* [PATCH 1/2] mm: Move lru_to_page to mm.h @ 2018-11-29 7:52 Nikolay Borisov 2018-11-29 7:52 ` [PATCH 2/2] fs: Don't open-code lru_to_page Nikolay Borisov 0 siblings, 1 reply; 9+ messages in thread From: Nikolay Borisov @ 2018-11-29 7:52 UTC (permalink / raw) To: linux-kernel Cc: Nikolay Borisov, Andrew Morton, Matthew Wilcox, Michal Hocko, Dan Williams, Kirill A. Shutemov, Pavel Tatashin, Souptick Joarder, Logan Gunthorpe, Keith Busch, Tony Luck, Ingo Molnar, linux-mm There are multiple places in the kernel which opencode this helper, this patch moves it to the more generic mm.h header in preparation for using it. No functional changes. Signed-off-by: Nikolay Borisov <nborisov@suse.com> --- include/linux/mm.h | 2 ++ include/linux/mm_inline.h | 3 --- 2 files changed, 2 insertions(+), 3 deletions(-) diff --git a/include/linux/mm.h b/include/linux/mm.h index 5411de93a363..47b4aa5bba93 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -146,6 +146,8 @@ extern int overcommit_kbytes_handler(struct ctl_table *, int, void __user *, /* test whether an address (unsigned long or pointer) is aligned to PAGE_SIZE */ #define PAGE_ALIGNED(addr) IS_ALIGNED((unsigned long)(addr), PAGE_SIZE) +#define lru_to_page(head) (list_entry((head)->prev, struct page, lru)) + /* * Linux kernel virtual memory manager primitives. * The idea being to have a "virtual" mm in the same way diff --git a/include/linux/mm_inline.h b/include/linux/mm_inline.h index 10191c28fc04..04ec454d44ce 100644 --- a/include/linux/mm_inline.h +++ b/include/linux/mm_inline.h @@ -124,7 +124,4 @@ static __always_inline enum lru_list page_lru(struct page *page) } return lru; } - -#define lru_to_page(head) (list_entry((head)->prev, struct page, lru)) - #endif -- 2.17.1 ^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH 2/2] fs: Don't open-code lru_to_page 2018-11-29 7:52 [PATCH 1/2] mm: Move lru_to_page to mm.h Nikolay Borisov @ 2018-11-29 7:52 ` Nikolay Borisov 2018-11-29 8:18 ` Michal Hocko 2018-11-30 9:11 ` Yan, Zheng 0 siblings, 2 replies; 9+ messages in thread From: Nikolay Borisov @ 2018-11-29 7:52 UTC (permalink / raw) To: linux-kernel Cc: Nikolay Borisov, David Howells, Chris Mason, Josef Bacik, David Sterba, Yan, Zheng, Sage Weil, Ilya Dryomov, Steve French, Theodore Ts'o, Andreas Dilger, Mark Fasheh, Joel Becker, Mike Marshall, Martin Brandenburg, Andrew Morton, Mike Rapoport, Matthew Wilcox, Randy Dunlap, YueHaibing, Shakeel Butt, Dan Williams, Michal Hocko, linux-afs, linux-btrfs, ceph-devel, linux-cifs, samba-technical, linux-ext4, ocfs2-devel, devel, linux-mm There are a bunch of filesystems which essentially open-code lru_to_page helper. Change them to using the helper. No functional changes. Signed-off-by: Nikolay Borisov <nborisov@suse.com> --- Since this is a mostly mechanical change I've actually batched all of them in a single patch. fs/afs/file.c | 5 +++-- fs/btrfs/extent_io.c | 2 +- fs/ceph/addr.c | 5 ++--- fs/cifs/file.c | 3 ++- fs/ext4/readpage.c | 2 +- fs/ocfs2/aops.c | 3 ++- fs/orangefs/inode.c | 2 +- mm/swap.c | 2 +- 8 files changed, 13 insertions(+), 11 deletions(-) diff --git a/fs/afs/file.c b/fs/afs/file.c index d6bc3f5d784b..323ae9912203 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -17,6 +17,7 @@ #include <linux/writeback.h> #include <linux/gfp.h> #include <linux/task_io_accounting_ops.h> +#include <linux/mm.h> #include "internal.h" static int afs_file_mmap(struct file *file, struct vm_area_struct *vma); @@ -441,7 +442,7 @@ static int afs_readpages_one(struct file *file, struct address_space *mapping, /* Count the number of contiguous pages at the front of the list. Note * that the list goes prev-wards rather than next-wards. */ - first = list_entry(pages->prev, struct page, lru); + first = lru_to_page(pages); index = first->index + 1; n = 1; for (p = first->lru.prev; p != pages; p = p->prev) { @@ -473,7 +474,7 @@ static int afs_readpages_one(struct file *file, struct address_space *mapping, * page at the end of the file. */ do { - page = list_entry(pages->prev, struct page, lru); + page = lru_to_page(pages); list_del(&page->lru); index = page->index; if (add_to_page_cache_lru(page, mapping, index, diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 19f4b8fd654f..8332c5f4b1c3 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -4104,7 +4104,7 @@ int extent_readpages(struct address_space *mapping, struct list_head *pages, u64 prev_em_start = (u64)-1; for (page_idx = 0; page_idx < nr_pages; page_idx++) { - page = list_entry(pages->prev, struct page, lru); + page = lru_to_page(pages); prefetchw(&page->flags); list_del(&page->lru); diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 8eade7a993c1..5d0c05e288cc 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -306,7 +306,7 @@ static int start_read(struct inode *inode, struct ceph_rw_context *rw_ctx, struct ceph_osd_client *osdc = &ceph_inode_to_client(inode)->client->osdc; struct ceph_inode_info *ci = ceph_inode(inode); - struct page *page = list_entry(page_list->prev, struct page, lru); + struct page *page = lru_to_page(page_list); struct ceph_vino vino; struct ceph_osd_request *req; u64 off; @@ -333,8 +333,7 @@ static int start_read(struct inode *inode, struct ceph_rw_context *rw_ctx, if (got) ceph_put_cap_refs(ci, got); while (!list_empty(page_list)) { - page = list_entry(page_list->prev, - struct page, lru); + page = lru_to_page(page_list); list_del(&page->lru); put_page(page); } diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 74c33d5fafc8..b16a4d887d17 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -33,6 +33,7 @@ #include <linux/mount.h> #include <linux/slab.h> #include <linux/swap.h> +#include <linux/mm.h> #include <asm/div64.h> #include "cifsfs.h" #include "cifspdu.h" @@ -3975,7 +3976,7 @@ readpages_get_pages(struct address_space *mapping, struct list_head *page_list, INIT_LIST_HEAD(tmplist); - page = list_entry(page_list->prev, struct page, lru); + page = lru_to_page(page_list); /* * Lock the page and put it in the cache. Since no one else diff --git a/fs/ext4/readpage.c b/fs/ext4/readpage.c index f461d75ac049..6aa282ee455a 100644 --- a/fs/ext4/readpage.c +++ b/fs/ext4/readpage.c @@ -128,7 +128,7 @@ int ext4_mpage_readpages(struct address_space *mapping, prefetchw(&page->flags); if (pages) { - page = list_entry(pages->prev, struct page, lru); + page = lru_to_page(pages); list_del(&page->lru); if (add_to_page_cache_lru(page, mapping, page->index, readahead_gfp_mask(mapping))) diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index eb1ce30412dc..832c1759a09a 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -30,6 +30,7 @@ #include <linux/quotaops.h> #include <linux/blkdev.h> #include <linux/uio.h> +#include <linux/mm.h> #include <cluster/masklog.h> @@ -397,7 +398,7 @@ static int ocfs2_readpages(struct file *filp, struct address_space *mapping, * Check whether a remote node truncated this file - we just * drop out in that case as it's not worth handling here. */ - last = list_entry(pages->prev, struct page, lru); + last = lru_to_page(pages); start = (loff_t)last->index << PAGE_SHIFT; if (start >= i_size_read(inode)) goto out_unlock; diff --git a/fs/orangefs/inode.c b/fs/orangefs/inode.c index fe53381b26b1..f038235c64bd 100644 --- a/fs/orangefs/inode.c +++ b/fs/orangefs/inode.c @@ -77,7 +77,7 @@ static int orangefs_readpages(struct file *file, for (page_idx = 0; page_idx < nr_pages; page_idx++) { struct page *page; - page = list_entry(pages->prev, struct page, lru); + page = lru_to_page(pages); list_del(&page->lru); if (!add_to_page_cache(page, mapping, diff --git a/mm/swap.c b/mm/swap.c index aa483719922e..20b9e9d99652 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -126,7 +126,7 @@ void put_pages_list(struct list_head *pages) while (!list_empty(pages)) { struct page *victim; - victim = list_entry(pages->prev, struct page, lru); + victim = lru_to_page(pages); list_del(&victim->lru); put_page(victim); } -- 2.17.1 ^ permalink raw reply related [flat|nested] 9+ messages in thread
* Re: [PATCH 2/2] fs: Don't open-code lru_to_page 2018-11-29 7:52 ` [PATCH 2/2] fs: Don't open-code lru_to_page Nikolay Borisov @ 2018-11-29 8:18 ` Michal Hocko 2018-11-29 8:50 ` Nikolay Borisov 2018-11-29 10:12 ` David Hildenbrand 2018-11-30 9:11 ` Yan, Zheng 1 sibling, 2 replies; 9+ messages in thread From: Michal Hocko @ 2018-11-29 8:18 UTC (permalink / raw) To: Nikolay Borisov Cc: linux-kernel, David Howells, Chris Mason, Josef Bacik, David Sterba, Yan, Zheng, Sage Weil, Ilya Dryomov, Steve French, Theodore Ts'o, Andreas Dilger, Mark Fasheh, Joel Becker, Mike Marshall, Martin Brandenburg, Andrew Morton, Mike Rapoport, Matthew Wilcox, Randy Dunlap, YueHaibing, Shakeel Butt, Dan Williams, linux-afs, linux-btrfs, ceph-devel, linux-cifs, samba-technical, linux-ext4, ocfs2-devel, devel, linux-mm On Thu 29-11-18 09:52:57, Nikolay Borisov wrote: > There are a bunch of filesystems which essentially open-code lru_to_page > helper. Change them to using the helper. No functional changes. I would just squash the two into a single patch. It makes the first one more obvious. Or is there any reason to have them separate? > Signed-off-by: Nikolay Borisov <nborisov@suse.com> Acked-by: Michal Hocko <mhocko@suse.com> > --- > > Since this is a mostly mechanical change I've actually batched all of them in > a single patch. > > fs/afs/file.c | 5 +++-- > fs/btrfs/extent_io.c | 2 +- > fs/ceph/addr.c | 5 ++--- > fs/cifs/file.c | 3 ++- > fs/ext4/readpage.c | 2 +- > fs/ocfs2/aops.c | 3 ++- > fs/orangefs/inode.c | 2 +- > mm/swap.c | 2 +- > 8 files changed, 13 insertions(+), 11 deletions(-) > > diff --git a/fs/afs/file.c b/fs/afs/file.c > index d6bc3f5d784b..323ae9912203 100644 > --- a/fs/afs/file.c > +++ b/fs/afs/file.c > @@ -17,6 +17,7 @@ > #include <linux/writeback.h> > #include <linux/gfp.h> > #include <linux/task_io_accounting_ops.h> > +#include <linux/mm.h> > #include "internal.h" > > static int afs_file_mmap(struct file *file, struct vm_area_struct *vma); > @@ -441,7 +442,7 @@ static int afs_readpages_one(struct file *file, struct address_space *mapping, > /* Count the number of contiguous pages at the front of the list. Note > * that the list goes prev-wards rather than next-wards. > */ > - first = list_entry(pages->prev, struct page, lru); > + first = lru_to_page(pages); > index = first->index + 1; > n = 1; > for (p = first->lru.prev; p != pages; p = p->prev) { > @@ -473,7 +474,7 @@ static int afs_readpages_one(struct file *file, struct address_space *mapping, > * page at the end of the file. > */ > do { > - page = list_entry(pages->prev, struct page, lru); > + page = lru_to_page(pages); > list_del(&page->lru); > index = page->index; > if (add_to_page_cache_lru(page, mapping, index, > diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c > index 19f4b8fd654f..8332c5f4b1c3 100644 > --- a/fs/btrfs/extent_io.c > +++ b/fs/btrfs/extent_io.c > @@ -4104,7 +4104,7 @@ int extent_readpages(struct address_space *mapping, struct list_head *pages, > u64 prev_em_start = (u64)-1; > > for (page_idx = 0; page_idx < nr_pages; page_idx++) { > - page = list_entry(pages->prev, struct page, lru); > + page = lru_to_page(pages); > > prefetchw(&page->flags); > list_del(&page->lru); > diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c > index 8eade7a993c1..5d0c05e288cc 100644 > --- a/fs/ceph/addr.c > +++ b/fs/ceph/addr.c > @@ -306,7 +306,7 @@ static int start_read(struct inode *inode, struct ceph_rw_context *rw_ctx, > struct ceph_osd_client *osdc = > &ceph_inode_to_client(inode)->client->osdc; > struct ceph_inode_info *ci = ceph_inode(inode); > - struct page *page = list_entry(page_list->prev, struct page, lru); > + struct page *page = lru_to_page(page_list); > struct ceph_vino vino; > struct ceph_osd_request *req; > u64 off; > @@ -333,8 +333,7 @@ static int start_read(struct inode *inode, struct ceph_rw_context *rw_ctx, > if (got) > ceph_put_cap_refs(ci, got); > while (!list_empty(page_list)) { > - page = list_entry(page_list->prev, > - struct page, lru); > + page = lru_to_page(page_list); > list_del(&page->lru); > put_page(page); > } > diff --git a/fs/cifs/file.c b/fs/cifs/file.c > index 74c33d5fafc8..b16a4d887d17 100644 > --- a/fs/cifs/file.c > +++ b/fs/cifs/file.c > @@ -33,6 +33,7 @@ > #include <linux/mount.h> > #include <linux/slab.h> > #include <linux/swap.h> > +#include <linux/mm.h> > #include <asm/div64.h> > #include "cifsfs.h" > #include "cifspdu.h" > @@ -3975,7 +3976,7 @@ readpages_get_pages(struct address_space *mapping, struct list_head *page_list, > > INIT_LIST_HEAD(tmplist); > > - page = list_entry(page_list->prev, struct page, lru); > + page = lru_to_page(page_list); > > /* > * Lock the page and put it in the cache. Since no one else > diff --git a/fs/ext4/readpage.c b/fs/ext4/readpage.c > index f461d75ac049..6aa282ee455a 100644 > --- a/fs/ext4/readpage.c > +++ b/fs/ext4/readpage.c > @@ -128,7 +128,7 @@ int ext4_mpage_readpages(struct address_space *mapping, > > prefetchw(&page->flags); > if (pages) { > - page = list_entry(pages->prev, struct page, lru); > + page = lru_to_page(pages); > list_del(&page->lru); > if (add_to_page_cache_lru(page, mapping, page->index, > readahead_gfp_mask(mapping))) > diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c > index eb1ce30412dc..832c1759a09a 100644 > --- a/fs/ocfs2/aops.c > +++ b/fs/ocfs2/aops.c > @@ -30,6 +30,7 @@ > #include <linux/quotaops.h> > #include <linux/blkdev.h> > #include <linux/uio.h> > +#include <linux/mm.h> > > #include <cluster/masklog.h> > > @@ -397,7 +398,7 @@ static int ocfs2_readpages(struct file *filp, struct address_space *mapping, > * Check whether a remote node truncated this file - we just > * drop out in that case as it's not worth handling here. > */ > - last = list_entry(pages->prev, struct page, lru); > + last = lru_to_page(pages); > start = (loff_t)last->index << PAGE_SHIFT; > if (start >= i_size_read(inode)) > goto out_unlock; > diff --git a/fs/orangefs/inode.c b/fs/orangefs/inode.c > index fe53381b26b1..f038235c64bd 100644 > --- a/fs/orangefs/inode.c > +++ b/fs/orangefs/inode.c > @@ -77,7 +77,7 @@ static int orangefs_readpages(struct file *file, > for (page_idx = 0; page_idx < nr_pages; page_idx++) { > struct page *page; > > - page = list_entry(pages->prev, struct page, lru); > + page = lru_to_page(pages); > list_del(&page->lru); > if (!add_to_page_cache(page, > mapping, > diff --git a/mm/swap.c b/mm/swap.c > index aa483719922e..20b9e9d99652 100644 > --- a/mm/swap.c > +++ b/mm/swap.c > @@ -126,7 +126,7 @@ void put_pages_list(struct list_head *pages) > while (!list_empty(pages)) { > struct page *victim; > > - victim = list_entry(pages->prev, struct page, lru); > + victim = lru_to_page(pages); > list_del(&victim->lru); > put_page(victim); > } > -- > 2.17.1 -- Michal Hocko SUSE Labs ^ permalink raw reply [flat|nested] 9+ messages in thread
* [PATCH 2/2] fs: Don't open-code lru_to_page 2018-11-29 8:18 ` Michal Hocko @ 2018-11-29 8:50 ` Nikolay Borisov 2018-11-29 9:04 ` Michal Hocko ` (2 more replies) 2018-11-29 10:12 ` David Hildenbrand 1 sibling, 3 replies; 9+ messages in thread From: Nikolay Borisov @ 2018-11-29 8:50 UTC (permalink / raw) To: Michal Hocko Cc: linux-kernel, David Howells, Chris Mason, Josef Bacik, David Sterba, Yan, Zheng, Sage Weil, Ilya Dryomov, Steve French, Theodore Ts'o, Andreas Dilger, Mark Fasheh, Joel Becker, Mike Marshall, Martin Brandenburg, Andrew Morton, Mike Rapoport, Matthew Wilcox, Randy Dunlap, YueHaibing, Shakeel Butt, Dan Williams, linux-afs, linux-btrfs, ceph-devel, linux-cifs, samba-technical, linux-ext4, ocfs2-devel, devel, linux-mm On 29.11.18 г. 10:18 ч., Michal Hocko wrote: > On Thu 29-11-18 09:52:57, Nikolay Borisov wrote: >> There are a bunch of filesystems which essentially open-code lru_to_page >> helper. Change them to using the helper. No functional changes. > > I would just squash the two into a single patch. It makes the first one > more obvious. Or is there any reason to have them separate? No reason, just didn't know how people would react so that's why I chose to send as two separate. If I squash them who would be the best person to take them ? > >> Signed-off-by: Nikolay Borisov <nborisov@suse.com> > > Acked-by: Michal Hocko <mhocko@suse.com> > >> --- >> >> Since this is a mostly mechanical change I've actually batched all of them in >> a single patch. >> >> fs/afs/file.c | 5 +++-- >> fs/btrfs/extent_io.c | 2 +- >> fs/ceph/addr.c | 5 ++--- >> fs/cifs/file.c | 3 ++- >> fs/ext4/readpage.c | 2 +- >> fs/ocfs2/aops.c | 3 ++- >> fs/orangefs/inode.c | 2 +- >> mm/swap.c | 2 +- >> 8 files changed, 13 insertions(+), 11 deletions(-) >> >> diff --git a/fs/afs/file.c b/fs/afs/file.c >> index d6bc3f5d784b..323ae9912203 100644 >> --- a/fs/afs/file.c >> +++ b/fs/afs/file.c >> @@ -17,6 +17,7 @@ >> #include <linux/writeback.h> >> #include <linux/gfp.h> >> #include <linux/task_io_accounting_ops.h> >> +#include <linux/mm.h> >> #include "internal.h" >> >> static int afs_file_mmap(struct file *file, struct vm_area_struct *vma); >> @@ -441,7 +442,7 @@ static int afs_readpages_one(struct file *file, struct address_space *mapping, >> /* Count the number of contiguous pages at the front of the list. Note >> * that the list goes prev-wards rather than next-wards. >> */ >> - first = list_entry(pages->prev, struct page, lru); >> + first = lru_to_page(pages); >> index = first->index + 1; >> n = 1; >> for (p = first->lru.prev; p != pages; p = p->prev) { >> @@ -473,7 +474,7 @@ static int afs_readpages_one(struct file *file, struct address_space *mapping, >> * page at the end of the file. >> */ >> do { >> - page = list_entry(pages->prev, struct page, lru); >> + page = lru_to_page(pages); >> list_del(&page->lru); >> index = page->index; >> if (add_to_page_cache_lru(page, mapping, index, >> diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c >> index 19f4b8fd654f..8332c5f4b1c3 100644 >> --- a/fs/btrfs/extent_io.c >> +++ b/fs/btrfs/extent_io.c >> @@ -4104,7 +4104,7 @@ int extent_readpages(struct address_space *mapping, struct list_head *pages, >> u64 prev_em_start = (u64)-1; >> >> for (page_idx = 0; page_idx < nr_pages; page_idx++) { >> - page = list_entry(pages->prev, struct page, lru); >> + page = lru_to_page(pages); >> >> prefetchw(&page->flags); >> list_del(&page->lru); >> diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c >> index 8eade7a993c1..5d0c05e288cc 100644 >> --- a/fs/ceph/addr.c >> +++ b/fs/ceph/addr.c >> @@ -306,7 +306,7 @@ static int start_read(struct inode *inode, struct ceph_rw_context *rw_ctx, >> struct ceph_osd_client *osdc = >> &ceph_inode_to_client(inode)->client->osdc; >> struct ceph_inode_info *ci = ceph_inode(inode); >> - struct page *page = list_entry(page_list->prev, struct page, lru); >> + struct page *page = lru_to_page(page_list); >> struct ceph_vino vino; >> struct ceph_osd_request *req; >> u64 off; >> @@ -333,8 +333,7 @@ static int start_read(struct inode *inode, struct ceph_rw_context *rw_ctx, >> if (got) >> ceph_put_cap_refs(ci, got); >> while (!list_empty(page_list)) { >> - page = list_entry(page_list->prev, >> - struct page, lru); >> + page = lru_to_page(page_list); >> list_del(&page->lru); >> put_page(page); >> } >> diff --git a/fs/cifs/file.c b/fs/cifs/file.c >> index 74c33d5fafc8..b16a4d887d17 100644 >> --- a/fs/cifs/file.c >> +++ b/fs/cifs/file.c >> @@ -33,6 +33,7 @@ >> #include <linux/mount.h> >> #include <linux/slab.h> >> #include <linux/swap.h> >> +#include <linux/mm.h> >> #include <asm/div64.h> >> #include "cifsfs.h" >> #include "cifspdu.h" >> @@ -3975,7 +3976,7 @@ readpages_get_pages(struct address_space *mapping, struct list_head *page_list, >> >> INIT_LIST_HEAD(tmplist); >> >> - page = list_entry(page_list->prev, struct page, lru); >> + page = lru_to_page(page_list); >> >> /* >> * Lock the page and put it in the cache. Since no one else >> diff --git a/fs/ext4/readpage.c b/fs/ext4/readpage.c >> index f461d75ac049..6aa282ee455a 100644 >> --- a/fs/ext4/readpage.c >> +++ b/fs/ext4/readpage.c >> @@ -128,7 +128,7 @@ int ext4_mpage_readpages(struct address_space *mapping, >> >> prefetchw(&page->flags); >> if (pages) { >> - page = list_entry(pages->prev, struct page, lru); >> + page = lru_to_page(pages); >> list_del(&page->lru); >> if (add_to_page_cache_lru(page, mapping, page->index, >> readahead_gfp_mask(mapping))) >> diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c >> index eb1ce30412dc..832c1759a09a 100644 >> --- a/fs/ocfs2/aops.c >> +++ b/fs/ocfs2/aops.c >> @@ -30,6 +30,7 @@ >> #include <linux/quotaops.h> >> #include <linux/blkdev.h> >> #include <linux/uio.h> >> +#include <linux/mm.h> >> >> #include <cluster/masklog.h> >> >> @@ -397,7 +398,7 @@ static int ocfs2_readpages(struct file *filp, struct address_space *mapping, >> * Check whether a remote node truncated this file - we just >> * drop out in that case as it's not worth handling here. >> */ >> - last = list_entry(pages->prev, struct page, lru); >> + last = lru_to_page(pages); >> start = (loff_t)last->index << PAGE_SHIFT; >> if (start >= i_size_read(inode)) >> goto out_unlock; >> diff --git a/fs/orangefs/inode.c b/fs/orangefs/inode.c >> index fe53381b26b1..f038235c64bd 100644 >> --- a/fs/orangefs/inode.c >> +++ b/fs/orangefs/inode.c >> @@ -77,7 +77,7 @@ static int orangefs_readpages(struct file *file, >> for (page_idx = 0; page_idx < nr_pages; page_idx++) { >> struct page *page; >> >> - page = list_entry(pages->prev, struct page, lru); >> + page = lru_to_page(pages); >> list_del(&page->lru); >> if (!add_to_page_cache(page, >> mapping, >> diff --git a/mm/swap.c b/mm/swap.c >> index aa483719922e..20b9e9d99652 100644 >> --- a/mm/swap.c >> +++ b/mm/swap.c >> @@ -126,7 +126,7 @@ void put_pages_list(struct list_head *pages) >> while (!list_empty(pages)) { >> struct page *victim; >> >> - victim = list_entry(pages->prev, struct page, lru); >> + victim = lru_to_page(pages); >> list_del(&victim->lru); >> put_page(victim); >> } >> -- >> 2.17.1 > ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH 2/2] fs: Don't open-code lru_to_page 2018-11-29 8:50 ` Nikolay Borisov @ 2018-11-29 9:04 ` Michal Hocko 2018-11-29 11:17 ` Mike Rapoport 2018-11-29 11:38 ` Pankaj Gupta 2 siblings, 0 replies; 9+ messages in thread From: Michal Hocko @ 2018-11-29 9:04 UTC (permalink / raw) To: Nikolay Borisov Cc: linux-kernel, David Howells, Chris Mason, Josef Bacik, David Sterba, Yan, Zheng, Sage Weil, Ilya Dryomov, Steve French, Theodore Ts'o, Andreas Dilger, Mark Fasheh, Joel Becker, Mike Marshall, Martin Brandenburg, Andrew Morton, Mike Rapoport, Matthew Wilcox, Randy Dunlap, YueHaibing, Shakeel Butt, Dan Williams, linux-afs, linux-btrfs, ceph-devel, linux-cifs, samba-technical, linux-ext4, ocfs2-devel, devel, linux-mm On Thu 29-11-18 10:50:08, Nikolay Borisov wrote: > > > On 29.11.18 г. 10:18 ч., Michal Hocko wrote: > > On Thu 29-11-18 09:52:57, Nikolay Borisov wrote: > >> There are a bunch of filesystems which essentially open-code lru_to_page > >> helper. Change them to using the helper. No functional changes. > > > > I would just squash the two into a single patch. It makes the first one > > more obvious. Or is there any reason to have them separate? > > No reason, just didn't know how people would react so that's why I chose > to send as two separate. This is a matter of taste I guess. But I usually prefer to have callers along with a new helper in a single patch. This is not a new helper per-se but doing it the same way seems reasonable to me. Not that I would insist of course. You can use my ack for both patch in case you decide to leave it as is. > If I squash them who would be the best person to take them ? Sounds like a mmotm material to me. -- Michal Hocko SUSE Labs ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH 2/2] fs: Don't open-code lru_to_page 2018-11-29 8:50 ` Nikolay Borisov 2018-11-29 9:04 ` Michal Hocko @ 2018-11-29 11:17 ` Mike Rapoport 2018-11-29 11:38 ` Pankaj Gupta 2 siblings, 0 replies; 9+ messages in thread From: Mike Rapoport @ 2018-11-29 11:17 UTC (permalink / raw) To: Nikolay Borisov Cc: Michal Hocko, linux-kernel, David Howells, Chris Mason, Josef Bacik, David Sterba, Yan, Zheng, Sage Weil, Ilya Dryomov, Steve French, Theodore Ts'o, Andreas Dilger, Mark Fasheh, Joel Becker, Mike Marshall, Martin Brandenburg, Andrew Morton, Mike Rapoport, Matthew Wilcox, Randy Dunlap, YueHaibing, Shakeel Butt, Dan Williams, linux-afs, linux-btrfs, ceph-devel, linux-cifs, samba-technical, linux-ext4, ocfs2-devel, devel, linux-mm On Thu, Nov 29, 2018 at 10:50:08AM +0200, Nikolay Borisov wrote: > > > On 29.11.18 г. 10:18 ч., Michal Hocko wrote: > > On Thu 29-11-18 09:52:57, Nikolay Borisov wrote: > >> There are a bunch of filesystems which essentially open-code lru_to_page > >> helper. Change them to using the helper. No functional changes. > > > > I would just squash the two into a single patch. It makes the first one > > more obvious. Or is there any reason to have them separate? > > No reason, just didn't know how people would react so that's why I chose > to send as two separate. I'm with Michal on this, a single patch seems better here. Yet, for either way: Reviewed-by: Mike Rapoport <rppt@linux.ibm.com> > If I squash them who would be the best person to take them ? > > > > >> Signed-off-by: Nikolay Borisov <nborisov@suse.com> > > > > Acked-by: Michal Hocko <mhocko@suse.com> > > > >> --- > >> > >> Since this is a mostly mechanical change I've actually batched all of them in > >> a single patch. > >> > >> fs/afs/file.c | 5 +++-- > >> fs/btrfs/extent_io.c | 2 +- > >> fs/ceph/addr.c | 5 ++--- > >> fs/cifs/file.c | 3 ++- > >> fs/ext4/readpage.c | 2 +- > >> fs/ocfs2/aops.c | 3 ++- > >> fs/orangefs/inode.c | 2 +- > >> mm/swap.c | 2 +- > >> 8 files changed, 13 insertions(+), 11 deletions(-) > >> > >> diff --git a/fs/afs/file.c b/fs/afs/file.c > >> index d6bc3f5d784b..323ae9912203 100644 > >> --- a/fs/afs/file.c > >> +++ b/fs/afs/file.c > >> @@ -17,6 +17,7 @@ > >> #include <linux/writeback.h> > >> #include <linux/gfp.h> > >> #include <linux/task_io_accounting_ops.h> > >> +#include <linux/mm.h> > >> #include "internal.h" > >> > >> static int afs_file_mmap(struct file *file, struct vm_area_struct *vma); > >> @@ -441,7 +442,7 @@ static int afs_readpages_one(struct file *file, struct address_space *mapping, > >> /* Count the number of contiguous pages at the front of the list. Note > >> * that the list goes prev-wards rather than next-wards. > >> */ > >> - first = list_entry(pages->prev, struct page, lru); > >> + first = lru_to_page(pages); > >> index = first->index + 1; > >> n = 1; > >> for (p = first->lru.prev; p != pages; p = p->prev) { > >> @@ -473,7 +474,7 @@ static int afs_readpages_one(struct file *file, struct address_space *mapping, > >> * page at the end of the file. > >> */ > >> do { > >> - page = list_entry(pages->prev, struct page, lru); > >> + page = lru_to_page(pages); > >> list_del(&page->lru); > >> index = page->index; > >> if (add_to_page_cache_lru(page, mapping, index, > >> diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c > >> index 19f4b8fd654f..8332c5f4b1c3 100644 > >> --- a/fs/btrfs/extent_io.c > >> +++ b/fs/btrfs/extent_io.c > >> @@ -4104,7 +4104,7 @@ int extent_readpages(struct address_space *mapping, struct list_head *pages, > >> u64 prev_em_start = (u64)-1; > >> > >> for (page_idx = 0; page_idx < nr_pages; page_idx++) { > >> - page = list_entry(pages->prev, struct page, lru); > >> + page = lru_to_page(pages); > >> > >> prefetchw(&page->flags); > >> list_del(&page->lru); > >> diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c > >> index 8eade7a993c1..5d0c05e288cc 100644 > >> --- a/fs/ceph/addr.c > >> +++ b/fs/ceph/addr.c > >> @@ -306,7 +306,7 @@ static int start_read(struct inode *inode, struct ceph_rw_context *rw_ctx, > >> struct ceph_osd_client *osdc = > >> &ceph_inode_to_client(inode)->client->osdc; > >> struct ceph_inode_info *ci = ceph_inode(inode); > >> - struct page *page = list_entry(page_list->prev, struct page, lru); > >> + struct page *page = lru_to_page(page_list); > >> struct ceph_vino vino; > >> struct ceph_osd_request *req; > >> u64 off; > >> @@ -333,8 +333,7 @@ static int start_read(struct inode *inode, struct ceph_rw_context *rw_ctx, > >> if (got) > >> ceph_put_cap_refs(ci, got); > >> while (!list_empty(page_list)) { > >> - page = list_entry(page_list->prev, > >> - struct page, lru); > >> + page = lru_to_page(page_list); > >> list_del(&page->lru); > >> put_page(page); > >> } > >> diff --git a/fs/cifs/file.c b/fs/cifs/file.c > >> index 74c33d5fafc8..b16a4d887d17 100644 > >> --- a/fs/cifs/file.c > >> +++ b/fs/cifs/file.c > >> @@ -33,6 +33,7 @@ > >> #include <linux/mount.h> > >> #include <linux/slab.h> > >> #include <linux/swap.h> > >> +#include <linux/mm.h> > >> #include <asm/div64.h> > >> #include "cifsfs.h" > >> #include "cifspdu.h" > >> @@ -3975,7 +3976,7 @@ readpages_get_pages(struct address_space *mapping, struct list_head *page_list, > >> > >> INIT_LIST_HEAD(tmplist); > >> > >> - page = list_entry(page_list->prev, struct page, lru); > >> + page = lru_to_page(page_list); > >> > >> /* > >> * Lock the page and put it in the cache. Since no one else > >> diff --git a/fs/ext4/readpage.c b/fs/ext4/readpage.c > >> index f461d75ac049..6aa282ee455a 100644 > >> --- a/fs/ext4/readpage.c > >> +++ b/fs/ext4/readpage.c > >> @@ -128,7 +128,7 @@ int ext4_mpage_readpages(struct address_space *mapping, > >> > >> prefetchw(&page->flags); > >> if (pages) { > >> - page = list_entry(pages->prev, struct page, lru); > >> + page = lru_to_page(pages); > >> list_del(&page->lru); > >> if (add_to_page_cache_lru(page, mapping, page->index, > >> readahead_gfp_mask(mapping))) > >> diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c > >> index eb1ce30412dc..832c1759a09a 100644 > >> --- a/fs/ocfs2/aops.c > >> +++ b/fs/ocfs2/aops.c > >> @@ -30,6 +30,7 @@ > >> #include <linux/quotaops.h> > >> #include <linux/blkdev.h> > >> #include <linux/uio.h> > >> +#include <linux/mm.h> > >> > >> #include <cluster/masklog.h> > >> > >> @@ -397,7 +398,7 @@ static int ocfs2_readpages(struct file *filp, struct address_space *mapping, > >> * Check whether a remote node truncated this file - we just > >> * drop out in that case as it's not worth handling here. > >> */ > >> - last = list_entry(pages->prev, struct page, lru); > >> + last = lru_to_page(pages); > >> start = (loff_t)last->index << PAGE_SHIFT; > >> if (start >= i_size_read(inode)) > >> goto out_unlock; > >> diff --git a/fs/orangefs/inode.c b/fs/orangefs/inode.c > >> index fe53381b26b1..f038235c64bd 100644 > >> --- a/fs/orangefs/inode.c > >> +++ b/fs/orangefs/inode.c > >> @@ -77,7 +77,7 @@ static int orangefs_readpages(struct file *file, > >> for (page_idx = 0; page_idx < nr_pages; page_idx++) { > >> struct page *page; > >> > >> - page = list_entry(pages->prev, struct page, lru); > >> + page = lru_to_page(pages); > >> list_del(&page->lru); > >> if (!add_to_page_cache(page, > >> mapping, > >> diff --git a/mm/swap.c b/mm/swap.c > >> index aa483719922e..20b9e9d99652 100644 > >> --- a/mm/swap.c > >> +++ b/mm/swap.c > >> @@ -126,7 +126,7 @@ void put_pages_list(struct list_head *pages) > >> while (!list_empty(pages)) { > >> struct page *victim; > >> > >> - victim = list_entry(pages->prev, struct page, lru); > >> + victim = lru_to_page(pages); > >> list_del(&victim->lru); > >> put_page(victim); > >> } > >> -- > >> 2.17.1 > > > -- Sincerely yours, Mike. ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH 2/2] fs: Don't open-code lru_to_page 2018-11-29 8:50 ` Nikolay Borisov 2018-11-29 9:04 ` Michal Hocko 2018-11-29 11:17 ` Mike Rapoport @ 2018-11-29 11:38 ` Pankaj Gupta 2 siblings, 0 replies; 9+ messages in thread From: Pankaj Gupta @ 2018-11-29 11:38 UTC (permalink / raw) To: Nikolay Borisov Cc: Michal Hocko, linux-kernel, David Howells, Chris Mason, Josef Bacik, David Sterba, Zheng Yan, Sage Weil, Ilya Dryomov, Steve French, Theodore Ts'o, Andreas Dilger, Mark Fasheh, Joel Becker, Mike Marshall, Martin Brandenburg, Andrew Morton, Mike Rapoport, Matthew Wilcox, Randy Dunlap, YueHaibing, Shakeel Butt, Dan Williams, linux-afs, linux-btrfs, ceph-devel, linux-cifs, samba-technical, linux-ext4, ocfs2-devel, devel, linux-mm > >> There are a bunch of filesystems which essentially open-code lru_to_page > >> helper. Change them to using the helper. No functional changes. > > > > I would just squash the two into a single patch. It makes the first one > > more obvious. Or is there any reason to have them separate? > > No reason, just didn't know how people would react so that's why I chose > to send as two separate. > > If I squash them who would be the best person to take them ? For squashed version: Acked-by: Pankaj gupta <pagupta@redhat.com> > > > > >> Signed-off-by: Nikolay Borisov <nborisov@suse.com> > > > > Acked-by: Michal Hocko <mhocko@suse.com> > > > >> --- > >> > >> Since this is a mostly mechanical change I've actually batched all of them > >> in > >> a single patch. > >> > >> fs/afs/file.c | 5 +++-- > >> fs/btrfs/extent_io.c | 2 +- > >> fs/ceph/addr.c | 5 ++--- > >> fs/cifs/file.c | 3 ++- > >> fs/ext4/readpage.c | 2 +- > >> fs/ocfs2/aops.c | 3 ++- > >> fs/orangefs/inode.c | 2 +- > >> mm/swap.c | 2 +- > >> 8 files changed, 13 insertions(+), 11 deletions(-) > >> > >> diff --git a/fs/afs/file.c b/fs/afs/file.c > >> index d6bc3f5d784b..323ae9912203 100644 > >> --- a/fs/afs/file.c > >> +++ b/fs/afs/file.c > >> @@ -17,6 +17,7 @@ > >> #include <linux/writeback.h> > >> #include <linux/gfp.h> > >> #include <linux/task_io_accounting_ops.h> > >> +#include <linux/mm.h> > >> #include "internal.h" > >> > >> static int afs_file_mmap(struct file *file, struct vm_area_struct *vma); > >> @@ -441,7 +442,7 @@ static int afs_readpages_one(struct file *file, struct > >> address_space *mapping, > >> /* Count the number of contiguous pages at the front of the list. Note > >> * that the list goes prev-wards rather than next-wards. > >> */ > >> - first = list_entry(pages->prev, struct page, lru); > >> + first = lru_to_page(pages); > >> index = first->index + 1; > >> n = 1; > >> for (p = first->lru.prev; p != pages; p = p->prev) { > >> @@ -473,7 +474,7 @@ static int afs_readpages_one(struct file *file, struct > >> address_space *mapping, > >> * page at the end of the file. > >> */ > >> do { > >> - page = list_entry(pages->prev, struct page, lru); > >> + page = lru_to_page(pages); > >> list_del(&page->lru); > >> index = page->index; > >> if (add_to_page_cache_lru(page, mapping, index, > >> diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c > >> index 19f4b8fd654f..8332c5f4b1c3 100644 > >> --- a/fs/btrfs/extent_io.c > >> +++ b/fs/btrfs/extent_io.c > >> @@ -4104,7 +4104,7 @@ int extent_readpages(struct address_space *mapping, > >> struct list_head *pages, > >> u64 prev_em_start = (u64)-1; > >> > >> for (page_idx = 0; page_idx < nr_pages; page_idx++) { > >> - page = list_entry(pages->prev, struct page, lru); > >> + page = lru_to_page(pages); > >> > >> prefetchw(&page->flags); > >> list_del(&page->lru); > >> diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c > >> index 8eade7a993c1..5d0c05e288cc 100644 > >> --- a/fs/ceph/addr.c > >> +++ b/fs/ceph/addr.c > >> @@ -306,7 +306,7 @@ static int start_read(struct inode *inode, struct > >> ceph_rw_context *rw_ctx, > >> struct ceph_osd_client *osdc = > >> &ceph_inode_to_client(inode)->client->osdc; > >> struct ceph_inode_info *ci = ceph_inode(inode); > >> - struct page *page = list_entry(page_list->prev, struct page, lru); > >> + struct page *page = lru_to_page(page_list); > >> struct ceph_vino vino; > >> struct ceph_osd_request *req; > >> u64 off; > >> @@ -333,8 +333,7 @@ static int start_read(struct inode *inode, struct > >> ceph_rw_context *rw_ctx, > >> if (got) > >> ceph_put_cap_refs(ci, got); > >> while (!list_empty(page_list)) { > >> - page = list_entry(page_list->prev, > >> - struct page, lru); > >> + page = lru_to_page(page_list); > >> list_del(&page->lru); > >> put_page(page); > >> } > >> diff --git a/fs/cifs/file.c b/fs/cifs/file.c > >> index 74c33d5fafc8..b16a4d887d17 100644 > >> --- a/fs/cifs/file.c > >> +++ b/fs/cifs/file.c > >> @@ -33,6 +33,7 @@ > >> #include <linux/mount.h> > >> #include <linux/slab.h> > >> #include <linux/swap.h> > >> +#include <linux/mm.h> > >> #include <asm/div64.h> > >> #include "cifsfs.h" > >> #include "cifspdu.h" > >> @@ -3975,7 +3976,7 @@ readpages_get_pages(struct address_space *mapping, > >> struct list_head *page_list, > >> > >> INIT_LIST_HEAD(tmplist); > >> > >> - page = list_entry(page_list->prev, struct page, lru); > >> + page = lru_to_page(page_list); > >> > >> /* > >> * Lock the page and put it in the cache. Since no one else > >> diff --git a/fs/ext4/readpage.c b/fs/ext4/readpage.c > >> index f461d75ac049..6aa282ee455a 100644 > >> --- a/fs/ext4/readpage.c > >> +++ b/fs/ext4/readpage.c > >> @@ -128,7 +128,7 @@ int ext4_mpage_readpages(struct address_space > >> *mapping, > >> > >> prefetchw(&page->flags); > >> if (pages) { > >> - page = list_entry(pages->prev, struct page, lru); > >> + page = lru_to_page(pages); > >> list_del(&page->lru); > >> if (add_to_page_cache_lru(page, mapping, page->index, > >> readahead_gfp_mask(mapping))) > >> diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c > >> index eb1ce30412dc..832c1759a09a 100644 > >> --- a/fs/ocfs2/aops.c > >> +++ b/fs/ocfs2/aops.c > >> @@ -30,6 +30,7 @@ > >> #include <linux/quotaops.h> > >> #include <linux/blkdev.h> > >> #include <linux/uio.h> > >> +#include <linux/mm.h> > >> > >> #include <cluster/masklog.h> > >> > >> @@ -397,7 +398,7 @@ static int ocfs2_readpages(struct file *filp, struct > >> address_space *mapping, > >> * Check whether a remote node truncated this file - we just > >> * drop out in that case as it's not worth handling here. > >> */ > >> - last = list_entry(pages->prev, struct page, lru); > >> + last = lru_to_page(pages); > >> start = (loff_t)last->index << PAGE_SHIFT; > >> if (start >= i_size_read(inode)) > >> goto out_unlock; > >> diff --git a/fs/orangefs/inode.c b/fs/orangefs/inode.c > >> index fe53381b26b1..f038235c64bd 100644 > >> --- a/fs/orangefs/inode.c > >> +++ b/fs/orangefs/inode.c > >> @@ -77,7 +77,7 @@ static int orangefs_readpages(struct file *file, > >> for (page_idx = 0; page_idx < nr_pages; page_idx++) { > >> struct page *page; > >> > >> - page = list_entry(pages->prev, struct page, lru); > >> + page = lru_to_page(pages); > >> list_del(&page->lru); > >> if (!add_to_page_cache(page, > >> mapping, > >> diff --git a/mm/swap.c b/mm/swap.c > >> index aa483719922e..20b9e9d99652 100644 > >> --- a/mm/swap.c > >> +++ b/mm/swap.c > >> @@ -126,7 +126,7 @@ void put_pages_list(struct list_head *pages) > >> while (!list_empty(pages)) { > >> struct page *victim; > >> > >> - victim = list_entry(pages->prev, struct page, lru); > >> + victim = lru_to_page(pages); > >> list_del(&victim->lru); > >> put_page(victim); > >> } > >> -- > >> 2.17.1 > > > > ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH 2/2] fs: Don't open-code lru_to_page 2018-11-29 8:18 ` Michal Hocko 2018-11-29 8:50 ` Nikolay Borisov @ 2018-11-29 10:12 ` David Hildenbrand 1 sibling, 0 replies; 9+ messages in thread From: David Hildenbrand @ 2018-11-29 10:12 UTC (permalink / raw) To: Michal Hocko, Nikolay Borisov Cc: linux-kernel, David Howells, Chris Mason, Josef Bacik, David Sterba, Yan, Zheng, Sage Weil, Ilya Dryomov, Steve French, Theodore Ts'o, Andreas Dilger, Mark Fasheh, Joel Becker, Mike Marshall, Martin Brandenburg, Andrew Morton, Mike Rapoport, Matthew Wilcox, Randy Dunlap, YueHaibing, Shakeel Butt, Dan Williams, linux-afs, linux-btrfs, ceph-devel, linux-cifs, samba-technical, linux-ext4, ocfs2-devel, devel, linux-mm On 29.11.18 09:18, Michal Hocko wrote: > On Thu 29-11-18 09:52:57, Nikolay Borisov wrote: >> There are a bunch of filesystems which essentially open-code lru_to_page >> helper. Change them to using the helper. No functional changes. > > I would just squash the two into a single patch. It makes the first one > more obvious. Or is there any reason to have them separate? > >> Signed-off-by: Nikolay Borisov <nborisov@suse.com> > > Acked-by: Michal Hocko <mhocko@suse.com> mm_inline.h is 99.9% about lru and there is barely anything about lru in mm.h. However this simple macro seems to differ from the other inlined functions. So to the squashed patch Reviewed-by: David Hildenbrand <david@redhat.com> -- Thanks, David / dhildenb ^ permalink raw reply [flat|nested] 9+ messages in thread
* Re: [PATCH 2/2] fs: Don't open-code lru_to_page 2018-11-29 7:52 ` [PATCH 2/2] fs: Don't open-code lru_to_page Nikolay Borisov 2018-11-29 8:18 ` Michal Hocko @ 2018-11-30 9:11 ` Yan, Zheng 1 sibling, 0 replies; 9+ messages in thread From: Yan, Zheng @ 2018-11-30 9:11 UTC (permalink / raw) To: nborisov Cc: Linux Kernel Mailing List, David Howells, Chris Mason, josef, David Sterba, Zheng Yan, Sage Weil, Ilya Dryomov, sfrench, Theodore Ts'o, adilger.kernel, mark, jlbec, hubcap, martin, Andrew Morton, rppt, willy, rdunlap, yuehaibing, shakeelb, dan.j.williams, mhocko, linux-afs, linux-btrfs, ceph-devel, linux-cifs, samba-technical, linux-ext4, ocfs2-devel, devel, linux-mm On Thu, Nov 29, 2018 at 3:56 PM Nikolay Borisov <nborisov@suse.com> wrote: > > There are a bunch of filesystems which essentially open-code lru_to_page > helper. Change them to using the helper. No functional changes. > > Signed-off-by: Nikolay Borisov <nborisov@suse.com> > --- > > Since this is a mostly mechanical change I've actually batched all of them in > a single patch. > > fs/afs/file.c | 5 +++-- > fs/btrfs/extent_io.c | 2 +- > fs/ceph/addr.c | 5 ++--- > fs/cifs/file.c | 3 ++- > fs/ext4/readpage.c | 2 +- > fs/ocfs2/aops.c | 3 ++- > fs/orangefs/inode.c | 2 +- > mm/swap.c | 2 +- > 8 files changed, 13 insertions(+), 11 deletions(-) > > diff --git a/fs/afs/file.c b/fs/afs/file.c > index d6bc3f5d784b..323ae9912203 100644 > --- a/fs/afs/file.c > +++ b/fs/afs/file.c > @@ -17,6 +17,7 @@ > #include <linux/writeback.h> > #include <linux/gfp.h> > #include <linux/task_io_accounting_ops.h> > +#include <linux/mm.h> > #include "internal.h" > > static int afs_file_mmap(struct file *file, struct vm_area_struct *vma); > @@ -441,7 +442,7 @@ static int afs_readpages_one(struct file *file, struct address_space *mapping, > /* Count the number of contiguous pages at the front of the list. Note > * that the list goes prev-wards rather than next-wards. > */ > - first = list_entry(pages->prev, struct page, lru); > + first = lru_to_page(pages); > index = first->index + 1; > n = 1; > for (p = first->lru.prev; p != pages; p = p->prev) { > @@ -473,7 +474,7 @@ static int afs_readpages_one(struct file *file, struct address_space *mapping, > * page at the end of the file. > */ > do { > - page = list_entry(pages->prev, struct page, lru); > + page = lru_to_page(pages); > list_del(&page->lru); > index = page->index; > if (add_to_page_cache_lru(page, mapping, index, > diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c > index 19f4b8fd654f..8332c5f4b1c3 100644 > --- a/fs/btrfs/extent_io.c > +++ b/fs/btrfs/extent_io.c > @@ -4104,7 +4104,7 @@ int extent_readpages(struct address_space *mapping, struct list_head *pages, > u64 prev_em_start = (u64)-1; > > for (page_idx = 0; page_idx < nr_pages; page_idx++) { > - page = list_entry(pages->prev, struct page, lru); > + page = lru_to_page(pages); > > prefetchw(&page->flags); > list_del(&page->lru); > diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c > index 8eade7a993c1..5d0c05e288cc 100644 > --- a/fs/ceph/addr.c > +++ b/fs/ceph/addr.c > @@ -306,7 +306,7 @@ static int start_read(struct inode *inode, struct ceph_rw_context *rw_ctx, > struct ceph_osd_client *osdc = > &ceph_inode_to_client(inode)->client->osdc; > struct ceph_inode_info *ci = ceph_inode(inode); > - struct page *page = list_entry(page_list->prev, struct page, lru); > + struct page *page = lru_to_page(page_list); > struct ceph_vino vino; > struct ceph_osd_request *req; > u64 off; > @@ -333,8 +333,7 @@ static int start_read(struct inode *inode, struct ceph_rw_context *rw_ctx, > if (got) > ceph_put_cap_refs(ci, got); > while (!list_empty(page_list)) { > - page = list_entry(page_list->prev, > - struct page, lru); > + page = lru_to_page(page_list); > list_del(&page->lru); > put_page(page); > } For the ceph patch Acked-by:: "Yan, Zheng" <zyan@redhat.com > diff --git a/fs/cifs/file.c b/fs/cifs/file.c > index 74c33d5fafc8..b16a4d887d17 100644 > --- a/fs/cifs/file.c > +++ b/fs/cifs/file.c > @@ -33,6 +33,7 @@ > #include <linux/mount.h> > #include <linux/slab.h> > #include <linux/swap.h> > +#include <linux/mm.h> > #include <asm/div64.h> > #include "cifsfs.h" > #include "cifspdu.h" > @@ -3975,7 +3976,7 @@ readpages_get_pages(struct address_space *mapping, struct list_head *page_list, > > INIT_LIST_HEAD(tmplist); > > - page = list_entry(page_list->prev, struct page, lru); > + page = lru_to_page(page_list); > > /* > * Lock the page and put it in the cache. Since no one else > diff --git a/fs/ext4/readpage.c b/fs/ext4/readpage.c > index f461d75ac049..6aa282ee455a 100644 > --- a/fs/ext4/readpage.c > +++ b/fs/ext4/readpage.c > @@ -128,7 +128,7 @@ int ext4_mpage_readpages(struct address_space *mapping, > > prefetchw(&page->flags); > if (pages) { > - page = list_entry(pages->prev, struct page, lru); > + page = lru_to_page(pages); > list_del(&page->lru); > if (add_to_page_cache_lru(page, mapping, page->index, > readahead_gfp_mask(mapping))) > diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c > index eb1ce30412dc..832c1759a09a 100644 > --- a/fs/ocfs2/aops.c > +++ b/fs/ocfs2/aops.c > @@ -30,6 +30,7 @@ > #include <linux/quotaops.h> > #include <linux/blkdev.h> > #include <linux/uio.h> > +#include <linux/mm.h> > > #include <cluster/masklog.h> > > @@ -397,7 +398,7 @@ static int ocfs2_readpages(struct file *filp, struct address_space *mapping, > * Check whether a remote node truncated this file - we just > * drop out in that case as it's not worth handling here. > */ > - last = list_entry(pages->prev, struct page, lru); > + last = lru_to_page(pages); > start = (loff_t)last->index << PAGE_SHIFT; > if (start >= i_size_read(inode)) > goto out_unlock; > diff --git a/fs/orangefs/inode.c b/fs/orangefs/inode.c > index fe53381b26b1..f038235c64bd 100644 > --- a/fs/orangefs/inode.c > +++ b/fs/orangefs/inode.c > @@ -77,7 +77,7 @@ static int orangefs_readpages(struct file *file, > for (page_idx = 0; page_idx < nr_pages; page_idx++) { > struct page *page; > > - page = list_entry(pages->prev, struct page, lru); > + page = lru_to_page(pages); > list_del(&page->lru); > if (!add_to_page_cache(page, > mapping, > diff --git a/mm/swap.c b/mm/swap.c > index aa483719922e..20b9e9d99652 100644 > --- a/mm/swap.c > +++ b/mm/swap.c > @@ -126,7 +126,7 @@ void put_pages_list(struct list_head *pages) > while (!list_empty(pages)) { > struct page *victim; > > - victim = list_entry(pages->prev, struct page, lru); > + victim = lru_to_page(pages); > list_del(&victim->lru); > put_page(victim); > } > -- > 2.17.1 > ^ permalink raw reply [flat|nested] 9+ messages in thread
end of thread, other threads:[~2018-11-30 9:11 UTC | newest] Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2018-11-29 7:52 [PATCH 1/2] mm: Move lru_to_page to mm.h Nikolay Borisov 2018-11-29 7:52 ` [PATCH 2/2] fs: Don't open-code lru_to_page Nikolay Borisov 2018-11-29 8:18 ` Michal Hocko 2018-11-29 8:50 ` Nikolay Borisov 2018-11-29 9:04 ` Michal Hocko 2018-11-29 11:17 ` Mike Rapoport 2018-11-29 11:38 ` Pankaj Gupta 2018-11-29 10:12 ` David Hildenbrand 2018-11-30 9:11 ` Yan, Zheng
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).