All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] mm: Functions may simplify the use of return values
@ 2022-05-07 15:06 Li kunyu
  2022-05-09  7:19 ` Peter Zijlstra
  0 siblings, 1 reply; 3+ messages in thread
From: Li kunyu @ 2022-05-07 15:06 UTC (permalink / raw)
  To: dave.hansen, luto, peterz, tglx, mingo, bp, x86, hpa, akpm
  Cc: linux-kernel, linux-mm, Li kunyu

p4d_clear_huge may be optimized for void return type and function usage.
vunmap_p4d_range function saves a few steps here.

Signed-off-by: Li kunyu <kunyu@nfschina.com>
---
 arch/x86/mm/pgtable.c   |  3 +--
 include/linux/pgtable.h | 12 +++---------
 mm/vmalloc.c            |  7 ++-----
 3 files changed, 6 insertions(+), 16 deletions(-)

diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c
index 3481b35cb4ec..9222b6832c36 100644
--- a/arch/x86/mm/pgtable.c
+++ b/arch/x86/mm/pgtable.c
@@ -676,9 +676,8 @@ int p4d_set_huge(p4d_t *p4d, phys_addr_t addr, pgprot_t prot)
  *
  * No 512GB pages yet -- always return 0
  */
-int p4d_clear_huge(p4d_t *p4d)
+void p4d_clear_huge(p4d_t *p4d)
 {
-	return 0;
 }
 #endif
 
diff --git a/include/linux/pgtable.h b/include/linux/pgtable.h
index f4f4077b97aa..5448b5b10241 100644
--- a/include/linux/pgtable.h
+++ b/include/linux/pgtable.h
@@ -1394,16 +1394,13 @@ static inline int pmd_protnone(pmd_t pmd)
 
 #ifndef __PAGETABLE_P4D_FOLDED
 int p4d_set_huge(p4d_t *p4d, phys_addr_t addr, pgprot_t prot);
-int p4d_clear_huge(p4d_t *p4d);
+void p4d_clear_huge(p4d_t *p4d);
 #else
 static inline int p4d_set_huge(p4d_t *p4d, phys_addr_t addr, pgprot_t prot)
 {
 	return 0;
 }
-static inline int p4d_clear_huge(p4d_t *p4d)
-{
-	return 0;
-}
+static inline void p4d_clear_huge(p4d_t *p4d) { }
 #endif /* !__PAGETABLE_P4D_FOLDED */
 
 int pud_set_huge(pud_t *pud, phys_addr_t addr, pgprot_t prot);
@@ -1426,10 +1423,7 @@ static inline int pmd_set_huge(pmd_t *pmd, phys_addr_t addr, pgprot_t prot)
 {
 	return 0;
 }
-static inline int p4d_clear_huge(p4d_t *p4d)
-{
-	return 0;
-}
+static inline void p4d_clear_huge(p4d_t *p4d) { }
 static inline int pud_clear_huge(pud_t *pud)
 {
 	return 0;
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index cadfbb5155ea..fa915f3cb75e 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -389,18 +389,15 @@ static void vunmap_p4d_range(pgd_t *pgd, unsigned long addr, unsigned long end,
 {
 	p4d_t *p4d;
 	unsigned long next;
-	int cleared;
 
 	p4d = p4d_offset(pgd, addr);
 	do {
 		next = p4d_addr_end(addr, end);
 
-		cleared = p4d_clear_huge(p4d);
-		if (cleared || p4d_bad(*p4d))
+		p4d_clear_huge(p4d);
+		if (p4d_bad(*p4d))
 			*mask |= PGTBL_P4D_MODIFIED;
 
-		if (cleared)
-			continue;
 		if (p4d_none_or_clear_bad(p4d))
 			continue;
 		vunmap_pud_range(p4d, addr, next, mask);
-- 
2.18.2


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH] mm: Functions may simplify the use of return values
  2022-05-07 15:06 [PATCH] mm: Functions may simplify the use of return values Li kunyu
@ 2022-05-09  7:19 ` Peter Zijlstra
  2022-05-09 11:39   ` Li kunyu
  0 siblings, 1 reply; 3+ messages in thread
From: Peter Zijlstra @ 2022-05-09  7:19 UTC (permalink / raw)
  To: Li kunyu
  Cc: dave.hansen, luto, tglx, mingo, bp, x86, hpa, akpm, linux-kernel,
	linux-mm

On Sat, May 07, 2022 at 11:06:30PM +0800, Li kunyu wrote:
> p4d_clear_huge may be optimized for void return type and function usage.
> vunmap_p4d_range function saves a few steps here.

This makes it different from all the other p??_clear_huge() functions,
which seems like an exceedingly bad idea.

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH] mm: Functions may simplify the use of return values
  2022-05-09  7:19 ` Peter Zijlstra
@ 2022-05-09 11:39   ` Li kunyu
  0 siblings, 0 replies; 3+ messages in thread
From: Li kunyu @ 2022-05-09 11:39 UTC (permalink / raw)
  To: peterz
  Cc: akpm, bp, dave.hansen, hpa, kunyu, linux-kernel, linux-mm, luto,
	mingo, tglx, x86


I estimate that this function should not add code in the next year or two.
I consider and optimize this function in terms of code execution efficiency.
There may be a better way.


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2022-05-09 11:39 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-05-07 15:06 [PATCH] mm: Functions may simplify the use of return values Li kunyu
2022-05-09  7:19 ` Peter Zijlstra
2022-05-09 11:39   ` Li kunyu

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.