From patchwork Mon May 16 20:36:20 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Schumaker, Anna" X-Patchwork-Id: 12851428 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 05ABCC4332F for ; Mon, 16 May 2022 21:01:44 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1348605AbiEPVBk (ORCPT ); Mon, 16 May 2022 17:01:40 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45264 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1349201AbiEPVBI (ORCPT ); Mon, 16 May 2022 17:01:08 -0400 Received: from dfw.source.kernel.org (dfw.source.kernel.org [139.178.84.217]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 3055BB858 for ; Mon, 16 May 2022 13:36:26 -0700 (PDT) Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by dfw.source.kernel.org (Postfix) with ESMTPS id A08ED614B6 for ; Mon, 16 May 2022 20:36:25 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id C1BA9C34119; Mon, 16 May 2022 20:36:24 +0000 (UTC) From: Anna.Schumaker@Netapp.com To: steved@redhat.com, linux-nfs@vger.kernel.org Cc: Anna.Schumaker@Netapp.com Subject: [PATCH v1 3/5] NFS: Replace the READ_PLUS decoding code Date: Mon, 16 May 2022 16:36:20 -0400 Message-Id: <20220516203622.2605713-4-Anna.Schumaker@Netapp.com> X-Mailer: git-send-email 2.36.1 In-Reply-To: <20220516203622.2605713-1-Anna.Schumaker@Netapp.com> References: <20220516203622.2605713-1-Anna.Schumaker@Netapp.com> MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org From: Anna Schumaker We now take a 2-step process that allows us to place data and hole segments directly at their final position in the xdr_stream without needing to do a bunch of redundant copies to expand holes. Due to the variable lengths of each segment, the xdr metadata might cross page boundaries which I account for by setting a small scratch buffer so xdr_inline_decode() won't fail. Signed-off-by: Anna Schumaker --- fs/nfs/nfs42xdr.c | 168 ++++++++++++++++++++++++---------------------- 1 file changed, 87 insertions(+), 81 deletions(-) diff --git a/fs/nfs/nfs42xdr.c b/fs/nfs/nfs42xdr.c index 271e5f92ed01..5aedc4bf5313 100644 --- a/fs/nfs/nfs42xdr.c +++ b/fs/nfs/nfs42xdr.c @@ -1025,73 +1025,84 @@ static int decode_deallocate(struct xdr_stream *xdr, struct nfs42_falloc_res *re return decode_op_hdr(xdr, OP_DEALLOCATE); } -static int decode_read_plus_data(struct xdr_stream *xdr, - struct nfs_pgio_args *args, - struct nfs_pgio_res *res) -{ - uint32_t count, recvd; +struct read_plus_segment { + enum data_content4 type; uint64_t offset; + union { + struct { + uint64_t length; + } hole; + + struct { + uint32_t length; + unsigned int from; + } data; + }; +}; + +static inline uint64_t read_plus_segment_length(struct read_plus_segment *seg) +{ + return seg->type == NFS4_CONTENT_DATA ? seg->data.length : seg->hole.length; +} + +static int decode_read_plus_segment(struct xdr_stream *xdr, + struct read_plus_segment *seg) +{ __be32 *p; - p = xdr_inline_decode(xdr, 8 + 4); + p = xdr_inline_decode(xdr, 4); if (!p) - return 1; - - p = xdr_decode_hyper(p, &offset); - count = be32_to_cpup(p); - recvd = xdr_align_data(xdr, res->count, xdr_align_size(count)); - if (recvd > count) - recvd = count; - if (res->count + recvd > args->count) { - if (args->count > res->count) - res->count += args->count - res->count; - return 1; - } - res->count += recvd; - if (count > recvd) - return 1; + return -EIO; + seg->type = be32_to_cpup(p++); + + p = xdr_inline_decode(xdr, seg->type == NFS4_CONTENT_DATA ? 12 : 16); + if (!p) + return -EIO; + p = xdr_decode_hyper(p, &seg->offset); + + if (seg->type == NFS4_CONTENT_DATA) { + struct xdr_buf buf; + uint32_t len = be32_to_cpup(p); + + seg->data.length = len; + seg->data.from = xdr_stream_pos(xdr); + + if (!xdr_stream_subsegment(xdr, &buf, xdr_align_size(len))) + return -EIO; + } else if (seg->type == NFS4_CONTENT_HOLE) { + xdr_decode_hyper(p, &seg->hole.length); + } else + return -EINVAL; return 0; } -static int decode_read_plus_hole(struct xdr_stream *xdr, - struct nfs_pgio_args *args, - struct nfs_pgio_res *res, uint32_t *eof) +static int process_read_plus_segment(struct xdr_stream *xdr, + struct nfs_pgio_args *args, + struct nfs_pgio_res *res, + struct read_plus_segment *seg) { - uint64_t offset, length, recvd; - __be32 *p; + unsigned long offset = seg->offset; + unsigned long length = read_plus_segment_length(seg); + unsigned int bufpos; - p = xdr_inline_decode(xdr, 8 + 8); - if (!p) - return 1; - - p = xdr_decode_hyper(p, &offset); - p = xdr_decode_hyper(p, &length); - if (offset != args->offset + res->count) { - /* Server returned an out-of-sequence extent */ - if (offset > args->offset + res->count || - offset + length < args->offset + res->count) { - dprintk("NFS: server returned out of sequence extent: " - "offset/size = %llu/%llu != expected %llu\n", - (unsigned long long)offset, - (unsigned long long)length, - (unsigned long long)(args->offset + - res->count)); - return 1; - } - length -= args->offset + res->count - offset; + if (offset + length < args->offset) + return 0; + else if (offset > args->offset + args->count) { + res->eof = 0; + return 0; + } else if (offset < args->offset) { + length -= (args->offset - offset); + offset = args->offset; + } else if (offset + length > args->offset + args->count) { + length = (args->offset + args->count) - offset; + res->eof = 0; } - if (length + res->count > args->count) { - *eof = 0; - if (unlikely(res->count >= args->count)) - return 1; - length = args->count - res->count; - } - recvd = xdr_expand_hole(xdr, res->count, length); - res->count += recvd; - if (recvd < length) - return 1; - return 0; + bufpos = xdr->buf->head[0].iov_len + (offset - args->offset); + if (seg->type == NFS4_CONTENT_HOLE) + return xdr_stream_zero(xdr, bufpos, length); + else + return xdr_stream_move_segment(xdr, seg->data.from, bufpos, length); } static int decode_read_plus(struct xdr_stream *xdr, struct nfs_pgio_res *res) @@ -1099,8 +1110,10 @@ static int decode_read_plus(struct xdr_stream *xdr, struct nfs_pgio_res *res) struct nfs_pgio_header *hdr = container_of(res, struct nfs_pgio_header, res); struct nfs_pgio_args *args = &hdr->args; - uint32_t eof, segments, type; + uint32_t segments; + struct read_plus_segment *segs; int status, i; + char scratch_buf[16]; __be32 *p; status = decode_op_hdr(xdr, OP_READ_PLUS); @@ -1112,38 +1125,31 @@ static int decode_read_plus(struct xdr_stream *xdr, struct nfs_pgio_res *res) return -EIO; res->count = 0; - eof = be32_to_cpup(p++); + res->eof = be32_to_cpup(p++); segments = be32_to_cpup(p++); if (segments == 0) - goto out; + return status; + segs = kmalloc_array(segments, sizeof(*segs), GFP_KERNEL); + if (!segs) + return -ENOMEM; + + xdr_set_scratch_buffer(xdr, &scratch_buf, 32); + status = -EIO; for (i = 0; i < segments; i++) { - p = xdr_inline_decode(xdr, 4); - if (!p) - goto early_out; - - type = be32_to_cpup(p++); - if (type == NFS4_CONTENT_DATA) - status = decode_read_plus_data(xdr, args, res); - else if (type == NFS4_CONTENT_HOLE) - status = decode_read_plus_hole(xdr, args, res, &eof); - else - return -EINVAL; - + status = decode_read_plus_segment(xdr, &segs[i]); if (status < 0) - return status; - if (status > 0) - goto early_out; + goto out; } + xdr_set_pagelen(xdr, xdr_align_size(args->count)); + for (i = segments; i > 0; i--) + res->count += process_read_plus_segment(xdr, args, res, &segs[i-1]); + status = 0; + out: - res->eof = eof; - return 0; -early_out: - if (unlikely(!i)) - return -EIO; - res->eof = 0; - return 0; + kfree(segs); + return status; } static int decode_seek(struct xdr_stream *xdr, struct nfs42_seek_res *res)