diff for duplicates of <20160831213607.GA6921@linux.intel.com> diff --git a/a/1.txt b/N1/1.txt index 0920f14..db85b6c 100644 --- a/a/1.txt +++ b/N1/1.txt @@ -78,3 +78,7 @@ On Wed, Aug 31, 2016 at 08:20:48PM +0000, Kani, Toshimitsu wrote: > Can you please take a look? Yep, thanks for the report. +_______________________________________________ +Linux-nvdimm mailing list +Linux-nvdimm@lists.01.org +https://lists.01.org/mailman/listinfo/linux-nvdimm diff --git a/a/content_digest b/N1/content_digest index a1aa253..30daa61 100644 --- a/a/content_digest +++ b/N1/content_digest @@ -1,24 +1,23 @@ "ref\020160823220419.11717-1-ross.zwisler@linux.intel.com\0" "ref\020160830230150.GA12173@linux.intel.com\0" "ref\01472674799.2092.19.camel@hpe.com\0" - "ref\01472674799.2092.19.camel-ZPxbGqLxI0U@public.gmane.org\0" - "From\0Ross Zwisler <ross.zwisler-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>\0" + "From\0Ross Zwisler <ross.zwisler@linux.intel.com>\0" "Subject\0Re: [PATCH v2 0/9] re-enable DAX PMD support\0" "Date\0Wed, 31 Aug 2016 15:36:07 -0600\0" "To\0Kani" - " Toshimitsu <toshi.kani-ZPxbGqLxI0U@public.gmane.org>\0" - "Cc\0tytso-3s7WtUTddSA@public.gmane.org <tytso-3s7WtUTddSA@public.gmane.org>" - akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org <akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org> - mawilcox-0li6OtcxBFHby3iVrkZq2A@public.gmane.org <mawilcox-0li6OtcxBFHby3iVrkZq2A@public.gmane.org> - linux-nvdimm-hn68Rpc1hR1g9hUCZPvPmw@public.gmane.org <linux-nvdimm-hn68Rpc1hR1g9hUCZPvPmw@public.gmane.org> - david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org <david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org> - linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org <linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org> - linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org <linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org> - adilger.kernel-m1MBpc4rdrD3fQ9qLvQP4Q@public.gmane.org <adilger.kernel-m1MBpc4rdrD3fQ9qLvQP4Q@public.gmane.org> - viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org <viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org> - jack-IBi9RG/b67k@public.gmane.org <jack-IBi9RG/b67k@public.gmane.org> - linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org <linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org> - " linux-ext4-u79uwXL29TY76Z2rM5mHXA@public.gmane.org <linux-ext4-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>\0" + " Toshimitsu <toshi.kani@hpe.com>\0" + "Cc\0tytso@mit.edu <tytso@mit.edu>" + akpm@linux-foundation.org <akpm@linux-foundation.org> + mawilcox@microsoft.com <mawilcox@microsoft.com> + linux-nvdimm@lists.01.org <linux-nvdimm@lists.01.org> + david@fromorbit.com <david@fromorbit.com> + linux-kernel@vger.kernel.org <linux-kernel@vger.kernel.org> + linux-mm@kvack.org <linux-mm@kvack.org> + adilger.kernel@dilger.ca <adilger.kernel@dilger.ca> + viro@zeniv.linux.org.uk <viro@zeniv.linux.org.uk> + jack@suse.com <jack@suse.com> + linux-fsdevel@vger.kernel.org <linux-fsdevel@vger.kernel.org> + " linux-ext4@vger.kernel.org <linux-ext4@vger.kernel.org>\0" "\00:1\0" "b\0" "On Wed, Aug 31, 2016 at 08:20:48PM +0000, Kani, Toshimitsu wrote:\n" @@ -100,6 +99,10 @@ "> \n" "> Can you please take a look?\n" "\n" - Yep, thanks for the report. + "Yep, thanks for the report.\n" + "_______________________________________________\n" + "Linux-nvdimm mailing list\n" + "Linux-nvdimm@lists.01.org\n" + https://lists.01.org/mailman/listinfo/linux-nvdimm -fb7b47181b4bca9d9feaa7b813533219a247f4a52ae84389b16d9529803a8def +5f613268ab73202b1410f21f1d52e8bd7725b0aeed112024e8c928acde758b6a
diff --git a/a/1.txt b/N2/1.txt index 0920f14..07ec76f 100644 --- a/a/1.txt +++ b/N2/1.txt @@ -3,60 +3,60 @@ On Wed, Aug 31, 2016 at 08:20:48PM +0000, Kani, Toshimitsu wrote: > > On Tue, Aug 23, 2016 at 04:04:10PM -0600, Ross Zwisler wrote: > > > > > > DAX PMDs have been disabled since Jan Kara introduced DAX radix -> > > tree based locking. This series allows DAX PMDs to participate in +> > > tree based locking.��This series allows DAX PMDs to participate in > > > the DAX radix tree based locking scheme so that they can be re- > > > enabled. > > > > > > Changes since v1: -> > > - PMD entry locking is now done based on the starting offset of +> > > �- PMD entry locking is now done based on the starting offset of > > > the PMD entry, rather than on the radix tree slot which was > > > unreliable. (Jan) -> > > - Fixed the one issue I could find with hole punch. As far as I +> > > �- Fixed the one issue I could find with hole punch.��As far as I > > > can tell hole punch now works correctly for both PMD and PTE DAX > > > entries, 4k zero pages and huge zero pages. -> > > - Fixed the way that ext2 returns the size of holes in +> > > �- Fixed the way that ext2 returns the size of holes in > > > ext2_get_block(). (Jan) -> > > - Made the 'wait_table' global variable static in respnse to a +> > > �- Made the 'wait_table' global variable static in respnse to a > > > sparse warning. -> > > - Fixed some more inconsitent usage between the names 'ret' and +> > > �- Fixed some more inconsitent usage between the names 'ret' and > > > 'entry' for radix tree entry variables. > > > > > > Ross Zwisler (9): -> > > ext4: allow DAX writeback for hole punch -> > > ext2: tell DAX the size of allocation holes -> > > ext4: tell DAX the size of allocation holes -> > > dax: remove buffer_size_valid() -> > > dax: make 'wait_table' global variable static -> > > dax: consistent variable naming for DAX entries -> > > dax: coordinate locking for offsets in PMD range -> > > dax: re-enable DAX PMD support -> > > dax: remove "depends on BROKEN" from FS_DAX_PMD +> > > � ext4: allow DAX writeback for hole punch +> > > � ext2: tell DAX the size of allocation holes +> > > � ext4: tell DAX the size of allocation holes +> > > � dax: remove buffer_size_valid() +> > > � dax: make 'wait_table' global variable static +> > > � dax: consistent variable naming for DAX entries +> > > � dax: coordinate locking for offsets in PMD range +> > > � dax: re-enable DAX PMD support +> > > � dax: remove "depends on BROKEN" from FS_DAX_PMD > > > -> > > fs/Kconfig | 1 - -> > > fs/dax.c | 297 +++++++++++++++++++++++++++++----------- +> > > �fs/Kconfig����������|���1 - +> > > �fs/dax.c������������| 297 +++++++++++++++++++++++++++++----------- > > > ------------ -> > > fs/ext2/inode.c | 3 + -> > > fs/ext4/inode.c | 7 +- -> > > include/linux/dax.h | 29 ++++- -> > > mm/filemap.c | 6 +- -> > > 6 files changed, 201 insertions(+), 142 deletions(-) +> > > �fs/ext2/inode.c�����|���3 + +> > > �fs/ext4/inode.c�����|���7 +- +> > > �include/linux/dax.h |��29 ++++- +> > > �mm/filemap.c��������|���6 +- +> > > �6 files changed, 201 insertions(+), 142 deletions(-) > > > -> > > -- +> > > --� > > > 2.9.0 > > -> > Ping on this series? Any objections or comments? +> > Ping on this series?��Any objections or comments? > > Hi Ross, > > I am seeing a major performance loss in fio mmap test with this patch- -> set applied. This happens with or without my patches [1] applied on -> top of yours. Without my patches, dax_pmd_fault() falls back to the +> set applied. �This happens with or without my patches [1] applied on +> top of yours. �Without my patches,�dax_pmd_fault() falls back to the > pte handler since an mmap'ed address is not 2MB-aligned. > > I have attached three test results. -> o rc4.log - 4.8.0-rc4 (base) -> o non-pmd.log - 4.8.0-rc4 + your patchset (fall back to pte) -> o pmd.log - 4.8.0-rc4 + your patchset + my patchset (use pmd maps) +> �o rc4.log - 4.8.0-rc4 (base) +> �o non-pmd.log - 4.8.0-rc4 + your patchset (fall back to pte) +> �o pmd.log - 4.8.0-rc4 + your patchset + my patchset (use pmd maps) > > My test steps are as follows. > @@ -78,3 +78,9 @@ On Wed, Aug 31, 2016 at 08:20:48PM +0000, Kani, Toshimitsu wrote: > Can you please take a look? Yep, thanks for the report. + +-- +To unsubscribe, send a message with 'unsubscribe linux-mm' in +the body to majordomo@kvack.org. For more info on Linux MM, +see: http://www.linux-mm.org/ . +Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a> diff --git a/a/content_digest b/N2/content_digest index a1aa253..a725993 100644 --- a/a/content_digest +++ b/N2/content_digest @@ -1,24 +1,25 @@ "ref\020160823220419.11717-1-ross.zwisler@linux.intel.com\0" "ref\020160830230150.GA12173@linux.intel.com\0" "ref\01472674799.2092.19.camel@hpe.com\0" - "ref\01472674799.2092.19.camel-ZPxbGqLxI0U@public.gmane.org\0" - "From\0Ross Zwisler <ross.zwisler-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>\0" + "From\0Ross Zwisler <ross.zwisler@linux.intel.com>\0" "Subject\0Re: [PATCH v2 0/9] re-enable DAX PMD support\0" "Date\0Wed, 31 Aug 2016 15:36:07 -0600\0" "To\0Kani" - " Toshimitsu <toshi.kani-ZPxbGqLxI0U@public.gmane.org>\0" - "Cc\0tytso-3s7WtUTddSA@public.gmane.org <tytso-3s7WtUTddSA@public.gmane.org>" - akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org <akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org> - mawilcox-0li6OtcxBFHby3iVrkZq2A@public.gmane.org <mawilcox-0li6OtcxBFHby3iVrkZq2A@public.gmane.org> - linux-nvdimm-hn68Rpc1hR1g9hUCZPvPmw@public.gmane.org <linux-nvdimm-hn68Rpc1hR1g9hUCZPvPmw@public.gmane.org> - david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org <david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org> - linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org <linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org> - linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org <linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org> - adilger.kernel-m1MBpc4rdrD3fQ9qLvQP4Q@public.gmane.org <adilger.kernel-m1MBpc4rdrD3fQ9qLvQP4Q@public.gmane.org> - viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org <viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org> - jack-IBi9RG/b67k@public.gmane.org <jack-IBi9RG/b67k@public.gmane.org> - linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org <linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org> - " linux-ext4-u79uwXL29TY76Z2rM5mHXA@public.gmane.org <linux-ext4-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>\0" + " Toshimitsu <toshi.kani@hpe.com>\0" + "Cc\0ross.zwisler@linux.intel.com <ross.zwisler@linux.intel.com>" + linux-kernel@vger.kernel.org <linux-kernel@vger.kernel.org> + linux-mm@kvack.org <linux-mm@kvack.org> + adilger.kernel@dilger.ca <adilger.kernel@dilger.ca> + viro@zeniv.linux.org.uk <viro@zeniv.linux.org.uk> + dan.j.williams@intel.com <dan.j.williams@intel.com> + akpm@linux-foundation.org <akpm@linux-foundation.org> + mawilcox@microsoft.com <mawilcox@microsoft.com> + linux-nvdimm@lists.01.org <linux-nvdimm@lists.01.org> + linux-fsdevel@vger.kernel.org <linux-fsdevel@vger.kernel.org> + jack@suse.com <jack@suse.com> + tytso@mit.edu <tytso@mit.edu> + linux-ext4@vger.kernel.org <linux-ext4@vger.kernel.org> + " david@fromorbit.com <david@fromorbit.com>\0" "\00:1\0" "b\0" "On Wed, Aug 31, 2016 at 08:20:48PM +0000, Kani, Toshimitsu wrote:\n" @@ -26,60 +27,60 @@ "> > On Tue, Aug 23, 2016 at 04:04:10PM -0600, Ross Zwisler wrote:\n" "> > > \n" "> > > DAX PMDs have been disabled since Jan Kara introduced DAX radix\n" - "> > > tree based locking.\302\240\302\240This series allows DAX PMDs to participate in\n" + "> > > tree based locking.\303\257\302\277\302\275\303\257\302\277\302\275This series allows DAX PMDs to participate in\n" "> > > the DAX radix tree based locking scheme so that they can be re-\n" "> > > enabled.\n" "> > > \n" "> > > Changes since v1:\n" - "> > > \302\240- PMD entry locking is now done based on the starting offset of\n" + "> > > \303\257\302\277\302\275- PMD entry locking is now done based on the starting offset of\n" "> > > the PMD entry, rather than on the radix tree slot which was\n" "> > > unreliable. (Jan)\n" - "> > > \302\240- Fixed the one issue I could find with hole punch.\302\240\302\240As far as I\n" + "> > > \303\257\302\277\302\275- Fixed the one issue I could find with hole punch.\303\257\302\277\302\275\303\257\302\277\302\275As far as I\n" "> > > can tell hole punch now works correctly for both PMD and PTE DAX\n" "> > > entries, 4k zero pages and huge zero pages.\n" - "> > > \302\240- Fixed the way that ext2 returns the size of holes in\n" + "> > > \303\257\302\277\302\275- Fixed the way that ext2 returns the size of holes in\n" "> > > ext2_get_block(). (Jan)\n" - "> > > \302\240- Made the 'wait_table' global variable static in respnse to a\n" + "> > > \303\257\302\277\302\275- Made the 'wait_table' global variable static in respnse to a\n" "> > > sparse warning.\n" - "> > > \302\240- Fixed some more inconsitent usage between the names 'ret' and\n" + "> > > \303\257\302\277\302\275- Fixed some more inconsitent usage between the names 'ret' and\n" "> > > 'entry' for radix tree entry variables.\n" "> > > \n" "> > > Ross Zwisler (9):\n" - "> > > \302\240 ext4: allow DAX writeback for hole punch\n" - "> > > \302\240 ext2: tell DAX the size of allocation holes\n" - "> > > \302\240 ext4: tell DAX the size of allocation holes\n" - "> > > \302\240 dax: remove buffer_size_valid()\n" - "> > > \302\240 dax: make 'wait_table' global variable static\n" - "> > > \302\240 dax: consistent variable naming for DAX entries\n" - "> > > \302\240 dax: coordinate locking for offsets in PMD range\n" - "> > > \302\240 dax: re-enable DAX PMD support\n" - "> > > \302\240 dax: remove \"depends on BROKEN\" from FS_DAX_PMD\n" + "> > > \303\257\302\277\302\275 ext4: allow DAX writeback for hole punch\n" + "> > > \303\257\302\277\302\275 ext2: tell DAX the size of allocation holes\n" + "> > > \303\257\302\277\302\275 ext4: tell DAX the size of allocation holes\n" + "> > > \303\257\302\277\302\275 dax: remove buffer_size_valid()\n" + "> > > \303\257\302\277\302\275 dax: make 'wait_table' global variable static\n" + "> > > \303\257\302\277\302\275 dax: consistent variable naming for DAX entries\n" + "> > > \303\257\302\277\302\275 dax: coordinate locking for offsets in PMD range\n" + "> > > \303\257\302\277\302\275 dax: re-enable DAX PMD support\n" + "> > > \303\257\302\277\302\275 dax: remove \"depends on BROKEN\" from FS_DAX_PMD\n" "> > > \n" - "> > > \302\240fs/Kconfig\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240|\302\240\302\240\302\2401 -\n" - "> > > \302\240fs/dax.c\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240| 297 +++++++++++++++++++++++++++++-----------\n" + "> > > \303\257\302\277\302\275fs/Kconfig\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275|\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\2751 -\n" + "> > > \303\257\302\277\302\275fs/dax.c\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275| 297 +++++++++++++++++++++++++++++-----------\n" "> > > ------------\n" - "> > > \302\240fs/ext2/inode.c\302\240\302\240\302\240\302\240\302\240|\302\240\302\240\302\2403 +\n" - "> > > \302\240fs/ext4/inode.c\302\240\302\240\302\240\302\240\302\240|\302\240\302\240\302\2407 +-\n" - "> > > \302\240include/linux/dax.h |\302\240\302\24029 ++++-\n" - "> > > \302\240mm/filemap.c\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240|\302\240\302\240\302\2406 +-\n" - "> > > \302\2406 files changed, 201 insertions(+), 142 deletions(-)\n" + "> > > \303\257\302\277\302\275fs/ext2/inode.c\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275|\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\2753 +\n" + "> > > \303\257\302\277\302\275fs/ext4/inode.c\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275|\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\2757 +-\n" + "> > > \303\257\302\277\302\275include/linux/dax.h |\303\257\302\277\302\275\303\257\302\277\302\27529 ++++-\n" + "> > > \303\257\302\277\302\275mm/filemap.c\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\275|\303\257\302\277\302\275\303\257\302\277\302\275\303\257\302\277\302\2756 +-\n" + "> > > \303\257\302\277\302\2756 files changed, 201 insertions(+), 142 deletions(-)\n" "> > > \n" - "> > > --\302\240\n" + "> > > --\303\257\302\277\302\275\n" "> > > 2.9.0\n" "> > \n" - "> > Ping on this series?\302\240\302\240Any objections or comments?\n" + "> > Ping on this series?\303\257\302\277\302\275\303\257\302\277\302\275Any objections or comments?\n" "> \n" "> Hi Ross,\n" "> \n" "> I am seeing a major performance loss in fio mmap test with this patch-\n" - "> set applied. \302\240This happens with or without my patches [1] applied on\n" - "> top of yours. \302\240Without my patches,\302\240dax_pmd_fault() falls back to the\n" + "> set applied. \303\257\302\277\302\275This happens with or without my patches [1] applied on\n" + "> top of yours. \303\257\302\277\302\275Without my patches,\303\257\302\277\302\275dax_pmd_fault() falls back to the\n" "> pte handler since an mmap'ed address is not 2MB-aligned.\n" "> \n" "> I have attached three test results.\n" - "> \302\240o rc4.log - 4.8.0-rc4 (base)\n" - "> \302\240o non-pmd.log - 4.8.0-rc4 + your patchset (fall back to pte)\n" - "> \302\240o pmd.log - 4.8.0-rc4 + your patchset + my patchset (use pmd maps)\n" + "> \303\257\302\277\302\275o rc4.log - 4.8.0-rc4 (base)\n" + "> \303\257\302\277\302\275o non-pmd.log - 4.8.0-rc4 + your patchset (fall back to pte)\n" + "> \303\257\302\277\302\275o pmd.log - 4.8.0-rc4 + your patchset + my patchset (use pmd maps)\n" "> \n" "> My test steps are as follows.\n" "> \n" @@ -100,6 +101,12 @@ "> \n" "> Can you please take a look?\n" "\n" - Yep, thanks for the report. + "Yep, thanks for the report.\n" + "\n" + "--\n" + "To unsubscribe, send a message with 'unsubscribe linux-mm' in\n" + "the body to majordomo@kvack.org. For more info on Linux MM,\n" + "see: http://www.linux-mm.org/ .\n" + "Don't email: <a href=mailto:\"dont@kvack.org\"> email@kvack.org </a>" -fb7b47181b4bca9d9feaa7b813533219a247f4a52ae84389b16d9529803a8def +029f2ee6c85203cb936db678c54469100e6f24138b04d98aedcdd760e5d61510
diff --git a/a/1.txt b/N3/1.txt index 0920f14..7840fb2 100644 --- a/a/1.txt +++ b/N3/1.txt @@ -3,60 +3,60 @@ On Wed, Aug 31, 2016 at 08:20:48PM +0000, Kani, Toshimitsu wrote: > > On Tue, Aug 23, 2016 at 04:04:10PM -0600, Ross Zwisler wrote: > > > > > > DAX PMDs have been disabled since Jan Kara introduced DAX radix -> > > tree based locking. This series allows DAX PMDs to participate in +> > > tree based locking. This series allows DAX PMDs to participate in > > > the DAX radix tree based locking scheme so that they can be re- > > > enabled. > > > > > > Changes since v1: -> > > - PMD entry locking is now done based on the starting offset of +> > > - PMD entry locking is now done based on the starting offset of > > > the PMD entry, rather than on the radix tree slot which was > > > unreliable. (Jan) -> > > - Fixed the one issue I could find with hole punch. As far as I +> > > - Fixed the one issue I could find with hole punch. As far as I > > > can tell hole punch now works correctly for both PMD and PTE DAX > > > entries, 4k zero pages and huge zero pages. -> > > - Fixed the way that ext2 returns the size of holes in +> > > - Fixed the way that ext2 returns the size of holes in > > > ext2_get_block(). (Jan) -> > > - Made the 'wait_table' global variable static in respnse to a +> > > - Made the 'wait_table' global variable static in respnse to a > > > sparse warning. -> > > - Fixed some more inconsitent usage between the names 'ret' and +> > > - Fixed some more inconsitent usage between the names 'ret' and > > > 'entry' for radix tree entry variables. > > > > > > Ross Zwisler (9): -> > > ext4: allow DAX writeback for hole punch -> > > ext2: tell DAX the size of allocation holes -> > > ext4: tell DAX the size of allocation holes -> > > dax: remove buffer_size_valid() -> > > dax: make 'wait_table' global variable static -> > > dax: consistent variable naming for DAX entries -> > > dax: coordinate locking for offsets in PMD range -> > > dax: re-enable DAX PMD support -> > > dax: remove "depends on BROKEN" from FS_DAX_PMD +> > > ext4: allow DAX writeback for hole punch +> > > ext2: tell DAX the size of allocation holes +> > > ext4: tell DAX the size of allocation holes +> > > dax: remove buffer_size_valid() +> > > dax: make 'wait_table' global variable static +> > > dax: consistent variable naming for DAX entries +> > > dax: coordinate locking for offsets in PMD range +> > > dax: re-enable DAX PMD support +> > > dax: remove "depends on BROKEN" from FS_DAX_PMD > > > -> > > fs/Kconfig | 1 - -> > > fs/dax.c | 297 +++++++++++++++++++++++++++++----------- +> > > fs/Kconfig | 1 - +> > > fs/dax.c | 297 +++++++++++++++++++++++++++++----------- > > > ------------ -> > > fs/ext2/inode.c | 3 + -> > > fs/ext4/inode.c | 7 +- -> > > include/linux/dax.h | 29 ++++- -> > > mm/filemap.c | 6 +- -> > > 6 files changed, 201 insertions(+), 142 deletions(-) +> > > fs/ext2/inode.c | 3 + +> > > fs/ext4/inode.c | 7 +- +> > > include/linux/dax.h | 29 ++++- +> > > mm/filemap.c | 6 +- +> > > 6 files changed, 201 insertions(+), 142 deletions(-) > > > -> > > -- +> > > -- > > > 2.9.0 > > -> > Ping on this series? Any objections or comments? +> > Ping on this series? Any objections or comments? > > Hi Ross, > > I am seeing a major performance loss in fio mmap test with this patch- -> set applied. This happens with or without my patches [1] applied on -> top of yours. Without my patches, dax_pmd_fault() falls back to the +> set applied. This happens with or without my patches [1] applied on +> top of yours. Without my patches, dax_pmd_fault() falls back to the > pte handler since an mmap'ed address is not 2MB-aligned. > > I have attached three test results. -> o rc4.log - 4.8.0-rc4 (base) -> o non-pmd.log - 4.8.0-rc4 + your patchset (fall back to pte) -> o pmd.log - 4.8.0-rc4 + your patchset + my patchset (use pmd maps) +> o rc4.log - 4.8.0-rc4 (base) +> o non-pmd.log - 4.8.0-rc4 + your patchset (fall back to pte) +> o pmd.log - 4.8.0-rc4 + your patchset + my patchset (use pmd maps) > > My test steps are as follows. > @@ -78,3 +78,9 @@ On Wed, Aug 31, 2016 at 08:20:48PM +0000, Kani, Toshimitsu wrote: > Can you please take a look? Yep, thanks for the report. + +-- +To unsubscribe, send a message with 'unsubscribe linux-mm' in +the body to majordomo@kvack.org. For more info on Linux MM, +see: http://www.linux-mm.org/ . +Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a> diff --git a/a/content_digest b/N3/content_digest index a1aa253..1bb81d7 100644 --- a/a/content_digest +++ b/N3/content_digest @@ -1,24 +1,25 @@ "ref\020160823220419.11717-1-ross.zwisler@linux.intel.com\0" "ref\020160830230150.GA12173@linux.intel.com\0" "ref\01472674799.2092.19.camel@hpe.com\0" - "ref\01472674799.2092.19.camel-ZPxbGqLxI0U@public.gmane.org\0" - "From\0Ross Zwisler <ross.zwisler-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>\0" + "From\0Ross Zwisler <ross.zwisler@linux.intel.com>\0" "Subject\0Re: [PATCH v2 0/9] re-enable DAX PMD support\0" "Date\0Wed, 31 Aug 2016 15:36:07 -0600\0" "To\0Kani" - " Toshimitsu <toshi.kani-ZPxbGqLxI0U@public.gmane.org>\0" - "Cc\0tytso-3s7WtUTddSA@public.gmane.org <tytso-3s7WtUTddSA@public.gmane.org>" - akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org <akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org> - mawilcox-0li6OtcxBFHby3iVrkZq2A@public.gmane.org <mawilcox-0li6OtcxBFHby3iVrkZq2A@public.gmane.org> - linux-nvdimm-hn68Rpc1hR1g9hUCZPvPmw@public.gmane.org <linux-nvdimm-hn68Rpc1hR1g9hUCZPvPmw@public.gmane.org> - david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org <david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org> - linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org <linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org> - linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org <linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org> - adilger.kernel-m1MBpc4rdrD3fQ9qLvQP4Q@public.gmane.org <adilger.kernel-m1MBpc4rdrD3fQ9qLvQP4Q@public.gmane.org> - viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org <viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org> - jack-IBi9RG/b67k@public.gmane.org <jack-IBi9RG/b67k@public.gmane.org> - linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org <linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org> - " linux-ext4-u79uwXL29TY76Z2rM5mHXA@public.gmane.org <linux-ext4-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>\0" + " Toshimitsu <toshi.kani@hpe.com>\0" + "Cc\0ross.zwisler@linux.intel.com <ross.zwisler@linux.intel.com>" + linux-kernel@vger.kernel.org <linux-kernel@vger.kernel.org> + linux-mm@kvack.org <linux-mm@kvack.org> + adilger.kernel@dilger.ca <adilger.kernel@dilger.ca> + viro@zeniv.linux.org.uk <viro@zeniv.linux.org.uk> + dan.j.williams@intel.com <dan.j.williams@intel.com> + akpm@linux-foundation.org <akpm@linux-foundation.org> + mawilcox@microsoft.com <mawilcox@microsoft.com> + linux-nvdimm@lists.01.org <linux-nvdimm@lists.01.org> + linux-fsdevel@vger.kernel.org <linux-fsdevel@vger.kernel.org> + jack@suse.com <jack@suse.com> + tytso@mit.edu <tytso@mit.edu> + linux-ext4@vger.kernel.org <linux-ext4@vger.kernel.org> + " david@fromorbit.com <david@fromorbit.com>\0" "\00:1\0" "b\0" "On Wed, Aug 31, 2016 at 08:20:48PM +0000, Kani, Toshimitsu wrote:\n" @@ -26,60 +27,60 @@ "> > On Tue, Aug 23, 2016 at 04:04:10PM -0600, Ross Zwisler wrote:\n" "> > > \n" "> > > DAX PMDs have been disabled since Jan Kara introduced DAX radix\n" - "> > > tree based locking.\302\240\302\240This series allows DAX PMDs to participate in\n" + "> > > tree based locking. This series allows DAX PMDs to participate in\n" "> > > the DAX radix tree based locking scheme so that they can be re-\n" "> > > enabled.\n" "> > > \n" "> > > Changes since v1:\n" - "> > > \302\240- PMD entry locking is now done based on the starting offset of\n" + "> > > - PMD entry locking is now done based on the starting offset of\n" "> > > the PMD entry, rather than on the radix tree slot which was\n" "> > > unreliable. (Jan)\n" - "> > > \302\240- Fixed the one issue I could find with hole punch.\302\240\302\240As far as I\n" + "> > > - Fixed the one issue I could find with hole punch. As far as I\n" "> > > can tell hole punch now works correctly for both PMD and PTE DAX\n" "> > > entries, 4k zero pages and huge zero pages.\n" - "> > > \302\240- Fixed the way that ext2 returns the size of holes in\n" + "> > > - Fixed the way that ext2 returns the size of holes in\n" "> > > ext2_get_block(). (Jan)\n" - "> > > \302\240- Made the 'wait_table' global variable static in respnse to a\n" + "> > > - Made the 'wait_table' global variable static in respnse to a\n" "> > > sparse warning.\n" - "> > > \302\240- Fixed some more inconsitent usage between the names 'ret' and\n" + "> > > - Fixed some more inconsitent usage between the names 'ret' and\n" "> > > 'entry' for radix tree entry variables.\n" "> > > \n" "> > > Ross Zwisler (9):\n" - "> > > \302\240 ext4: allow DAX writeback for hole punch\n" - "> > > \302\240 ext2: tell DAX the size of allocation holes\n" - "> > > \302\240 ext4: tell DAX the size of allocation holes\n" - "> > > \302\240 dax: remove buffer_size_valid()\n" - "> > > \302\240 dax: make 'wait_table' global variable static\n" - "> > > \302\240 dax: consistent variable naming for DAX entries\n" - "> > > \302\240 dax: coordinate locking for offsets in PMD range\n" - "> > > \302\240 dax: re-enable DAX PMD support\n" - "> > > \302\240 dax: remove \"depends on BROKEN\" from FS_DAX_PMD\n" + "> > > ext4: allow DAX writeback for hole punch\n" + "> > > ext2: tell DAX the size of allocation holes\n" + "> > > ext4: tell DAX the size of allocation holes\n" + "> > > dax: remove buffer_size_valid()\n" + "> > > dax: make 'wait_table' global variable static\n" + "> > > dax: consistent variable naming for DAX entries\n" + "> > > dax: coordinate locking for offsets in PMD range\n" + "> > > dax: re-enable DAX PMD support\n" + "> > > dax: remove \"depends on BROKEN\" from FS_DAX_PMD\n" "> > > \n" - "> > > \302\240fs/Kconfig\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240|\302\240\302\240\302\2401 -\n" - "> > > \302\240fs/dax.c\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240| 297 +++++++++++++++++++++++++++++-----------\n" + "> > > fs/Kconfig | 1 -\n" + "> > > fs/dax.c | 297 +++++++++++++++++++++++++++++-----------\n" "> > > ------------\n" - "> > > \302\240fs/ext2/inode.c\302\240\302\240\302\240\302\240\302\240|\302\240\302\240\302\2403 +\n" - "> > > \302\240fs/ext4/inode.c\302\240\302\240\302\240\302\240\302\240|\302\240\302\240\302\2407 +-\n" - "> > > \302\240include/linux/dax.h |\302\240\302\24029 ++++-\n" - "> > > \302\240mm/filemap.c\302\240\302\240\302\240\302\240\302\240\302\240\302\240\302\240|\302\240\302\240\302\2406 +-\n" - "> > > \302\2406 files changed, 201 insertions(+), 142 deletions(-)\n" + "> > > fs/ext2/inode.c | 3 +\n" + "> > > fs/ext4/inode.c | 7 +-\n" + "> > > include/linux/dax.h | 29 ++++-\n" + "> > > mm/filemap.c | 6 +-\n" + "> > > 6 files changed, 201 insertions(+), 142 deletions(-)\n" "> > > \n" - "> > > --\302\240\n" + "> > > -- \n" "> > > 2.9.0\n" "> > \n" - "> > Ping on this series?\302\240\302\240Any objections or comments?\n" + "> > Ping on this series? Any objections or comments?\n" "> \n" "> Hi Ross,\n" "> \n" "> I am seeing a major performance loss in fio mmap test with this patch-\n" - "> set applied. \302\240This happens with or without my patches [1] applied on\n" - "> top of yours. \302\240Without my patches,\302\240dax_pmd_fault() falls back to the\n" + "> set applied. This happens with or without my patches [1] applied on\n" + "> top of yours. Without my patches, dax_pmd_fault() falls back to the\n" "> pte handler since an mmap'ed address is not 2MB-aligned.\n" "> \n" "> I have attached three test results.\n" - "> \302\240o rc4.log - 4.8.0-rc4 (base)\n" - "> \302\240o non-pmd.log - 4.8.0-rc4 + your patchset (fall back to pte)\n" - "> \302\240o pmd.log - 4.8.0-rc4 + your patchset + my patchset (use pmd maps)\n" + "> o rc4.log - 4.8.0-rc4 (base)\n" + "> o non-pmd.log - 4.8.0-rc4 + your patchset (fall back to pte)\n" + "> o pmd.log - 4.8.0-rc4 + your patchset + my patchset (use pmd maps)\n" "> \n" "> My test steps are as follows.\n" "> \n" @@ -100,6 +101,12 @@ "> \n" "> Can you please take a look?\n" "\n" - Yep, thanks for the report. + "Yep, thanks for the report.\n" + "\n" + "--\n" + "To unsubscribe, send a message with 'unsubscribe linux-mm' in\n" + "the body to majordomo@kvack.org. For more info on Linux MM,\n" + "see: http://www.linux-mm.org/ .\n" + "Don't email: <a href=mailto:\"dont@kvack.org\"> email@kvack.org </a>" -fb7b47181b4bca9d9feaa7b813533219a247f4a52ae84389b16d9529803a8def +c6764301cf34335bbbe450e4adb3393a18a79359b88216ab6e5b50412518d5e5
diff --git a/a/content_digest b/N4/content_digest index a1aa253..8991952 100644 --- a/a/content_digest +++ b/N4/content_digest @@ -1,24 +1,25 @@ "ref\020160823220419.11717-1-ross.zwisler@linux.intel.com\0" "ref\020160830230150.GA12173@linux.intel.com\0" "ref\01472674799.2092.19.camel@hpe.com\0" - "ref\01472674799.2092.19.camel-ZPxbGqLxI0U@public.gmane.org\0" - "From\0Ross Zwisler <ross.zwisler-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>\0" + "From\0Ross Zwisler <ross.zwisler@linux.intel.com>\0" "Subject\0Re: [PATCH v2 0/9] re-enable DAX PMD support\0" "Date\0Wed, 31 Aug 2016 15:36:07 -0600\0" "To\0Kani" - " Toshimitsu <toshi.kani-ZPxbGqLxI0U@public.gmane.org>\0" - "Cc\0tytso-3s7WtUTddSA@public.gmane.org <tytso-3s7WtUTddSA@public.gmane.org>" - akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org <akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org> - mawilcox-0li6OtcxBFHby3iVrkZq2A@public.gmane.org <mawilcox-0li6OtcxBFHby3iVrkZq2A@public.gmane.org> - linux-nvdimm-hn68Rpc1hR1g9hUCZPvPmw@public.gmane.org <linux-nvdimm-hn68Rpc1hR1g9hUCZPvPmw@public.gmane.org> - david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org <david-FqsqvQoI3Ljby3iVrkZq2A@public.gmane.org> - linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org <linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org> - linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org <linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org> - adilger.kernel-m1MBpc4rdrD3fQ9qLvQP4Q@public.gmane.org <adilger.kernel-m1MBpc4rdrD3fQ9qLvQP4Q@public.gmane.org> - viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org <viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn@public.gmane.org> - jack-IBi9RG/b67k@public.gmane.org <jack-IBi9RG/b67k@public.gmane.org> - linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org <linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org> - " linux-ext4-u79uwXL29TY76Z2rM5mHXA@public.gmane.org <linux-ext4-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>\0" + " Toshimitsu <toshi.kani@hpe.com>\0" + "Cc\0ross.zwisler@linux.intel.com <ross.zwisler@linux.intel.com>" + linux-kernel@vger.kernel.org <linux-kernel@vger.kernel.org> + linux-mm@kvack.org <linux-mm@kvack.org> + adilger.kernel@dilger.ca <adilger.kernel@dilger.ca> + viro@zeniv.linux.org.uk <viro@zeniv.linux.org.uk> + dan.j.williams@intel.com <dan.j.williams@intel.com> + akpm@linux-foundation.org <akpm@linux-foundation.org> + mawilcox@microsoft.com <mawilcox@microsoft.com> + linux-nvdimm@lists.01.org <linux-nvdimm@ml01.01.org> + linux-fsdevel@vger.kernel.org <linux-fsdevel@vger.kernel.org> + jack@suse.com <jack@suse.com> + tytso@mit.edu <tytso@mit.edu> + linux-ext4@vger.kernel.org <linux-ext4@vger.kernel.org> + " david@fromorbit.com <david@fromorbit.com>\0" "\00:1\0" "b\0" "On Wed, Aug 31, 2016 at 08:20:48PM +0000, Kani, Toshimitsu wrote:\n" @@ -102,4 +103,4 @@ "\n" Yep, thanks for the report. -fb7b47181b4bca9d9feaa7b813533219a247f4a52ae84389b16d9529803a8def +9d7e920e69a6054c86bff2d21f4b8472bb0553f1c3ec4f9c036d5e5f2529d995
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.