{"id":807080,"url":"http://patchwork.ozlabs.org/api/1.0/patches/807080/?format=json","project":{"id":14,"url":"http://patchwork.ozlabs.org/api/1.0/projects/14/?format=json","name":"QEMU Development","link_name":"qemu-devel","list_id":"qemu-devel.nongnu.org","list_email":"qemu-devel@nongnu.org","web_url":"","scm_url":"","webscm_url":""},"msgid":"<20170829122745.14309-2-stefanha@redhat.com>","date":"2017-08-29T12:27:43","name":"[v2,1/3] nbd-client: avoid read_reply_co entry if send failed","commit_ref":null,"pull_url":null,"state":"new","archived":false,"hash":"1b126bc77adce028fda3099e3f0c36a237bfbea7","submitter":{"id":17227,"url":"http://patchwork.ozlabs.org/api/1.0/people/17227/?format=json","name":"Stefan Hajnoczi","email":"stefanha@redhat.com"},"delegate":null,"mbox":"http://patchwork.ozlabs.org/project/qemu-devel/patch/20170829122745.14309-2-stefanha@redhat.com/mbox/","series":[{"id":359,"url":"http://patchwork.ozlabs.org/api/1.0/series/359/?format=json","date":"2017-08-29T12:27:42","name":"nbd-client: enter read_reply_co during init to avoid crash","version":2,"mbox":"http://patchwork.ozlabs.org/series/359/mbox/"}],"check":"pending","checks":"http://patchwork.ozlabs.org/api/patches/807080/checks/","tags":{},"headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":"patchwork-incoming@bilbo.ozlabs.org","Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=nongnu.org\n\t(client-ip=2001:4830:134:3::11; helo=lists.gnu.org;\n\tenvelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org;\n\treceiver=)","ext-mx02.extmail.prod.ext.phx2.redhat.com;\n\tdmarc=none (p=none dis=none) header.from=redhat.com","ext-mx02.extmail.prod.ext.phx2.redhat.com;\n\tspf=fail smtp.mailfrom=stefanha@redhat.com"],"Received":["from lists.gnu.org (lists.gnu.org [IPv6:2001:4830:134:3::11])\n\t(using TLSv1 with cipher AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xhSfp58PKz9sNc\n\tfor ;\n\tTue, 29 Aug 2017 22:31:14 +1000 (AEST)","from localhost ([::1]:44718 helo=lists.gnu.org)\n\tby lists.gnu.org with esmtp (Exim 4.71) (envelope-from\n\t)\n\tid 1dmffk-0007ot-Kg\n\tfor incoming@patchwork.ozlabs.org; Tue, 29 Aug 2017 08:31:12 -0400","from eggs.gnu.org ([2001:4830:134:3::10]:43065)\n\tby lists.gnu.org with esmtp (Exim 4.71)\n\t(envelope-from ) id 1dmfcf-0005Ze-Kx\n\tfor qemu-devel@nongnu.org; Tue, 29 Aug 2017 08:28:07 -0400","from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)\n\t(envelope-from ) id 1dmfce-0002kS-G1\n\tfor qemu-devel@nongnu.org; Tue, 29 Aug 2017 08:28:01 -0400","from mx1.redhat.com ([209.132.183.28]:18537)\n\tby eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32)\n\t(Exim 4.71) (envelope-from )\n\tid 1dmfcb-0002ih-Bo; Tue, 29 Aug 2017 08:27:57 -0400","from smtp.corp.redhat.com\n\t(int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12])\n\t(using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits))\n\t(No client certificate requested)\n\tby mx1.redhat.com (Postfix) with ESMTPS id 4EA1780C03;\n\tTue, 29 Aug 2017 12:27:56 +0000 (UTC)","from localhost (ovpn-117-175.ams2.redhat.com [10.36.117.175])\n\tby smtp.corp.redhat.com (Postfix) with ESMTP id 4F33970651;\n\tTue, 29 Aug 2017 12:27:52 +0000 (UTC)"],"DMARC-Filter":"OpenDMARC Filter v1.3.2 mx1.redhat.com 4EA1780C03","From":"Stefan Hajnoczi ","To":"","Date":"Tue, 29 Aug 2017 13:27:43 +0100","Message-Id":"<20170829122745.14309-2-stefanha@redhat.com>","In-Reply-To":"<20170829122745.14309-1-stefanha@redhat.com>","References":"<20170829122745.14309-1-stefanha@redhat.com>","X-Scanned-By":"MIMEDefang 2.79 on 10.5.11.12","X-Greylist":"Sender IP whitelisted, not delayed by milter-greylist-4.5.16\n\t(mx1.redhat.com [10.5.110.26]);\n\tTue, 29 Aug 2017 12:27:56 +0000 (UTC)","X-detected-operating-system":"by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic]\n\t[fuzzy]","X-Received-From":"209.132.183.28","Subject":"[Qemu-devel] [PATCH v2 1/3] nbd-client: avoid read_reply_co entry\n\tif send failed","X-BeenThere":"qemu-devel@nongnu.org","X-Mailman-Version":"2.1.21","Precedence":"list","List-Id":"","List-Unsubscribe":",\n\t","List-Archive":"","List-Post":"","List-Help":"","List-Subscribe":",\n\t","Cc":"Kevin Wolf ,\n\tVladimir Sementsov-Ogievskiy ,\n\tqemu-block@nongnu.org, Stefan Hajnoczi ,\n\tPaolo Bonzini ","Errors-To":"qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org","Sender":"\"Qemu-devel\"\n\t"},"content":"The following segfault is encountered if the NBD server closes the UNIX\ndomain socket immediately after negotiation:\n\n Program terminated with signal SIGSEGV, Segmentation fault.\n #0 aio_co_schedule (ctx=0x0, co=0xd3c0ff2ef0) at util/async.c:441\n 441 QSLIST_INSERT_HEAD_ATOMIC(&ctx->scheduled_coroutines,\n (gdb) bt\n #0 0x000000d3c01a50f8 in aio_co_schedule (ctx=0x0, co=0xd3c0ff2ef0) at util/async.c:441\n #1 0x000000d3c012fa90 in nbd_coroutine_end (bs=bs@entry=0xd3c0fec650, request=) at block/nbd-client.c:207\n #2 0x000000d3c012fb58 in nbd_client_co_preadv (bs=0xd3c0fec650, offset=0, bytes=, qiov=0x7ffc10a91b20, flags=0) at block/nbd-client.c:237\n #3 0x000000d3c0128e63 in bdrv_driver_preadv (bs=bs@entry=0xd3c0fec650, offset=offset@entry=0, bytes=bytes@entry=512, qiov=qiov@entry=0x7ffc10a91b20, flags=0) at block/io.c:836\n #4 0x000000d3c012c3e0 in bdrv_aligned_preadv (child=child@entry=0xd3c0ff51d0, req=req@entry=0x7f31885d6e90, offset=offset@entry=0, bytes=bytes@entry=512, align=align@entry=1, qiov=qiov@entry=0x7ffc10a91b20, f\n+lags=0) at block/io.c:1086\n #5 0x000000d3c012c6b8 in bdrv_co_preadv (child=0xd3c0ff51d0, offset=offset@entry=0, bytes=bytes@entry=512, qiov=qiov@entry=0x7ffc10a91b20, flags=flags@entry=0) at block/io.c:1182\n #6 0x000000d3c011cc17 in blk_co_preadv (blk=0xd3c0ff4f80, offset=0, bytes=512, qiov=0x7ffc10a91b20, flags=0) at block/block-backend.c:1032\n #7 0x000000d3c011ccec in blk_read_entry (opaque=0x7ffc10a91b40) at block/block-backend.c:1079\n #8 0x000000d3c01bbb96 in coroutine_trampoline (i0=, i1=) at util/coroutine-ucontext.c:79\n #9 0x00007f3196cb8600 in __start_context () at /lib64/libc.so.6\n\nThe problem is that nbd_client_init() uses\nnbd_client_attach_aio_context() -> aio_co_schedule(new_context,\nclient->read_reply_co). Execution of read_reply_co is deferred to a BH\nwhich doesn't run until later.\n\nIn the mean time blk_co_preadv() can be called and nbd_coroutine_end()\ncalls aio_wake() on read_reply_co. At this point in time\nread_reply_co's ctx isn't set because it has never been entered yet.\n\nThis patch simplifies the nbd_co_send_request() ->\nnbd_co_receive_reply() -> nbd_coroutine_end() lifecycle to just\nnbd_co_send_request() -> nbd_co_receive_reply(). The request is \"ended\"\nif an error occurs at any point. Callers no longer have to invoke\nnbd_coroutine_end().\n\nThis cleanup also eliminates the segfault because we don't call\naio_co_schedule() to wake up s->read_reply_co if sending the request\nfailed. It is only necessary to wake up s->read_reply_co if a reply was\nreceived.\n\nNote this only happens with UNIX domain sockets on Linux. It doesn't\nseem possible to reproduce this with TCP sockets.\n\nSuggested-by: Paolo Bonzini \nSigned-off-by: Stefan Hajnoczi \n---\n block/nbd-client.c | 25 +++++++++----------------\n 1 file changed, 9 insertions(+), 16 deletions(-)","diff":"diff --git a/block/nbd-client.c b/block/nbd-client.c\nindex 25bcaa2346..ea728fffc8 100644\n--- a/block/nbd-client.c\n+++ b/block/nbd-client.c\n@@ -144,12 +144,12 @@ static int nbd_co_send_request(BlockDriverState *bs,\n request->handle = INDEX_TO_HANDLE(s, i);\n \n if (s->quit) {\n- qemu_co_mutex_unlock(&s->send_mutex);\n- return -EIO;\n+ rc = -EIO;\n+ goto err;\n }\n if (!s->ioc) {\n- qemu_co_mutex_unlock(&s->send_mutex);\n- return -EPIPE;\n+ rc = -EPIPE;\n+ goto err;\n }\n \n if (qiov) {\n@@ -166,8 +166,13 @@ static int nbd_co_send_request(BlockDriverState *bs,\n } else {\n rc = nbd_send_request(s->ioc, request);\n }\n+\n+err:\n if (rc < 0) {\n s->quit = true;\n+ s->requests[i].coroutine = NULL;\n+ s->in_flight--;\n+ qemu_co_queue_next(&s->free_sema);\n }\n qemu_co_mutex_unlock(&s->send_mutex);\n return rc;\n@@ -201,13 +206,6 @@ static void nbd_co_receive_reply(NBDClientSession *s,\n /* Tell the read handler to read another header. */\n s->reply.handle = 0;\n }\n-}\n-\n-static void nbd_coroutine_end(BlockDriverState *bs,\n- NBDRequest *request)\n-{\n- NBDClientSession *s = nbd_get_client_session(bs);\n- int i = HANDLE_TO_INDEX(s, request->handle);\n \n s->requests[i].coroutine = NULL;\n \n@@ -243,7 +241,6 @@ int nbd_client_co_preadv(BlockDriverState *bs, uint64_t offset,\n } else {\n nbd_co_receive_reply(client, &request, &reply, qiov);\n }\n- nbd_coroutine_end(bs, &request);\n return -reply.error;\n }\n \n@@ -272,7 +269,6 @@ int nbd_client_co_pwritev(BlockDriverState *bs, uint64_t offset,\n } else {\n nbd_co_receive_reply(client, &request, &reply, NULL);\n }\n- nbd_coroutine_end(bs, &request);\n return -reply.error;\n }\n \n@@ -306,7 +302,6 @@ int nbd_client_co_pwrite_zeroes(BlockDriverState *bs, int64_t offset,\n } else {\n nbd_co_receive_reply(client, &request, &reply, NULL);\n }\n- nbd_coroutine_end(bs, &request);\n return -reply.error;\n }\n \n@@ -330,7 +325,6 @@ int nbd_client_co_flush(BlockDriverState *bs)\n } else {\n nbd_co_receive_reply(client, &request, &reply, NULL);\n }\n- nbd_coroutine_end(bs, &request);\n return -reply.error;\n }\n \n@@ -355,7 +349,6 @@ int nbd_client_co_pdiscard(BlockDriverState *bs, int64_t offset, int bytes)\n } else {\n nbd_co_receive_reply(client, &request, &reply, NULL);\n }\n- nbd_coroutine_end(bs, &request);\n return -reply.error;\n \n }\n","prefixes":["v2","1/3"]}