From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from zimbra13.linbit.com (zimbra.linbit.com [212.69.161.123]) by mail09.linbit.com (LINBIT Mail Daemon) with ESMTP id EFC9B10583BF for ; Fri, 8 Nov 2013 15:59:21 +0100 (CET) Received: from localhost (localhost [127.0.0.1]) by zimbra13.linbit.com (Postfix) with ESMTP id C93B644A2C3 for ; Fri, 8 Nov 2013 15:59:21 +0100 (CET) Received: from zimbra13.linbit.com ([127.0.0.1]) by localhost (zimbra13.linbit.com [127.0.0.1]) (amavisd-new, port 10032) with ESMTP id IzawxkhSw38j for ; Fri, 8 Nov 2013 15:59:21 +0100 (CET) Received: from localhost (localhost [127.0.0.1]) by zimbra13.linbit.com (Postfix) with ESMTP id 8DFDE44A3C9 for ; Fri, 8 Nov 2013 15:59:21 +0100 (CET) Received: from zimbra13.linbit.com ([127.0.0.1]) by localhost (zimbra13.linbit.com [127.0.0.1]) (amavisd-new, port 10026) with ESMTP id X5GgHr3Xh57b for ; Fri, 8 Nov 2013 15:59:21 +0100 (CET) Received: from soda.linbit (tuerlsteher.linbit.com [86.59.100.100]) by zimbra13.linbit.com (Postfix) with ESMTP id 71C0C44A2C3 for ; Fri, 8 Nov 2013 15:59:21 +0100 (CET) Resent-Message-ID: <20131108145921.GH4144@soda.linbit> Received: from mail-pa0-f49.google.com (mail-pa0-f49.google.com [209.85.220.49]) (using TLSv1 with cipher RC4-SHA (128/128 bits)) (No client certificate requested) by mail09.linbit.com (LINBIT Mail Daemon) with ESMTPS id A3FB910578A3 for ; Tue, 29 Oct 2013 21:24:26 +0100 (CET) Received: by mail-pa0-f49.google.com with SMTP id lj1so552931pab.22 for ; Tue, 29 Oct 2013 13:24:24 -0700 (PDT) From: Kent Overstreet To: axboe@kernel.dk Date: Tue, 29 Oct 2013 13:18:01 -0700 Message-Id: <1383077896-4132-9-git-send-email-kmo@daterainc.com> In-Reply-To: <1383077896-4132-1-git-send-email-kmo@daterainc.com> References: <1383077896-4132-1-git-send-email-kmo@daterainc.com> Cc: nbd-general@lists.sourceforge.net, Paul Clements , linux-kernel@vger.kernel.org, hch@infradead.org, Kent Overstreet , tj@kernel.org, drbd-user@lists.linbit.com, Lars Ellenberg Subject: [Drbd-dev] [PATCH 08/23] block: Immutable bio vecs List-Id: "*Coordination* of development, patches, contributions -- *Questions* \(even to developers\) go to drbd-user, please." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , This adds a mechanism by which we can advance a bio by an arbitrary number of bytes without modifying the biovec: bio->bi_iter.bi_bvec_done indicates the number of bytes completed in the current bvec. Various driver code still needs to be updated to not refer to the bvec directly before we can use this for interesting things, like efficient bio splitting. Signed-off-by: Kent Overstreet Cc: Jens Axboe Cc: Lars Ellenberg Cc: Paul Clements Cc: drbd-user@lists.linbit.com Cc: nbd-general@lists.sourceforge.net --- drivers/block/drbd/drbd_main.c | 4 +-- drivers/block/nbd.c | 2 +- fs/bio.c | 27 ++------------ include/linux/bio.h | 81 +++++++++++++++++++++++++++++++++++++----- include/linux/blk_types.h | 2 ++ include/linux/blkdev.h | 4 +-- 6 files changed, 82 insertions(+), 38 deletions(-) diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c index 1589ea4..fee8b51 100644 --- a/drivers/block/drbd/drbd_main.c +++ b/drivers/block/drbd/drbd_main.c @@ -1546,7 +1546,7 @@ static int _drbd_send_bio(struct drbd_conf *mdev, struct bio *bio) err = _drbd_no_send_page(mdev, bvec.bv_page, bvec.bv_offset, bvec.bv_len, - bio_iter_last(bio, iter) + bio_iter_last(bvec, iter) ? 0 : MSG_MORE); if (err) return err; @@ -1565,7 +1565,7 @@ static int _drbd_send_zc_bio(struct drbd_conf *mdev, struct bio *bio) err = _drbd_send_page(mdev, bvec.bv_page, bvec.bv_offset, bvec.bv_len, - bio_iter_last(bio, iter) ? 0 : MSG_MORE); + bio_iter_last(bvec, iter) ? 0 : MSG_MORE); if (err) return err; } diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index aa362f4..55298db 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -278,7 +278,7 @@ static int nbd_send_req(struct nbd_device *nbd, struct request *req) */ rq_for_each_segment(bvec, req, iter) { flags = 0; - if (!rq_iter_last(req, iter)) + if (!rq_iter_last(bvec, iter)) flags = MSG_MORE; dprintk(DBG_TX, "%s: request %p: sending %d bytes data\n", nbd->disk->disk_name, req, bvec.bv_len); diff --git a/fs/bio.c b/fs/bio.c index eca05c7..b39436a 100644 --- a/fs/bio.c +++ b/fs/bio.c @@ -532,13 +532,11 @@ void __bio_clone(struct bio *bio, struct bio *bio_src) * most users will be overriding ->bi_bdev with a new target, * so we don't set nor calculate new physical/hw segment counts here */ - bio->bi_iter.bi_sector = bio_src->bi_iter.bi_sector; bio->bi_bdev = bio_src->bi_bdev; bio->bi_flags |= 1 << BIO_CLONED; bio->bi_rw = bio_src->bi_rw; bio->bi_vcnt = bio_src->bi_vcnt; - bio->bi_iter.bi_size = bio_src->bi_iter.bi_size; - bio->bi_iter.bi_idx = bio_src->bi_iter.bi_idx; + bio->bi_iter = bio_src->bi_iter; } EXPORT_SYMBOL(__bio_clone); @@ -808,28 +806,7 @@ void bio_advance(struct bio *bio, unsigned bytes) if (bio_integrity(bio)) bio_integrity_advance(bio, bytes); - bio->bi_iter.bi_sector += bytes >> 9; - bio->bi_iter.bi_size -= bytes; - - if (bio->bi_rw & BIO_NO_ADVANCE_ITER_MASK) - return; - - while (bytes) { - if (unlikely(bio->bi_iter.bi_idx >= bio->bi_vcnt)) { - WARN_ONCE(1, "bio idx %d >= vcnt %d\n", - bio->bi_iter.bi_idx, bio->bi_vcnt); - break; - } - - if (bytes >= bio_iovec(bio).bv_len) { - bytes -= bio_iovec(bio).bv_len; - bio->bi_iter.bi_idx++; - } else { - bio_iovec(bio).bv_len -= bytes; - bio_iovec(bio).bv_offset += bytes; - bytes = 0; - } - } + bio_advance_iter(bio, &bio->bi_iter, bytes); } EXPORT_SYMBOL(bio_advance); diff --git a/include/linux/bio.h b/include/linux/bio.h index 5724feb..151868e 100644 --- a/include/linux/bio.h +++ b/include/linux/bio.h @@ -64,11 +64,38 @@ #define bio_iovec_idx(bio, idx) (&((bio)->bi_io_vec[(idx)])) #define __bio_iovec(bio) bio_iovec_idx((bio), (bio)->bi_iter.bi_idx) -#define bio_iter_iovec(bio, iter) ((bio)->bi_io_vec[(iter).bi_idx]) +#define __bvec_iter_bvec(bvec, iter) (&(bvec)[(iter).bi_idx]) -#define bio_page(bio) (bio_iovec((bio)).bv_page) -#define bio_offset(bio) (bio_iovec((bio)).bv_offset) -#define bio_iovec(bio) (*__bio_iovec(bio)) +#define bvec_iter_page(bvec, iter) \ + (__bvec_iter_bvec((bvec), (iter))->bv_page) + +#define bvec_iter_len(bvec, iter) \ + min((iter).bi_size, \ + __bvec_iter_bvec((bvec), (iter))->bv_len - (iter).bi_bvec_done) + +#define bvec_iter_offset(bvec, iter) \ + (__bvec_iter_bvec((bvec), (iter))->bv_offset + (iter).bi_bvec_done) + +#define bvec_iter_bvec(bvec, iter) \ +((struct bio_vec) { \ + .bv_page = bvec_iter_page((bvec), (iter)), \ + .bv_len = bvec_iter_len((bvec), (iter)), \ + .bv_offset = bvec_iter_offset((bvec), (iter)), \ +}) + +#define bio_iter_iovec(bio, iter) \ + bvec_iter_bvec((bio)->bi_io_vec, (iter)) + +#define bio_iter_page(bio, iter) \ + bvec_iter_page((bio)->bi_io_vec, (iter)) +#define bio_iter_len(bio, iter) \ + bvec_iter_len((bio)->bi_io_vec, (iter)) +#define bio_iter_offset(bio, iter) \ + bvec_iter_offset((bio)->bi_io_vec, (iter)) + +#define bio_page(bio) bio_iter_page((bio), (bio)->bi_iter) +#define bio_offset(bio) bio_iter_offset((bio), (bio)->bi_iter) +#define bio_iovec(bio) bio_iter_iovec((bio), (bio)->bi_iter) #define bio_segments(bio) ((bio)->bi_vcnt - (bio)->bi_iter.bi_idx) #define bio_sectors(bio) ((bio)->bi_iter.bi_size >> 9) @@ -145,16 +172,54 @@ static inline void *bio_data(struct bio *bio) bvl = bio_iovec_idx((bio), (i)), i < (bio)->bi_vcnt; \ i++) +static inline void bvec_iter_advance(struct bio_vec *bv, struct bvec_iter *iter, + unsigned bytes) +{ + WARN_ONCE(bytes > iter->bi_size, + "Attempted to advance past end of bvec iter\n"); + + while (bytes) { + unsigned len = min(bytes, bvec_iter_len(bv, *iter)); + + bytes -= len; + iter->bi_size -= len; + iter->bi_bvec_done += len; + + if (iter->bi_bvec_done == __bvec_iter_bvec(bv, *iter)->bv_len) { + iter->bi_bvec_done = 0; + iter->bi_idx++; + } + } +} + +#define for_each_bvec(bvl, bio_vec, iter, start) \ + for ((iter) = start; \ + (bvl) = bvec_iter_bvec((bio_vec), (iter)), \ + (iter).bi_size; \ + bvec_iter_advance((bio_vec), &(iter), (bvl).bv_len)) + + +static inline void bio_advance_iter(struct bio *bio, struct bvec_iter *iter, + unsigned bytes) +{ + iter->bi_sector += bytes >> 9; + + if (bio->bi_rw & BIO_NO_ADVANCE_ITER_MASK) + iter->bi_size -= bytes; + else + bvec_iter_advance(bio->bi_io_vec, iter, bytes); +} + #define __bio_for_each_segment(bvl, bio, iter, start) \ for (iter = (start); \ - bvl = bio_iter_iovec((bio), (iter)), \ - (iter).bi_idx < (bio)->bi_vcnt; \ - (iter).bi_idx++) + (iter).bi_size && \ + ((bvl = bio_iter_iovec((bio), (iter))), 1); \ + bio_advance_iter((bio), &(iter), (bvl).bv_len)) #define bio_for_each_segment(bvl, bio, iter) \ __bio_for_each_segment(bvl, bio, iter, (bio)->bi_iter) -#define bio_iter_last(bio, iter) ((iter).bi_idx == (bio)->bi_vcnt - 1) +#define bio_iter_last(bvec, iter) ((iter).bi_size == (bvec).bv_len) /* * get a reference to a bio, so it won't disappear. the intended use is diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index d46e8a6..72f1274 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -34,6 +34,8 @@ struct bvec_iter { unsigned int bi_size; /* residual I/O count */ unsigned int bi_idx; /* current index into bvl_vec */ + + unsigned int bi_bvec_done; }; /* diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index a436249..9874af4 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -714,9 +714,9 @@ struct req_iterator { __rq_for_each_bio(_iter.bio, _rq) \ bio_for_each_segment(bvl, _iter.bio, _iter.iter) -#define rq_iter_last(rq, _iter) \ +#define rq_iter_last(bvec, _iter) \ (_iter.bio->bi_next == NULL && \ - bio_iter_last(_iter.bio, _iter.iter)) + bio_iter_last(bvec, _iter.iter)) #ifndef ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE # error "You should define ARCH_IMPLEMENTS_FLUSH_DCACHE_PAGE for your platform" -- 1.8.4.rc3