From patchwork Thu Jan 23 16:45:10 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Dr. David Alan Gilbert" X-Patchwork-Id: 11348989 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 4D0471398 for ; Thu, 23 Jan 2020 18:42:40 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 21BDA21D7E for ; Thu, 23 Jan 2020 18:42:40 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (1024-bit key) header.d=redhat.com header.i=@redhat.com header.b="Cv0zY1Ng" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 21BDA21D7E Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=redhat.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=qemu-devel-bounces+patchwork-qemu-devel=patchwork.kernel.org@nongnu.org Received: from localhost ([::1]:34980 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1iuhR8-0006G8-JV for patchwork-qemu-devel@patchwork.kernel.org; Thu, 23 Jan 2020 13:42:38 -0500 Received: from eggs.gnu.org ([2001:470:142:3::10]:41129) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1iufdh-0004iy-P3 for qemu-devel@nongnu.org; Thu, 23 Jan 2020 11:47:31 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1iufde-0006Oc-9M for qemu-devel@nongnu.org; Thu, 23 Jan 2020 11:47:29 -0500 Received: from us-smtp-delivery-1.mimecast.com ([205.139.110.120]:40172 helo=us-smtp-1.mimecast.com) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1iufde-0006OF-66 for qemu-devel@nongnu.org; Thu, 23 Jan 2020 11:47:26 -0500 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1579798045; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=P1S7BvPGij6D+BUScO+a/R/uy1LHQuGCVOlGGCLFdJM=; b=Cv0zY1NgKevD178Kts1mdLoqGTOrXCP4qVDH/ZJmOrWvBMo4GbTIMKslu79FSEeqFiIRcE gTDpiaX+64ka1Zq6dWBgIg1dkCz9QjaO9r5YOvQOzLAfSR6WOizS0O/GUYVKlLj+voaY0B cYZ8yHwuMXH3EtfVZw2oDkik/Mj5LA0= 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-242-AUYU0EJzPiKDxgQ-7qYGoQ-1; Thu, 23 Jan 2020 11:47:23 -0500 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 mimecast-mx01.redhat.com (Postfix) with ESMTPS id 17618101027A for ; Thu, 23 Jan 2020 16:47:23 +0000 (UTC) Received: from dgilbert-t580.localhost (ovpn-116-110.ams2.redhat.com [10.36.116.110]) by smtp.corp.redhat.com (Postfix) with ESMTP id 6504738F; Thu, 23 Jan 2020 16:47:22 +0000 (UTC) From: "Dr. David Alan Gilbert (git)" To: qemu-devel@nongnu.org, stefanha@redhat.com Subject: [PULL 028/108] virtiofsd: Fast path for virtio read Date: Thu, 23 Jan 2020 16:45:10 +0000 Message-Id: <20200123164630.91498-29-dgilbert@redhat.com> In-Reply-To: <20200123164630.91498-1-dgilbert@redhat.com> References: <20200123164630.91498-1-dgilbert@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.84 on 10.5.11.23 X-MC-Unique: AUYU0EJzPiKDxgQ-7qYGoQ-1 X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] X-Received-From: 205.139.110.120 X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+patchwork-qemu-devel=patchwork.kernel.org@nongnu.org Sender: "Qemu-devel" From: "Dr. David Alan Gilbert" Readv the data straight into the guests buffer. Signed-off-by: Dr. David Alan Gilbert With fix by: Signed-off-by: Eryu Guan Reviewed-by: Masayoshi Mizuma Signed-off-by: Dr. David Alan Gilbert --- tools/virtiofsd/fuse_lowlevel.c | 5 + tools/virtiofsd/fuse_virtio.c | 162 ++++++++++++++++++++++++++++++++ tools/virtiofsd/fuse_virtio.h | 4 + 3 files changed, 171 insertions(+) diff --git a/tools/virtiofsd/fuse_lowlevel.c b/tools/virtiofsd/fuse_lowlevel.c index 380d93bd01..4f4684d942 100644 --- a/tools/virtiofsd/fuse_lowlevel.c +++ b/tools/virtiofsd/fuse_lowlevel.c @@ -475,6 +475,11 @@ static int fuse_send_data_iov_fallback(struct fuse_session *se, return fuse_send_msg(se, ch, iov, iov_count); } + if (fuse_lowlevel_is_virtio(se) && buf->count == 1 && + buf->buf[0].flags == (FUSE_BUF_IS_FD | FUSE_BUF_FD_SEEK)) { + return virtio_send_data_iov(se, ch, iov, iov_count, buf, len); + } + abort(); /* Will have taken vhost path */ return 0; } diff --git a/tools/virtiofsd/fuse_virtio.c b/tools/virtiofsd/fuse_virtio.c index f1adeb6345..7e2711b504 100644 --- a/tools/virtiofsd/fuse_virtio.c +++ b/tools/virtiofsd/fuse_virtio.c @@ -230,6 +230,168 @@ err: return ret; } +/* + * Callback from fuse_send_data_iov_* when it's virtio and the buffer + * is a single FD with FUSE_BUF_IS_FD | FUSE_BUF_FD_SEEK + * We need send the iov and then the buffer. + * Return 0 on success + */ +int virtio_send_data_iov(struct fuse_session *se, struct fuse_chan *ch, + struct iovec *iov, int count, struct fuse_bufvec *buf, + size_t len) +{ + int ret = 0; + VuVirtqElement *elem; + VuVirtq *q; + + assert(count >= 1); + assert(iov[0].iov_len >= sizeof(struct fuse_out_header)); + + struct fuse_out_header *out = iov[0].iov_base; + /* TODO: Endianness! */ + + size_t iov_len = iov_size(iov, count); + size_t tosend_len = iov_len + len; + + out->len = tosend_len; + + fuse_log(FUSE_LOG_DEBUG, "%s: count=%d len=%zd iov_len=%zd\n", __func__, + count, len, iov_len); + + /* unique == 0 is notification which we don't support */ + assert(out->unique); + + /* For virtio we always have ch */ + assert(ch); + assert(!ch->qi->reply_sent); + elem = ch->qi->qe; + q = &ch->qi->virtio_dev->dev.vq[ch->qi->qidx]; + + /* The 'in' part of the elem is to qemu */ + unsigned int in_num = elem->in_num; + struct iovec *in_sg = elem->in_sg; + size_t in_len = iov_size(in_sg, in_num); + fuse_log(FUSE_LOG_DEBUG, "%s: elem %d: with %d in desc of length %zd\n", + __func__, elem->index, in_num, in_len); + + /* + * The elem should have room for a 'fuse_out_header' (out from fuse) + * plus the data based on the len in the header. + */ + if (in_len < sizeof(struct fuse_out_header)) { + fuse_log(FUSE_LOG_ERR, "%s: elem %d too short for out_header\n", + __func__, elem->index); + ret = E2BIG; + goto err; + } + if (in_len < tosend_len) { + fuse_log(FUSE_LOG_ERR, "%s: elem %d too small for data len %zd\n", + __func__, elem->index, tosend_len); + ret = E2BIG; + goto err; + } + + /* TODO: Limit to 'len' */ + + /* First copy the header data from iov->in_sg */ + copy_iov(iov, count, in_sg, in_num, iov_len); + + /* + * Build a copy of the the in_sg iov so we can skip bits in it, + * including changing the offsets + */ + struct iovec *in_sg_cpy = calloc(sizeof(struct iovec), in_num); + assert(in_sg_cpy); + memcpy(in_sg_cpy, in_sg, sizeof(struct iovec) * in_num); + /* These get updated as we skip */ + struct iovec *in_sg_ptr = in_sg_cpy; + int in_sg_cpy_count = in_num; + + /* skip over parts of in_sg that contained the header iov */ + size_t skip_size = iov_len; + + size_t in_sg_left = 0; + do { + while (skip_size != 0 && in_sg_cpy_count) { + if (skip_size >= in_sg_ptr[0].iov_len) { + skip_size -= in_sg_ptr[0].iov_len; + in_sg_ptr++; + in_sg_cpy_count--; + } else { + in_sg_ptr[0].iov_len -= skip_size; + in_sg_ptr[0].iov_base += skip_size; + break; + } + } + + int i; + for (i = 0, in_sg_left = 0; i < in_sg_cpy_count; i++) { + in_sg_left += in_sg_ptr[i].iov_len; + } + fuse_log(FUSE_LOG_DEBUG, + "%s: after skip skip_size=%zd in_sg_cpy_count=%d " + "in_sg_left=%zd\n", + __func__, skip_size, in_sg_cpy_count, in_sg_left); + ret = preadv(buf->buf[0].fd, in_sg_ptr, in_sg_cpy_count, + buf->buf[0].pos); + + if (ret == -1) { + ret = errno; + fuse_log(FUSE_LOG_DEBUG, "%s: preadv failed (%m) len=%zd\n", + __func__, len); + free(in_sg_cpy); + goto err; + } + fuse_log(FUSE_LOG_DEBUG, "%s: preadv ret=%d len=%zd\n", __func__, + ret, len); + if (ret < len && ret) { + fuse_log(FUSE_LOG_DEBUG, "%s: ret < len\n", __func__); + /* Skip over this much next time around */ + skip_size = ret; + buf->buf[0].pos += ret; + len -= ret; + + /* Lets do another read */ + continue; + } + if (!ret) { + /* EOF case? */ + fuse_log(FUSE_LOG_DEBUG, "%s: !ret in_sg_left=%zd\n", __func__, + in_sg_left); + break; + } + if (ret != len) { + fuse_log(FUSE_LOG_DEBUG, "%s: ret!=len\n", __func__); + ret = EIO; + free(in_sg_cpy); + goto err; + } + in_sg_left -= ret; + len -= ret; + } while (in_sg_left); + free(in_sg_cpy); + + /* Need to fix out->len on EOF */ + if (len) { + struct fuse_out_header *out_sg = in_sg[0].iov_base; + + tosend_len -= len; + out_sg->len = tosend_len; + } + + ret = 0; + + vu_queue_push(&se->virtio_dev->dev, q, elem, tosend_len); + vu_queue_notify(&se->virtio_dev->dev, q); + +err: + if (ret == 0) { + ch->qi->reply_sent = true; + } + + return ret; +} + /* Thread function for individual queues, created when a queue is 'started' */ static void *fv_queue_thread(void *opaque) { diff --git a/tools/virtiofsd/fuse_virtio.h b/tools/virtiofsd/fuse_virtio.h index 135a14875a..cc676b9193 100644 --- a/tools/virtiofsd/fuse_virtio.h +++ b/tools/virtiofsd/fuse_virtio.h @@ -26,4 +26,8 @@ int virtio_loop(struct fuse_session *se); int virtio_send_msg(struct fuse_session *se, struct fuse_chan *ch, struct iovec *iov, int count); +int virtio_send_data_iov(struct fuse_session *se, struct fuse_chan *ch, + struct iovec *iov, int count, + struct fuse_bufvec *buf, size_t len); + #endif