Re: [f2fs-dev] [PATCH] f2fs: initialize page->private when using for our internal use

[Chao Yu]

On 2021/7/5 16:56, Jaegeuk Kim wrote:
> On 07/05, Chao Yu wrote:
> > On 2021/7/5 13:22, Jaegeuk Kim wrote:
> > > We need to guarantee it's initially zero. Otherwise, it'll hurt entire flag
> > > operations.
> >
> > Oops, I didn't get the point, shouldn't .private be zero after page was
> > just allocated by filesystem? What's the case we will encounter stall
> > private data left in page?
>
> I'm seeing f2fs_migrate_page() has the newpage with some value without Private
> flag. That causes a kernel panic later due to wrong private flag used in f2fs.

I'm not familiar with that part of codes, so Cc mm mailing list for help.

My question is newpage in .migrate_page() may contain non-zero value in .private
field but w/o setting PagePrivate flag, is it a normal case?

Thanks,

> > Cc Matthew Wilcox.
> >
> > Thanks,
> >
> > > Fixes: b763f3bedc2d ("f2fs: restructure f2fs page.private layout")
> > > Signed-off-by: Jaegeuk Kim <jaegeuk@xxxxxxxxxx>
> > > ---
> > >    fs/f2fs/data.c | 2 ++
> > >    fs/f2fs/f2fs.h | 5 ++++-
> > >    2 files changed, 6 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
> > > index 3a01a1b50104..d2cf48c5a2e4 100644
> > > --- a/fs/f2fs/data.c
> > > +++ b/fs/f2fs/data.c
> > > @@ -3819,6 +3819,8 @@ int f2fs_migrate_page(struct address_space *mapping,
> > >    		get_page(newpage);
> > >    	}
> > > +	/* guarantee to start from no stale private field */
> > > +	set_page_private(newpage, 0);
> > >    	if (PagePrivate(page)) {
> > >    		set_page_private(newpage, page_private(page));
> > >    		SetPagePrivate(newpage);
> > > diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
> > > index 65befc68d88e..ee8eb33e2c25 100644
> > > --- a/fs/f2fs/f2fs.h
> > > +++ b/fs/f2fs/f2fs.h
> > > @@ -1331,7 +1331,8 @@ enum {
> > >    #define PAGE_PRIVATE_GET_FUNC(name, flagname) \
> > >    static inline bool page_private_##name(struct page *page) \
> > >    { \
> > > -	return test_bit(PAGE_PRIVATE_NOT_POINTER, &page_private(page)) && \
> > > +	return PagePrivate(page) && \
> > > +		test_bit(PAGE_PRIVATE_NOT_POINTER, &page_private(page)) && \
> > >    		test_bit(PAGE_PRIVATE_##flagname, &page_private(page)); \
> > >    }
> > > @@ -1341,6 +1342,7 @@ static inline void set_page_private_##name(struct page *page) \
> > >    	if (!PagePrivate(page)) { \
> > >    		get_page(page); \
> > >    		SetPagePrivate(page); \
> > > +		set_page_private(page, 0); \
> > >    	} \
> > >    	set_bit(PAGE_PRIVATE_NOT_POINTER, &page_private(page)); \
> > >    	set_bit(PAGE_PRIVATE_##flagname, &page_private(page)); \
> > > @@ -1392,6 +1394,7 @@ static inline void set_page_private_data(struct page *page, unsigned long data)
> > >    	if (!PagePrivate(page)) {
> > >    		get_page(page);
> > >    		SetPagePrivate(page);
> > > +		set_page_private(page, 0);
> > >    	}
> > >    	set_bit(PAGE_PRIVATE_NOT_POINTER, &page_private(page));
> > >    	page_private(page) |= data << PAGE_PRIVATE_MAX;