project-base/server-skynet-source-3rd-jemalloc
2
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

ARM: Don't extend bit LG_VADDR to compute high address bits.

Browse Source 
In userspace ARM on Linux, zero-ing the high bits is the correct way to do this.
This doesn't fix the fact that we currently set LG_VADDR to 48 on ARM, when in
fact larger virtual address sizes are coming soon.  We'll cross that bridge when
we come to it.
This commit is contained in:
David Goldblatt 2017-09-29 13:54:08 -07:00 committed by David Goldblatt
parent 0720192a32
commit 7a8bc7172b
1 changed files with 12 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
include/jemalloc/internal/rtree.h
View File

@ -178,9 +178,21 @@ rtree_leaf_elm_bits_read(tsdn_t *tsdn, rtree_t *rtree, rtree_leaf_elm_t *elm,
JEMALLOC_ALWAYS_INLINE extent_t * JEMALLOC_ALWAYS_INLINE extent_t *
rtree_leaf_elm_bits_extent_get(uintptr_t bits) { rtree_leaf_elm_bits_extent_get(uintptr_t bits) {
# ifdef __aarch64__
/*
* aarch64 doesn't sign extend the highest virtual address bit to set
* the higher ones. Instead, the high bits gets zeroed.
*/
uintptr_t high_bit_mask = ((uintptr_t)1 << LG_VADDR) - 1;
/* Mask off the slab bit. */
uintptr_t low_bit_mask = ~(uintptr_t)1;
uintptr_t mask = high_bit_mask & low_bit_mask;
return (extent_t *)(bits & mask);
# else
/* Restore sign-extended high bits, mask slab bit. */ /* Restore sign-extended high bits, mask slab bit. */
return (extent_t *)((uintptr_t)((intptr_t)(bits << RTREE_NHIB) >> return (extent_t *)((uintptr_t)((intptr_t)(bits << RTREE_NHIB) >>
RTREE_NHIB) & ~((uintptr_t)0x1)); RTREE_NHIB) & ~((uintptr_t)0x1));
# endif
} }
JEMALLOC_ALWAYS_INLINE szind_t JEMALLOC_ALWAYS_INLINE szind_t