summaryrefslogtreecommitdiffstats
path: root/mm
diff options
context:
space:
mode:
authorHuang Pei <huangpei@loongson.cn>2021-02-25 17:22:49 -0800
committerLinus Torvalds <torvalds@linux-foundation.org>2021-02-26 09:41:05 -0800
commitf685a533a7fab35c5d069dcd663f59c8e4171a75 (patch)
tree5962d85a0301fdb9c4e9077956f7ffd8e43f29aa /mm
parentdd23e8098f33a55b22b869bc7fc0a795ccbb9f87 (diff)
downloadlinux-f685a533a7fab35c5d069dcd663f59c8e4171a75.tar.gz
linux-f685a533a7fab35c5d069dcd663f59c8e4171a75.tar.bz2
linux-f685a533a7fab35c5d069dcd663f59c8e4171a75.zip
MIPS: make userspace mapping young by default
MIPS page fault path(except huge page) takes 3 exceptions (1 TLB Miss + 2 TLB Invalid), butthe second TLB Invalid exception is just triggered by __update_tlb from do_page_fault writing tlb without _PAGE_VALID set. With this patch, user space mapping prot is made young by default (with both _PAGE_VALID and _PAGE_YOUNG set), and it only take 1 TLB Miss + 1 TLB Invalid exception Remove pte_sw_mkyoung without polluting MM code and make page fault delay of MIPS on par with other architecture Link: https://lkml.kernel.org/r/20210204013942.8398-1-huangpei@loongson.cn Signed-off-by: Huang Pei <huangpei@loongson.cn> Reviewed-by: Nicholas Piggin <npiggin@gmail.com> Acked-by: <huangpei@loongson.cn> Acked-by: Thomas Bogendoerfer <tsbogend@alpha.franken.de> Cc: Christophe Leroy <christophe.leroy@csgroup.eu> Cc: <ambrosehua@gmail.com> Cc: Bibo Mao <maobibo@loongson.cn> Cc: Jiaxun Yang <jiaxun.yang@flygoat.com> Cc: Paul Burton <paulburton@kernel.org> Cc: Li Xuefeng <lixuefeng@loongson.cn> Cc: Yang Tiezhu <yangtiezhu@loongson.cn> Cc: Gao Juxin <gaojuxin@loongson.cn> Cc: Fuxin Zhang <zhangfx@lemote.com> Cc: Huacai Chen <chenhc@lemote.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'mm')
-rw-r--r--mm/memory.c4
1 files changed, 0 insertions, 4 deletions
diff --git a/mm/memory.c b/mm/memory.c
index 784249f3307b..c8e357627318 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2902,7 +2902,6 @@ static vm_fault_t wp_page_copy(struct vm_fault *vmf)
}
flush_cache_page(vma, vmf->address, pte_pfn(vmf->orig_pte));
entry = mk_pte(new_page, vma->vm_page_prot);
- entry = pte_sw_mkyoung(entry);
entry = maybe_mkwrite(pte_mkdirty(entry), vma);
/*
@@ -3560,7 +3559,6 @@ static vm_fault_t do_anonymous_page(struct vm_fault *vmf)
__SetPageUptodate(page);
entry = mk_pte(page, vma->vm_page_prot);
- entry = pte_sw_mkyoung(entry);
if (vma->vm_flags & VM_WRITE)
entry = pte_mkwrite(pte_mkdirty(entry));
@@ -3745,8 +3743,6 @@ void do_set_pte(struct vm_fault *vmf, struct page *page, unsigned long addr)
if (prefault && arch_wants_old_prefaulted_pte())
entry = pte_mkold(entry);
- else
- entry = pte_sw_mkyoung(entry);
if (write)
entry = maybe_mkwrite(pte_mkdirty(entry), vma);