| Message ID | 20200416154606.306-2-willy@infradead.org (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Make PageWriteback use the PageLocked optimisation | expand |
On Thu, Apr 16, 2020 at 08:46:02AM -0700, Matthew Wilcox wrote: > From: "Matthew Wilcox (Oracle)" <willy@infradead.org> > > This local definition hasn't been used since commit 84c6591103db > ("locking/atomics, asm-generic/bitops/lock.h: Rewrite using > atomic_fetch_*()") which provided a default definition. > > Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org> > Reviewed-by: William Kucharski <william.kucharski@oracle.com> > Cc: Will Deacon <will@kernel.org> > --- > mm/filemap.c | 23 ----------------------- > 1 file changed, 23 deletions(-) Ok, for my own curiosity I tried building for Alpha because I couldn't for the life of me figure it out, and behold: mm/filemap.c: In function 'unlock_page': mm/filemap.c:1271:6: error: implicit declaration of function 'clear_bit_unlock_is_negative_byte' [-Werror=implicit-function-declaration] if (clear_bit_unlock_is_negative_byte(PG_locked, &page->flags)) ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ cc1: some warnings being treated as errors make[1]: *** [scripts/Makefile.build:267: mm/filemap.o] Error 1 make[1]: *** Waiting for unfinished jobs.... make: *** [Makefile:1722: mm] Error 2 make: *** Waiting for unfinished jobs.... I had to enable CONFIG_SMP, so maybe the robot doesn't do that? Anyway, it's somewhat reassuring that it broke, if not unfortunate at the same time! Will
On Thu, Apr 16, 2020 at 06:02:24PM +0100, Will Deacon wrote: > On Thu, Apr 16, 2020 at 08:46:02AM -0700, Matthew Wilcox wrote: > > From: "Matthew Wilcox (Oracle)" <willy@infradead.org> > > This local definition hasn't been used since commit 84c6591103db > > ("locking/atomics, asm-generic/bitops/lock.h: Rewrite using > > atomic_fetch_*()") which provided a default definition. > > Ok, for my own curiosity I tried building for Alpha because I couldn't for > the life of me figure it out, and behold: > > mm/filemap.c: In function 'unlock_page': > mm/filemap.c:1271:6: error: implicit declaration of function 'clear_bit_unlock_is_negative_byte' [-Werror=implicit-function-declaration] > if (clear_bit_unlock_is_negative_byte(PG_locked, &page->flags)) > ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > cc1: some warnings being treated as errors > > I had to enable CONFIG_SMP, so maybe the robot doesn't do that? > > Anyway, it's somewhat reassuring that it broke, if not unfortunate at the same > time! Thanks! The robot says it built two alpha configs, randconfig-a001-20200325 and defconfig. I imagine neither has SMP set. kbuild people, please can you add SMP and non-SMP options to the configs you test?
diff --git a/mm/filemap.c b/mm/filemap.c index 23a051a7ef0f..e475117e89eb 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1248,29 +1248,6 @@ void add_page_wait_queue(struct page *page, wait_queue_entry_t *waiter) } EXPORT_SYMBOL_GPL(add_page_wait_queue); -#ifndef clear_bit_unlock_is_negative_byte - -/* - * PG_waiters is the high bit in the same byte as PG_lock. - * - * On x86 (and on many other architectures), we can clear PG_lock and - * test the sign bit at the same time. But if the architecture does - * not support that special operation, we just do this all by hand - * instead. - * - * The read of PG_waiters has to be after (or concurrently with) PG_locked - * being cleared, but a memory barrier should be unneccssary since it is - * in the same byte as PG_locked. - */ -static inline bool clear_bit_unlock_is_negative_byte(long nr, volatile void *mem) -{ - clear_bit_unlock(nr, mem); - /* smp_mb__after_atomic(); */ - return test_bit(PG_waiters, mem); -} - -#endif - /** * unlock_page - unlock a locked page * @page: the page