From patchwork Thu Oct 11 07:21:33 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Fam Zheng X-Patchwork-Id: 10636229 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id A432E16B1 for ; Thu, 11 Oct 2018 07:23:05 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 84D792AEE0 for ; Thu, 11 Oct 2018 07:23:05 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 785012AF07; Thu, 11 Oct 2018 07:23:05 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.9 required=2.0 tests=BAYES_00,MAILING_LIST_MULTI, RCVD_IN_DNSWL_HI autolearn=ham version=3.3.1 Received: from lists.gnu.org (lists.gnu.org [208.118.235.17]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id E4EE72AEE0 for ; Thu, 11 Oct 2018 07:23:04 +0000 (UTC) Received: from localhost ([::1]:60914 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gAVJI-0007oJ-21 for patchwork-qemu-devel@patchwork.kernel.org; Thu, 11 Oct 2018 03:23:04 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:49232) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gAVI3-0006ij-DL for qemu-devel@nongnu.org; Thu, 11 Oct 2018 03:21:48 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1gAVI2-000375-BB for qemu-devel@nongnu.org; Thu, 11 Oct 2018 03:21:47 -0400 Received: from mx1.redhat.com ([209.132.183.28]:42174) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1gAVHz-00035e-U0; Thu, 11 Oct 2018 03:21:44 -0400 Received: from smtp.corp.redhat.com (int-mx08.intmail.prod.int.phx2.redhat.com [10.5.11.23]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 40754A53BA; Thu, 11 Oct 2018 07:21:43 +0000 (UTC) Received: from magic.redhat.com (ovpn-12-151.pek2.redhat.com [10.72.12.151]) by smtp.corp.redhat.com (Postfix) with ESMTP id B176017CEC; Thu, 11 Oct 2018 07:21:39 +0000 (UTC) From: Fam Zheng To: qemu-devel@nongnu.org Date: Thu, 11 Oct 2018 15:21:33 +0800 Message-Id: <20181011072135.588-2-famz@redhat.com> In-Reply-To: <20181011072135.588-1-famz@redhat.com> References: <20181011072135.588-1-famz@redhat.com> X-Scanned-By: MIMEDefang 2.84 on 10.5.11.23 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.26]); Thu, 11 Oct 2018 07:21:43 +0000 (UTC) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] X-Received-From: 209.132.183.28 Subject: [Qemu-devel] [PATCH v5 1/3] file-posix: Skip effectiveless OFD lock operations X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Kevin Wolf , qemu-block@nongnu.org, Max Reitz Errors-To: qemu-devel-bounces+patchwork-qemu-devel=patchwork.kernel.org@nongnu.org Sender: "Qemu-devel" X-Virus-Scanned: ClamAV using ClamSMTP If we know we've already locked the bytes, don't do it again; similarly don't unlock a byte if we haven't locked it. This doesn't change the behavior, but fixes a corner case explained below. Libvirt had an error handling bug that an image can get its (ownership, file mode, SELinux) permissions changed (RHBZ 1584982) by mistake behind QEMU. Specifically, an image in use by Libvirt VM has: $ ls -lhZ b.img -rw-r--r--. qemu qemu system_u:object_r:svirt_image_t:s0:c600,c690 b.img Trying to attach it a second time won't work because of image locking. And after the error, it becomes: $ ls -lhZ b.img -rw-r--r--. root root system_u:object_r:virt_image_t:s0 b.img Then, we won't be able to do OFD lock operations with the existing fd. In other words, the code such as in blk_detach_dev: blk_set_perm(blk, 0, BLK_PERM_ALL, &error_abort); can abort() QEMU, out of environmental changes. This patch is an easy fix to this and the change is regardlessly reasonable, so do it. Signed-off-by: Fam Zheng Reviewed-by: Max Reitz --- block/file-posix.c | 54 +++++++++++++++++++++++++++++++++++++--------- 1 file changed, 44 insertions(+), 10 deletions(-) diff --git a/block/file-posix.c b/block/file-posix.c index 2da3a76355..cf5eb98caa 100644 --- a/block/file-posix.c +++ b/block/file-posix.c @@ -152,6 +152,11 @@ typedef struct BDRVRawState { uint64_t perm; uint64_t shared_perm; + /* The perms bits whose corresponding bytes are already locked in + * s->lock_fd. */ + uint64_t locked_perm; + uint64_t locked_shared_perm; + #ifdef CONFIG_XFS bool is_xfs:1; #endif @@ -680,43 +685,72 @@ typedef enum { * file; if @unlock == true, also unlock the unneeded bytes. * @shared_perm_lock_bits is the mask of all permissions that are NOT shared. */ -static int raw_apply_lock_bytes(int fd, +static int raw_apply_lock_bytes(BDRVRawState *s, int fd, uint64_t perm_lock_bits, uint64_t shared_perm_lock_bits, bool unlock, Error **errp) { int ret; int i; + uint64_t locked_perm, locked_shared_perm; + + if (s) { + locked_perm = s->locked_perm; + locked_shared_perm = s->locked_shared_perm; + } else { + /* + * We don't have the previous bits, just lock/unlock for each of the + * requested bits. + */ + if (unlock) { + locked_perm = BLK_PERM_ALL; + locked_shared_perm = BLK_PERM_ALL; + } else { + locked_perm = 0; + locked_shared_perm = 0; + } + } PERM_FOREACH(i) { int off = RAW_LOCK_PERM_BASE + i; - if (perm_lock_bits & (1ULL << i)) { + uint64_t bit = (1ULL << i); + if ((perm_lock_bits & bit) && !(locked_perm & bit)) { ret = qemu_lock_fd(fd, off, 1, false); if (ret) { error_setg(errp, "Failed to lock byte %d", off); return ret; + } else if (s) { + s->locked_perm |= bit; } - } else if (unlock) { + } else if (unlock && (locked_perm & bit) && !(perm_lock_bits & bit)) { ret = qemu_unlock_fd(fd, off, 1); if (ret) { error_setg(errp, "Failed to unlock byte %d", off); return ret; + } else if (s) { + s->locked_perm &= ~bit; } } } PERM_FOREACH(i) { int off = RAW_LOCK_SHARED_BASE + i; - if (shared_perm_lock_bits & (1ULL << i)) { + uint64_t bit = (1ULL << i); + if ((shared_perm_lock_bits & bit) && !(locked_shared_perm & bit)) { ret = qemu_lock_fd(fd, off, 1, false); if (ret) { error_setg(errp, "Failed to lock byte %d", off); return ret; + } else if (s) { + s->locked_shared_perm |= bit; } - } else if (unlock) { + } else if (unlock && (locked_shared_perm & bit) && + !(shared_perm_lock_bits & bit)) { ret = qemu_unlock_fd(fd, off, 1); if (ret) { error_setg(errp, "Failed to unlock byte %d", off); return ret; + } else if (s) { + s->locked_shared_perm &= ~bit; } } } @@ -784,7 +818,7 @@ static int raw_handle_perm_lock(BlockDriverState *bs, switch (op) { case RAW_PL_PREPARE: - ret = raw_apply_lock_bytes(s->lock_fd, s->perm | new_perm, + ret = raw_apply_lock_bytes(s, s->lock_fd, s->perm | new_perm, ~s->shared_perm | ~new_shared, false, errp); if (!ret) { @@ -799,7 +833,7 @@ static int raw_handle_perm_lock(BlockDriverState *bs, op = RAW_PL_ABORT; /* fall through to unlock bytes. */ case RAW_PL_ABORT: - raw_apply_lock_bytes(s->lock_fd, s->perm, ~s->shared_perm, + raw_apply_lock_bytes(s, s->lock_fd, s->perm, ~s->shared_perm, true, &local_err); if (local_err) { /* Theoretically the above call only unlocks bytes and it cannot @@ -809,7 +843,7 @@ static int raw_handle_perm_lock(BlockDriverState *bs, } break; case RAW_PL_COMMIT: - raw_apply_lock_bytes(s->lock_fd, new_perm, ~new_shared, + raw_apply_lock_bytes(s, s->lock_fd, new_perm, ~new_shared, true, &local_err); if (local_err) { /* Theoretically the above call only unlocks bytes and it cannot @@ -2213,7 +2247,7 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp) shared = BLK_PERM_ALL & ~BLK_PERM_RESIZE; /* Step one: Take locks */ - result = raw_apply_lock_bytes(fd, perm, ~shared, false, errp); + result = raw_apply_lock_bytes(NULL, fd, perm, ~shared, false, errp); if (result < 0) { goto out_close; } @@ -2257,7 +2291,7 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp) } out_unlock: - raw_apply_lock_bytes(fd, 0, 0, true, &local_err); + raw_apply_lock_bytes(NULL, fd, 0, 0, true, &local_err); if (local_err) { /* The above call should not fail, and if it does, that does * not mean the whole creation operation has failed. So