forked from Minki/linux
ff03b88467
The failing address alignment to HPAGE_MASK in do_exception(), for hugetlb faults, was useless from the beginning. With 2 GB hugepage support it became wrong, but w/o further negative impact. Now it could have negative performance impact because it breaks the cacheline optimization for process_huge_page(). Therefore, remove it. Note that we still have failing address alignment by HW to PAGE_SIZE, for all page faults, not just hugetlb faults. So this patch will not fix UFFD_FEATURE_EXACT_ADDRESS for userfaultfd handling. It will just move the failing address for hugetlb faults a bit closer to the real address, at 4K page granularity, similar to normal page faults. Reviewed-by: Heiko Carstens <hca@linux.ibm.com> Signed-off-by: Gerald Schaefer <gerald.schaefer@linux.ibm.com> Signed-off-by: Vasily Gorbik <gor@linux.ibm.com> |
||
|---|---|---|
| .. | ||
| cmm.c | ||
| dump_pagetables.c | ||
| extable.c | ||
| extmem.c | ||
| fault.c | ||
| gmap.c | ||
| hugetlbpage.c | ||
| init.c | ||
| kasan_init.c | ||
| maccess.c | ||
| Makefile | ||
| mmap.c | ||
| page-states.c | ||
| pageattr.c | ||
| pgalloc.c | ||
| pgtable.c | ||
| vmem.c | ||