Sign Up
Log In
Log In
or
Sign Up
Places
All Projects
Status Monitor
Collapse sidebar
SUSE:SLE-15-SP7:GA
xen.15892
xsa307.patch
Overview
Repositories
Revisions
Requests
Users
Attributes
Meta
File xsa307.patch of Package xen.15892
x86+Arm32: make find_next_{,zero_}bit() have well defined behavior These functions getting used with the 2nd and 3rd arguments being equal wasn't well defined: Arm64 reliably returns the value of the 2nd argument in this case, while on x86 for bitmaps up to 64 bits wide the return value was undefined (due to the undefined behavior of a shift of a value by the number of bits it's wide) when the incoming value was 64. On Arm32 an actual out of bounds access would happen when the size/offset value is a multiple of 32; if this access doesn't fault, the return value would have been sufficiently correct afaict. Make the functions consistently tolerate the last two arguments being equal (and in fact the 3rd argument being greater or equal to the 2nd), in favor of finding and fixing all the use sites that violate the original more strict assumption. This is XSA-307. Signed-off-by: Jan Beulich <jbeulich@suse.com> Acked-by: Julien Grall <julien@xen.org> --- The most obvious (albeit still indirect) exposure to guests is evtchn_check_pollers(), which imo makes this a security issue at least for Arm32. This was originally already discussed between (at least) Andrew and me, and I don't really recall who brought up the issue first. Note that Arm's Linux origin of the code may call for syncing publication with them. Then again I don't want to tell them just to see them go public ahead of us. --- a/xen/arch/arm/arm32/lib/findbit.S +++ b/xen/arch/arm/arm32/lib/findbit.S @@ -42,8 +42,8 @@ ENDPROC(_find_first_zero_bit_le) * Prototype: int find_next_zero_bit(void *addr, unsigned int maxbit, int offset) */ ENTRY(_find_next_zero_bit_le) - teq r1, #0 - beq 3b + cmp r1, r2 + bls 3b ands ip, r2, #7 beq 1b @ If new byte, goto old routine ARM( ldrb r3, [r0, r2, lsr #3] ) @@ -83,8 +83,8 @@ ENDPROC(_find_first_bit_le) * Prototype: int find_next_zero_bit(void *addr, unsigned int maxbit, int offset) */ ENTRY(_find_next_bit_le) - teq r1, #0 - beq 3b + cmp r1, r2 + bls 3b ands ip, r2, #7 beq 1b @ If new byte, goto old routine ARM( ldrb r3, [r0, r2, lsr #3] ) @@ -117,8 +117,8 @@ ENTRY(_find_first_zero_bit_be) ENDPROC(_find_first_zero_bit_be) ENTRY(_find_next_zero_bit_be) - teq r1, #0 - beq 3b + cmp r1, r2 + bls 3b ands ip, r2, #7 beq 1b @ If new byte, goto old routine eor r3, r2, #0x18 @ big endian byte ordering @@ -151,8 +151,8 @@ ENTRY(_find_first_bit_be) ENDPROC(_find_first_bit_be) ENTRY(_find_next_bit_be) - teq r1, #0 - beq 3b + cmp r1, r2 + bls 3b ands ip, r2, #7 beq 1b @ If new byte, goto old routine eor r3, r2, #0x18 @ big endian byte ordering --- a/xen/include/asm-x86/bitops.h +++ b/xen/include/asm-x86/bitops.h @@ -340,7 +340,7 @@ static always_inline unsigned int __scan const unsigned long *a__ = (addr); \ unsigned int s__ = (size); \ unsigned int o__ = (off); \ - if ( __builtin_constant_p(size) && !s__ ) \ + if ( o__ >= s__ ) \ r__ = s__; \ else if ( __builtin_constant_p(size) && s__ <= BITS_PER_LONG ) \ r__ = o__ + __scanbit(*(const unsigned long *)(a__) >> o__, s__); \ @@ -372,7 +372,7 @@ static always_inline unsigned int __scan const unsigned long *a__ = (addr); \ unsigned int s__ = (size); \ unsigned int o__ = (off); \ - if ( __builtin_constant_p(size) && !s__ ) \ + if ( o__ >= s__ ) \ r__ = s__; \ else if ( __builtin_constant_p(size) && s__ <= BITS_PER_LONG ) \ r__ = o__ + __scanbit(~*(const unsigned long *)(a__) >> o__, s__); \
Locations
Projects
Search
Status Monitor
Help
OpenBuildService.org
Documentation
API Documentation
Code of Conduct
Contact
Support
@OBShq
Terms
openSUSE Build Service is sponsored by
The Open Build Service is an
openSUSE project
.
Sign Up
Log In
Places
Places
All Projects
Status Monitor