X-Git-Url: https://git.whamcloud.com/?a=blobdiff_plain;f=libcfs%2Flibcfs%2Ffail.c;h=0478f4a8128146b3c8e36d667fb7618af0df4f05;hb=76bea6ca661609e3788d1cfe7e08e4a63af0a349;hp=d653ee658d8736560ce048b7519835520072f1cb;hpb=a7f5b2e19d5645c3a54e957bdc20fa10cacf0724;p=fs%2Flustre-release.git diff --git a/libcfs/libcfs/fail.c b/libcfs/libcfs/fail.c index d653ee6..0478f4a 100644 --- a/libcfs/libcfs/fail.c +++ b/libcfs/libcfs/fail.c @@ -1,6 +1,4 @@ -/* -*- mode: c; c-basic-offset: 8; indent-tabs-mode: nil; -*- - * vim:expandtab:shiftwidth=8:tabstop=8: - * +/* * GPL HEADER START * * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. @@ -18,117 +16,130 @@ * You should have received a copy of the GNU General Public License * version 2 along with this program; If not, see http://www.gnu.org/licenses * - * Please contact Oracle Corporation, Inc., 500 Oracle Parkway, Redwood Shores, - * CA 94065 USA or visit www.oracle.com if you need additional information or - * have any questions. - * * GPL HEADER END */ /* * Copyright (c) 2007, 2010, Oracle and/or its affiliates. All rights reserved. * Use is subject to license terms. + * + * Copyright (c) 2012, 2015, Intel Corporation. */ /* * This file is part of Lustre, http://www.lustre.org/ * Lustre is a trademark of Oracle Corporation, Inc. */ -#ifndef __KERNEL__ -#include -#else +#include #include -#endif unsigned long cfs_fail_loc = 0; unsigned int cfs_fail_val = 0; -cfs_waitq_t cfs_race_waitq; +int cfs_fail_err; +DECLARE_WAIT_QUEUE_HEAD(cfs_race_waitq); int cfs_race_state; -CFS_EXPORT_SYMBOL(cfs_fail_loc); -CFS_EXPORT_SYMBOL(cfs_fail_val); -CFS_EXPORT_SYMBOL(cfs_race_waitq); -CFS_EXPORT_SYMBOL(cfs_race_state); +EXPORT_SYMBOL(cfs_fail_loc); +EXPORT_SYMBOL(cfs_fail_val); +EXPORT_SYMBOL(cfs_fail_err); +EXPORT_SYMBOL(cfs_race_waitq); +EXPORT_SYMBOL(cfs_race_state); int __cfs_fail_check_set(__u32 id, __u32 value, int set) { - static cfs_atomic_t cfs_fail_count = CFS_ATOMIC_INIT(0); - - LASSERT(!(id & CFS_FAIL_ONCE)); - - if ((cfs_fail_loc & (CFS_FAILED | CFS_FAIL_ONCE)) == - (CFS_FAILED | CFS_FAIL_ONCE)) { - cfs_atomic_set(&cfs_fail_count, 0); /* paranoia */ - return 0; - } - - /* Fail 1/cfs_fail_val times */ - if (cfs_fail_loc & CFS_FAIL_RAND) { - if (cfs_fail_val < 2 || cfs_rand() % cfs_fail_val > 0) - return 0; - } - - /* Skip the first cfs_fail_val, then fail */ - if (cfs_fail_loc & CFS_FAIL_SKIP) { - if (cfs_atomic_inc_return(&cfs_fail_count) <= cfs_fail_val) - return 0; - } - - /* Fail cfs_fail_val times, overridden by FAIL_ONCE */ - if (cfs_fail_loc & CFS_FAIL_SOME && - (!(cfs_fail_loc & CFS_FAIL_ONCE) || cfs_fail_val <= 1)) { - int count = cfs_atomic_inc_return(&cfs_fail_count); - - if (count >= cfs_fail_val) { - cfs_set_bit(CFS_FAIL_ONCE_BIT, &cfs_fail_loc); - cfs_atomic_set(&cfs_fail_count, 0); - /* we are lost race to increase */ - if (count > cfs_fail_val) - return 0; - } - } - - if ((set == CFS_FAIL_LOC_ORSET || set == CFS_FAIL_LOC_RESET) && - (value & CFS_FAIL_ONCE)) - cfs_set_bit(CFS_FAIL_ONCE_BIT, &cfs_fail_loc); - /* Lost race to set CFS_FAILED_BIT. */ - if (cfs_test_and_set_bit(CFS_FAILED_BIT, &cfs_fail_loc)) { - /* If CFS_FAIL_ONCE is valid, only one process can fail, - * otherwise multi-process can fail at the same time. */ - if (cfs_fail_loc & CFS_FAIL_ONCE) - return 0; - } - - switch (set) { - case CFS_FAIL_LOC_NOSET: - break; - case CFS_FAIL_LOC_ORSET: - cfs_fail_loc |= value & ~(CFS_FAILED | CFS_FAIL_ONCE); - break; - case CFS_FAIL_LOC_RESET: - cfs_fail_loc = value; - break; - default: - LASSERTF(0, "called with bad set %u\n", set); - break; - } - - return 1; + static atomic_t cfs_fail_count = ATOMIC_INIT(0); + + LASSERT(!(id & CFS_FAIL_ONCE)); + + if ((cfs_fail_loc & (CFS_FAILED | CFS_FAIL_ONCE)) == + (CFS_FAILED | CFS_FAIL_ONCE)) { + atomic_set(&cfs_fail_count, 0); /* paranoia */ + return 0; + } + + /* Fail 1/cfs_fail_val times */ + if (cfs_fail_loc & CFS_FAIL_RAND) { + if (cfs_fail_val < 2 || prandom_u32_max(cfs_fail_val) > 0) + return 0; + } + + /* Skip the first cfs_fail_val, then fail */ + if (cfs_fail_loc & CFS_FAIL_SKIP) { + if (atomic_inc_return(&cfs_fail_count) <= cfs_fail_val) + return 0; + } + + /* check cfs_fail_val... */ + if (set == CFS_FAIL_LOC_VALUE) { + if (cfs_fail_val != -1 && cfs_fail_val != value) + return 0; + } + + /* Fail cfs_fail_val times, overridden by FAIL_ONCE */ + if (cfs_fail_loc & CFS_FAIL_SOME && + (!(cfs_fail_loc & CFS_FAIL_ONCE) || cfs_fail_val <= 1)) { + int count = atomic_inc_return(&cfs_fail_count); + + if (count >= cfs_fail_val) { + set_bit(CFS_FAIL_ONCE_BIT, &cfs_fail_loc); + atomic_set(&cfs_fail_count, 0); + /* we are lost race to increase */ + if (count > cfs_fail_val) + return 0; + } + } + + /* Take into account the current call for FAIL_ONCE for ORSET only, + * as RESET is a new fail_loc, it does not change the current call */ + if ((set == CFS_FAIL_LOC_ORSET) && (value & CFS_FAIL_ONCE)) + set_bit(CFS_FAIL_ONCE_BIT, &cfs_fail_loc); + /* Lost race to set CFS_FAILED_BIT. */ + if (test_and_set_bit(CFS_FAILED_BIT, &cfs_fail_loc)) { + /* If CFS_FAIL_ONCE is valid, only one process can fail, + * otherwise multi-process can fail at the same time. */ + if (cfs_fail_loc & CFS_FAIL_ONCE) + return 0; + } + + switch (set) { + case CFS_FAIL_LOC_NOSET: + case CFS_FAIL_LOC_VALUE: + break; + case CFS_FAIL_LOC_ORSET: + cfs_fail_loc |= value & ~(CFS_FAILED | CFS_FAIL_ONCE); + break; + case CFS_FAIL_LOC_RESET: + cfs_fail_loc = value; + atomic_set(&cfs_fail_count, 0); + break; + default: + LASSERTF(0, "called with bad set %u\n", set); + break; + } + + return 1; } -CFS_EXPORT_SYMBOL(__cfs_fail_check_set); +EXPORT_SYMBOL(__cfs_fail_check_set); int __cfs_fail_timeout_set(__u32 id, __u32 value, int ms, int set) { - int ret = 0; - - ret = __cfs_fail_check_set(id, value, set); - if (ret) { - CERROR("cfs_fail_timeout id %x sleeping for %dms\n", - id, ms); - cfs_schedule_timeout_and_set_state(CFS_TASK_UNINT, - cfs_time_seconds(ms) / 1000); - cfs_set_current_state(CFS_TASK_RUNNING); - CERROR("cfs_fail_timeout id %x awake\n", id); - } - return ret; + ktime_t till = ktime_add_ms(ktime_get(), ms); + int ret = 0; + + ret = __cfs_fail_check_set(id, value, set); + if (ret && likely(ms > 0)) { + CERROR("cfs_fail_timeout id %x sleeping for %dms\n", id, ms); + while (ktime_before(ktime_get(), till)) { + schedule_timeout_uninterruptible(cfs_time_seconds(1) + / 10); + set_current_state(TASK_RUNNING); + if (!cfs_fail_loc) { + CERROR("cfs_fail_timeout interrupted\n"); + break; + } + } + if (cfs_fail_loc) + CERROR("cfs_fail_timeout id %x awake\n", id); + } + return ret; } -CFS_EXPORT_SYMBOL(__cfs_fail_timeout_set); +EXPORT_SYMBOL(__cfs_fail_timeout_set);