powerpc: ppc64 address space capped at 32TB, mmap randomisation disabled
PATCH 1/1 powerpc: ppc64 address space capped at 32TB, mmap randomisation disabled
Commit fba2369e6ceb (mm: use vm_unmapped_area() on powerpc architecture) has a bug in slice_scan_available() where we compare an unsigned long (high_slices) against a shifted int. As a result, comparisons against the top 32 bits of high_slices (representing the top 32TB) always returns 0 and the top of our mmap region is clamped at 32TB
This also breaks mmap randomisation since the randomised address is always up near the top of the address space and it gets clamped down to 32TB.
---
arch/powerpc/mm/slice.c +1 -1
@@ -258,7 +258,7 @@ static bool slice_scan_available(unsigned long addr, slice =GET_HIGH_SLICE_INDEX (addr); *boundary_addr = (slice + end) ? ((slice + end) <<SLICE_HIGH_SHIFT ) :SLICE_LOW_TOP ; -return !!(available.high_slices & (1u << slice)); +return !!(available.high_slices & (1ul << slice)); } }
Quoted 20 lines · Anton Blanchard
Commit
fba2369e6ceb(mm: usevm_unmapped_area()on powerpc architecture) has a bug inslice_scan_available()where we compare an unsigned long (high_slices) against a shifted int. As a result, comparisons against the top 32 bits ofhigh_slices(representing the top 32TB) always returns 0 and the top of our mmap region is clamped at 32TBThis also breaks mmap randomisation since the randomised address is always up near the top of the address space and it gets clamped down to 32TB.
Cc [email protected] # v3.10+Signed-off-by Anton Blanchard <[email protected]>---
arch/powerpc/mm/slice.c +0
-0
arch/powerpc/mm/slice.c static bool slice_scan_available(unsigned long addr, · line 264
Quoted 8 lines
arch/powerpc/mm/slice.c +1
-1@@ -258,7 +258,7 @@ static bool slice_scan_available(unsigned long addr, slice =GET_HIGH_SLICE_INDEX (addr); *boundary_addr = (slice + end) ? ((slice + end) <<SLICE_HIGH_SHIFT ) :SLICE_LOW_TOP ; -return !!(available.high_slices & (1u << slice)); +return !!(available.high_slices & (1ul << slice)); } }
Good catch, sorry about that...
Signature · 2 lines
Michel "Walken" Lespinasse A program is never fully debugged until the last user dies.