From: "Darrick J. Wong" <darrick.wong@oracle.com> To: david@fromorbit.com, darrick.wong@oracle.com Cc: linux-btrfs@vger.kernel.org, fstests@vger.kernel.org, xfs@oss.sgi.com Subject: [PATCH 12/32] dio: unwritten conversion bug tests Date: Thu, 11 Feb 2016 15:40:37 -0800 [thread overview] Message-ID: <20160211234037.2202.93885.stgit@birch.djwong.org> (raw) In-Reply-To: <20160211233916.2202.40961.stgit@birch.djwong.org> Check that we don't expose old disk contents when a directio write to an unwritten extent fails due to IO errors. This primarily affects XFS and ext4. Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> --- .gitignore | 1 src/aio-dio-regress/aiocp.c | 489 +++++++++++++++++++++++++++++++++++++++++++ tests/generic/250 | 100 +++++++++ tests/generic/250.out | 10 + tests/generic/252 | 103 +++++++++ tests/generic/252.out | 10 + tests/generic/group | 2 7 files changed, 715 insertions(+) create mode 100644 src/aio-dio-regress/aiocp.c create mode 100755 tests/generic/250 create mode 100644 tests/generic/250.out create mode 100755 tests/generic/252 create mode 100644 tests/generic/252.out diff --git a/.gitignore b/.gitignore index a6f47d3..bbe7c1a 100644 --- a/.gitignore +++ b/.gitignore @@ -112,6 +112,7 @@ /src/aio-dio-regress/aio-free-ring-with-bogus-nr-pages /src/aio-dio-regress/aio-io-setup-with-nonwritable-context-pointer /src/aio-dio-regress/aio-last-ref-held-by-io +/src/aio-dio-regress/aiocp /src/aio-dio-regress/aiodio_sparse2 /src/aio-dio-regress/aio-dio-eof-race /src/cloner diff --git a/src/aio-dio-regress/aiocp.c b/src/aio-dio-regress/aiocp.c new file mode 100644 index 0000000..1abff9c --- /dev/null +++ b/src/aio-dio-regress/aiocp.c @@ -0,0 +1,489 @@ +/* + * Copyright (c) 2004 Daniel McNeil <daniel@osdl.org> + * 2004 Open Source Development Lab + * 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; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will 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 to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Module: .c + */ + +/* + * Change History: + * + * + * version of copy command using async i/o + * From: Stephen Hemminger <shemminger@osdl.org> + * Modified by Daniel McNeil <daniel@osdl.org> for testing aio. + * - added -a alignment + * - added -b blksize option + * _ added -s size option + * - added -f open_flag option + * - added -w (no write) option (reads from source only) + * - added -n (num aio) option + * - added -z (zero dest) opton (writes zeros to dest only) + * - added -D delay_ms option + * - 2/2004 Marty Ridgeway (mridge@us.ibm.com) Changes to adapt to LTP + * + * Copy file by using a async I/O state machine. + * 1. Start read request + * 2. When read completes turn it into a write request + * 3. When write completes decrement counter and free resources + * + * + * Usage: aiocp [-b blksize] -n [num_aio] [-w] [-z] [-s filesize] + * [-f DIRECT|TRUNC|CREAT|SYNC|LARGEFILE] src dest + */ + +//#define _GNU_SOURCE +//#define DEBUG 1 +#undef DEBUG + +#include <unistd.h> +#include <stdio.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <sys/param.h> +#include <fcntl.h> +#include <errno.h> +#include <stdlib.h> +#include <sys/select.h> + +#include <libaio.h> + +#define AIO_BLKSIZE (64*1024) +#define AIO_MAXIO 32 + +static int aio_blksize = AIO_BLKSIZE; +static int aio_maxio = AIO_MAXIO; + +static int busy = 0; // # of I/O's in flight +static int tocopy = 0; // # of blocks left to copy +static int srcfd; // source fd +static int dstfd = -1; // destination file descriptor +static const char *dstname = NULL; +static const char *srcname = NULL; +static int source_open_flag = O_RDONLY; /* open flags on source file */ +static int dest_open_flag = O_WRONLY; /* open flags on dest file */ +static int no_write; /* do not write */ +static int zero; /* write zero's only */ + +static int debug; +static int count_io_q_waits; /* how many time io_queue_wait called */ + +struct iocb **iocb_free; /* array of pointers to iocb */ +int iocb_free_count; /* current free count */ +int alignment = 512; /* buffer alignment */ + +struct timeval delay; /* delay between i/o */ + +int init_iocb(int n, int iosize) +{ + void *buf; + int i; + + if ((iocb_free = malloc(n * sizeof(struct iocb *))) == 0) { + return -1; + } + + for (i = 0; i < n; i++) { + if (!(iocb_free[i] = (struct iocb *) malloc(sizeof(struct iocb)))) + return -1; + if (posix_memalign(&buf, alignment, iosize)) + return -1; + if (debug > 1) { + printf("buf allocated at 0x%p, align:%d\n", + buf, alignment); + } + if (zero) { + /* + * We are writing zero's to dstfd + */ + memset(buf, 0, iosize); + } + io_prep_pread(iocb_free[i], -1, buf, iosize, 0); + } + iocb_free_count = i; + return 0; +} + +struct iocb *alloc_iocb() +{ + if (!iocb_free_count) + return 0; + return iocb_free[--iocb_free_count]; +} + +void free_iocb(struct iocb *io) +{ + iocb_free[iocb_free_count++] = io; +} + +/* + * io_wait_run() - wait for an io_event and then call the callback. + */ +int io_wait_run(io_context_t ctx, struct timespec *to) +{ + struct io_event events[aio_maxio]; + struct io_event *ep; + int ret, n; + + /* + * get up to aio_maxio events at a time. + */ + ret = n = io_getevents(ctx, 1, aio_maxio, events, to); + + /* + * Call the callback functions for each event. + */ + for (ep = events; n-- > 0; ep++) { + io_callback_t cb = (io_callback_t)ep->data; + struct iocb *iocb = (struct iocb *)ep->obj; + + cb(ctx, iocb, ep->res, ep->res2); + } + return ret; +} + +/* Fatal error handler */ +static void io_error(const char *func, int rc) +{ + if (rc == -ENOSYS) + fprintf(stderr, "AIO not in this kernel\n"); + else if (rc < 0) + fprintf(stderr, "%s: %s\n", func, strerror(-rc)); + else + fprintf(stderr, "%s: error %d\n", func, rc); + + if (dstfd > 0) + close(dstfd); + if (dstname && dest_open_flag & O_CREAT) + unlink(dstname); + exit(1); +} + +/* + * Write complete callback. + * Adjust counts and free resources + */ +static void wr_done(io_context_t ctx, struct iocb *iocb, long res, long res2) +{ + if (res2 != 0) { + io_error("aio write", res2); + } + if (res != iocb->u.c.nbytes) { + fprintf(stderr, "write missed bytes expect %lu got %ld\n", + iocb->u.c.nbytes, res2); + exit(1); + } + --tocopy; + --busy; + free_iocb(iocb); + if (debug) + write(2, "w", 1); +} + +/* + * Read complete callback. + * Change read iocb into a write iocb and start it. + */ +static void rd_done(io_context_t ctx, struct iocb *iocb, long res, long res2) +{ + /* library needs accessors to look at iocb? */ + int iosize = iocb->u.c.nbytes; + char *buf = iocb->u.c.buf; + off_t offset = iocb->u.c.offset; + + if (res2 != 0) + io_error("aio read", res2); + if (res != iosize) { + fprintf(stderr, "read missing bytes expect %lu got %ld\n", + iocb->u.c.nbytes, res); + exit(1); + } + + + /* turn read into write */ + if (no_write) { + --tocopy; + --busy; + free_iocb(iocb); + } else { + io_prep_pwrite(iocb, dstfd, buf, iosize, offset); + io_set_callback(iocb, wr_done); + if (1 != (res = io_submit(ctx, 1, &iocb))) + io_error("io_submit write", res); + } + if (debug) + write(2, "r", 1); + if (debug > 1) + printf("%d", iosize); +} + +void usage() +{ + fprintf(stderr, + "Usage: aiocp [-a align] [-s size] [-b blksize] [-n num_io]" + " [-f open_flag] SOURCE DEST\n" + "This copies from SOURCE to DEST using AIO.\n\n" + "Usage: aiocp [options] -w SOURCE\n" + "This does sequential AIO reads (no writes).\n\n" + "Usage: aiocp [options] -z DEST\n" + "This does sequential AIO writes of zeros.\n"); + + exit(1); +} + +/* + * Scale value by kilo, mega, or giga. + */ +long long scale_by_kmg(long long value, char scale) +{ + switch (scale) { + case 'g': + case 'G': + value *= 1024; + case 'm': + case 'M': + value *= 1024; + case 'k': + case 'K': + value *= 1024; + break; + case '\0': + break; + default: + usage(); + break; + } + return value; +} + +int main(int argc, char *const *argv) +{ + struct stat st; + off_t length = 0, offset = 0; + io_context_t myctx; + int c; + extern char *optarg; + extern int optind, opterr, optopt; + + while ((c = getopt(argc, argv, "a:b:df:n:s:wzD:")) != -1) { + char *endp; + + switch (c) { + case 'a': /* alignment of data buffer */ + alignment = strtol(optarg, &endp, 0); + alignment = (long)scale_by_kmg((long long)alignment, + *endp); + break; + case 'f': /* use these open flags */ + if (strcmp(optarg, "LARGEFILE") == 0 || + strcmp(optarg, "O_LARGEFILE") == 0) { + source_open_flag |= O_LARGEFILE; + dest_open_flag |= O_LARGEFILE; + } else if (strcmp(optarg, "TRUNC") == 0 || + strcmp(optarg, "O_TRUNC") == 0) { + dest_open_flag |= O_TRUNC; + } else if (strcmp(optarg, "SYNC") == 0 || + strcmp(optarg, "O_SYNC") == 0) { + dest_open_flag |= O_SYNC | O_NONBLOCK; + } else if (strcmp(optarg, "DIRECT") == 0 || + strcmp(optarg, "O_DIRECT") == 0) { + source_open_flag |= O_DIRECT; + dest_open_flag |= O_DIRECT; + } else if (strncmp(optarg, "CREAT", 5) == 0 || + strncmp(optarg, "O_CREAT", 5) == 0) { + dest_open_flag |= O_CREAT; + } + break; + case 'd': + debug++; + break; + case 'D': + delay.tv_usec = atoi(optarg); + break; + case 'b': /* block size */ + aio_blksize = strtol(optarg, &endp, 0); + aio_blksize = (long)scale_by_kmg((long long)aio_blksize, *endp); + break; + + case 'n': /* num io */ + aio_maxio = strtol(optarg, &endp, 0); + break; + case 's': /* size to transfer */ + length = strtoll(optarg, &endp, 0); + length = scale_by_kmg(length, *endp); + break; + case 'w': /* no write */ + no_write = 1; + break; + case 'z': /* write zero's */ + zero = 1; + break; + + default: + usage(); + } + } + + argc -= optind; + argv += optind; + +#ifndef DEBUG + if (argc < 1) { + usage(); + } +#else + source_open_flag |= O_DIRECT; + dest_open_flag |= O_DIRECT; + aio_blksize = 1; + aio_maxio=1; + srcname = "junkdata"; + dstname = "ff2"; +#endif + if (!zero) { +#ifndef DEBUG + if ((srcfd = open(srcname = *argv, source_open_flag)) < 0) { +#else + if ((srcfd = open(srcname, source_open_flag)) < 0) { +#endif + perror(srcname); + exit(1); + } + argv++; + argc--; + if (fstat(srcfd, &st) < 0) { + perror("fstat"); + exit(1); + } + if (length == 0) + length = st.st_size; + } + + if (!no_write) { + /* + * We are either copying or writing zeros to dstname + */ +#ifndef DEBUG + if (argc < 1) { + usage(); + } + if ((dstfd = open(dstname = *argv, dest_open_flag, 0666)) < 0) { +#else + if ((dstfd = open(dstname, dest_open_flag, 0666)) < 0) { +#endif + perror(dstname); + exit(1); + } + if (zero) { + /* + * get size of dest, if we are zeroing it. + * TODO: handle devices. + */ + if (fstat(dstfd, &st) < 0) { + perror("fstat"); + exit(1); + } + if (length == 0) + length = st.st_size; + } + } + + /* initialize state machine */ + memset(&myctx, 0, sizeof(myctx)); + io_queue_init(aio_maxio, &myctx); + tocopy = howmany(length, aio_blksize); +printf("tocopy=%d len=%d blk=%d\n", tocopy, length, aio_blksize); + if (init_iocb(aio_maxio, aio_blksize) < 0) { + fprintf(stderr, "Error allocating the i/o buffers\n"); + exit(1); + } + + while (tocopy > 0) { + int i, rc; + /* Submit as many reads as once as possible upto aio_maxio */ + int n = MIN(MIN(aio_maxio - busy, aio_maxio), + howmany(length - offset, aio_blksize)); + if (n > 0) { + struct iocb *ioq[n]; + + for (i = 0; i < n; i++) { + struct iocb *io = alloc_iocb(); + int iosize = MIN(length - offset, aio_blksize); + + if (zero) { + /* + * We are writing zero's to dstfd + */ + io_prep_pwrite(io, dstfd, io->u.c.buf, + iosize, offset); + io_set_callback(io, wr_done); + } else { + io_prep_pread(io, srcfd, io->u.c.buf, + iosize, offset); + io_set_callback(io, rd_done); + } + ioq[i] = io; + offset += iosize; + } + + rc = io_submit(myctx, n, ioq); + if (rc < 0) + io_error("io_submit", rc); + + busy += n; + if (debug > 1) + printf("io_submit(%d) busy:%d\n", n, busy); + if (delay.tv_usec) { + struct timeval t = delay; + (void)select(0,0,0,0,&t); + } + } + + /* + * We have submitted all the i/o requests. Wait for at least one to complete + * and call the callbacks. + */ + count_io_q_waits++; + rc = io_wait_run(myctx, 0); + if (rc < 0) + io_error("io_wait_run", rc); + + if (debug > 1) { + printf("io_wait_run: rc == %d\n", rc); + printf("busy:%d aio_maxio:%d tocopy:%d\n", + busy, aio_maxio, tocopy); + } + } + + if (srcfd != -1) + close(srcfd); + if (dstfd != -1) + close(dstfd); + exit(0); +} + +/* + * Results look like: + * [alanm@toolbox ~/MOT3]$ ../taio -d kernel-source-2.4.8-0.4g.ppc.rpm abc + * rrrrrrrrrrrrrrrwwwrwrrwwrrwrwwrrwrwrwwrrwrwrrrrwwrwwwrrwrrrwwwwwwwwwwwwwwwww + * rrrrrrrrrrrrrrwwwrrwrwrwrwrrwwwwwwwwwwwwwwrrrrrrrrrrrrrrrrrrwwwwrwrwwrwrwrwr + * wrrrrrrrwwwwwwwwwwwwwrrrwrrrwrrwrwwwwwwwwwwrrrrwwrwrrrrrrrrrrrwwwwwwwwwwwrww + * wwwrrrrrrrrwwrrrwwrwrwrwwwrrrrrrrwwwrrwwwrrwrwwwwwwwwrrrrrrrwwwrrrrrrrwwwwww + * wwwwwwwrwrrrrrrrrwrrwrrwrrwrwrrrwrrrwrrrwrwwwwwwwwwwwwwwwwwwrrrwwwrrrrrrrrrr + * rrwrrrrrrwrrwwwwwwwwwwwwwwwwrwwwrrwrwwrrrrrrrrrrrrrrrrrrrwwwwwwwwwwwwwwwwwww + * rrrrrwrrwrwrwrrwrrrwwwwwwwwrrrrwrrrwrwwrwrrrwrrwrrrrwwwwwwwrwrwwwwrwwrrrwrrr + * rrrwwwwwwwrrrrwwrrrrrrrrrrrrwrwrrrrwwwwwwwwwwwwwwrwrrrrwwwwrwrrrrwrwwwrrrwww + * rwwrrrrrrrwrrrrrrrrrrrrwwwwrrrwwwrwrrwwwwwwwwwwwwwwwwwwwwwrrrrrrrwwwwwwwrw + */ diff --git a/tests/generic/250 b/tests/generic/250 new file mode 100755 index 0000000..b0b175a --- /dev/null +++ b/tests/generic/250 @@ -0,0 +1,100 @@ +#! /bin/bash +# FS QA Test No. 250 +# +# Create an unwritten extent, set up dm-error, try a DIO write, then +# make sure we can't read back old disk contents. +# +#----------------------------------------------------------------------- +# Copyright (c) 2016, Oracle and/or its affiliates. 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 +#----------------------------------------------------------------------- + +seq=`basename $0` +seqres=$RESULT_DIR/$seq +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 + +_cleanup() +{ + cd / + rm -rf $tmp.* $testdir + _dmerror_cleanup +} + +# get standard environment, filters and checks +. ./common/rc +. ./common/filter +. ./common/dmerror + +# real QA test starts here +_supported_os Linux +_require_scratch +_require_dm_target error +_require_xfs_io_command "falloc" + +rm -f $seqres.full + + +echo "Format and mount" +$XFS_IO_PROG -d -c "pwrite -S 0x69 -b 1048576 0 $((64 * 1048576))" $SCRATCH_DEV >> $seqres.full +_scratch_mkfs_sized $((64 * 1048576)) > $seqres.full 2>&1 +_dmerror_init +_dmerror_mount >> $seqres.full 2>&1 +_dmerror_unmount +_dmerror_mount + +testdir=$SCRATCH_MNT/test-$seq +mkdir $testdir + +blksz=65536 +nr=640 +bufnr=128 +filesize=$((blksz * nr)) +bufsize=$((blksz * bufnr)) + +_require_fs_space $SCRATCH_MNT $((filesize / 1024 * 5 / 4)) + +echo "Create the original files" +$XFS_IO_PROG -f -c "falloc 0 $filesize" $testdir/file2 >> $seqres.full +_dmerror_unmount +_dmerror_mount + +echo "Compare files" +md5sum $testdir/file2 | _filter_scratch + +echo "CoW and unmount" +$XFS_IO_PROG -f -c "pwrite -S 0x63 $bufsize 1" $testdir/file2 >> $seqres.full +sync +_dmerror_load_error_table +$XFS_IO_PROG -d -f -c "pwrite -S 0x63 -b $bufsize 0 $filesize" $testdir/file2 >> $seqres.full +_dmerror_load_working_table +_dmerror_unmount +_dmerror_mount + +echo "Compare files" +md5sum $testdir/file2 | _filter_scratch + +echo "Check for damage" +_dmerror_unmount +_dmerror_cleanup +_repair_scratch_fs >> $seqres.full + +# success, all done +status=0 +exit diff --git a/tests/generic/250.out b/tests/generic/250.out new file mode 100644 index 0000000..710c80e --- /dev/null +++ b/tests/generic/250.out @@ -0,0 +1,10 @@ +QA output created by 250 +Format and mount +Create the original files +Compare files +ec8bb3b24d5b0f1b5bdf8c8f0f541ee6 SCRATCH_MNT/test-250/file2 +CoW and unmount +pwrite64: Input/output error +Compare files +3ed86318f4ff8da26c1c2a6e3041f9be SCRATCH_MNT/test-250/file2 +Check for damage diff --git a/tests/generic/252 b/tests/generic/252 new file mode 100755 index 0000000..fc9a723 --- /dev/null +++ b/tests/generic/252 @@ -0,0 +1,103 @@ +#! /bin/bash +# FS QA Test No. 252 +# +# Create an unwritten extent, set up dm-error, try an AIO DIO write, then +# make sure we can't read back old disk contents. +# +#----------------------------------------------------------------------- +# Copyright (c) 2016, Oracle and/or its affiliates. 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 +#----------------------------------------------------------------------- + +seq=`basename $0` +seqres=$RESULT_DIR/$seq +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 + +_cleanup() +{ + cd / + rm -rf $tmp.* $testdir $TEST_DIR/moo + _dmerror_cleanup +} + +# get standard environment, filters and checks +. ./common/rc +. ./common/filter +. ./common/dmerror + +# real QA test starts here +_supported_os Linux +_require_scratch +_require_dm_target error +_require_xfs_io_command "falloc" +_require_test_program "aio-dio-regress/aiocp" +AIO_TEST="src/aio-dio-regress/aiocp" + +rm -f $seqres.full + + +echo "Format and mount" +$XFS_IO_PROG -d -c "pwrite -S 0x69 -b 1048576 0 $((64 * 1048576))" $SCRATCH_DEV >> $seqres.full +_scratch_mkfs_sized $((64 * 1048576)) > $seqres.full 2>&1 +_dmerror_init +_dmerror_mount >> $seqres.full 2>&1 +_dmerror_unmount +_dmerror_mount + +testdir=$SCRATCH_MNT/test-$seq +mkdir $testdir + +blksz=65536 +nr=640 +bufnr=128 +filesize=$((blksz * nr)) +bufsize=$((blksz * bufnr)) + +_require_fs_space $SCRATCH_MNT $((filesize / 1024 * 5 / 4)) + +echo "Create the original files" +$XFS_IO_PROG -f -c "falloc 0 $filesize" $testdir/file2 >> $seqres.full +_dmerror_unmount +_dmerror_mount + +echo "Compare files" +md5sum $testdir/file2 | _filter_scratch + +echo "CoW and unmount" +$XFS_IO_PROG -f -c "pwrite -S 0x63 $bufsize 1" $testdir/file2 >> $seqres.full +$XFS_IO_PROG -f -c "pwrite -S 0x63 -b $bufsize 0 $filesize" $TEST_DIR/moo >> $seqres.full +sync +_dmerror_load_error_table +$AIO_TEST -f DIRECT -b $bufsize $TEST_DIR/moo $testdir/file2 >> $seqres.full +_dmerror_load_working_table +_dmerror_unmount +_dmerror_mount + +echo "Compare files" +md5sum $testdir/file2 | _filter_scratch + +echo "Check for damage" +_dmerror_unmount +_dmerror_cleanup +_repair_scratch_fs >> $seqres.full + +# success, all done +status=0 +exit diff --git a/tests/generic/252.out b/tests/generic/252.out new file mode 100644 index 0000000..3bc78c2 --- /dev/null +++ b/tests/generic/252.out @@ -0,0 +1,10 @@ +QA output created by 252 +Format and mount +Create the original files +Compare files +ec8bb3b24d5b0f1b5bdf8c8f0f541ee6 SCRATCH_MNT/test-252/file2 +CoW and unmount +write missed bytes expect 8388608 got 0 +Compare files +3ed86318f4ff8da26c1c2a6e3041f9be SCRATCH_MNT/test-252/file2 +Check for damage diff --git a/tests/generic/group b/tests/generic/group index 860ff4a..0e4e7d3 100644 --- a/tests/generic/group +++ b/tests/generic/group @@ -252,7 +252,9 @@ 247 auto quick rw 248 auto quick rw 249 auto quick rw +250 auto quick prealloc rw 251 ioctl trim +252 auto quick prealloc rw 255 auto quick prealloc 256 auto quick 257 dir auto quick
WARNING: multiple messages have this Message-ID (diff)
From: "Darrick J. Wong" <darrick.wong@oracle.com> To: david@fromorbit.com, darrick.wong@oracle.com Cc: fstests@vger.kernel.org, linux-btrfs@vger.kernel.org, xfs@oss.sgi.com Subject: [PATCH 12/32] dio: unwritten conversion bug tests Date: Thu, 11 Feb 2016 15:40:37 -0800 [thread overview] Message-ID: <20160211234037.2202.93885.stgit@birch.djwong.org> (raw) In-Reply-To: <20160211233916.2202.40961.stgit@birch.djwong.org> Check that we don't expose old disk contents when a directio write to an unwritten extent fails due to IO errors. This primarily affects XFS and ext4. Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> --- .gitignore | 1 src/aio-dio-regress/aiocp.c | 489 +++++++++++++++++++++++++++++++++++++++++++ tests/generic/250 | 100 +++++++++ tests/generic/250.out | 10 + tests/generic/252 | 103 +++++++++ tests/generic/252.out | 10 + tests/generic/group | 2 7 files changed, 715 insertions(+) create mode 100644 src/aio-dio-regress/aiocp.c create mode 100755 tests/generic/250 create mode 100644 tests/generic/250.out create mode 100755 tests/generic/252 create mode 100644 tests/generic/252.out diff --git a/.gitignore b/.gitignore index a6f47d3..bbe7c1a 100644 --- a/.gitignore +++ b/.gitignore @@ -112,6 +112,7 @@ /src/aio-dio-regress/aio-free-ring-with-bogus-nr-pages /src/aio-dio-regress/aio-io-setup-with-nonwritable-context-pointer /src/aio-dio-regress/aio-last-ref-held-by-io +/src/aio-dio-regress/aiocp /src/aio-dio-regress/aiodio_sparse2 /src/aio-dio-regress/aio-dio-eof-race /src/cloner diff --git a/src/aio-dio-regress/aiocp.c b/src/aio-dio-regress/aiocp.c new file mode 100644 index 0000000..1abff9c --- /dev/null +++ b/src/aio-dio-regress/aiocp.c @@ -0,0 +1,489 @@ +/* + * Copyright (c) 2004 Daniel McNeil <daniel@osdl.org> + * 2004 Open Source Development Lab + * 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; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will 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 to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + * + * Module: .c + */ + +/* + * Change History: + * + * + * version of copy command using async i/o + * From: Stephen Hemminger <shemminger@osdl.org> + * Modified by Daniel McNeil <daniel@osdl.org> for testing aio. + * - added -a alignment + * - added -b blksize option + * _ added -s size option + * - added -f open_flag option + * - added -w (no write) option (reads from source only) + * - added -n (num aio) option + * - added -z (zero dest) opton (writes zeros to dest only) + * - added -D delay_ms option + * - 2/2004 Marty Ridgeway (mridge@us.ibm.com) Changes to adapt to LTP + * + * Copy file by using a async I/O state machine. + * 1. Start read request + * 2. When read completes turn it into a write request + * 3. When write completes decrement counter and free resources + * + * + * Usage: aiocp [-b blksize] -n [num_aio] [-w] [-z] [-s filesize] + * [-f DIRECT|TRUNC|CREAT|SYNC|LARGEFILE] src dest + */ + +//#define _GNU_SOURCE +//#define DEBUG 1 +#undef DEBUG + +#include <unistd.h> +#include <stdio.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <sys/param.h> +#include <fcntl.h> +#include <errno.h> +#include <stdlib.h> +#include <sys/select.h> + +#include <libaio.h> + +#define AIO_BLKSIZE (64*1024) +#define AIO_MAXIO 32 + +static int aio_blksize = AIO_BLKSIZE; +static int aio_maxio = AIO_MAXIO; + +static int busy = 0; // # of I/O's in flight +static int tocopy = 0; // # of blocks left to copy +static int srcfd; // source fd +static int dstfd = -1; // destination file descriptor +static const char *dstname = NULL; +static const char *srcname = NULL; +static int source_open_flag = O_RDONLY; /* open flags on source file */ +static int dest_open_flag = O_WRONLY; /* open flags on dest file */ +static int no_write; /* do not write */ +static int zero; /* write zero's only */ + +static int debug; +static int count_io_q_waits; /* how many time io_queue_wait called */ + +struct iocb **iocb_free; /* array of pointers to iocb */ +int iocb_free_count; /* current free count */ +int alignment = 512; /* buffer alignment */ + +struct timeval delay; /* delay between i/o */ + +int init_iocb(int n, int iosize) +{ + void *buf; + int i; + + if ((iocb_free = malloc(n * sizeof(struct iocb *))) == 0) { + return -1; + } + + for (i = 0; i < n; i++) { + if (!(iocb_free[i] = (struct iocb *) malloc(sizeof(struct iocb)))) + return -1; + if (posix_memalign(&buf, alignment, iosize)) + return -1; + if (debug > 1) { + printf("buf allocated at 0x%p, align:%d\n", + buf, alignment); + } + if (zero) { + /* + * We are writing zero's to dstfd + */ + memset(buf, 0, iosize); + } + io_prep_pread(iocb_free[i], -1, buf, iosize, 0); + } + iocb_free_count = i; + return 0; +} + +struct iocb *alloc_iocb() +{ + if (!iocb_free_count) + return 0; + return iocb_free[--iocb_free_count]; +} + +void free_iocb(struct iocb *io) +{ + iocb_free[iocb_free_count++] = io; +} + +/* + * io_wait_run() - wait for an io_event and then call the callback. + */ +int io_wait_run(io_context_t ctx, struct timespec *to) +{ + struct io_event events[aio_maxio]; + struct io_event *ep; + int ret, n; + + /* + * get up to aio_maxio events at a time. + */ + ret = n = io_getevents(ctx, 1, aio_maxio, events, to); + + /* + * Call the callback functions for each event. + */ + for (ep = events; n-- > 0; ep++) { + io_callback_t cb = (io_callback_t)ep->data; + struct iocb *iocb = (struct iocb *)ep->obj; + + cb(ctx, iocb, ep->res, ep->res2); + } + return ret; +} + +/* Fatal error handler */ +static void io_error(const char *func, int rc) +{ + if (rc == -ENOSYS) + fprintf(stderr, "AIO not in this kernel\n"); + else if (rc < 0) + fprintf(stderr, "%s: %s\n", func, strerror(-rc)); + else + fprintf(stderr, "%s: error %d\n", func, rc); + + if (dstfd > 0) + close(dstfd); + if (dstname && dest_open_flag & O_CREAT) + unlink(dstname); + exit(1); +} + +/* + * Write complete callback. + * Adjust counts and free resources + */ +static void wr_done(io_context_t ctx, struct iocb *iocb, long res, long res2) +{ + if (res2 != 0) { + io_error("aio write", res2); + } + if (res != iocb->u.c.nbytes) { + fprintf(stderr, "write missed bytes expect %lu got %ld\n", + iocb->u.c.nbytes, res2); + exit(1); + } + --tocopy; + --busy; + free_iocb(iocb); + if (debug) + write(2, "w", 1); +} + +/* + * Read complete callback. + * Change read iocb into a write iocb and start it. + */ +static void rd_done(io_context_t ctx, struct iocb *iocb, long res, long res2) +{ + /* library needs accessors to look at iocb? */ + int iosize = iocb->u.c.nbytes; + char *buf = iocb->u.c.buf; + off_t offset = iocb->u.c.offset; + + if (res2 != 0) + io_error("aio read", res2); + if (res != iosize) { + fprintf(stderr, "read missing bytes expect %lu got %ld\n", + iocb->u.c.nbytes, res); + exit(1); + } + + + /* turn read into write */ + if (no_write) { + --tocopy; + --busy; + free_iocb(iocb); + } else { + io_prep_pwrite(iocb, dstfd, buf, iosize, offset); + io_set_callback(iocb, wr_done); + if (1 != (res = io_submit(ctx, 1, &iocb))) + io_error("io_submit write", res); + } + if (debug) + write(2, "r", 1); + if (debug > 1) + printf("%d", iosize); +} + +void usage() +{ + fprintf(stderr, + "Usage: aiocp [-a align] [-s size] [-b blksize] [-n num_io]" + " [-f open_flag] SOURCE DEST\n" + "This copies from SOURCE to DEST using AIO.\n\n" + "Usage: aiocp [options] -w SOURCE\n" + "This does sequential AIO reads (no writes).\n\n" + "Usage: aiocp [options] -z DEST\n" + "This does sequential AIO writes of zeros.\n"); + + exit(1); +} + +/* + * Scale value by kilo, mega, or giga. + */ +long long scale_by_kmg(long long value, char scale) +{ + switch (scale) { + case 'g': + case 'G': + value *= 1024; + case 'm': + case 'M': + value *= 1024; + case 'k': + case 'K': + value *= 1024; + break; + case '\0': + break; + default: + usage(); + break; + } + return value; +} + +int main(int argc, char *const *argv) +{ + struct stat st; + off_t length = 0, offset = 0; + io_context_t myctx; + int c; + extern char *optarg; + extern int optind, opterr, optopt; + + while ((c = getopt(argc, argv, "a:b:df:n:s:wzD:")) != -1) { + char *endp; + + switch (c) { + case 'a': /* alignment of data buffer */ + alignment = strtol(optarg, &endp, 0); + alignment = (long)scale_by_kmg((long long)alignment, + *endp); + break; + case 'f': /* use these open flags */ + if (strcmp(optarg, "LARGEFILE") == 0 || + strcmp(optarg, "O_LARGEFILE") == 0) { + source_open_flag |= O_LARGEFILE; + dest_open_flag |= O_LARGEFILE; + } else if (strcmp(optarg, "TRUNC") == 0 || + strcmp(optarg, "O_TRUNC") == 0) { + dest_open_flag |= O_TRUNC; + } else if (strcmp(optarg, "SYNC") == 0 || + strcmp(optarg, "O_SYNC") == 0) { + dest_open_flag |= O_SYNC | O_NONBLOCK; + } else if (strcmp(optarg, "DIRECT") == 0 || + strcmp(optarg, "O_DIRECT") == 0) { + source_open_flag |= O_DIRECT; + dest_open_flag |= O_DIRECT; + } else if (strncmp(optarg, "CREAT", 5) == 0 || + strncmp(optarg, "O_CREAT", 5) == 0) { + dest_open_flag |= O_CREAT; + } + break; + case 'd': + debug++; + break; + case 'D': + delay.tv_usec = atoi(optarg); + break; + case 'b': /* block size */ + aio_blksize = strtol(optarg, &endp, 0); + aio_blksize = (long)scale_by_kmg((long long)aio_blksize, *endp); + break; + + case 'n': /* num io */ + aio_maxio = strtol(optarg, &endp, 0); + break; + case 's': /* size to transfer */ + length = strtoll(optarg, &endp, 0); + length = scale_by_kmg(length, *endp); + break; + case 'w': /* no write */ + no_write = 1; + break; + case 'z': /* write zero's */ + zero = 1; + break; + + default: + usage(); + } + } + + argc -= optind; + argv += optind; + +#ifndef DEBUG + if (argc < 1) { + usage(); + } +#else + source_open_flag |= O_DIRECT; + dest_open_flag |= O_DIRECT; + aio_blksize = 1; + aio_maxio=1; + srcname = "junkdata"; + dstname = "ff2"; +#endif + if (!zero) { +#ifndef DEBUG + if ((srcfd = open(srcname = *argv, source_open_flag)) < 0) { +#else + if ((srcfd = open(srcname, source_open_flag)) < 0) { +#endif + perror(srcname); + exit(1); + } + argv++; + argc--; + if (fstat(srcfd, &st) < 0) { + perror("fstat"); + exit(1); + } + if (length == 0) + length = st.st_size; + } + + if (!no_write) { + /* + * We are either copying or writing zeros to dstname + */ +#ifndef DEBUG + if (argc < 1) { + usage(); + } + if ((dstfd = open(dstname = *argv, dest_open_flag, 0666)) < 0) { +#else + if ((dstfd = open(dstname, dest_open_flag, 0666)) < 0) { +#endif + perror(dstname); + exit(1); + } + if (zero) { + /* + * get size of dest, if we are zeroing it. + * TODO: handle devices. + */ + if (fstat(dstfd, &st) < 0) { + perror("fstat"); + exit(1); + } + if (length == 0) + length = st.st_size; + } + } + + /* initialize state machine */ + memset(&myctx, 0, sizeof(myctx)); + io_queue_init(aio_maxio, &myctx); + tocopy = howmany(length, aio_blksize); +printf("tocopy=%d len=%d blk=%d\n", tocopy, length, aio_blksize); + if (init_iocb(aio_maxio, aio_blksize) < 0) { + fprintf(stderr, "Error allocating the i/o buffers\n"); + exit(1); + } + + while (tocopy > 0) { + int i, rc; + /* Submit as many reads as once as possible upto aio_maxio */ + int n = MIN(MIN(aio_maxio - busy, aio_maxio), + howmany(length - offset, aio_blksize)); + if (n > 0) { + struct iocb *ioq[n]; + + for (i = 0; i < n; i++) { + struct iocb *io = alloc_iocb(); + int iosize = MIN(length - offset, aio_blksize); + + if (zero) { + /* + * We are writing zero's to dstfd + */ + io_prep_pwrite(io, dstfd, io->u.c.buf, + iosize, offset); + io_set_callback(io, wr_done); + } else { + io_prep_pread(io, srcfd, io->u.c.buf, + iosize, offset); + io_set_callback(io, rd_done); + } + ioq[i] = io; + offset += iosize; + } + + rc = io_submit(myctx, n, ioq); + if (rc < 0) + io_error("io_submit", rc); + + busy += n; + if (debug > 1) + printf("io_submit(%d) busy:%d\n", n, busy); + if (delay.tv_usec) { + struct timeval t = delay; + (void)select(0,0,0,0,&t); + } + } + + /* + * We have submitted all the i/o requests. Wait for at least one to complete + * and call the callbacks. + */ + count_io_q_waits++; + rc = io_wait_run(myctx, 0); + if (rc < 0) + io_error("io_wait_run", rc); + + if (debug > 1) { + printf("io_wait_run: rc == %d\n", rc); + printf("busy:%d aio_maxio:%d tocopy:%d\n", + busy, aio_maxio, tocopy); + } + } + + if (srcfd != -1) + close(srcfd); + if (dstfd != -1) + close(dstfd); + exit(0); +} + +/* + * Results look like: + * [alanm@toolbox ~/MOT3]$ ../taio -d kernel-source-2.4.8-0.4g.ppc.rpm abc + * rrrrrrrrrrrrrrrwwwrwrrwwrrwrwwrrwrwrwwrrwrwrrrrwwrwwwrrwrrrwwwwwwwwwwwwwwwww + * rrrrrrrrrrrrrrwwwrrwrwrwrwrrwwwwwwwwwwwwwwrrrrrrrrrrrrrrrrrrwwwwrwrwwrwrwrwr + * wrrrrrrrwwwwwwwwwwwwwrrrwrrrwrrwrwwwwwwwwwwrrrrwwrwrrrrrrrrrrrwwwwwwwwwwwrww + * wwwrrrrrrrrwwrrrwwrwrwrwwwrrrrrrrwwwrrwwwrrwrwwwwwwwwrrrrrrrwwwrrrrrrrwwwwww + * wwwwwwwrwrrrrrrrrwrrwrrwrrwrwrrrwrrrwrrrwrwwwwwwwwwwwwwwwwwwrrrwwwrrrrrrrrrr + * rrwrrrrrrwrrwwwwwwwwwwwwwwwwrwwwrrwrwwrrrrrrrrrrrrrrrrrrrwwwwwwwwwwwwwwwwwww + * rrrrrwrrwrwrwrrwrrrwwwwwwwwrrrrwrrrwrwwrwrrrwrrwrrrrwwwwwwwrwrwwwwrwwrrrwrrr + * rrrwwwwwwwrrrrwwrrrrrrrrrrrrwrwrrrrwwwwwwwwwwwwwwrwrrrrwwwwrwrrrrwrwwwrrrwww + * rwwrrrrrrrwrrrrrrrrrrrrwwwwrrrwwwrwrrwwwwwwwwwwwwwwwwwwwwwrrrrrrrwwwwwwwrw + */ diff --git a/tests/generic/250 b/tests/generic/250 new file mode 100755 index 0000000..b0b175a --- /dev/null +++ b/tests/generic/250 @@ -0,0 +1,100 @@ +#! /bin/bash +# FS QA Test No. 250 +# +# Create an unwritten extent, set up dm-error, try a DIO write, then +# make sure we can't read back old disk contents. +# +#----------------------------------------------------------------------- +# Copyright (c) 2016, Oracle and/or its affiliates. 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 +#----------------------------------------------------------------------- + +seq=`basename $0` +seqres=$RESULT_DIR/$seq +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 + +_cleanup() +{ + cd / + rm -rf $tmp.* $testdir + _dmerror_cleanup +} + +# get standard environment, filters and checks +. ./common/rc +. ./common/filter +. ./common/dmerror + +# real QA test starts here +_supported_os Linux +_require_scratch +_require_dm_target error +_require_xfs_io_command "falloc" + +rm -f $seqres.full + + +echo "Format and mount" +$XFS_IO_PROG -d -c "pwrite -S 0x69 -b 1048576 0 $((64 * 1048576))" $SCRATCH_DEV >> $seqres.full +_scratch_mkfs_sized $((64 * 1048576)) > $seqres.full 2>&1 +_dmerror_init +_dmerror_mount >> $seqres.full 2>&1 +_dmerror_unmount +_dmerror_mount + +testdir=$SCRATCH_MNT/test-$seq +mkdir $testdir + +blksz=65536 +nr=640 +bufnr=128 +filesize=$((blksz * nr)) +bufsize=$((blksz * bufnr)) + +_require_fs_space $SCRATCH_MNT $((filesize / 1024 * 5 / 4)) + +echo "Create the original files" +$XFS_IO_PROG -f -c "falloc 0 $filesize" $testdir/file2 >> $seqres.full +_dmerror_unmount +_dmerror_mount + +echo "Compare files" +md5sum $testdir/file2 | _filter_scratch + +echo "CoW and unmount" +$XFS_IO_PROG -f -c "pwrite -S 0x63 $bufsize 1" $testdir/file2 >> $seqres.full +sync +_dmerror_load_error_table +$XFS_IO_PROG -d -f -c "pwrite -S 0x63 -b $bufsize 0 $filesize" $testdir/file2 >> $seqres.full +_dmerror_load_working_table +_dmerror_unmount +_dmerror_mount + +echo "Compare files" +md5sum $testdir/file2 | _filter_scratch + +echo "Check for damage" +_dmerror_unmount +_dmerror_cleanup +_repair_scratch_fs >> $seqres.full + +# success, all done +status=0 +exit diff --git a/tests/generic/250.out b/tests/generic/250.out new file mode 100644 index 0000000..710c80e --- /dev/null +++ b/tests/generic/250.out @@ -0,0 +1,10 @@ +QA output created by 250 +Format and mount +Create the original files +Compare files +ec8bb3b24d5b0f1b5bdf8c8f0f541ee6 SCRATCH_MNT/test-250/file2 +CoW and unmount +pwrite64: Input/output error +Compare files +3ed86318f4ff8da26c1c2a6e3041f9be SCRATCH_MNT/test-250/file2 +Check for damage diff --git a/tests/generic/252 b/tests/generic/252 new file mode 100755 index 0000000..fc9a723 --- /dev/null +++ b/tests/generic/252 @@ -0,0 +1,103 @@ +#! /bin/bash +# FS QA Test No. 252 +# +# Create an unwritten extent, set up dm-error, try an AIO DIO write, then +# make sure we can't read back old disk contents. +# +#----------------------------------------------------------------------- +# Copyright (c) 2016, Oracle and/or its affiliates. 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 +#----------------------------------------------------------------------- + +seq=`basename $0` +seqres=$RESULT_DIR/$seq +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 + +_cleanup() +{ + cd / + rm -rf $tmp.* $testdir $TEST_DIR/moo + _dmerror_cleanup +} + +# get standard environment, filters and checks +. ./common/rc +. ./common/filter +. ./common/dmerror + +# real QA test starts here +_supported_os Linux +_require_scratch +_require_dm_target error +_require_xfs_io_command "falloc" +_require_test_program "aio-dio-regress/aiocp" +AIO_TEST="src/aio-dio-regress/aiocp" + +rm -f $seqres.full + + +echo "Format and mount" +$XFS_IO_PROG -d -c "pwrite -S 0x69 -b 1048576 0 $((64 * 1048576))" $SCRATCH_DEV >> $seqres.full +_scratch_mkfs_sized $((64 * 1048576)) > $seqres.full 2>&1 +_dmerror_init +_dmerror_mount >> $seqres.full 2>&1 +_dmerror_unmount +_dmerror_mount + +testdir=$SCRATCH_MNT/test-$seq +mkdir $testdir + +blksz=65536 +nr=640 +bufnr=128 +filesize=$((blksz * nr)) +bufsize=$((blksz * bufnr)) + +_require_fs_space $SCRATCH_MNT $((filesize / 1024 * 5 / 4)) + +echo "Create the original files" +$XFS_IO_PROG -f -c "falloc 0 $filesize" $testdir/file2 >> $seqres.full +_dmerror_unmount +_dmerror_mount + +echo "Compare files" +md5sum $testdir/file2 | _filter_scratch + +echo "CoW and unmount" +$XFS_IO_PROG -f -c "pwrite -S 0x63 $bufsize 1" $testdir/file2 >> $seqres.full +$XFS_IO_PROG -f -c "pwrite -S 0x63 -b $bufsize 0 $filesize" $TEST_DIR/moo >> $seqres.full +sync +_dmerror_load_error_table +$AIO_TEST -f DIRECT -b $bufsize $TEST_DIR/moo $testdir/file2 >> $seqres.full +_dmerror_load_working_table +_dmerror_unmount +_dmerror_mount + +echo "Compare files" +md5sum $testdir/file2 | _filter_scratch + +echo "Check for damage" +_dmerror_unmount +_dmerror_cleanup +_repair_scratch_fs >> $seqres.full + +# success, all done +status=0 +exit diff --git a/tests/generic/252.out b/tests/generic/252.out new file mode 100644 index 0000000..3bc78c2 --- /dev/null +++ b/tests/generic/252.out @@ -0,0 +1,10 @@ +QA output created by 252 +Format and mount +Create the original files +Compare files +ec8bb3b24d5b0f1b5bdf8c8f0f541ee6 SCRATCH_MNT/test-252/file2 +CoW and unmount +write missed bytes expect 8388608 got 0 +Compare files +3ed86318f4ff8da26c1c2a6e3041f9be SCRATCH_MNT/test-252/file2 +Check for damage diff --git a/tests/generic/group b/tests/generic/group index 860ff4a..0e4e7d3 100644 --- a/tests/generic/group +++ b/tests/generic/group @@ -252,7 +252,9 @@ 247 auto quick rw 248 auto quick rw 249 auto quick rw +250 auto quick prealloc rw 251 ioctl trim +252 auto quick prealloc rw 255 auto quick prealloc 256 auto quick 257 dir auto quick _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs
next prev parent reply other threads:[~2016-02-11 23:40 UTC|newest] Thread overview: 81+ messages / expand[flat|nested] mbox.gz Atom feed top 2016-02-11 23:39 [PATCH v4.2 00/32] xfstests: test the nfs/cifs/btrfs/xfs reflink/dedupe ioctls Darrick J. Wong 2016-02-11 23:39 ` Darrick J. Wong 2016-02-11 23:39 ` [PATCH 01/32] generic/182: this is a dedupe test, check for dedupe Darrick J. Wong 2016-02-11 23:39 ` Darrick J. Wong 2016-02-11 23:39 ` [PATCH 02/32] xfstests: filter whitespace in 128 and 132 Darrick J. Wong 2016-02-11 23:39 ` Darrick J. Wong 2016-02-11 23:39 ` [PATCH 03/32] xfstests: make _scratch_mkfs_blocksized usable Darrick J. Wong 2016-02-11 23:39 ` Darrick J. Wong 2016-02-11 23:39 ` [PATCH 04/32] reflink: remove redundant filesystem checks from the end of the tests Darrick J. Wong 2016-02-11 23:39 ` Darrick J. Wong 2016-02-11 23:39 ` [PATCH 05/32] reflink: remove unnecessary umounts from test code Darrick J. Wong 2016-02-11 23:39 ` Darrick J. Wong 2016-02-11 23:39 ` [PATCH 06/32] common/dmerror: add some more dmerror routines Darrick J. Wong 2016-02-11 23:39 ` Darrick J. Wong 2016-02-11 23:40 ` [PATCH 07/32] common: create _require_test_program to look for programs in src/ Darrick J. Wong 2016-02-11 23:40 ` Darrick J. Wong 2016-02-11 23:40 ` [PATCH 08/32] common: provide a method to repair the scratch fs Darrick J. Wong 2016-02-11 23:40 ` Darrick J. Wong 2016-02-11 23:40 ` [PATCH 09/32] reflink: refactor mixed block creation code Darrick J. Wong 2016-02-11 23:40 ` Darrick J. Wong 2016-02-11 23:40 ` [PATCH 10/32] reflink: fix style problems in existing tests Darrick J. Wong 2016-02-11 23:40 ` [PATCH 11/32] reflink: add _require_odirect to the directio tests Darrick J. Wong 2016-02-11 23:40 ` Darrick J. Wong 2016-02-11 23:40 ` Darrick J. Wong [this message] 2016-02-11 23:40 ` [PATCH 12/32] dio: unwritten conversion bug tests Darrick J. Wong 2016-02-12 3:52 ` Dave Chinner 2016-02-12 3:52 ` Dave Chinner 2016-02-12 15:22 ` Theodore Ts'o 2016-02-12 15:22 ` Theodore Ts'o 2016-02-12 22:49 ` Dave Chinner 2016-02-12 22:49 ` Dave Chinner 2016-02-12 17:40 ` Darrick J. Wong 2016-02-12 17:40 ` Darrick J. Wong 2016-02-11 23:40 ` [PATCH 13/32] reflink: test intersecting CoW and falloc/fpunch/fzero/fcollapse/finsert/ftrunc Darrick J. Wong 2016-02-11 23:40 ` Darrick J. Wong 2016-02-11 23:40 ` [PATCH 14/32] reflink: test CoW behavior with IO errors Darrick J. Wong 2016-02-11 23:40 ` Darrick J. Wong 2016-02-11 23:40 ` [PATCH 15/32] reflink: test CoW operations against the source file Darrick J. Wong 2016-02-11 23:40 ` Darrick J. Wong 2016-02-11 23:41 ` [PATCH 16/32] reflink: ensure that we can handle reflinking a lot of extents Darrick J. Wong 2016-02-11 23:41 ` Darrick J. Wong 2016-02-11 23:41 ` [PATCH 17/32] xfs: more reflink tests Darrick J. Wong 2016-02-11 23:41 ` Darrick J. Wong 2016-02-11 23:41 ` [PATCH 18/32] xfs/122: support refcount/rmap data structures Darrick J. Wong 2016-02-11 23:41 ` Darrick J. Wong 2016-02-11 23:41 ` [PATCH 19/32] xfs: test fragmentation characteristics of copy-on-write Darrick J. Wong 2016-02-11 23:41 ` Darrick J. Wong 2016-02-11 23:41 ` [PATCH 20/32] reflink: high offset reflink and dedupe tests Darrick J. Wong 2016-02-11 23:41 ` Darrick J. Wong 2016-02-11 23:41 ` [PATCH 21/32] reflink: test xfs cow behavior when the filesystem crashes Darrick J. Wong 2016-02-11 23:41 ` Darrick J. Wong 2016-02-11 23:41 ` [PATCH 22/32] reflink: test quota accounting Darrick J. Wong 2016-02-11 23:41 ` Darrick J. Wong 2016-02-11 23:41 ` [PATCH 23/32] reflink: test CoW across a mixed range of block types with cowextsize set Darrick J. Wong 2016-02-11 23:41 ` Darrick J. Wong 2016-02-11 23:42 ` [PATCH 24/32] xfs: test the automatic cowextsize extent garbage collector Darrick J. Wong 2016-02-11 23:42 ` Darrick J. Wong 2016-02-11 23:42 ` [PATCH 25/32] xfs: test rmapbt functionality Darrick J. Wong 2016-02-11 23:42 ` Darrick J. Wong 2016-02-11 23:42 ` [PATCH 26/32] reflink: test aio copy on write Darrick J. Wong 2016-02-11 23:42 ` Darrick J. Wong 2016-02-11 23:42 ` [PATCH 27/32] xfs: aio cow tests Darrick J. Wong 2016-02-11 23:42 ` Darrick J. Wong 2016-02-11 23:42 ` [PATCH 28/32] xfs: test xfs_getbmapx behavior with shared extents Darrick J. Wong 2016-02-11 23:42 ` Darrick J. Wong 2016-02-11 23:42 ` [PATCH 29/32] reflink: test reflink+cow+enospc all at the same time Darrick J. Wong 2016-02-11 23:42 ` Darrick J. Wong 2016-02-11 23:42 ` [PATCH 30/32] xfs: ensure that we don't crash when freeing the ag reservations on a ro mount Darrick J. Wong 2016-02-11 23:42 ` Darrick J. Wong 2016-02-11 23:42 ` [PATCH 31/32] xfs: test source CoW across mixed block types with cowextsz set Darrick J. Wong 2016-02-11 23:42 ` Darrick J. Wong 2016-02-11 23:42 ` [PATCH 32/32] reflink: kick the slow tests out of the 'quick' group Darrick J. Wong 2016-02-11 23:42 ` Darrick J. Wong 2016-02-12 17:41 ` [PATCH 33/32] aiocp: fix the gcc warnings Darrick J. Wong 2016-02-12 17:41 ` Darrick J. Wong 2016-02-13 1:26 ` [PATCH v4.2 00/32] xfstests: test the nfs/cifs/btrfs/xfs reflink/dedupe ioctls Dave Chinner 2016-02-13 1:26 ` Dave Chinner 2016-02-13 4:38 ` Darrick J. Wong 2016-02-13 4:38 ` Darrick J. Wong 2016-02-13 8:03 ` Christoph Hellwig 2016-02-13 8:03 ` Christoph Hellwig
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20160211234037.2202.93885.stgit@birch.djwong.org \ --to=darrick.wong@oracle.com \ --cc=david@fromorbit.com \ --cc=fstests@vger.kernel.org \ --cc=linux-btrfs@vger.kernel.org \ --cc=xfs@oss.sgi.com \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
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.