freebsd-skq/tests/sys/aio/lio_kqueue_test.c
jhb be47bc68fb Refactor the AIO subsystem to permit file-type-specific handling and
improve cancellation robustness.

Introduce a new file operation, fo_aio_queue, which is responsible for
queueing and completing an asynchronous I/O request for a given file.
The AIO subystem now exports library of routines to manipulate AIO
requests as well as the ability to run a handler function in the
"default" pool of AIO daemons to service a request.

A default implementation for file types which do not include an
fo_aio_queue method queues requests to the "default" pool invoking the
fo_read or fo_write methods as before.

The AIO subsystem permits file types to install a private "cancel"
routine when a request is queued to permit safe dequeueing and cleanup
of cancelled requests.

Sockets now use their own pool of AIO daemons and service per-socket
requests in FIFO order.  Socket requests will not block indefinitely
permitting timely cancellation of all requests.

Due to the now-tight coupling of the AIO subsystem with file types,
the AIO subsystem is now a standard part of all kernels.  The VFS_AIO
kernel option and aio.ko module are gone.

Many file types may block indefinitely in their fo_read or fo_write
callbacks resulting in a hung AIO daemon.  This can result in hung
user processes (when processes attempt to cancel all outstanding
requests during exit) or a hung system.  To protect against this, AIO
requests are only permitted for known "safe" files by default.  AIO
requests for all file types can be enabled by setting the new
vfs.aio.enable_usafe sysctl to a non-zero value.  The AIO tests have
been updated to skip operations on unsafe file types if the sysctl is
zero.

Currently, AIO requests on sockets and raw disks are considered safe
and are enabled by default.  aio_mlock() is also enabled by default.

Reviewed by:	cem, jilles
Discussed with:	kib (earlier version)
Sponsored by:	Chelsio Communications
Differential Revision:	https://reviews.freebsd.org/D5289
2016-03-01 18:12:14 +00:00

239 lines
6.2 KiB
C

/*-
* Copyright (C) 2005 IronPort Systems, Inc. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* $FreeBSD$
*/
/*
* Note: it is a good idea to run this against a physical drive to
* exercise the physio fast path (ie. lio_kqueue /dev/<something safe>)
* This will ensure op's counting is correct. It is currently broken.
*
* Also note that LIO & kqueue is not implemented in FreeBSD yet, LIO
* is also broken with respect to op's and some paths.
*
* A patch to make this work is at:
* http://www.ambrisko.com/doug/listio_kqueue/listio_kqueue.patch
*/
#include <sys/types.h>
#include <sys/event.h>
#include <sys/time.h>
#include <aio.h>
#include <fcntl.h>
#include <err.h>
#include <errno.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include "freebsd_test_suite/macros.h"
#include "local.h"
#define PATH_TEMPLATE "aio.XXXXXXXXXX"
#define LIO_MAX 5
#define MAX_IOCBS LIO_MAX * 16
#define MAX_RUNS 300
int
main(int argc, char *argv[])
{
int fd;
struct aiocb *iocb[MAX_IOCBS];
struct aiocb **lio[LIO_MAX], **kq_lio;
int i, result, run, error, j, k;
char buffer[32768];
int kq;
struct kevent ke, kq_returned;
struct timespec ts;
struct sigevent sig;
time_t time1, time2;
char *file, pathname[sizeof(PATH_TEMPLATE)];
int tmp_file = 0, failed = 0;
PLAIN_REQUIRE_KERNEL_MODULE("aio", 0);
PLAIN_REQUIRE_UNSAFE_AIO(0);
kq = kqueue();
if (kq < 0)
err(1, "kqeueue(2) failed");
if (argc == 1) {
strcpy(pathname, PATH_TEMPLATE);
fd = mkstemp(pathname);
file = pathname;
tmp_file = 1;
} else {
file = argv[1];
fd = open(file, O_RDWR|O_CREAT, 0666);
}
if (fd < 0)
err(1, "can't open %s", argv[1]);
#ifdef DEBUG
printf("Hello kq %d fd %d\n", kq, fd);
#endif
for (run = 0; run < MAX_RUNS; run++) {
#ifdef DEBUG
printf("Run %d\n", run);
#endif
for (j = 0; j < LIO_MAX; j++) {
lio[j] =
malloc(sizeof(struct aiocb *) * MAX_IOCBS/LIO_MAX);
for (i = 0; i < MAX_IOCBS / LIO_MAX; i++) {
k = (MAX_IOCBS / LIO_MAX * j) + i;
lio[j][i] = iocb[k] =
calloc(1, sizeof(struct aiocb));
iocb[k]->aio_nbytes = sizeof(buffer);
iocb[k]->aio_buf = buffer;
iocb[k]->aio_fildes = fd;
iocb[k]->aio_offset
= iocb[k]->aio_nbytes * k * (run + 1);
#ifdef DEBUG
printf("hello iocb[k] %d\n",
iocb[k]->aio_offset);
#endif
iocb[k]->aio_lio_opcode = LIO_WRITE;
}
sig.sigev_notify_kqueue = kq;
sig.sigev_value.sival_ptr = lio[j];
sig.sigev_notify = SIGEV_KEVENT;
time(&time1);
result = lio_listio(LIO_NOWAIT, lio[j],
MAX_IOCBS / LIO_MAX, &sig);
error = errno;
time(&time2);
#ifdef DEBUG
printf("Time %d %d %d result -> %d\n",
time1, time2, time2-time1, result);
#endif
if (result != 0) {
errno = error;
err(1, "FAIL: Result %d iteration %d\n",
result, j);
}
#ifdef DEBUG
printf("write %d is at %p\n", j, lio[j]);
#endif
}
for (i = 0; i < LIO_MAX; i++) {
for (j = LIO_MAX - 1; j >=0; j--) {
if (lio[j])
break;
}
for (;;) {
bzero(&ke, sizeof(ke));
bzero(&kq_returned, sizeof(ke));
ts.tv_sec = 0;
ts.tv_nsec = 1;
#ifdef DEBUG
printf("FOO lio %d -> %p\n", j, lio[j]);
#endif
EV_SET(&ke, (uintptr_t)lio[j],
EVFILT_LIO, EV_ONESHOT, 0, 0, iocb[j]);
result = kevent(kq, NULL, 0,
&kq_returned, 1, &ts);
error = errno;
if (result < 0) {
perror("kevent error: ");
}
kq_lio = kq_returned.udata;
#ifdef DEBUG
printf("kevent %d %d errno %d return.ident %p "
"return.data %p return.udata %p %p\n",
i, result, error,
kq_returned.ident, kq_returned.data,
kq_returned.udata,
lio[j]);
#endif
if (kq_lio)
break;
#ifdef DEBUG
printf("Try again\n");
#endif
}
#ifdef DEBUG
printf("lio %p\n", lio);
#endif
for (j = 0; j < LIO_MAX; j++) {
if (lio[j] == kq_lio)
break;
}
if (j == LIO_MAX)
errx(1, "FAIL: ");
#ifdef DEBUG
printf("Error Result for %d is %d\n", j, result);
#endif
if (result < 0) {
printf("FAIL: run %d, operation %d result %d \n", run, LIO_MAX - i -1, result);
failed++;
} else
printf("PASS: run %d, operation %d result %d \n", run, LIO_MAX - i -1, result);
for (k = 0; k < MAX_IOCBS / LIO_MAX; k++) {
result = aio_return(kq_lio[k]);
#ifdef DEBUG
printf("Return Resulto for %d %d is %d\n", j, k, result);
#endif
if (result != sizeof(buffer)) {
printf("FAIL: run %d, operation %d sub-opt %d result %d (errno=%d) should be %zu\n",
run, LIO_MAX - i -1, k, result, errno, sizeof(buffer));
} else {
printf("PASS: run %d, operation %d sub-opt %d result %d\n",
run, LIO_MAX - i -1, k, result);
}
}
#ifdef DEBUG
printf("\n");
#endif
for (k = 0; k < MAX_IOCBS / LIO_MAX; k++)
free(lio[j][k]);
free(lio[j]);
lio[j] = NULL;
}
}
#ifdef DEBUG
printf("Done\n");
#endif
if (tmp_file)
unlink(pathname);
if (failed)
errx(1, "FAIL: %d testcases failed", failed);
else
errx(0, "PASS: All\n");
}