get_new_data_page should release inode page when we encounter any
error in its procedure, but there is one error path didn't cover
this, fix it.
Signed-off-by: Chao Yu <[email protected]>
---
fs/f2fs/data.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 08dfdc6..ea8898b 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -397,8 +397,10 @@ struct page *get_new_data_page(struct inode *inode,
int err;
repeat:
page = grab_cache_page(mapping, index);
- if (!page)
+ if (!page) {
+ f2fs_put_page(ipage, 1);
return ERR_PTR(-ENOMEM);
+ }
set_new_dnode(&dn, inode, ipage, NULL, 0);
err = f2fs_reserve_block(&dn, index);
--
2.4.2
Hi Chao,
On Thu, Jul 09, 2015 at 06:20:08PM +0800, Chao Yu wrote:
> get_new_data_page should release inode page when we encounter any
> error in its procedure, but there is one error path didn't cover
> this, fix it.
Nice catch.
But, I think we should fix its caller:
in init_inode_metadata(),
err = make_empty_dir();
if (err)
goto put_error;
---------------
Thanks,
>
> Signed-off-by: Chao Yu <[email protected]>
> ---
> fs/f2fs/data.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
> index 08dfdc6..ea8898b 100644
> --- a/fs/f2fs/data.c
> +++ b/fs/f2fs/data.c
> @@ -397,8 +397,10 @@ struct page *get_new_data_page(struct inode *inode,
> int err;
> repeat:
> page = grab_cache_page(mapping, index);
> - if (!page)
> + if (!page) {
> + f2fs_put_page(ipage, 1);
> return ERR_PTR(-ENOMEM);
> + }
>
> set_new_dnode(&dn, inode, ipage, NULL, 0);
> err = f2fs_reserve_block(&dn, index);
> --
> 2.4.2
Hi Jaegeuk,
> -----Original Message-----
> From: Jaegeuk Kim [mailto:[email protected]]
> Sent: Saturday, July 11, 2015 8:17 AM
> To: Chao Yu; Chao Yu
> Cc: [email protected]; [email protected];
> [email protected]; [email protected]
> Subject: Re: [f2fs-dev] [PATCH] f2fs: fix to release inode page in get_new_data_page
>
> Hi Chao,
>
> On Thu, Jul 09, 2015 at 06:20:08PM +0800, Chao Yu wrote:
> > get_new_data_page should release inode page when we encounter any
> > error in its procedure, but there is one error path didn't cover
> > this, fix it.
>
> Nice catch.
> But, I think we should fix its caller:
>
> in init_inode_metadata(),
> err = make_empty_dir();
> if (err)
> goto put_error;
> ---------------
Previously, I fixed in the same way, but I got an oops when I test the
patch with xfstest suit, it shows we will meet an error in this call
path IIRC:
->f2fs_mkdir
->__f2fs_add_link
->init_inode_metadata
->make_empty_dir
->get_new_data_page
->f2fs_reserve_block
->reserve_new_block
->inc_valid_block_count
return -ENOSPC;
->f2fs_put_dnode
->f2fs_put_page(ipage, 1)
put_error:
->f2fs_put_page(ipage, 1);
f2fs_bug_on(F2FS_P_SB(page), !PageLocked(page));
And I don't think we should change error handling method of f2fs_put_dnode
for just fixing this issue.
How do you think?
Thanks,
> Thanks,
>
> >
> > Signed-off-by: Chao Yu <[email protected]>
> > ---
> > fs/f2fs/data.c | 4 +++-
> > 1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
> > index 08dfdc6..ea8898b 100644
> > --- a/fs/f2fs/data.c
> > +++ b/fs/f2fs/data.c
> > @@ -397,8 +397,10 @@ struct page *get_new_data_page(struct inode *inode,
> > int err;
> > repeat:
> > page = grab_cache_page(mapping, index);
> > - if (!page)
> > + if (!page) {
> > + f2fs_put_page(ipage, 1);
> > return ERR_PTR(-ENOMEM);
> > + }
> >
> > set_new_dnode(&dn, inode, ipage, NULL, 0);
> > err = f2fs_reserve_block(&dn, index);
> > --
> > 2.4.2
>
> ------------------------------------------------------------------------------
> Don't Limit Your Business. Reach for the Cloud.
> GigeNET's Cloud Solutions provide you with the tools and support that
> you need to offload your IT needs and focus on growing your business.
> Configured For All Businesses. Start Your Cloud Today.
> https://www.gigenetcloud.com/
> _______________________________________________
> Linux-f2fs-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
Hi Chao,
On Sat, Jul 11, 2015 at 10:02:51AM +0800, Chao Yu wrote:
> Hi Jaegeuk,
>
> > -----Original Message-----
> > From: Jaegeuk Kim [mailto:[email protected]]
> > Sent: Saturday, July 11, 2015 8:17 AM
> > To: Chao Yu; Chao Yu
> > Cc: [email protected]; [email protected];
> > [email protected]; [email protected]
> > Subject: Re: [f2fs-dev] [PATCH] f2fs: fix to release inode page in get_new_data_page
> >
> > Hi Chao,
> >
> > On Thu, Jul 09, 2015 at 06:20:08PM +0800, Chao Yu wrote:
> > > get_new_data_page should release inode page when we encounter any
> > > error in its procedure, but there is one error path didn't cover
> > > this, fix it.
> >
> > Nice catch.
> > But, I think we should fix its caller:
> >
> > in init_inode_metadata(),
> > err = make_empty_dir();
> > if (err)
> > goto put_error;
> > ---------------
>
> Previously, I fixed in the same way, but I got an oops when I test the
> patch with xfstest suit, it shows we will meet an error in this call
> path IIRC:
>
> ->f2fs_mkdir
> ->__f2fs_add_link
> ->init_inode_metadata
> ->make_empty_dir
> ->get_new_data_page
> ->f2fs_reserve_block
> ->reserve_new_block
> ->inc_valid_block_count
> return -ENOSPC;
> ->f2fs_put_dnode
> ->f2fs_put_page(ipage, 1)
>
> put_error:
> ->f2fs_put_page(ipage, 1);
> f2fs_bug_on(F2FS_P_SB(page), !PageLocked(page));
>
> And I don't think we should change error handling method of f2fs_put_dnode
> for just fixing this issue.
>
> How do you think?
Indeed. I cannot think about other clean way for now.
Instead, how about adding this description in the patch and some comments in
the codes?
Thanks,
> -----Original Message-----
> From: Jaegeuk Kim [mailto:[email protected]]
> Sent: Tuesday, July 14, 2015 7:26 AM
> To: Chao Yu
> Cc: [email protected]; [email protected]
> Subject: Re: [f2fs-dev] [PATCH] f2fs: fix to release inode page in get_new_data_page
[snip]
> > And I don't think we should change error handling method of f2fs_put_dnode
> > for just fixing this issue.
> >
> > How do you think?
>
> Indeed. I cannot think about other clean way for now.
> Instead, how about adding this description in the patch and some comments in
> the codes?
OK, Please help to review the following patch. :)
Thanks,