From patchwork Sun Aug 9 06:30:37 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Zorro Lang X-Patchwork-Id: 11706555 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 9537313B1 for ; Sun, 9 Aug 2020 06:31:10 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 79C0220729 for ; Sun, 9 Aug 2020 06:31:10 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="ele+KVFF" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1725988AbgHIGbK (ORCPT ); Sun, 9 Aug 2020 02:31:10 -0400 Received: from us-smtp-2.mimecast.com ([205.139.110.61]:34069 "EHLO us-smtp-delivery-1.mimecast.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1726199AbgHIGbI (ORCPT ); Sun, 9 Aug 2020 02:31:08 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1596954666; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=Jg2onTdkbTF7vO+lIBSmKY0XWjQFIp7tSGuQvPchvXM=; b=ele+KVFF6A7ZK+kKLFfNnwFF6dRCIuki/4cD3eVEBWBTOu5k8XFQ3uR9neYFbfEq2TOkqD RCQWzGFa3sXjZOpEnxU3nwGpkmTAyjnQpIYY/hARQLk/bMTHA0KrkfpHkEEY3A6/0nU/zd PIUImcMvZF7O2GNo+ZnEplh1GL2uCmg= Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-148-45lqglAFMPy6-UM2V7d2yw-1; Sun, 09 Aug 2020 02:30:50 -0400 X-MC-Unique: 45lqglAFMPy6-UM2V7d2yw-1 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.phx2.redhat.com [10.5.11.16]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 9203D100A8C1; Sun, 9 Aug 2020 06:30:49 +0000 (UTC) Received: from localhost.localdomain.com (ovpn-12-33.pek2.redhat.com [10.72.12.33]) by smtp.corp.redhat.com (Postfix) with ESMTP id B31C25C6DD; Sun, 9 Aug 2020 06:30:47 +0000 (UTC) From: Zorro Lang To: fstests@vger.kernel.org Cc: io-uring@vger.kernel.org, jmoyer@redhat.com Subject: [PATCH v2 1/4] fsstress: add IO_URING read and write operations Date: Sun, 9 Aug 2020 14:30:37 +0800 Message-Id: <20200809063040.15521-2-zlang@redhat.com> In-Reply-To: <20200809063040.15521-1-zlang@redhat.com> References: <20200809063040.15521-1-zlang@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.16 Sender: fstests-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: fstests@vger.kernel.org IO_URING is a new feature of curent linux kernel, add basic IO_URING read/write into fsstess to cover this kind of IO testing. Signed-off-by: Zorro Lang --- README | 4 +- configure.ac | 1 + include/builddefs.in | 1 + ltp/Makefile | 5 ++ ltp/fsstress.c | 139 ++++++++++++++++++++++++++++++++++++++++- m4/Makefile | 1 + m4/package_liburing.m4 | 4 ++ 7 files changed, 152 insertions(+), 3 deletions(-) create mode 100644 m4/package_liburing.m4 diff --git a/README b/README index d0e23fcd..ae0f804d 100644 --- a/README +++ b/README @@ -8,13 +8,13 @@ _______________________ sudo apt-get install xfslibs-dev uuid-dev libtool-bin \ e2fsprogs automake gcc libuuid1 quota attr libattr1-dev make \ libacl1-dev libaio-dev xfsprogs libgdbm-dev gawk fio dbench \ - uuid-runtime python sqlite3 + uuid-runtime python sqlite3 liburing-dev For Fedora, RHEL, or CentOS: yum install acl attr automake bc dbench dump e2fsprogs fio \ gawk gcc indent libtool lvm2 make psmisc quota sed \ xfsdump xfsprogs \ libacl-devel libattr-devel libaio-devel libuuid-devel \ - xfsprogs-devel btrfs-progs-devel python sqlite + xfsprogs-devel btrfs-progs-devel python sqlite liburing-devel (Older distributions may require xfsprogs-qa-devel as well.) (Note that for RHEL and CentOS, you may need the EPEL repo.) - run make diff --git a/configure.ac b/configure.ac index 4bb50b32..8922c47e 100644 --- a/configure.ac +++ b/configure.ac @@ -61,6 +61,7 @@ AC_PACKAGE_NEED_ACLINIT_LIBACL AC_PACKAGE_WANT_GDBM AC_PACKAGE_WANT_AIO +AC_PACKAGE_WANT_URING AC_PACKAGE_WANT_DMAPI AC_PACKAGE_WANT_LINUX_FIEMAP_H AC_PACKAGE_WANT_FALLOCATE diff --git a/include/builddefs.in b/include/builddefs.in index e7894b1a..fded3230 100644 --- a/include/builddefs.in +++ b/include/builddefs.in @@ -61,6 +61,7 @@ RPM_VERSION = @rpm_version@ ENABLE_SHARED = @enable_shared@ HAVE_DB = @have_db@ HAVE_AIO = @have_aio@ +HAVE_URING = @have_uring@ HAVE_FALLOCATE = @have_fallocate@ HAVE_OPEN_BY_HANDLE_AT = @have_open_by_handle_at@ HAVE_DMAPI = @have_dmapi@ diff --git a/ltp/Makefile b/ltp/Makefile index ebf40336..198d930f 100644 --- a/ltp/Makefile +++ b/ltp/Makefile @@ -24,6 +24,11 @@ LCFLAGS += -DAIO LLDLIBS += -laio -lpthread endif +ifeq ($(HAVE_URING), true) +LCFLAGS += -DURING +LLDLIBS += -luring +endif + ifeq ($(HAVE_LIBBTRFSUTIL), true) LLDLIBS += -lbtrfsutil endif diff --git a/ltp/fsstress.c b/ltp/fsstress.c index 709fdeec..a4188e1c 100644 --- a/ltp/fsstress.c +++ b/ltp/fsstress.c @@ -30,6 +30,11 @@ #include io_context_t io_ctx; #endif +#ifdef URING +#include +#define URING_ENTRIES 1 +struct io_uring ring; +#endif #include #include @@ -139,6 +144,8 @@ typedef enum { OP_TRUNCATE, OP_UNLINK, OP_UNRESVSP, + OP_URING_READ, + OP_URING_WRITE, OP_WRITE, OP_WRITEV, OP_LAST @@ -267,6 +274,8 @@ void sync_f(int, long); void truncate_f(int, long); void unlink_f(int, long); void unresvsp_f(int, long); +void uring_read_f(int, long); +void uring_write_f(int, long); void write_f(int, long); void writev_f(int, long); char *xattr_flag_to_string(int); @@ -335,6 +344,8 @@ opdesc_t ops[] = { { OP_TRUNCATE, "truncate", truncate_f, 2, 1 }, { OP_UNLINK, "unlink", unlink_f, 1, 1 }, { OP_UNRESVSP, "unresvsp", unresvsp_f, 1, 1 }, + { OP_URING_READ, "uring_read", uring_read_f, 1, 0 }, + { OP_URING_WRITE, "uring_write", uring_write_f, 1, 1 }, { OP_WRITE, "write", write_f, 4, 1 }, { OP_WRITEV, "writev", writev_f, 4, 1 }, }, *ops_end; @@ -692,6 +703,12 @@ int main(int argc, char **argv) fprintf(stderr, "io_setup failed"); exit(1); } +#endif +#ifdef URING + if (io_uring_queue_init(URING_ENTRIES, &ring, 0)) { + fprintf(stderr, "io_uring_queue_init failed\n"); + exit(1); + } #endif for (i = 0; !loops || (i < loops); i++) doproc(); @@ -701,7 +718,9 @@ int main(int argc, char **argv) return 1; } #endif - +#ifdef URING + io_uring_queue_exit(&ring); +#endif cleanup_flist(); free(freq_table); return 0; @@ -2170,6 +2189,108 @@ do_aio_rw(int opno, long r, int flags) } #endif +#ifdef URING +void +do_uring_rw(int opno, long r, int flags) +{ + char *buf; + int e; + pathname_t f; + int fd; + size_t len; + int64_t lr; + off64_t off; + struct stat64 stb; + int v; + char st[1024]; + struct io_uring_sqe *sqe; + struct io_uring_cqe *cqe; + struct iovec iovec; + int iswrite = (flags & (O_WRONLY | O_RDWR)) ? 1 : 0; + + init_pathname(&f); + if (!get_fname(FT_REGFILE, r, &f, NULL, NULL, &v)) { + if (v) + printf("%d/%d: do_uring_rw - no filename\n", procid, opno); + goto uring_out3; + } + fd = open_path(&f, flags); + e = fd < 0 ? errno : 0; + check_cwd(); + if (fd < 0) { + if (v) + printf("%d/%d: do_uring_rw - open %s failed %d\n", + procid, opno, f.path, e); + goto uring_out3; + } + if (fstat64(fd, &stb) < 0) { + if (v) + printf("%d/%d: do_uring_rw - fstat64 %s failed %d\n", + procid, opno, f.path, errno); + goto uring_out2; + } + inode_info(st, sizeof(st), &stb, v); + if (!iswrite && stb.st_size == 0) { + if (v) + printf("%d/%d: do_uring_rw - %s%s zero size\n", procid, opno, + f.path, st); + goto uring_out2; + } + sqe = io_uring_get_sqe(&ring); + if (!sqe) { + if (v) + printf("%d/%d: do_uring_rw - io_uring_get_sqe failed\n", + procid, opno); + goto uring_out2; + } + lr = ((int64_t)random() << 32) + random(); + len = (random() % FILELEN_MAX) + 1; + buf = malloc(len); + if (!buf) { + if (v) + printf("%d/%d: do_uring_rw - malloc failed\n", + procid, opno); + goto uring_out2; + } + iovec.iov_base = buf; + iovec.iov_len = len; + if (iswrite) { + off = (off64_t)(lr % MIN(stb.st_size + (1024 * 1024), MAXFSIZE)); + off %= maxfsize; + memset(buf, nameseq & 0xff, len); + io_uring_prep_writev(sqe, fd, &iovec, 1, off); + } else { + off = (off64_t)(lr % stb.st_size); + io_uring_prep_readv(sqe, fd, &iovec, 1, off); + } + + if ((e = io_uring_submit(&ring)) != 1) { + if (v) + printf("%d/%d: %s - io_uring_submit failed %d\n", procid, opno, + iswrite ? "uring_write" : "uring_read", e); + goto uring_out1; + } + if ((e = io_uring_wait_cqe(&ring, &cqe)) < 0) { + if (v) + printf("%d/%d: %s - io_uring_wait_cqe failed %d\n", procid, opno, + iswrite ? "uring_write" : "uring_read", e); + goto uring_out1; + } + if (v) + printf("%d/%d: %s %s%s [%lld, %d(res=%d)] %d\n", + procid, opno, iswrite ? "uring_write" : "uring_read", + f.path, st, (long long)off, (int)len, cqe->res, e); + io_uring_cqe_seen(&ring, cqe); + + uring_out1: + free(buf); + uring_out2: + close(fd); + uring_out3: + free_pathname(&f); +} +#endif + void aread_f(int opno, long r) { @@ -5044,6 +5165,22 @@ unresvsp_f(int opno, long r) close(fd); } +void +uring_read_f(int opno, long r) +{ +#ifdef URING + do_uring_rw(opno, r, O_RDONLY); +#endif +} + +void +uring_write_f(int opno, long r) +{ +#ifdef URING + do_uring_rw(opno, r, O_WRONLY); +#endif +} + void write_f(int opno, long r) { diff --git a/m4/Makefile b/m4/Makefile index 7fbff822..0352534d 100644 --- a/m4/Makefile +++ b/m4/Makefile @@ -14,6 +14,7 @@ LSRCFILES = \ package_dmapidev.m4 \ package_globals.m4 \ package_libcdev.m4 \ + package_liburing.m4 \ package_ncurses.m4 \ package_pthread.m4 \ package_ssldev.m4 \ diff --git a/m4/package_liburing.m4 b/m4/package_liburing.m4 new file mode 100644 index 00000000..c92cc02a --- /dev/null +++ b/m4/package_liburing.m4 @@ -0,0 +1,4 @@ +AC_DEFUN([AC_PACKAGE_WANT_URING], + [ AC_CHECK_HEADERS(liburing.h, [ have_uring=true ], [ have_uring=false ]) + AC_SUBST(have_uring) + ]) From patchwork Sun Aug 9 06:30:38 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Zorro Lang X-Patchwork-Id: 11706549 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 49BBA138C for ; Sun, 9 Aug 2020 06:30:58 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 30BE520729 for ; Sun, 9 Aug 2020 06:30:58 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="X1pVvkoe" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726175AbgHIGa5 (ORCPT ); Sun, 9 Aug 2020 02:30:57 -0400 Received: from us-smtp-delivery-124.mimecast.com ([216.205.24.124]:41577 "EHLO us-smtp-delivery-124.mimecast.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726098AbgHIGa5 (ORCPT ); Sun, 9 Aug 2020 02:30:57 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1596954656; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=HZ3iGdp7OPMi+6kd9idtublBxHGrjAF6VXW9ZEuiXd8=; b=X1pVvkoebFA2L23Mb+QxpEv03dW/DG/lds5fH+W1pvDXIvUqHUyI7BUHAqQ31ycamT7W4L CEaOD2gWchMlUUYyIWCPnXpLQ7wF0t+xnDLFZhvH+upYLursXXSVckAxkmUXUN0RkKRi3X GTv9loqnk3fYF/RETCxalAJ4wz+MUaU= Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-355-PmCrZAXOM9-BDZdJmV-bVQ-1; Sun, 09 Aug 2020 02:30:54 -0400 X-MC-Unique: PmCrZAXOM9-BDZdJmV-bVQ-1 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.phx2.redhat.com [10.5.11.16]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 13D5A107B806; Sun, 9 Aug 2020 06:30:53 +0000 (UTC) Received: from localhost.localdomain.com (ovpn-12-33.pek2.redhat.com [10.72.12.33]) by smtp.corp.redhat.com (Postfix) with ESMTP id 7F91F5C6DD; Sun, 9 Aug 2020 06:30:51 +0000 (UTC) From: Zorro Lang To: fstests@vger.kernel.org Cc: io-uring@vger.kernel.org, jmoyer@redhat.com Subject: [PATCH v2 2/4] fsstress: reduce the number of events when io_setup Date: Sun, 9 Aug 2020 14:30:38 +0800 Message-Id: <20200809063040.15521-3-zlang@redhat.com> In-Reply-To: <20200809063040.15521-1-zlang@redhat.com> References: <20200809063040.15521-1-zlang@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.16 Sender: fstests-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: fstests@vger.kernel.org The original number(128) of aio events for io_setup too big. When try to run lots of fsstress processes(e.g. -p 1000) always hit io_setup EAGAIN error, due to the nr_events exceeds the limit of available events. Due to each fsstress process only does once libaio read/write operation each time. So reduce the aio events number to 1, to make more fsstress processes can do AIO test. Signed-off-by: Zorro Lang --- ltp/fsstress.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/ltp/fsstress.c b/ltp/fsstress.c index a4188e1c..0e7be6bb 100644 --- a/ltp/fsstress.c +++ b/ltp/fsstress.c @@ -28,6 +28,7 @@ #endif #ifdef AIO #include +#define AIO_ENTRIES 1 io_context_t io_ctx; #endif #ifdef URING @@ -699,8 +700,8 @@ int main(int argc, char **argv) } procid = i; #ifdef AIO - if (io_setup(128, &io_ctx) != 0) { - fprintf(stderr, "io_setup failed"); + if (io_setup(AIO_ENTRIES, &io_ctx) != 0) { + fprintf(stderr, "io_setup failed\n"); exit(1); } #endif From patchwork Sun Aug 9 06:30:39 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Zorro Lang X-Patchwork-Id: 11706551 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id CAB9A138C for ; Sun, 9 Aug 2020 06:31:02 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id B3B3920771 for ; Sun, 9 Aug 2020 06:31:02 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="OfUhiNri" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726098AbgHIGbC (ORCPT ); Sun, 9 Aug 2020 02:31:02 -0400 Received: from us-smtp-1.mimecast.com ([207.211.31.81]:29016 "EHLO us-smtp-delivery-1.mimecast.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1725988AbgHIGbC (ORCPT ); Sun, 9 Aug 2020 02:31:02 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1596954661; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=X7WZddFwwKHu/gIcw1qCLu59BphJ+2feFSV3X2iMviU=; b=OfUhiNrixesQ2/0GzjsEBAxE9AQyJDPoVE970QyYr/bMTDrSv5fWVKnT9MU26+VojSpwPa Msi+k0RiXJSWjD7PNrnUKBSrc93fyRA8eIWic3S+eHG8HQSDek2d+uGDPDpuagsz549/mN pdg3fhKUsW0AN5DIObfrNZfZoiWDJww= Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-443-zdRvRjmNMbynuK3VCKnpuQ-1; Sun, 09 Aug 2020 02:30:57 -0400 X-MC-Unique: zdRvRjmNMbynuK3VCKnpuQ-1 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.phx2.redhat.com [10.5.11.16]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 35EE58015CE; Sun, 9 Aug 2020 06:30:56 +0000 (UTC) Received: from localhost.localdomain.com (ovpn-12-33.pek2.redhat.com [10.72.12.33]) by smtp.corp.redhat.com (Postfix) with ESMTP id A9C9F5F9DC; Sun, 9 Aug 2020 06:30:54 +0000 (UTC) From: Zorro Lang To: fstests@vger.kernel.org Cc: io-uring@vger.kernel.org, jmoyer@redhat.com Subject: [PATCH v2 3/4] fsstress: fix memory leak in do_aio_rw Date: Sun, 9 Aug 2020 14:30:39 +0800 Message-Id: <20200809063040.15521-4-zlang@redhat.com> In-Reply-To: <20200809063040.15521-1-zlang@redhat.com> References: <20200809063040.15521-1-zlang@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.16 Sender: fstests-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: fstests@vger.kernel.org If io_submit or io_getevents fails, the do_aio_rw() won't free the "buf" and cause memory leak. Signed-off-by: Zorro Lang --- ltp/fsstress.c | 31 ++++++++++++++++--------------- 1 file changed, 16 insertions(+), 15 deletions(-) diff --git a/ltp/fsstress.c b/ltp/fsstress.c index 0e7be6bb..4b255bf6 100644 --- a/ltp/fsstress.c +++ b/ltp/fsstress.c @@ -2099,8 +2099,7 @@ do_aio_rw(int opno, long r, int flags) if (!get_fname(FT_REGFILE, r, &f, NULL, NULL, &v)) { if (v) printf("%d/%d: do_aio_rw - no filename\n", procid, opno); - free_pathname(&f); - return; + goto aio_out3; } fd = open_path(&f, flags|O_DIRECT); e = fd < 0 ? errno : 0; @@ -2109,16 +2108,13 @@ do_aio_rw(int opno, long r, int flags) if (v) printf("%d/%d: do_aio_rw - open %s failed %d\n", procid, opno, f.path, e); - free_pathname(&f); - return; + goto aio_out3; } if (fstat64(fd, &stb) < 0) { if (v) printf("%d/%d: do_aio_rw - fstat64 %s failed %d\n", procid, opno, f.path, errno); - free_pathname(&f); - close(fd); - return; + goto aio_out2; } inode_info(st, sizeof(st), &stb, v); if (!iswrite && stb.st_size == 0) { @@ -2150,6 +2146,12 @@ do_aio_rw(int opno, long r, int flags) else if (len > diob.d_maxiosz) len = diob.d_maxiosz; buf = memalign(diob.d_mem, len); + if (!buf) { + if (v) + printf("%d/%d: do_aio_rw - memalign failed\n", + procid, opno); + goto aio_out2; + } if (iswrite) { off = (off64_t)(lr % MIN(stb.st_size + (1024 * 1024), MAXFSIZE)); @@ -2166,27 +2168,26 @@ do_aio_rw(int opno, long r, int flags) if (v) printf("%d/%d: %s - io_submit failed %d\n", procid, opno, iswrite ? "awrite" : "aread", e); - free_pathname(&f); - close(fd); - return; + goto aio_out1; } if ((e = io_getevents(io_ctx, 1, 1, &event, NULL)) != 1) { if (v) printf("%d/%d: %s - io_getevents failed %d\n", procid, opno, iswrite ? "awrite" : "aread", e); - free_pathname(&f); - close(fd); - return; + goto aio_out1; } e = event.res != len ? event.res2 : 0; - free(buf); if (v) printf("%d/%d: %s %s%s [%lld,%d] %d\n", procid, opno, iswrite ? "awrite" : "aread", f.path, st, (long long)off, (int)len, e); - free_pathname(&f); + aio_out1: + free(buf); + aio_out2: close(fd); + aio_out3: + free_pathname(&f); } #endif From patchwork Sun Aug 9 06:30:40 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Zorro Lang X-Patchwork-Id: 11706553 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 70333138C for ; Sun, 9 Aug 2020 06:31:05 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 4F9AF20771 for ; Sun, 9 Aug 2020 06:31:05 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="ap0hZYTc" Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726207AbgHIGbE (ORCPT ); Sun, 9 Aug 2020 02:31:04 -0400 Received: from us-smtp-1.mimecast.com ([207.211.31.81]:50526 "EHLO us-smtp-delivery-1.mimecast.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1725988AbgHIGbE (ORCPT ); Sun, 9 Aug 2020 02:31:04 -0400 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1596954662; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=+ELvlo8LS2gOVwdB1fdETxuQKV4QUKLJ++eGuDLSmDU=; b=ap0hZYTcIdWnEzk775o7YuSZLSofBZ2UHwRzZ54EvAxC5OAKfdrc6xiY6tSvLxYg3CNl5B aQGQltvUXoTgHLhiTCMtVzZc+Igt7eM8iDhv0Tm00B4yp0kK1Ig9vRfHT37QWmpyFu4iuT WNCn119csBTFF6my4O/iNDJGh0WoFPQ= Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-409-4UTok5RnM_KZ82AJAoV-dQ-1; Sun, 09 Aug 2020 02:31:00 -0400 X-MC-Unique: 4UTok5RnM_KZ82AJAoV-dQ-1 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.phx2.redhat.com [10.5.11.16]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 4FA898015CE; Sun, 9 Aug 2020 06:30:59 +0000 (UTC) Received: from localhost.localdomain.com (ovpn-12-33.pek2.redhat.com [10.72.12.33]) by smtp.corp.redhat.com (Postfix) with ESMTP id C507E1C4; Sun, 9 Aug 2020 06:30:57 +0000 (UTC) From: Zorro Lang To: fstests@vger.kernel.org Cc: io-uring@vger.kernel.org, jmoyer@redhat.com Subject: [PATCH v2 4/4] fsx: add IO_URING test Date: Sun, 9 Aug 2020 14:30:40 +0800 Message-Id: <20200809063040.15521-5-zlang@redhat.com> In-Reply-To: <20200809063040.15521-1-zlang@redhat.com> References: <20200809063040.15521-1-zlang@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.16 Sender: fstests-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: fstests@vger.kernel.org New IO_URING test for fsx, use -U option to enable IO_URING test. Signed-off-by: Zorro Lang --- ltp/fsx.c | 158 +++++++++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 144 insertions(+), 14 deletions(-) diff --git a/ltp/fsx.c b/ltp/fsx.c index 7c76655a..7475bf9d 100644 --- a/ltp/fsx.c +++ b/ltp/fsx.c @@ -34,6 +34,9 @@ #ifdef AIO #include #endif +#ifdef URING +#include +#endif #include #ifndef MAP_FILE @@ -176,21 +179,17 @@ int integrity = 0; /* -i flag */ int fsxgoodfd = 0; int o_direct; /* -Z */ int aio = 0; +int uring = 0; int mark_nr = 0; int page_size; int page_mask; int mmap_mask; -#ifdef AIO -int aio_rw(int rw, int fd, char *buf, unsigned len, unsigned offset); +int fsx_rw(int rw, int fd, char *buf, unsigned len, unsigned offset); #define READ 0 #define WRITE 1 -#define fsxread(a,b,c,d) aio_rw(READ, a,b,c,d) -#define fsxwrite(a,b,c,d) aio_rw(WRITE, a,b,c,d) -#else -#define fsxread(a,b,c,d) read(a,b,c) -#define fsxwrite(a,b,c,d) write(a,b,c) -#endif +#define fsxread(a,b,c,d) fsx_rw(READ, a,b,c,d) +#define fsxwrite(a,b,c,d) fsx_rw(WRITE, a,b,c,d) const char *replayops = NULL; const char *recordops = NULL; @@ -2242,7 +2241,7 @@ void usage(void) { fprintf(stdout, "usage: %s", - "fsx [-dknqxABEFJLOWZ] [-b opnum] [-c Prob] [-g filldata] [-i logdev] [-j logid] [-l flen] [-m start:end] [-o oplen] [-p progressinterval] [-r readbdy] [-s style] [-t truncbdy] [-w writebdy] [-D startingop] [-N numops] [-P dirpath] [-S seed] fname\n\ + "fsx [-dknqxBEFJLOWZ][-A|-U] [-b opnum] [-c Prob] [-g filldata] [-i logdev] [-j logid] [-l flen] [-m start:end] [-o oplen] [-p progressinterval] [-r readbdy] [-s style] [-t truncbdy] [-w writebdy] [-D startingop] [-N numops] [-P dirpath] [-S seed] fname\n\ -b opnum: beginning operation number (default 1)\n\ -c P: 1 in P chance of file close+open at each op (default infinity)\n\ -d: debug output for all operations\n\ @@ -2265,7 +2264,10 @@ usage(void) -y synchronize changes to a file\n" #ifdef AIO -" -A: Use the AIO system calls\n" +" -A: Use the AIO system calls, -A excludes -U\n" +#endif +#ifdef URING +" -U: Use the IO_URING system calls, -U excludes -A\n" #endif " -D startingop: debug output starting at specified operation\n" #ifdef HAVE_LINUX_FALLOC_H @@ -2425,13 +2427,131 @@ out_error: errno = -ret; return -1; } +#endif + +#ifdef URING +struct io_uring ring; +#define URING_ENTRIES 1024 +int +uring_setup() +{ + int ret; + + ret = io_uring_queue_init(URING_ENTRIES, &ring, 0); + if (ret != 0) { + fprintf(stderr, "uring_setup: io_uring_queue_init failed: %s\n", + strerror(ret)); + return -1; + } + return 0; +} -int aio_rw(int rw, int fd, char *buf, unsigned len, unsigned offset) +int +__uring_rw(int rw, int fd, char *buf, unsigned len, unsigned offset) { + struct io_uring_sqe *sqe; + struct io_uring_cqe *cqe; + struct iovec iovec; int ret; + int res, res2 = 0; + char *p = buf; + unsigned l = len; + unsigned o = offset; + + + /* + * Due to io_uring tries non-blocking IOs (especially read), that + * always cause 'normal' short reading. To avoid this short read + * fail, try to loop read/write (escpecilly read) data. + */ + uring_loop: + sqe = io_uring_get_sqe(&ring); + if (!sqe) { + fprintf(stderr, "uring_rw: io_uring_get_sqe failed: %s\n", + strerror(errno)); + return -1; + } + + iovec.iov_base = p; + iovec.iov_len = l; + if (rw == READ) { + io_uring_prep_readv(sqe, fd, &iovec, 1, o); + } else { + io_uring_prep_writev(sqe, fd, &iovec, 1, o); + } + + ret = io_uring_submit(&ring); + if (ret != 1) { + fprintf(stderr, "errcode=%d\n", -ret); + fprintf(stderr, "uring %s: io_uring_submit failed: %s\n", + rw == READ ? "read":"write", strerror(-ret)); + goto uring_error; + } + + ret = io_uring_wait_cqe(&ring, &cqe); + if (ret < 0) { + if (ret == 0) + fprintf(stderr, "uring %s: no events available\n", + rw == READ ? "read":"write"); + else { + fprintf(stderr, "errcode=%d\n", -ret); + fprintf(stderr, "uring %s: io_uring_wait_cqe failed: %s\n", + rw == READ ? "read":"write", strerror(-ret)); + } + goto uring_error; + } + res = cqe->res; + io_uring_cqe_seen(&ring, cqe); + + res2 += res; + if (len != res2) { + if (res > 0) { + o += res; + l -= res; + p += res; + if (l > 0) + goto uring_loop; + } else if (res < 0) { + ret = res; + fprintf(stderr, "errcode=%d\n", -ret); + fprintf(stderr, "uring %s: io_uring failed: %s\n", + rw == READ ? "read":"write", strerror(-ret)); + goto uring_error; + } else { + fprintf(stderr, "uring %s bad io length: %d instead of %u\n", + rw == READ ? "read":"write", res2, len); + } + } + return res2; + + uring_error: + /* + * The caller expects error return in traditional libc + * convention, i.e. -1 and the errno set to error. + */ + errno = -ret; + return -1; +} +#endif + +int fsx_rw(int rw, int fd, char *buf, unsigned len, unsigned offset) +{ + int ret = -1; if (aio) { +#ifdef AIO ret = __aio_rw(rw, fd, buf, len, offset); +#elif + fprintf(stderr, "io_rw: need AIO support!\n"); + exit(111); +#endif + } else if (uring) { +#ifdef URING + ret = __uring_rw(rw, fd, buf, len, offset); +#elif + fprintf(stderr, "io_rw: need IO_URING support!\n"); + exit(111); +#endif } else { if (rw == READ) ret = read(fd, buf, len); @@ -2441,8 +2561,6 @@ int aio_rw(int rw, int fd, char *buf, unsigned len, unsigned offset) return ret; } -#endif - #define test_fallocate(mode) __test_fallocate(mode, #mode) int @@ -2496,7 +2614,7 @@ main(int argc, char **argv) setvbuf(stdout, (char *)0, _IOLBF, 0); /* line buffered stdout */ while ((ch = getopt_long(argc, argv, - "b:c:dfg:i:j:kl:m:no:p:qr:s:t:w:xyABD:EFJKHzCILN:OP:RS:WXZ", + "b:c:dfg:i:j:kl:m:no:p:qr:s:t:w:xyABD:EFJKHzCILN:OP:RS:UWXZ", longopts, NULL)) != EOF) switch (ch) { case 'b': @@ -2604,6 +2722,9 @@ main(int argc, char **argv) case 'A': aio = 1; break; + case 'U': + uring = 1; + break; case 'D': debugstart = getnum(optarg, &endp); if (debugstart < 1) @@ -2694,6 +2815,11 @@ main(int argc, char **argv) if (argc != 1) usage(); + if (aio && uring) { + fprintf(stderr, "-A and -U shouldn't be used together\n"); + usage(); + } + if (integrity && !dirpath) { fprintf(stderr, "option -i requires -P \n"); usage(); @@ -2784,6 +2910,10 @@ main(int argc, char **argv) if (aio) aio_setup(); #endif +#ifdef URING + if (uring) + uring_setup(); +#endif if (!(o_flags & O_TRUNC)) { off_t ret;