Thanks Eric, as we move fwd, these patch will be better. --- Cheers! Akshay Lal On Wed, Sep 8, 2010 at 11:51 AM, Eric Sandeen <sandeen@xxxxxxxxxxx> wrote: > Akshay Lal wrote: >> There was a missing fi at the the end of the _check_ext4_eof_flag method. >> >> Also corrected some indentation. > > Thanks, I've committed this version. > > In the future, keeping patch resubmissions tidy is helpful, i.e. > > Subject: [PATCH Vx] subsystem: description > > patch description > > Signed-off-by: > --- > other information that won't go into the commit > > patch itself > > ... without the other cruft, forwarded snippets, etc :) > > see also Documentation/SubmittingPatches > in the kernel tree. > > Thanks, > > -Eric > >> Signed-off-by: Akshay Lal <alal@xxxxxxxxxx> >> --------------------------------------------------------------------------------------- >> Updated patch: >> --------------------------------------------------------------------------------------- >> From e6906071ab6c0ad38d3ee0a463b5c7944e71fd00 Mon Sep 17 00:00:00 2001 >> From: Akshay Lal <alal@xxxxxxxxxx> >> Date: Tue, 7 Sep 2010 12:54:33 -0700 >> Subject: [PATCH] Test to ensure that the EOFBLOCK_FL gets set/unset correctly. >> >> As found by Theodore Ts'o: >> If a 128K file is falloc'ed using the KEEP_SIZE flag, and then >> write exactly 128K, the EOFBLOCK_FL doesn't get cleared correctly. >> This is bad since it forces e2fsck to complain about that inode. >> If you have a large number of inodes that are written with fallocate >> using KEEP_SIZE, and then fill them up to their expected size, >> e2fsck will potentially complain about a _huge_ number of inodes. >> This would also cause a huge increase in the time taken by e2fsck >> to complete its check. >> >> Test scenarios covered: >> 1. Fallocating X bytes and writing Y (Y<X) (buffered and direct io) >> 2. Fallocating X bytes and writing Y (Y=X) (buffered and direct io) >> 3. Fallocating X bytes and writing Y (Y>X) (buffered and direct io) >> >> These test cases exercise the normal and edge case conditions using >> falloc (and KEEP_SIZE). >> >> Ref: http://thread.gmane.org/gmane.comp.file-systems.ext4/20682 >> --- >> 243 | 178 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ >> 243.out | 13 +++++ >> group | 1 + >> 3 files changed, 192 insertions(+), 0 deletions(-) >> create mode 100644 243 >> create mode 100644 243.out >> >> diff --git a/243 b/243 >> new file mode 100644 >> index 0000000..8b2e647 >> --- /dev/null >> +++ b/243 >> @@ -0,0 +1,178 @@ >> +#! /bin/bash >> +# FS QA Test No. 243 >> +# >> +# Test to ensure that the EOFBLOCK_FL gets set/unset correctly. >> +# >> +# As found by Theodore Ts'o: >> +# If a 128K file is falloc'ed using the KEEP_SIZE flag, and then >> +# write exactly 128K, the EOFBLOCK_FL doesn't get cleared correctly. >> +# This is bad since it forces e2fsck to complain about that inode. >> +# If you have a large number of inodes that are written with fallocate >> +# using KEEP_SIZE, and then fill them up to their expected size, >> +# e2fsck will potentially complain about a _huge_ number of inodes. >> +# This would also cause a huge increase in the time taken by e2fsck >> +# to complete its check. >> +# >> +# Test scenarios covered: >> +# 1. Fallocating X bytes and writing Y (Y<X) (buffered and direct io) >> +# 2. Fallocating X bytes and writing Y (Y=X) (buffered and direct io) >> +# 3. Fallocating X bytes and writing Y (Y>X) (buffered and direct io) >> +# >> +# These test cases exercise the normal and edge case conditions using >> +# falloc (and KEEP_SIZE). >> +# >> +# Ref: http://thread.gmane.org/gmane.comp.file-systems.ext4/20682 >> +# >> +#----------------------------------------------------------------------- >> +# Copyright (c) 2010 Google, Inc. All Rights Reserved. >> +# >> +# This program is free software; you can redistribute it and/or >> +# modify it under the terms of the GNU General Public License as >> +# published by the Free Software Foundation. >> +# >> +# This program is distributed in the hope that it would be useful, >> +# but WITHOUT ANY WARRANTY; without even the implied warranty of >> +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the >> +# GNU General Public License for more details. >> +# >> +# You should have received a copy of the GNU General Public License >> +# along with this program; if not, write the Free Software Foundation, >> +# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA >> +#----------------------------------------------------------------------- >> +# >> +# creator >> +owner=alal@xxxxxxxxxx >> + >> +seq=`basename $0` >> +echo "QA output created by $seq" >> + >> +here=`pwd` >> +tmp=/tmp/$$ >> +status=1 # failure is the default! >> +trap "_cleanup; exit \$status" 0 1 2 3 15 >> + >> +# Test specific macros. >> +BIT_NOT_SET=0 # inode flag - 0x400000 bit is not set. >> +BIT_SET=1 # inode flag - 0x400000 bit is set. >> + >> +# Generic test cleanup function. >> +_cleanup() >> +{ >> + cd / >> + rm -f $tmp.* >> +} >> + >> +# Ext4 uses the EOFBLOCKS_FL bit when fallocating blocks with KEEP_SIZE >> +# enabled. The only time this bit should be set is when extending the allocated >> +# blocks further than what the i_size represents. In the situations wherein the >> +# i_size covers all allocated blocks, this bit should be cleared. >> + >> +# Checks the state of the sample file in the filesystem and returns whether >> +# the inode flag 0x400000 is set or not. >> +_check_ext4_eof_flag() >> +{ >> + # Check whether EOFBLOCK_FL is set. >> + # For ext4 filesystems: use debugfs to check if EOFBLOCKS_FL is set. >> + # Other filesystems: do nothing. The default fsck at the end of the test >> + # should catch any potential errors. >> + if [ "${FSTYP}" == "ext4" ]; then >> + bit_set=1 >> + >> + # Unmount the ${TEST_DEV} >> + umount ${TEST_DEV} >> + >> + # Run debugfs to gather file_parameters - specifically iflags. >> + file_params=`debugfs ${TEST_DEV} -R "stat ${1}" 2>&1 | grep -e Flags:` >> + iflags=${file_params#*Flags: } >> + >> + # Ensure that the iflags value was parsed correctly. >> + if [ -z ${iflags} ]; then >> + echo "iFlags value was not parsed successfully." >> $seq.full >> + status=1 >> + exit ${status} >> + fi >> + >> + # Check if EOFBLOCKS_FL is set. >> + if ((${iflags} & 0x400000)); then >> + echo "EOFBLOCK_FL bit is set." >> $seq.full >> + bit_set=1 >> + else >> + echo "EOFBLOCK_FL bit is not set." >> $seq.full >> + bit_set=0 >> + fi >> + >> + # Check current bit state to expected value. >> + if [ ${bit_set} -ne ${2} ]; then >> + echo "Error: Current bit state incorrect." >> $seq.full >> + status=1 >> + exit ${status} >> + fi >> + >> + # Mount the ${TEST_DEV} >> + mount ${TEST_DEV} -t ${FSTYP} ${TEST_DIR} >> + fi >> +} >> + >> +# Get standard environment, filters and checks. >> +. ./common.rc >> +. ./common.filter >> + >> +# Prerequisites for the test run. >> +_supported_fs ext4 xfs btrfs gfs2 >> +_supported_os Linux >> +_require_xfs_io_falloc >> + >> +# Real QA test starts here. >> +rm -f $seq.full >> + >> +# Begin test cases. >> +echo "Test 1: Fallocate 40960 bytes and write 4096 bytes (buffered io)." \ >> + >> $seq.full >> +${XFS_IO_PROG} -F -f \ >> + -c 'falloc -k 0 40960' \ >> + -c 'pwrite 0 4096' \ >> + ${TEST_DIR}/test_1 | _filter_xfs_io_unique >> +_check_ext4_eof_flag test_1 ${BIT_SET} >> + >> +echo "Test 2: Fallocate 40960 bytes and write 4096 bytes (direct io)." \ >> + >> $seq.full >> +${XFS_IO_PROG} -F -f -d \ >> + -c 'falloc -k 0 40960' \ >> + -c 'pwrite 0 4096' \ >> + ${TEST_DIR}/test_2 | _filter_xfs_io_unique >> +_check_ext4_eof_flag test_2 ${BIT_SET} >> + >> +echo "Test 3: Fallocate 40960 bytes and write 40960 bytes (buffered io)." \ >> + >> $seq.full >> +${XFS_IO_PROG} -F -f \ >> + -c 'falloc -k 0 40960' \ >> + -c 'pwrite 0 40960' \ >> + ${TEST_DIR}/test_3 | _filter_xfs_io_unique >> +_check_ext4_eof_flag test_3 ${BIT_NOT_SET} >> + >> +echo "Test 4: Fallocate 40960 bytes and write 40960 bytes (direct io)." \ >> + >> $seq.full >> +${XFS_IO_PROG} -F -f -d \ >> + -c 'falloc -k 0 40960' \ >> + -c 'pwrite 0 40960' \ >> + ${TEST_DIR}/test_4 | _filter_xfs_io_unique >> +_check_ext4_eof_flag test_4 ${BIT_NOT_SET} >> + >> +echo "Test 5: Fallocate 128k, seek 256k and write 4k block (buffered io)." \ >> + >> $seq.full >> +${XFS_IO_PROG} -F -f \ >> + -c 'falloc -k 0 128k' \ >> + -c 'pwrite 256k 4k' \ >> + ${TEST_DIR}/test_5 | _filter_xfs_io_unique >> +_check_ext4_eof_flag test_5 ${BIT_NOT_SET} >> + >> +echo "Test 6: Fallocate 128k, seek to 256k and write a 4k block (direct io)." \ >> + >> $seq.full >> +${XFS_IO_PROG} -F -f -d \ >> + -c 'falloc -k 0 128k' \ >> + -c 'pwrite 256k 4k' \ >> + ${TEST_DIR}/test_6 | _filter_xfs_io_unique >> +_check_ext4_eof_flag test_6 ${BIT_NOT_SET} >> + >> +status=0 >> +exit ${status} >> diff --git a/243.out b/243.out >> new file mode 100644 >> index 0000000..290a005 >> --- /dev/null >> +++ b/243.out >> @@ -0,0 +1,13 @@ >> +QA output created by 243 >> +wrote 4096/4096 bytes at offset 0 >> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) >> +wrote 4096/4096 bytes at offset 0 >> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) >> +wrote 40960/40960 bytes at offset 0 >> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) >> +wrote 40960/40960 bytes at offset 0 >> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) >> +wrote 4096/4096 bytes at offset 262144 >> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) >> +wrote 4096/4096 bytes at offset 262144 >> +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) >> diff --git a/group b/group >> index ff16bb3..e6dab13 100644 >> --- a/group >> +++ b/group >> @@ -356,3 +356,4 @@ deprecated >> 240 auto aio quick rw >> 241 auto >> 242 auto quick prealloc >> +243 auto quick prealloc > > _______________________________________________ xfs mailing list xfs@xxxxxxxxxxx http://oss.sgi.com/mailman/listinfo/xfs
