From patchwork Wed Oct 31 22:13:09 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Long Li X-Patchwork-Id: 991735 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Authentication-Results: ozlabs.org; spf=none (mailfrom) smtp.mailfrom=vger.kernel.org (client-ip=209.132.180.67; helo=vger.kernel.org; envelope-from=linux-cifs-owner@vger.kernel.org; receiver=) Authentication-Results: ozlabs.org; dmarc=none (p=none dis=none) header.from=linuxonhyperv.com Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 42ljL30zpnz9s8J for ; Thu, 1 Nov 2018 09:14:19 +1100 (AEDT) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726023AbeKAHOR (ORCPT ); Thu, 1 Nov 2018 03:14:17 -0400 Received: from a2nlsmtp01-03.prod.iad2.secureserver.net ([198.71.225.37]:42514 "EHLO a2nlsmtp01-03.prod.iad2.secureserver.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725755AbeKAHOR (ORCPT ); Thu, 1 Nov 2018 03:14:17 -0400 Received: from linuxonhyperv2.linuxonhyperv.com ([107.180.71.197]) by : HOSTING RELAY : with ESMTP id HyjkgVEW1Wzu3HyjkgRF8R; Wed, 31 Oct 2018 15:13:16 -0700 x-originating-ip: 107.180.71.197 Received: from longli by linuxonhyperv2.linuxonhyperv.com with local (Exim 4.91) (envelope-from ) id 1gHyjk-0000gY-4B; Wed, 31 Oct 2018 15:13:16 -0700 From: Long Li To: Steve French , linux-cifs@vger.kernel.org, samba-technical@lists.samba.org, linux-kernel@vger.kernel.org Cc: Long Li Subject: [Patch v4 1/3] CIFS: Add support for direct I/O read Date: Wed, 31 Oct 2018 22:13:09 +0000 Message-Id: <20181031221311.2596-1-longli@linuxonhyperv.com> X-Mailer: git-send-email 2.18.0 Reply-To: longli@microsoft.com X-CMAE-Envelope: MS4wfDU3diU7whGWatn5LsZ3I4dWVsqR2P23Wz3bSGWz0cVGgTggzm6VM+9qqJBuFy3KqLITh4Bd6UGTmXHWZLhVxOWlrWll7tuwST8eHaj3E05CFLnCez1i blM0r3L8X/rNzD0s65Yo0pQ/sGxAjhcmIcdE+B04NhwRswvUzc6q4HHJemnNk6kUeP4lg0araJm6bqpkK+LrHvJaYamHze7PcV8DVX1TTlfHVG59FM9SbeOt fPeceDrNW9uJqHfEIUnF7Ir04zdQ1M/Wwhpb0o+w+iIj+H4je+5986bqEYCv5TWU2Rh/IXD+PLVlSpVXYkafhYmsdVvIVfmbbWJG6x5qxxfVm8qJazl6kYyA JXiBKBbH+R6wgN+YCpCxZfQ9XV1YRw== Sender: linux-cifs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-cifs@vger.kernel.org From: Long Li With direct I/O read, we transfer the data directly from transport layer to the user data buffer. Change in v3: add support for kernel AIO Change in v4: Refactor common read code to __cifs_readv for direct and non-direct I/O. Retry on direct I/O failure. Signed-off-by: Long Li --- fs/cifs/cifsfs.h | 1 + fs/cifs/cifsglob.h | 5 ++ fs/cifs/file.c | 219 +++++++++++++++++++++++++++++++++++++++++++---------- 3 files changed, 186 insertions(+), 39 deletions(-) diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 5f02318..7fba9aa 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -102,6 +102,7 @@ extern int cifs_open(struct inode *inode, struct file *file); extern int cifs_close(struct inode *inode, struct file *file); extern int cifs_closedir(struct inode *inode, struct file *file); extern ssize_t cifs_user_readv(struct kiocb *iocb, struct iov_iter *to); +extern ssize_t cifs_direct_readv(struct kiocb *iocb, struct iov_iter *to); extern ssize_t cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to); extern ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from); extern ssize_t cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from); diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 7f62c98..52248dd 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -1146,6 +1146,11 @@ struct cifs_aio_ctx { unsigned int len; unsigned int total_len; bool should_dirty; + /* + * Indicates if this aio_ctx is for direct_io, + * If yes, iter is a copy of the user passed iov_iter + */ + bool direct_io; }; struct cifs_readdata; diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 87eece6..daab878 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -2965,7 +2965,6 @@ cifs_uncached_readdata_release(struct kref *refcount) kref_put(&rdata->ctx->refcount, cifs_aio_ctx_release); for (i = 0; i < rdata->nr_pages; i++) { put_page(rdata->pages[i]); - rdata->pages[i] = NULL; } cifs_readdata_release(refcount); } @@ -3092,6 +3091,63 @@ cifs_uncached_copy_into_pages(struct TCP_Server_Info *server, return uncached_fill_pages(server, rdata, iter, iter->count); } +static int cifs_resend_rdata(struct cifs_readdata *rdata, + struct list_head *rdata_list, + struct cifs_aio_ctx *ctx) +{ + int wait_retry = 0; + unsigned int rsize, credits; + int rc; + struct TCP_Server_Info *server = tlink_tcon(rdata->cfile->tlink)->ses->server; + + /* + * Try to resend this rdata, waiting for credits up to 3 seconds. + * Note: we are attempting to resend the whole rdata not in segments + */ + do { + rc = server->ops->wait_mtu_credits(server, rdata->bytes, + &rsize, &credits); + + if (rc) + break; + + if (rsize < rdata->bytes) { + add_credits_and_wake_if(server, credits, 0); + msleep(1000); + wait_retry++; + } + } while (rsize < rdata->bytes && wait_retry < 3); + + /* + * If we can't find enough credits to send this rdata + * release the rdata and return failure, this will pass + * whatever I/O amount we have finished to VFS. + */ + if (rsize < rdata->bytes) { + rc = -EBUSY; + goto out; + } + + rc = -EAGAIN; + while (rc == -EAGAIN) + if (!rdata->cfile->invalidHandle || + !(rc = cifs_reopen_file(rdata->cfile, true))) + rc = server->ops->async_readv(rdata); + + if (!rc) { + /* Add to aio pending list */ + list_add_tail(&rdata->list, rdata_list); + return 0; + } + + add_credits_and_wake_if(server, rdata->credits, 0); +out: + kref_put(&rdata->refcount, + cifs_uncached_readdata_release); + + return rc; +} + static int cifs_send_async_read(loff_t offset, size_t len, struct cifsFileInfo *open_file, struct cifs_sb_info *cifs_sb, struct list_head *rdata_list, @@ -3103,6 +3159,9 @@ cifs_send_async_read(loff_t offset, size_t len, struct cifsFileInfo *open_file, int rc; pid_t pid; struct TCP_Server_Info *server; + struct page **pagevec; + size_t start; + struct iov_iter direct_iov = ctx->iter; server = tlink_tcon(open_file->tlink)->ses->server; @@ -3111,6 +3170,9 @@ cifs_send_async_read(loff_t offset, size_t len, struct cifsFileInfo *open_file, else pid = current->tgid; + if (ctx->direct_io) + iov_iter_advance(&direct_iov, offset - ctx->pos); + do { rc = server->ops->wait_mtu_credits(server, cifs_sb->rsize, &rsize, &credits); @@ -3118,20 +3180,56 @@ cifs_send_async_read(loff_t offset, size_t len, struct cifsFileInfo *open_file, break; cur_len = min_t(const size_t, len, rsize); - npages = DIV_ROUND_UP(cur_len, PAGE_SIZE); - /* allocate a readdata struct */ - rdata = cifs_readdata_alloc(npages, + if (ctx->direct_io) { + + cur_len = iov_iter_get_pages_alloc( + &direct_iov, &pagevec, + cur_len, &start); + if (cur_len < 0) { + cifs_dbg(VFS, + "couldn't get user pages (cur_len=%zd)" + " iter type %d" + " iov_offset %zd count %zd\n", + cur_len, direct_iov.type, direct_iov.iov_offset, + direct_iov.count); + dump_stack(); + break; + } + iov_iter_advance(&direct_iov, cur_len); + + rdata = cifs_readdata_direct_alloc( + pagevec, cifs_uncached_readv_complete); + if (!rdata) { + add_credits_and_wake_if(server, credits, 0); + rc = -ENOMEM; + break; + } + + npages = (cur_len + start + PAGE_SIZE-1) / PAGE_SIZE; + rdata->page_offset = start; + rdata->tailsz = npages > 1 ? + cur_len-(PAGE_SIZE-start)-(npages-2)*PAGE_SIZE : + cur_len; + + } else { + + npages = DIV_ROUND_UP(cur_len, PAGE_SIZE); + /* allocate a readdata struct */ + rdata = cifs_readdata_alloc(npages, cifs_uncached_readv_complete); - if (!rdata) { - add_credits_and_wake_if(server, credits, 0); - rc = -ENOMEM; - break; - } + if (!rdata) { + add_credits_and_wake_if(server, credits, 0); + rc = -ENOMEM; + break; + } - rc = cifs_read_allocate_pages(rdata, npages); - if (rc) - goto error; + rc = cifs_read_allocate_pages(rdata, npages); + if (rc) + goto error; + + rdata->tailsz = PAGE_SIZE; + } rdata->cfile = cifsFileInfo_get(open_file); rdata->nr_pages = npages; @@ -3139,7 +3237,6 @@ cifs_send_async_read(loff_t offset, size_t len, struct cifsFileInfo *open_file, rdata->bytes = cur_len; rdata->pid = pid; rdata->pagesz = PAGE_SIZE; - rdata->tailsz = PAGE_SIZE; rdata->read_into_pages = cifs_uncached_read_into_pages; rdata->copy_into_pages = cifs_uncached_copy_into_pages; rdata->credits = credits; @@ -3153,9 +3250,11 @@ cifs_send_async_read(loff_t offset, size_t len, struct cifsFileInfo *open_file, if (rc) { add_credits_and_wake_if(server, rdata->credits, 0); kref_put(&rdata->refcount, - cifs_uncached_readdata_release); - if (rc == -EAGAIN) + cifs_uncached_readdata_release); + if (rc == -EAGAIN) { + iov_iter_revert(&direct_iov, cur_len); continue; + } break; } @@ -3211,45 +3310,62 @@ collect_uncached_read_data(struct cifs_aio_ctx *ctx) * reading. */ if (got_bytes && got_bytes < rdata->bytes) { - rc = cifs_readdata_to_iov(rdata, to); + rc = 0; + if (!ctx->direct_io) + rc = cifs_readdata_to_iov(rdata, to); if (rc) { kref_put(&rdata->refcount, - cifs_uncached_readdata_release); + cifs_uncached_readdata_release); continue; } } - rc = cifs_send_async_read( + if (ctx->direct_io) { + /* + * Re-use rdata as this is a + * direct I/O + */ + rc = cifs_resend_rdata( + rdata, + &tmp_list, ctx); + } else { + rc = cifs_send_async_read( rdata->offset + got_bytes, rdata->bytes - got_bytes, rdata->cfile, cifs_sb, &tmp_list, ctx); + kref_put(&rdata->refcount, + cifs_uncached_readdata_release); + } + list_splice(&tmp_list, &ctx->list); - kref_put(&rdata->refcount, - cifs_uncached_readdata_release); goto again; } else if (rdata->result) rc = rdata->result; - else + else if (!ctx->direct_io) rc = cifs_readdata_to_iov(rdata, to); /* if there was a short read -- discard anything left */ if (rdata->got_bytes && rdata->got_bytes < rdata->bytes) rc = -ENODATA; + + ctx->total_len += rdata->got_bytes; } list_del_init(&rdata->list); kref_put(&rdata->refcount, cifs_uncached_readdata_release); } - for (i = 0; i < ctx->npages; i++) { - if (ctx->should_dirty) - set_page_dirty(ctx->bv[i].bv_page); - put_page(ctx->bv[i].bv_page); - } + if (!ctx->direct_io) { + for (i = 0; i < ctx->npages; i++) { + if (ctx->should_dirty) + set_page_dirty(ctx->bv[i].bv_page); + put_page(ctx->bv[i].bv_page); + } - ctx->total_len = ctx->len - iov_iter_count(to); + ctx->total_len = ctx->len - iov_iter_count(to); + } cifs_stats_bytes_read(tcon, ctx->total_len); @@ -3267,18 +3383,27 @@ collect_uncached_read_data(struct cifs_aio_ctx *ctx) complete(&ctx->done); } -ssize_t cifs_user_readv(struct kiocb *iocb, struct iov_iter *to) +ssize_t __cifs_readv(struct kiocb *iocb, struct iov_iter *to, bool direct) { - struct file *file = iocb->ki_filp; - ssize_t rc; size_t len; - ssize_t total_read = 0; - loff_t offset = iocb->ki_pos; + struct file *file = iocb->ki_filp; struct cifs_sb_info *cifs_sb; - struct cifs_tcon *tcon; struct cifsFileInfo *cfile; + struct cifs_tcon *tcon; + ssize_t rc, total_read = 0; + loff_t offset = iocb->ki_pos; struct cifs_aio_ctx *ctx; + /* + * iov_iter_get_pages_alloc() doesn't work with ITER_KVEC, + * fall back to data copy read path + * this could be improved by getting pages directly in ITER_KVEC + */ + if (direct && to->type & ITER_KVEC) { + cifs_dbg(FYI, "use non-direct cifs_user_readv for kvec I/O\n"); + direct = false; + } + len = iov_iter_count(to); if (!len) return 0; @@ -3305,14 +3430,20 @@ ssize_t cifs_user_readv(struct kiocb *iocb, struct iov_iter *to) if (to->type == ITER_IOVEC) ctx->should_dirty = true; - rc = setup_aio_ctx_iter(ctx, to, READ); - if (rc) { - kref_put(&ctx->refcount, cifs_aio_ctx_release); - return rc; + if (direct) { + ctx->pos = offset; + ctx->direct_io = true; + ctx->iter = *to; + ctx->len = len; + } else { + rc = setup_aio_ctx_iter(ctx, to, READ); + if (rc) { + kref_put(&ctx->refcount, cifs_aio_ctx_release); + return rc; + } + len = ctx->len; } - len = ctx->len; - /* grab a lock here due to read response handlers can access ctx */ mutex_lock(&ctx->aio_mutex); @@ -3354,6 +3485,16 @@ ssize_t cifs_user_readv(struct kiocb *iocb, struct iov_iter *to) return rc; } +ssize_t cifs_direct_readv(struct kiocb *iocb, struct iov_iter *to) +{ + return __cifs_readv(iocb, to, true); +} + +ssize_t cifs_user_readv(struct kiocb *iocb, struct iov_iter *to) +{ + return __cifs_readv(iocb, to, false); +} + ssize_t cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to) { From patchwork Wed Oct 31 22:13:10 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Long Li X-Patchwork-Id: 991738 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Authentication-Results: ozlabs.org; spf=none (mailfrom) smtp.mailfrom=vger.kernel.org (client-ip=209.132.180.67; helo=vger.kernel.org; envelope-from=linux-cifs-owner@vger.kernel.org; receiver=) Authentication-Results: ozlabs.org; dmarc=none (p=none dis=none) header.from=linuxonhyperv.com Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 42ljbM2dX4z9s0t for ; Thu, 1 Nov 2018 09:25:51 +1100 (AEDT) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1725812AbeKAHZv (ORCPT ); Thu, 1 Nov 2018 03:25:51 -0400 Received: from a2nlsmtp01-04.prod.iad2.secureserver.net ([198.71.225.38]:40652 "EHLO a2nlsmtp01-04.prod.iad2.secureserver.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725755AbeKAHZv (ORCPT ); Thu, 1 Nov 2018 03:25:51 -0400 X-Greylist: delayed 678 seconds by postgrey-1.27 at vger.kernel.org; Thu, 01 Nov 2018 03:25:50 EDT Received: from linuxonhyperv2.linuxonhyperv.com ([107.180.71.197]) by : HOSTING RELAY : with ESMTP id Hyk0gaSwUCT0vHyk0gWfTC; Wed, 31 Oct 2018 15:13:32 -0700 x-originating-ip: 107.180.71.197 Received: from longli by linuxonhyperv2.linuxonhyperv.com with local (Exim 4.91) (envelope-from ) id 1gHyk0-0000ge-DM; Wed, 31 Oct 2018 15:13:32 -0700 From: Long Li To: Steve French , linux-cifs@vger.kernel.org, samba-technical@lists.samba.org, linux-kernel@vger.kernel.org Cc: Long Li Subject: [Patch v4 2/3] CIFS: Add support for direct I/O write Date: Wed, 31 Oct 2018 22:13:10 +0000 Message-Id: <20181031221311.2596-2-longli@linuxonhyperv.com> X-Mailer: git-send-email 2.18.0 In-Reply-To: <20181031221311.2596-1-longli@linuxonhyperv.com> References: <20181031221311.2596-1-longli@linuxonhyperv.com> Reply-To: longli@microsoft.com X-CMAE-Envelope: MS4wfD0nyaf9yJp8tqibuk+c6xuZTH+7qHSuaJlIgvX0CPtkUCddZqdpp4qycklYnWC3ZR+ZdbQbhFp+BmepjRnv0dDpT3KvMv+DXSFEO2IZsO6IHV4qQt+u Dh7d0tOU+mpObRw1HXA29E7aVfkTvGOBId7jz8NaMiFYI1lk3WFHwk3EhBEQd8m/9H3AMexqmc2jExanyWuEUrFzFIWuLzb6HmYZwuOK5uJupGhnW/zfw7Xu xYLfIP8Ai0DtoXYxz5fprqngmQY2r6s2wT2OuWcapxeAfwKcsYVArIJORX9NgcvTiekRk4h4//x9LgyEoD4b5305q5xyohIgnjFt7qJgAmo1gXPdObShSVJZ UvByYx7WovZpKMIKa23PZLIyDBgXgw== Sender: linux-cifs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-cifs@vger.kernel.org From: Long Li With direct I/O write, user supplied buffers are pinned to the memory and data are transferred directly from user buffers to the transport layer. Change in v3: add support for kernel AIO Change in v4: Refactor common write code to __cifs_writev for direct and non-direct I/O. Retry on direct I/O failure. Signed-off-by: Long Li --- fs/cifs/cifsfs.h | 1 + fs/cifs/file.c | 194 +++++++++++++++++++++++++++++++++++++++++++------------ 2 files changed, 154 insertions(+), 41 deletions(-) diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 7fba9aa..e9c5103 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -105,6 +105,7 @@ extern ssize_t cifs_user_readv(struct kiocb *iocb, struct iov_iter *to); extern ssize_t cifs_direct_readv(struct kiocb *iocb, struct iov_iter *to); extern ssize_t cifs_strict_readv(struct kiocb *iocb, struct iov_iter *to); extern ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from); +extern ssize_t cifs_direct_writev(struct kiocb *iocb, struct iov_iter *from); extern ssize_t cifs_strict_writev(struct kiocb *iocb, struct iov_iter *from); extern int cifs_lock(struct file *, int, struct file_lock *); extern int cifs_fsync(struct file *, loff_t, loff_t, int); diff --git a/fs/cifs/file.c b/fs/cifs/file.c index daab878..1a41c04 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -2524,6 +2524,55 @@ wdata_fill_from_iovec(struct cifs_writedata *wdata, struct iov_iter *from, } static int +cifs_resend_wdata(struct cifs_writedata *wdata, struct list_head *wdata_list, struct cifs_aio_ctx *ctx) +{ + int wait_retry = 0; + unsigned int wsize, credits; + int rc; + struct TCP_Server_Info *server = tlink_tcon(wdata->cfile->tlink)->ses->server; + + /* + * Try to resend this wdata, waiting for credits up to 3 seconds. + * Note: we are attempting to resend the whole wdata not in segments + */ + do { + rc = server->ops->wait_mtu_credits(server, wdata->bytes, &wsize, &credits); + + if (rc) + break; + + if (wsize < wdata->bytes) { + add_credits_and_wake_if(server, credits, 0); + msleep(1000); + wait_retry++; + } + } while (wsize < wdata->bytes && wait_retry < 3); + + if (wsize < wdata->bytes) { + rc = -EBUSY; + goto out; + } + + rc = -EAGAIN; + while (rc == -EAGAIN) + if (!wdata->cfile->invalidHandle || + !(rc = cifs_reopen_file(wdata->cfile, false))) + rc = server->ops->async_writev(wdata, + cifs_uncached_writedata_release); + + if (!rc) { + list_add_tail(&wdata->list, wdata_list); + return 0; + } + + add_credits_and_wake_if(server, wdata->credits, 0); +out: + kref_put(&wdata->refcount, cifs_uncached_writedata_release); + + return rc; +} + +static int cifs_write_from_iter(loff_t offset, size_t len, struct iov_iter *from, struct cifsFileInfo *open_file, struct cifs_sb_info *cifs_sb, struct list_head *wdata_list, @@ -2537,6 +2586,8 @@ cifs_write_from_iter(loff_t offset, size_t len, struct iov_iter *from, loff_t saved_offset = offset; pid_t pid; struct TCP_Server_Info *server; + struct page **pagevec; + size_t start; if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_RWPIDFORWARD) pid = open_file->pid; @@ -2553,38 +2604,74 @@ cifs_write_from_iter(loff_t offset, size_t len, struct iov_iter *from, if (rc) break; - nr_pages = get_numpages(wsize, len, &cur_len); - wdata = cifs_writedata_alloc(nr_pages, + if (ctx->direct_io) { + cur_len = iov_iter_get_pages_alloc( + from, &pagevec, wsize, &start); + if (cur_len < 0) { + cifs_dbg(VFS, + "direct_writev couldn't get user pages " + "(rc=%zd) iter type %d iov_offset %zd count" + " %zd\n", + cur_len, from->type, + from->iov_offset, from->count); + dump_stack(); + break; + } + iov_iter_advance(from, cur_len); + + nr_pages = (cur_len + start + PAGE_SIZE - 1) / PAGE_SIZE; + + wdata = cifs_writedata_direct_alloc(pagevec, cifs_uncached_writev_complete); - if (!wdata) { - rc = -ENOMEM; - add_credits_and_wake_if(server, credits, 0); - break; - } + if (!wdata) { + rc = -ENOMEM; + add_credits_and_wake_if(server, credits, 0); + break; + } - rc = cifs_write_allocate_pages(wdata->pages, nr_pages); - if (rc) { - kfree(wdata); - add_credits_and_wake_if(server, credits, 0); - break; - } - num_pages = nr_pages; - rc = wdata_fill_from_iovec(wdata, from, &cur_len, &num_pages); - if (rc) { - for (i = 0; i < nr_pages; i++) - put_page(wdata->pages[i]); - kfree(wdata); - add_credits_and_wake_if(server, credits, 0); - break; - } + wdata->page_offset = start; + wdata->tailsz = + nr_pages > 1 ? + cur_len - (PAGE_SIZE - start) - + (nr_pages - 2) * PAGE_SIZE : + cur_len; + } else { + nr_pages = get_numpages(wsize, len, &cur_len); + wdata = cifs_writedata_alloc(nr_pages, + cifs_uncached_writev_complete); + if (!wdata) { + rc = -ENOMEM; + add_credits_and_wake_if(server, credits, 0); + break; + } - /* - * Bring nr_pages down to the number of pages we actually used, - * and free any pages that we didn't use. - */ - for ( ; nr_pages > num_pages; nr_pages--) - put_page(wdata->pages[nr_pages - 1]); + rc = cifs_write_allocate_pages(wdata->pages, nr_pages); + if (rc) { + kfree(wdata); + add_credits_and_wake_if(server, credits, 0); + break; + } + + num_pages = nr_pages; + rc = wdata_fill_from_iovec(wdata, from, &cur_len, &num_pages); + if (rc) { + for (i = 0; i < nr_pages; i++) + put_page(wdata->pages[i]); + kfree(wdata); + add_credits_and_wake_if(server, credits, 0); + break; + } + + /* + * Bring nr_pages down to the number of pages we actually used, + * and free any pages that we didn't use. + */ + for ( ; nr_pages > num_pages; nr_pages--) + put_page(wdata->pages[nr_pages - 1]); + + wdata->tailsz = cur_len - ((nr_pages - 1) * PAGE_SIZE); + } wdata->sync_mode = WB_SYNC_ALL; wdata->nr_pages = nr_pages; @@ -2593,7 +2680,6 @@ cifs_write_from_iter(loff_t offset, size_t len, struct iov_iter *from, wdata->pid = pid; wdata->bytes = cur_len; wdata->pagesz = PAGE_SIZE; - wdata->tailsz = cur_len - ((nr_pages - 1) * PAGE_SIZE); wdata->credits = credits; wdata->ctx = ctx; kref_get(&ctx->refcount); @@ -2668,13 +2754,17 @@ static void collect_uncached_write_data(struct cifs_aio_ctx *ctx) INIT_LIST_HEAD(&tmp_list); list_del_init(&wdata->list); - iov_iter_advance(&tmp_from, + if (ctx->direct_io) + rc = cifs_resend_wdata(wdata, &tmp_list, ctx); + else { + iov_iter_advance(&tmp_from, wdata->offset - ctx->pos); - rc = cifs_write_from_iter(wdata->offset, + rc = cifs_write_from_iter(wdata->offset, wdata->bytes, &tmp_from, ctx->cfile, cifs_sb, &tmp_list, ctx); + } list_splice(&tmp_list, &ctx->list); @@ -2687,8 +2777,9 @@ static void collect_uncached_write_data(struct cifs_aio_ctx *ctx) kref_put(&wdata->refcount, cifs_uncached_writedata_release); } - for (i = 0; i < ctx->npages; i++) - put_page(ctx->bv[i].bv_page); + if (!ctx->direct_io) + for (i = 0; i < ctx->npages; i++) + put_page(ctx->bv[i].bv_page); cifs_stats_bytes_written(tcon, ctx->total_len); set_bit(CIFS_INO_INVALID_MAPPING, &CIFS_I(dentry->d_inode)->flags); @@ -2703,7 +2794,7 @@ static void collect_uncached_write_data(struct cifs_aio_ctx *ctx) complete(&ctx->done); } -ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from) +static ssize_t __cifs_writev(struct kiocb *iocb, struct iov_iter *from, bool direct) { struct file *file = iocb->ki_filp; ssize_t total_written = 0; @@ -2712,13 +2803,18 @@ ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from) struct cifs_sb_info *cifs_sb; struct cifs_aio_ctx *ctx; struct iov_iter saved_from = *from; + size_t len = iov_iter_count(from); int rc; /* - * BB - optimize the way when signing is disabled. We can drop this - * extra memory-to-memory copying and use iovec buffers for constructing - * write request. + * iov_iter_get_pages_alloc doesn't work with ITER_KVEC. + * In this case, fall back to non-direct write function. + * this could be improved by getting pages directly in ITER_KVEC */ + if (direct && from->type & ITER_KVEC) { + cifs_dbg(FYI, "use non-direct cifs_writev for kvec I/O\n"); + direct = false; + } rc = generic_write_checks(iocb, from); if (rc <= 0) @@ -2742,10 +2838,16 @@ ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from) ctx->pos = iocb->ki_pos; - rc = setup_aio_ctx_iter(ctx, from, WRITE); - if (rc) { - kref_put(&ctx->refcount, cifs_aio_ctx_release); - return rc; + if (direct) { + ctx->direct_io = true; + ctx->iter = *from; + ctx->len = len; + } else { + rc = setup_aio_ctx_iter(ctx, from, WRITE); + if (rc) { + kref_put(&ctx->refcount, cifs_aio_ctx_release); + return rc; + } } /* grab a lock here due to read response handlers can access ctx */ @@ -2795,6 +2897,16 @@ ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from) return total_written; } +ssize_t cifs_direct_writev(struct kiocb *iocb, struct iov_iter *from) +{ + return __cifs_writev(iocb, from, true); +} + +ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from) +{ + return __cifs_writev(iocb, from, false); +} + static ssize_t cifs_writev(struct kiocb *iocb, struct iov_iter *from) { From patchwork Wed Oct 31 22:13:11 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Long Li X-Patchwork-Id: 991736 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Authentication-Results: ozlabs.org; spf=none (mailfrom) smtp.mailfrom=vger.kernel.org (client-ip=209.132.180.67; helo=vger.kernel.org; envelope-from=linux-cifs-owner@vger.kernel.org; receiver=) Authentication-Results: ozlabs.org; dmarc=none (p=none dis=none) header.from=linuxonhyperv.com Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id 42ljLT4TYZz9sBQ for ; Thu, 1 Nov 2018 09:14:41 +1100 (AEDT) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726897AbeKAHOf (ORCPT ); Thu, 1 Nov 2018 03:14:35 -0400 Received: from a2nlsmtp01-05.prod.iad2.secureserver.net ([198.71.225.49]:46978 "EHLO a2nlsmtp01-05.prod.iad2.secureserver.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725755AbeKAHOf (ORCPT ); Thu, 1 Nov 2018 03:14:35 -0400 Received: from linuxonhyperv2.linuxonhyperv.com ([107.180.71.197]) by : HOSTING RELAY : with ESMTP id Hyk2g4ZDjbOGjHyk2gzy2j; Wed, 31 Oct 2018 15:13:34 -0700 x-originating-ip: 107.180.71.197 Received: from longli by linuxonhyperv2.linuxonhyperv.com with local (Exim 4.91) (envelope-from ) id 1gHyk1-0000gk-VS; Wed, 31 Oct 2018 15:13:34 -0700 From: Long Li To: Steve French , linux-cifs@vger.kernel.org, samba-technical@lists.samba.org, linux-kernel@vger.kernel.org Cc: Long Li Subject: [Patch v4 3/3] CIFS: Add direct I/O functions to file_operations Date: Wed, 31 Oct 2018 22:13:11 +0000 Message-Id: <20181031221311.2596-3-longli@linuxonhyperv.com> X-Mailer: git-send-email 2.18.0 In-Reply-To: <20181031221311.2596-1-longli@linuxonhyperv.com> References: <20181031221311.2596-1-longli@linuxonhyperv.com> Reply-To: longli@microsoft.com X-CMAE-Envelope: MS4wfGEI+e0OWB/Cv8nU4vELvGFJ0VfqsDNujUeSkcNDOBjrLEpJQPs6PPQIYRmdIWBmrhOe/cQ7BU7OXLSrwTRidz+0T9IIKxL3r7QEwOp2tAZ4ZbXkcMNl A+33LDBTREt0vZzB5TPQ2z+3rMqbKzItg4LrgZrO7WJ756IcomaWIXD8034SRmAwZ6EpFB0P0Zt09cGf2JRqGszeMinTZKHGgIshmP0FzDTlz78s5eP3BqrQ 5fe/1hxvqyNtt4pPwTPOgBkekb6K8MZ9DjZCKUYFYboF7vvWTBNQ+DkVI7y70zEXxs98X8ByThRu5u6CKtioWOEZa6QUtxYmfDcBj5x9ePFwER6VgPnO6gL4 /VObBeSZraG2/cDVBvzE/n1Lynigsw== Sender: linux-cifs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-cifs@vger.kernel.org From: Long Li With direct read/write functions implemented, add them to file_operations. Dircet I/O is used under two conditions: 1. When mounting with "cache=none", CIFS uses direct I/O for all user file data transfer. 2. When opening a file with O_DIRECT, CIFS uses direct I/O for all data transfer on this file. Signed-off-by: Long Li --- fs/cifs/cifsfs.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 62f1662..f18091b 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -1113,9 +1113,8 @@ const struct file_operations cifs_file_strict_ops = { }; const struct file_operations cifs_file_direct_ops = { - /* BB reevaluate whether they can be done with directio, no cache */ - .read_iter = cifs_user_readv, - .write_iter = cifs_user_writev, + .read_iter = cifs_direct_readv, + .write_iter = cifs_direct_writev, .open = cifs_open, .release = cifs_close, .lock = cifs_lock, @@ -1169,9 +1168,8 @@ const struct file_operations cifs_file_strict_nobrl_ops = { }; const struct file_operations cifs_file_direct_nobrl_ops = { - /* BB reevaluate whether they can be done with directio, no cache */ - .read_iter = cifs_user_readv, - .write_iter = cifs_user_writev, + .read_iter = cifs_direct_readv, + .write_iter = cifs_direct_writev, .open = cifs_open, .release = cifs_close, .fsync = cifs_fsync,