From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932495AbbBZOEp (ORCPT ); Thu, 26 Feb 2015 09:04:45 -0500 Received: from mx1.redhat.com ([209.132.183.28]:50739 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932726AbbBZOCn (ORCPT ); Thu, 26 Feb 2015 09:02:43 -0500 Organization: Red Hat UK Ltd. Registered Address: Red Hat UK Ltd, Amberley Place, 107-111 Peascod Street, Windsor, Berkshire, SI4 1TE, United Kingdom. Registered in England and Wales under Company Registration No. 3798903 Subject: [PATCH 07/13] FS-Cache: Permit fscache_cancel_op() to cancel in-progress operations too From: David Howells To: linux-cachefs@redhat.com Cc: dhowells@redhat.com, linux-fsdevel@vger.kernel.org, linux-nfs@vger.kernel.org, linux-kernel@vger.kernel.org Date: Thu, 26 Feb 2015 14:02:39 +0000 Message-ID: <20150226140239.2387.13014.stgit@warthog.procyon.org.uk> In-Reply-To: <20150226140155.2387.70579.stgit@warthog.procyon.org.uk> References: <20150226140155.2387.70579.stgit@warthog.procyon.org.uk> User-Agent: StGit/0.17.1-dirty MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Currently, fscache_cancel_op() only cancels pending operations - attempts to cancel in-progress operations are ignored. This leads to a problem in fscache_wait_for_operation_activation() whereby the wait is terminated, but the object has been killed. The check at the end of the function now triggers because it's no longer contingent on the cache having produced an I/O error since the commit that fixed the logic error in fscache_object_is_dead(). The result of the check is that it tries to cancel the operation - but since the object may not be pending by this point, the cancellation request may be ignored - with the result that the the object is just put by the caller and fscache_put_operation has an assertion failure because the operation isn't in either the COMPLETE or the CANCELLED states. To fix this, we permit in-progress ops to be cancelled under some circumstances. The bug results in an oops that looks something like this: FS-Cache: fscache_wait_for_operation_activation() = -ENOBUFS [obj dead 3] FS-Cache: FS-Cache: Assertion failed FS-Cache: 3 == 5 is false ------------[ cut here ]------------ kernel BUG at ../fs/fscache/operation.c:432! ... RIP: 0010:[] fscache_put_operation+0xf2/0x2cd Call Trace: [] __fscache_read_or_alloc_pages+0x2ec/0x3b3 [] __nfs_readpages_from_fscache+0x59/0xbf [nfs] [] nfs_readpages+0x10c/0x185 [nfs] [] ? alloc_pages_current+0x119/0x13e [] ? __page_cache_alloc+0xfb/0x10a [] __do_page_cache_readahead+0x188/0x22c [] ondemand_readahead+0x29e/0x2af [] page_cache_sync_readahead+0x38/0x3a [] generic_file_read_iter+0x1a2/0x55a [] ? nfs_revalidate_mapping+0xd6/0x288 [nfs] [] nfs_file_read+0x49/0x70 [nfs] [] new_sync_read+0x78/0x9c [] __vfs_read+0x13/0x38 [] vfs_read+0x95/0x121 [] SyS_read+0x4c/0x8a [] system_call_fastpath+0x12/0x17 Signed-off-by: David Howells --- fs/fscache/internal.h | 3 ++- fs/fscache/operation.c | 20 +++++++++++++++++--- fs/fscache/page.c | 4 ++-- 3 files changed, 21 insertions(+), 6 deletions(-) diff --git a/fs/fscache/internal.h b/fs/fscache/internal.h index 3063a58b7d3d..87c4544ec912 100644 --- a/fs/fscache/internal.h +++ b/fs/fscache/internal.h @@ -125,7 +125,8 @@ extern int fscache_submit_exclusive_op(struct fscache_object *, extern int fscache_submit_op(struct fscache_object *, struct fscache_operation *); extern int fscache_cancel_op(struct fscache_operation *, - void (*)(struct fscache_operation *)); + void (*)(struct fscache_operation *), + bool); extern void fscache_cancel_all_ops(struct fscache_object *); extern void fscache_abort_object(struct fscache_object *); extern void fscache_start_operations(struct fscache_object *); diff --git a/fs/fscache/operation.c b/fs/fscache/operation.c index 18658fffbba1..67594a8d791a 100644 --- a/fs/fscache/operation.c +++ b/fs/fscache/operation.c @@ -312,9 +312,11 @@ void fscache_start_operations(struct fscache_object *object) * cancel an operation that's pending on an object */ int fscache_cancel_op(struct fscache_operation *op, - void (*do_cancel)(struct fscache_operation *)) + void (*do_cancel)(struct fscache_operation *), + bool cancel_in_progress_op) { struct fscache_object *object = op->object; + bool put = false; int ret; _enter("OBJ%x OP%x}", op->object->debug_id, op->debug_id); @@ -328,8 +330,19 @@ int fscache_cancel_op(struct fscache_operation *op, ret = -EBUSY; if (op->state == FSCACHE_OP_ST_PENDING) { ASSERT(!list_empty(&op->pend_link)); - fscache_stat(&fscache_n_op_cancelled); list_del_init(&op->pend_link); + put = true; + fscache_stat(&fscache_n_op_cancelled); + if (do_cancel) + do_cancel(op); + op->state = FSCACHE_OP_ST_CANCELLED; + if (test_bit(FSCACHE_OP_EXCLUSIVE, &op->flags)) + object->n_exclusive--; + if (test_and_clear_bit(FSCACHE_OP_WAITING, &op->flags)) + wake_up_bit(&op->flags, FSCACHE_OP_WAITING); + ret = 0; + } else if (op->state == FSCACHE_OP_ST_IN_PROGRESS && cancel_in_progress_op) { + fscache_stat(&fscache_n_op_cancelled); if (do_cancel) do_cancel(op); op->state = FSCACHE_OP_ST_CANCELLED; @@ -337,10 +350,11 @@ int fscache_cancel_op(struct fscache_operation *op, object->n_exclusive--; if (test_and_clear_bit(FSCACHE_OP_WAITING, &op->flags)) wake_up_bit(&op->flags, FSCACHE_OP_WAITING); - fscache_put_operation(op); ret = 0; } + if (put) + fscache_put_operation(op); spin_unlock(&object->lock); _leave(" = %d", ret); return ret; diff --git a/fs/fscache/page.c b/fs/fscache/page.c index d0805e31361c..433cae927eca 100644 --- a/fs/fscache/page.c +++ b/fs/fscache/page.c @@ -359,7 +359,7 @@ int fscache_wait_for_operation_activation(struct fscache_object *object, fscache_stat(stat_op_waits); if (wait_on_bit(&op->flags, FSCACHE_OP_WAITING, TASK_INTERRUPTIBLE) != 0) { - ret = fscache_cancel_op(op, do_cancel); + ret = fscache_cancel_op(op, do_cancel, false); if (ret == 0) return -ERESTARTSYS; @@ -380,7 +380,7 @@ check_if_dead: if (unlikely(fscache_object_is_dying(object) || fscache_cache_is_broken(object))) { enum fscache_operation_state state = op->state; - fscache_cancel_op(op, do_cancel); + fscache_cancel_op(op, do_cancel, true); if (stat_object_dead) fscache_stat(stat_object_dead); _leave(" = -ENOBUFS [obj dead %d]", state);