2022-11-25 07:15:54

by Wupeng Ma

[permalink] [raw]
Subject: [PATCH -next 1/1] mm/memory-failure.c: Cleanup in unpoison_memory

From: Ma Wupeng <[email protected]>

If freeit it true, the value of ret must be zero, there is no need to
check the value of freeit after label unlock_mutex.

We can drop variable freeit to do this cleanup.

Signed-off-by: Ma Wupeng <[email protected]>
---
mm/memory-failure.c | 6 ++----
1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index 2e62940c7bae..c77a9e37e27e 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -2338,7 +2338,6 @@ int unpoison_memory(unsigned long pfn)
struct page *page;
struct page *p;
int ret = -EBUSY;
- int freeit = 0;
unsigned long count = 1;
bool huge = false;
static DEFINE_RATELIMIT_STATE(unpoison_rs, DEFAULT_RATELIMIT_INTERVAL,
@@ -2413,10 +2412,9 @@ int unpoison_memory(unsigned long pfn)
goto unlock_mutex;
}
}
- freeit = !!TestClearPageHWPoison(p);

put_page(page);
- if (freeit) {
+ if (TestClearPageHWPoison(p)) {
put_page(page);
ret = 0;
}
@@ -2424,7 +2422,7 @@ int unpoison_memory(unsigned long pfn)

unlock_mutex:
mutex_unlock(&mf_mutex);
- if (!ret || freeit) {
+ if (!ret) {
if (!huge)
num_poisoned_pages_sub(pfn, 1);
unpoison_pr_info("Unpoison: Software-unpoisoned page %#lx\n",
--
2.25.1


Subject: Re: [PATCH -next 1/1] mm/memory-failure.c: Cleanup in unpoison_memory

On Fri, Nov 25, 2022 at 02:54:44PM +0800, Wupeng Ma wrote:
> From: Ma Wupeng <[email protected]>
>
> If freeit it true, the value of ret must be zero, there is no need to
> check the value of freeit after label unlock_mutex.
>
> We can drop variable freeit to do this cleanup.
>
> Signed-off-by: Ma Wupeng <[email protected]>

Hi Wupeng,

> ---
> mm/memory-failure.c | 6 ++----
> 1 file changed, 2 insertions(+), 4 deletions(-)
>
> diff --git a/mm/memory-failure.c b/mm/memory-failure.c
> index 2e62940c7bae..c77a9e37e27e 100644
> --- a/mm/memory-failure.c
> +++ b/mm/memory-failure.c
> @@ -2338,7 +2338,6 @@ int unpoison_memory(unsigned long pfn)
> struct page *page;
> struct page *p;
> int ret = -EBUSY;
> - int freeit = 0;
> unsigned long count = 1;
> bool huge = false;
> static DEFINE_RATELIMIT_STATE(unpoison_rs, DEFAULT_RATELIMIT_INTERVAL,
> @@ -2413,10 +2412,9 @@ int unpoison_memory(unsigned long pfn)
> goto unlock_mutex;
> }
> }
> - freeit = !!TestClearPageHWPoison(p);
>
> put_page(page);
> - if (freeit) {
> + if (TestClearPageHWPoison(p)) {

This reorders put_page() and TestClearPageHWPoison(), but when we run
into this else block, the target page or hugepage should have refcount > 1,
so it does not cause any behavioral change. So I'm fine with it.

Looks good to me. Thank you for the patch.

Acked-by: Naoya Horiguchi <[email protected]>

> put_page(page);
> ret = 0;
> }
> @@ -2424,7 +2422,7 @@ int unpoison_memory(unsigned long pfn)
>
> unlock_mutex:
> mutex_unlock(&mf_mutex);
> - if (!ret || freeit) {
> + if (!ret) {
> if (!huge)
> num_poisoned_pages_sub(pfn, 1);
> unpoison_pr_info("Unpoison: Software-unpoisoned page %#lx\n",
> --
> 2.25.1

2022-11-26 02:00:07

by Miaohe Lin

[permalink] [raw]
Subject: Re: [PATCH -next 1/1] mm/memory-failure.c: Cleanup in unpoison_memory

On 2022/11/25 14:54, Wupeng Ma wrote:
> From: Ma Wupeng <[email protected]>
>
> If freeit it true, the value of ret must be zero, there is no need to

s/it/is/

> check the value of freeit after label unlock_mutex.
>
> We can drop variable freeit to do this cleanup.
>
> Signed-off-by: Ma Wupeng <[email protected]>

Reviewed-by: Miaohe Lin <[email protected]>

Thanks,
Miaohe Lin