diff mbox series

[v3,4/4] btrfs: avoid allocating unnecessary page pointers

Message ID 20200825054808.16241-5-wqu@suse.com (mailing list archive)
State New, archived
Headers show
Series btrfs: basic refactor of btrfs_buffered_write() | expand

Commit Message

Qu Wenruo Aug. 25, 2020, 5:48 a.m. UTC 
Commit 142349f541d0 ("btrfs: lower the dirty balance poll interval")
introduced one limit which is definitely suspicious:

- ensure we always have 8 pages allocated
  The 8 lower limit looks pretty strange, this means even we're just
  writing 4K, we will allocate page pointers for 8 pages no matter what.
  To me, this 8 pages look more like a upper limit.

This 8 pages upper limit looks so incorrect that my eyes alawys correct
it into "min(, 8)" other than "max(, 8)".

This patch will use a fixed size (SZ_64K) other than page numbers to
setup the upper limit.
Also, with comment added to show why we need a upper limit.

Signed-off-by: Qu Wenruo <wqu@suse.com>
---
 fs/btrfs/file.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

Comments

kernel test robot Aug. 25, 2020, 7:46 a.m. UTC | #1 
Hi Qu,

Thank you for the patch! Perhaps something to improve:

[auto build test WARNING on kdave/for-next]
[also build test WARNING on next-20200824]
[cannot apply to v5.9-rc2]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]

url:    https://github.com/0day-ci/linux/commits/Qu-Wenruo/btrfs-basic-refactor-of-btrfs_buffered_write/20200825-135114
base:   https://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux.git for-next
config: s390-randconfig-r012-20200825 (attached as .config)
compiler: clang version 12.0.0 (https://github.com/llvm/llvm-project 77e5a195f818b9ace91f7b12ab948b21d7918238)
reproduce (this is a W=1 build):
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # install s390 cross compiling tool for clang build
        # apt-get install binutils-s390x-linux-gnu
        # save the attached .config to linux build tree
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross ARCH=s390 

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>

All warnings (new ones prefixed by >>):

   include/uapi/linux/swab.h:19:12: note: expanded from macro '___constant_swab32'
           (((__u32)(x) & (__u32)0x000000ffUL) << 24) |            \
                     ^
   In file included from fs/btrfs/file.c:11:
   In file included from include/linux/backing-dev.h:15:
   In file included from include/linux/blkdev.h:25:
   In file included from include/linux/scatterlist.h:9:
   In file included from arch/s390/include/asm/io.h:72:
   include/asm-generic/io.h:490:61: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
           val = __le32_to_cpu((__le32 __force)__raw_readl(PCI_IOBASE + addr));
                                                           ~~~~~~~~~~ ^
   include/uapi/linux/byteorder/big_endian.h:34:59: note: expanded from macro '__le32_to_cpu'
   #define __le32_to_cpu(x) __swab32((__force __u32)(__le32)(x))
                                                             ^
   include/uapi/linux/swab.h:119:21: note: expanded from macro '__swab32'
           ___constant_swab32(x) :                 \
                              ^
   include/uapi/linux/swab.h:20:12: note: expanded from macro '___constant_swab32'
           (((__u32)(x) & (__u32)0x0000ff00UL) <<  8) |            \
                     ^
   In file included from fs/btrfs/file.c:11:
   In file included from include/linux/backing-dev.h:15:
   In file included from include/linux/blkdev.h:25:
   In file included from include/linux/scatterlist.h:9:
   In file included from arch/s390/include/asm/io.h:72:
   include/asm-generic/io.h:490:61: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
           val = __le32_to_cpu((__le32 __force)__raw_readl(PCI_IOBASE + addr));
                                                           ~~~~~~~~~~ ^
   include/uapi/linux/byteorder/big_endian.h:34:59: note: expanded from macro '__le32_to_cpu'
   #define __le32_to_cpu(x) __swab32((__force __u32)(__le32)(x))
                                                             ^
   include/uapi/linux/swab.h:119:21: note: expanded from macro '__swab32'
           ___constant_swab32(x) :                 \
                              ^
   include/uapi/linux/swab.h:21:12: note: expanded from macro '___constant_swab32'
           (((__u32)(x) & (__u32)0x00ff0000UL) >>  8) |            \
                     ^
   In file included from fs/btrfs/file.c:11:
   In file included from include/linux/backing-dev.h:15:
   In file included from include/linux/blkdev.h:25:
   In file included from include/linux/scatterlist.h:9:
   In file included from arch/s390/include/asm/io.h:72:
   include/asm-generic/io.h:490:61: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
           val = __le32_to_cpu((__le32 __force)__raw_readl(PCI_IOBASE + addr));
                                                           ~~~~~~~~~~ ^
   include/uapi/linux/byteorder/big_endian.h:34:59: note: expanded from macro '__le32_to_cpu'
   #define __le32_to_cpu(x) __swab32((__force __u32)(__le32)(x))
                                                             ^
   include/uapi/linux/swab.h:119:21: note: expanded from macro '__swab32'
           ___constant_swab32(x) :                 \
                              ^
   include/uapi/linux/swab.h:22:12: note: expanded from macro '___constant_swab32'
           (((__u32)(x) & (__u32)0xff000000UL) >> 24)))
                     ^
   In file included from fs/btrfs/file.c:11:
   In file included from include/linux/backing-dev.h:15:
   In file included from include/linux/blkdev.h:25:
   In file included from include/linux/scatterlist.h:9:
   In file included from arch/s390/include/asm/io.h:72:
   include/asm-generic/io.h:490:61: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
           val = __le32_to_cpu((__le32 __force)__raw_readl(PCI_IOBASE + addr));
                                                           ~~~~~~~~~~ ^
   include/uapi/linux/byteorder/big_endian.h:34:59: note: expanded from macro '__le32_to_cpu'
   #define __le32_to_cpu(x) __swab32((__force __u32)(__le32)(x))
                                                             ^
   include/uapi/linux/swab.h:120:12: note: expanded from macro '__swab32'
           __fswab32(x))
                     ^
   In file included from fs/btrfs/file.c:11:
   In file included from include/linux/backing-dev.h:15:
   In file included from include/linux/blkdev.h:25:
   In file included from include/linux/scatterlist.h:9:
   In file included from arch/s390/include/asm/io.h:72:
   include/asm-generic/io.h:501:33: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
           __raw_writeb(value, PCI_IOBASE + addr);
                               ~~~~~~~~~~ ^
   include/asm-generic/io.h:511:59: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
           __raw_writew((u16 __force)cpu_to_le16(value), PCI_IOBASE + addr);
                                                         ~~~~~~~~~~ ^
   include/asm-generic/io.h:521:59: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
           __raw_writel((u32 __force)cpu_to_le32(value), PCI_IOBASE + addr);
                                                         ~~~~~~~~~~ ^
   include/asm-generic/io.h:609:20: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
           readsb(PCI_IOBASE + addr, buffer, count);
                  ~~~~~~~~~~ ^
   include/asm-generic/io.h:617:20: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
           readsw(PCI_IOBASE + addr, buffer, count);
                  ~~~~~~~~~~ ^
   include/asm-generic/io.h:625:20: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
           readsl(PCI_IOBASE + addr, buffer, count);
                  ~~~~~~~~~~ ^
   include/asm-generic/io.h:634:21: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
           writesb(PCI_IOBASE + addr, buffer, count);
                   ~~~~~~~~~~ ^
   include/asm-generic/io.h:643:21: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
           writesw(PCI_IOBASE + addr, buffer, count);
                   ~~~~~~~~~~ ^
   include/asm-generic/io.h:652:21: warning: performing pointer arithmetic on a null pointer has undefined behavior [-Wnull-pointer-arithmetic]
           writesl(PCI_IOBASE + addr, buffer, count);
                   ~~~~~~~~~~ ^
>> fs/btrfs/file.c:1571:13: warning: comparison of distinct pointer types ('typeof (nr_pages) *' (aka 'int *') and 'typeof (65536 / ((1UL) << 12)) *' (aka 'unsigned long *')) [-Wcompare-distinct-pointer-types]
           nr_pages = min(nr_pages, SZ_64K / PAGE_SIZE);
                      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   include/linux/kernel.h:883:19: note: expanded from macro 'min'
   #define min(x, y)       __careful_cmp(x, y, <)
                           ^~~~~~~~~~~~~~~~~~~~~~
   include/linux/kernel.h:874:24: note: expanded from macro '__careful_cmp'
           __builtin_choose_expr(__safe_cmp(x, y), \
                                 ^~~~~~~~~~~~~~~~
   include/linux/kernel.h:864:4: note: expanded from macro '__safe_cmp'
                   (__typecheck(x, y) && __no_side_effects(x, y))
                    ^~~~~~~~~~~~~~~~~
   include/linux/kernel.h:850:29: note: expanded from macro '__typecheck'
                   (!!(sizeof((typeof(x) *)1 == (typeof(y) *)1)))
                              ~~~~~~~~~~~~~~ ^  ~~~~~~~~~~~~~~
   21 warnings generated.

# https://github.com/0day-ci/linux/commit/a73ab37ebab960522a0b353a6f20c8094ab911c5
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Qu-Wenruo/btrfs-basic-refactor-of-btrfs_buffered_write/20200825-135114
git checkout a73ab37ebab960522a0b353a6f20c8094ab911c5
vim +1571 fs/btrfs/file.c

  1548	
  1549	/* Helper to get how many pages we should alloc for the batch */
  1550	static int calc_nr_pages(loff_t pos, struct iov_iter *iov)
  1551	{
  1552		int nr_pages;
  1553	
  1554		/*
  1555		 * Try to cover the full iov range, as btrfs metadata/data reserve
  1556		 * and release can be pretty slow, thus the more pages we process in
  1557		 * one batch the better.
  1558		 */
  1559		nr_pages = (round_up(pos + iov_iter_count(iov), PAGE_SIZE) -
  1560			    round_down(pos, PAGE_SIZE)) / PAGE_SIZE;
  1561	
  1562		nr_pages = min(nr_pages, current->nr_dirtied_pause -
  1563					 current->nr_dirtied);
  1564	
  1565		/*
  1566		 * Limit the batch to 64K, too large batch may lead to higher memory
  1567		 * pressure and increase the possibility of short-copy.
  1568		 * With more and more short-copy, the benefit of batch copy would be
  1569		 * hugely reduced, as we will fall back to page-by-page copy.
  1570		 */
> 1571		nr_pages = min(nr_pages, SZ_64K / PAGE_SIZE);
  1572		return nr_pages;
  1573	}
  1574	

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
kernel test robot Aug. 25, 2020, 7:57 a.m. UTC | #2 
Hi Qu,

Thank you for the patch! Perhaps something to improve:

[auto build test WARNING on kdave/for-next]
[also build test WARNING on next-20200824]
[cannot apply to v5.9-rc2]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]

url:    https://github.com/0day-ci/linux/commits/Qu-Wenruo/btrfs-basic-refactor-of-btrfs_buffered_write/20200825-135114
base:   https://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux.git for-next
config: x86_64-randconfig-s022-20200825 (attached as .config)
compiler: gcc-9 (Debian 9.3.0-15) 9.3.0
reproduce:
        # apt-get install sparse
        # sparse version: v0.6.2-191-g10164920-dirty
        # save the attached .config to linux build tree
        make W=1 C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' ARCH=x86_64 

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>


sparse warnings: (new ones prefixed by >>)

>> fs/btrfs/file.c:1571:20: sparse: sparse: incompatible types in comparison expression (different type sizes):
>> fs/btrfs/file.c:1571:20: sparse:    int *
>> fs/btrfs/file.c:1571:20: sparse:    unsigned long *

# https://github.com/0day-ci/linux/commit/a73ab37ebab960522a0b353a6f20c8094ab911c5
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Qu-Wenruo/btrfs-basic-refactor-of-btrfs_buffered_write/20200825-135114
git checkout a73ab37ebab960522a0b353a6f20c8094ab911c5
vim +1571 fs/btrfs/file.c

  1548	
  1549	/* Helper to get how many pages we should alloc for the batch */
  1550	static int calc_nr_pages(loff_t pos, struct iov_iter *iov)
  1551	{
  1552		int nr_pages;
  1553	
  1554		/*
  1555		 * Try to cover the full iov range, as btrfs metadata/data reserve
  1556		 * and release can be pretty slow, thus the more pages we process in
  1557		 * one batch the better.
  1558		 */
  1559		nr_pages = (round_up(pos + iov_iter_count(iov), PAGE_SIZE) -
  1560			    round_down(pos, PAGE_SIZE)) / PAGE_SIZE;
  1561	
  1562		nr_pages = min(nr_pages, current->nr_dirtied_pause -
  1563					 current->nr_dirtied);
  1564	
  1565		/*
  1566		 * Limit the batch to 64K, too large batch may lead to higher memory
  1567		 * pressure and increase the possibility of short-copy.
  1568		 * With more and more short-copy, the benefit of batch copy would be
  1569		 * hugely reduced, as we will fall back to page-by-page copy.
  1570		 */
> 1571		nr_pages = min(nr_pages, SZ_64K / PAGE_SIZE);
  1572		return nr_pages;
  1573	}
  1574	

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
kernel test robot Aug. 26, 2020, 12:31 p.m. UTC | #3 
Hi Qu,

Thank you for the patch! Perhaps something to improve:

[auto build test WARNING on kdave/for-next]
[also build test WARNING on next-20200826]
[cannot apply to btrfs/next v5.9-rc2]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]

url:    https://github.com/0day-ci/linux/commits/Qu-Wenruo/btrfs-basic-refactor-of-btrfs_buffered_write/20200825-135114
base:   https://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux.git for-next
config: riscv-randconfig-r033-20200826 (attached as .config)
compiler: clang version 12.0.0 (https://github.com/llvm/llvm-project 7cfcecece0e0430937cf529ce74d3a071a4dedc6)
reproduce (this is a W=1 build):
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # install riscv cross compiling tool for clang build
        # apt-get install binutils-riscv64-linux-gnu
        # save the attached .config to linux build tree
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross ARCH=riscv 

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>

All warnings (new ones prefixed by >>):

>> fs/btrfs/file.c:1571:13: warning: comparison of distinct pointer types ('typeof (nr_pages) *' (aka 'int *') and 'typeof (65536 / ((1UL) << (12))) *' (aka 'unsigned long *')) [-Wcompare-distinct-pointer-types]
           nr_pages = min(nr_pages, SZ_64K / PAGE_SIZE);
                      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
   include/linux/kernel.h:883:19: note: expanded from macro 'min'
   #define min(x, y)       __careful_cmp(x, y, <)
                           ^~~~~~~~~~~~~~~~~~~~~~
   include/linux/kernel.h:874:24: note: expanded from macro '__careful_cmp'
           __builtin_choose_expr(__safe_cmp(x, y), \
                                 ^~~~~~~~~~~~~~~~
   include/linux/kernel.h:864:4: note: expanded from macro '__safe_cmp'
                   (__typecheck(x, y) && __no_side_effects(x, y))
                    ^~~~~~~~~~~~~~~~~
   include/linux/kernel.h:850:29: note: expanded from macro '__typecheck'
                   (!!(sizeof((typeof(x) *)1 == (typeof(y) *)1)))
                              ~~~~~~~~~~~~~~ ^  ~~~~~~~~~~~~~~
   1 warning generated.

# https://github.com/0day-ci/linux/commit/a73ab37ebab960522a0b353a6f20c8094ab911c5
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Qu-Wenruo/btrfs-basic-refactor-of-btrfs_buffered_write/20200825-135114
git checkout a73ab37ebab960522a0b353a6f20c8094ab911c5
vim +1571 fs/btrfs/file.c

  1548	
  1549	/* Helper to get how many pages we should alloc for the batch */
  1550	static int calc_nr_pages(loff_t pos, struct iov_iter *iov)
  1551	{
  1552		int nr_pages;
  1553	
  1554		/*
  1555		 * Try to cover the full iov range, as btrfs metadata/data reserve
  1556		 * and release can be pretty slow, thus the more pages we process in
  1557		 * one batch the better.
  1558		 */
  1559		nr_pages = (round_up(pos + iov_iter_count(iov), PAGE_SIZE) -
  1560			    round_down(pos, PAGE_SIZE)) / PAGE_SIZE;
  1561	
  1562		nr_pages = min(nr_pages, current->nr_dirtied_pause -
  1563					 current->nr_dirtied);
  1564	
  1565		/*
  1566		 * Limit the batch to 64K, too large batch may lead to higher memory
  1567		 * pressure and increase the possibility of short-copy.
  1568		 * With more and more short-copy, the benefit of batch copy would be
  1569		 * hugely reduced, as we will fall back to page-by-page copy.
  1570		 */
> 1571		nr_pages = min(nr_pages, SZ_64K / PAGE_SIZE);
  1572		return nr_pages;
  1573	}
  1574	

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
diff mbox series

Patch

diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c
index 67d2368a8fa6..de6d1c313042 100644
--- a/fs/btrfs/file.c
+++ b/fs/btrfs/file.c
@@ -1561,7 +1561,14 @@  static int calc_nr_pages(loff_t pos, struct iov_iter *iov)
 
 	nr_pages = min(nr_pages, current->nr_dirtied_pause -
 				 current->nr_dirtied);
-	nr_pages = max(nr_pages, 8);
+
+	/*
+	 * Limit the batch to 64K, too large batch may lead to higher memory
+	 * pressure and increase the possibility of short-copy.
+	 * With more and more short-copy, the benefit of batch copy would be
+	 * hugely reduced, as we will fall back to page-by-page copy.
+	 */
+	nr_pages = min(nr_pages, SZ_64K / PAGE_SIZE);
 	return nr_pages;
 }