[LTP] [PATCH v2] Create dio_read.c test
Cyril Hrubis
chrubis@suse.cz
Thu Jan 13 14:22:44 CET 2022
Hi!
Pushed with some changes, thanks.
> diff --git a/runtest/ltp-aiodio.part4 b/runtest/ltp-aiodio.part4
> index ef1cfdac6..88a1b492b 100644
> --- a/runtest/ltp-aiodio.part4
> +++ b/runtest/ltp-aiodio.part4
> @@ -61,8 +61,8 @@ DIT001 dio_truncate
> DIT002 dio_truncate
> #Running read_checkzero
> #gread_checkzero
> -#Running ltp-diorh
> -DOR000 ltp-diorh $TMPDIR/aiodio.$$/file2
> -DOR001 ltp-diorh $TMPDIR/aiodio.$$/file3
> -DOR002 ltp-diorh $TMPDIR/aiodio.$$/file4
> -DOR003 ltp-diorh $TMPDIR/aiodio.$$/file5
> +#Running dio_read
> +DOR000 dio_read
> +DOR001 dio_read
> +DOR002 dio_read
> +DOR003 dio_read
I've adjusted these to run with different number of writers and numbers
of iterations so that the runtime more or less matches the original test
runtime.
> diff --git a/testcases/kernel/io/ltp-aiodio/.gitignore b/testcases/kernel/io/ltp-aiodio/.gitignore
> index 8da8e946b..f5f20d57e 100644
> --- a/testcases/kernel/io/ltp-aiodio/.gitignore
> +++ b/testcases/kernel/io/ltp-aiodio/.gitignore
> @@ -5,6 +5,6 @@
> /dio_append
> /dio_sparse
> /dio_truncate
> +/dio_read
> /dirty
> -/ltp-diorh
> /read_checkzero
> diff --git a/testcases/kernel/io/ltp-aiodio/dio_read.c b/testcases/kernel/io/ltp-aiodio/dio_read.c
> new file mode 100644
> index 000000000..900f1e70d
> --- /dev/null
> +++ b/testcases/kernel/io/ltp-aiodio/dio_read.c
> @@ -0,0 +1,176 @@
> +// SPDX-License-Identifier: GPL-2.0-or-later
> +/*
> + * Copyright (C) 2021 SUSE LLC Andrea Cervesato <andrea.cervesato@suse.com>
> + */
> +
> +/*\
> + * [Description]
> + *
> + * Create a file using buffered writes while other processes are doing
> + * O_DIRECT reads and check if the buffer reads always see zero.
> + */
> +
> +#define _GNU_SOURCE
> +
> +#include <unistd.h>
> +#include <limits.h>
> +#include <string.h>
> +#include <fcntl.h>
> +#include <sys/mman.h>
> +#include "tst_test.h"
> +#include "common.h"
> +
> +static char *str_numchildren;
> +static char *str_writesize;
> +static char *str_readsize;
> +static char *str_filesize;
> +
> +static char *filename = "file.bin";
> +static int numchildren = 100;
> +static long long writesize = 32 * 1024 * 1024;
> +static long long readsize = 32 * 1024 * 1024;
> +static long long filesize = 128 * 1024 * 1024;
> +static int *children_completed;
> +static char *iobuf;
> +static int filedesc;
Renamed this just to fd as that is more idiomatic in C code.
> +static void do_buffered_writes(int fd, char *bufptr, long long fsize, long long wsize, int pattern)
> +{
> + long long offset;
> + long long w;
> +
> + memset(bufptr, pattern, wsize);
> +
> + tst_res(TINFO, "child %i writing file", getpid());
> +
> + for (offset = 0; offset + wsize <= fsize; offset += wsize) {
> + w = pwrite(fd, bufptr, wsize, offset);
> + if (w < 0)
> + tst_brk(TBROK, "pwrite: %s", tst_strerrno(-w));
> + if (w != wsize)
> + tst_brk(TBROK, "pwrite: wrote %lld bytes out of %lld", w, wsize);
> +
> + SAFE_FSYNC(fd);
> + }
> +}
> +
> +static int do_direct_reads(char *filename, char *bufptr, long long fsize, long long rsize)
> +{
> + int fd;
> + long long offset;
> + long long w;
> + int fail = 0;
> +
> + while ((fd = open(filename, O_RDONLY | O_DIRECT, 0666)) < 0)
> + usleep(100);
As the file is created in the test setup now there is no need to loop
here, so I've changed it to SAFE_OPEN()
> + while (1) {
> + for (offset = 0; offset + rsize < fsize; offset += rsize) {
> + char *bufoff;
> +
> + if (*children_completed >= numchildren)
I've added a counter and message here that prints how many iterations
the readed did and it's about 10 for me in most cases, which only shows
that the synchronization is really important since we would have been
wasting most of the time the other way.
> + goto exit;
> +
> + w = pread(fd, bufptr, rsize, offset);
> + if (w < 0)
> + tst_brk(TBROK, "pread: %s", tst_strerrno(-w));
> + if (w != rsize)
> + tst_brk(TBROK, "pread: read %lld bytes out of %lld", w, rsize);
> +
> + bufoff = check_zero(bufptr, rsize);
> + if (bufoff) {
> + fail = 1;
> + goto exit;
> + }
> + }
> + }
> +
> +exit:
> + SAFE_CLOSE(fd);
> +
> + return fail;
> +}
> +
> +static void setup(void)
> +{
> + struct stat sb;
> + long long buffsize;
> + long long alignment;
> +
> + if (tst_parse_int(str_numchildren, &numchildren, 1, INT_MAX))
> + tst_brk(TBROK, "Invalid number of children '%s'", str_numchildren);
> +
> + if (tst_parse_filesize(str_filesize, &filesize, 1, LLONG_MAX))
> + tst_brk(TBROK, "Invalid file size '%s'", str_filesize);
> +
> + if (tst_parse_filesize(str_writesize, &writesize, 1, filesize))
> + tst_brk(TBROK, "Invalid write blocks size '%s'", str_writesize);
> +
> + if (tst_parse_filesize(str_readsize, &readsize, 1, filesize))
> + tst_brk(TBROK, "Invalid read blocks size '%s'", str_readsize);
> +
> + SAFE_STAT(".", &sb);
> + alignment = sb.st_blksize;
> +
> + buffsize = readsize;
> + if (writesize > readsize)
> + buffsize = writesize;
> +
> + iobuf = SAFE_MEMALIGN(alignment, buffsize);
> +
> + children_completed = SAFE_MMAP(NULL, sizeof(int), PROT_READ | PROT_WRITE, MAP_SHARED | MAP_ANONYMOUS, -1, 0);
> +
> + filedesc = SAFE_OPEN(filename, O_CREAT | O_TRUNC | O_RDWR, 0666);
> +}
> +
> +static void cleanup(void)
> +{
> + SAFE_CLOSE(filedesc);
> +}
> +
> +static void run(void)
> +{
> + int i;
> + int fail;
> +
> + // Fill the file with a known pattern so that the blocks
> + // on disk can be detected if they become exposed
> + do_buffered_writes(filedesc, iobuf, filesize, writesize, 1);
> + SAFE_FSYNC(filedesc);
> + SAFE_FTRUNCATE(filedesc, 0);
> + SAFE_FSYNC(filedesc);
> +
> + SAFE_FTRUNCATE(filedesc, filesize);
> +
> + *children_completed = 0;
> +
> + for (i = 0; i < numchildren; i++) {
> + if (!SAFE_FORK()) {
> + do_buffered_writes(filedesc, iobuf, filesize, writesize, 0);
> + (*children_completed)++;
This actually caused a rare timeouts. The problem is that we do run this
increment concurenlty from numchildren processes and if it happens that
two of them finish at the exactly same time they would both fetch value
N and store N+1, which means that the sum would be one less than
expected and the reader will spin forever.
WHat we do have to do here is to use memory barriers so that the fetch
and store are atomic. Happily we do have atomic operations in the test
library exactly for this purpose so I've changed this line to:
tst_atomic_add_return(1, children_completed);
With that the lockups and timeouts are no longer happening.
> + return;
> + }
> + }
> +
> + fail = do_direct_reads(filename, iobuf, filesize, readsize);
> +
> + if (fail)
> + tst_res(TFAIL, "Non zero bytes read");
> + else
> + tst_res(TPASS, "All bytes read were zeroed");
> +}
> +
> +static struct tst_test test = {
> + .test_all = run,
> + .setup = setup,
> + .cleanup = cleanup,
> + .needs_tmpdir = 1,
> + .forks_child = 1,
> + .options = (struct tst_option[]) {
> + {"n:", &str_numchildren, "Number of threads (default 100)"},
> + {"w:", &str_writesize, "Size of writing blocks (default 32M)"},
> + {"r:", &str_readsize, "Size of reading blocks (default 32M)"},
> + {"s:", &str_filesize, "File size (default 128M)"},
> + {}
> + },
> +};
> diff --git a/testcases/kernel/io/ltp-aiodio/ltp-diorh.c b/testcases/kernel/io/ltp-aiodio/ltp-diorh.c
> deleted file mode 100644
> index 3bdf62388..000000000
> --- a/testcases/kernel/io/ltp-aiodio/ltp-diorh.c
> +++ /dev/null
> @@ -1,176 +0,0 @@
> -/*
> - * Copyright (C) 2003,2004 Red Hat, Inc. All rights reserved.
> - *
> - * The contents of this file may be used under the terms of the GNU
> - * General Public License version 2 (the "GPL")
> - *
> - * Author: Stephen C. Tweedie <sct@redhat.com>
> - * 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., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
> - *
> - * Module: .c
> - */
> -
> -/*
> - * Change History:
> - *
> - * 2/2004 Marty Ridgeway (mridge@us.ibm.com) Changes to adapt to LTP
> - *
> - */
> -
> -#define _XOPEN_SOURCE 600
> -#define _GNU_SOURCE
> -#define MAX_ITERATIONS 250
> -
> -#include <unistd.h>
> -#include <stdlib.h>
> -#include <stdio.h>
> -#include <string.h>
> -#include <errno.h>
> -#include <fcntl.h>
> -#include <sys/mman.h>
> -#include <sys/wait.h>
> -
> -#define BIGSIZE 128*1024*1024
> -#define READSIZE 32*1024*1024
> -#define WRITESIZE 32*1024*1024
> -
> -int pagesize;
> -char *iobuf;
> -int pass = 0;
> -
> -void assert(const char *what, int assertion)
> -{
> - if (assertion)
> - return;
> - perror(what);
> - exit(1);
> -}
> -
> -void do_buffered_writes(int fd, int pattern)
> -{
> - int rc;
> - int offset;
> -
> - memset(iobuf, pattern, WRITESIZE);
> - for (offset = 0; offset + WRITESIZE <= BIGSIZE; offset += WRITESIZE) {
> - rc = pwrite(fd, iobuf, WRITESIZE, offset);
> - assert("pwrite", rc >= 0);
> - if (rc != WRITESIZE) {
> - fprintf(stderr, "Pass %d: short write (%d out of %d)\n",
> - pass, rc, WRITESIZE);
> - exit(1);
> - }
> - fsync(fd);
> - }
> -}
> -
> -int do_direct_reads(char *filename)
> -{
> - int fd;
> - int offset;
> - int rc, i;
> - int *p;
> -
> - fd = open(filename, O_DIRECT | O_RDONLY, 0);
> - assert("open", fd >= 0);
> -
> - for (offset = 0; offset + READSIZE <= BIGSIZE; offset += READSIZE) {
> - rc = pread(fd, iobuf, READSIZE, offset);
> - assert("pread", rc >= 0);
> - if (rc != READSIZE) {
> - fprintf(stderr, "Pass: %d short read (%d out of %d)\n",
> - pass, rc, READSIZE);
> - exit(1);
> - }
> - for (i = 0, p = (int *)iobuf; i < READSIZE; i += 4) {
> - if (*p) {
> - fprintf(stderr,
> - "Pass: %d Found data (%08x) at offset %d+%d\n",
> - pass, *p, offset, i);
> - close(fd);
> - return 1;
> - }
> - p++;
> - }
> - }
> - close(fd);
> - return 0;
> -}
> -
> -int main(int argc, char *argv[])
> -{
> - char *filename;
> - int fd;
> - int pid;
> - int err;
> - int bufsize;
> -
> - if (argc != 2) {
> - fprintf(stderr, "Needs a filename as an argument.\n");
> - exit(1);
> - }
> -
> - filename = argv[1];
> -
> - pagesize = getpagesize();
> - bufsize = READSIZE;
> - if (WRITESIZE > READSIZE)
> - bufsize = WRITESIZE;
> - err = posix_memalign((void **)&iobuf, pagesize, bufsize);
> - if (err) {
> - fprintf(stderr, "Error allocating %d aligned bytes.\n",
> - bufsize);
> - exit(1);
> - }
> -
> - fd = open(filename, O_CREAT | O_TRUNC | O_RDWR, 0666);
> - assert("open", fd >= 0);
> -
> - do {
> -
> - assert("ftruncate", ftruncate(fd, BIGSIZE) == 0);
> - fsync(fd);
> -
> - pid = fork();
> - assert("fork", pid >= 0);
> -
> - if (!pid) {
> - do_buffered_writes(fd, 0);
> - exit(0);
> - }
> -
> - err = do_direct_reads(filename);
> -
> - wait4(pid, NULL, WNOHANG, 0);
> -
> - if (err)
> - break;
> -
> - /* Fill the file with a known pattern so that the blocks
> - * on disk can be detected if they become exposed. */
> - do_buffered_writes(fd, 1);
> - fsync(fd);
> -
> - assert("ftruncate", ftruncate(fd, 0) == 0);
> - fsync(fd);
> - } while (pass++ < MAX_ITERATIONS);
> -
> - if (!err) {
> - fprintf(stdout, "ltp-diorh: Completed %d iterations OK \n",
> - pass);
> - }
> -
> - return err;
> -}
> --
> 2.34.1
>
>
> --
> Mailing list info: https://lists.linux.it/listinfo/ltp
--
Cyril Hrubis
chrubis@suse.cz
More information about the ltp
mailing list