From patchwork Mon Aug 8 02:01:01 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Ashish Mittal X-Patchwork-Id: 9266611 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id DD1416048F for ; Mon, 8 Aug 2016 02:02:26 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id C818027165 for ; Mon, 8 Aug 2016 02:02:26 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id BC85527D5D; Mon, 8 Aug 2016 02:02:26 +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=-6.8 required=2.0 tests=BAYES_00, DKIM_ADSP_CUSTOM_MED, DKIM_SIGNED, FREEMAIL_FROM, RCVD_IN_DNSWL_HI, T_DKIM_INVALID 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 EE48F27165 for ; Mon, 8 Aug 2016 02:02:23 +0000 (UTC) Received: from localhost ([::1]:54488 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bWZtW-0002zP-8I for patchwork-qemu-devel@patchwork.kernel.org; Sun, 07 Aug 2016 22:02:22 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:47755) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bWZt8-0002z6-62 for qemu-devel@nongnu.org; Sun, 07 Aug 2016 22:02:03 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bWZt1-0005R3-RD for qemu-devel@nongnu.org; Sun, 07 Aug 2016 22:01:57 -0400 Received: from mail-pf0-x241.google.com ([2607:f8b0:400e:c00::241]:34376) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bWZt1-0005Qw-29 for qemu-devel@nongnu.org; Sun, 07 Aug 2016 22:01:51 -0400 Received: by mail-pf0-x241.google.com with SMTP id g202so24094233pfb.1 for ; Sun, 07 Aug 2016 19:01:50 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=from:to:cc:subject:date:message-id; bh=i0B2FATO7uOmZ2/Mb4WzIJtF//P0jpRpvJSvia3WXcs=; b=mOXoIfXrJE6vwPQJZbsUjLt9KWydf3HSxjDp7u7ud4JBEKi34MgC2HqDah/hfO0M2a IUNiUNeHhS2ix/8QRDVjipSv2gHr2dUl2T0RtZOgBMCKpg4XT5bsvJLxHPHak05cVw/l zBj//WghunICV4a0zSvSofQRaPK/0IULLLJdmkCnK8KlD6Kbw3c9lvHXpSEeAIo/m7VN 8GHEysAyBb869qEh/2zea1uqwVh7YEu7F4avJ9EujMwjYTJ9A/QHc2a6O0euo1Zy2zjd 9gH4sJjbIKS+6B63zfZTMW3Iyv5IIO1nOH6zNrF7ZuVlINz2kXYtG9/1+XRDcfz73Wrr erDA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:from:to:cc:subject:date:message-id; bh=i0B2FATO7uOmZ2/Mb4WzIJtF//P0jpRpvJSvia3WXcs=; b=bvSwKfBM0KHPmwUWk/0mfkw9hOrgdt74rR5DiH/9EU8Md6f0HNX7wekQs5mRH+9ZDV ImK+Vl3iHq0S/kjQWtTt9O1k/m4Mti89UrR4EMl00+O6pVmbJOkGTu3HTt2BzB4kYIyM oh6ld40yQDWSAALqCVZIHZzNi8/o5W894rqTB/5w7KV7J9PrY7Eq29Tk085sR7f7fQkB so/6qOModI4TKS2Z64qTb3eEOyzfBGP94Jh7kwxJWOgzyRAAXWhnwregotRukbHUL3Yr rD9kXgiqnvfof5o8RJzJMYMh+DiQRFNWm7fQfPuvErgR/04YZCXRaU7S9k9T9nSdXlbO 0A+Q== X-Gm-Message-State: AEkoouuYjqDQvo/iKZYRwUecvtUP9OdVsKTDvukGEnE0fjROj34GDaXNALydOqruaFbhNQ== X-Received: by 10.98.158.78 with SMTP id s75mr158225354pfd.137.1470621708873; Sun, 07 Aug 2016 19:01:48 -0700 (PDT) Received: from localhost.localdomain.localdomain (c-71-202-29-219.hsd1.ca.comcast.net. [71.202.29.219]) by smtp.gmail.com with ESMTPSA id v26sm43191394pfi.41.2016.08.07.19.01.47 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Sun, 07 Aug 2016 19:01:48 -0700 (PDT) From: Ashish Mittal X-Google-Original-From: Ashish Mittal To: qemu-devel@nongnu.org, pbonzini@redhat.com, kwolf@redhat.com, armbru@redhat.com, berrange@redhat.com, ashish.mittal@veritas.com, stefanha@gmail.com Date: Sun, 7 Aug 2016 19:01:01 -0700 Message-Id: <1470621661-21877-1-git-send-email-ashish.mittal@veritas.com> X-Mailer: git-send-email 2.5.5 X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] X-Received-From: 2607:f8b0:400e:c00::241 Subject: [Qemu-devel] [PATCH v2 RFC] block/vxhs: Initial commit to add Veritas HyperScale VxHS block device support 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: Ketan.Nilangekar@veritas.com, Abhijit.Dey@veritas.com Errors-To: qemu-devel-bounces+patchwork-qemu-devel=patchwork.kernel.org@nongnu.org Sender: "Qemu-devel" X-Virus-Scanned: ClamAV using ClamSMTP This patch adds support for a new block device type called "vxhs". Source code for the library that this code loads can be downloaded from: https://github.com/MittalAshish/libqnio.git Version 2 patch submission fixes the following issues: (1) Removed code to dlopen library. We now check if libqnio is installed during configure, and directly link with it. (2) Changed file headers to mention GPLv2-or-later license. (3) Removed unnecessary type casts and inlines. (4) Removed custom tokenize function and modified code to use g_strsplit. (5) Replaced malloc/free with g_new/g_free and removed code that checks for memory allocation failure conditions. (6) Removed some block ops implementations that were place-holders only. (7) Removed all custom debug messages. Added new messages in block/trace-events (8) Other miscellaneous corrections. TODO: QAPI changes and (fixes to) review comments from Stefan. Signed-off-by: Ashish Mittal --- block/Makefile.objs | 2 + block/trace-events | 40 ++ block/vxhs.c | 1199 +++++++++++++++++++++++++++++++++++++++++++++++++++ block/vxhs.h | 294 +++++++++++++ configure | 50 +++ 5 files changed, 1585 insertions(+) create mode 100644 block/vxhs.c create mode 100644 block/vxhs.h diff --git a/block/Makefile.objs b/block/Makefile.objs index 2593a2f..5f83305 100644 --- a/block/Makefile.objs +++ b/block/Makefile.objs @@ -20,6 +20,7 @@ block-obj-$(CONFIG_RBD) += rbd.o block-obj-$(CONFIG_GLUSTERFS) += gluster.o block-obj-$(CONFIG_ARCHIPELAGO) += archipelago.o block-obj-$(CONFIG_LIBSSH2) += ssh.o +block-obj-$(CONFIG_VXHS) += vxhs.o block-obj-y += accounting.o dirty-bitmap.o block-obj-y += write-threshold.o @@ -43,3 +44,4 @@ block-obj-m += dmg.o dmg.o-libs := $(BZIP2_LIBS) qcow.o-libs := -lz linux-aio.o-libs := -laio +vxhs.o-libs := $(VXHS_LIBS) diff --git a/block/trace-events b/block/trace-events index 978ef4f..8f02725 100644 --- a/block/trace-events +++ b/block/trace-events @@ -114,3 +114,43 @@ qed_aio_write_data(void *s, void *acb, int ret, uint64_t offset, size_t len) "s qed_aio_write_prefill(void *s, void *acb, uint64_t start, size_t len, uint64_t offset) "s %p acb %p start %"PRIu64" len %zu offset %"PRIu64 qed_aio_write_postfill(void *s, void *acb, uint64_t start, size_t len, uint64_t offset) "s %p acb %p start %"PRIu64" len %zu offset %"PRIu64 qed_aio_write_main(void *s, void *acb, int ret, uint64_t offset, size_t len) "s %p acb %p ret %d offset %"PRIu64" len %zu" + +# block/vxhs.c +vxhs_bdrv_init(int version) "Registering VxHS %d AIO driver" +vxhs_iio_callback(int error, int reason) "ctx is NULL: error %d, reason %d" +vxhs_setup_qnio(void *s) "Context to HyperScale IO manager = %p" +vxhs_setup_qnio_nwerror(char c) "Could not initialize the network channel. Bailing out%c" +vxhs_open_device_cmdline(const char *vxhs_uri) "Original command line : %s" +vxhs_iio_callback_iofail(int err, int reason, void *acb, int seg) "Read/Write failed: error %d, reason %d, acb %p, segment %d" +vxhs_iio_callback_retry(char *guid, void *acb) "vDisk %s, added acb %p to retry queue (5)" +vxhs_iio_callback_chnlfail(int error) "QNIO channel failed, no i/o (%d)" +vxhs_iio_callback_fail(int r, void *acb, int seg, uint64_t size, int err) " ALERT: reason = %d , acb = %p, acb->segments = %d, acb->size = %lu Error = %d" +vxhs_fail_aio(char * guid, void *acb) "vDisk %s, failing acb %p" +vxhs_iio_callback_ready(char *vd, int err) "async vxhs_iio_callback: IRP_VDISK_CHECK_IO_FAILOVER_READY completed for vdisk %s with error %d" +vxhs_iio_callback_chnfail(int err, int error) "QNIO channel failed, no i/o %d, %d" +vxhs_iio_callback_unknwn(int opcode, int err) "unexpected opcode %d, errno %d" +vxhs_open_device_qnio(char *of_vsa_addr) "Could not open an QNIO connection to: %s" +vxhs_create(const char *filename) "vxhs_create: came in to open file = %s" +vxhs_open(const char *filename) "vxhs_vxhs_open: came in to open file = %s" +vxhs_open_fail(int ret) "Could not open the device. Error = %d" +vxhs_open_epipe(char c) "Could not create a pipe for device. bailing out%c" +vxhs_aio_rw(char *guid, int iodir, uint64_t size, uint64_t offset) "vDisk %s, vDisk device is in failed state iodir = %d size = %lu offset = %lu" +vxhs_aio_rw_retry(char *guid, void *acb, int queue) "vDisk %s, added acb %p to retry queue(%d)" +vxhs_aio_rw_invalid(int req) "Invalid I/O request iodir %d" +vxhs_aio_rw_ioerr(char *guid, int iodir, uint64_t size, uint64_t off, void *acb, int seg, int ret, int err) "IO ERROR (vDisk %s) FOR : Read/Write = %d size = %lu offset = %lu ACB = %p Segments = %d. Error = %d, errno = %d" +vxhs_co_flush(char *guid, int ret, int err) "vDisk (%s) Flush ioctl failed ret = %d errno = %d" +vxhs_co_flush_iocnt(int iocnt) "In the flush the IO count = %d" +vxhs_get_vdisk_stat(char *guid, int ret, int err) "vDisk (%s) stat ioctl failed, ret = %d, errno = %d" +vxhs_switch_storage_agent(char *ip, char *guid) "Query host %s for vdisk %s" +vxhs_switch_storage_agent_failed(char *ip, char *guid, int res, int err) "Query to host %s for vdisk %s failed, res = %d, errno = %d" +vxhs_failover_ioctl_cb(char *ip, char *guid) "Switched to storage server host-IP %s for vdisk %s" +vxhs_failover_ioctl_cb_retry(char *guid) "failover_ioctl_cb: keep looking for io target for vdisk %s" +vxhs_failover_io(char *vdisk) "I/O Failover starting for vDisk %s" +vxhs_reopen_vdisk(char *ip) "Failed to connect to storage agent on host-ip %s" +vxhs_reopen_vdisk_openfail(char *fname) "Failed to open vdisk device: %s" +vxhs_build_io_target_list(int num, char *uri) "About to parse URI %d : %s" +vxhs_handle_queued_ios(void *acb, int res) "Restarted acb %p res %d" +vxhs_restart_aio(int dir, int res, int err) "IO ERROR FOR: Read/Write = %d Error = %d, errno = %d" +vxhs_complete_aio(void *acb, uint64_t ret) "aio failed acb %p ret %ld" +vxhs_aio_rw_iofail(char *guid) "vDisk %s, I/O operation failed." +vxhs_aio_rw_devfail(char *guid, int dir, uint64_t size, uint64_t off) "vDisk %s, vDisk device failed iodir = %d size = %lu offset = %lu" diff --git a/block/vxhs.c b/block/vxhs.c new file mode 100644 index 0000000..9a82757 --- /dev/null +++ b/block/vxhs.c @@ -0,0 +1,1199 @@ +/* + * QEMU Block driver for Veritas HyperScale (VxHS) + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + * + */ + +#include "vxhs.h" +#include +#include "trace.h" + +/* qnio client ioapi_ctx */ +static void __attribute__((unused)) *global_qnio_ctx; + +/* insure init once */ +static pthread_mutex_t __attribute__((unused)) of_global_ctx_lock = + PTHREAD_MUTEX_INITIALIZER; + +/* HyperScale Driver Version */ +static const int __attribute__((unused)) vxhs_drv_version = 8895; + +/* vdisk prefix to pass to qnio */ +static const char vdisk_prefix[] = "/dev/of/vdisk"; + +void vxhs_inc_acb_segment_count(void *ptr, int count) +{ + VXHSAIOCB *acb = ptr; + BDRVVXHSState *s = acb->common.bs->opaque; + + VXHS_SPIN_LOCK(s->vdisk_acb_lock); + acb->segments += count; + VXHS_SPIN_UNLOCK(s->vdisk_acb_lock); +} + +void vxhs_dec_acb_segment_count(void *ptr, int count) +{ + VXHSAIOCB *acb = ptr; + BDRVVXHSState *s = acb->common.bs->opaque; + + VXHS_SPIN_LOCK(s->vdisk_acb_lock); + acb->segments -= count; + VXHS_SPIN_UNLOCK(s->vdisk_acb_lock); +} + +int vxhs_dec_and_get_acb_segment_count(void *ptr, int count) +{ + VXHSAIOCB *acb = ptr; + BDRVVXHSState *s = acb->common.bs->opaque; + int segcount = 0; + + + VXHS_SPIN_LOCK(s->vdisk_acb_lock); + acb->segments -= count; + segcount = acb->segments; + VXHS_SPIN_UNLOCK(s->vdisk_acb_lock); + + return segcount; +} + +void vxhs_set_acb_buffer(void *ptr, void *buffer) +{ + VXHSAIOCB *acb = ptr; + + acb->buffer = buffer; +} + +void vxhs_inc_vdisk_iocount(void *ptr, uint32_t count) +{ + BDRVVXHSState *s = ptr; + + VXHS_SPIN_LOCK(s->vdisk_lock); + s->vdisk_aio_count += count; + VXHS_SPIN_UNLOCK(s->vdisk_lock); +} + +void vxhs_dec_vdisk_iocount(void *ptr, uint32_t count) +{ + BDRVVXHSState *s = ptr; + + VXHS_SPIN_LOCK(s->vdisk_lock); + s->vdisk_aio_count -= count; + VXHS_SPIN_UNLOCK(s->vdisk_lock); +} + +uint32_t vxhs_get_vdisk_iocount(void *ptr) +{ + BDRVVXHSState *s = ptr; + uint32_t count = 0; + + VXHS_SPIN_LOCK(s->vdisk_lock); + count = s->vdisk_aio_count; + VXHS_SPIN_UNLOCK(s->vdisk_lock); + + return count; +} + +void vxhs_iio_callback(uint32_t rfd, uint32_t reason, void *ctx, void *m) +{ + VXHSAIOCB *acb = NULL; + BDRVVXHSState *s = NULL; + int rv = 0; + int segcount = 0; + uint32_t error = 0; + uint32_t opcode = 0; + + assert(m); + if (m) { + /* TODO: need common get message attrs, not two separate lib calls */ + error = qemu_iio_extract_msg_error(m); + opcode = qemu_iio_extract_msg_opcode(m); + } + switch (opcode) { + case IRP_READ_REQUEST: + case IRP_WRITE_REQUEST: + + /* + * ctx is VXHSAIOCB* + * ctx is NULL if error is VXERROR_CHANNEL_HUP or reason is IIO_REASON_HUP + */ + if (ctx) { + acb = ctx; + s = acb->common.bs->opaque; + } else { + trace_vxhs_iio_callback(error, reason); + goto out; + } + + if (error) { + trace_vxhs_iio_callback_iofail(error, reason, acb, acb->segments); + + if (reason == IIO_REASON_DONE || reason == IIO_REASON_EVENT) { + /* + * Storage agent failed while I/O was in progress + * Fail over only if the qnio channel dropped, indicating + * storage agent failure. Don't fail over in response to other + * I/O errors such as disk failure. + */ + if (error == VXERROR_RETRY_ON_SOURCE || error == VXERROR_HUP || + error == VXERROR_CHANNEL_HUP || error == -1) { + /* + * Start vDisk IO failover once callback is + * called against all the pending IOs. + * If vDisk has no redundency enabled + * then IO failover routine will mark + * the vDisk failed and fail all the + * AIOs without retry (stateless vDisk) + */ + VXHS_SPIN_LOCK(s->vdisk_lock); + if (!OF_VDISK_IOFAILOVER_IN_PROGRESS(s)) { + OF_VDISK_SET_IOFAILOVER_IN_PROGRESS(s); + } + /* + * Check if this acb is already queued before. + * It is possible in case if I/Os are submitted + * in multiple segments (QNIO_MAX_IO_SIZE). + */ + VXHS_SPIN_LOCK(s->vdisk_acb_lock); + if (!OF_AIOCB_FLAGS_QUEUED(acb)) { + QSIMPLEQ_INSERT_TAIL(&s->vdisk_aio_retryq, + acb, retry_entry); + OF_AIOCB_FLAGS_SET_QUEUED(acb); + s->vdisk_aio_retry_qd++; + trace_vxhs_iio_callback_retry(s->vdisk_guid, acb); + } + segcount = --acb->segments; + VXHS_SPIN_UNLOCK(s->vdisk_acb_lock); + /* + * Decrement AIO count only when callback is called + * against all the segments of aiocb. + */ + if (segcount == 0 && --s->vdisk_aio_count == 0) { + /* + * Start vDisk I/O failover + */ + VXHS_SPIN_UNLOCK(s->vdisk_lock); + /* + * TODO: + * Need to explore further if it is possible to optimize + * the failover operation on Virtual-Machine (global) + * specific rather vDisk specific. + */ + vxhs_failover_io(s); + goto out; + } + VXHS_SPIN_UNLOCK(s->vdisk_lock); + goto out; + } + } else if (reason == IIO_REASON_HUP) { + /* + * Channel failed, spontaneous notification, + * not in response to I/O + */ + trace_vxhs_iio_callback_chnlfail(error); + /* + * TODO: Start channel failover when no I/O is outstanding + */ + goto out; + } else { + trace_vxhs_iio_callback_fail(reason, acb, acb->segments, acb->size, error); + } + } + /* + * Set error into acb if not set. In case if acb is being + * submitted in multiple segments then need to set the error + * only once. + * + * Once acb done callback is called for the last segment + * then acb->ret return status will be sent back to the + * caller. + */ + VXHS_SPIN_LOCK(s->vdisk_acb_lock); + if (error && !acb->ret) { + acb->ret = error; + } + --acb->segments; + segcount = acb->segments; + assert(segcount >= 0); + VXHS_SPIN_UNLOCK(s->vdisk_acb_lock); + /* + * Check if all the outstanding I/Os are done against acb. + * If yes then send signal for AIO completion. + */ + if (segcount == 0) { + rv = qemu_write_full(s->fds[VDISK_FD_WRITE], &acb, sizeof(acb)); + if (rv != sizeof(acb)) { + error_report("VXHS AIO completion failed: %s", strerror(errno)); + abort(); + } + } + break; + + case IRP_VDISK_CHECK_IO_FAILOVER_READY: + /* ctx is BDRVVXHSState* */ + assert(ctx); + trace_vxhs_iio_callback_ready(((BDRVVXHSState *)ctx)->vdisk_guid, error); + vxhs_failover_ioctl_cb(error, ctx); + break; + + default: + if (reason == IIO_REASON_HUP) { + /* + * Channel failed, spontaneous notification, + * not in response to I/O + */ + trace_vxhs_iio_callback_chnfail(error, errno); + /* + * TODO: Start channel failover when no I/O is outstanding + */ + } else { + trace_vxhs_iio_callback_unknwn(opcode, error); + } + break; + } +out: + return; +} + +void vxhs_complete_aio(VXHSAIOCB *acb, BDRVVXHSState *s) +{ + BlockCompletionFunc *cb = acb->common.cb; + void *opaque = acb->common.opaque; + int ret = 0; + + if (acb->ret != 0) { + trace_vxhs_complete_aio(acb, acb->ret); + /* + * We mask all the IO errors generically as EIO for upper layers + * Right now our IO Manager uses non standard error codes. Instead + * of confusing upper layers with incorrect interpretation we are + * doing this workaround. + */ + ret = (-EIO); + } + /* + * Copy back contents from stablization buffer into original iovector + * before returning the IO + */ + if (acb->buffer != NULL) { + qemu_iovec_from_buf(acb->qiov, 0, acb->buffer, acb->qiov->size); + free(acb->buffer); + acb->buffer = NULL; + } + vxhs_dec_vdisk_iocount(s, 1); + acb->aio_done = VXHS_IO_COMPLETED; + qemu_aio_unref(acb); + cb(opaque, ret); +} + +/* + * This is the HyperScale event handler registered to QEMU. + * It is invoked when any IO gets completed and written on pipe + * by callback called from QNIO thread context. Then it marks + * the AIO as completed, and releases HyperScale AIO callbacks. + */ +void vxhs_aio_event_reader(void *opaque) +{ + BDRVVXHSState *s = opaque; + ssize_t ret; + + do { + char *p = (char *)&s->qnio_event_acb; + + ret = read(s->fds[VDISK_FD_READ], p + s->event_reader_pos, + sizeof(s->qnio_event_acb) - s->event_reader_pos); + if (ret > 0) { + s->event_reader_pos += ret; + if (s->event_reader_pos == sizeof(s->qnio_event_acb)) { + s->event_reader_pos = 0; + vxhs_complete_aio(s->qnio_event_acb, s); + } + } + } while (ret < 0 && errno == EINTR); +} + +/* + * QEMU calls this to check if there are any pending IO on vDisk. + * It will wait in a loop until all the AIOs are completed. + */ +int vxhs_aio_flush_cb(void *opaque) +{ + BDRVVXHSState *s = opaque; + + return vxhs_get_vdisk_iocount(s); +} + +/* + * This will be called by QEMU while booting for each vDisk. + * bs->opaque will be allocated by QEMU upper block layer before + * calling open. It will load all the QNIO operations from + * qemuqnio library and call QNIO operation to create channel to + * do IO on vDisk. It parses the URI, gets the hostname, vDisk + * path and then sets HyperScale event handler to QEMU. + */ +void *vxhs_setup_qnio(void) +{ + void *qnio_ctx = NULL; + + qnio_ctx = qemu_iio_init(vxhs_iio_callback); + + if (qnio_ctx != NULL) { + trace_vxhs_setup_qnio(qnio_ctx); + } else { + trace_vxhs_setup_qnio_nwerror('.'); + } + + return qnio_ctx; +} + +int vxhs_open_device(const char *vxhs_uri, int *cfd, int *rfd, + BDRVVXHSState *s) +{ + char *file_name; + char *of_vsa_addr; + int ret = 0; + gchar **target_list; + + pthread_mutex_lock(&of_global_ctx_lock); + if (global_qnio_ctx == NULL) { + global_qnio_ctx = vxhs_setup_qnio(); + if (global_qnio_ctx == NULL) { + pthread_mutex_unlock(&of_global_ctx_lock); + return -1; + } + } + pthread_mutex_unlock(&of_global_ctx_lock); + + *cfd = -1; + + of_vsa_addr = g_new0(char, OF_MAX_SERVER_ADDR); + file_name = g_new0(char, OF_MAX_FILE_LEN); + + /* + * The steps below need to done by all the block drivers in QEMU which + * support AIO. Need to create pipe for communicating b/w two threads + * in different context. And set handler for read event when IO completion + * is reported by non-QEMU context. + */ + trace_vxhs_open_device_cmdline(vxhs_uri); + target_list = g_strsplit(vxhs_uri, "%7D", 0); + assert(target_list != NULL && target_list[0] != NULL); + vxhs_build_io_target_list(s, target_list); + + snprintf(file_name, OF_MAX_FILE_LEN, "%s%s", vdisk_prefix, s->vdisk_guid); + snprintf(of_vsa_addr, OF_MAX_SERVER_ADDR, "of://%s:%d", + s->vdisk_hostinfo[s->vdisk_cur_host_idx].hostip, + s->vdisk_hostinfo[s->vdisk_cur_host_idx].port); + + *cfd = qemu_open_iio_conn(global_qnio_ctx, of_vsa_addr, 0); + if (*cfd < 0) { + trace_vxhs_open_device_qnio(of_vsa_addr); + ret = -EIO; + goto out; + } + *rfd = qemu_iio_devopen(global_qnio_ctx, *cfd, file_name, 0); + s->aio_context = qemu_get_aio_context(); + +out: + /* uri is still in use, cleaned up in close */ + if (file_name != NULL) { + g_free(file_name); + } + if (of_vsa_addr != NULL) { + g_free(of_vsa_addr); + } + g_strfreev(target_list); + return ret; +} + +int vxhs_create(const char *filename, + QemuOpts *options, Error **errp) +{ + int ret = 0; + int qemu_cfd = 0; + int qemu_rfd = 0; + BDRVVXHSState s; + + trace_vxhs_create(filename); + ret = vxhs_open_device(filename, &qemu_cfd, &qemu_rfd, &s); + + return ret; +} + +static QemuOptsList runtime_opts = { + .name = "vxhs", + .head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head), + .desc = { + { + .name = "filename", + .type = QEMU_OPT_STRING, + .help = "URI to the Veritas HyperScale image", + }, + { /* end of list */ } + }, +}; + +int vxhs_open(BlockDriverState *bs, QDict *options, + int bdrv_flags, Error **errp) +{ + BDRVVXHSState *s = bs->opaque; + int ret = 0; + int qemu_qnio_cfd = 0; + int qemu_rfd = 0; + QemuOpts *opts; + Error *local_err = NULL; + const char *vxhs_uri; + + opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort); + qemu_opts_absorb_qdict(opts, options, &local_err); + if (local_err) { + error_propagate(errp, local_err); + ret = -EINVAL; + goto out; + } + + vxhs_uri = qemu_opt_get(opts, "filename"); + + memset(s, 0, sizeof(*s)); + trace_vxhs_open(vxhs_uri); + ret = vxhs_open_device(vxhs_uri, &qemu_qnio_cfd, &qemu_rfd, s); + if (ret != 0) { + trace_vxhs_open_fail(ret); + return ret; + } + s->qnio_ctx = global_qnio_ctx; + s->vdisk_hostinfo[0].qnio_cfd = qemu_qnio_cfd; + s->vdisk_hostinfo[0].vdisk_rfd = qemu_rfd; + s->vdisk_size = 0; + QSIMPLEQ_INIT(&s->vdisk_aio_retryq); + + ret = qemu_pipe(s->fds); + if (ret < 0) { + trace_vxhs_open_epipe('.'); + ret = -errno; + goto out; + } + fcntl(s->fds[VDISK_FD_READ], F_SETFL, O_NONBLOCK); + + aio_set_fd_handler(s->aio_context, s->fds[VDISK_FD_READ], + false, vxhs_aio_event_reader, NULL, s); + + /* + * Allocate/Initialize the spin-locks. + * + * NOTE: + * Since spin lock is being allocated + * dynamically hence moving acb struct + * specific lock to BDRVVXHSState + * struct. The reason being, + * we don't want the overhead of spin + * lock being dynamically allocated and + * freed for every AIO. + */ + s->vdisk_lock = VXHS_SPIN_LOCK_ALLOC; + s->vdisk_acb_lock = VXHS_SPIN_LOCK_ALLOC; + + return 0; + +out: + if (s->vdisk_hostinfo[0].vdisk_rfd >= 0) { + qemu_iio_devclose(s->qnio_ctx, 0, + s->vdisk_hostinfo[0].vdisk_rfd); + } + /* never close qnio_cfd */ + trace_vxhs_open_fail(ret); + return ret; +} + +static const AIOCBInfo vxhs_aiocb_info = { + .aiocb_size = sizeof(VXHSAIOCB) +}; + +/* + * This is called in QNIO thread context when IO done + * on IO Manager and QNIO client received the data or + * ACK. It notify another event handler thread running in QEMU context + * by writing on the pipe + */ +void vxhs_finish_aiocb(ssize_t ret, void *arg) +{ + VXHSAIOCB *acb = arg; + BlockDriverState *bs = acb->common.bs; + BDRVVXHSState *s = bs->opaque; + int rv; + + acb->ret = ret; + rv = qemu_write_full(s->fds[VDISK_FD_WRITE], &acb, sizeof(acb)); + if (rv != sizeof(acb)) { + error_report("VXHS AIO completion failed: %s", + strerror(errno)); + abort(); + } +} + +/* + * This allocates QEMU-VXHS callback for each IO + * and is passed to QNIO. When QNIO completes the work, + * it will be passed back through the callback. + */ +BlockAIOCB *vxhs_aio_rw(BlockDriverState *bs, + int64_t sector_num, QEMUIOVector *qiov, + int nb_sectors, + BlockCompletionFunc *cb, + void *opaque, int iodir) +{ + VXHSAIOCB *acb = NULL; + BDRVVXHSState *s = bs->opaque; + size_t size; + uint64_t offset; + int iio_flags = 0; + int ret = 0; + + offset = sector_num * BDRV_SECTOR_SIZE; + size = nb_sectors * BDRV_SECTOR_SIZE; + + acb = qemu_aio_get(&vxhs_aiocb_info, bs, cb, opaque); + /* + * Setup or initialize VXHSAIOCB. + * Every single field should be initialized since + * acb will be picked up from the slab without + * initializing with zero. + */ + acb->io_offset = offset; + acb->size = size; + acb->ret = 0; + acb->flags = 0; + acb->aio_done = VXHS_IO_INPROGRESS; + acb->segments = 0; + acb->buffer = 0; + acb->qiov = qiov; + acb->direction = iodir; + + VXHS_SPIN_LOCK(s->vdisk_lock); + if (OF_VDISK_FAILED(s)) { + trace_vxhs_aio_rw(s->vdisk_guid, iodir, size, offset); + VXHS_SPIN_UNLOCK(s->vdisk_lock); + goto errout; + } + if (OF_VDISK_IOFAILOVER_IN_PROGRESS(s)) { + QSIMPLEQ_INSERT_TAIL(&s->vdisk_aio_retryq, acb, retry_entry); + s->vdisk_aio_retry_qd++; + OF_AIOCB_FLAGS_SET_QUEUED(acb); + VXHS_SPIN_UNLOCK(s->vdisk_lock); + trace_vxhs_aio_rw_retry(s->vdisk_guid, acb, 1); + goto out; + } + s->vdisk_aio_count++; + VXHS_SPIN_UNLOCK(s->vdisk_lock); + + iio_flags = (IIO_FLAG_DONE | IIO_FLAG_ASYNC); + + switch (iodir) { + case VDISK_AIO_WRITE: + vxhs_inc_acb_segment_count(acb, 1); + ret = qemu_iio_writev(s->qnio_ctx, + s->vdisk_hostinfo[s->vdisk_cur_host_idx].vdisk_rfd, + qiov->iov, qiov->niov, offset, (void *)acb, iio_flags); + break; + case VDISK_AIO_READ: + vxhs_inc_acb_segment_count(acb, 1); + ret = qemu_iio_readv(s->qnio_ctx, + s->vdisk_hostinfo[s->vdisk_cur_host_idx].vdisk_rfd, + qiov->iov, qiov->niov, offset, (void *)acb, iio_flags); + break; + default: + trace_vxhs_aio_rw_invalid(iodir); + goto errout; + } + + if (ret != 0) { + trace_vxhs_aio_rw_ioerr( + s->vdisk_guid, iodir, size, offset, + acb, acb->segments, ret, errno); + /* + * Don't retry I/Os against vDisk having no + * redundency or statefull storage on compute + * + * TODO: Revisit this code path to see if any + * particular error needs to be handled. + * At this moment failing the I/O. + */ + VXHS_SPIN_LOCK(s->vdisk_lock); + if (s->vdisk_nhosts == 1) { + trace_vxhs_aio_rw_iofail(s->vdisk_guid); + s->vdisk_aio_count--; + vxhs_dec_acb_segment_count(acb, 1); + VXHS_SPIN_UNLOCK(s->vdisk_lock); + goto errout; + } + if (OF_VDISK_FAILED(s)) { + trace_vxhs_aio_rw_devfail( + s->vdisk_guid, iodir, size, offset); + s->vdisk_aio_count--; + vxhs_dec_acb_segment_count(acb, 1); + VXHS_SPIN_UNLOCK(s->vdisk_lock); + goto errout; + } + if (OF_VDISK_IOFAILOVER_IN_PROGRESS(s)) { + /* + * Queue all incoming io requests after failover starts. + * Number of requests that can arrive is limited by io queue depth + * so an app blasting independent ios will not exhaust memory. + */ + QSIMPLEQ_INSERT_TAIL(&s->vdisk_aio_retryq, acb, retry_entry); + s->vdisk_aio_retry_qd++; + OF_AIOCB_FLAGS_SET_QUEUED(acb); + s->vdisk_aio_count--; + vxhs_dec_acb_segment_count(acb, 1); + VXHS_SPIN_UNLOCK(s->vdisk_lock); + trace_vxhs_aio_rw_retry(s->vdisk_guid, acb, 2); + goto out; + } + OF_VDISK_SET_IOFAILOVER_IN_PROGRESS(s); + QSIMPLEQ_INSERT_TAIL(&s->vdisk_aio_retryq, acb, retry_entry); + s->vdisk_aio_retry_qd++; + OF_AIOCB_FLAGS_SET_QUEUED(acb); + vxhs_dec_acb_segment_count(acb, 1); + trace_vxhs_aio_rw_retry(s->vdisk_guid, acb, 3); + /* + * Start I/O failover if there is no active + * AIO within vxhs block driver. + */ + if (--s->vdisk_aio_count == 0) { + VXHS_SPIN_UNLOCK(s->vdisk_lock); + /* + * Start IO failover + */ + vxhs_failover_io(s); + goto out; + } + VXHS_SPIN_UNLOCK(s->vdisk_lock); + } + +out: + return &acb->common; + +errout: + qemu_aio_unref(acb); + return NULL; +} + +BlockAIOCB *vxhs_aio_readv(BlockDriverState *bs, + int64_t sector_num, QEMUIOVector *qiov, + int nb_sectors, + BlockCompletionFunc *cb, void *opaque) +{ + return vxhs_aio_rw(bs, sector_num, qiov, nb_sectors, + cb, opaque, VDISK_AIO_READ); +} + +BlockAIOCB *vxhs_aio_writev(BlockDriverState *bs, + int64_t sector_num, QEMUIOVector *qiov, + int nb_sectors, + BlockCompletionFunc *cb, void *opaque) +{ + return vxhs_aio_rw(bs, sector_num, qiov, nb_sectors, + cb, opaque, VDISK_AIO_WRITE); +} + +/* + * This is called by QEMU when a flush gets triggered from within + * a guest at the block layer, either for IDE or SCSI disks. + */ +int vxhs_co_flush(BlockDriverState *bs) +{ + BDRVVXHSState *s = bs->opaque; + uint64_t size = 0; + int ret = 0; + uint32_t iocount = 0; + + ret = qemu_iio_ioctl(s->qnio_ctx, + s->vdisk_hostinfo[s->vdisk_cur_host_idx].vdisk_rfd, + VDISK_AIO_FLUSH, &size, NULL, IIO_FLAG_SYNC); + + if (ret < 0) { + /* + * Currently not handling the flush ioctl + * failure because of network connection + * disconnect. Since all the writes are + * commited into persistent storage hence + * this flush call is noop and we can safely + * return success status to the caller. + * + * If any write failure occurs for inflight + * write AIO because of network disconnect + * then anyway IO failover will be triggered. + */ + trace_vxhs_co_flush(s->vdisk_guid, ret, errno); + ret = 0; + } + + iocount = vxhs_get_vdisk_iocount(s); + if (iocount > 0) { + trace_vxhs_co_flush_iocnt(iocount); + } + + return ret; +} + +/* + * This is called by guest or QEMU to free blocks. + * When block freed when files deleted in the guest, fstrim utility + * can be used to pass the hints to the block layer if the disk supports + * TRIM. It send WRITE_SAME SCSI command to QEMU virtio-scsi layer, which + * call bdrv_aio_discard interface. + */ +coroutine_fn int vxhs_co_pdiscard(BlockDriverState *bs, + int64_t sector_num, int nb_sectors) +{ + int64_t off, size; + + off = sector_num * BDRV_SECTOR_SIZE; + size = nb_sectors * BDRV_SECTOR_SIZE; + + vxhsErr("We are faking the discard for range off = %lu " + "for %lu bytes\n", off, size); + vxhsErr("returning from discard\n"); + + return 0; +} + +unsigned long vxhs_get_vdisk_stat(BDRVVXHSState *s) +{ + void *ctx = NULL; + int flags = 0; + unsigned long vdisk_size = 0; + int ret = 0; + + ret = qemu_iio_ioctl(s->qnio_ctx, + s->vdisk_hostinfo[s->vdisk_cur_host_idx].vdisk_rfd, + VDISK_STAT, &vdisk_size, ctx, flags); + + if (ret < 0) { + trace_vxhs_get_vdisk_stat(s->vdisk_guid, ret, errno); + } + + return vdisk_size; +} + +/* + * Returns the size of vDisk in bytes. This is required + * by QEMU block upper block layer so that it is visible + * to guest. + */ +int64_t vxhs_getlength(BlockDriverState *bs) +{ + BDRVVXHSState *s = bs->opaque; + unsigned long vdisk_size = 0; + + if (s->vdisk_size > 0) { + vdisk_size = s->vdisk_size; + } else { + /* + * Fetch the vDisk size using stat ioctl + */ + vdisk_size = vxhs_get_vdisk_stat(s); + if (vdisk_size > 0) { + s->vdisk_size = vdisk_size; + } + } + + if (vdisk_size > 0) { + return (int64_t)vdisk_size; /* return size in bytes */ + } else { + return -EIO; + } +} + +/* + * Returns actual blocks allocated for the vDisk. + * This is required by qemu-img utility. + */ +int64_t vxhs_get_allocated_blocks(BlockDriverState *bs) +{ + BDRVVXHSState *s = bs->opaque; + unsigned long vdisk_size = 0; + + if (s->vdisk_size > 0) { + vdisk_size = s->vdisk_size; + } else { + /* + * TODO: + * Once HyperScale storage-virtualizer provides + * actual physical allocation of blocks then + * fetch that information and return back to the + * caller but for now just get the full size. + */ + vdisk_size = vxhs_get_vdisk_stat(s); + if (vdisk_size > 0) { + s->vdisk_size = vdisk_size; + } + } + + if (vdisk_size > 0) { + return (int64_t)vdisk_size; /* return size in bytes */ + } else { + return -EIO; + } +} + +void vxhs_close(BlockDriverState *bs) +{ + BDRVVXHSState *s = bs->opaque; + + close(s->fds[VDISK_FD_READ]); + close(s->fds[VDISK_FD_WRITE]); + + /* + * never close channel - not ref counted, will + * close for all vdisks + */ + if (s->vdisk_hostinfo[s->vdisk_cur_host_idx].vdisk_rfd >= 0) { + qemu_iio_devclose(s->qnio_ctx, 0, + s->vdisk_hostinfo[s->vdisk_cur_host_idx].vdisk_rfd); + } + if (s->vdisk_lock) { + VXHS_SPIN_LOCK_DESTROY(s->vdisk_lock); + s->vdisk_lock = NULL; + } + if (s->vdisk_acb_lock) { + VXHS_SPIN_LOCK_DESTROY(s->vdisk_acb_lock); + s->vdisk_acb_lock = NULL; + } + + /* + * TODO: Verify that all the resources were relinguished. + */ +} + +/* + * If errors are consistent with storage agent failure: + * - Try to reconnect in case error is transient or storage agent restarted. + * - Currently failover is being triggered on per vDisk basis. There is + * a scope of further optimization where failover can be global (per VM). + * - In case of network (storage agent) failure, for all the vDisks, having + * no redundency, I/Os will be failed without attempting for I/O failover + * because of stateless nature of vDisk. + * - If local or source storage agent is down then send an ioctl to remote + * storage agent to check if remote storage agent in a state to accept + * application I/Os. + * - Once remote storage agent is ready to accept I/O, start I/O shipping. + * - If I/Os cannot be serviced then vDisk will be marked failed so that + * new incoming I/Os are returned with failure immediately. + * - If vDisk I/O failover is in progress then all new/inflight I/Os will + * queued and will be restarted or failed based on failover operation + * is successful or not. + * - I/O failover can be started either in I/O forward or I/O backward + * path. + * - I/O failover will be started as soon as all the pending acb(s) + * are queued and there is no pending I/O count. + * - If I/O failover couldn't be completed within QNIO_CONNECT_TIMOUT_SECS + * then vDisk will be marked failed and all I/Os will be completed with + * error. + */ + +int vxhs_switch_storage_agent(BDRVVXHSState *s) +{ + int res = 0; + int flags = (IIO_FLAG_ASYNC | IIO_FLAG_DONE); + + trace_vxhs_switch_storage_agent( + s->vdisk_hostinfo[s->vdisk_ask_failover_idx].hostip, + s->vdisk_guid); + + res = vxhs_reopen_vdisk(s, s->vdisk_ask_failover_idx); + if (res == 0) { + res = qemu_iio_ioctl(s->qnio_ctx, + s->vdisk_hostinfo[s->vdisk_ask_failover_idx].vdisk_rfd, + VDISK_CHECK_IO_FAILOVER_READY, NULL, s, flags); + } + if (res != 0) { + trace_vxhs_switch_storage_agent_failed( + s->vdisk_hostinfo[s->vdisk_ask_failover_idx].hostip, + s->vdisk_guid, res, errno); + /* + * TODO: calling vxhs_failover_ioctl_cb from here ties up the qnio epoll + * loop if qemu_iio_ioctl fails synchronously (-1) for all hosts in io + * target list. + */ + + /* try next host */ + vxhs_failover_ioctl_cb(res, s); + } + return res; +} + +void vxhs_failover_ioctl_cb(int res, void *ctx) +{ + BDRVVXHSState *s = ctx; + + if (res == 0) { + /* found failover target */ + s->vdisk_cur_host_idx = s->vdisk_ask_failover_idx; + s->vdisk_ask_failover_idx = 0; + trace_vxhs_failover_ioctl_cb( + s->vdisk_hostinfo[s->vdisk_cur_host_idx].hostip, + s->vdisk_guid); + VXHS_SPIN_LOCK(s->vdisk_lock); + OF_VDISK_RESET_IOFAILOVER_IN_PROGRESS(s); + VXHS_SPIN_UNLOCK(s->vdisk_lock); + vxhs_handle_queued_ios(s); + } else { + /* keep looking */ + trace_vxhs_failover_ioctl_cb_retry(s->vdisk_guid); + s->vdisk_ask_failover_idx++; + if (s->vdisk_ask_failover_idx == s->vdisk_nhosts) { + /* pause and cycle through list again */ + sleep(QNIO_CONNECT_RETRY_SECS); + s->vdisk_ask_failover_idx = 0; + } + res = vxhs_switch_storage_agent(s); + } +} + +int vxhs_failover_io(BDRVVXHSState *s) +{ + int res = 0; + + trace_vxhs_failover_io(s->vdisk_guid); + + s->vdisk_ask_failover_idx = 0; + res = vxhs_switch_storage_agent(s); + + return res; +} + +/* + * Try to reopen the vDisk on one of the available hosts + * If vDisk reopen is successful on any of the host then + * check if that node is ready to accept I/O. + */ +int vxhs_reopen_vdisk(BDRVVXHSState *s, int index) +{ + char *of_vsa_addr = NULL; + char *file_name = NULL; + int res = 0; + + /* + * Don't close the channel if it was opened + * before successfully. It will be handled + * within iio* api if the same channel open + * fd is reused. + * + * close stale vdisk device remote fd since + * it is invalid after channel disconnect. + */ + if (s->vdisk_hostinfo[index].vdisk_rfd >= 0) { + qemu_iio_devclose(s->qnio_ctx, 0, + s->vdisk_hostinfo[index].vdisk_rfd); + s->vdisk_hostinfo[index].vdisk_rfd = -1; + } + /* + * build storage agent address and vdisk device name strings + */ + of_vsa_addr = g_new0(char, OF_MAX_SERVER_ADDR); + file_name = g_new0(char, OF_MAX_FILE_LEN); + snprintf(file_name, OF_MAX_FILE_LEN, "%s%s", vdisk_prefix, s->vdisk_guid); + snprintf(of_vsa_addr, OF_MAX_SERVER_ADDR, "of://%s:%d", + s->vdisk_hostinfo[index].hostip, s->vdisk_hostinfo[index].port); + /* + * open qnio channel to storage agent if not opened before. + */ + if (s->vdisk_hostinfo[index].qnio_cfd < 0) { + s->vdisk_hostinfo[index].qnio_cfd = + qemu_open_iio_conn(global_qnio_ctx, of_vsa_addr, 0); + if (s->vdisk_hostinfo[index].qnio_cfd < 0) { + trace_vxhs_reopen_vdisk(s->vdisk_hostinfo[index].hostip); + res = ENODEV; + goto out; + } + } + /* + * open vdisk device + */ + s->vdisk_hostinfo[index].vdisk_rfd = + qemu_iio_devopen(global_qnio_ctx, + s->vdisk_hostinfo[index].qnio_cfd, file_name, 0); + if (s->vdisk_hostinfo[index].vdisk_rfd < 0) { + trace_vxhs_reopen_vdisk_openfail(file_name); + res = EIO; + goto out; + } +out: + if (of_vsa_addr) { + g_free(of_vsa_addr); + } + if (file_name) { + g_free(file_name); + } + return res; +} + +/* + * vxhs_build_io_target_list: Initialize io target list with ip addresses of + * local storage agent and reflection target storage agents. The local storage + * agent ip is the efficient internal address in the uri, e.g. 192.168.0.2. + * The local storage agent address is stored at index 0. The reflection target + * ips, are the E-W data network addresses of the reflection node agents, also + * extracted from the uri. + */ +int vxhs_build_io_target_list(BDRVVXHSState *s, char **filenames) +{ + URI *uri = NULL; + int i = 0; + + for (i = 0; filenames[i] != NULL && *filenames[i]; i++) { + trace_vxhs_build_io_target_list(i+1, filenames[i]); + uri = uri_parse(filenames[i]); + assert(uri != NULL && uri->server != NULL); + s->vdisk_hostinfo[i].hostip = g_new0(char, strlen(uri->server)); + strncpy((s->vdisk_hostinfo[i].hostip), uri->server, IP_ADDR_LEN); + s->vdisk_hostinfo[i].port = uri->port; + s->vdisk_hostinfo[i].qnio_cfd = -1; + s->vdisk_hostinfo[i].vdisk_rfd = -1; + if (i==0 && (strstr(uri->path, "vxhs") == NULL)) { + s->vdisk_guid = g_new0(char, strlen(uri->path)+3); + strcpy((s->vdisk_guid), uri->path); + strcat((s->vdisk_guid), "}"); + } + uri_free(uri); + } + s->vdisk_nhosts = i; + s->vdisk_cur_host_idx = 0; + + return 0; +} + +int vxhs_handle_queued_ios(BDRVVXHSState *s) +{ + VXHSAIOCB *acb = NULL; + int res = 0; + + VXHS_SPIN_LOCK(s->vdisk_lock); + while ((acb = QSIMPLEQ_FIRST(&s->vdisk_aio_retryq)) != NULL) { + /* + * Before we process the acb, check whether I/O failover + * started again due to failback or cascading failure. + */ + if (OF_VDISK_IOFAILOVER_IN_PROGRESS(s)) { + VXHS_SPIN_UNLOCK(s->vdisk_lock); + goto out; + } + QSIMPLEQ_REMOVE_HEAD(&s->vdisk_aio_retryq, retry_entry); + s->vdisk_aio_retry_qd--; + OF_AIOCB_FLAGS_RESET_QUEUED(acb); + if (OF_VDISK_FAILED(s)) { + VXHS_SPIN_UNLOCK(s->vdisk_lock); + vxhs_fail_aio(acb, EIO); + VXHS_SPIN_LOCK(s->vdisk_lock); + } else { + VXHS_SPIN_UNLOCK(s->vdisk_lock); + res = vxhs_restart_aio(acb); + trace_vxhs_handle_queued_ios(acb, res); + VXHS_SPIN_LOCK(s->vdisk_lock); + if (res) { + QSIMPLEQ_INSERT_TAIL(&s->vdisk_aio_retryq, + acb, retry_entry); + OF_AIOCB_FLAGS_SET_QUEUED(acb); + VXHS_SPIN_UNLOCK(s->vdisk_lock); + goto out; + } + } + } + VXHS_SPIN_UNLOCK(s->vdisk_lock); +out: + return res; +} + +int vxhs_restart_aio(VXHSAIOCB *acb) +{ + BDRVVXHSState *s = NULL; + int iio_flags = 0; + int res = 0; + + s = acb->common.bs->opaque; + + if (acb->direction == VDISK_AIO_WRITE) { + vxhs_inc_vdisk_iocount(s, 1); + vxhs_inc_acb_segment_count(acb, 1); + iio_flags = (IIO_FLAG_DONE | IIO_FLAG_ASYNC); + res = qemu_iio_writev(s->qnio_ctx, + s->vdisk_hostinfo[s->vdisk_cur_host_idx].vdisk_rfd, + acb->qiov->iov, acb->qiov->niov, + acb->io_offset, (void *)acb, iio_flags); + } + + if (acb->direction == VDISK_AIO_READ) { + vxhs_inc_vdisk_iocount(s, 1); + vxhs_inc_acb_segment_count(acb, 1); + iio_flags = (IIO_FLAG_DONE | IIO_FLAG_ASYNC); + res = qemu_iio_readv(s->qnio_ctx, + s->vdisk_hostinfo[s->vdisk_cur_host_idx].vdisk_rfd, + acb->qiov->iov, acb->qiov->niov, + acb->io_offset, (void *)acb, iio_flags); + } + + if (res != 0) { + vxhs_dec_vdisk_iocount(s, 1); + vxhs_dec_acb_segment_count(acb, 1); + trace_vxhs_restart_aio(acb->direction, res, errno); + } + + return res; +} + +void vxhs_fail_aio(VXHSAIOCB *acb, int err) +{ + BDRVVXHSState *s = NULL; + int segcount = 0; + int rv = 0; + + s = acb->common.bs->opaque; + + trace_vxhs_fail_aio(s->vdisk_guid, acb); + if (!acb->ret) { + acb->ret = err; + } + VXHS_SPIN_LOCK(s->vdisk_acb_lock); + segcount = acb->segments; + VXHS_SPIN_UNLOCK(s->vdisk_acb_lock); + if (segcount == 0) { + /* + * Complete the io request + */ + rv = qemu_write_full(s->fds[VDISK_FD_WRITE], &acb, sizeof(acb)); + if (rv != sizeof(acb)) { + error_report("VXHS AIO completion failed: %s", + strerror(errno)); + abort(); + } + } +} + +static BlockDriver bdrv_vxhs = { + .format_name = "vxhs", + .protocol_name = "vxhs", + .instance_size = sizeof(BDRVVXHSState), + .bdrv_file_open = vxhs_open, + .bdrv_create = vxhs_create, + .bdrv_close = vxhs_close, + .bdrv_getlength = vxhs_getlength, + .bdrv_get_allocated_file_size = vxhs_get_allocated_blocks, + .bdrv_aio_readv = vxhs_aio_readv, + .bdrv_aio_writev = vxhs_aio_writev, + .bdrv_co_flush_to_disk = vxhs_co_flush, + .bdrv_co_pdiscard = vxhs_co_pdiscard, +}; + +void bdrv_vxhs_init(void) +{ + trace_vxhs_bdrv_init(vxhs_drv_version); + bdrv_register(&bdrv_vxhs); +} + +/* + * The line below is how our drivier is initialized. + * DO NOT TOUCH IT + */ +block_init(bdrv_vxhs_init); diff --git a/block/vxhs.h b/block/vxhs.h new file mode 100644 index 0000000..2979f76 --- /dev/null +++ b/block/vxhs.h @@ -0,0 +1,294 @@ +/* + * QEMU Block driver for Veritas HyperScale (VxHS) + * + * This work is licensed under the terms of the GNU GPL, version 2 or later. + * See the COPYING file in the top-level directory. + * + */ + +#ifndef VXHSD_H +#define VXHSD_H + +#include +#include +#include + +#include "qemu/osdep.h" +#include "qapi/error.h" +#include "qemu/error-report.h" +#include "block/block_int.h" +#include "qemu/uri.h" +#include "qemu/queue.h" + +#define vxhsErr(fmt, ...) { \ + time_t t = time(0); \ + char buf[9] = {0}; \ + strftime(buf, 9, "%H:%M:%S", localtime(&t)); \ + fprintf(stderr, "[%s: %lu] %d: %s():\t", buf, pthread_self(), \ + __LINE__, __func__); \ + fprintf(stderr, fmt, ## __VA_ARGS__); \ +} + +#define OF_GUID_STR_LEN 40 +#define OF_GUID_STR_SZ (OF_GUID_STR_LEN + 1) +#define QNIO_CONNECT_RETRY_SECS 5 +#define QNIO_CONNECT_TIMOUT_SECS 120 + +/* constants from io_qnio.h */ +#define IIO_REASON_DONE 0x00000004 +#define IIO_REASON_EVENT 0x00000008 +#define IIO_REASON_HUP 0x00000010 + +/* + * IO specific flags + */ +#define IIO_FLAG_ASYNC 0x00000001 +#define IIO_FLAG_DONE 0x00000010 +#define IIO_FLAG_SYNC 0 + +/* constants from error.h */ +#define VXERROR_RETRY_ON_SOURCE 44 +#define VXERROR_HUP 901 +#define VXERROR_CHANNEL_HUP 903 + +/* constants from iomgr.h and opcode.h */ +#define IRP_READ_REQUEST 0x1FFF +#define IRP_WRITE_REQUEST 0x2FFF +#define IRP_VDISK_CHECK_IO_FAILOVER_READY 2020 + +/* Lock specific macros */ +#define VXHS_SPIN_LOCK_ALLOC \ + (qemu_ck_initialize_lock()) +#define VXHS_SPIN_LOCK(lock) \ + (qemu_ck_spin_lock(lock)) +#define VXHS_SPIN_UNLOCK(lock) \ + (qemu_ck_spin_unlock(lock)) +#define VXHS_SPIN_LOCK_DESTROY(lock) \ + (qemu_ck_destroy_lock(lock)) + +typedef enum { + VDISK_AIO_READ, + VDISK_AIO_WRITE, + VDISK_STAT, + VDISK_TRUNC, + VDISK_AIO_FLUSH, + VDISK_AIO_RECLAIM, + VDISK_GET_GEOMETRY, + VDISK_CHECK_IO_FAILOVER_READY, + VDISK_AIO_LAST_CMD +} VDISKAIOCmd; + +typedef enum { + VXHS_IO_INPROGRESS, + VXHS_IO_COMPLETED, + VXHS_IO_ERROR +} VXHSIOState; + + +typedef void *qemu_aio_ctx_t; +typedef void (*qnio_callback_t)(ssize_t retval, void *arg); + +#define VDISK_FD_READ 0 +#define VDISK_FD_WRITE 1 + +#define QNIO_VDISK_NONE 0x00 +#define QNIO_VDISK_CREATE 0x01 + +/* max IO size supported by QEMU NIO lib */ +#define QNIO_MAX_IO_SIZE 4194304 + +#define IP_ADDR_LEN 20 +#define OF_MAX_FILE_LEN 1024 +#define OF_MAX_SERVER_ADDR 1024 +#define MAX_HOSTS 4 + +/* + * Opcodes for making IOCTL on QEMU NIO library + */ +#define BASE_OPCODE_SHARED 1000 +#define BASE_OPCODE_DAL 2000 +#define IRP_VDISK_STAT (BASE_OPCODE_SHARED + 5) +#define IRP_VDISK_GET_GEOMETRY (BASE_OPCODE_DAL + 17) +#define IRP_VDISK_READ_PARTITION (BASE_OPCODE_DAL + 18) +#define IRP_VDISK_FLUSH (BASE_OPCODE_DAL + 19) + +/* + * BDRVVXHSState specific flags + */ +#define OF_VDISK_FLAGS_STATE_ACTIVE 0x0000000000000001 +#define OF_VDISK_FLAGS_STATE_FAILED 0x0000000000000002 +#define OF_VDISK_FLAGS_IOFAILOVER_IN_PROGRESS 0x0000000000000004 + +#define OF_VDISK_ACTIVE(s) \ + ((s)->vdisk_flags & OF_VDISK_FLAGS_STATE_ACTIVE) +#define OF_VDISK_SET_ACTIVE(s) \ + ((s)->vdisk_flags |= OF_VDISK_FLAGS_STATE_ACTIVE) +#define OF_VDISK_RESET_ACTIVE(s) \ + ((s)->vdisk_flags &= ~OF_VDISK_FLAGS_STATE_ACTIVE) + +#define OF_VDISK_FAILED(s) \ + ((s)->vdisk_flags & OF_VDISK_FLAGS_STATE_FAILED) +#define OF_VDISK_SET_FAILED(s) \ + ((s)->vdisk_flags |= OF_VDISK_FLAGS_STATE_FAILED) +#define OF_VDISK_RESET_FAILED(s) \ + ((s)->vdisk_flags &= ~OF_VDISK_FLAGS_STATE_FAILED) + +#define OF_VDISK_IOFAILOVER_IN_PROGRESS(s) \ + ((s)->vdisk_flags & OF_VDISK_FLAGS_IOFAILOVER_IN_PROGRESS) +#define OF_VDISK_SET_IOFAILOVER_IN_PROGRESS(s) \ + ((s)->vdisk_flags |= OF_VDISK_FLAGS_IOFAILOVER_IN_PROGRESS) +#define OF_VDISK_RESET_IOFAILOVER_IN_PROGRESS(s) \ + ((s)->vdisk_flags &= ~OF_VDISK_FLAGS_IOFAILOVER_IN_PROGRESS) + +/* + * VXHSAIOCB specific flags + */ +#define OF_ACB_QUEUED 0x00000001 + +#define OF_AIOCB_FLAGS_QUEUED(a) \ + ((a)->flags & OF_ACB_QUEUED) +#define OF_AIOCB_FLAGS_SET_QUEUED(a) \ + ((a)->flags |= OF_ACB_QUEUED) +#define OF_AIOCB_FLAGS_RESET_QUEUED(a) \ + ((a)->flags &= ~OF_ACB_QUEUED) + +typedef struct qemu2qnio_ctx { + uint32_t qnio_flag; + uint64_t qnio_size; + char *qnio_channel; + char *target; + qnio_callback_t qnio_cb; +} qemu2qnio_ctx_t; + +typedef qemu2qnio_ctx_t qnio2qemu_ctx_t; + +typedef struct LibQNIOSymbol { + const char *name; + gpointer *addr; +} LibQNIOSymbol; + +typedef void (*iio_cb_t) (uint32_t rfd, uint32_t reason, void *ctx, + void *reply); + +/* + * HyperScale AIO callbacks structure + */ +typedef struct VXHSAIOCB { + BlockAIOCB common; + size_t ret; + size_t size; + QEMUBH *bh; + int aio_done; + int segments; + int flags; + size_t io_offset; + QEMUIOVector *qiov; + void *buffer; + int direction; /* IO direction (r/w) */ + QSIMPLEQ_ENTRY(VXHSAIOCB) retry_entry; +} VXHSAIOCB; + +typedef struct VXHSvDiskHostsInfo { + int qnio_cfd; /* Channel FD */ + int vdisk_rfd; /* vDisk remote FD */ + char *hostip; /* Host's IP addresses */ + int port; /* Host's port number */ +} VXHSvDiskHostsInfo; + +/* + * Structure per vDisk maintained for state + */ +typedef struct BDRVVXHSState { + int fds[2]; + int64_t vdisk_size; + int64_t vdisk_blocks; + int64_t vdisk_flags; + int vdisk_aio_count; + int event_reader_pos; + VXHSAIOCB *qnio_event_acb; + void *qnio_ctx; + void *vdisk_lock; /* Lock to protect BDRVVXHSState */ + void *vdisk_acb_lock; /* Protects ACB */ + VXHSvDiskHostsInfo vdisk_hostinfo[MAX_HOSTS]; /* Per host info */ + int vdisk_nhosts; /* Total number of hosts */ + int vdisk_cur_host_idx; /* IOs are being shipped to */ + int vdisk_ask_failover_idx; /*asking permsn to ship io*/ + QSIMPLEQ_HEAD(aio_retryq, VXHSAIOCB) vdisk_aio_retryq; + int vdisk_aio_retry_qd; + char *vdisk_guid; + AioContext *aio_context; +} BDRVVXHSState; + +int vxhs_load_iio_ops(void); +void bdrv_vxhs_init(void); +void *vxhs_initialize(void); +void *vxhs_setup_qnio(void); +int qemu_qnio_fini(BDRVVXHSState *s); +void vxhs_iio_callback(uint32_t rfd, uint32_t reason, void *ctx, void *m); +int qemu_qnio_init(BDRVVXHSState *s, const char *vxhs_uri); +void vxhs_aio_event_reader(void *opaque); +void vxhs_complete_aio(VXHSAIOCB *acb, BDRVVXHSState *s); +int vxhs_aio_flush_cb(void *opaque); +void vxhs_finish_aiocb(ssize_t ret, void *arg); +unsigned long vxhs_get_vdisk_stat(BDRVVXHSState *s); +int vxhs_open(BlockDriverState *bs, QDict *options, + int bdrv_flags, Error **errp); +int vxhs_create(const char *filename, QemuOpts *options, + Error **errp); +int vxhs_open_device(const char *vxhs_uri, int *cfd, int *rfd, + BDRVVXHSState *s); +void vxhs_close(BlockDriverState *bs); +void vxhs_aio_cancel(BlockAIOCB *blockacb); +int vxhs_truncate(BlockDriverState *bs, int64_t offset); +int qemu_submit_io(BDRVVXHSState *s, struct iovec *iov, int64_t niov, + int64_t offset, int cmd, qemu_aio_ctx_t acb); +BlockAIOCB *vxhs_aio_flush(BlockDriverState *bs, + BlockCompletionFunc *cb, void *opaque); +BlockAIOCB *vxhs_aio_pdiscard(BlockDriverState *bs, int64_t sector_num, + int nb_sectors, + BlockCompletionFunc *cb, + void *opaque); +BlockAIOCB *vxhs_aio_readv(BlockDriverState *bs, int64_t sector_num, + QEMUIOVector *qiov, int nb_sectors, + BlockCompletionFunc *cb, void *opaque); +BlockAIOCB *vxhs_aio_writev(BlockDriverState *bs, int64_t sector_num, + QEMUIOVector *qiov, int nb_sectors, + BlockCompletionFunc *cb, + void *opaque); +coroutine_fn int vxhs_co_read(BlockDriverState *bs, int64_t sector_num, + uint8_t *buf, int nb_sectors); +coroutine_fn int vxhs_co_write(BlockDriverState *bs, int64_t sector_num, + const uint8_t *buf, int nb_sectors); +int64_t vxhs_get_allocated_blocks(BlockDriverState *bs); +BlockAIOCB *vxhs_aio_rw(BlockDriverState *bs, int64_t sector_num, + QEMUIOVector *qiov, int nb_sectors, + BlockCompletionFunc *cb, + void *opaque, int write); +int vxhs_co_flush(BlockDriverState *bs); +coroutine_fn int vxhs_co_pdiscard(BlockDriverState *bs, + int64_t sector_num, int nb_sectors); +int vxhs_has_zero_init(BlockDriverState *bs); +int64_t vxhs_getlength(BlockDriverState *bs); +void vxhs_inc_vdisk_iocount(void *ptr, uint32_t delta); +void vxhs_dec_vdisk_iocount(void *ptr, uint32_t delta); +uint32_t vxhs_get_vdisk_iocount(void *ptr); +void vxhs_inc_acb_segment_count(void *ptr, int count); +void vxhs_dec_acb_segment_count(void *ptr, int count); +int vxhs_dec_and_get_acb_segment_count(void *ptr, int count); +void vxhs_set_acb_buffer(void *ptr, void *buffer); +int vxhs_sync_rw(BlockDriverState *bs, int64_t sector_num, + const uint8_t *buf, int nb_sectors, int write); +int vxhs_failover_io(BDRVVXHSState *s); +int vxhs_reopen_vdisk(BDRVVXHSState *s, + int hostinfo_index); +int vxhs_switch_storage_agent(BDRVVXHSState *s); +int vxhs_check_io_failover_ready(BDRVVXHSState *s, + int hostinfo_index); +int vxhs_build_io_target_list(BDRVVXHSState *s, char **filenames); +int vxhs_handle_queued_ios(BDRVVXHSState *s); +int vxhs_restart_aio(VXHSAIOCB *acb); +void vxhs_fail_aio(VXHSAIOCB *acb, int err); +void vxhs_failover_ioctl_cb(int res, void *ctx); +char *vxhs_string_iterate(char *p, const char *d, const size_t len); + +#endif diff --git a/configure b/configure index f57fcc6..c4f6261 100755 --- a/configure +++ b/configure @@ -320,6 +320,7 @@ vhdx="" numa="" tcmalloc="no" jemalloc="no" +vxhs="yes" # parse CC options first for opt do @@ -1150,6 +1151,11 @@ for opt do ;; --enable-jemalloc) jemalloc="yes" ;; + --disable-vxhs) vxhs="no" + ;; + --enable-vxhs) vxhs="yes" + ;; + *) echo "ERROR: unknown option $opt" echo "Try '$0 --help' for more information" @@ -1380,6 +1386,7 @@ disabled with --disable-FEATURE, default is enabled if available: numa libnuma support tcmalloc tcmalloc support jemalloc jemalloc support + vxhs Veritas HyperScale vDisk backend support NOTE: The object files are built at the place where configure is launched EOF @@ -4543,6 +4550,43 @@ if do_cc -nostdlib -Wl,-r -Wl,--no-relax -o $TMPMO $TMPO; then fi ########################################## +# Veritas HyperScale block driver VxHS +# Check if libqnio is installed +if test "$vxhs" != "no" ; then + cat > $TMPC < +#include + +void vxhs_inc_acb_segment_count(void *acb, int count); +void vxhs_dec_acb_segment_count(void *acb, int count); +void vxhs_set_acb_buffer(void *ptr, void *buffer); + +void vxhs_inc_acb_segment_count(void *ptr, int count) +{ +} +void vxhs_dec_acb_segment_count(void *ptr, int count) +{ +} +void vxhs_set_acb_buffer(void *ptr, void *buffer) +{ +} +int main(void) { + qemu_ck_initialize_lock(); + return 0; +} +EOF + vxhs_libs="-lqnioshim -lqnio" + if compile_prog "" "$vxhs_libs" ; then + vxhs=yes + else + if test "$vxhs" = "yes" ; then + feature_not_found "vxhs block device" "Install libqnio. See github" + fi + vxhs=no + fi +fi + +########################################## # End of CC checks # After here, no more $cc or $ld runs @@ -5479,6 +5523,12 @@ if test "$pthread_setname_np" = "yes" ; then echo "CONFIG_PTHREAD_SETNAME_NP=y" >> $config_host_mak fi +if test "$vxhs" = "yes" ; then + echo "CONFIG_VXHS=y" >> $config_host_mak + echo "VXHS_CFLAGS=$vxhs_cflags" >> $config_host_mak + echo "VXHS_LIBS=$vxhs_libs" >> $config_host_mak +fi + if test "$tcg_interpreter" = "yes"; then QEMU_INCLUDES="-I\$(SRC_PATH)/tcg/tci $QEMU_INCLUDES" elif test "$ARCH" = "sparc64" ; then