALIGN(VMALLOC_END, SZ_256M) - 1) which will have
value 0xffffffff and will be suitable for the comparison.

Reported-by: Andreas Schwab <[email protected]>
Signed-off-by: Christophe Leroy <[email protected]>
Fixes: c49643319715 ("powerpc/32s: Only leave NX unset on segments used for modules")
---
arch/powerpc/mm/book3s32/mmu.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/mm/book3s32/mmu.c b/arch/powerpc/mm/book3s32/mmu.c
index 82ae9e06a773..d426eaf76bb0 100644
--- a/arch/powerpc/mm/book3s32/mmu.c
+++ b/arch/powerpc/mm/book3s32/mmu.c
@@ -194,12 +194,12 @@ static bool is_module_segment(unsigned long addr)
#ifdef MODULES_VADDR
if (addr < ALIGN_DOWN(MODULES_VADDR, SZ_256M))
return false;
- if (addr >= ALIGN(MODULES_END, SZ_256M))
+ if (addr > ALIGN(MODULES_END, SZ_256M) - 1)
return false;
#else
if (addr < ALIGN_DOWN(VMALLOC_START, SZ_256M))
return false;
- if (addr >= ALIGN(VMALLOC_END, SZ_256M))
+ if (addr > ALIGN(VMALLOC_END, SZ_256M) - 1)
return false;
#endif
return true;
--
2.25.0

2020-08-21 09:21:03

by Andreas Schwab

Subject: Re: [PATCH] powerpc/32s: Fix module loading failure when VMALLOC_END is over 0xf0000000

On Aug 21 2020, Christophe Leroy wrote:

> In is_module_segment(), when VMALLOC_END is over 0xf0000000,
> ALIGN(VMALLOC_END, SZ_256M) has value 0.
>
> In that case, addr >= ALIGN(VMALLOC_END, SZ_256M) is always
> true then is_module_segment() always returns false.
>
> Use (ALIGN(VMALLOC_END, SZ_256M) - 1) which will have
> value 0xffffffff and will be suitable for the comparison.
>
> Reported-by: Andreas Schwab <[email protected]>
> Signed-off-by: Christophe Leroy <[email protected]>
> Fixes: c49643319715 ("powerpc/32s: Only leave NX unset on segments used for modules")

Thanks, that fixes the crash.

Tested-by: Andreas Schwab <[email protected]>

Andreas.

--
Andreas Schwab, [email protected]
GPG Key fingerprint = 7578 EB47 D4E5 4D69 2510 2552 DF73 E780 A9DA AEC1
"And now for something completely different."

2020-08-27 07:48:00

by Michael Ellerman

Subject: Re: [PATCH] powerpc/32s: Fix module loading failure when VMALLOC_END is over 0xf0000000

On Fri, 21 Aug 2020 07:15:25 +0000 (UTC), Christophe Leroy wrote:
> In is_module_segment(), when VMALLOC_END is over 0xf0000000,
> ALIGN(VMALLOC_END, SZ_256M) has value 0.
>
> In that case, addr >= ALIGN(VMALLOC_END, SZ_256M) is always
> true then is_module_segment() always returns false.
>
> Use (ALIGN(VMALLOC_END, SZ_256M) - 1) which will have
> value 0xffffffff and will be suitable for the comparison.

Applied to powerpc/fixes.

[1/1] powerpc/32s: Fix module loading failure when VMALLOC_END is over 0xf0000000
https://git.kernel.org/powerpc/c/541cebb51f3422d4f2c6cb95c1e5cc3dcc9e5021

cheers