diff options
Diffstat (limited to 'fs/netfs')
| -rw-r--r-- | fs/netfs/Kconfig | 18 | ||||
| -rw-r--r-- | fs/netfs/Makefile | 10 | ||||
| -rw-r--r-- | fs/netfs/buffered_read.c | 623 | ||||
| -rw-r--r-- | fs/netfs/buffered_write.c | 1175 | ||||
| -rw-r--r-- | fs/netfs/direct_read.c | 166 | ||||
| -rw-r--r-- | fs/netfs/direct_write.c | 74 | ||||
| -rw-r--r-- | fs/netfs/fscache_cache.c | 2 | ||||
| -rw-r--r-- | fs/netfs/fscache_cookie.c | 6 | ||||
| -rw-r--r-- | fs/netfs/fscache_io.c | 15 | ||||
| -rw-r--r-- | fs/netfs/fscache_main.c | 1 | ||||
| -rw-r--r-- | fs/netfs/fscache_volume.c | 17 | ||||
| -rw-r--r-- | fs/netfs/internal.h | 117 | ||||
| -rw-r--r-- | fs/netfs/io.c | 787 | ||||
| -rw-r--r-- | fs/netfs/iterator.c | 50 | ||||
| -rw-r--r-- | fs/netfs/locking.c | 25 | ||||
| -rw-r--r-- | fs/netfs/main.c | 68 | ||||
| -rw-r--r-- | fs/netfs/misc.c | 219 | ||||
| -rw-r--r-- | fs/netfs/objects.c | 113 | ||||
| -rw-r--r-- | fs/netfs/output.c | 478 | ||||
| -rw-r--r-- | fs/netfs/read_collect.c | 706 | ||||
| -rw-r--r-- | fs/netfs/read_pgpriv2.c | 227 | ||||
| -rw-r--r-- | fs/netfs/read_retry.c | 313 | ||||
| -rw-r--r-- | fs/netfs/read_single.c | 195 | ||||
| -rw-r--r-- | fs/netfs/rolling_buffer.c | 222 | ||||
| -rw-r--r-- | fs/netfs/stats.c | 49 | ||||
| -rw-r--r-- | fs/netfs/write_collect.c | 550 | ||||
| -rw-r--r-- | fs/netfs/write_issue.c | 927 | ||||
| -rw-r--r-- | fs/netfs/write_retry.c | 234 |
28 files changed, 4734 insertions, 2653 deletions
diff --git a/fs/netfs/Kconfig b/fs/netfs/Kconfig index bec805e0c44c..7701c037c328 100644 --- a/fs/netfs/Kconfig +++ b/fs/netfs/Kconfig @@ -22,6 +22,14 @@ config NETFS_STATS between CPUs. On the other hand, the stats are very useful for debugging purposes. Saying 'Y' here is recommended. +config NETFS_DEBUG + bool "Enable dynamic debugging netfslib and FS-Cache" + depends on NETFS_SUPPORT + help + This permits debugging to be dynamically enabled in the local caching + management module. If this is set, the debugging output may be + enabled by setting bits in /sys/module/netfs/parameters/debug. + config FSCACHE bool "General filesystem local caching manager" depends on NETFS_SUPPORT @@ -50,13 +58,3 @@ config FSCACHE_STATS debugging purposes. Saying 'Y' here is recommended. See Documentation/filesystems/caching/fscache.rst for more information. - -config FSCACHE_DEBUG - bool "Debug FS-Cache" - depends on FSCACHE - help - This permits debugging to be dynamically enabled in the local caching - management module. If this is set, the debugging output may be - enabled by setting bits in /sys/modules/fscache/parameter/debug. - - See Documentation/filesystems/caching/fscache.rst for more information. diff --git a/fs/netfs/Makefile b/fs/netfs/Makefile index d4d1d799819e..b43188d64bd8 100644 --- a/fs/netfs/Makefile +++ b/fs/netfs/Makefile @@ -5,13 +5,19 @@ netfs-y := \ buffered_write.o \ direct_read.o \ direct_write.o \ - io.o \ iterator.o \ locking.o \ main.o \ misc.o \ objects.o \ - output.o + read_collect.o \ + read_pgpriv2.o \ + read_retry.o \ + read_single.o \ + rolling_buffer.o \ + write_collect.o \ + write_issue.o \ + write_retry.o netfs-$(CONFIG_NETFS_STATS) += stats.o diff --git a/fs/netfs/buffered_read.c b/fs/netfs/buffered_read.c index 3298c29b5548..0d1b6d35ff3b 100644 --- a/fs/netfs/buffered_read.c +++ b/fs/netfs/buffered_read.c @@ -9,114 +9,10 @@ #include <linux/task_io_accounting_ops.h> #include "internal.h" -/* - * Unlock the folios in a read operation. We need to set PG_fscache on any - * folios we're going to write back before we unlock them. - */ -void netfs_rreq_unlock_folios(struct netfs_io_request *rreq) -{ - struct netfs_io_subrequest *subreq; - struct netfs_folio *finfo; - struct folio *folio; - pgoff_t start_page = rreq->start / PAGE_SIZE; - pgoff_t last_page = ((rreq->start + rreq->len) / PAGE_SIZE) - 1; - size_t account = 0; - bool subreq_failed = false; - - XA_STATE(xas, &rreq->mapping->i_pages, start_page); - - if (test_bit(NETFS_RREQ_FAILED, &rreq->flags)) { - __clear_bit(NETFS_RREQ_COPY_TO_CACHE, &rreq->flags); - list_for_each_entry(subreq, &rreq->subrequests, rreq_link) { - __clear_bit(NETFS_SREQ_COPY_TO_CACHE, &subreq->flags); - } - } - - /* Walk through the pagecache and the I/O request lists simultaneously. - * We may have a mixture of cached and uncached sections and we only - * really want to write out the uncached sections. This is slightly - * complicated by the possibility that we might have huge pages with a - * mixture inside. - */ - subreq = list_first_entry(&rreq->subrequests, - struct netfs_io_subrequest, rreq_link); - subreq_failed = (subreq->error < 0); - - trace_netfs_rreq(rreq, netfs_rreq_trace_unlock); - - rcu_read_lock(); - xas_for_each(&xas, folio, last_page) { - loff_t pg_end; - bool pg_failed = false; - bool folio_started; - - if (xas_retry(&xas, folio)) - continue; - - pg_end = folio_pos(folio) + folio_size(folio) - 1; - - folio_started = false; - for (;;) { - loff_t sreq_end; - - if (!subreq) { - pg_failed = true; - break; - } - if (!folio_started && test_bit(NETFS_SREQ_COPY_TO_CACHE, &subreq->flags)) { - trace_netfs_folio(folio, netfs_folio_trace_copy_to_cache); - folio_start_fscache(folio); - folio_started = true; - } - pg_failed |= subreq_failed; - sreq_end = subreq->start + subreq->len - 1; - if (pg_end < sreq_end) - break; - - account += subreq->transferred; - if (!list_is_last(&subreq->rreq_link, &rreq->subrequests)) { - subreq = list_next_entry(subreq, rreq_link); - subreq_failed = (subreq->error < 0); - } else { - subreq = NULL; - subreq_failed = false; - } - - if (pg_end == sreq_end) - break; - } - - if (!pg_failed) { - flush_dcache_folio(folio); - finfo = netfs_folio_info(folio); - if (finfo) { - trace_netfs_folio(folio, netfs_folio_trace_filled_gaps); - if (finfo->netfs_group) - folio_change_private(folio, finfo->netfs_group); - else - folio_detach_private(folio); - kfree(finfo); - } - folio_mark_uptodate(folio); - } - - if (!test_bit(NETFS_RREQ_DONT_UNLOCK_FOLIOS, &rreq->flags)) { - if (folio->index == rreq->no_unlock_folio && - test_bit(NETFS_RREQ_NO_UNLOCK_FOLIO, &rreq->flags)) - _debug("no unlock"); - else - folio_unlock(folio); - } - } - rcu_read_unlock(); - - task_io_account_read(account); - if (rreq->netfs_ops->done) - rreq->netfs_ops->done(rreq); -} - static void netfs_cache_expand_readahead(struct netfs_io_request *rreq, - loff_t *_start, size_t *_len, loff_t i_size) + unsigned long long *_start, + unsigned long long *_len, + unsigned long long i_size) { struct netfs_cache_resources *cres = &rreq->cache_resources; @@ -167,6 +63,262 @@ static int netfs_begin_cache_read(struct netfs_io_request *rreq, struct netfs_in return fscache_begin_read_operation(&rreq->cache_resources, netfs_i_cookie(ctx)); } +/* + * netfs_prepare_read_iterator - Prepare the subreq iterator for I/O + * @subreq: The subrequest to be set up + * + * Prepare the I/O iterator representing the read buffer on a subrequest for + * the filesystem to use for I/O (it can be passed directly to a socket). This + * is intended to be called from the ->issue_read() method once the filesystem + * has trimmed the request to the size it wants. + * + * Returns the limited size if successful and -ENOMEM if insufficient memory + * available. + * + * [!] NOTE: This must be run in the same thread as ->issue_read() was called + * in as we access the readahead_control struct. + */ +static ssize_t netfs_prepare_read_iterator(struct netfs_io_subrequest *subreq) +{ + struct netfs_io_request *rreq = subreq->rreq; + size_t rsize = subreq->len; + + if (subreq->source == NETFS_DOWNLOAD_FROM_SERVER) + rsize = umin(rsize, rreq->io_streams[0].sreq_max_len); + + if (rreq->ractl) { + /* If we don't have sufficient folios in the rolling buffer, + * extract a folioq's worth from the readahead region at a time + * into the buffer. Note that this acquires a ref on each page + * that we will need to release later - but we don't want to do + * that until after we've started the I/O. + */ + struct folio_batch put_batch; + + folio_batch_init(&put_batch); + while (rreq->submitted < subreq->start + rsize) { + ssize_t added; + + added = rolling_buffer_load_from_ra(&rreq->buffer, rreq->ractl, + &put_batch); + if (added < 0) + return added; + rreq->submitted += added; + } + folio_batch_release(&put_batch); + } + + subreq->len = rsize; + if (unlikely(rreq->io_streams[0].sreq_max_segs)) { + size_t limit = netfs_limit_iter(&rreq->buffer.iter, 0, rsize, + rreq->io_streams[0].sreq_max_segs); + + if (limit < rsize) { + subreq->len = limit; + trace_netfs_sreq(subreq, netfs_sreq_trace_limited); + } + } + + subreq->io_iter = rreq->buffer.iter; + + iov_iter_truncate(&subreq->io_iter, subreq->len); + rolling_buffer_advance(&rreq->buffer, subreq->len); + return subreq->len; +} + +static enum netfs_io_source netfs_cache_prepare_read(struct netfs_io_request *rreq, + struct netfs_io_subrequest *subreq, + loff_t i_size) +{ + struct netfs_cache_resources *cres = &rreq->cache_resources; + enum netfs_io_source source; + + if (!cres->ops) + return NETFS_DOWNLOAD_FROM_SERVER; + source = cres->ops->prepare_read(subreq, i_size); + trace_netfs_sreq(subreq, netfs_sreq_trace_prepare); + return source; + +} + +/* + * Issue a read against the cache. + * - Eats the caller's ref on subreq. + */ +static void netfs_read_cache_to_pagecache(struct netfs_io_request *rreq, + struct netfs_io_subrequest *subreq) +{ + struct netfs_cache_resources *cres = &rreq->cache_resources; + + netfs_stat(&netfs_n_rh_read); + cres->ops->read(cres, subreq->start, &subreq->io_iter, NETFS_READ_HOLE_IGNORE, + netfs_cache_read_terminated, subreq); +} + +static void netfs_queue_read(struct netfs_io_request *rreq, + struct netfs_io_subrequest *subreq, + bool last_subreq) +{ + struct netfs_io_stream *stream = &rreq->io_streams[0]; + + __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + + /* We add to the end of the list whilst the collector may be walking + * the list. The collector only goes nextwards and uses the lock to + * remove entries off of the front. + */ + spin_lock(&rreq->lock); + list_add_tail(&subreq->rreq_link, &stream->subrequests); + if (list_is_first(&subreq->rreq_link, &stream->subrequests)) { + stream->front = subreq; + if (!stream->active) { + stream->collected_to = stream->front->start; + /* Store list pointers before active flag */ + smp_store_release(&stream->active, true); + } + } + + if (last_subreq) { + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags); + } + + spin_unlock(&rreq->lock); +} + +static void netfs_issue_read(struct netfs_io_request *rreq, + struct netfs_io_subrequest *subreq) +{ + switch (subreq->source) { + case NETFS_DOWNLOAD_FROM_SERVER: + rreq->netfs_ops->issue_read(subreq); + break; + case NETFS_READ_FROM_CACHE: + netfs_read_cache_to_pagecache(rreq, subreq); + break; + default: + __set_bit(NETFS_SREQ_CLEAR_TAIL, &subreq->flags); + subreq->error = 0; + iov_iter_zero(subreq->len, &subreq->io_iter); + subreq->transferred = subreq->len; + netfs_read_subreq_terminated(subreq); + break; + } +} + +/* + * Perform a read to the pagecache from a series of sources of different types, + * slicing up the region to be read according to available cache blocks and + * network rsize. + */ +static void netfs_read_to_pagecache(struct netfs_io_request *rreq) +{ + struct netfs_inode *ictx = netfs_inode(rreq->inode); + unsigned long long start = rreq->start; + ssize_t size = rreq->len; + int ret = 0; + + do { + struct netfs_io_subrequest *subreq; + enum netfs_io_source source = NETFS_SOURCE_UNKNOWN; + ssize_t slice; + + subreq = netfs_alloc_subrequest(rreq); + if (!subreq) { + ret = -ENOMEM; + break; + } + + subreq->start = start; + subreq->len = size; + + source = netfs_cache_prepare_read(rreq, subreq, rreq->i_size); + subreq->source = source; + if (source == NETFS_DOWNLOAD_FROM_SERVER) { + unsigned long long zp = umin(ictx->zero_point, rreq->i_size); + size_t len = subreq->len; + + if (unlikely(rreq->origin == NETFS_READ_SINGLE)) + zp = rreq->i_size; + if (subreq->start >= zp) { + subreq->source = source = NETFS_FILL_WITH_ZEROES; + goto fill_with_zeroes; + } + + if (len > zp - subreq->start) + len = zp - subreq->start; + if (len == 0) { + pr_err("ZERO-LEN READ: R=%08x[%x] l=%zx/%zx s=%llx z=%llx i=%llx", + rreq->debug_id, subreq->debug_index, + subreq->len, size, + subreq->start, ictx->zero_point, rreq->i_size); + break; + } + subreq->len = len; + + netfs_stat(&netfs_n_rh_download); + if (rreq->netfs_ops->prepare_read) { + ret = rreq->netfs_ops->prepare_read(subreq); + if (ret < 0) { + subreq->error = ret; + /* Not queued - release both refs. */ + netfs_put_subrequest(subreq, false, + netfs_sreq_trace_put_cancel); + netfs_put_subrequest(subreq, false, + netfs_sreq_trace_put_cancel); + break; + } + trace_netfs_sreq(subreq, netfs_sreq_trace_prepare); + } + goto issue; + } + + fill_with_zeroes: + if (source == NETFS_FILL_WITH_ZEROES) { + subreq->source = NETFS_FILL_WITH_ZEROES; + trace_netfs_sreq(subreq, netfs_sreq_trace_submit); + netfs_stat(&netfs_n_rh_zero); + goto issue; + } + + if (source == NETFS_READ_FROM_CACHE) { + trace_netfs_sreq(subreq, netfs_sreq_trace_submit); + goto issue; + } + + pr_err("Unexpected read source %u\n", source); + WARN_ON_ONCE(1); + break; + + issue: + slice = netfs_prepare_read_iterator(subreq); + if (slice < 0) { + ret = slice; + subreq->error = ret; + trace_netfs_sreq(subreq, netfs_sreq_trace_cancel); + /* Not queued - release both refs. */ + netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_cancel); + netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_cancel); + break; + } + size -= slice; + start += slice; + + netfs_queue_read(rreq, subreq, size <= 0); + netfs_issue_read(rreq, subreq); + cond_resched(); + } while (size > 0); + + if (unlikely(size > 0)) { + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags); + netfs_wake_read_collector(rreq); + } + + /* Defer error return as we may need to wait for outstanding I/O. */ + cmpxchg(&rreq->error, 0, ret); +} + /** * netfs_readahead - Helper to manage a read request * @ractl: The description of the readahead request @@ -185,22 +337,19 @@ static int netfs_begin_cache_read(struct netfs_io_request *rreq, struct netfs_in void netfs_readahead(struct readahead_control *ractl) { struct netfs_io_request *rreq; - struct netfs_inode *ctx = netfs_inode(ractl->mapping->host); + struct netfs_inode *ictx = netfs_inode(ractl->mapping->host); + unsigned long long start = readahead_pos(ractl); + size_t size = readahead_length(ractl); int ret; - _enter("%lx,%x", readahead_index(ractl), readahead_count(ractl)); - - if (readahead_count(ractl) == 0) - return; - - rreq = netfs_alloc_request(ractl->mapping, ractl->file, - readahead_pos(ractl), - readahead_length(ractl), + rreq = netfs_alloc_request(ractl->mapping, ractl->file, start, size, NETFS_READAHEAD); if (IS_ERR(rreq)) return; - ret = netfs_begin_cache_read(rreq, ctx); + __set_bit(NETFS_RREQ_OFFLOAD_COLLECTION, &rreq->flags); + + ret = netfs_begin_cache_read(rreq, ictx); if (ret == -ENOMEM || ret == -EINTR || ret == -ERESTARTSYS) goto cleanup_free; @@ -210,18 +359,13 @@ void netfs_readahead(struct readahead_control *ractl) netfs_rreq_expand(rreq, ractl); - /* Set up the output buffer */ - iov_iter_xarray(&rreq->iter, ITER_DEST, &ractl->mapping->i_pages, - rreq->start, rreq->len); - - /* Drop the refs on the folios here rather than in the cache or - * filesystem. The locks will be dropped in netfs_rreq_unlock(). - */ - while (readahead_folio(ractl)) - ; + rreq->ractl = ractl; + rreq->submitted = rreq->start; + if (rolling_buffer_init(&rreq->buffer, rreq->debug_id, ITER_DEST) < 0) + goto cleanup_free; + netfs_read_to_pagecache(rreq); - netfs_begin_read(rreq, false); - netfs_put_request(rreq, false, netfs_rreq_trace_put_return); + netfs_put_request(rreq, true, netfs_rreq_trace_put_return); return; cleanup_free: @@ -230,6 +374,112 @@ cleanup_free: } EXPORT_SYMBOL(netfs_readahead); +/* + * Create a rolling buffer with a single occupying folio. + */ +static int netfs_create_singular_buffer(struct netfs_io_request *rreq, struct folio *folio, + unsigned int rollbuf_flags) +{ + ssize_t added; + + if (rolling_buffer_init(&rreq->buffer, rreq->debug_id, ITER_DEST) < 0) + return -ENOMEM; + + added = rolling_buffer_append(&rreq->buffer, folio, rollbuf_flags); + if (added < 0) + return added; + rreq->submitted = rreq->start + added; + rreq->ractl = (struct readahead_control *)1UL; + return 0; +} + +/* + * Read into gaps in a folio partially filled by a streaming write. + */ +static int netfs_read_gaps(struct file *file, struct folio *folio) +{ + struct netfs_io_request *rreq; + struct address_space *mapping = folio->mapping; + struct netfs_folio *finfo = netfs_folio_info(folio); + struct netfs_inode *ctx = netfs_inode(mapping->host); + struct folio *sink = NULL; + struct bio_vec *bvec; + unsigned int from = finfo->dirty_offset; + unsigned int to = from + finfo->dirty_len; + unsigned int off = 0, i = 0; + size_t flen = folio_size(folio); + size_t nr_bvec = flen / PAGE_SIZE + 2; + size_t part; + int ret; + + _enter("%lx", folio->index); + + rreq = netfs_alloc_request(mapping, file, folio_pos(folio), flen, NETFS_READ_GAPS); + if (IS_ERR(rreq)) { + ret = PTR_ERR(rreq); + goto alloc_error; + } + + ret = netfs_begin_cache_read(rreq, ctx); + if (ret == -ENOMEM || ret == -EINTR || ret == -ERESTARTSYS) + goto discard; + + netfs_stat(&netfs_n_rh_read_folio); + trace_netfs_read(rreq, rreq->start, rreq->len, netfs_read_trace_read_gaps); + + /* Fiddle the buffer so that a gap at the beginning and/or a gap at the + * end get copied to, but the middle is discarded. + */ + ret = -ENOMEM; + bvec = kmalloc_array(nr_bvec, sizeof(*bvec), GFP_KERNEL); + if (!bvec) + goto discard; + + sink = folio_alloc(GFP_KERNEL, 0); + if (!sink) { + kfree(bvec); + goto discard; + } + + trace_netfs_folio(folio, netfs_folio_trace_read_gaps); + + rreq->direct_bv = bvec; + rreq->direct_bv_count = nr_bvec; + if (from > 0) { + bvec_set_folio(&bvec[i++], folio, from, 0); + off = from; + } + while (off < to) { + part = min_t(size_t, to - off, PAGE_SIZE); + bvec_set_folio(&bvec[i++], sink, part, 0); + off += part; + } + if (to < flen) + bvec_set_folio(&bvec[i++], folio, flen - to, to); + iov_iter_bvec(&rreq->buffer.iter, ITER_DEST, bvec, i, rreq->len); + rreq->submitted = rreq->start + flen; + + netfs_read_to_pagecache(rreq); + + if (sink) + folio_put(sink); + + ret = netfs_wait_for_read(rreq); + if (ret >= 0) { + flush_dcache_folio(folio); + folio_mark_uptodate(folio); + } + folio_unlock(folio); + netfs_put_request(rreq, false, netfs_rreq_trace_put_return); + return ret < 0 ? ret : 0; + +discard: + netfs_put_request(rreq, false, netfs_rreq_trace_put_discard); +alloc_error: + folio_unlock(folio); + return ret; +} + /** * netfs_read_folio - Helper to manage a read_folio request * @file: The file to read from @@ -249,13 +499,17 @@ int netfs_read_folio(struct file *file, struct folio *folio) struct address_space *mapping = folio->mapping; struct netfs_io_request *rreq; struct netfs_inode *ctx = netfs_inode(mapping->host); - struct folio *sink = NULL; int ret; + if (folio_test_dirty(folio)) { + trace_netfs_folio(folio, netfs_folio_trace_read_gaps); + return netfs_read_gaps(file, folio); + } + _enter("%lx", folio->index); rreq = netfs_alloc_request(mapping, file, - folio_file_pos(folio), folio_size(folio), + folio_pos(folio), folio_size(folio), NETFS_READPAGE); if (IS_ERR(rreq)) { ret = PTR_ERR(rreq); @@ -266,58 +520,16 @@ int netfs_read_folio(struct file *file, struct folio *folio) if (ret == -ENOMEM || ret == -EINTR || ret == -ERESTARTSYS) goto discard; - netfs_stat(&netfs_n_rh_readpage); + netfs_stat(&netfs_n_rh_read_folio); trace_netfs_read(rreq, rreq->start, rreq->len, netfs_read_trace_readpage); /* Set up the output buffer */ - if (folio_test_dirty(folio)) { - /* Handle someone trying to read from an unflushed streaming - * write. We fiddle the buffer so that a gap at the beginning - * and/or a gap at the end get copied to, but the middle is - * discarded. - */ - struct netfs_folio *finfo = netfs_folio_info(folio); - struct bio_vec *bvec; - unsigned int from = finfo->dirty_offset; - unsigned int to = from + finfo->dirty_len; - unsigned int off = 0, i = 0; - size_t flen = folio_size(folio); - size_t nr_bvec = flen / PAGE_SIZE + 2; - size_t part; - - ret = -ENOMEM; - bvec = kmalloc_array(nr_bvec, sizeof(*bvec), GFP_KERNEL); - if (!bvec) - goto discard; - - sink = folio_alloc(GFP_KERNEL, 0); - if (!sink) - goto discard; - - trace_netfs_folio(folio, netfs_folio_trace_read_gaps); - - rreq->direct_bv = bvec; - rreq->direct_bv_count = nr_bvec; - if (from > 0) { - bvec_set_folio(&bvec[i++], folio, from, 0); - off = from; - } - while (off < to) { - part = min_t(size_t, to - off, PAGE_SIZE); - bvec_set_folio(&bvec[i++], sink, part, 0); - off += part; - } - if (to < flen) - bvec_set_folio(&bvec[i++], folio, flen - to, to); - iov_iter_bvec(&rreq->iter, ITER_DEST, bvec, i, rreq->len); - } else { - iov_iter_xarray(&rreq->iter, ITER_DEST, &mapping->i_pages, - rreq->start, rreq->len); - } + ret = netfs_create_singular_buffer(rreq, folio, 0); + if (ret < 0) + goto discard; - ret = netfs_begin_read(rreq, true); - if (sink) - folio_put(sink); + netfs_read_to_pagecache(rreq); + ret = netfs_wait_for_read(rreq); netfs_put_request(rreq, false, netfs_rreq_trace_put_return); return ret < 0 ? ret : 0; @@ -355,7 +567,7 @@ static bool netfs_skip_folio_read(struct folio *folio, loff_t pos, size_t len, if (unlikely(always_fill)) { if (pos - offset + len <= i_size) return false; /* Page entirely before EOF */ - zero_user_segment(&folio->page, 0, plen); + folio_zero_segment(folio, 0, plen); folio_mark_uptodate(folio); return true; } @@ -374,12 +586,12 @@ static bool netfs_skip_folio_read(struct folio *folio, loff_t pos, size_t len, return false; zero_out: - zero_user_segments(&folio->page, 0, offset, offset + len, plen); + folio_zero_segments(folio, 0, offset, offset + len, plen); return true; } /** - * netfs_write_begin - Helper to prepare for writing + * netfs_write_begin - Helper to prepare for writing [DEPRECATED] * @ctx: The netfs context * @file: The file to read from * @mapping: The mapping to read from @@ -390,13 +602,10 @@ zero_out: * * Pre-read data for a write-begin request by drawing data from the cache if * possible, or the netfs if not. Space beyond the EOF is zero-filled. - * Multiple I/O requests from different sources will get munged together. If - * necessary, the readahead window can be expanded in either direction to a - * more convenient alighment for RPC efficiency or to make storage in the cache - * feasible. + * Multiple I/O requests from different sources will get munged together. * * The calling netfs must provide a table of operations, only one of which, - * issue_op, is mandatory. + * issue_read, is mandatory. * * The check_write_begin() operation can be provided to check for and flush * conflicting writes once the folio is grabbed and locked. It is passed a @@ -410,6 +619,9 @@ zero_out: * inode before calling this. * * This is usable whether or not caching is enabled. + * + * Note that this should be considered deprecated and netfs_perform_write() + * used instead. */ int netfs_write_begin(struct netfs_inode *ctx, struct file *file, struct address_space *mapping, @@ -421,8 +633,6 @@ int netfs_write_begin(struct netfs_inode *ctx, pgoff_t index = pos >> PAGE_SHIFT; int ret; - DEFINE_READAHEAD(ractl, file, NULL, mapping, index); - retry: folio = __filemap_get_folio(mapping, index, FGP_WRITEBEGIN, mapping_gfp_mask(mapping)); @@ -443,7 +653,7 @@ retry: if (folio_test_uptodate(folio)) goto have_folio; - /* If the page is beyond the EOF, we want to clear it - unless it's + /* If the folio is beyond the EOF, we want to clear it - unless it's * within the cache granule containing the EOF, in which case we need * to preload the granule. */ @@ -454,7 +664,7 @@ retry: } rreq = netfs_alloc_request(mapping, file, - folio_file_pos(folio), folio_size(folio), + folio_pos(folio), folio_size(folio), NETFS_READ_FOR_WRITE); if (IS_ERR(rreq)) { ret = PTR_ERR(rreq); @@ -470,28 +680,19 @@ retry: netfs_stat(&netfs_n_rh_write_begin); trace_netfs_read(rreq, pos, len, netfs_read_trace_write_begin); - /* Expand the request to meet caching requirements and download - * preferences. - */ - ractl._nr_pages = folio_nr_pages(folio); - netfs_rreq_expand(rreq, &ractl); - /* Set up the output buffer */ - iov_iter_xarray(&rreq->iter, ITER_DEST, &mapping->i_pages, - rreq->start, rreq->len); - - /* We hold the folio locks, so we can drop the references */ - folio_get(folio); - while (readahead_folio(&ractl)) - ; + ret = netfs_create_singular_buffer(rreq, folio, 0); + if (ret < 0) + goto error_put; - ret = netfs_begin_read(rreq, true); + netfs_read_to_pagecache(rreq); + ret = netfs_wait_for_read(rreq); if (ret < 0) goto error; netfs_put_request(rreq, false, netfs_rreq_trace_put_return); have_folio: - ret = folio_wait_fscache_killable(folio); + ret = folio_wait_private_2_killable(folio); if (ret < 0) goto error; have_folio_no_wait: @@ -512,7 +713,7 @@ error: EXPORT_SYMBOL(netfs_write_begin); /* - * Preload the data into a page we're proposing to write into. + * Preload the data into a folio we're proposing to write into. */ int netfs_prefetch_for_write(struct file *file, struct folio *folio, size_t offset, size_t len) @@ -545,12 +746,14 @@ int netfs_prefetch_for_write(struct file *file, struct folio *folio, trace_netfs_read(rreq, start, flen, netfs_read_trace_prefetch_for_write); /* Set up the output buffer */ - iov_iter_xarray(&rreq->iter, ITER_DEST, &mapping->i_pages, - rreq->start, rreq->len); + ret = netfs_create_singular_buffer(rreq, folio, NETFS_ROLLBUF_PAGECACHE_MARK); + if (ret < 0) + goto error_put; - ret = netfs_begin_read(rreq, true); + netfs_read_to_pagecache(rreq); + ret = netfs_wait_for_read(rreq); netfs_put_request(rreq, false, netfs_rreq_trace_put_return); - return ret; + return ret < 0 ? ret : 0; error_put: netfs_put_request(rreq, false, netfs_rreq_trace_put_discard); diff --git a/fs/netfs/buffered_write.c b/fs/netfs/buffered_write.c index 9a0d32e4b422..b4826360a411 100644 --- a/fs/netfs/buffered_write.c +++ b/fs/netfs/buffered_write.c @@ -1,5 +1,5 @@ // SPDX-License-Identifier: GPL-2.0-only -/* Network filesystem high-level write support. +/* Network filesystem high-level buffered write support. * * Copyright (C) 2023 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) @@ -13,104 +13,22 @@ #include <linux/pagevec.h> #include "internal.h" -/* - * Determined write method. Adjust netfs_folio_traces if this is changed. - */ -enum netfs_how_to_modify { - NETFS_FOLIO_IS_UPTODATE, /* Folio is uptodate already */ - NETFS_JUST_PREFETCH, /* We have to read the folio anyway */ - NETFS_WHOLE_FOLIO_MODIFY, /* We're going to overwrite the whole folio */ - NETFS_MODIFY_AND_CLEAR, /* We can assume there is no data to be downloaded. */ - NETFS_STREAMING_WRITE, /* Store incomplete data in non-uptodate page. */ - NETFS_STREAMING_WRITE_CONT, /* Continue streaming write. */ - NETFS_FLUSH_CONTENT, /* Flush incompatible content. */ -}; - -static void netfs_cleanup_buffered_write(struct netfs_io_request *wreq); - -static void netfs_set_group(struct folio *folio, struct netfs_group *netfs_group) +static void __netfs_set_group(struct folio *folio, struct netfs_group *netfs_group) { - if (netfs_group && !folio_get_private(folio)) + if (netfs_group) folio_attach_private(folio, netfs_get_group(netfs_group)); } -#if IS_ENABLED(CONFIG_FSCACHE) -static void netfs_folio_start_fscache(bool caching, struct folio *folio) -{ - if (caching) - folio_start_fscache(folio); -} -#else -static void netfs_folio_start_fscache(bool caching, struct folio *folio) -{ -} -#endif - -/* - * Decide how we should modify a folio. We might be attempting to do - * write-streaming, in which case we don't want to a local RMW cycle if we can - * avoid it. If we're doing local caching or content crypto, we award that - * priority over avoiding RMW. If the file is open readably, then we also - * assume that we may want to read what we wrote. - */ -static enum netfs_how_to_modify netfs_how_to_modify(struct netfs_inode *ctx, - struct file *file, - struct folio *folio, - void *netfs_group, - size_t flen, - size_t offset, - size_t len, - bool maybe_trouble) +static void netfs_set_group(struct folio *folio, struct netfs_group *netfs_group) { - struct netfs_folio *finfo = netfs_folio_info(folio); - loff_t pos = folio_file_pos(folio); - - _enter(""); - - if (netfs_folio_group(folio) != netfs_group) - return NETFS_FLUSH_CONTENT; - - if (folio_test_uptodate(folio)) - return NETFS_FOLIO_IS_UPTODATE; + void *priv = folio_get_private(folio); - if (pos >= ctx->zero_point) - return NETFS_MODIFY_AND_CLEAR; - - if (!maybe_trouble && offset == 0 && len >= flen) - return NETFS_WHOLE_FOLIO_MODIFY; - - if (file->f_mode & FMODE_READ) - goto no_write_streaming; - if (test_bit(NETFS_ICTX_NO_WRITE_STREAMING, &ctx->flags)) - goto no_write_streaming; - - if (netfs_is_cache_enabled(ctx)) { - /* We don't want to get a streaming write on a file that loses - * caching service temporarily because the backing store got - * culled. - */ - if (!test_bit(NETFS_ICTX_NO_WRITE_STREAMING, &ctx->flags)) - set_bit(NETFS_ICTX_NO_WRITE_STREAMING, &ctx->flags); - goto no_write_streaming; - } - - if (!finfo) - return NETFS_STREAMING_WRITE; - - /* We can continue a streaming write only if it continues on from the - * previous. If it overlaps, we must flush lest we suffer a partial - * copy and disjoint dirty regions. - */ - if (offset == finfo->dirty_offset + finfo->dirty_len) - return NETFS_STREAMING_WRITE_CONT; - return NETFS_FLUSH_CONTENT; - -no_write_streaming: - if (finfo) { - netfs_stat(&netfs_n_wh_wstream_conflict); - return NETFS_FLUSH_CONTENT; + if (unlikely(priv != netfs_group)) { + if (netfs_group && (!priv || priv == NETFS_FOLIO_COPY_TO_CACHE)) + folio_attach_private(folio, netfs_get_group(netfs_group)); + else if (!netfs_group && priv == NETFS_FOLIO_COPY_TO_CACHE) + folio_detach_private(folio); } - return NETFS_JUST_PREFETCH; } /* @@ -130,17 +48,48 @@ static struct folio *netfs_grab_folio_for_write(struct address_space *mapping, mapping_gfp_mask(mapping)); } +/* + * Update i_size and estimate the update to i_blocks to reflect the additional + * data written into the pagecache until we can find out from the server what + * the values actually are. + */ +static void netfs_update_i_size(struct netfs_inode *ctx, struct inode *inode, + loff_t i_size, loff_t pos, size_t copied) +{ + blkcnt_t add; + size_t gap; + + if (ctx->ops->update_i_size) { + ctx->ops->update_i_size(inode, pos); + return; + } + + i_size_write(inode, pos); +#if IS_ENABLED(CONFIG_FSCACHE) + fscache_update_cookie(ctx->cache, NULL, &pos); +#endif + + gap = SECTOR_SIZE - (i_size & (SECTOR_SIZE - 1)); + if (copied > gap) { + add = DIV_ROUND_UP(copied - gap, SECTOR_SIZE); + + inode->i_blocks = min_t(blkcnt_t, + DIV_ROUND_UP(pos, SECTOR_SIZE), + inode->i_blocks + add); + } +} + /** * netfs_perform_write - Copy data into the pagecache. * @iocb: The operation parameters * @iter: The source buffer - * @netfs_group: Grouping for dirty pages (eg. ceph snaps). + * @netfs_group: Grouping for dirty folios (eg. ceph snaps). * - * Copy data into pagecache pages attached to the inode specified by @iocb. + * Copy data into pagecache folios attached to the inode specified by @iocb. * The caller must hold appropriate inode locks. * - * Dirty pages are tagged with a netfs_folio struct if they're not up to date - * to indicate the range modified. Dirty pages may also be tagged with a + * Dirty folios are tagged with a netfs_folio struct if they're not up to date + * to indicate the range modified. Dirty folios may also be tagged with a * netfs-specific grouping such that data from an old group gets flushed before * a new one is started. */ @@ -159,28 +108,24 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, .range_end = iocb->ki_pos + iter->count, }; struct netfs_io_request *wreq = NULL; - struct netfs_folio *finfo; - struct folio *folio; - enum netfs_how_to_modify howto; - enum netfs_folio_trace trace; - unsigned int bdp_flags = (iocb->ki_flags & IOCB_SYNC) ? 0: BDP_ASYNC; - ssize_t written = 0, ret; - loff_t i_size, pos = iocb->ki_pos, from, to; - size_t max_chunk = PAGE_SIZE << MAX_PAGECACHE_ORDER; + struct folio *folio = NULL, *writethrough = NULL; + unsigned int bdp_flags = (iocb->ki_flags & IOCB_NOWAIT) ? BDP_ASYNC : 0; + ssize_t written = 0, ret, ret2; + loff_t i_size, pos = iocb->ki_pos; + size_t max_chunk = mapping_max_folio_size(mapping); bool maybe_trouble = false; if (unlikely(test_bit(NETFS_ICTX_WRITETHROUGH, &ctx->flags) || iocb->ki_flags & (IOCB_DSYNC | IOCB_SYNC)) ) { - if (pos < i_size_read(inode)) { - ret = filemap_write_and_wait_range(mapping, pos, pos + iter->count); - if (ret < 0) { - goto out; - } - } - wbc_attach_fdatawrite_inode(&wbc, mapping->host); + ret = filemap_write_and_wait_range(mapping, pos, pos + iter->count); + if (ret < 0) { + wbc_detach_inode(&wbc); + goto out; + } + wreq = netfs_begin_writethrough(iocb, iter->count); if (IS_ERR(wreq)) { wbc_detach_inode(&wbc); @@ -190,19 +135,20 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, } if (!is_sync_kiocb(iocb)) wreq->iocb = iocb; - wreq->cleanup = netfs_cleanup_buffered_write; + netfs_stat(&netfs_n_wh_writethrough); + } else { + netfs_stat(&netfs_n_wh_buffered_write); } do { + struct netfs_folio *finfo; + struct netfs_group *group; + unsigned long long fpos; size_t flen; size_t offset; /* Offset into pagecache folio */ size_t part; /* Bytes to write to folio */ size_t copied; /* Bytes copied from user */ - ret = balance_dirty_pages_ratelimited_flags(mapping, bdp_flags); - if (unlikely(ret < 0)) - break; - offset = pos & (max_chunk - 1); part = min(max_chunk - offset, iov_iter_count(iter)); @@ -228,95 +174,125 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, } flen = folio_size(folio); - offset = pos & (flen - 1); + fpos = folio_pos(folio); + offset = pos - fpos; part = min_t(size_t, flen - offset, part); + /* Wait for writeback to complete. The writeback engine owns + * the info in folio->private and may change it until it + * removes the WB mark. + */ + if (folio_get_private(folio) && + folio_wait_writeback_killable(folio)) { + ret = written ? -EINTR : -ERESTARTSYS; + goto error_folio_unlock; + } + if (signal_pending(current)) { ret = written ? -EINTR : -ERESTARTSYS; goto error_folio_unlock; } - /* See if we need to prefetch the area we're going to modify. - * We need to do this before we get a lock on the folio in case - * there's more than one writer competing for the same cache - * block. + /* Decide how we should modify a folio. We might be attempting + * to do write-streaming, in which case we don't want to a + * local RMW cycle if we can avoid it. If we're doing local + * caching or content crypto, we award that priority over + * avoiding RMW. If the file is open readably, then we also + * assume that we may want to read what we wrote. */ - howto = netfs_how_to_modify(ctx, file, folio, netfs_group, - flen, offset, part, maybe_trouble); - _debug("howto %u", howto); - switch (howto) { - case NETFS_JUST_PREFETCH: - ret = netfs_prefetch_for_write(file, folio, offset, part); - if (ret < 0) { - _debug("prefetch = %zd", ret); - goto error_folio_unlock; - } - break; - case NETFS_FOLIO_IS_UPTODATE: - case NETFS_WHOLE_FOLIO_MODIFY: - case NETFS_STREAMING_WRITE_CONT: - break; - case NETFS_MODIFY_AND_CLEAR: - zero_user_segment(&folio->page, 0, offset); - break; - case NETFS_STREAMING_WRITE: - ret = -EIO; - if (WARN_ON(folio_get_private(folio))) - goto error_folio_unlock; - break; - case NETFS_FLUSH_CONTENT: - trace_netfs_folio(folio, netfs_flush_content); - from = folio_pos(folio); - to = from + folio_size(folio) - 1; - folio_unlock(folio); - folio_put(folio); - ret = filemap_write_and_wait_range(mapping, from, to); - if (ret < 0) - goto error_folio_unlock; - continue; + finfo = netfs_folio_info(folio); + group = netfs_folio_group(folio); + + if (unlikely(group != netfs_group) && + group != NETFS_FOLIO_COPY_TO_CACHE) + goto flush_content; + + if (folio_test_uptodate(folio)) { + if (mapping_writably_mapped(mapping)) + flush_dcache_folio(folio); + copied = copy_folio_from_iter_atomic(folio, offset, part, iter); + if (unlikely(copied == 0)) + goto copy_failed; + netfs_set_group(folio, netfs_group); + trace_netfs_folio(folio, netfs_folio_is_uptodate); + goto copied; } - if (mapping_writably_mapped(mapping)) - flush_dcache_folio(folio); - - copied = copy_folio_from_iter_atomic(folio, offset, part, iter); - - flush_dcache_folio(folio); - - /* Deal with a (partially) failed copy */ - if (copied == 0) { - ret = -EFAULT; - goto error_folio_unlock; + /* If the page is above the zero-point then we assume that the + * server would just return a block of zeros or a short read if + * we try to read it. + */ + if (fpos >= ctx->zero_point) { + folio_zero_segment(folio, 0, offset); + copied = copy_folio_from_iter_atomic(folio, offset, part, iter); + if (unlikely(copied == 0)) + goto copy_failed; + folio_zero_segment(folio, offset + copied, flen); + __netfs_set_group(folio, netfs_group); + folio_mark_uptodate(folio); + trace_netfs_folio(folio, netfs_modify_and_clear); + goto copied; } - trace = (enum netfs_folio_trace)howto; - switch (howto) { - case NETFS_FOLIO_IS_UPTODATE: - case NETFS_JUST_PREFETCH: - netfs_set_group(folio, netfs_group); - break; - case NETFS_MODIFY_AND_CLEAR: - zero_user_segment(&folio->page, offset + copied, flen); - netfs_set_group(folio, netfs_group); - folio_mark_uptodate(folio); - break; - case NETFS_WHOLE_FOLIO_MODIFY: + /* See if we can write a whole folio in one go. */ + if (!maybe_trouble && offset == 0 && part >= flen) { + copied = copy_folio_from_iter_atomic(folio, offset, part, iter); + if (unlikely(copied == 0)) + goto copy_failed; if (unlikely(copied < part)) { maybe_trouble = true; iov_iter_revert(iter, copied); copied = 0; + folio_unlock(folio); goto retry; } - netfs_set_group(folio, netfs_group); + __netfs_set_group(folio, netfs_group); folio_mark_uptodate(folio); - break; - case NETFS_STREAMING_WRITE: + trace_netfs_folio(folio, netfs_whole_folio_modify); + goto copied; + } + + /* We don't want to do a streaming write on a file that loses + * caching service temporarily because the backing store got + * culled and we don't really want to get a streaming write on + * a file that's open for reading as ->read_folio() then has to + * be able to flush it. + */ + if ((file->f_mode & FMODE_READ) || + netfs_is_cache_enabled(ctx)) { + if (finfo) { + netfs_stat(&netfs_n_wh_wstream_conflict); + goto flush_content; + } + ret = netfs_prefetch_for_write(file, folio, offset, part); + if (ret < 0) { + _debug("prefetch = %zd", ret); + goto error_folio_unlock; + } + /* Note that copy-to-cache may have been set. */ + + copied = copy_folio_from_iter_atomic(folio, offset, part, iter); + if (unlikely(copied == 0)) + goto copy_failed; + netfs_set_group(folio, netfs_group); + trace_netfs_folio(folio, netfs_just_prefetch); + goto copied; + } + + if (!finfo) { + ret = -EIO; + if (WARN_ON(folio_get_private(folio))) + goto error_folio_unlock; + copied = copy_folio_from_iter_atomic(folio, offset, part, iter); + if (unlikely(copied == 0)) + goto copy_failed; if (offset == 0 && copied == flen) { - netfs_set_group(folio, netfs_group); + __netfs_set_group(folio, netfs_group); folio_mark_uptodate(folio); - trace = netfs_streaming_filled_page; - break; + trace_netfs_folio(folio, netfs_streaming_filled_page); + goto copied; } + finfo = kzalloc(sizeof(*finfo), GFP_KERNEL); if (!finfo) { iov_iter_revert(iter, copied); @@ -328,9 +304,18 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, finfo->dirty_len = copied; folio_attach_private(folio, (void *)((unsigned long)finfo | NETFS_FOLIO_INFO)); - break; - case NETFS_STREAMING_WRITE_CONT: - finfo = netfs_folio_info(folio); + trace_netfs_folio(folio, netfs_streaming_write); + goto copied; + } + + /* We can continue a streaming write only if it continues on + * from the previous. If it overlaps, we must flush lest we + * suffer a partial copy and disjoint dirty regions. + */ + if (offset == finfo->dirty_offset + finfo->dirty_len) { + copied = copy_folio_from_iter_atomic(folio, offset, part, iter); + if (unlikely(copied == 0)) + goto copy_failed; finfo->dirty_len += copied; if (finfo->dirty_offset == 0 && finfo->dirty_len == flen) { if (finfo->netfs_group) @@ -339,72 +324,78 @@ ssize_t netfs_perform_write(struct kiocb *iocb, struct iov_iter *iter, folio_detach_private(folio); folio_mark_uptodate(folio); kfree(finfo); - trace = netfs_streaming_cont_filled_page; + trace_netfs_folio(folio, netfs_streaming_cont_filled_page); + } else { + trace_netfs_folio(folio, netfs_streaming_write_cont); } - break; - default: - WARN(true, "Unexpected modify type %u ix=%lx\n", - howto, folio->index); - ret = -EIO; - goto error_folio_unlock; + goto copied; } - trace_netfs_folio(folio, trace); + /* Incompatible write; flush the folio and try again. */ + flush_content: + trace_netfs_folio(folio, netfs_flush_content); + folio_unlock(folio); + folio_put(folio); + ret = filemap_write_and_wait_range(mapping, fpos, fpos + flen - 1); + if (ret < 0) + goto error_folio_unlock; + continue; + + copied: + flush_dcache_folio(folio); /* Update the inode size if we moved the EOF marker */ - i_size = i_size_read(inode); pos += copied; - if (pos > i_size) { - if (ctx->ops->update_i_size) { - ctx->ops->update_i_size(inode, pos); - } else { - i_size_write(inode, pos); -#if IS_ENABLED(CONFIG_FSCACHE) - fscache_update_cookie(ctx->cache, NULL, &pos); -#endif - } - } + i_size = i_size_read(inode); + if (pos > i_size) + netfs_update_i_size(ctx, inode, i_size, pos, copied); written += copied; if (likely(!wreq)) { folio_mark_dirty(folio); + folio_unlock(folio); } else { - if (folio_test_dirty(folio)) - /* Sigh. mmap. */ - folio_clear_dirty_for_io(folio); - /* We make multiple writes to the folio... */ - if (!folio_test_writeback(folio)) { - folio_wait_fscache(folio); - folio_start_writeback(folio); - folio_start_fscache(folio); - if (wreq->iter.count == 0) - trace_netfs_folio(folio, netfs_folio_trace_wthru); - else - trace_netfs_folio(folio, netfs_folio_trace_wthru_plus); - } - netfs_advance_writethrough(wreq, copied, - offset + copied == flen); + netfs_advance_writethrough(wreq, &wbc, folio, copied, + offset + copied == flen, + &writethrough); + /* Folio unlocked */ } retry: - folio_unlock(folio); folio_put(folio); folio = NULL; + ret = balance_dirty_pages_ratelimited_flags(mapping, bdp_flags); + if (unlikely(ret < 0)) + break; + cond_resched(); } while (iov_iter_count(iter)); out: + if (likely(written)) { + /* Set indication that ctime and mtime got updated in case + * close is deferred. + */ + set_bit(NETFS_ICTX_MODIFIED_ATTR, &ctx->flags); + if (unlikely(ctx->ops->post_modify)) + ctx->ops->post_modify(inode); + } + if (unlikely(wreq)) { - ret = netfs_end_writethrough(wreq, iocb); + ret2 = netfs_end_writethrough(wreq, &wbc, writethrough); wbc_detach_inode(&wbc); - if (ret == -EIOCBQUEUED) - return ret; + if (ret2 == -EIOCBQUEUED) + return ret2; + if (ret == 0) + ret = ret2; } iocb->ki_pos += written; _leave(" = %zd [%zd]", written, ret); return written ? written : ret; +copy_failed: + ret = -EFAULT; error_folio_unlock: folio_unlock(folio); folio_put(folio); @@ -416,7 +407,7 @@ EXPORT_SYMBOL(netfs_perform_write); * netfs_buffered_write_iter_locked - write data to a file * @iocb: IO state structure (file, offset, etc.) * @from: iov_iter with data to write - * @netfs_group: Grouping for dirty pages (eg. ceph snaps). + * @netfs_group: Grouping for dirty folios (eg. ceph snaps). * * This function does all the work needed for actually writing data to a * file. It does all basic checks, removes SUID from the file, updates @@ -500,37 +491,44 @@ EXPORT_SYMBOL(netfs_file_write_iter); /* * Notification that a previously read-only page is about to become writable. - * Note that the caller indicates a single page of a multipage folio. + * The caller indicates the precise page that needs to be written to, but + * we only track group on a per-folio basis, so we block more often than + * we might otherwise. */ vm_fault_t netfs_page_mkwrite(struct vm_fault *vmf, struct netfs_group *netfs_group) { + struct netfs_group *group; struct folio *folio = page_folio(vmf->page); struct file *file = vmf->vma->vm_file; + struct address_space *mapping = file->f_mapping; struct inode *inode = file_inode(file); - vm_fault_t ret = VM_FAULT_RETRY; + struct netfs_inode *ictx = netfs_inode(inode); + vm_fault_t ret = VM_FAULT_NOPAGE; int err; _enter("%lx", folio->index); sb_start_pagefault(inode->i_sb); - if (folio_wait_writeback_killable(folio)) - goto out; - if (folio_lock_killable(folio) < 0) goto out; + if (folio->mapping != mapping) + goto unlock; + if (folio_wait_writeback_killable(folio) < 0) + goto unlock; /* Can we see a streaming write here? */ if (WARN_ON(!folio_test_uptodate(folio))) { - ret = VM_FAULT_SIGBUS | VM_FAULT_LOCKED; - goto out; + ret = VM_FAULT_SIGBUS; + goto unlock; } - if (netfs_folio_group(folio) != netfs_group) { + group = netfs_folio_group(folio); + if (group != netfs_group && group != NETFS_FOLIO_COPY_TO_CACHE) { folio_unlock(folio); - err = filemap_fdatawait_range(inode->i_mapping, - folio_pos(folio), - folio_pos(folio) + folio_size(folio)); + err = filemap_fdatawrite_range(mapping, + folio_pos(folio), + folio_pos(folio) + folio_size(folio)); switch (err) { case 0: ret = VM_FAULT_RETRY; @@ -550,708 +548,15 @@ vm_fault_t netfs_page_mkwrite(struct vm_fault *vmf, struct netfs_group *netfs_gr trace_netfs_folio(folio, netfs_folio_trace_mkwrite); netfs_set_group(folio, netfs_group); file_update_time(file); + set_bit(NETFS_ICTX_MODIFIED_ATTR, &ictx->flags); + if (ictx->ops->post_modify) + ictx->ops->post_modify(inode); ret = VM_FAULT_LOCKED; out: sb_end_pagefault(inode->i_sb); return ret; -} -EXPORT_SYMBOL(netfs_page_mkwrite); - -/* - * Kill all the pages in the given range - */ -static void netfs_kill_pages(struct address_space *mapping, - loff_t start, loff_t len) -{ - struct folio *folio; - pgoff_t index = start / PAGE_SIZE; - pgoff_t last = (start + len - 1) / PAGE_SIZE, next; - - _enter("%llx-%llx", start, start + len - 1); - - do { - _debug("kill %lx (to %lx)", index, last); - - folio = filemap_get_folio(mapping, index); - if (IS_ERR(folio)) { - next = index + 1; - continue; - } - - next = folio_next_index(folio); - - trace_netfs_folio(folio, netfs_folio_trace_kill); - folio_clear_uptodate(folio); - if (folio_test_fscache(folio)) - folio_end_fscache(folio); - folio_end_writeback(folio); - folio_lock(folio); - generic_error_remove_folio(mapping, folio); - folio_unlock(folio); - folio_put(folio); - - } while (index = next, index <= last); - - _leave(""); -} - -/* - * Redirty all the pages in a given range. - */ -static void netfs_redirty_pages(struct address_space *mapping, - loff_t start, loff_t len) -{ - struct folio *folio; - pgoff_t index = start / PAGE_SIZE; - pgoff_t last = (start + len - 1) / PAGE_SIZE, next; - - _enter("%llx-%llx", start, start + len - 1); - - do { - _debug("redirty %llx @%llx", len, start); - - folio = filemap_get_folio(mapping, index); - if (IS_ERR(folio)) { - next = index + 1; - continue; - } - - next = folio_next_index(folio); - trace_netfs_folio(folio, netfs_folio_trace_redirty); - filemap_dirty_folio(mapping, folio); - if (folio_test_fscache(folio)) - folio_end_fscache(folio); - folio_end_writeback(folio); - folio_put(folio); - } while (index = next, index <= last); - - balance_dirty_pages_ratelimited(mapping); - - _leave(""); -} - -/* - * Completion of write to server - */ -static void netfs_pages_written_back(struct netfs_io_request *wreq) -{ - struct address_space *mapping = wreq->mapping; - struct netfs_folio *finfo; - struct netfs_group *group = NULL; - struct folio *folio; - pgoff_t last; - int gcount = 0; - - XA_STATE(xas, &mapping->i_pages, wreq->start / PAGE_SIZE); - - _enter("%llx-%llx", wreq->start, wreq->start + wreq->len); - - rcu_read_lock(); - - last = (wreq->start + wreq->len - 1) / PAGE_SIZE; - xas_for_each(&xas, folio, last) { - WARN(!folio_test_writeback(folio), - "bad %zx @%llx page %lx %lx\n", - wreq->len, wreq->start, folio->index, last); - - if ((finfo = netfs_folio_info(folio))) { - /* Streaming writes cannot be redirtied whilst under - * writeback, so discard the streaming record. - */ - folio_detach_private(folio); - group = finfo->netfs_group; - gcount++; - trace_netfs_folio(folio, netfs_folio_trace_clear_s); - kfree(finfo); - } else if ((group = netfs_folio_group(folio))) { - /* Need to detach the group pointer if the page didn't - * get redirtied. If it has been redirtied, then it - * must be within the same group. - */ - if (folio_test_dirty(folio)) { - trace_netfs_folio(folio, netfs_folio_trace_redirtied); - goto end_wb; - } - if (folio_trylock(folio)) { - if (!folio_test_dirty(folio)) { - folio_detach_private(folio); - gcount++; - trace_netfs_folio(folio, netfs_folio_trace_clear_g); - } else { - trace_netfs_folio(folio, netfs_folio_trace_redirtied); - } - folio_unlock(folio); - goto end_wb; - } - - xas_pause(&xas); - rcu_read_unlock(); - folio_lock(folio); - if (!folio_test_dirty(folio)) { - folio_detach_private(folio); - gcount++; - trace_netfs_folio(folio, netfs_folio_trace_clear_g); - } else { - trace_netfs_folio(folio, netfs_folio_trace_redirtied); - } - folio_unlock(folio); - rcu_read_lock(); - } else { - trace_netfs_folio(folio, netfs_folio_trace_clear); - } - end_wb: - if (folio_test_fscache(folio)) - folio_end_fscache(folio); - xas_advance(&xas, folio_next_index(folio) - 1); - folio_end_writeback(folio); - } - - rcu_read_unlock(); - netfs_put_group_many(group, gcount); - _leave(""); -} - -/* - * Deal with the disposition of the folios that are under writeback to close - * out the operation. - */ -static void netfs_cleanup_buffered_write(struct netfs_io_request *wreq) -{ - struct address_space *mapping = wreq->mapping; - - _enter(""); - - switch (wreq->error) { - case 0: - netfs_pages_written_back(wreq); - break; - - default: - pr_notice("R=%08x Unexpected error %d\n", wreq->debug_id, wreq->error); - fallthrough; - case -EACCES: - case -EPERM: - case -ENOKEY: - case -EKEYEXPIRED: - case -EKEYREJECTED: - case -EKEYREVOKED: - case -ENETRESET: - case -EDQUOT: - case -ENOSPC: - netfs_redirty_pages(mapping, wreq->start, wreq->len); - break; - - case -EROFS: - case -EIO: - case -EREMOTEIO: - case -EFBIG: - case -ENOENT: - case -ENOMEDIUM: - case -ENXIO: - netfs_kill_pages(mapping, wreq->start, wreq->len); - break; - } - - if (wreq->error) - mapping_set_error(mapping, wreq->error); - if (wreq->netfs_ops->done) - wreq->netfs_ops->done(wreq); -} - -/* - * Extend the region to be written back to include subsequent contiguously - * dirty pages if possible, but don't sleep while doing so. - * - * If this page holds new content, then we can include filler zeros in the - * writeback. - */ -static void netfs_extend_writeback(struct address_space *mapping, - struct netfs_group *group, - struct xa_state *xas, - long *_count, - loff_t start, - loff_t max_len, - bool caching, - size_t *_len, - size_t *_top) -{ - struct netfs_folio *finfo; - struct folio_batch fbatch; - struct folio *folio; - unsigned int i; - pgoff_t index = (start + *_len) / PAGE_SIZE; - size_t len; - void *priv; - bool stop = true; - - folio_batch_init(&fbatch); - - do { - /* Firstly, we gather up a batch of contiguous dirty pages - * under the RCU read lock - but we can't clear the dirty flags - * there if any of those pages are mapped. - */ - rcu_read_lock(); - - xas_for_each(xas, folio, ULONG_MAX) { - stop = true; - if (xas_retry(xas, folio)) - continue; - if (xa_is_value(folio)) - break; - if (folio->index != index) { - xas_reset(xas); - break; - } - - if (!folio_try_get_rcu(folio)) { - xas_reset(xas); - continue; - } - - /* Has the folio moved or been split? */ - if (unlikely(folio != xas_reload(xas))) { - folio_put(folio); - xas_reset(xas); - break; - } - - if (!folio_trylock(folio)) { - folio_put(folio); - xas_reset(xas); - break; - } - if (!folio_test_dirty(folio) || - folio_test_writeback(folio) || - folio_test_fscache(folio)) { - folio_unlock(folio); - folio_put(folio); - xas_reset(xas); - break; - } - - stop = false; - len = folio_size(folio); - priv = folio_get_private(folio); - if ((const struct netfs_group *)priv != group) { - stop = true; - finfo = netfs_folio_info(folio); - if (finfo->netfs_group != group || - finfo->dirty_offset > 0) { - folio_unlock(folio); - folio_put(folio); - xas_reset(xas); - break; - } - len = finfo->dirty_len; - } - - *_top += folio_size(folio); - index += folio_nr_pages(folio); - *_count -= folio_nr_pages(folio); - *_len += len; - if (*_len >= max_len || *_count <= 0) - stop = true; - - if (!folio_batch_add(&fbatch, folio)) - break; - if (stop) - break; - } - - xas_pause(xas); - rcu_read_unlock(); - - /* Now, if we obtained any folios, we can shift them to being - * writable and mark them for caching. - */ - if (!folio_batch_count(&fbatch)) - break; - - for (i = 0; i < folio_batch_count(&fbatch); i++) { - folio = fbatch.folios[i]; - trace_netfs_folio(folio, netfs_folio_trace_store_plus); - - if (!folio_clear_dirty_for_io(folio)) - BUG(); - folio_start_writeback(folio); - netfs_folio_start_fscache(caching, folio); - folio_unlock(folio); - } - - folio_batch_release(&fbatch); - cond_resched(); - } while (!stop); -} - -/* - * Synchronously write back the locked page and any subsequent non-locked dirty - * pages. - */ -static ssize_t netfs_write_back_from_locked_folio(struct address_space *mapping, - struct writeback_control *wbc, - struct netfs_group *group, - struct xa_state *xas, - struct folio *folio, - unsigned long long start, - unsigned long long end) -{ - struct netfs_io_request *wreq; - struct netfs_folio *finfo; - struct netfs_inode *ctx = netfs_inode(mapping->host); - unsigned long long i_size = i_size_read(&ctx->inode); - size_t len, max_len; - bool caching = netfs_is_cache_enabled(ctx); - long count = wbc->nr_to_write; - int ret; - - _enter(",%lx,%llx-%llx,%u", folio->index, start, end, caching); - - wreq = netfs_alloc_request(mapping, NULL, start, folio_size(folio), - NETFS_WRITEBACK); - if (IS_ERR(wreq)) { - folio_unlock(folio); - return PTR_ERR(wreq); - } - - if (!folio_clear_dirty_for_io(folio)) - BUG(); - folio_start_writeback(folio); - netfs_folio_start_fscache(caching, folio); - - count -= folio_nr_pages(folio); - - /* Find all consecutive lockable dirty pages that have contiguous - * written regions, stopping when we find a page that is not - * immediately lockable, is not dirty or is missing, or we reach the - * end of the range. - */ - trace_netfs_folio(folio, netfs_folio_trace_store); - - len = wreq->len; - finfo = netfs_folio_info(folio); - if (finfo) { - start += finfo->dirty_offset; - if (finfo->dirty_offset + finfo->dirty_len != len) { - len = finfo->dirty_len; - goto cant_expand; - } - len = finfo->dirty_len; - } - - if (start < i_size) { - /* Trim the write to the EOF; the extra data is ignored. Also - * put an upper limit on the size of a single storedata op. - */ - max_len = 65536 * 4096; - max_len = min_t(unsigned long long, max_len, end - start + 1); - max_len = min_t(unsigned long long, max_len, i_size - start); - - if (len < max_len) - netfs_extend_writeback(mapping, group, xas, &count, start, - max_len, caching, &len, &wreq->upper_len); - } - -cant_expand: - len = min_t(unsigned long long, len, i_size - start); - - /* We now have a contiguous set of dirty pages, each with writeback - * set; the first page is still locked at this point, but all the rest - * have been unlocked. - */ +unlock: folio_unlock(folio); - wreq->start = start; - wreq->len = len; - - if (start < i_size) { - _debug("write back %zx @%llx [%llx]", len, start, i_size); - - /* Speculatively write to the cache. We have to fix this up - * later if the store fails. - */ - wreq->cleanup = netfs_cleanup_buffered_write; - - iov_iter_xarray(&wreq->iter, ITER_SOURCE, &mapping->i_pages, start, - wreq->upper_len); - __set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); - ret = netfs_begin_write(wreq, true, netfs_write_trace_writeback); - if (ret == 0 || ret == -EIOCBQUEUED) - wbc->nr_to_write -= len / PAGE_SIZE; - } else { - _debug("write discard %zx @%llx [%llx]", len, start, i_size); - - /* The dirty region was entirely beyond the EOF. */ - fscache_clear_page_bits(mapping, start, len, caching); - netfs_pages_written_back(wreq); - ret = 0; - } - - netfs_put_request(wreq, false, netfs_rreq_trace_put_return); - _leave(" = 1"); - return 1; -} - -/* - * Write a region of pages back to the server - */ -static ssize_t netfs_writepages_begin(struct address_space *mapping, - struct writeback_control *wbc, - struct netfs_group *group, - struct xa_state *xas, - unsigned long long *_start, - unsigned long long end) -{ - const struct netfs_folio *finfo; - struct folio *folio; - unsigned long long start = *_start; - ssize_t ret; - void *priv; - int skips = 0; - - _enter("%llx,%llx,", start, end); - -search_again: - /* Find the first dirty page in the group. */ - rcu_read_lock(); - - for (;;) { - folio = xas_find_marked(xas, end / PAGE_SIZE, PAGECACHE_TAG_DIRTY); - if (xas_retry(xas, folio) || xa_is_value(folio)) - continue; - if (!folio) - break; - - if (!folio_try_get_rcu(folio)) { - xas_reset(xas); - continue; - } - - if (unlikely(folio != xas_reload(xas))) { - folio_put(folio); - xas_reset(xas); - continue; - } - - /* Skip any dirty folio that's not in the group of interest. */ - priv = folio_get_private(folio); - if ((const struct netfs_group *)priv != group) { - finfo = netfs_folio_info(folio); - if (finfo->netfs_group != group) { - folio_put(folio); - continue; - } - } - - xas_pause(xas); - break; - } - rcu_read_unlock(); - if (!folio) - return 0; - - start = folio_pos(folio); /* May regress with THPs */ - - _debug("wback %lx", folio->index); - - /* At this point we hold neither the i_pages lock nor the page lock: - * the page may be truncated or invalidated (changing page->mapping to - * NULL), or even swizzled back from swapper_space to tmpfs file - * mapping - */ -lock_again: - if (wbc->sync_mode != WB_SYNC_NONE) { - ret = folio_lock_killable(folio); - if (ret < 0) - return ret; - } else { - if (!folio_trylock(folio)) - goto search_again; - } - - if (folio->mapping != mapping || - !folio_test_dirty(folio)) { - start += folio_size(folio); - folio_unlock(folio); - goto search_again; - } - - if (folio_test_writeback(folio) || - folio_test_fscache(folio)) { - folio_unlock(folio); - if (wbc->sync_mode != WB_SYNC_NONE) { - folio_wait_writeback(folio); -#ifdef CONFIG_FSCACHE - folio_wait_fscache(folio); -#endif - goto lock_again; - } - - start += folio_size(folio); - if (wbc->sync_mode == WB_SYNC_NONE) { - if (skips >= 5 || need_resched()) { - ret = 0; - goto out; - } - skips++; - } - goto search_again; - } - - ret = netfs_write_back_from_locked_folio(mapping, wbc, group, xas, - folio, start, end); -out: - if (ret > 0) - *_start = start + ret; - _leave(" = %zd [%llx]", ret, *_start); - return ret; -} - -/* - * Write a region of pages back to the server - */ -static int netfs_writepages_region(struct address_space *mapping, - struct writeback_control *wbc, - struct netfs_group *group, - unsigned long long *_start, - unsigned long long end) -{ - ssize_t ret; - - XA_STATE(xas, &mapping->i_pages, *_start / PAGE_SIZE); - - do { - ret = netfs_writepages_begin(mapping, wbc, group, &xas, - _start, end); - if (ret > 0 && wbc->nr_to_write > 0) - cond_resched(); - } while (ret > 0 && wbc->nr_to_write > 0); - - return ret > 0 ? 0 : ret; -} - -/* - * write some of the pending data back to the server - */ -int netfs_writepages(struct address_space *mapping, - struct writeback_control *wbc) -{ - struct netfs_group *group = NULL; - loff_t start, end; - int ret; - - _enter(""); - - /* We have to be careful as we can end up racing with setattr() - * truncating the pagecache since the caller doesn't take a lock here - * to prevent it. - */ - - if (wbc->range_cyclic && mapping->writeback_index) { - start = mapping->writeback_index * PAGE_SIZE; - ret = netfs_writepages_region(mapping, wbc, group, - &start, LLONG_MAX); - if (ret < 0) - goto out; - - if (wbc->nr_to_write <= 0) { - mapping->writeback_index = start / PAGE_SIZE; - goto out; - } - - start = 0; - end = mapping->writeback_index * PAGE_SIZE; - mapping->writeback_index = 0; - ret = netfs_writepages_region(mapping, wbc, group, &start, end); - if (ret == 0) - mapping->writeback_index = start / PAGE_SIZE; - } else if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) { - start = 0; - ret = netfs_writepages_region(mapping, wbc, group, - &start, LLONG_MAX); - if (wbc->nr_to_write > 0 && ret == 0) - mapping->writeback_index = start / PAGE_SIZE; - } else { - start = wbc->range_start; - ret = netfs_writepages_region(mapping, wbc, group, - &start, wbc->range_end); - } - -out: - _leave(" = %d", ret); - return ret; -} -EXPORT_SYMBOL(netfs_writepages); - -/* - * Deal with the disposition of a laundered folio. - */ -static void netfs_cleanup_launder_folio(struct netfs_io_request *wreq) -{ - if (wreq->error) { - pr_notice("R=%08x Laundering error %d\n", wreq->debug_id, wreq->error); - mapping_set_error(wreq->mapping, wreq->error); - } -} - -/** - * netfs_launder_folio - Clean up a dirty folio that's being invalidated - * @folio: The folio to clean - * - * This is called to write back a folio that's being invalidated when an inode - * is getting torn down. Ideally, writepages would be used instead. - */ -int netfs_launder_folio(struct folio *folio) -{ - struct netfs_io_request *wreq; - struct address_space *mapping = folio->mapping; - struct netfs_folio *finfo = netfs_folio_info(folio); - struct netfs_group *group = netfs_folio_group(folio); - struct bio_vec bvec; - unsigned long long i_size = i_size_read(mapping->host); - unsigned long long start = folio_pos(folio); - size_t offset = 0, len; - int ret = 0; - - if (finfo) { - offset = finfo->dirty_offset; - start += offset; - len = finfo->dirty_len; - } else { - len = folio_size(folio); - } - len = min_t(unsigned long long, len, i_size - start); - - wreq = netfs_alloc_request(mapping, NULL, start, len, NETFS_LAUNDER_WRITE); - if (IS_ERR(wreq)) { - ret = PTR_ERR(wreq); - goto out; - } - - if (!folio_clear_dirty_for_io(folio)) - goto out_put; - - trace_netfs_folio(folio, netfs_folio_trace_launder); - - _debug("launder %llx-%llx", start, start + len - 1); - - /* Speculatively write to the cache. We have to fix this up later if - * the store fails. - */ - wreq->cleanup = netfs_cleanup_launder_folio; - - bvec_set_folio(&bvec, folio, len, offset); - iov_iter_bvec(&wreq->iter, ITER_SOURCE, &bvec, 1, len); - __set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); - ret = netfs_begin_write(wreq, true, netfs_write_trace_launder); - -out_put: - folio_detach_private(folio); - netfs_put_group(group); - kfree(finfo); - netfs_put_request(wreq, false, netfs_rreq_trace_put_return); -out: - folio_wait_fscache(folio); - _leave(" = %d", ret); - return ret; + goto out; } -EXPORT_SYMBOL(netfs_launder_folio); +EXPORT_SYMBOL(netfs_page_mkwrite); diff --git a/fs/netfs/direct_read.c b/fs/netfs/direct_read.c index ad4370b3935d..5e3f0aeb51f3 100644 --- a/fs/netfs/direct_read.c +++ b/fs/netfs/direct_read.c @@ -16,6 +16,149 @@ #include <linux/netfs.h> #include "internal.h" +static void netfs_prepare_dio_read_iterator(struct netfs_io_subrequest *subreq) +{ + struct netfs_io_request *rreq = subreq->rreq; + size_t rsize; + + rsize = umin(subreq->len, rreq->io_streams[0].sreq_max_len); + subreq->len = rsize; + + if (unlikely(rreq->io_streams[0].sreq_max_segs)) { + size_t limit = netfs_limit_iter(&rreq->buffer.iter, 0, rsize, + rreq->io_streams[0].sreq_max_segs); + + if (limit < rsize) { + subreq->len = limit; + trace_netfs_sreq(subreq, netfs_sreq_trace_limited); + } + } + + trace_netfs_sreq(subreq, netfs_sreq_trace_prepare); + + subreq->io_iter = rreq->buffer.iter; + iov_iter_truncate(&subreq->io_iter, subreq->len); + iov_iter_advance(&rreq->buffer.iter, subreq->len); +} + +/* + * Perform a read to a buffer from the server, slicing up the region to be read + * according to the network rsize. + */ +static int netfs_dispatch_unbuffered_reads(struct netfs_io_request *rreq) +{ + struct netfs_io_stream *stream = &rreq->io_streams[0]; + unsigned long long start = rreq->start; + ssize_t size = rreq->len; + int ret = 0; + + do { + struct netfs_io_subrequest *subreq; + ssize_t slice; + + subreq = netfs_alloc_subrequest(rreq); + if (!subreq) { + ret = -ENOMEM; + break; + } + + subreq->source = NETFS_DOWNLOAD_FROM_SERVER; + subreq->start = start; + subreq->len = size; + + __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + + spin_lock(&rreq->lock); + list_add_tail(&subreq->rreq_link, &stream->subrequests); + if (list_is_first(&subreq->rreq_link, &stream->subrequests)) { + stream->front = subreq; + if (!stream->active) { + stream->collected_to = stream->front->start; + /* Store list pointers before active flag */ + smp_store_release(&stream->active, true); + } + } + trace_netfs_sreq(subreq, netfs_sreq_trace_added); + spin_unlock(&rreq->lock); + + netfs_stat(&netfs_n_rh_download); + if (rreq->netfs_ops->prepare_read) { + ret = rreq->netfs_ops->prepare_read(subreq); + if (ret < 0) { + netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_cancel); + break; + } + } + + netfs_prepare_dio_read_iterator(subreq); + slice = subreq->len; + size -= slice; + start += slice; + rreq->submitted += slice; + if (size <= 0) { + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags); + } + + rreq->netfs_ops->issue_read(subreq); + + if (test_bit(NETFS_RREQ_PAUSE, &rreq->flags)) + netfs_wait_for_pause(rreq); + if (test_bit(NETFS_RREQ_FAILED, &rreq->flags)) + break; + if (test_bit(NETFS_RREQ_BLOCKED, &rreq->flags) && + test_bit(NETFS_RREQ_NONBLOCK, &rreq->flags)) + break; + cond_resched(); + } while (size > 0); + + if (unlikely(size > 0)) { + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags); + netfs_wake_read_collector(rreq); + } + + return ret; +} + +/* + * Perform a read to an application buffer, bypassing the pagecache and the + * local disk cache. + */ +static ssize_t netfs_unbuffered_read(struct netfs_io_request *rreq, bool sync) +{ + ssize_t ret; + + _enter("R=%x %llx-%llx", + rreq->debug_id, rreq->start, rreq->start + rreq->len - 1); + + if (rreq->len == 0) { + pr_err("Zero-sized read [R=%x]\n", rreq->debug_id); + return -EIO; + } + + // TODO: Use bounce buffer if requested + + inode_dio_begin(rreq->inode); + + ret = netfs_dispatch_unbuffered_reads(rreq); + + if (!rreq->submitted) { + netfs_put_request(rreq, false, netfs_rreq_trace_put_no_submit); + inode_dio_end(rreq->inode); + ret = 0; + goto out; + } + + if (sync) + ret = netfs_wait_for_read(rreq); + else + ret = -EIOCBQUEUED; +out: + _leave(" = %zd", ret); + return ret; +} + /** * netfs_unbuffered_read_iter_locked - Perform an unbuffered or direct I/O read * @iocb: The I/O control descriptor describing the read @@ -26,12 +169,12 @@ * * The caller must hold any appropriate locks. */ -static ssize_t netfs_unbuffered_read_iter_locked(struct kiocb *iocb, struct iov_iter *iter) +ssize_t netfs_unbuffered_read_iter_locked(struct kiocb *iocb, struct iov_iter *iter) { struct netfs_io_request *rreq; ssize_t ret; size_t orig_count = iov_iter_count(iter); - bool async = !is_sync_kiocb(iocb); + bool sync = is_sync_kiocb(iocb); _enter(""); @@ -62,15 +205,15 @@ static ssize_t netfs_unbuffered_read_iter_locked(struct kiocb *iocb, struct iov_ * the request. */ if (user_backed_iter(iter)) { - ret = netfs_extract_user_iter(iter, rreq->len, &rreq->iter, 0); + ret = netfs_extract_user_iter(iter, rreq->len, &rreq->buffer.iter, 0); if (ret < 0) goto out; - rreq->direct_bv = (struct bio_vec *)rreq->iter.bvec; + rreq->direct_bv = (struct bio_vec *)rreq->buffer.iter.bvec; rreq->direct_bv_count = ret; rreq->direct_bv_unpin = iov_iter_extract_will_pin(iter); - rreq->len = iov_iter_count(&rreq->iter); + rreq->len = iov_iter_count(&rreq->buffer.iter); } else { - rreq->iter = *iter; + rreq->buffer.iter = *iter; rreq->len = orig_count; rreq->direct_bv_unpin = false; iov_iter_advance(iter, orig_count); @@ -78,13 +221,15 @@ static ssize_t netfs_unbuffered_read_iter_locked(struct kiocb *iocb, struct iov_ // TODO: Set up bounce buffer if needed - if (async) + if (!sync) { rreq->iocb = iocb; + __set_bit(NETFS_RREQ_OFFLOAD_COLLECTION, &rreq->flags); + } - ret = netfs_begin_read(rreq, is_sync_kiocb(iocb)); + ret = netfs_unbuffered_read(rreq, sync); if (ret < 0) goto out; /* May be -EIOCBQUEUED */ - if (!async) { + if (sync) { // TODO: Copy from bounce buffer iocb->ki_pos += rreq->transferred; ret = rreq->transferred; @@ -94,10 +239,9 @@ out: netfs_put_request(rreq, false, netfs_rreq_trace_put_return); if (ret > 0) orig_count -= ret; - if (ret != -EIOCBQUEUED) - iov_iter_revert(iter, orig_count - iov_iter_count(iter)); return ret; } +EXPORT_SYMBOL(netfs_unbuffered_read_iter_locked); /** * netfs_unbuffered_read_iter - Perform an unbuffered or direct I/O read diff --git a/fs/netfs/direct_write.c b/fs/netfs/direct_write.c index bee047e20f5d..42ce53cc216e 100644 --- a/fs/netfs/direct_write.c +++ b/fs/netfs/direct_write.c @@ -12,7 +12,7 @@ static void netfs_cleanup_dio_write(struct netfs_io_request *wreq) { struct inode *inode = wreq->inode; - unsigned long long end = wreq->start + wreq->len; + unsigned long long end = wreq->start + wreq->transferred; if (!wreq->error && i_size_read(inode) < end) { @@ -27,13 +27,14 @@ static void netfs_cleanup_dio_write(struct netfs_io_request *wreq) * Perform an unbuffered write where we may have to do an RMW operation on an * encrypted file. This can also be used for direct I/O writes. */ -static ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov_iter *iter, +ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov_iter *iter, struct netfs_group *netfs_group) { struct netfs_io_request *wreq; unsigned long long start = iocb->ki_pos; unsigned long long end = start + iov_iter_count(iter); ssize_t ret, n; + size_t len = iov_iter_count(iter); bool async = !is_sync_kiocb(iocb); _enter(""); @@ -46,13 +47,17 @@ static ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov _debug("uw %llx-%llx", start, end); - wreq = netfs_alloc_request(iocb->ki_filp->f_mapping, iocb->ki_filp, - start, end - start, - iocb->ki_flags & IOCB_DIRECT ? - NETFS_DIO_WRITE : NETFS_UNBUFFERED_WRITE); + wreq = netfs_create_write_req(iocb->ki_filp->f_mapping, iocb->ki_filp, start, + iocb->ki_flags & IOCB_DIRECT ? + NETFS_DIO_WRITE : NETFS_UNBUFFERED_WRITE); if (IS_ERR(wreq)) return PTR_ERR(wreq); + wreq->io_streams[0].avail = true; + trace_netfs_write(wreq, (iocb->ki_flags & IOCB_DIRECT ? + netfs_write_trace_dio_write : + netfs_write_trace_unbuffered_write)); + { /* If this is an async op and we're not using a bounce buffer, * we have to save the source buffer as the iterator is only @@ -62,23 +67,27 @@ static ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov * allocate a sufficiently large bvec array and may shorten the * request. */ - if (async || user_backed_iter(iter)) { - n = netfs_extract_user_iter(iter, wreq->len, &wreq->iter, 0); + if (user_backed_iter(iter)) { + n = netfs_extract_user_iter(iter, len, &wreq->buffer.iter, 0); if (n < 0) { ret = n; goto out; } - wreq->direct_bv = (struct bio_vec *)wreq->iter.bvec; + wreq->direct_bv = (struct bio_vec *)wreq->buffer.iter.bvec; wreq->direct_bv_count = n; wreq->direct_bv_unpin = iov_iter_extract_will_pin(iter); - wreq->len = iov_iter_count(&wreq->iter); } else { - wreq->iter = *iter; + /* If this is a kernel-generated async DIO request, + * assume that any resources the iterator points to + * (eg. a bio_vec array) will persist till the end of + * the op. + */ + wreq->buffer.iter = *iter; } - - wreq->io_iter = wreq->iter; } + __set_bit(NETFS_RREQ_USE_IO_ITER, &wreq->flags); + /* Copy the data into the bounce buffer and encrypt it. */ // TODO @@ -86,11 +95,9 @@ static ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov __set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); if (async) wreq->iocb = iocb; + wreq->len = iov_iter_count(&wreq->buffer.iter); wreq->cleanup = netfs_cleanup_dio_write; - ret = netfs_begin_write(wreq, is_sync_kiocb(iocb), - iocb->ki_flags & IOCB_DIRECT ? - netfs_write_trace_dio_write : - netfs_write_trace_unbuffered_write); + ret = netfs_unbuffered_write(wreq, is_sync_kiocb(iocb), wreq->len); if (ret < 0) { _debug("begin = %zd", ret); goto out; @@ -100,9 +107,7 @@ static ssize_t netfs_unbuffered_write_iter_locked(struct kiocb *iocb, struct iov trace_netfs_rreq(wreq, netfs_rreq_trace_wait_ip); wait_on_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS, TASK_UNINTERRUPTIBLE); - ret = wreq->error; - _debug("waited = %zd", ret); if (ret == 0) { ret = wreq->transferred; iocb->ki_pos += ret; @@ -115,6 +120,7 @@ out: netfs_put_request(wreq, false, netfs_rreq_trace_put_return); return ret; } +EXPORT_SYMBOL(netfs_unbuffered_write_iter_locked); /** * netfs_unbuffered_write_iter - Unbuffered write to a file @@ -132,18 +138,20 @@ out: ssize_t netfs_unbuffered_write_iter(struct kiocb *iocb, struct iov_iter *from) { struct file *file = iocb->ki_filp; - struct inode *inode = file->f_mapping->host; + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; struct netfs_inode *ictx = netfs_inode(inode); - unsigned long long end; ssize_t ret; + loff_t pos = iocb->ki_pos; + unsigned long long end = pos + iov_iter_count(from) - 1; - _enter("%llx,%zx,%llx", iocb->ki_pos, iov_iter_count(from), i_size_read(inode)); + _enter("%llx,%zx,%llx", pos, iov_iter_count(from), i_size_read(inode)); if (!iov_iter_count(from)) return 0; trace_netfs_write_iter(iocb, from); - netfs_stat(&netfs_n_rh_dio_write); + netfs_stat(&netfs_n_wh_dio_write); ret = netfs_start_io_direct(inode); if (ret < 0) @@ -157,7 +165,25 @@ ssize_t netfs_unbuffered_write_iter(struct kiocb *iocb, struct iov_iter *from) ret = file_update_time(file); if (ret < 0) goto out; - ret = kiocb_invalidate_pages(iocb, iov_iter_count(from)); + if (iocb->ki_flags & IOCB_NOWAIT) { + /* We could block if there are any pages in the range. */ + ret = -EAGAIN; + if (filemap_range_has_page(mapping, pos, end)) + if (filemap_invalidate_inode(inode, true, pos, end)) + goto out; + } else { + ret = filemap_write_and_wait_range(mapping, pos, end); + if (ret < 0) + goto out; + } + + /* + * After a write we want buffered reads to be sure to go to disk to get + * the new data. We invalidate clean cached page from the region we're + * about to write. We do this *before* the write so that we can return + * without clobbering -EIOCBQUEUED from ->direct_IO(). + */ + ret = filemap_invalidate_inode(inode, true, pos, end); if (ret < 0) goto out; end = iocb->ki_pos + iov_iter_count(from); diff --git a/fs/netfs/fscache_cache.c b/fs/netfs/fscache_cache.c index 9397ed39b0b4..8f70f8da064b 100644 --- a/fs/netfs/fscache_cache.c +++ b/fs/netfs/fscache_cache.c @@ -372,7 +372,7 @@ void fscache_withdraw_cache(struct fscache_cache *cache) EXPORT_SYMBOL(fscache_withdraw_cache); #ifdef CONFIG_PROC_FS -static const char fscache_cache_states[NR__FSCACHE_CACHE_STATE] = "-PAEW"; +static const char fscache_cache_states[NR__FSCACHE_CACHE_STATE] __nonstring = "-PAEW"; /* * Generate a list of caches in /proc/fs/fscache/caches diff --git a/fs/netfs/fscache_cookie.c b/fs/netfs/fscache_cookie.c index bce2492186d0..3d56fc73435f 100644 --- a/fs/netfs/fscache_cookie.c +++ b/fs/netfs/fscache_cookie.c @@ -29,7 +29,7 @@ static LIST_HEAD(fscache_cookie_lru); static DEFINE_SPINLOCK(fscache_cookie_lru_lock); DEFINE_TIMER(fscache_cookie_lru_timer, fscache_cookie_lru_timed_out); static DECLARE_WORK(fscache_cookie_lru_work, fscache_cookie_lru_worker); -static const char fscache_cookie_states[FSCACHE_COOKIE_STATE__NR] = "-LCAIFUWRD"; +static const char fscache_cookie_states[FSCACHE_COOKIE_STATE__NR] __nonstring = "-LCAIFUWRD"; static unsigned int fscache_lru_cookie_timeout = 10 * HZ; void fscache_print_cookie(struct fscache_cookie *cookie, char prefix) @@ -741,6 +741,10 @@ again_locked: spin_lock(&cookie->lock); } if (test_bit(FSCACHE_COOKIE_DO_LRU_DISCARD, &cookie->flags)) { + if (atomic_read(&cookie->n_accesses) != 0) + /* still being accessed: postpone it */ + break; + __fscache_set_cookie_state(cookie, FSCACHE_COOKIE_STATE_LRU_DISCARDING); wake = true; diff --git a/fs/netfs/fscache_io.c b/fs/netfs/fscache_io.c index 43a651ed8264..b1722a82c03d 100644 --- a/fs/netfs/fscache_io.c +++ b/fs/netfs/fscache_io.c @@ -9,7 +9,6 @@ #include <linux/uio.h> #include <linux/bvec.h> #include <linux/slab.h> -#include <linux/uio.h> #include "internal.h" /** @@ -166,6 +165,7 @@ struct fscache_write_request { loff_t start; size_t len; bool set_bits; + bool using_pgpriv2; netfs_io_terminated_t term_func; void *term_func_priv; }; @@ -182,7 +182,7 @@ void __fscache_clear_page_bits(struct address_space *mapping, rcu_read_lock(); xas_for_each(&xas, page, last) { - end_page_fscache(page); + folio_end_private_2(page_folio(page)); } rcu_read_unlock(); } @@ -197,8 +197,9 @@ static void fscache_wreq_done(void *priv, ssize_t transferred_or_error, { struct fscache_write_request *wreq = priv; - fscache_clear_page_bits(wreq->mapping, wreq->start, wreq->len, - wreq->set_bits); + if (wreq->using_pgpriv2) + fscache_clear_page_bits(wreq->mapping, wreq->start, wreq->len, + wreq->set_bits); if (wreq->term_func) wreq->term_func(wreq->term_func_priv, transferred_or_error, @@ -212,7 +213,7 @@ void __fscache_write_to_cache(struct fscache_cookie *cookie, loff_t start, size_t len, loff_t i_size, netfs_io_terminated_t term_func, void *term_func_priv, - bool cond) + bool using_pgpriv2, bool cond) { struct fscache_write_request *wreq; struct netfs_cache_resources *cres; @@ -230,6 +231,7 @@ void __fscache_write_to_cache(struct fscache_cookie *cookie, wreq->mapping = mapping; wreq->start = start; wreq->len = len; + wreq->using_pgpriv2 = using_pgpriv2; wreq->set_bits = cond; wreq->term_func = term_func; wreq->term_func_priv = term_func_priv; @@ -257,7 +259,8 @@ abandon_end: abandon_free: kfree(wreq); abandon: - fscache_clear_page_bits(mapping, start, len, cond); + if (using_pgpriv2) + fscache_clear_page_bits(mapping, start, len, cond); if (term_func) term_func(term_func_priv, ret, false); } diff --git a/fs/netfs/fscache_main.c b/fs/netfs/fscache_main.c index 42e98bb523e3..49849005eb7c 100644 --- a/fs/netfs/fscache_main.c +++ b/fs/netfs/fscache_main.c @@ -103,6 +103,7 @@ void __exit fscache_exit(void) kmem_cache_destroy(fscache_cookie_jar); fscache_proc_cleanup(); + timer_shutdown_sync(&fscache_cookie_lru_timer); destroy_workqueue(fscache_wq); pr_notice("FS-Cache unloaded\n"); } diff --git a/fs/netfs/fscache_volume.c b/fs/netfs/fscache_volume.c index cdf991bdd9de..ced14ac78cc1 100644 --- a/fs/netfs/fscache_volume.c +++ b/fs/netfs/fscache_volume.c @@ -27,6 +27,19 @@ struct fscache_volume *fscache_get_volume(struct fscache_volume *volume, return volume; } +struct fscache_volume *fscache_try_get_volume(struct fscache_volume *volume, + enum fscache_volume_trace where) +{ + int ref; + + if (!__refcount_inc_not_zero(&volume->ref, &ref)) + return NULL; + + trace_fscache_volume(volume->debug_id, ref + 1, where); + return volume; +} +EXPORT_SYMBOL(fscache_try_get_volume); + static void fscache_see_volume(struct fscache_volume *volume, enum fscache_volume_trace where) { @@ -309,8 +322,7 @@ maybe_wait: } return; no_wait: - clear_bit_unlock(FSCACHE_VOLUME_CREATING, &volume->flags); - wake_up_bit(&volume->flags, FSCACHE_VOLUME_CREATING); + clear_and_wake_up_bit(FSCACHE_VOLUME_CREATING, &volume->flags); } /* @@ -420,6 +432,7 @@ void fscache_put_volume(struct fscache_volume *volume, fscache_free_volume(volume); } } +EXPORT_SYMBOL(fscache_put_volume); /* * Relinquish a volume representation cookie. diff --git a/fs/netfs/internal.h b/fs/netfs/internal.h index ec7045d24400..1c4f953c3d68 100644 --- a/fs/netfs/internal.h +++ b/fs/netfs/internal.h @@ -7,6 +7,7 @@ #include <linux/slab.h> #include <linux/seq_file.h> +#include <linux/folio_queue.h> #include <linux/netfs.h> #include <linux/fscache.h> #include <linux/fscache-cache.h> @@ -22,21 +23,18 @@ /* * buffered_read.c */ -void netfs_rreq_unlock_folios(struct netfs_io_request *rreq); +void netfs_cache_read_terminated(void *priv, ssize_t transferred_or_error, bool was_async); int netfs_prefetch_for_write(struct file *file, struct folio *folio, size_t offset, size_t len); /* - * io.c - */ -int netfs_begin_read(struct netfs_io_request *rreq, bool sync); - -/* * main.c */ extern unsigned int netfs_debug; extern struct list_head netfs_io_requests; extern spinlock_t netfs_proc_lock; +extern mempool_t netfs_request_pool; +extern mempool_t netfs_subrequest_pool; #ifdef CONFIG_PROC_FS static inline void netfs_proc_add_rreq(struct netfs_io_request *rreq) @@ -61,15 +59,9 @@ static inline void netfs_proc_del_rreq(struct netfs_io_request *rreq) {} /* * misc.c */ -#define NETFS_FLAG_PUT_MARK BIT(0) -#define NETFS_FLAG_PAGECACHE_MARK BIT(1) -int netfs_xa_store_and_mark(struct xarray *xa, unsigned long index, - struct folio *folio, unsigned int flags, - gfp_t gfp_mask); -int netfs_add_folios_to_buffer(struct xarray *buffer, - struct address_space *mapping, - pgoff_t index, pgoff_t to, gfp_t gfp_mask); -void netfs_clear_buffer(struct xarray *buffer); +struct folio_queue *netfs_buffer_make_space(struct netfs_io_request *rreq, + enum netfs_folioq_trace trace); +void netfs_reset_iter(struct netfs_io_subrequest *subreq); /* * objects.c @@ -90,23 +82,43 @@ static inline void netfs_see_request(struct netfs_io_request *rreq, trace_netfs_rreq_ref(rreq->debug_id, refcount_read(&rreq->ref), what); } +static inline void netfs_see_subrequest(struct netfs_io_subrequest *subreq, + enum netfs_sreq_ref_trace what) +{ + trace_netfs_sreq_ref(subreq->rreq->debug_id, subreq->debug_index, + refcount_read(&subreq->ref), what); +} + /* - * output.c + * read_collect.c */ -int netfs_begin_write(struct netfs_io_request *wreq, bool may_wait, - enum netfs_write_trace what); -struct netfs_io_request *netfs_begin_writethrough(struct kiocb *iocb, size_t len); -int netfs_advance_writethrough(struct netfs_io_request *wreq, size_t copied, bool to_page_end); -int netfs_end_writethrough(struct netfs_io_request *wreq, struct kiocb *iocb); +void netfs_read_collection_worker(struct work_struct *work); +void netfs_wake_read_collector(struct netfs_io_request *rreq); +void netfs_cache_read_terminated(void *priv, ssize_t transferred_or_error, bool was_async); +ssize_t netfs_wait_for_read(struct netfs_io_request *rreq); +void netfs_wait_for_pause(struct netfs_io_request *rreq); + +/* + * read_pgpriv2.c + */ +void netfs_pgpriv2_copy_to_cache(struct netfs_io_request *rreq, struct folio *folio); +void netfs_pgpriv2_end_copy_to_cache(struct netfs_io_request *rreq); +bool netfs_pgpriv2_unlock_copied_folios(struct netfs_io_request *wreq); + +/* + * read_retry.c + */ +void netfs_retry_reads(struct netfs_io_request *rreq); +void netfs_unlock_abandoned_read_pages(struct netfs_io_request *rreq); /* * stats.c */ #ifdef CONFIG_NETFS_STATS extern atomic_t netfs_n_rh_dio_read; -extern atomic_t netfs_n_rh_dio_write; extern atomic_t netfs_n_rh_readahead; -extern atomic_t netfs_n_rh_readpage; +extern atomic_t netfs_n_rh_read_folio; +extern atomic_t netfs_n_rh_read_single; extern atomic_t netfs_n_rh_rreq; extern atomic_t netfs_n_rh_sreq; extern atomic_t netfs_n_rh_download; @@ -123,6 +135,13 @@ extern atomic_t netfs_n_rh_write_begin; extern atomic_t netfs_n_rh_write_done; extern atomic_t netfs_n_rh_write_failed; extern atomic_t netfs_n_rh_write_zskip; +extern atomic_t netfs_n_rh_retry_read_req; +extern atomic_t netfs_n_rh_retry_read_subreq; +extern atomic_t netfs_n_wh_buffered_write; +extern atomic_t netfs_n_wh_writethrough; +extern atomic_t netfs_n_wh_dio_write; +extern atomic_t netfs_n_wh_writepages; +extern atomic_t netfs_n_wh_copy_to_cache; extern atomic_t netfs_n_wh_wstream_conflict; extern atomic_t netfs_n_wh_upload; extern atomic_t netfs_n_wh_upload_done; @@ -130,6 +149,11 @@ extern atomic_t netfs_n_wh_upload_failed; extern atomic_t netfs_n_wh_write; extern atomic_t netfs_n_wh_write_done; extern atomic_t netfs_n_wh_write_failed; +extern atomic_t netfs_n_wh_retry_write_req; +extern atomic_t netfs_n_wh_retry_write_subreq; +extern atomic_t netfs_n_wb_lock_skip; +extern atomic_t netfs_n_wb_lock_wait; +extern atomic_t netfs_n_folioq; int netfs_stats_show(struct seq_file *m, void *v); @@ -149,6 +173,41 @@ static inline void netfs_stat_d(atomic_t *stat) #endif /* + * write_collect.c + */ +int netfs_folio_written_back(struct folio *folio); +void netfs_write_collection_worker(struct work_struct *work); +void netfs_wake_write_collector(struct netfs_io_request *wreq, bool was_async); + +/* + * write_issue.c + */ +struct netfs_io_request *netfs_create_write_req(struct address_space *mapping, + struct file *file, + loff_t start, + enum netfs_io_origin origin); +void netfs_reissue_write(struct netfs_io_stream *stream, + struct netfs_io_subrequest *subreq, + struct iov_iter *source); +void netfs_issue_write(struct netfs_io_request *wreq, + struct netfs_io_stream *stream); +size_t netfs_advance_write(struct netfs_io_request *wreq, + struct netfs_io_stream *stream, + loff_t start, size_t len, bool to_eof); +struct netfs_io_request *netfs_begin_writethrough(struct kiocb *iocb, size_t len); +int netfs_advance_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, + struct folio *folio, size_t copied, bool to_page_end, + struct folio **writethrough_cache); +int netfs_end_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, + struct folio *writethrough_cache); +int netfs_unbuffered_write(struct netfs_io_request *wreq, bool may_wait, size_t len); + +/* + * write_retry.c + */ +void netfs_retry_writes(struct netfs_io_request *wreq); + +/* * Miscellaneous functions. */ static inline bool netfs_is_cache_enabled(struct netfs_inode *ctx) @@ -168,7 +227,7 @@ static inline bool netfs_is_cache_enabled(struct netfs_inode *ctx) */ static inline struct netfs_group *netfs_get_group(struct netfs_group *netfs_group) { - if (netfs_group) + if (netfs_group && netfs_group != NETFS_FOLIO_COPY_TO_CACHE) refcount_inc(&netfs_group->ref); return netfs_group; } @@ -178,7 +237,9 @@ static inline struct netfs_group *netfs_get_group(struct netfs_group *netfs_grou */ static inline void netfs_put_group(struct netfs_group *netfs_group) { - if (netfs_group && refcount_dec_and_test(&netfs_group->ref)) + if (netfs_group && + netfs_group != NETFS_FOLIO_COPY_TO_CACHE && + refcount_dec_and_test(&netfs_group->ref)) netfs_group->free(netfs_group); } @@ -187,7 +248,9 @@ static inline void netfs_put_group(struct netfs_group *netfs_group) */ static inline void netfs_put_group_many(struct netfs_group *netfs_group, int nr) { - if (netfs_group && refcount_sub_and_test(nr, &netfs_group->ref)) + if (netfs_group && + netfs_group != NETFS_FOLIO_COPY_TO_CACHE && + refcount_sub_and_test(nr, &netfs_group->ref)) netfs_group->free(netfs_group); } @@ -326,8 +389,6 @@ extern const struct seq_operations fscache_volumes_seq_ops; struct fscache_volume *fscache_get_volume(struct fscache_volume *volume, enum fscache_volume_trace where); -void fscache_put_volume(struct fscache_volume *volume, - enum fscache_volume_trace where); bool fscache_begin_volume_access(struct fscache_volume *volume, struct fscache_cookie *cookie, enum fscache_access_trace why); diff --git a/fs/netfs/io.c b/fs/netfs/io.c deleted file mode 100644 index 4261ad6c55b6..000000000000 --- a/fs/netfs/io.c +++ /dev/null @@ -1,787 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-or-later -/* Network filesystem high-level read support. - * - * Copyright (C) 2021 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - */ - -#include <linux/module.h> -#include <linux/export.h> -#include <linux/fs.h> -#include <linux/mm.h> -#include <linux/pagemap.h> -#include <linux/slab.h> -#include <linux/uio.h> -#include <linux/sched/mm.h> -#include <linux/task_io_accounting_ops.h> -#include "internal.h" - -/* - * Clear the unread part of an I/O request. - */ -static void netfs_clear_unread(struct netfs_io_subrequest *subreq) -{ - iov_iter_zero(iov_iter_count(&subreq->io_iter), &subreq->io_iter); -} - -static void netfs_cache_read_terminated(void *priv, ssize_t transferred_or_error, - bool was_async) -{ - struct netfs_io_subrequest *subreq = priv; - - netfs_subreq_terminated(subreq, transferred_or_error, was_async); -} - -/* - * Issue a read against the cache. - * - Eats the caller's ref on subreq. - */ -static void netfs_read_from_cache(struct netfs_io_request *rreq, - struct netfs_io_subrequest *subreq, - enum netfs_read_from_hole read_hole) -{ - struct netfs_cache_resources *cres = &rreq->cache_resources; - - netfs_stat(&netfs_n_rh_read); - cres->ops->read(cres, subreq->start, &subreq->io_iter, read_hole, - netfs_cache_read_terminated, subreq); -} - -/* - * Fill a subrequest region with zeroes. - */ -static void netfs_fill_with_zeroes(struct netfs_io_request *rreq, - struct netfs_io_subrequest *subreq) -{ - netfs_stat(&netfs_n_rh_zero); - __set_bit(NETFS_SREQ_CLEAR_TAIL, &subreq->flags); - netfs_subreq_terminated(subreq, 0, false); -} - -/* - * Ask the netfs to issue a read request to the server for us. - * - * The netfs is expected to read from subreq->pos + subreq->transferred to - * subreq->pos + subreq->len - 1. It may not backtrack and write data into the - * buffer prior to the transferred point as it might clobber dirty data - * obtained from the cache. - * - * Alternatively, the netfs is allowed to indicate one of two things: - * - * - NETFS_SREQ_SHORT_READ: A short read - it will get called again to try and - * make progress. - * - * - NETFS_SREQ_CLEAR_TAIL: A short read - the rest of the buffer will be - * cleared. - */ -static void netfs_read_from_server(struct netfs_io_request *rreq, - struct netfs_io_subrequest *subreq) -{ - netfs_stat(&netfs_n_rh_download); - - if (rreq->origin != NETFS_DIO_READ && - iov_iter_count(&subreq->io_iter) != subreq->len - subreq->transferred) - pr_warn("R=%08x[%u] ITER PRE-MISMATCH %zx != %zx-%zx %lx\n", - rreq->debug_id, subreq->debug_index, - iov_iter_count(&subreq->io_iter), subreq->len, - subreq->transferred, subreq->flags); - rreq->netfs_ops->issue_read(subreq); -} - -/* - * Release those waiting. - */ -static void netfs_rreq_completed(struct netfs_io_request *rreq, bool was_async) -{ - trace_netfs_rreq(rreq, netfs_rreq_trace_done); - netfs_clear_subrequests(rreq, was_async); - netfs_put_request(rreq, was_async, netfs_rreq_trace_put_complete); -} - -/* - * Deal with the completion of writing the data to the cache. We have to clear - * the PG_fscache bits on the folios involved and release the caller's ref. - * - * May be called in softirq mode and we inherit a ref from the caller. - */ -static void netfs_rreq_unmark_after_write(struct netfs_io_request *rreq, - bool was_async) -{ - struct netfs_io_subrequest *subreq; - struct folio *folio; - pgoff_t unlocked = 0; - bool have_unlocked = false; - - rcu_read_lock(); - - list_for_each_entry(subreq, &rreq->subrequests, rreq_link) { - XA_STATE(xas, &rreq->mapping->i_pages, subreq->start / PAGE_SIZE); - - xas_for_each(&xas, folio, (subreq->start + subreq->len - 1) / PAGE_SIZE) { - if (xas_retry(&xas, folio)) - continue; - - /* We might have multiple writes from the same huge - * folio, but we mustn't unlock a folio more than once. - */ - if (have_unlocked && folio->index <= unlocked) - continue; - unlocked = folio_next_index(folio) - 1; - trace_netfs_folio(folio, netfs_folio_trace_end_copy); - folio_end_fscache(folio); - have_unlocked = true; - } - } - - rcu_read_unlock(); - netfs_rreq_completed(rreq, was_async); -} - -static void netfs_rreq_copy_terminated(void *priv, ssize_t transferred_or_error, - bool was_async) -{ - struct netfs_io_subrequest *subreq = priv; - struct netfs_io_request *rreq = subreq->rreq; - - if (IS_ERR_VALUE(transferred_or_error)) { - netfs_stat(&netfs_n_rh_write_failed); - trace_netfs_failure(rreq, subreq, transferred_or_error, - netfs_fail_copy_to_cache); - } else { - netfs_stat(&netfs_n_rh_write_done); - } - - trace_netfs_sreq(subreq, netfs_sreq_trace_write_term); - - /* If we decrement nr_copy_ops to 0, the ref belongs to us. */ - if (atomic_dec_and_test(&rreq->nr_copy_ops)) - netfs_rreq_unmark_after_write(rreq, was_async); - - netfs_put_subrequest(subreq, was_async, netfs_sreq_trace_put_terminated); -} - -/* - * Perform any outstanding writes to the cache. We inherit a ref from the - * caller. - */ -static void netfs_rreq_do_write_to_cache(struct netfs_io_request *rreq) -{ - struct netfs_cache_resources *cres = &rreq->cache_resources; - struct netfs_io_subrequest *subreq, *next, *p; - struct iov_iter iter; - int ret; - - trace_netfs_rreq(rreq, netfs_rreq_trace_copy); - - /* We don't want terminating writes trying to wake us up whilst we're - * still going through the list. - */ - atomic_inc(&rreq->nr_copy_ops); - - list_for_each_entry_safe(subreq, p, &rreq->subrequests, rreq_link) { - if (!test_bit(NETFS_SREQ_COPY_TO_CACHE, &subreq->flags)) { - list_del_init(&subreq->rreq_link); - netfs_put_subrequest(subreq, false, - netfs_sreq_trace_put_no_copy); - } - } - - list_for_each_entry(subreq, &rreq->subrequests, rreq_link) { - /* Amalgamate adjacent writes */ - while (!list_is_last(&subreq->rreq_link, &rreq->subrequests)) { - next = list_next_entry(subreq, rreq_link); - if (next->start != subreq->start + subreq->len) - break; - subreq->len += next->len; - list_del_init(&next->rreq_link); - netfs_put_subrequest(next, false, - netfs_sreq_trace_put_merged); - } - - ret = cres->ops->prepare_write(cres, &subreq->start, &subreq->len, - subreq->len, rreq->i_size, true); - if (ret < 0) { - trace_netfs_failure(rreq, subreq, ret, netfs_fail_prepare_write); - trace_netfs_sreq(subreq, netfs_sreq_trace_write_skip); - continue; - } - - iov_iter_xarray(&iter, ITER_SOURCE, &rreq->mapping->i_pages, - subreq->start, subreq->len); - - atomic_inc(&rreq->nr_copy_ops); - netfs_stat(&netfs_n_rh_write); - netfs_get_subrequest(subreq, netfs_sreq_trace_get_copy_to_cache); - trace_netfs_sreq(subreq, netfs_sreq_trace_write); - cres->ops->write(cres, subreq->start, &iter, - netfs_rreq_copy_terminated, subreq); - } - - /* If we decrement nr_copy_ops to 0, the usage ref belongs to us. */ - if (atomic_dec_and_test(&rreq->nr_copy_ops)) - netfs_rreq_unmark_after_write(rreq, false); -} - -static void netfs_rreq_write_to_cache_work(struct work_struct *work) -{ - struct netfs_io_request *rreq = - container_of(work, struct netfs_io_request, work); - - netfs_rreq_do_write_to_cache(rreq); -} - -static void netfs_rreq_write_to_cache(struct netfs_io_request *rreq) -{ - rreq->work.func = netfs_rreq_write_to_cache_work; - if (!queue_work(system_unbound_wq, &rreq->work)) - BUG(); -} - -/* - * Handle a short read. - */ -static void netfs_rreq_short_read(struct netfs_io_request *rreq, - struct netfs_io_subrequest *subreq) -{ - __clear_bit(NETFS_SREQ_SHORT_IO, &subreq->flags); - __set_bit(NETFS_SREQ_SEEK_DATA_READ, &subreq->flags); - - netfs_stat(&netfs_n_rh_short_read); - trace_netfs_sreq(subreq, netfs_sreq_trace_resubmit_short); - - netfs_get_subrequest(subreq, netfs_sreq_trace_get_short_read); - atomic_inc(&rreq->nr_outstanding); - if (subreq->source == NETFS_READ_FROM_CACHE) - netfs_read_from_cache(rreq, subreq, NETFS_READ_HOLE_CLEAR); - else - netfs_read_from_server(rreq, subreq); -} - -/* - * Reset the subrequest iterator prior to resubmission. - */ -static void netfs_reset_subreq_iter(struct netfs_io_request *rreq, - struct netfs_io_subrequest *subreq) -{ - size_t remaining = subreq->len - subreq->transferred; - size_t count = iov_iter_count(&subreq->io_iter); - - if (count == remaining) - return; - - _debug("R=%08x[%u] ITER RESUB-MISMATCH %zx != %zx-%zx-%llx %x\n", - rreq->debug_id, subreq->debug_index, - iov_iter_count(&subreq->io_iter), subreq->transferred, - subreq->len, rreq->i_size, - subreq->io_iter.iter_type); - - if (count < remaining) - iov_iter_revert(&subreq->io_iter, remaining - count); - else - iov_iter_advance(&subreq->io_iter, count - remaining); -} - -/* - * Resubmit any short or failed operations. Returns true if we got the rreq - * ref back. - */ -static bool netfs_rreq_perform_resubmissions(struct netfs_io_request *rreq) -{ - struct netfs_io_subrequest *subreq; - - WARN_ON(in_interrupt()); - - trace_netfs_rreq(rreq, netfs_rreq_trace_resubmit); - - /* We don't want terminating submissions trying to wake us up whilst - * we're still going through the list. - */ - atomic_inc(&rreq->nr_outstanding); - - __clear_bit(NETFS_RREQ_INCOMPLETE_IO, &rreq->flags); - list_for_each_entry(subreq, &rreq->subrequests, rreq_link) { - if (subreq->error) { - if (subreq->source != NETFS_READ_FROM_CACHE) - break; - subreq->source = NETFS_DOWNLOAD_FROM_SERVER; - subreq->error = 0; - netfs_stat(&netfs_n_rh_download_instead); - trace_netfs_sreq(subreq, netfs_sreq_trace_download_instead); - netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); - atomic_inc(&rreq->nr_outstanding); - netfs_reset_subreq_iter(rreq, subreq); - netfs_read_from_server(rreq, subreq); - } else if (test_bit(NETFS_SREQ_SHORT_IO, &subreq->flags)) { - netfs_rreq_short_read(rreq, subreq); - } - } - - /* If we decrement nr_outstanding to 0, the usage ref belongs to us. */ - if (atomic_dec_and_test(&rreq->nr_outstanding)) - return true; - - wake_up_var(&rreq->nr_outstanding); - return false; -} - -/* - * Check to see if the data read is still valid. - */ -static void netfs_rreq_is_still_valid(struct netfs_io_request *rreq) -{ - struct netfs_io_subrequest *subreq; - - if (!rreq->netfs_ops->is_still_valid || - rreq->netfs_ops->is_still_valid(rreq)) - return; - - list_for_each_entry(subreq, &rreq->subrequests, rreq_link) { - if (subreq->source == NETFS_READ_FROM_CACHE) { - subreq->error = -ESTALE; - __set_bit(NETFS_RREQ_INCOMPLETE_IO, &rreq->flags); - } - } -} - -/* - * Determine how much we can admit to having read from a DIO read. - */ -static void netfs_rreq_assess_dio(struct netfs_io_request *rreq) -{ - struct netfs_io_subrequest *subreq; - unsigned int i; - size_t transferred = 0; - - for (i = 0; i < rreq->direct_bv_count; i++) - flush_dcache_page(rreq->direct_bv[i].bv_page); - - list_for_each_entry(subreq, &rreq->subrequests, rreq_link) { - if (subreq->error || subreq->transferred == 0) - break; - transferred += subreq->transferred; - if (subreq->transferred < subreq->len) - break; - } - - for (i = 0; i < rreq->direct_bv_count; i++) - flush_dcache_page(rreq->direct_bv[i].bv_page); - - rreq->transferred = transferred; - task_io_account_read(transferred); - - if (rreq->iocb) { - rreq->iocb->ki_pos += transferred; - if (rreq->iocb->ki_complete) - rreq->iocb->ki_complete( - rreq->iocb, rreq->error ? rreq->error : transferred); - } - if (rreq->netfs_ops->done) - rreq->netfs_ops->done(rreq); - inode_dio_end(rreq->inode); -} - -/* - * Assess the state of a read request and decide what to do next. - * - * Note that we could be in an ordinary kernel thread, on a workqueue or in - * softirq context at this point. We inherit a ref from the caller. - */ -static void netfs_rreq_assess(struct netfs_io_request *rreq, bool was_async) -{ - trace_netfs_rreq(rreq, netfs_rreq_trace_assess); - -again: - netfs_rreq_is_still_valid(rreq); - - if (!test_bit(NETFS_RREQ_FAILED, &rreq->flags) && - test_bit(NETFS_RREQ_INCOMPLETE_IO, &rreq->flags)) { - if (netfs_rreq_perform_resubmissions(rreq)) - goto again; - return; - } - - if (rreq->origin != NETFS_DIO_READ) - netfs_rreq_unlock_folios(rreq); - else - netfs_rreq_assess_dio(rreq); - - trace_netfs_rreq(rreq, netfs_rreq_trace_wake_ip); - clear_bit_unlock(NETFS_RREQ_IN_PROGRESS, &rreq->flags); - wake_up_bit(&rreq->flags, NETFS_RREQ_IN_PROGRESS); - - if (test_bit(NETFS_RREQ_COPY_TO_CACHE, &rreq->flags)) - return netfs_rreq_write_to_cache(rreq); - - netfs_rreq_completed(rreq, was_async); -} - -static void netfs_rreq_work(struct work_struct *work) -{ - struct netfs_io_request *rreq = - container_of(work, struct netfs_io_request, work); - netfs_rreq_assess(rreq, false); -} - -/* - * Handle the completion of all outstanding I/O operations on a read request. - * We inherit a ref from the caller. - */ -static void netfs_rreq_terminated(struct netfs_io_request *rreq, - bool was_async) -{ - if (test_bit(NETFS_RREQ_INCOMPLETE_IO, &rreq->flags) && - was_async) { - if (!queue_work(system_unbound_wq, &rreq->work)) - BUG(); - } else { - netfs_rreq_assess(rreq, was_async); - } -} - -/** - * netfs_subreq_terminated - Note the termination of an I/O operation. - * @subreq: The I/O request that has terminated. - * @transferred_or_error: The amount of data transferred or an error code. - * @was_async: The termination was asynchronous - * - * This tells the read helper that a contributory I/O operation has terminated, - * one way or another, and that it should integrate the results. - * - * The caller indicates in @transferred_or_error the outcome of the operation, - * supplying a positive value to indicate the number of bytes transferred, 0 to - * indicate a failure to transfer anything that should be retried or a negative - * error code. The helper will look after reissuing I/O operations as - * appropriate and writing downloaded data to the cache. - * - * If @was_async is true, the caller might be running in softirq or interrupt - * context and we can't sleep. - */ -void netfs_subreq_terminated(struct netfs_io_subrequest *subreq, - ssize_t transferred_or_error, - bool was_async) -{ - struct netfs_io_request *rreq = subreq->rreq; - int u; - - _enter("R=%x[%x]{%llx,%lx},%zd", - rreq->debug_id, subreq->debug_index, - subreq->start, subreq->flags, transferred_or_error); - - switch (subreq->source) { - case NETFS_READ_FROM_CACHE: - netfs_stat(&netfs_n_rh_read_done); - break; - case NETFS_DOWNLOAD_FROM_SERVER: - netfs_stat(&netfs_n_rh_download_done); - break; - default: - break; - } - - if (IS_ERR_VALUE(transferred_or_error)) { - subreq->error = transferred_or_error; - trace_netfs_failure(rreq, subreq, transferred_or_error, - netfs_fail_read); - goto failed; - } - - if (WARN(transferred_or_error > subreq->len - subreq->transferred, - "Subreq overread: R%x[%x] %zd > %zu - %zu", - rreq->debug_id, subreq->debug_index, - transferred_or_error, subreq->len, subreq->transferred)) - transferred_or_error = subreq->len - subreq->transferred; - - subreq->error = 0; - subreq->transferred += transferred_or_error; - if (subreq->transferred < subreq->len) - goto incomplete; - -complete: - __clear_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags); - if (test_bit(NETFS_SREQ_COPY_TO_CACHE, &subreq->flags)) - set_bit(NETFS_RREQ_COPY_TO_CACHE, &rreq->flags); - -out: - trace_netfs_sreq(subreq, netfs_sreq_trace_terminated); - - /* If we decrement nr_outstanding to 0, the ref belongs to us. */ - u = atomic_dec_return(&rreq->nr_outstanding); - if (u == 0) - netfs_rreq_terminated(rreq, was_async); - else if (u == 1) - wake_up_var(&rreq->nr_outstanding); - - netfs_put_subrequest(subreq, was_async, netfs_sreq_trace_put_terminated); - return; - -incomplete: - if (test_bit(NETFS_SREQ_CLEAR_TAIL, &subreq->flags)) { - netfs_clear_unread(subreq); - subreq->transferred = subreq->len; - goto complete; - } - - if (transferred_or_error == 0) { - if (__test_and_set_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags)) { - subreq->error = -ENODATA; - goto failed; - } - } else { - __clear_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags); - } - - __set_bit(NETFS_SREQ_SHORT_IO, &subreq->flags); - set_bit(NETFS_RREQ_INCOMPLETE_IO, &rreq->flags); - goto out; - -failed: - if (subreq->source == NETFS_READ_FROM_CACHE) { - netfs_stat(&netfs_n_rh_read_failed); - set_bit(NETFS_RREQ_INCOMPLETE_IO, &rreq->flags); - } else { - netfs_stat(&netfs_n_rh_download_failed); - set_bit(NETFS_RREQ_FAILED, &rreq->flags); - rreq->error = subreq->error; - } - goto out; -} -EXPORT_SYMBOL(netfs_subreq_terminated); - -static enum netfs_io_source netfs_cache_prepare_read(struct netfs_io_subrequest *subreq, - loff_t i_size) -{ - struct netfs_io_request *rreq = subreq->rreq; - struct netfs_cache_resources *cres = &rreq->cache_resources; - - if (cres->ops) - return cres->ops->prepare_read(subreq, i_size); - if (subreq->start >= rreq->i_size) - return NETFS_FILL_WITH_ZEROES; - return NETFS_DOWNLOAD_FROM_SERVER; -} - -/* - * Work out what sort of subrequest the next one will be. - */ -static enum netfs_io_source -netfs_rreq_prepare_read(struct netfs_io_request *rreq, - struct netfs_io_subrequest *subreq, - struct iov_iter *io_iter) -{ - enum netfs_io_source source = NETFS_DOWNLOAD_FROM_SERVER; - struct netfs_inode *ictx = netfs_inode(rreq->inode); - size_t lsize; - - _enter("%llx-%llx,%llx", subreq->start, subreq->start + subreq->len, rreq->i_size); - - if (rreq->origin != NETFS_DIO_READ) { - source = netfs_cache_prepare_read(subreq, rreq->i_size); - if (source == NETFS_INVALID_READ) - goto out; - } - - if (source == NETFS_DOWNLOAD_FROM_SERVER) { - /* Call out to the netfs to let it shrink the request to fit - * its own I/O sizes and boundaries. If it shinks it here, it - * will be called again to make simultaneous calls; if it wants - * to make serial calls, it can indicate a short read and then - * we will call it again. - */ - if (rreq->origin != NETFS_DIO_READ) { - if (subreq->start >= ictx->zero_point) { - source = NETFS_FILL_WITH_ZEROES; - goto set; - } - if (subreq->len > ictx->zero_point - subreq->start) - subreq->len = ictx->zero_point - subreq->start; - } - if (subreq->len > rreq->i_size - subreq->start) - subreq->len = rreq->i_size - subreq->start; - if (rreq->rsize && subreq->len > rreq->rsize) - subreq->len = rreq->rsize; - - if (rreq->netfs_ops->clamp_length && - !rreq->netfs_ops->clamp_length(subreq)) { - source = NETFS_INVALID_READ; - goto out; - } - - if (subreq->max_nr_segs) { - lsize = netfs_limit_iter(io_iter, 0, subreq->len, - subreq->max_nr_segs); - if (subreq->len > lsize) { - subreq->len = lsize; - trace_netfs_sreq(subreq, netfs_sreq_trace_limited); - } - } - } - -set: - if (subreq->len > rreq->len) - pr_warn("R=%08x[%u] SREQ>RREQ %zx > %zx\n", - rreq->debug_id, subreq->debug_index, - subreq->len, rreq->len); - - if (WARN_ON(subreq->len == 0)) { - source = NETFS_INVALID_READ; - goto out; - } - - subreq->source = source; - trace_netfs_sreq(subreq, netfs_sreq_trace_prepare); - - subreq->io_iter = *io_iter; - iov_iter_truncate(&subreq->io_iter, subreq->len); - iov_iter_advance(io_iter, subreq->len); -out: - subreq->source = source; - trace_netfs_sreq(subreq, netfs_sreq_trace_prepare); - return source; -} - -/* - * Slice off a piece of a read request and submit an I/O request for it. - */ -static bool netfs_rreq_submit_slice(struct netfs_io_request *rreq, - struct iov_iter *io_iter, - unsigned int *_debug_index) -{ - struct netfs_io_subrequest *subreq; - enum netfs_io_source source; - - subreq = netfs_alloc_subrequest(rreq); - if (!subreq) - return false; - - subreq->debug_index = (*_debug_index)++; - subreq->start = rreq->start + rreq->submitted; - subreq->len = io_iter->count; - - _debug("slice %llx,%zx,%zx", subreq->start, subreq->len, rreq->submitted); - list_add_tail(&subreq->rreq_link, &rreq->subrequests); - - /* Call out to the cache to find out what it can do with the remaining - * subset. It tells us in subreq->flags what it decided should be done - * and adjusts subreq->len down if the subset crosses a cache boundary. - * - * Then when we hand the subset, it can choose to take a subset of that - * (the starts must coincide), in which case, we go around the loop - * again and ask it to download the next piece. - */ - source = netfs_rreq_prepare_read(rreq, subreq, io_iter); - if (source == NETFS_INVALID_READ) - goto subreq_failed; - - atomic_inc(&rreq->nr_outstanding); - - rreq->submitted += subreq->len; - - trace_netfs_sreq(subreq, netfs_sreq_trace_submit); - switch (source) { - case NETFS_FILL_WITH_ZEROES: - netfs_fill_with_zeroes(rreq, subreq); - break; - case NETFS_DOWNLOAD_FROM_SERVER: - netfs_read_from_server(rreq, subreq); - break; - case NETFS_READ_FROM_CACHE: - netfs_read_from_cache(rreq, subreq, NETFS_READ_HOLE_IGNORE); - break; - default: - BUG(); - } - - return true; - -subreq_failed: - rreq->error = subreq->error; - netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_failed); - return false; -} - -/* - * Begin the process of reading in a chunk of data, where that data may be - * stitched together from multiple sources, including multiple servers and the - * local cache. - */ -int netfs_begin_read(struct netfs_io_request *rreq, bool sync) -{ - struct iov_iter io_iter; - unsigned int debug_index = 0; - int ret; - - _enter("R=%x %llx-%llx", - rreq->debug_id, rreq->start, rreq->start + rreq->len - 1); - - if (rreq->len == 0) { - pr_err("Zero-sized read [R=%x]\n", rreq->debug_id); - return -EIO; - } - - if (rreq->origin == NETFS_DIO_READ) - inode_dio_begin(rreq->inode); - - // TODO: Use bounce buffer if requested - rreq->io_iter = rreq->iter; - - INIT_WORK(&rreq->work, netfs_rreq_work); - - /* Chop the read into slices according to what the cache and the netfs - * want and submit each one. - */ - netfs_get_request(rreq, netfs_rreq_trace_get_for_outstanding); - atomic_set(&rreq->nr_outstanding, 1); - io_iter = rreq->io_iter; - do { - _debug("submit %llx + %zx >= %llx", - rreq->start, rreq->submitted, rreq->i_size); - if (rreq->origin == NETFS_DIO_READ && - rreq->start + rreq->submitted >= rreq->i_size) - break; - if (!netfs_rreq_submit_slice(rreq, &io_iter, &debug_index)) - break; - if (test_bit(NETFS_RREQ_BLOCKED, &rreq->flags) && - test_bit(NETFS_RREQ_NONBLOCK, &rreq->flags)) - break; - - } while (rreq->submitted < rreq->len); - - if (!rreq->submitted) { - netfs_put_request(rreq, false, netfs_rreq_trace_put_no_submit); - if (rreq->origin == NETFS_DIO_READ) - inode_dio_end(rreq->inode); - ret = 0; - goto out; - } - - if (sync) { - /* Keep nr_outstanding incremented so that the ref always - * belongs to us, and the service code isn't punted off to a - * random thread pool to process. Note that this might start - * further work, such as writing to the cache. - */ - wait_var_event(&rreq->nr_outstanding, - atomic_read(&rreq->nr_outstanding) == 1); - if (atomic_dec_and_test(&rreq->nr_outstanding)) - netfs_rreq_assess(rreq, false); - - trace_netfs_rreq(rreq, netfs_rreq_trace_wait_ip); - wait_on_bit(&rreq->flags, NETFS_RREQ_IN_PROGRESS, - TASK_UNINTERRUPTIBLE); - - ret = rreq->error; - if (ret == 0 && rreq->submitted < rreq->len && - rreq->origin != NETFS_DIO_READ) { - trace_netfs_failure(rreq, NULL, ret, netfs_fail_short_read); - ret = -EIO; - } - } else { - /* If we decrement nr_outstanding to 0, the ref belongs to us. */ - if (atomic_dec_and_test(&rreq->nr_outstanding)) - netfs_rreq_assess(rreq, false); - ret = -EIOCBQUEUED; - } - -out: - return ret; -} diff --git a/fs/netfs/iterator.c b/fs/netfs/iterator.c index b781bbbf1d8d..72a435e5fc6d 100644 --- a/fs/netfs/iterator.c +++ b/fs/netfs/iterator.c @@ -188,9 +188,59 @@ static size_t netfs_limit_xarray(const struct iov_iter *iter, size_t start_offse return min(span, max_size); } +/* + * Select the span of a folio queue iterator we're going to use. Limit it by + * both maximum size and maximum number of segments. Returns the size of the + * span in bytes. + */ +static size_t netfs_limit_folioq(const struct iov_iter *iter, size_t start_offset, + size_t max_size, size_t max_segs) +{ + const struct folio_queue *folioq = iter->folioq; + unsigned int nsegs = 0; + unsigned int slot = iter->folioq_slot; + size_t span = 0, n = iter->count; + + if (WARN_ON(!iov_iter_is_folioq(iter)) || + WARN_ON(start_offset > n) || + n == 0) + return 0; + max_size = umin(max_size, n - start_offset); + + if (slot >= folioq_nr_slots(folioq)) { + folioq = folioq->next; + slot = 0; + } + + start_offset += iter->iov_offset; + do { + size_t flen = folioq_folio_size(folioq, slot); + + if (start_offset < flen) { + span += flen - start_offset; + nsegs++; + start_offset = 0; + } else { + start_offset -= flen; + } + if (span >= max_size || nsegs >= max_segs) + break; + + slot++; + if (slot >= folioq_nr_slots(folioq)) { + folioq = folioq->next; + slot = 0; + } + } while (folioq); + + return umin(span, max_size); +} + size_t netfs_limit_iter(const struct iov_iter *iter, size_t start_offset, size_t max_size, size_t max_segs) { + if (iov_iter_is_folioq(iter)) + return netfs_limit_folioq(iter, start_offset, max_size, max_segs); if (iov_iter_is_bvec(iter)) return netfs_limit_bvec(iter, start_offset, max_size, max_segs); if (iov_iter_is_xarray(iter)) diff --git a/fs/netfs/locking.c b/fs/netfs/locking.c index 75dc52a49b3a..2249ecd09d0a 100644 --- a/fs/netfs/locking.c +++ b/fs/netfs/locking.c @@ -19,25 +19,13 @@ * Must be called under a lock that serializes taking new references * to i_dio_count, usually by inode->i_mutex. */ -static int inode_dio_wait_interruptible(struct inode *inode) +static int netfs_inode_dio_wait_interruptible(struct inode *inode) { - if (!atomic_read(&inode->i_dio_count)) + if (inode_dio_finished(inode)) return 0; - wait_queue_head_t *wq = bit_waitqueue(&inode->i_state, __I_DIO_WAKEUP); - DEFINE_WAIT_BIT(q, &inode->i_state, __I_DIO_WAKEUP); - - for (;;) { - prepare_to_wait(wq, &q.wq_entry, TASK_INTERRUPTIBLE); - if (!atomic_read(&inode->i_dio_count)) - break; - if (signal_pending(current)) - break; - schedule(); - } - finish_wait(wq, &q.wq_entry); - - return atomic_read(&inode->i_dio_count) ? -ERESTARTSYS : 0; + inode_dio_wait_interruptible(inode); + return !inode_dio_finished(inode) ? -ERESTARTSYS : 0; } /* Call with exclusively locked inode->i_rwsem */ @@ -46,7 +34,7 @@ static int netfs_block_o_direct(struct netfs_inode *ictx) if (!test_bit(NETFS_ICTX_ODIRECT, &ictx->flags)) return 0; clear_bit(NETFS_ICTX_ODIRECT, &ictx->flags); - return inode_dio_wait_interruptible(&ictx->inode); + return netfs_inode_dio_wait_interruptible(&ictx->inode); } /** @@ -121,6 +109,7 @@ int netfs_start_io_write(struct inode *inode) up_write(&inode->i_rwsem); return -ERESTARTSYS; } + downgrade_write(&inode->i_rwsem); return 0; } EXPORT_SYMBOL(netfs_start_io_write); @@ -135,7 +124,7 @@ EXPORT_SYMBOL(netfs_start_io_write); void netfs_end_io_write(struct inode *inode) __releases(inode->i_rwsem) { - up_write(&inode->i_rwsem); + up_read(&inode->i_rwsem); } EXPORT_SYMBOL(netfs_end_io_write); diff --git a/fs/netfs/main.c b/fs/netfs/main.c index 5e77618a7940..70ecc8f5f210 100644 --- a/fs/netfs/main.c +++ b/fs/netfs/main.c @@ -7,6 +7,7 @@ #include <linux/module.h> #include <linux/export.h> +#include <linux/mempool.h> #include <linux/proc_fs.h> #include <linux/seq_file.h> #include "internal.h" @@ -23,6 +24,11 @@ unsigned netfs_debug; module_param_named(debug, netfs_debug, uint, S_IWUSR | S_IRUGO); MODULE_PARM_DESC(netfs_debug, "Netfs support debugging mask"); +static struct kmem_cache *netfs_request_slab; +static struct kmem_cache *netfs_subrequest_slab; +mempool_t netfs_request_pool; +mempool_t netfs_subrequest_pool; + #ifdef CONFIG_PROC_FS LIST_HEAD(netfs_io_requests); DEFINE_SPINLOCK(netfs_proc_lock); @@ -30,13 +36,16 @@ DEFINE_SPINLOCK(netfs_proc_lock); static const char *netfs_origins[nr__netfs_io_origin] = { [NETFS_READAHEAD] = "RA", [NETFS_READPAGE] = "RP", + [NETFS_READ_GAPS] = "RG", + [NETFS_READ_SINGLE] = "R1", [NETFS_READ_FOR_WRITE] = "RW", + [NETFS_DIO_READ] = "DR", [NETFS_WRITEBACK] = "WB", + [NETFS_WRITEBACK_SINGLE] = "W1", [NETFS_WRITETHROUGH] = "WT", - [NETFS_LAUNDER_WRITE] = "LW", [NETFS_UNBUFFERED_WRITE] = "UW", - [NETFS_DIO_READ] = "DR", [NETFS_DIO_WRITE] = "DW", + [NETFS_PGPRIV2_COPY_TO_CACHE] = "2C", }; /* @@ -56,13 +65,13 @@ static int netfs_requests_seq_show(struct seq_file *m, void *v) rreq = list_entry(v, struct netfs_io_request, proc_link); seq_printf(m, - "%08x %s %3d %2lx %4d %3d @%04llx %zx/%zx", + "%08x %s %3d %2lx %4ld %3d @%04llx %llx/%llx", rreq->debug_id, netfs_origins[rreq->origin], refcount_read(&rreq->ref), rreq->flags, rreq->error, - atomic_read(&rreq->nr_outstanding), + 0, rreq->start, rreq->submitted, rreq->len); seq_putc(m, '\n'); return 0; @@ -98,25 +107,58 @@ static int __init netfs_init(void) { int ret = -ENOMEM; + netfs_request_slab = kmem_cache_create("netfs_request", + sizeof(struct netfs_io_request), 0, + SLAB_HWCACHE_ALIGN | SLAB_ACCOUNT, + NULL); + if (!netfs_request_slab) + goto error_req; + + if (mempool_init_slab_pool(&netfs_request_pool, 100, netfs_request_slab) < 0) + goto error_reqpool; + + netfs_subrequest_slab = kmem_cache_create("netfs_subrequest", + sizeof(struct netfs_io_subrequest) + 16, 0, + SLAB_HWCACHE_ALIGN | SLAB_ACCOUNT, + NULL); + if (!netfs_subrequest_slab) + goto error_subreq; + + if (mempool_init_slab_pool(&netfs_subrequest_pool, 100, netfs_subrequest_slab) < 0) + goto error_subreqpool; + +#ifdef CONFIG_PROC_FS if (!proc_mkdir("fs/netfs", NULL)) - goto error; + goto error_proc; if (!proc_create_seq("fs/netfs/requests", S_IFREG | 0444, NULL, &netfs_requests_seq_ops)) - goto error_proc; + goto error_procfile; +#endif #ifdef CONFIG_FSCACHE_STATS if (!proc_create_single("fs/netfs/stats", S_IFREG | 0444, NULL, netfs_stats_show)) - goto error_proc; + goto error_procfile; #endif ret = fscache_init(); if (ret < 0) - goto error_proc; + goto error_fscache; return 0; +error_fscache: +#ifdef CONFIG_PROC_FS +error_procfile: + remove_proc_subtree("fs/netfs", NULL); error_proc: - remove_proc_entry("fs/netfs", NULL); -error: +#endif + mempool_exit(&netfs_subrequest_pool); +error_subreqpool: + kmem_cache_destroy(netfs_subrequest_slab); +error_subreq: + mempool_exit(&netfs_request_pool); +error_reqpool: + kmem_cache_destroy(netfs_request_slab); +error_req: return ret; } fs_initcall(netfs_init); @@ -124,6 +166,10 @@ fs_initcall(netfs_init); static void __exit netfs_exit(void) { fscache_exit(); - remove_proc_entry("fs/netfs", NULL); + remove_proc_subtree("fs/netfs", NULL); + mempool_exit(&netfs_subrequest_pool); + kmem_cache_destroy(netfs_subrequest_slab); + mempool_exit(&netfs_request_pool); + kmem_cache_destroy(netfs_request_slab); } module_exit(netfs_exit); diff --git a/fs/netfs/misc.c b/fs/netfs/misc.c index 90051ced8e2a..7099aa07737a 100644 --- a/fs/netfs/misc.c +++ b/fs/netfs/misc.c @@ -8,85 +8,118 @@ #include <linux/swap.h> #include "internal.h" -/* - * Attach a folio to the buffer and maybe set marks on it to say that we need - * to put the folio later and twiddle the pagecache flags. - */ -int netfs_xa_store_and_mark(struct xarray *xa, unsigned long index, - struct folio *folio, unsigned int flags, - gfp_t gfp_mask) -{ - XA_STATE_ORDER(xas, xa, index, folio_order(folio)); - -retry: - xas_lock(&xas); - for (;;) { - xas_store(&xas, folio); - if (!xas_error(&xas)) - break; - xas_unlock(&xas); - if (!xas_nomem(&xas, gfp_mask)) - return xas_error(&xas); - goto retry; - } - - if (flags & NETFS_FLAG_PUT_MARK) - xas_set_mark(&xas, NETFS_BUF_PUT_MARK); - if (flags & NETFS_FLAG_PAGECACHE_MARK) - xas_set_mark(&xas, NETFS_BUF_PAGECACHE_MARK); - xas_unlock(&xas); - return xas_error(&xas); -} - -/* - * Create the specified range of folios in the buffer attached to the read - * request. The folios are marked with NETFS_BUF_PUT_MARK so that we know that - * these need freeing later. +/** + * netfs_alloc_folioq_buffer - Allocate buffer space into a folio queue + * @mapping: Address space to set on the folio (or NULL). + * @_buffer: Pointer to the folio queue to add to (may point to a NULL; updated). + * @_cur_size: Current size of the buffer (updated). + * @size: Target size of the buffer. + * @gfp: The allocation constraints. */ -int netfs_add_folios_to_buffer(struct xarray *buffer, - struct address_space *mapping, - pgoff_t index, pgoff_t to, gfp_t gfp_mask) +int netfs_alloc_folioq_buffer(struct address_space *mapping, + struct folio_queue **_buffer, + size_t *_cur_size, ssize_t size, gfp_t gfp) { - struct folio *folio; - int ret; + struct folio_queue *tail = *_buffer, *p; - if (to + 1 == index) /* Page range is inclusive */ + size = round_up(size, PAGE_SIZE); + if (*_cur_size >= size) return 0; + if (tail) + while (tail->next) + tail = tail->next; + do { - /* TODO: Figure out what order folio can be allocated here */ - folio = filemap_alloc_folio(readahead_gfp_mask(mapping), 0); + struct folio *folio; + int order = 0, slot; + + if (!tail || folioq_full(tail)) { + p = netfs_folioq_alloc(0, GFP_NOFS, netfs_trace_folioq_alloc_buffer); + if (!p) + return -ENOMEM; + if (tail) { + tail->next = p; + p->prev = tail; + } else { + *_buffer = p; + } + tail = p; + } + + if (size - *_cur_size > PAGE_SIZE) + order = umin(ilog2(size - *_cur_size) - PAGE_SHIFT, + MAX_PAGECACHE_ORDER); + + folio = folio_alloc(gfp, order); + if (!folio && order > 0) + folio = folio_alloc(gfp, 0); if (!folio) return -ENOMEM; - folio->index = index; - ret = netfs_xa_store_and_mark(buffer, index, folio, - NETFS_FLAG_PUT_MARK, gfp_mask); - if (ret < 0) { - folio_put(folio); - return ret; - } - index += folio_nr_pages(folio); - } while (index <= to && index != 0); + folio->mapping = mapping; + folio->index = *_cur_size / PAGE_SIZE; + trace_netfs_folio(folio, netfs_folio_trace_alloc_buffer); + slot = folioq_append_mark(tail, folio); + *_cur_size += folioq_folio_size(tail, slot); + } while (*_cur_size < size); return 0; } +EXPORT_SYMBOL(netfs_alloc_folioq_buffer); -/* - * Clear an xarray buffer, putting a ref on the folios that have - * NETFS_BUF_PUT_MARK set. +/** + * netfs_free_folioq_buffer - Free a folio queue. + * @fq: The start of the folio queue to free + * + * Free up a chain of folio_queues and, if marked, the marked folios they point + * to. */ -void netfs_clear_buffer(struct xarray *buffer) +void netfs_free_folioq_buffer(struct folio_queue *fq) { - struct folio *folio; - XA_STATE(xas, buffer, 0); + struct folio_queue *next; + struct folio_batch fbatch; + + folio_batch_init(&fbatch); - rcu_read_lock(); - xas_for_each_marked(&xas, folio, ULONG_MAX, NETFS_BUF_PUT_MARK) { - folio_put(folio); + for (; fq; fq = next) { + for (int slot = 0; slot < folioq_count(fq); slot++) { + struct folio *folio = folioq_folio(fq, slot); + + if (!folio || + !folioq_is_marked(fq, slot)) + continue; + + trace_netfs_folio(folio, netfs_folio_trace_put); + if (folio_batch_add(&fbatch, folio)) + folio_batch_release(&fbatch); + } + + netfs_stat_d(&netfs_n_folioq); + next = fq->next; + kfree(fq); } - rcu_read_unlock(); - xa_destroy(buffer); + + folio_batch_release(&fbatch); +} +EXPORT_SYMBOL(netfs_free_folioq_buffer); + +/* + * Reset the subrequest iterator to refer just to the region remaining to be + * read. The iterator may or may not have been advanced by socket ops or + * extraction ops to an extent that may or may not match the amount actually + * read. + */ +void netfs_reset_iter(struct netfs_io_subrequest *subreq) +{ + struct iov_iter *io_iter = &subreq->io_iter; + size_t remain = subreq->len - subreq->transferred; + + if (io_iter->count > remain) + iov_iter_advance(io_iter, io_iter->count - remain); + else if (io_iter->count < remain) + iov_iter_revert(io_iter, remain - io_iter->count); + iov_iter_truncate(&subreq->io_iter, remain); } /** @@ -177,12 +210,22 @@ EXPORT_SYMBOL(netfs_clear_inode_writeback); */ void netfs_invalidate_folio(struct folio *folio, size_t offset, size_t length) { - struct netfs_folio *finfo = NULL; + struct netfs_folio *finfo; + struct netfs_inode *ctx = netfs_inode(folio_inode(folio)); size_t flen = folio_size(folio); _enter("{%lx},%zx,%zx", folio->index, offset, length); - folio_wait_fscache(folio); + if (offset == 0 && length == flen) { + unsigned long long i_size = i_size_read(&ctx->inode); + unsigned long long fpos = folio_pos(folio), end; + + end = umin(fpos + flen, i_size); + if (fpos < i_size && end > ctx->zero_point) + ctx->zero_point = end; + } + + folio_wait_private_2(folio); /* [DEPRECATED] */ if (!folio_test_private(folio)) return; @@ -196,18 +239,34 @@ void netfs_invalidate_folio(struct folio *folio, size_t offset, size_t length) /* We have a partially uptodate page from a streaming write. */ unsigned int fstart = finfo->dirty_offset; unsigned int fend = fstart + finfo->dirty_len; - unsigned int end = offset + length; + unsigned int iend = offset + length; if (offset >= fend) return; - if (end <= fstart) + if (iend <= fstart) + return; + + /* The invalidation region overlaps the data. If the region + * covers the start of the data, we either move along the start + * or just erase the data entirely. + */ + if (offset <= fstart) { + if (iend >= fend) + goto erase_completely; + /* Move the start of the data. */ + finfo->dirty_len = fend - iend; + finfo->dirty_offset = offset; + return; + } + + /* Reduce the length of the data if the invalidation region + * covers the tail part. + */ + if (iend >= fend) { + finfo->dirty_len = offset - fstart; return; - if (offset <= fstart && end >= fend) - goto erase_completely; - if (offset <= fstart && end > fstart) - goto reduce_len; - if (offset > fstart && end >= fend) - goto move_start; + } + /* A partial write was split. The caller has already zeroed * it, so just absorb the hole. */ @@ -220,12 +279,6 @@ erase_completely: folio_clear_uptodate(folio); kfree(finfo); return; -reduce_len: - finfo->dirty_len = offset + length - finfo->dirty_offset; - return; -move_start: - finfo->dirty_len -= offset - finfo->dirty_offset; - finfo->dirty_offset = offset; } EXPORT_SYMBOL(netfs_invalidate_folio); @@ -242,18 +295,20 @@ bool netfs_release_folio(struct folio *folio, gfp_t gfp) struct netfs_inode *ctx = netfs_inode(folio_inode(folio)); unsigned long long end; - end = folio_pos(folio) + folio_size(folio); + if (folio_test_dirty(folio)) + return false; + + end = umin(folio_pos(folio) + folio_size(folio), i_size_read(&ctx->inode)); if (end > ctx->zero_point) ctx->zero_point = end; if (folio_test_private(folio)) return false; - if (folio_test_fscache(folio)) { + if (unlikely(folio_test_private_2(folio))) { /* [DEPRECATED] */ if (current_is_kswapd() || !(gfp & __GFP_FS)) return false; - folio_wait_fscache(folio); + folio_wait_private_2(folio); } - fscache_note_page_release(netfs_i_cookie(ctx)); return true; } diff --git a/fs/netfs/objects.c b/fs/netfs/objects.c index 610ceb5bd86c..dc6b41ef18b0 100644 --- a/fs/netfs/objects.c +++ b/fs/netfs/objects.c @@ -6,6 +6,8 @@ */ #include <linux/slab.h> +#include <linux/mempool.h> +#include <linux/delay.h> #include "internal.h" /* @@ -20,43 +22,59 @@ struct netfs_io_request *netfs_alloc_request(struct address_space *mapping, struct inode *inode = file ? file_inode(file) : mapping->host; struct netfs_inode *ctx = netfs_inode(inode); struct netfs_io_request *rreq; - bool is_unbuffered = (origin == NETFS_UNBUFFERED_WRITE || - origin == NETFS_DIO_READ || - origin == NETFS_DIO_WRITE); - bool cached = !is_unbuffered && netfs_is_cache_enabled(ctx); + mempool_t *mempool = ctx->ops->request_pool ?: &netfs_request_pool; + struct kmem_cache *cache = mempool->pool_data; int ret; - rreq = kzalloc(ctx->ops->io_request_size ?: sizeof(struct netfs_io_request), - GFP_KERNEL); - if (!rreq) - return ERR_PTR(-ENOMEM); + for (;;) { + rreq = mempool_alloc(mempool, GFP_KERNEL); + if (rreq) + break; + msleep(10); + } + memset(rreq, 0, kmem_cache_size(cache)); rreq->start = start; rreq->len = len; - rreq->upper_len = len; rreq->origin = origin; rreq->netfs_ops = ctx->ops; rreq->mapping = mapping; rreq->inode = inode; rreq->i_size = i_size_read(inode); rreq->debug_id = atomic_inc_return(&debug_ids); - INIT_LIST_HEAD(&rreq->subrequests); - INIT_WORK(&rreq->work, NULL); + rreq->wsize = INT_MAX; + rreq->io_streams[0].sreq_max_len = ULONG_MAX; + rreq->io_streams[0].sreq_max_segs = 0; + spin_lock_init(&rreq->lock); + INIT_LIST_HEAD(&rreq->io_streams[0].subrequests); + INIT_LIST_HEAD(&rreq->io_streams[1].subrequests); + init_waitqueue_head(&rreq->waitq); refcount_set(&rreq->ref, 1); + if (origin == NETFS_READAHEAD || + origin == NETFS_READPAGE || + origin == NETFS_READ_GAPS || + origin == NETFS_READ_SINGLE || + origin == NETFS_READ_FOR_WRITE || + origin == NETFS_DIO_READ) { + INIT_WORK(&rreq->work, netfs_read_collection_worker); + rreq->io_streams[0].avail = true; + } else { + INIT_WORK(&rreq->work, netfs_write_collection_worker); + } + __set_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags); - if (cached) - __set_bit(NETFS_RREQ_WRITE_TO_CACHE, &rreq->flags); if (file && file->f_flags & O_NONBLOCK) __set_bit(NETFS_RREQ_NONBLOCK, &rreq->flags); if (rreq->netfs_ops->init_request) { ret = rreq->netfs_ops->init_request(rreq, file); if (ret < 0) { - kfree(rreq); + mempool_free(rreq, rreq->netfs_ops->request_pool ?: &netfs_request_pool); return ERR_PTR(ret); } } + atomic_inc(&ctx->io_count); trace_netfs_rreq_ref(rreq->debug_id, 1, netfs_rreq_trace_new); netfs_proc_add_rreq(rreq); netfs_stat(&netfs_n_rh_rreq); @@ -74,20 +92,34 @@ void netfs_get_request(struct netfs_io_request *rreq, enum netfs_rreq_ref_trace void netfs_clear_subrequests(struct netfs_io_request *rreq, bool was_async) { struct netfs_io_subrequest *subreq; - - while (!list_empty(&rreq->subrequests)) { - subreq = list_first_entry(&rreq->subrequests, - struct netfs_io_subrequest, rreq_link); - list_del(&subreq->rreq_link); - netfs_put_subrequest(subreq, was_async, - netfs_sreq_trace_put_clear); + struct netfs_io_stream *stream; + int s; + + for (s = 0; s < ARRAY_SIZE(rreq->io_streams); s++) { + stream = &rreq->io_streams[s]; + while (!list_empty(&stream->subrequests)) { + subreq = list_first_entry(&stream->subrequests, + struct netfs_io_subrequest, rreq_link); + list_del(&subreq->rreq_link); + netfs_put_subrequest(subreq, was_async, + netfs_sreq_trace_put_clear); + } } } +static void netfs_free_request_rcu(struct rcu_head *rcu) +{ + struct netfs_io_request *rreq = container_of(rcu, struct netfs_io_request, rcu); + + mempool_free(rreq, rreq->netfs_ops->request_pool ?: &netfs_request_pool); + netfs_stat_d(&netfs_n_rh_rreq); +} + static void netfs_free_request(struct work_struct *work) { struct netfs_io_request *rreq = container_of(work, struct netfs_io_request, work); + struct netfs_inode *ictx = netfs_inode(rreq->inode); unsigned int i; trace_netfs_rreq(rreq, netfs_rreq_trace_free); @@ -106,8 +138,11 @@ static void netfs_free_request(struct work_struct *work) } kvfree(rreq->direct_bv); } - kfree_rcu(rreq, rcu); - netfs_stat_d(&netfs_n_rh_rreq); + rolling_buffer_clear(&rreq->buffer); + + if (atomic_dec_and_test(&ictx->io_count)) + wake_up_var(&ictx->io_count); + call_rcu(&rreq->rcu, netfs_free_request_rcu); } void netfs_put_request(struct netfs_io_request *rreq, bool was_async, @@ -125,7 +160,7 @@ void netfs_put_request(struct netfs_io_request *rreq, bool was_async, if (was_async) { rreq->work.func = netfs_free_request; if (!queue_work(system_unbound_wq, &rreq->work)) - BUG(); + WARN_ON(1); } else { netfs_free_request(&rreq->work); } @@ -139,19 +174,25 @@ void netfs_put_request(struct netfs_io_request *rreq, bool was_async, struct netfs_io_subrequest *netfs_alloc_subrequest(struct netfs_io_request *rreq) { struct netfs_io_subrequest *subreq; - - subreq = kzalloc(rreq->netfs_ops->io_subrequest_size ?: - sizeof(struct netfs_io_subrequest), - GFP_KERNEL); - if (subreq) { - INIT_WORK(&subreq->work, NULL); - INIT_LIST_HEAD(&subreq->rreq_link); - refcount_set(&subreq->ref, 2); - subreq->rreq = rreq; - netfs_get_request(rreq, netfs_rreq_trace_get_subreq); - netfs_stat(&netfs_n_rh_sreq); + mempool_t *mempool = rreq->netfs_ops->subrequest_pool ?: &netfs_subrequest_pool; + struct kmem_cache *cache = mempool->pool_data; + + for (;;) { + subreq = mempool_alloc(rreq->netfs_ops->subrequest_pool ?: &netfs_subrequest_pool, + GFP_KERNEL); + if (subreq) + break; + msleep(10); } + memset(subreq, 0, kmem_cache_size(cache)); + INIT_WORK(&subreq->work, NULL); + INIT_LIST_HEAD(&subreq->rreq_link); + refcount_set(&subreq->ref, 2); + subreq->rreq = rreq; + subreq->debug_index = atomic_inc_return(&rreq->subreq_counter); + netfs_get_request(rreq, netfs_rreq_trace_get_subreq); + netfs_stat(&netfs_n_rh_sreq); return subreq; } @@ -173,7 +214,7 @@ static void netfs_free_subrequest(struct netfs_io_subrequest *subreq, trace_netfs_sreq(subreq, netfs_sreq_trace_free); if (rreq->netfs_ops->free_subrequest) rreq->netfs_ops->free_subrequest(subreq); - kfree(subreq); + mempool_free(subreq, rreq->netfs_ops->subrequest_pool ?: &netfs_subrequest_pool); netfs_stat_d(&netfs_n_rh_sreq); netfs_put_request(rreq, was_async, netfs_rreq_trace_put_subreq); } diff --git a/fs/netfs/output.c b/fs/netfs/output.c deleted file mode 100644 index 625eb68f3e5a..000000000000 --- a/fs/netfs/output.c +++ /dev/null @@ -1,478 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* Network filesystem high-level write support. - * - * Copyright (C) 2023 Red Hat, Inc. All Rights Reserved. - * Written by David Howells (dhowells@redhat.com) - */ - -#include <linux/fs.h> -#include <linux/mm.h> -#include <linux/pagemap.h> -#include <linux/slab.h> -#include <linux/writeback.h> -#include <linux/pagevec.h> -#include "internal.h" - -/** - * netfs_create_write_request - Create a write operation. - * @wreq: The write request this is storing from. - * @dest: The destination type - * @start: Start of the region this write will modify - * @len: Length of the modification - * @worker: The worker function to handle the write(s) - * - * Allocate a write operation, set it up and add it to the list on a write - * request. - */ -struct netfs_io_subrequest *netfs_create_write_request(struct netfs_io_request *wreq, - enum netfs_io_source dest, - loff_t start, size_t len, - work_func_t worker) -{ - struct netfs_io_subrequest *subreq; - - subreq = netfs_alloc_subrequest(wreq); - if (subreq) { - INIT_WORK(&subreq->work, worker); - subreq->source = dest; - subreq->start = start; - subreq->len = len; - subreq->debug_index = wreq->subreq_counter++; - - switch (subreq->source) { - case NETFS_UPLOAD_TO_SERVER: - netfs_stat(&netfs_n_wh_upload); - break; - case NETFS_WRITE_TO_CACHE: - netfs_stat(&netfs_n_wh_write); - break; - default: - BUG(); - } - - subreq->io_iter = wreq->io_iter; - iov_iter_advance(&subreq->io_iter, subreq->start - wreq->start); - iov_iter_truncate(&subreq->io_iter, subreq->len); - - trace_netfs_sreq_ref(wreq->debug_id, subreq->debug_index, - refcount_read(&subreq->ref), - netfs_sreq_trace_new); - atomic_inc(&wreq->nr_outstanding); - list_add_tail(&subreq->rreq_link, &wreq->subrequests); - trace_netfs_sreq(subreq, netfs_sreq_trace_prepare); - } - - return subreq; -} -EXPORT_SYMBOL(netfs_create_write_request); - -/* - * Process a completed write request once all the component operations have - * been completed. - */ -static void netfs_write_terminated(struct netfs_io_request *wreq, bool was_async) -{ - struct netfs_io_subrequest *subreq; - struct netfs_inode *ctx = netfs_inode(wreq->inode); - size_t transferred = 0; - - _enter("R=%x[]", wreq->debug_id); - - trace_netfs_rreq(wreq, netfs_rreq_trace_write_done); - - list_for_each_entry(subreq, &wreq->subrequests, rreq_link) { - if (subreq->error || subreq->transferred == 0) - break; - transferred += subreq->transferred; - if (subreq->transferred < subreq->len) - break; - } - wreq->transferred = transferred; - - list_for_each_entry(subreq, &wreq->subrequests, rreq_link) { - if (!subreq->error) - continue; - switch (subreq->source) { - case NETFS_UPLOAD_TO_SERVER: - /* Depending on the type of failure, this may prevent - * writeback completion unless we're in disconnected - * mode. - */ - if (!wreq->error) - wreq->error = subreq->error; - break; - - case NETFS_WRITE_TO_CACHE: - /* Failure doesn't prevent writeback completion unless - * we're in disconnected mode. - */ - if (subreq->error != -ENOBUFS) - ctx->ops->invalidate_cache(wreq); - break; - - default: - WARN_ON_ONCE(1); - if (!wreq->error) - wreq->error = -EIO; - return; - } - } - - wreq->cleanup(wreq); - - if (wreq->origin == NETFS_DIO_WRITE && - wreq->mapping->nrpages) { - pgoff_t first = wreq->start >> PAGE_SHIFT; - pgoff_t last = (wreq->start + wreq->transferred - 1) >> PAGE_SHIFT; - invalidate_inode_pages2_range(wreq->mapping, first, last); - } - - if (wreq->origin == NETFS_DIO_WRITE) - inode_dio_end(wreq->inode); - - _debug("finished"); - trace_netfs_rreq(wreq, netfs_rreq_trace_wake_ip); - clear_bit_unlock(NETFS_RREQ_IN_PROGRESS, &wreq->flags); - wake_up_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS); - - if (wreq->iocb) { - wreq->iocb->ki_pos += transferred; - if (wreq->iocb->ki_complete) - wreq->iocb->ki_complete( - wreq->iocb, wreq->error ? wreq->error : transferred); - } - - netfs_clear_subrequests(wreq, was_async); - netfs_put_request(wreq, was_async, netfs_rreq_trace_put_complete); -} - -/* - * Deal with the completion of writing the data to the cache. - */ -void netfs_write_subrequest_terminated(void *_op, ssize_t transferred_or_error, - bool was_async) -{ - struct netfs_io_subrequest *subreq = _op; - struct netfs_io_request *wreq = subreq->rreq; - unsigned int u; - - _enter("%x[%x] %zd", wreq->debug_id, subreq->debug_index, transferred_or_error); - - switch (subreq->source) { - case NETFS_UPLOAD_TO_SERVER: - netfs_stat(&netfs_n_wh_upload_done); - break; - case NETFS_WRITE_TO_CACHE: - netfs_stat(&netfs_n_wh_write_done); - break; - case NETFS_INVALID_WRITE: - break; - default: - BUG(); - } - - if (IS_ERR_VALUE(transferred_or_error)) { - subreq->error = transferred_or_error; - trace_netfs_failure(wreq, subreq, transferred_or_error, - netfs_fail_write); - goto failed; - } - - if (WARN(transferred_or_error > subreq->len - subreq->transferred, - "Subreq excess write: R%x[%x] %zd > %zu - %zu", - wreq->debug_id, subreq->debug_index, - transferred_or_error, subreq->len, subreq->transferred)) - transferred_or_error = subreq->len - subreq->transferred; - - subreq->error = 0; - subreq->transferred += transferred_or_error; - - if (iov_iter_count(&subreq->io_iter) != subreq->len - subreq->transferred) - pr_warn("R=%08x[%u] ITER POST-MISMATCH %zx != %zx-%zx %x\n", - wreq->debug_id, subreq->debug_index, - iov_iter_count(&subreq->io_iter), subreq->len, - subreq->transferred, subreq->io_iter.iter_type); - - if (subreq->transferred < subreq->len) - goto incomplete; - - __clear_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags); -out: - trace_netfs_sreq(subreq, netfs_sreq_trace_terminated); - - /* If we decrement nr_outstanding to 0, the ref belongs to us. */ - u = atomic_dec_return(&wreq->nr_outstanding); - if (u == 0) - netfs_write_terminated(wreq, was_async); - else if (u == 1) - wake_up_var(&wreq->nr_outstanding); - - netfs_put_subrequest(subreq, was_async, netfs_sreq_trace_put_terminated); - return; - -incomplete: - if (transferred_or_error == 0) { - if (__test_and_set_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags)) { - subreq->error = -ENODATA; - goto failed; - } - } else { - __clear_bit(NETFS_SREQ_NO_PROGRESS, &subreq->flags); - } - - __set_bit(NETFS_SREQ_SHORT_IO, &subreq->flags); - set_bit(NETFS_RREQ_INCOMPLETE_IO, &wreq->flags); - goto out; - -failed: - switch (subreq->source) { - case NETFS_WRITE_TO_CACHE: - netfs_stat(&netfs_n_wh_write_failed); - set_bit(NETFS_RREQ_INCOMPLETE_IO, &wreq->flags); - break; - case NETFS_UPLOAD_TO_SERVER: - netfs_stat(&netfs_n_wh_upload_failed); - set_bit(NETFS_RREQ_FAILED, &wreq->flags); - wreq->error = subreq->error; - break; - default: - break; - } - goto out; -} -EXPORT_SYMBOL(netfs_write_subrequest_terminated); - -static void netfs_write_to_cache_op(struct netfs_io_subrequest *subreq) -{ - struct netfs_io_request *wreq = subreq->rreq; - struct netfs_cache_resources *cres = &wreq->cache_resources; - - trace_netfs_sreq(subreq, netfs_sreq_trace_submit); - - cres->ops->write(cres, subreq->start, &subreq->io_iter, - netfs_write_subrequest_terminated, subreq); -} - -static void netfs_write_to_cache_op_worker(struct work_struct *work) -{ - struct netfs_io_subrequest *subreq = - container_of(work, struct netfs_io_subrequest, work); - - netfs_write_to_cache_op(subreq); -} - -/** - * netfs_queue_write_request - Queue a write request for attention - * @subreq: The write request to be queued - * - * Queue the specified write request for processing by a worker thread. We - * pass the caller's ref on the request to the worker thread. - */ -void netfs_queue_write_request(struct netfs_io_subrequest *subreq) -{ - if (!queue_work(system_unbound_wq, &subreq->work)) - netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_wip); -} -EXPORT_SYMBOL(netfs_queue_write_request); - -/* - * Set up a op for writing to the cache. - */ -static void netfs_set_up_write_to_cache(struct netfs_io_request *wreq) -{ - struct netfs_cache_resources *cres = &wreq->cache_resources; - struct netfs_io_subrequest *subreq; - struct netfs_inode *ctx = netfs_inode(wreq->inode); - struct fscache_cookie *cookie = netfs_i_cookie(ctx); - loff_t start = wreq->start; - size_t len = wreq->len; - int ret; - - if (!fscache_cookie_enabled(cookie)) { - clear_bit(NETFS_RREQ_WRITE_TO_CACHE, &wreq->flags); - return; - } - - _debug("write to cache"); - ret = fscache_begin_write_operation(cres, cookie); - if (ret < 0) - return; - - ret = cres->ops->prepare_write(cres, &start, &len, wreq->upper_len, - i_size_read(wreq->inode), true); - if (ret < 0) - return; - - subreq = netfs_create_write_request(wreq, NETFS_WRITE_TO_CACHE, start, len, - netfs_write_to_cache_op_worker); - if (!subreq) - return; - - netfs_write_to_cache_op(subreq); -} - -/* - * Begin the process of writing out a chunk of data. - * - * We are given a write request that holds a series of dirty regions and - * (partially) covers a sequence of folios, all of which are present. The - * pages must have been marked as writeback as appropriate. - * - * We need to perform the following steps: - * - * (1) If encrypting, create an output buffer and encrypt each block of the - * data into it, otherwise the output buffer will point to the original - * folios. - * - * (2) If the data is to be cached, set up a write op for the entire output - * buffer to the cache, if the cache wants to accept it. - * - * (3) If the data is to be uploaded (ie. not merely cached): - * - * (a) If the data is to be compressed, create a compression buffer and - * compress the data into it. - * - * (b) For each destination we want to upload to, set up write ops to write - * to that destination. We may need multiple writes if the data is not - * contiguous or the span exceeds wsize for a server. - */ -int netfs_begin_write(struct netfs_io_request *wreq, bool may_wait, - enum netfs_write_trace what) -{ - struct netfs_inode *ctx = netfs_inode(wreq->inode); - - _enter("R=%x %llx-%llx f=%lx", - wreq->debug_id, wreq->start, wreq->start + wreq->len - 1, - wreq->flags); - - trace_netfs_write(wreq, what); - if (wreq->len == 0 || wreq->iter.count == 0) { - pr_err("Zero-sized write [R=%x]\n", wreq->debug_id); - return -EIO; - } - - if (wreq->origin == NETFS_DIO_WRITE) - inode_dio_begin(wreq->inode); - - wreq->io_iter = wreq->iter; - - /* ->outstanding > 0 carries a ref */ - netfs_get_request(wreq, netfs_rreq_trace_get_for_outstanding); - atomic_set(&wreq->nr_outstanding, 1); - - /* Start the encryption/compression going. We can do that in the - * background whilst we generate a list of write ops that we want to - * perform. - */ - // TODO: Encrypt or compress the region as appropriate - - /* We need to write all of the region to the cache */ - if (test_bit(NETFS_RREQ_WRITE_TO_CACHE, &wreq->flags)) - netfs_set_up_write_to_cache(wreq); - - /* However, we don't necessarily write all of the region to the server. - * Caching of reads is being managed this way also. - */ - if (test_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags)) - ctx->ops->create_write_requests(wreq, wreq->start, wreq->len); - - if (atomic_dec_and_test(&wreq->nr_outstanding)) - netfs_write_terminated(wreq, false); - - if (!may_wait) - return -EIOCBQUEUED; - - wait_on_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS, - TASK_UNINTERRUPTIBLE); - return wreq->error; -} - -/* - * Begin a write operation for writing through the pagecache. - */ -struct netfs_io_request *netfs_begin_writethrough(struct kiocb *iocb, size_t len) -{ - struct netfs_io_request *wreq; - struct file *file = iocb->ki_filp; - - wreq = netfs_alloc_request(file->f_mapping, file, iocb->ki_pos, len, - NETFS_WRITETHROUGH); - if (IS_ERR(wreq)) - return wreq; - - trace_netfs_write(wreq, netfs_write_trace_writethrough); - - __set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); - iov_iter_xarray(&wreq->iter, ITER_SOURCE, &wreq->mapping->i_pages, wreq->start, 0); - wreq->io_iter = wreq->iter; - - /* ->outstanding > 0 carries a ref */ - netfs_get_request(wreq, netfs_rreq_trace_get_for_outstanding); - atomic_set(&wreq->nr_outstanding, 1); - return wreq; -} - -static void netfs_submit_writethrough(struct netfs_io_request *wreq, bool final) -{ - struct netfs_inode *ictx = netfs_inode(wreq->inode); - unsigned long long start; - size_t len; - - if (!test_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags)) - return; - - start = wreq->start + wreq->submitted; - len = wreq->iter.count - wreq->submitted; - if (!final) { - len /= wreq->wsize; /* Round to number of maximum packets */ - len *= wreq->wsize; - } - - ictx->ops->create_write_requests(wreq, start, len); - wreq->submitted += len; -} - -/* - * Advance the state of the write operation used when writing through the - * pagecache. Data has been copied into the pagecache that we need to append - * to the request. If we've added more than wsize then we need to create a new - * subrequest. - */ -int netfs_advance_writethrough(struct netfs_io_request *wreq, size_t copied, bool to_page_end) -{ - _enter("ic=%zu sb=%zu ws=%u cp=%zu tp=%u", - wreq->iter.count, wreq->submitted, wreq->wsize, copied, to_page_end); - - wreq->iter.count += copied; - wreq->io_iter.count += copied; - if (to_page_end && wreq->io_iter.count - wreq->submitted >= wreq->wsize) - netfs_submit_writethrough(wreq, false); - - return wreq->error; -} - -/* - * End a write operation used when writing through the pagecache. - */ -int netfs_end_writethrough(struct netfs_io_request *wreq, struct kiocb *iocb) -{ - int ret = -EIOCBQUEUED; - - _enter("ic=%zu sb=%zu ws=%u", - wreq->iter.count, wreq->submitted, wreq->wsize); - - if (wreq->submitted < wreq->io_iter.count) - netfs_submit_writethrough(wreq, true); - - if (atomic_dec_and_test(&wreq->nr_outstanding)) - netfs_write_terminated(wreq, false); - - if (is_sync_kiocb(iocb)) { - wait_on_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS, - TASK_UNINTERRUPTIBLE); - ret = wreq->error; - } - - netfs_put_request(wreq, false, netfs_rreq_trace_put_return); - return ret; -} diff --git a/fs/netfs/read_collect.c b/fs/netfs/read_collect.c new file mode 100644 index 000000000000..23c75755ad4e --- /dev/null +++ b/fs/netfs/read_collect.c @@ -0,0 +1,706 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Network filesystem read subrequest result collection, assessment and + * retrying. + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include <linux/export.h> +#include <linux/fs.h> +#include <linux/mm.h> +#include <linux/pagemap.h> +#include <linux/slab.h> +#include <linux/task_io_accounting_ops.h> +#include "internal.h" + +/* Notes made in the collector */ +#define HIT_PENDING 0x01 /* A front op was still pending */ +#define MADE_PROGRESS 0x04 /* Made progress cleaning up a stream or the folio set */ +#define BUFFERED 0x08 /* The pagecache needs cleaning up */ +#define NEED_RETRY 0x10 /* A front op requests retrying */ +#define COPY_TO_CACHE 0x40 /* Need to copy subrequest to cache */ +#define ABANDON_SREQ 0x80 /* Need to abandon untransferred part of subrequest */ + +/* + * Clear the unread part of an I/O request. + */ +static void netfs_clear_unread(struct netfs_io_subrequest *subreq) +{ + netfs_reset_iter(subreq); + WARN_ON_ONCE(subreq->len - subreq->transferred != iov_iter_count(&subreq->io_iter)); + iov_iter_zero(iov_iter_count(&subreq->io_iter), &subreq->io_iter); + if (subreq->start + subreq->transferred >= subreq->rreq->i_size) + __set_bit(NETFS_SREQ_HIT_EOF, &subreq->flags); +} + +/* + * Flush, mark and unlock a folio that's now completely read. If we want to + * cache the folio, we set the group to NETFS_FOLIO_COPY_TO_CACHE, mark it + * dirty and let writeback handle it. + */ +static void netfs_unlock_read_folio(struct netfs_io_request *rreq, + struct folio_queue *folioq, + int slot) +{ + struct netfs_folio *finfo; + struct folio *folio = folioq_folio(folioq, slot); + + if (unlikely(folio_pos(folio) < rreq->abandon_to)) { + trace_netfs_folio(folio, netfs_folio_trace_abandon); + goto just_unlock; + } + + flush_dcache_folio(folio); + folio_mark_uptodate(folio); + + if (!test_bit(NETFS_RREQ_USE_PGPRIV2, &rreq->flags)) { + finfo = netfs_folio_info(folio); + if (finfo) { + trace_netfs_folio(folio, netfs_folio_trace_filled_gaps); + if (finfo->netfs_group) + folio_change_private(folio, finfo->netfs_group); + else + folio_detach_private(folio); + kfree(finfo); + } + + if (test_bit(NETFS_RREQ_FOLIO_COPY_TO_CACHE, &rreq->flags)) { + if (!WARN_ON_ONCE(folio_get_private(folio) != NULL)) { + trace_netfs_folio(folio, netfs_folio_trace_copy_to_cache); + folio_attach_private(folio, NETFS_FOLIO_COPY_TO_CACHE); + folio_mark_dirty(folio); + } + } else { + trace_netfs_folio(folio, netfs_folio_trace_read_done); + } + + folioq_clear(folioq, slot); + } else { + // TODO: Use of PG_private_2 is deprecated. + if (test_bit(NETFS_RREQ_FOLIO_COPY_TO_CACHE, &rreq->flags)) + netfs_pgpriv2_copy_to_cache(rreq, folio); + } + +just_unlock: + if (!test_bit(NETFS_RREQ_DONT_UNLOCK_FOLIOS, &rreq->flags)) { + if (folio->index == rreq->no_unlock_folio && + test_bit(NETFS_RREQ_NO_UNLOCK_FOLIO, &rreq->flags)) { + _debug("no unlock"); + } else { + trace_netfs_folio(folio, netfs_folio_trace_read_unlock); + folio_unlock(folio); + } + } + + folioq_clear(folioq, slot); +} + +/* + * Unlock any folios we've finished with. + */ +static void netfs_read_unlock_folios(struct netfs_io_request *rreq, + unsigned int *notes) +{ + struct folio_queue *folioq = rreq->buffer.tail; + unsigned long long collected_to = rreq->collected_to; + unsigned int slot = rreq->buffer.first_tail_slot; + + if (rreq->cleaned_to >= rreq->collected_to) + return; + + // TODO: Begin decryption + + if (slot >= folioq_nr_slots(folioq)) { + folioq = rolling_buffer_delete_spent(&rreq->buffer); + if (!folioq) { + rreq->front_folio_order = 0; + return; + } + slot = 0; + } + + for (;;) { + struct folio *folio; + unsigned long long fpos, fend; + unsigned int order; + size_t fsize; + + if (*notes & COPY_TO_CACHE) + set_bit(NETFS_RREQ_FOLIO_COPY_TO_CACHE, &rreq->flags); + + folio = folioq_folio(folioq, slot); + if (WARN_ONCE(!folio_test_locked(folio), + "R=%08x: folio %lx is not locked\n", + rreq->debug_id, folio->index)) + trace_netfs_folio(folio, netfs_folio_trace_not_locked); + + order = folioq_folio_order(folioq, slot); + rreq->front_folio_order = order; + fsize = PAGE_SIZE << order; + fpos = folio_pos(folio); + fend = umin(fpos + fsize, rreq->i_size); + + trace_netfs_collect_folio(rreq, folio, fend, collected_to); + + /* Unlock any folio we've transferred all of. */ + if (collected_to < fend) + break; + + netfs_unlock_read_folio(rreq, folioq, slot); + WRITE_ONCE(rreq->cleaned_to, fpos + fsize); + *notes |= MADE_PROGRESS; + + clear_bit(NETFS_RREQ_FOLIO_COPY_TO_CACHE, &rreq->flags); + + /* Clean up the head folioq. If we clear an entire folioq, then + * we can get rid of it provided it's not also the tail folioq + * being filled by the issuer. + */ + folioq_clear(folioq, slot); + slot++; + if (slot >= folioq_nr_slots(folioq)) { + folioq = rolling_buffer_delete_spent(&rreq->buffer); + if (!folioq) + goto done; + slot = 0; + trace_netfs_folioq(folioq, netfs_trace_folioq_read_progress); + } + + if (fpos + fsize >= collected_to) + break; + } + + rreq->buffer.tail = folioq; +done: + rreq->buffer.first_tail_slot = slot; +} + +/* + * Collect and assess the results of various read subrequests. We may need to + * retry some of the results. + * + * Note that we have a sequence of subrequests, which may be drawing on + * different sources and may or may not be the same size or starting position + * and may not even correspond in boundary alignment. + */ +static void netfs_collect_read_results(struct netfs_io_request *rreq) +{ + struct netfs_io_subrequest *front, *remove; + struct netfs_io_stream *stream = &rreq->io_streams[0]; + unsigned int notes; + + _enter("%llx-%llx", rreq->start, rreq->start + rreq->len); + trace_netfs_rreq(rreq, netfs_rreq_trace_collect); + trace_netfs_collect(rreq); + +reassess: + if (rreq->origin == NETFS_READAHEAD || + rreq->origin == NETFS_READPAGE || + rreq->origin == NETFS_READ_FOR_WRITE) + notes = BUFFERED; + else + notes = 0; + + /* Remove completed subrequests from the front of the stream and + * advance the completion point. We stop when we hit something that's + * in progress. The issuer thread may be adding stuff to the tail + * whilst we're doing this. + */ + front = READ_ONCE(stream->front); + while (front) { + size_t transferred; + + trace_netfs_collect_sreq(rreq, front); + _debug("sreq [%x] %llx %zx/%zx", + front->debug_index, front->start, front->transferred, front->len); + + if (stream->collected_to < front->start) { + trace_netfs_collect_gap(rreq, stream, front->start, 'F'); + stream->collected_to = front->start; + } + + if (test_bit(NETFS_SREQ_IN_PROGRESS, &front->flags)) + notes |= HIT_PENDING; + smp_rmb(); /* Read counters after IN_PROGRESS flag. */ + transferred = READ_ONCE(front->transferred); + + /* If we can now collect the next folio, do so. We don't want + * to defer this as we have to decide whether we need to copy + * to the cache or not, and that may differ between adjacent + * subreqs. + */ + if (notes & BUFFERED) { + size_t fsize = PAGE_SIZE << rreq->front_folio_order; + + /* Clear the tail of a short read. */ + if (!(notes & HIT_PENDING) && + front->error == 0 && + transferred < front->len && + (test_bit(NETFS_SREQ_HIT_EOF, &front->flags) || + test_bit(NETFS_SREQ_CLEAR_TAIL, &front->flags))) { + netfs_clear_unread(front); + transferred = front->transferred = front->len; + trace_netfs_sreq(front, netfs_sreq_trace_clear); + } + + stream->collected_to = front->start + transferred; + rreq->collected_to = stream->collected_to; + + if (test_bit(NETFS_SREQ_COPY_TO_CACHE, &front->flags)) + notes |= COPY_TO_CACHE; + + if (test_bit(NETFS_SREQ_FAILED, &front->flags)) { + rreq->abandon_to = front->start + front->len; + front->transferred = front->len; + transferred = front->len; + trace_netfs_rreq(rreq, netfs_rreq_trace_set_abandon); + } + if (front->start + transferred >= rreq->cleaned_to + fsize || + test_bit(NETFS_SREQ_HIT_EOF, &front->flags)) + netfs_read_unlock_folios(rreq, ¬es); + } else { + stream->collected_to = front->start + transferred; + rreq->collected_to = stream->collected_to; + } + + /* Stall if the front is still undergoing I/O. */ + if (notes & HIT_PENDING) + break; + + if (test_bit(NETFS_SREQ_FAILED, &front->flags)) { + if (!stream->failed) { + stream->error = front->error; + rreq->error = front->error; + set_bit(NETFS_RREQ_FAILED, &rreq->flags); + stream->failed = true; + } + notes |= MADE_PROGRESS | ABANDON_SREQ; + } else if (test_bit(NETFS_SREQ_NEED_RETRY, &front->flags)) { + stream->need_retry = true; + notes |= NEED_RETRY | MADE_PROGRESS; + break; + } else { + if (!stream->failed) + stream->transferred = stream->collected_to - rreq->start; + notes |= MADE_PROGRESS; + } + + /* Remove if completely consumed. */ + stream->source = front->source; + spin_lock(&rreq->lock); + + remove = front; + trace_netfs_sreq(front, netfs_sreq_trace_discard); + list_del_init(&front->rreq_link); + front = list_first_entry_or_null(&stream->subrequests, + struct netfs_io_subrequest, rreq_link); + stream->front = front; + spin_unlock(&rreq->lock); + netfs_put_subrequest(remove, false, + notes & ABANDON_SREQ ? + netfs_sreq_trace_put_abandon : + netfs_sreq_trace_put_done); + } + + trace_netfs_collect_stream(rreq, stream); + trace_netfs_collect_state(rreq, rreq->collected_to, notes); + + if (!(notes & BUFFERED)) + rreq->cleaned_to = rreq->collected_to; + + if (notes & NEED_RETRY) + goto need_retry; + if ((notes & MADE_PROGRESS) && test_bit(NETFS_RREQ_PAUSE, &rreq->flags)) { + trace_netfs_rreq(rreq, netfs_rreq_trace_unpause); + clear_bit_unlock(NETFS_RREQ_PAUSE, &rreq->flags); + smp_mb__after_atomic(); /* Set PAUSE before task state */ + wake_up(&rreq->waitq); + } + + if (notes & MADE_PROGRESS) { + //cond_resched(); + goto reassess; + } + +out: + _leave(" = %x", notes); + return; + +need_retry: + /* Okay... We're going to have to retry parts of the stream. Note + * that any partially completed op will have had any wholly transferred + * folios removed from it. + */ + _debug("retry"); + netfs_retry_reads(rreq); + goto out; +} + +/* + * Do page flushing and suchlike after DIO. + */ +static void netfs_rreq_assess_dio(struct netfs_io_request *rreq) +{ + struct netfs_io_subrequest *subreq; + struct netfs_io_stream *stream = &rreq->io_streams[0]; + unsigned int i; + + /* Collect unbuffered reads and direct reads, adding up the transfer + * sizes until we find the first short or failed subrequest. + */ + list_for_each_entry(subreq, &stream->subrequests, rreq_link) { + rreq->transferred += subreq->transferred; + + if (subreq->transferred < subreq->len || + test_bit(NETFS_SREQ_FAILED, &subreq->flags)) { + rreq->error = subreq->error; + break; + } + } + + if (rreq->origin == NETFS_DIO_READ) { + for (i = 0; i < rreq->direct_bv_count; i++) { + flush_dcache_page(rreq->direct_bv[i].bv_page); + // TODO: cifs marks pages in the destination buffer + // dirty under some circumstances after a read. Do we + // need to do that too? + set_page_dirty(rreq->direct_bv[i].bv_page); + } + } + + if (rreq->iocb) { + rreq->iocb->ki_pos += rreq->transferred; + if (rreq->iocb->ki_complete) + rreq->iocb->ki_complete( + rreq->iocb, rreq->error ? rreq->error : rreq->transferred); + } + if (rreq->netfs_ops->done) + rreq->netfs_ops->done(rreq); + if (rreq->origin == NETFS_DIO_READ) + inode_dio_end(rreq->inode); +} + +/* + * Do processing after reading a monolithic single object. + */ +static void netfs_rreq_assess_single(struct netfs_io_request *rreq) +{ + struct netfs_io_stream *stream = &rreq->io_streams[0]; + + if (!rreq->error && stream->source == NETFS_DOWNLOAD_FROM_SERVER && + fscache_resources_valid(&rreq->cache_resources)) { + trace_netfs_rreq(rreq, netfs_rreq_trace_dirty); + netfs_single_mark_inode_dirty(rreq->inode); + } + + if (rreq->iocb) { + rreq->iocb->ki_pos += rreq->transferred; + if (rreq->iocb->ki_complete) + rreq->iocb->ki_complete( + rreq->iocb, rreq->error ? rreq->error : rreq->transferred); + } + if (rreq->netfs_ops->done) + rreq->netfs_ops->done(rreq); +} + +/* + * Perform the collection of subrequests and folios. + * + * Note that we're in normal kernel thread context at this point, possibly + * running on a workqueue. + */ +static void netfs_read_collection(struct netfs_io_request *rreq) +{ + struct netfs_io_stream *stream = &rreq->io_streams[0]; + + netfs_collect_read_results(rreq); + + /* We're done when the app thread has finished posting subreqs and the + * queue is empty. + */ + if (!test_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags)) + return; + smp_rmb(); /* Read ALL_QUEUED before subreq lists. */ + + if (!list_empty(&stream->subrequests)) + return; + + /* Okay, declare that all I/O is complete. */ + rreq->transferred = stream->transferred; + trace_netfs_rreq(rreq, netfs_rreq_trace_complete); + + //netfs_rreq_is_still_valid(rreq); + + switch (rreq->origin) { + case NETFS_DIO_READ: + case NETFS_READ_GAPS: + netfs_rreq_assess_dio(rreq); + break; + case NETFS_READ_SINGLE: + netfs_rreq_assess_single(rreq); + break; + default: + break; + } + task_io_account_read(rreq->transferred); + + trace_netfs_rreq(rreq, netfs_rreq_trace_wake_ip); + clear_and_wake_up_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags); + + trace_netfs_rreq(rreq, netfs_rreq_trace_done); + netfs_clear_subrequests(rreq, false); + netfs_unlock_abandoned_read_pages(rreq); + if (unlikely(rreq->copy_to_cache)) + netfs_pgpriv2_end_copy_to_cache(rreq); +} + +void netfs_read_collection_worker(struct work_struct *work) +{ + struct netfs_io_request *rreq = container_of(work, struct netfs_io_request, work); + + netfs_see_request(rreq, netfs_rreq_trace_see_work); + if (test_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags)) + netfs_read_collection(rreq); + netfs_put_request(rreq, false, netfs_rreq_trace_put_work); +} + +/* + * Wake the collection work item. + */ +void netfs_wake_read_collector(struct netfs_io_request *rreq) +{ + if (test_bit(NETFS_RREQ_OFFLOAD_COLLECTION, &rreq->flags) && + !test_bit(NETFS_RREQ_RETRYING, &rreq->flags)) { + if (!work_pending(&rreq->work)) { + netfs_get_request(rreq, netfs_rreq_trace_get_work); + if (!queue_work(system_unbound_wq, &rreq->work)) + netfs_put_request(rreq, true, netfs_rreq_trace_put_work_nq); + } + } else { + trace_netfs_rreq(rreq, netfs_rreq_trace_wake_queue); + wake_up(&rreq->waitq); + } +} + +/** + * netfs_read_subreq_progress - Note progress of a read operation. + * @subreq: The read request that has terminated. + * + * This tells the read side of netfs lib that a contributory I/O operation has + * made some progress and that it may be possible to unlock some folios. + * + * Before calling, the filesystem should update subreq->transferred to track + * the amount of data copied into the output buffer. + */ +void netfs_read_subreq_progress(struct netfs_io_subrequest *subreq) +{ + struct netfs_io_request *rreq = subreq->rreq; + struct netfs_io_stream *stream = &rreq->io_streams[0]; + size_t fsize = PAGE_SIZE << rreq->front_folio_order; + + trace_netfs_sreq(subreq, netfs_sreq_trace_progress); + + /* If we are at the head of the queue, wake up the collector, + * getting a ref to it if we were the ones to do so. + */ + if (subreq->start + subreq->transferred > rreq->cleaned_to + fsize && + (rreq->origin == NETFS_READAHEAD || + rreq->origin == NETFS_READPAGE || + rreq->origin == NETFS_READ_FOR_WRITE) && + list_is_first(&subreq->rreq_link, &stream->subrequests) + ) { + __set_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); + netfs_wake_read_collector(rreq); + } +} +EXPORT_SYMBOL(netfs_read_subreq_progress); + +/** + * netfs_read_subreq_terminated - Note the termination of an I/O operation. + * @subreq: The I/O request that has terminated. + * + * This tells the read helper that a contributory I/O operation has terminated, + * one way or another, and that it should integrate the results. + * + * The caller indicates the outcome of the operation through @subreq->error, + * supplying 0 to indicate a successful or retryable transfer (if + * NETFS_SREQ_NEED_RETRY is set) or a negative error code. The helper will + * look after reissuing I/O operations as appropriate and writing downloaded + * data to the cache. + * + * Before calling, the filesystem should update subreq->transferred to track + * the amount of data copied into the output buffer. + */ +void netfs_read_subreq_terminated(struct netfs_io_subrequest *subreq) +{ + struct netfs_io_request *rreq = subreq->rreq; + struct netfs_io_stream *stream = &rreq->io_streams[0]; + + switch (subreq->source) { + case NETFS_READ_FROM_CACHE: + netfs_stat(&netfs_n_rh_read_done); + break; + case NETFS_DOWNLOAD_FROM_SERVER: + netfs_stat(&netfs_n_rh_download_done); + break; + default: + break; + } + + /* Deal with retry requests, short reads and errors. If we retry + * but don't make progress, we abandon the attempt. + */ + if (!subreq->error && subreq->transferred < subreq->len) { + if (test_bit(NETFS_SREQ_HIT_EOF, &subreq->flags)) { + trace_netfs_sreq(subreq, netfs_sreq_trace_hit_eof); + } else if (test_bit(NETFS_SREQ_CLEAR_TAIL, &subreq->flags)) { + trace_netfs_sreq(subreq, netfs_sreq_trace_need_clear); + } else if (test_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) { + trace_netfs_sreq(subreq, netfs_sreq_trace_need_retry); + } else if (test_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags)) { + __set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + trace_netfs_sreq(subreq, netfs_sreq_trace_partial_read); + } else { + __set_bit(NETFS_SREQ_FAILED, &subreq->flags); + subreq->error = -ENODATA; + trace_netfs_sreq(subreq, netfs_sreq_trace_short); + } + } + + if (unlikely(subreq->error < 0)) { + trace_netfs_failure(rreq, subreq, subreq->error, netfs_fail_read); + if (subreq->source == NETFS_READ_FROM_CACHE) { + netfs_stat(&netfs_n_rh_read_failed); + __set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + } else { + netfs_stat(&netfs_n_rh_download_failed); + __set_bit(NETFS_SREQ_FAILED, &subreq->flags); + } + trace_netfs_rreq(rreq, netfs_rreq_trace_set_pause); + set_bit(NETFS_RREQ_PAUSE, &rreq->flags); + } + + trace_netfs_sreq(subreq, netfs_sreq_trace_terminated); + + clear_bit_unlock(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + smp_mb__after_atomic(); /* Clear IN_PROGRESS before task state */ + + /* If we are at the head of the queue, wake up the collector. */ + if (list_is_first(&subreq->rreq_link, &stream->subrequests) || + test_bit(NETFS_RREQ_RETRYING, &rreq->flags)) + netfs_wake_read_collector(rreq); + + netfs_put_subrequest(subreq, true, netfs_sreq_trace_put_terminated); +} +EXPORT_SYMBOL(netfs_read_subreq_terminated); + +/* + * Handle termination of a read from the cache. + */ +void netfs_cache_read_terminated(void *priv, ssize_t transferred_or_error, bool was_async) +{ + struct netfs_io_subrequest *subreq = priv; + + if (transferred_or_error > 0) { + subreq->error = 0; + if (transferred_or_error > 0) { + subreq->transferred += transferred_or_error; + __set_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); + } + } else { + subreq->error = transferred_or_error; + } + netfs_read_subreq_terminated(subreq); +} + +/* + * Wait for the read operation to complete, successfully or otherwise. + */ +ssize_t netfs_wait_for_read(struct netfs_io_request *rreq) +{ + struct netfs_io_subrequest *subreq; + struct netfs_io_stream *stream = &rreq->io_streams[0]; + DEFINE_WAIT(myself); + ssize_t ret; + + for (;;) { + trace_netfs_rreq(rreq, netfs_rreq_trace_wait_queue); + prepare_to_wait(&rreq->waitq, &myself, TASK_UNINTERRUPTIBLE); + + subreq = list_first_entry_or_null(&stream->subrequests, + struct netfs_io_subrequest, rreq_link); + if (subreq && + (!test_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags) || + test_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags))) { + __set_current_state(TASK_RUNNING); + netfs_read_collection(rreq); + continue; + } + + if (!test_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags)) + break; + + schedule(); + trace_netfs_rreq(rreq, netfs_rreq_trace_woke_queue); + } + + finish_wait(&rreq->waitq, &myself); + + ret = rreq->error; + if (ret == 0) { + ret = rreq->transferred; + switch (rreq->origin) { + case NETFS_DIO_READ: + case NETFS_READ_SINGLE: + ret = rreq->transferred; + break; + default: + if (rreq->submitted < rreq->len) { + trace_netfs_failure(rreq, NULL, ret, netfs_fail_short_read); + ret = -EIO; + } + break; + } + } + + return ret; +} + +/* + * Wait for a paused read operation to unpause or complete in some manner. + */ +void netfs_wait_for_pause(struct netfs_io_request *rreq) +{ + struct netfs_io_subrequest *subreq; + struct netfs_io_stream *stream = &rreq->io_streams[0]; + DEFINE_WAIT(myself); + + trace_netfs_rreq(rreq, netfs_rreq_trace_wait_pause); + + for (;;) { + trace_netfs_rreq(rreq, netfs_rreq_trace_wait_queue); + prepare_to_wait(&rreq->waitq, &myself, TASK_UNINTERRUPTIBLE); + + if (!test_bit(NETFS_RREQ_OFFLOAD_COLLECTION, &rreq->flags)) { + subreq = list_first_entry_or_null(&stream->subrequests, + struct netfs_io_subrequest, rreq_link); + if (subreq && + (!test_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags) || + test_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags))) { + __set_current_state(TASK_RUNNING); + netfs_read_collection(rreq); + continue; + } + } + + if (!test_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags) || + !test_bit(NETFS_RREQ_PAUSE, &rreq->flags)) + break; + + schedule(); + trace_netfs_rreq(rreq, netfs_rreq_trace_woke_queue); + } + + finish_wait(&rreq->waitq, &myself); +} diff --git a/fs/netfs/read_pgpriv2.c b/fs/netfs/read_pgpriv2.c new file mode 100644 index 000000000000..cf7727060215 --- /dev/null +++ b/fs/netfs/read_pgpriv2.c @@ -0,0 +1,227 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Read with PG_private_2 [DEPRECATED]. + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include <linux/export.h> +#include <linux/fs.h> +#include <linux/mm.h> +#include <linux/pagemap.h> +#include <linux/slab.h> +#include <linux/task_io_accounting_ops.h> +#include "internal.h" + +/* + * [DEPRECATED] Copy a folio to the cache with PG_private_2 set. + */ +static void netfs_pgpriv2_copy_folio(struct netfs_io_request *creq, struct folio *folio) +{ + struct netfs_io_stream *cache = &creq->io_streams[1]; + size_t fsize = folio_size(folio), flen = fsize; + loff_t fpos = folio_pos(folio), i_size; + bool to_eof = false; + + _enter(""); + + /* netfs_perform_write() may shift i_size around the page or from out + * of the page to beyond it, but cannot move i_size into or through the + * page since we have it locked. + */ + i_size = i_size_read(creq->inode); + + if (fpos >= i_size) { + /* mmap beyond eof. */ + _debug("beyond eof"); + folio_end_private_2(folio); + return; + } + + if (fpos + fsize > creq->i_size) + creq->i_size = i_size; + + if (flen > i_size - fpos) { + flen = i_size - fpos; + to_eof = true; + } else if (flen == i_size - fpos) { + to_eof = true; + } + + _debug("folio %zx %zx", flen, fsize); + + trace_netfs_folio(folio, netfs_folio_trace_store_copy); + + /* Attach the folio to the rolling buffer. */ + if (rolling_buffer_append(&creq->buffer, folio, 0) < 0) { + clear_bit(NETFS_RREQ_FOLIO_COPY_TO_CACHE, &creq->flags); + return; + } + + cache->submit_extendable_to = fsize; + cache->submit_off = 0; + cache->submit_len = flen; + + /* Attach the folio to one or more subrequests. For a big folio, we + * could end up with thousands of subrequests if the wsize is small - + * but we might need to wait during the creation of subrequests for + * network resources (eg. SMB credits). + */ + do { + ssize_t part; + + creq->buffer.iter.iov_offset = cache->submit_off; + + atomic64_set(&creq->issued_to, fpos + cache->submit_off); + cache->submit_extendable_to = fsize - cache->submit_off; + part = netfs_advance_write(creq, cache, fpos + cache->submit_off, + cache->submit_len, to_eof); + cache->submit_off += part; + if (part > cache->submit_len) + cache->submit_len = 0; + else + cache->submit_len -= part; + } while (cache->submit_len > 0); + + creq->buffer.iter.iov_offset = 0; + rolling_buffer_advance(&creq->buffer, fsize); + atomic64_set(&creq->issued_to, fpos + fsize); + + if (flen < fsize) + netfs_issue_write(creq, cache); +} + +/* + * [DEPRECATED] Set up copying to the cache. + */ +static struct netfs_io_request *netfs_pgpriv2_begin_copy_to_cache( + struct netfs_io_request *rreq, struct folio *folio) +{ + struct netfs_io_request *creq; + + if (!fscache_resources_valid(&rreq->cache_resources)) + goto cancel; + + creq = netfs_create_write_req(rreq->mapping, NULL, folio_pos(folio), + NETFS_PGPRIV2_COPY_TO_CACHE); + if (IS_ERR(creq)) + goto cancel; + + if (!creq->io_streams[1].avail) + goto cancel_put; + + trace_netfs_write(creq, netfs_write_trace_copy_to_cache); + netfs_stat(&netfs_n_wh_copy_to_cache); + rreq->copy_to_cache = creq; + return creq; + +cancel_put: + netfs_put_request(creq, false, netfs_rreq_trace_put_return); +cancel: + rreq->copy_to_cache = ERR_PTR(-ENOBUFS); + clear_bit(NETFS_RREQ_FOLIO_COPY_TO_CACHE, &rreq->flags); + return ERR_PTR(-ENOBUFS); +} + +/* + * [DEPRECATED] Mark page as requiring copy-to-cache using PG_private_2 and add + * it to the copy write request. + */ +void netfs_pgpriv2_copy_to_cache(struct netfs_io_request *rreq, struct folio *folio) +{ + struct netfs_io_request *creq = rreq->copy_to_cache; + + if (!creq) + creq = netfs_pgpriv2_begin_copy_to_cache(rreq, folio); + if (IS_ERR(creq)) + return; + + trace_netfs_folio(folio, netfs_folio_trace_copy_to_cache); + folio_start_private_2(folio); + netfs_pgpriv2_copy_folio(creq, folio); +} + +/* + * [DEPRECATED] End writing to the cache, flushing out any outstanding writes. + */ +void netfs_pgpriv2_end_copy_to_cache(struct netfs_io_request *rreq) +{ + struct netfs_io_request *creq = rreq->copy_to_cache; + + if (IS_ERR_OR_NULL(creq)) + return; + + netfs_issue_write(creq, &creq->io_streams[1]); + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &creq->flags); + + netfs_put_request(creq, false, netfs_rreq_trace_put_return); + creq->copy_to_cache = NULL; +} + +/* + * [DEPRECATED] Remove the PG_private_2 mark from any folios we've finished + * copying. + */ +bool netfs_pgpriv2_unlock_copied_folios(struct netfs_io_request *creq) +{ + struct folio_queue *folioq = creq->buffer.tail; + unsigned long long collected_to = creq->collected_to; + unsigned int slot = creq->buffer.first_tail_slot; + bool made_progress = false; + + if (slot >= folioq_nr_slots(folioq)) { + folioq = rolling_buffer_delete_spent(&creq->buffer); + slot = 0; + } + + for (;;) { + struct folio *folio; + unsigned long long fpos, fend; + size_t fsize, flen; + + folio = folioq_folio(folioq, slot); + if (WARN_ONCE(!folio_test_private_2(folio), + "R=%08x: folio %lx is not marked private_2\n", + creq->debug_id, folio->index)) + trace_netfs_folio(folio, netfs_folio_trace_not_under_wback); + + fpos = folio_pos(folio); + fsize = folio_size(folio); + flen = fsize; + + fend = min_t(unsigned long long, fpos + flen, creq->i_size); + + trace_netfs_collect_folio(creq, folio, fend, collected_to); + + /* Unlock any folio we've transferred all of. */ + if (collected_to < fend) + break; + + trace_netfs_folio(folio, netfs_folio_trace_end_copy); + folio_end_private_2(folio); + creq->cleaned_to = fpos + fsize; + made_progress = true; + + /* Clean up the head folioq. If we clear an entire folioq, then + * we can get rid of it provided it's not also the tail folioq + * being filled by the issuer. + */ + folioq_clear(folioq, slot); + slot++; + if (slot >= folioq_nr_slots(folioq)) { + folioq = rolling_buffer_delete_spent(&creq->buffer); + if (!folioq) + goto done; + slot = 0; + } + + if (fpos + fsize >= collected_to) + break; + } + + creq->buffer.tail = folioq; +done: + creq->buffer.first_tail_slot = slot; + return made_progress; +} diff --git a/fs/netfs/read_retry.c b/fs/netfs/read_retry.c new file mode 100644 index 000000000000..0f294b26e08c --- /dev/null +++ b/fs/netfs/read_retry.c @@ -0,0 +1,313 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Network filesystem read subrequest retrying. + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include <linux/fs.h> +#include <linux/slab.h> +#include "internal.h" + +static void netfs_reissue_read(struct netfs_io_request *rreq, + struct netfs_io_subrequest *subreq) +{ + __clear_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); + __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + netfs_stat(&netfs_n_rh_retry_read_subreq); + subreq->rreq->netfs_ops->issue_read(subreq); +} + +/* + * Go through the list of failed/short reads, retrying all retryable ones. We + * need to switch failed cache reads to network downloads. + */ +static void netfs_retry_read_subrequests(struct netfs_io_request *rreq) +{ + struct netfs_io_subrequest *subreq; + struct netfs_io_stream *stream = &rreq->io_streams[0]; + struct list_head *next; + + _enter("R=%x", rreq->debug_id); + + if (list_empty(&stream->subrequests)) + return; + + if (rreq->netfs_ops->retry_request) + rreq->netfs_ops->retry_request(rreq, NULL); + + /* If there's no renegotiation to do, just resend each retryable subreq + * up to the first permanently failed one. + */ + if (!rreq->netfs_ops->prepare_read && + !rreq->cache_resources.ops) { + list_for_each_entry(subreq, &stream->subrequests, rreq_link) { + if (test_bit(NETFS_SREQ_FAILED, &subreq->flags)) + break; + if (__test_and_clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) { + __clear_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); + subreq->retry_count++; + netfs_reset_iter(subreq); + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); + netfs_reissue_read(rreq, subreq); + } + } + return; + } + + /* Okay, we need to renegotiate all the download requests and flip any + * failed cache reads over to being download requests and negotiate + * those also. All fully successful subreqs have been removed from the + * list and any spare data from those has been donated. + * + * What we do is decant the list and rebuild it one subreq at a time so + * that we don't end up with donations jumping over a gap we're busy + * populating with smaller subrequests. In the event that the subreq + * we just launched finishes before we insert the next subreq, it'll + * fill in rreq->prev_donated instead. + * + * Note: Alternatively, we could split the tail subrequest right before + * we reissue it and fix up the donations under lock. + */ + next = stream->subrequests.next; + + do { + struct netfs_io_subrequest *from, *to, *tmp; + struct iov_iter source; + unsigned long long start, len; + size_t part; + bool boundary = false, subreq_superfluous = false; + + /* Go through the subreqs and find the next span of contiguous + * buffer that we then rejig (cifs, for example, needs the + * rsize renegotiating) and reissue. + */ + from = list_entry(next, struct netfs_io_subrequest, rreq_link); + to = from; + start = from->start + from->transferred; + len = from->len - from->transferred; + + _debug("from R=%08x[%x] s=%llx ctl=%zx/%zx", + rreq->debug_id, from->debug_index, + from->start, from->transferred, from->len); + + if (test_bit(NETFS_SREQ_FAILED, &from->flags) || + !test_bit(NETFS_SREQ_NEED_RETRY, &from->flags)) + goto abandon; + + list_for_each_continue(next, &stream->subrequests) { + subreq = list_entry(next, struct netfs_io_subrequest, rreq_link); + if (subreq->start + subreq->transferred != start + len || + test_bit(NETFS_SREQ_BOUNDARY, &subreq->flags) || + !test_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) + break; + to = subreq; + len += to->len; + } + + _debug(" - range: %llx-%llx %llx", start, start + len - 1, len); + + /* Determine the set of buffers we're going to use. Each + * subreq gets a subset of a single overall contiguous buffer. + */ + netfs_reset_iter(from); + source = from->io_iter; + source.count = len; + + /* Work through the sublist. */ + subreq = from; + list_for_each_entry_from(subreq, &stream->subrequests, rreq_link) { + if (!len) { + subreq_superfluous = true; + break; + } + subreq->source = NETFS_DOWNLOAD_FROM_SERVER; + subreq->start = start - subreq->transferred; + subreq->len = len + subreq->transferred; + __clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + __clear_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); + subreq->retry_count++; + + trace_netfs_sreq(subreq, netfs_sreq_trace_retry); + + /* Renegotiate max_len (rsize) */ + stream->sreq_max_len = subreq->len; + if (rreq->netfs_ops->prepare_read && + rreq->netfs_ops->prepare_read(subreq) < 0) { + trace_netfs_sreq(subreq, netfs_sreq_trace_reprep_failed); + __set_bit(NETFS_SREQ_FAILED, &subreq->flags); + goto abandon; + } + + part = umin(len, stream->sreq_max_len); + if (unlikely(stream->sreq_max_segs)) + part = netfs_limit_iter(&source, 0, part, stream->sreq_max_segs); + subreq->len = subreq->transferred + part; + subreq->io_iter = source; + iov_iter_truncate(&subreq->io_iter, part); + iov_iter_advance(&source, part); + len -= part; + start += part; + if (!len) { + if (boundary) + __set_bit(NETFS_SREQ_BOUNDARY, &subreq->flags); + } else { + __clear_bit(NETFS_SREQ_BOUNDARY, &subreq->flags); + } + + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); + netfs_reissue_read(rreq, subreq); + if (subreq == to) { + subreq_superfluous = false; + break; + } + } + + /* If we managed to use fewer subreqs, we can discard the + * excess; if we used the same number, then we're done. + */ + if (!len) { + if (!subreq_superfluous) + continue; + list_for_each_entry_safe_from(subreq, tmp, + &stream->subrequests, rreq_link) { + trace_netfs_sreq(subreq, netfs_sreq_trace_superfluous); + list_del(&subreq->rreq_link); + netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_done); + if (subreq == to) + break; + } + continue; + } + + /* We ran out of subrequests, so we need to allocate some more + * and insert them after. + */ + do { + subreq = netfs_alloc_subrequest(rreq); + if (!subreq) { + subreq = to; + goto abandon_after; + } + subreq->source = NETFS_DOWNLOAD_FROM_SERVER; + subreq->start = start; + subreq->len = len; + subreq->stream_nr = stream->stream_nr; + subreq->retry_count = 1; + + trace_netfs_sreq_ref(rreq->debug_id, subreq->debug_index, + refcount_read(&subreq->ref), + netfs_sreq_trace_new); + + list_add(&subreq->rreq_link, &to->rreq_link); + to = list_next_entry(to, rreq_link); + trace_netfs_sreq(subreq, netfs_sreq_trace_retry); + + stream->sreq_max_len = umin(len, rreq->rsize); + stream->sreq_max_segs = 0; + if (unlikely(stream->sreq_max_segs)) + part = netfs_limit_iter(&source, 0, part, stream->sreq_max_segs); + + netfs_stat(&netfs_n_rh_download); + if (rreq->netfs_ops->prepare_read(subreq) < 0) { + trace_netfs_sreq(subreq, netfs_sreq_trace_reprep_failed); + __set_bit(NETFS_SREQ_FAILED, &subreq->flags); + goto abandon; + } + + part = umin(len, stream->sreq_max_len); + subreq->len = subreq->transferred + part; + subreq->io_iter = source; + iov_iter_truncate(&subreq->io_iter, part); + iov_iter_advance(&source, part); + + len -= part; + start += part; + if (!len && boundary) { + __set_bit(NETFS_SREQ_BOUNDARY, &to->flags); + boundary = false; + } + + netfs_reissue_read(rreq, subreq); + } while (len); + + } while (!list_is_head(next, &stream->subrequests)); + + return; + + /* If we hit an error, fail all remaining incomplete subrequests */ +abandon_after: + if (list_is_last(&subreq->rreq_link, &stream->subrequests)) + return; + subreq = list_next_entry(subreq, rreq_link); +abandon: + list_for_each_entry_from(subreq, &stream->subrequests, rreq_link) { + if (!subreq->error && + !test_bit(NETFS_SREQ_FAILED, &subreq->flags) && + !test_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) + continue; + subreq->error = -ENOMEM; + __set_bit(NETFS_SREQ_FAILED, &subreq->flags); + __clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + } +} + +/* + * Retry reads. + */ +void netfs_retry_reads(struct netfs_io_request *rreq) +{ + struct netfs_io_subrequest *subreq; + struct netfs_io_stream *stream = &rreq->io_streams[0]; + DEFINE_WAIT(myself); + + netfs_stat(&netfs_n_rh_retry_read_req); + + set_bit(NETFS_RREQ_RETRYING, &rreq->flags); + + /* Wait for all outstanding I/O to quiesce before performing retries as + * we may need to renegotiate the I/O sizes. + */ + list_for_each_entry(subreq, &stream->subrequests, rreq_link) { + if (!test_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags)) + continue; + + trace_netfs_rreq(rreq, netfs_rreq_trace_wait_queue); + for (;;) { + prepare_to_wait(&rreq->waitq, &myself, TASK_UNINTERRUPTIBLE); + + if (!test_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags)) + break; + + trace_netfs_sreq(subreq, netfs_sreq_trace_wait_for); + schedule(); + trace_netfs_rreq(rreq, netfs_rreq_trace_woke_queue); + } + + finish_wait(&rreq->waitq, &myself); + } + clear_bit(NETFS_RREQ_RETRYING, &rreq->flags); + + trace_netfs_rreq(rreq, netfs_rreq_trace_resubmit); + netfs_retry_read_subrequests(rreq); +} + +/* + * Unlock any the pages that haven't been unlocked yet due to abandoned + * subrequests. + */ +void netfs_unlock_abandoned_read_pages(struct netfs_io_request *rreq) +{ + struct folio_queue *p; + + for (p = rreq->buffer.tail; p; p = p->next) { + for (int slot = 0; slot < folioq_count(p); slot++) { + struct folio *folio = folioq_folio(p, slot); + + if (folio && !folioq_is_marked2(p, slot)) { + trace_netfs_folio(folio, netfs_folio_trace_abandon); + folio_unlock(folio); + } + } + } +} diff --git a/fs/netfs/read_single.c b/fs/netfs/read_single.c new file mode 100644 index 000000000000..fea0ecdecc53 --- /dev/null +++ b/fs/netfs/read_single.c @@ -0,0 +1,195 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* Single, monolithic object support (e.g. AFS directory). + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include <linux/export.h> +#include <linux/fs.h> +#include <linux/mm.h> +#include <linux/pagemap.h> +#include <linux/slab.h> +#include <linux/uio.h> +#include <linux/sched/mm.h> +#include <linux/task_io_accounting_ops.h> +#include <linux/netfs.h> +#include "internal.h" + +/** + * netfs_single_mark_inode_dirty - Mark a single, monolithic object inode dirty + * @inode: The inode to mark + * + * Mark an inode that contains a single, monolithic object as dirty so that its + * writepages op will get called. If set, the SINGLE_NO_UPLOAD flag indicates + * that the object will only be written to the cache and not uploaded (e.g. AFS + * directory contents). + */ +void netfs_single_mark_inode_dirty(struct inode *inode) +{ + struct netfs_inode *ictx = netfs_inode(inode); + bool cache_only = test_bit(NETFS_ICTX_SINGLE_NO_UPLOAD, &ictx->flags); + bool caching = fscache_cookie_enabled(netfs_i_cookie(netfs_inode(inode))); + + if (cache_only && !caching) + return; + + mark_inode_dirty(inode); + + if (caching && !(inode->i_state & I_PINNING_NETFS_WB)) { + bool need_use = false; + + spin_lock(&inode->i_lock); + if (!(inode->i_state & I_PINNING_NETFS_WB)) { + inode->i_state |= I_PINNING_NETFS_WB; + need_use = true; + } + spin_unlock(&inode->i_lock); + + if (need_use) + fscache_use_cookie(netfs_i_cookie(ictx), true); + } + +} +EXPORT_SYMBOL(netfs_single_mark_inode_dirty); + +static int netfs_single_begin_cache_read(struct netfs_io_request *rreq, struct netfs_inode *ctx) +{ + return fscache_begin_read_operation(&rreq->cache_resources, netfs_i_cookie(ctx)); +} + +static void netfs_single_cache_prepare_read(struct netfs_io_request *rreq, + struct netfs_io_subrequest *subreq) +{ + struct netfs_cache_resources *cres = &rreq->cache_resources; + + if (!cres->ops) { + subreq->source = NETFS_DOWNLOAD_FROM_SERVER; + return; + } + subreq->source = cres->ops->prepare_read(subreq, rreq->i_size); + trace_netfs_sreq(subreq, netfs_sreq_trace_prepare); + +} + +static void netfs_single_read_cache(struct netfs_io_request *rreq, + struct netfs_io_subrequest *subreq) +{ + struct netfs_cache_resources *cres = &rreq->cache_resources; + + _enter("R=%08x[%x]", rreq->debug_id, subreq->debug_index); + netfs_stat(&netfs_n_rh_read); + cres->ops->read(cres, subreq->start, &subreq->io_iter, NETFS_READ_HOLE_FAIL, + netfs_cache_read_terminated, subreq); +} + +/* + * Perform a read to a buffer from the cache or the server. Only a single + * subreq is permitted as the object must be fetched in a single transaction. + */ +static int netfs_single_dispatch_read(struct netfs_io_request *rreq) +{ + struct netfs_io_stream *stream = &rreq->io_streams[0]; + struct netfs_io_subrequest *subreq; + int ret = 0; + + subreq = netfs_alloc_subrequest(rreq); + if (!subreq) + return -ENOMEM; + + subreq->source = NETFS_SOURCE_UNKNOWN; + subreq->start = 0; + subreq->len = rreq->len; + subreq->io_iter = rreq->buffer.iter; + + __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + + spin_lock(&rreq->lock); + list_add_tail(&subreq->rreq_link, &stream->subrequests); + trace_netfs_sreq(subreq, netfs_sreq_trace_added); + stream->front = subreq; + /* Store list pointers before active flag */ + smp_store_release(&stream->active, true); + spin_unlock(&rreq->lock); + + netfs_single_cache_prepare_read(rreq, subreq); + switch (subreq->source) { + case NETFS_DOWNLOAD_FROM_SERVER: + netfs_stat(&netfs_n_rh_download); + if (rreq->netfs_ops->prepare_read) { + ret = rreq->netfs_ops->prepare_read(subreq); + if (ret < 0) + goto cancel; + } + + rreq->netfs_ops->issue_read(subreq); + rreq->submitted += subreq->len; + break; + case NETFS_READ_FROM_CACHE: + trace_netfs_sreq(subreq, netfs_sreq_trace_submit); + netfs_single_read_cache(rreq, subreq); + rreq->submitted += subreq->len; + ret = 0; + break; + default: + pr_warn("Unexpected single-read source %u\n", subreq->source); + WARN_ON_ONCE(true); + ret = -EIO; + break; + } + + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &rreq->flags); + return ret; +cancel: + netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_cancel); + return ret; +} + +/** + * netfs_read_single - Synchronously read a single blob of pages. + * @inode: The inode to read from. + * @file: The file we're using to read or NULL. + * @iter: The buffer we're reading into. + * + * Fulfil a read request for a single monolithic object by drawing data from + * the cache if possible, or the netfs if not. The buffer may be larger than + * the file content; unused beyond the EOF will be zero-filled. The content + * will be read with a single I/O request (though this may be retried). + * + * The calling netfs must initialise a netfs context contiguous to the vfs + * inode before calling this. + * + * This is usable whether or not caching is enabled. If caching is enabled, + * the data will be stored as a single object into the cache. + */ +ssize_t netfs_read_single(struct inode *inode, struct file *file, struct iov_iter *iter) +{ + struct netfs_io_request *rreq; + struct netfs_inode *ictx = netfs_inode(inode); + ssize_t ret; + + rreq = netfs_alloc_request(inode->i_mapping, file, 0, iov_iter_count(iter), + NETFS_READ_SINGLE); + if (IS_ERR(rreq)) + return PTR_ERR(rreq); + + ret = netfs_single_begin_cache_read(rreq, ictx); + if (ret == -ENOMEM || ret == -EINTR || ret == -ERESTARTSYS) + goto cleanup_free; + + netfs_stat(&netfs_n_rh_read_single); + trace_netfs_read(rreq, 0, rreq->len, netfs_read_trace_read_single); + + rreq->buffer.iter = *iter; + netfs_single_dispatch_read(rreq); + + ret = netfs_wait_for_read(rreq); + netfs_put_request(rreq, true, netfs_rreq_trace_put_return); + return ret; + +cleanup_free: + netfs_put_request(rreq, false, netfs_rreq_trace_put_failed); + return ret; +} +EXPORT_SYMBOL(netfs_read_single); diff --git a/fs/netfs/rolling_buffer.c b/fs/netfs/rolling_buffer.c new file mode 100644 index 000000000000..207b6a326651 --- /dev/null +++ b/fs/netfs/rolling_buffer.c @@ -0,0 +1,222 @@ +// SPDX-License-Identifier: GPL-2.0-or-later +/* Rolling buffer helpers + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include <linux/bitops.h> +#include <linux/pagemap.h> +#include <linux/rolling_buffer.h> +#include <linux/slab.h> +#include "internal.h" + +static atomic_t debug_ids; + +/** + * netfs_folioq_alloc - Allocate a folio_queue struct + * @rreq_id: Associated debugging ID for tracing purposes + * @gfp: Allocation constraints + * @trace: Trace tag to indicate the purpose of the allocation + * + * Allocate, initialise and account the folio_queue struct and log a trace line + * to mark the allocation. + */ +struct folio_queue *netfs_folioq_alloc(unsigned int rreq_id, gfp_t gfp, + unsigned int /*enum netfs_folioq_trace*/ trace) +{ + struct folio_queue *fq; + + fq = kmalloc(sizeof(*fq), gfp); + if (fq) { + netfs_stat(&netfs_n_folioq); + folioq_init(fq, rreq_id); + fq->debug_id = atomic_inc_return(&debug_ids); + trace_netfs_folioq(fq, trace); + } + return fq; +} +EXPORT_SYMBOL(netfs_folioq_alloc); + +/** + * netfs_folioq_free - Free a folio_queue struct + * @folioq: The object to free + * @trace: Trace tag to indicate which free + * + * Free and unaccount the folio_queue struct. + */ +void netfs_folioq_free(struct folio_queue *folioq, + unsigned int /*enum netfs_trace_folioq*/ trace) +{ + trace_netfs_folioq(folioq, trace); + netfs_stat_d(&netfs_n_folioq); + kfree(folioq); +} +EXPORT_SYMBOL(netfs_folioq_free); + +/* + * Initialise a rolling buffer. We allocate an empty folio queue struct to so + * that the pointers can be independently driven by the producer and the + * consumer. + */ +int rolling_buffer_init(struct rolling_buffer *roll, unsigned int rreq_id, + unsigned int direction) +{ + struct folio_queue *fq; + + fq = netfs_folioq_alloc(rreq_id, GFP_NOFS, netfs_trace_folioq_rollbuf_init); + if (!fq) + return -ENOMEM; + + roll->head = fq; + roll->tail = fq; + iov_iter_folio_queue(&roll->iter, direction, fq, 0, 0, 0); + return 0; +} + +/* + * Add another folio_queue to a rolling buffer if there's no space left. + */ +int rolling_buffer_make_space(struct rolling_buffer *roll) +{ + struct folio_queue *fq, *head = roll->head; + + if (!folioq_full(head)) + return 0; + + fq = netfs_folioq_alloc(head->rreq_id, GFP_NOFS, netfs_trace_folioq_make_space); + if (!fq) + return -ENOMEM; + fq->prev = head; + + roll->head = fq; + if (folioq_full(head)) { + /* Make sure we don't leave the master iterator pointing to a + * block that might get immediately consumed. + */ + if (roll->iter.folioq == head && + roll->iter.folioq_slot == folioq_nr_slots(head)) { + roll->iter.folioq = fq; + roll->iter.folioq_slot = 0; + } + } + + /* Make sure the initialisation is stored before the next pointer. + * + * [!] NOTE: After we set head->next, the consumer is at liberty to + * immediately delete the old head. + */ + smp_store_release(&head->next, fq); + return 0; +} + +/* + * Decant the list of folios to read into a rolling buffer. + */ +ssize_t rolling_buffer_load_from_ra(struct rolling_buffer *roll, + struct readahead_control *ractl, + struct folio_batch *put_batch) +{ + struct folio_queue *fq; + struct page **vec; + int nr, ix, to; + ssize_t size = 0; + + if (rolling_buffer_make_space(roll) < 0) + return -ENOMEM; + + fq = roll->head; + vec = (struct page **)fq->vec.folios; + nr = __readahead_batch(ractl, vec + folio_batch_count(&fq->vec), + folio_batch_space(&fq->vec)); + ix = fq->vec.nr; + to = ix + nr; + fq->vec.nr = to; + for (; ix < to; ix++) { + struct folio *folio = folioq_folio(fq, ix); + unsigned int order = folio_order(folio); + + fq->orders[ix] = order; + size += PAGE_SIZE << order; + trace_netfs_folio(folio, netfs_folio_trace_read); + if (!folio_batch_add(put_batch, folio)) + folio_batch_release(put_batch); + } + WRITE_ONCE(roll->iter.count, roll->iter.count + size); + + /* Store the counter after setting the slot. */ + smp_store_release(&roll->next_head_slot, to); + return size; +} + +/* + * Append a folio to the rolling buffer. + */ +ssize_t rolling_buffer_append(struct rolling_buffer *roll, struct folio *folio, + unsigned int flags) +{ + ssize_t size = folio_size(folio); + int slot; + + if (rolling_buffer_make_space(roll) < 0) + return -ENOMEM; + + slot = folioq_append(roll->head, folio); + if (flags & ROLLBUF_MARK_1) + folioq_mark(roll->head, slot); + if (flags & ROLLBUF_MARK_2) + folioq_mark2(roll->head, slot); + + WRITE_ONCE(roll->iter.count, roll->iter.count + size); + + /* Store the counter after setting the slot. */ + smp_store_release(&roll->next_head_slot, slot); + return size; +} + +/* + * Delete a spent buffer from a rolling queue and return the next in line. We + * don't return the last buffer to keep the pointers independent, but return + * NULL instead. + */ +struct folio_queue *rolling_buffer_delete_spent(struct rolling_buffer *roll) +{ + struct folio_queue *spent = roll->tail, *next = READ_ONCE(spent->next); + + if (!next) + return NULL; + next->prev = NULL; + netfs_folioq_free(spent, netfs_trace_folioq_delete); + roll->tail = next; + return next; +} + +/* + * Clear out a rolling queue. Folios that have mark 1 set are put. + */ +void rolling_buffer_clear(struct rolling_buffer *roll) +{ + struct folio_batch fbatch; + struct folio_queue *p; + + folio_batch_init(&fbatch); + + while ((p = roll->tail)) { + roll->tail = p->next; + for (int slot = 0; slot < folioq_count(p); slot++) { + struct folio *folio = folioq_folio(p, slot); + + if (!folio) + continue; + if (folioq_is_marked(p, slot)) { + trace_netfs_folio(folio, netfs_folio_trace_put); + if (!folio_batch_add(&fbatch, folio)) + folio_batch_release(&fbatch); + } + } + + netfs_folioq_free(p, netfs_trace_folioq_clear); + } + + folio_batch_release(&fbatch); +} diff --git a/fs/netfs/stats.c b/fs/netfs/stats.c index deeba9f9dcf5..ab6b916addc4 100644 --- a/fs/netfs/stats.c +++ b/fs/netfs/stats.c @@ -10,9 +10,9 @@ #include "internal.h" atomic_t netfs_n_rh_dio_read; -atomic_t netfs_n_rh_dio_write; atomic_t netfs_n_rh_readahead; -atomic_t netfs_n_rh_readpage; +atomic_t netfs_n_rh_read_folio; +atomic_t netfs_n_rh_read_single; atomic_t netfs_n_rh_rreq; atomic_t netfs_n_rh_sreq; atomic_t netfs_n_rh_download; @@ -29,6 +29,13 @@ atomic_t netfs_n_rh_write_begin; atomic_t netfs_n_rh_write_done; atomic_t netfs_n_rh_write_failed; atomic_t netfs_n_rh_write_zskip; +atomic_t netfs_n_rh_retry_read_req; +atomic_t netfs_n_rh_retry_read_subreq; +atomic_t netfs_n_wh_buffered_write; +atomic_t netfs_n_wh_writethrough; +atomic_t netfs_n_wh_dio_write; +atomic_t netfs_n_wh_writepages; +atomic_t netfs_n_wh_copy_to_cache; atomic_t netfs_n_wh_wstream_conflict; atomic_t netfs_n_wh_upload; atomic_t netfs_n_wh_upload_done; @@ -36,41 +43,61 @@ atomic_t netfs_n_wh_upload_failed; atomic_t netfs_n_wh_write; atomic_t netfs_n_wh_write_done; atomic_t netfs_n_wh_write_failed; +atomic_t netfs_n_wh_retry_write_req; +atomic_t netfs_n_wh_retry_write_subreq; +atomic_t netfs_n_wb_lock_skip; +atomic_t netfs_n_wb_lock_wait; +atomic_t netfs_n_folioq; int netfs_stats_show(struct seq_file *m, void *v) { - seq_printf(m, "Netfs : DR=%u DW=%u RA=%u RP=%u WB=%u WBZ=%u\n", + seq_printf(m, "Reads : DR=%u RA=%u RF=%u RS=%u WB=%u WBZ=%u\n", atomic_read(&netfs_n_rh_dio_read), - atomic_read(&netfs_n_rh_dio_write), atomic_read(&netfs_n_rh_readahead), - atomic_read(&netfs_n_rh_readpage), + atomic_read(&netfs_n_rh_read_folio), + atomic_read(&netfs_n_rh_read_single), atomic_read(&netfs_n_rh_write_begin), atomic_read(&netfs_n_rh_write_zskip)); - seq_printf(m, "Netfs : ZR=%u sh=%u sk=%u\n", + seq_printf(m, "Writes : BW=%u WT=%u DW=%u WP=%u 2C=%u\n", + atomic_read(&netfs_n_wh_buffered_write), + atomic_read(&netfs_n_wh_writethrough), + atomic_read(&netfs_n_wh_dio_write), + atomic_read(&netfs_n_wh_writepages), + atomic_read(&netfs_n_wh_copy_to_cache)); + seq_printf(m, "ZeroOps: ZR=%u sh=%u sk=%u\n", atomic_read(&netfs_n_rh_zero), atomic_read(&netfs_n_rh_short_read), atomic_read(&netfs_n_rh_write_zskip)); - seq_printf(m, "Netfs : DL=%u ds=%u df=%u di=%u\n", + seq_printf(m, "DownOps: DL=%u ds=%u df=%u di=%u\n", atomic_read(&netfs_n_rh_download), atomic_read(&netfs_n_rh_download_done), atomic_read(&netfs_n_rh_download_failed), atomic_read(&netfs_n_rh_download_instead)); - seq_printf(m, "Netfs : RD=%u rs=%u rf=%u\n", + seq_printf(m, "CaRdOps: RD=%u rs=%u rf=%u\n", atomic_read(&netfs_n_rh_read), atomic_read(&netfs_n_rh_read_done), atomic_read(&netfs_n_rh_read_failed)); - seq_printf(m, "Netfs : UL=%u us=%u uf=%u\n", + seq_printf(m, "UpldOps: UL=%u us=%u uf=%u\n", atomic_read(&netfs_n_wh_upload), atomic_read(&netfs_n_wh_upload_done), atomic_read(&netfs_n_wh_upload_failed)); - seq_printf(m, "Netfs : WR=%u ws=%u wf=%u\n", + seq_printf(m, "CaWrOps: WR=%u ws=%u wf=%u\n", atomic_read(&netfs_n_wh_write), atomic_read(&netfs_n_wh_write_done), atomic_read(&netfs_n_wh_write_failed)); - seq_printf(m, "Netfs : rr=%u sr=%u wsc=%u\n", + seq_printf(m, "Retries: rq=%u rs=%u wq=%u ws=%u\n", + atomic_read(&netfs_n_rh_retry_read_req), + atomic_read(&netfs_n_rh_retry_read_subreq), + atomic_read(&netfs_n_wh_retry_write_req), + atomic_read(&netfs_n_wh_retry_write_subreq)); + seq_printf(m, "Objs : rr=%u sr=%u foq=%u wsc=%u\n", atomic_read(&netfs_n_rh_rreq), atomic_read(&netfs_n_rh_sreq), + atomic_read(&netfs_n_folioq), atomic_read(&netfs_n_wh_wstream_conflict)); + seq_printf(m, "WbLock : skip=%u wait=%u\n", + atomic_read(&netfs_n_wb_lock_skip), + atomic_read(&netfs_n_wb_lock_wait)); return fscache_stats_show(m); } EXPORT_SYMBOL(netfs_stats_show); diff --git a/fs/netfs/write_collect.c b/fs/netfs/write_collect.c new file mode 100644 index 000000000000..3fca59e6475d --- /dev/null +++ b/fs/netfs/write_collect.c @@ -0,0 +1,550 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Network filesystem write subrequest result collection, assessment + * and retrying. + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include <linux/export.h> +#include <linux/fs.h> +#include <linux/mm.h> +#include <linux/pagemap.h> +#include <linux/slab.h> +#include "internal.h" + +/* Notes made in the collector */ +#define HIT_PENDING 0x01 /* A front op was still pending */ +#define NEED_REASSESS 0x02 /* Need to loop round and reassess */ +#define MADE_PROGRESS 0x04 /* Made progress cleaning up a stream or the folio set */ +#define NEED_UNLOCK 0x08 /* The pagecache needs unlocking */ +#define NEED_RETRY 0x10 /* A front op requests retrying */ +#define SAW_FAILURE 0x20 /* One stream or hit a permanent failure */ + +static void netfs_dump_request(const struct netfs_io_request *rreq) +{ + pr_err("Request R=%08x r=%d fl=%lx or=%x e=%ld\n", + rreq->debug_id, refcount_read(&rreq->ref), rreq->flags, + rreq->origin, rreq->error); + pr_err(" st=%llx tsl=%zx/%llx/%llx\n", + rreq->start, rreq->transferred, rreq->submitted, rreq->len); + pr_err(" cci=%llx/%llx/%llx\n", + rreq->cleaned_to, rreq->collected_to, atomic64_read(&rreq->issued_to)); + pr_err(" iw=%pSR\n", rreq->netfs_ops->issue_write); + for (int i = 0; i < NR_IO_STREAMS; i++) { + const struct netfs_io_subrequest *sreq; + const struct netfs_io_stream *s = &rreq->io_streams[i]; + + pr_err(" str[%x] s=%x e=%d acnf=%u,%u,%u,%u\n", + s->stream_nr, s->source, s->error, + s->avail, s->active, s->need_retry, s->failed); + pr_err(" str[%x] ct=%llx t=%zx\n", + s->stream_nr, s->collected_to, s->transferred); + list_for_each_entry(sreq, &s->subrequests, rreq_link) { + pr_err(" sreq[%x:%x] sc=%u s=%llx t=%zx/%zx r=%d f=%lx\n", + sreq->stream_nr, sreq->debug_index, sreq->source, + sreq->start, sreq->transferred, sreq->len, + refcount_read(&sreq->ref), sreq->flags); + } + } +} + +/* + * Successful completion of write of a folio to the server and/or cache. Note + * that we are not allowed to lock the folio here on pain of deadlocking with + * truncate. + */ +int netfs_folio_written_back(struct folio *folio) +{ + enum netfs_folio_trace why = netfs_folio_trace_clear; + struct netfs_inode *ictx = netfs_inode(folio->mapping->host); + struct netfs_folio *finfo; + struct netfs_group *group = NULL; + int gcount = 0; + + if ((finfo = netfs_folio_info(folio))) { + /* Streaming writes cannot be redirtied whilst under writeback, + * so discard the streaming record. + */ + unsigned long long fend; + + fend = folio_pos(folio) + finfo->dirty_offset + finfo->dirty_len; + if (fend > ictx->zero_point) + ictx->zero_point = fend; + + folio_detach_private(folio); + group = finfo->netfs_group; + gcount++; + kfree(finfo); + why = netfs_folio_trace_clear_s; + goto end_wb; + } + + if ((group = netfs_folio_group(folio))) { + if (group == NETFS_FOLIO_COPY_TO_CACHE) { + why = netfs_folio_trace_clear_cc; + folio_detach_private(folio); + goto end_wb; + } + + /* Need to detach the group pointer if the page didn't get + * redirtied. If it has been redirtied, then it must be within + * the same group. + */ + why = netfs_folio_trace_redirtied; + if (!folio_test_dirty(folio)) { + folio_detach_private(folio); + gcount++; + why = netfs_folio_trace_clear_g; + } + } + +end_wb: + trace_netfs_folio(folio, why); + folio_end_writeback(folio); + return gcount; +} + +/* + * Unlock any folios we've finished with. + */ +static void netfs_writeback_unlock_folios(struct netfs_io_request *wreq, + unsigned int *notes) +{ + struct folio_queue *folioq = wreq->buffer.tail; + unsigned long long collected_to = wreq->collected_to; + unsigned int slot = wreq->buffer.first_tail_slot; + + if (WARN_ON_ONCE(!folioq)) { + pr_err("[!] Writeback unlock found empty rolling buffer!\n"); + netfs_dump_request(wreq); + return; + } + + if (wreq->origin == NETFS_PGPRIV2_COPY_TO_CACHE) { + if (netfs_pgpriv2_unlock_copied_folios(wreq)) + *notes |= MADE_PROGRESS; + return; + } + + if (slot >= folioq_nr_slots(folioq)) { + folioq = rolling_buffer_delete_spent(&wreq->buffer); + if (!folioq) + return; + slot = 0; + } + + for (;;) { + struct folio *folio; + struct netfs_folio *finfo; + unsigned long long fpos, fend; + size_t fsize, flen; + + folio = folioq_folio(folioq, slot); + if (WARN_ONCE(!folio_test_writeback(folio), + "R=%08x: folio %lx is not under writeback\n", + wreq->debug_id, folio->index)) + trace_netfs_folio(folio, netfs_folio_trace_not_under_wback); + + fpos = folio_pos(folio); + fsize = folio_size(folio); + finfo = netfs_folio_info(folio); + flen = finfo ? finfo->dirty_offset + finfo->dirty_len : fsize; + + fend = min_t(unsigned long long, fpos + flen, wreq->i_size); + + trace_netfs_collect_folio(wreq, folio, fend, collected_to); + + /* Unlock any folio we've transferred all of. */ + if (collected_to < fend) + break; + + wreq->nr_group_rel += netfs_folio_written_back(folio); + wreq->cleaned_to = fpos + fsize; + *notes |= MADE_PROGRESS; + + /* Clean up the head folioq. If we clear an entire folioq, then + * we can get rid of it provided it's not also the tail folioq + * being filled by the issuer. + */ + folioq_clear(folioq, slot); + slot++; + if (slot >= folioq_nr_slots(folioq)) { + folioq = rolling_buffer_delete_spent(&wreq->buffer); + if (!folioq) + goto done; + slot = 0; + } + + if (fpos + fsize >= collected_to) + break; + } + + wreq->buffer.tail = folioq; +done: + wreq->buffer.first_tail_slot = slot; +} + +/* + * Collect and assess the results of various write subrequests. We may need to + * retry some of the results - or even do an RMW cycle for content crypto. + * + * Note that we have a number of parallel, overlapping lists of subrequests, + * one to the server and one to the local cache for example, which may not be + * the same size or starting position and may not even correspond in boundary + * alignment. + */ +static void netfs_collect_write_results(struct netfs_io_request *wreq) +{ + struct netfs_io_subrequest *front, *remove; + struct netfs_io_stream *stream; + unsigned long long collected_to, issued_to; + unsigned int notes; + int s; + + _enter("%llx-%llx", wreq->start, wreq->start + wreq->len); + trace_netfs_collect(wreq); + trace_netfs_rreq(wreq, netfs_rreq_trace_collect); + +reassess_streams: + issued_to = atomic64_read(&wreq->issued_to); + smp_rmb(); + collected_to = ULLONG_MAX; + if (wreq->origin == NETFS_WRITEBACK || + wreq->origin == NETFS_WRITETHROUGH || + wreq->origin == NETFS_PGPRIV2_COPY_TO_CACHE) + notes = NEED_UNLOCK; + else + notes = 0; + + /* Remove completed subrequests from the front of the streams and + * advance the completion point on each stream. We stop when we hit + * something that's in progress. The issuer thread may be adding stuff + * to the tail whilst we're doing this. + */ + for (s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + /* Read active flag before list pointers */ + if (!smp_load_acquire(&stream->active)) + continue; + + front = stream->front; + while (front) { + trace_netfs_collect_sreq(wreq, front); + //_debug("sreq [%x] %llx %zx/%zx", + // front->debug_index, front->start, front->transferred, front->len); + + if (stream->collected_to < front->start) { + trace_netfs_collect_gap(wreq, stream, issued_to, 'F'); + stream->collected_to = front->start; + } + + /* Stall if the front is still undergoing I/O. */ + if (test_bit(NETFS_SREQ_IN_PROGRESS, &front->flags)) { + notes |= HIT_PENDING; + break; + } + smp_rmb(); /* Read counters after I-P flag. */ + + if (stream->failed) { + stream->collected_to = front->start + front->len; + notes |= MADE_PROGRESS | SAW_FAILURE; + goto cancel; + } + if (front->start + front->transferred > stream->collected_to) { + stream->collected_to = front->start + front->transferred; + stream->transferred = stream->collected_to - wreq->start; + notes |= MADE_PROGRESS; + } + if (test_bit(NETFS_SREQ_FAILED, &front->flags)) { + stream->failed = true; + stream->error = front->error; + if (stream->source == NETFS_UPLOAD_TO_SERVER) + mapping_set_error(wreq->mapping, front->error); + notes |= NEED_REASSESS | SAW_FAILURE; + break; + } + if (front->transferred < front->len) { + stream->need_retry = true; + notes |= NEED_RETRY | MADE_PROGRESS; + break; + } + + cancel: + /* Remove if completely consumed. */ + spin_lock(&wreq->lock); + + remove = front; + list_del_init(&front->rreq_link); + front = list_first_entry_or_null(&stream->subrequests, + struct netfs_io_subrequest, rreq_link); + stream->front = front; + spin_unlock(&wreq->lock); + netfs_put_subrequest(remove, false, + notes & SAW_FAILURE ? + netfs_sreq_trace_put_cancel : + netfs_sreq_trace_put_done); + } + + /* If we have an empty stream, we need to jump it forward + * otherwise the collection point will never advance. + */ + if (!front && issued_to > stream->collected_to) { + trace_netfs_collect_gap(wreq, stream, issued_to, 'E'); + stream->collected_to = issued_to; + } + + if (stream->collected_to < collected_to) + collected_to = stream->collected_to; + } + + if (collected_to != ULLONG_MAX && collected_to > wreq->collected_to) + wreq->collected_to = collected_to; + + for (s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (stream->active) + trace_netfs_collect_stream(wreq, stream); + } + + trace_netfs_collect_state(wreq, wreq->collected_to, notes); + + /* Unlock any folios that we have now finished with. */ + if (notes & NEED_UNLOCK) { + if (wreq->cleaned_to < wreq->collected_to) + netfs_writeback_unlock_folios(wreq, ¬es); + } else { + wreq->cleaned_to = wreq->collected_to; + } + + // TODO: Discard encryption buffers + + if (notes & NEED_RETRY) + goto need_retry; + if ((notes & MADE_PROGRESS) && test_bit(NETFS_RREQ_PAUSE, &wreq->flags)) { + trace_netfs_rreq(wreq, netfs_rreq_trace_unpause); + clear_bit_unlock(NETFS_RREQ_PAUSE, &wreq->flags); + smp_mb__after_atomic(); /* Set PAUSE before task state */ + wake_up(&wreq->waitq); + } + + if (notes & NEED_REASSESS) { + //cond_resched(); + goto reassess_streams; + } + if (notes & MADE_PROGRESS) { + //cond_resched(); + goto reassess_streams; + } + +out: + netfs_put_group_many(wreq->group, wreq->nr_group_rel); + wreq->nr_group_rel = 0; + _leave(" = %x", notes); + return; + +need_retry: + /* Okay... We're going to have to retry one or both streams. Note + * that any partially completed op will have had any wholly transferred + * folios removed from it. + */ + _debug("retry"); + netfs_retry_writes(wreq); + goto out; +} + +/* + * Perform the collection of subrequests, folios and encryption buffers. + */ +void netfs_write_collection_worker(struct work_struct *work) +{ + struct netfs_io_request *wreq = container_of(work, struct netfs_io_request, work); + struct netfs_inode *ictx = netfs_inode(wreq->inode); + size_t transferred; + int s; + + _enter("R=%x", wreq->debug_id); + + netfs_see_request(wreq, netfs_rreq_trace_see_work); + if (!test_bit(NETFS_RREQ_IN_PROGRESS, &wreq->flags)) { + netfs_put_request(wreq, false, netfs_rreq_trace_put_work); + return; + } + + netfs_collect_write_results(wreq); + + /* We're done when the app thread has finished posting subreqs and all + * the queues in all the streams are empty. + */ + if (!test_bit(NETFS_RREQ_ALL_QUEUED, &wreq->flags)) { + netfs_put_request(wreq, false, netfs_rreq_trace_put_work); + return; + } + smp_rmb(); /* Read ALL_QUEUED before lists. */ + + transferred = LONG_MAX; + for (s = 0; s < NR_IO_STREAMS; s++) { + struct netfs_io_stream *stream = &wreq->io_streams[s]; + if (!stream->active) + continue; + if (!list_empty(&stream->subrequests)) { + netfs_put_request(wreq, false, netfs_rreq_trace_put_work); + return; + } + if (stream->transferred < transferred) + transferred = stream->transferred; + } + + /* Okay, declare that all I/O is complete. */ + wreq->transferred = transferred; + trace_netfs_rreq(wreq, netfs_rreq_trace_write_done); + + if (wreq->io_streams[1].active && + wreq->io_streams[1].failed && + ictx->ops->invalidate_cache) { + /* Cache write failure doesn't prevent writeback completion + * unless we're in disconnected mode. + */ + ictx->ops->invalidate_cache(wreq); + } + + if (wreq->cleanup) + wreq->cleanup(wreq); + + if (wreq->origin == NETFS_DIO_WRITE && + wreq->mapping->nrpages) { + /* mmap may have got underfoot and we may now have folios + * locally covering the region we just wrote. Attempt to + * discard the folios, but leave in place any modified locally. + * ->write_iter() is prevented from interfering by the DIO + * counter. + */ + pgoff_t first = wreq->start >> PAGE_SHIFT; + pgoff_t last = (wreq->start + wreq->transferred - 1) >> PAGE_SHIFT; + invalidate_inode_pages2_range(wreq->mapping, first, last); + } + + if (wreq->origin == NETFS_DIO_WRITE) + inode_dio_end(wreq->inode); + + _debug("finished"); + trace_netfs_rreq(wreq, netfs_rreq_trace_wake_ip); + clear_and_wake_up_bit(NETFS_RREQ_IN_PROGRESS, &wreq->flags); + + if (wreq->iocb) { + size_t written = min(wreq->transferred, wreq->len); + wreq->iocb->ki_pos += written; + if (wreq->iocb->ki_complete) + wreq->iocb->ki_complete( + wreq->iocb, wreq->error ? wreq->error : written); + wreq->iocb = VFS_PTR_POISON; + } + + netfs_clear_subrequests(wreq, false); + netfs_put_request(wreq, false, netfs_rreq_trace_put_work_complete); +} + +/* + * Wake the collection work item. + */ +void netfs_wake_write_collector(struct netfs_io_request *wreq, bool was_async) +{ + if (!work_pending(&wreq->work)) { + netfs_get_request(wreq, netfs_rreq_trace_get_work); + if (!queue_work(system_unbound_wq, &wreq->work)) + netfs_put_request(wreq, was_async, netfs_rreq_trace_put_work_nq); + } +} + +/** + * netfs_write_subrequest_terminated - Note the termination of a write operation. + * @_op: The I/O request that has terminated. + * @transferred_or_error: The amount of data transferred or an error code. + * @was_async: The termination was asynchronous + * + * This tells the library that a contributory write I/O operation has + * terminated, one way or another, and that it should collect the results. + * + * The caller indicates in @transferred_or_error the outcome of the operation, + * supplying a positive value to indicate the number of bytes transferred or a + * negative error code. The library will look after reissuing I/O operations + * as appropriate and writing downloaded data to the cache. + * + * If @was_async is true, the caller might be running in softirq or interrupt + * context and we can't sleep. + * + * When this is called, ownership of the subrequest is transferred back to the + * library, along with a ref. + * + * Note that %_op is a void* so that the function can be passed to + * kiocb::term_func without the need for a casting wrapper. + */ +void netfs_write_subrequest_terminated(void *_op, ssize_t transferred_or_error, + bool was_async) +{ + struct netfs_io_subrequest *subreq = _op; + struct netfs_io_request *wreq = subreq->rreq; + struct netfs_io_stream *stream = &wreq->io_streams[subreq->stream_nr]; + + _enter("%x[%x] %zd", wreq->debug_id, subreq->debug_index, transferred_or_error); + + switch (subreq->source) { + case NETFS_UPLOAD_TO_SERVER: + netfs_stat(&netfs_n_wh_upload_done); + break; + case NETFS_WRITE_TO_CACHE: + netfs_stat(&netfs_n_wh_write_done); + break; + case NETFS_INVALID_WRITE: + break; + default: + BUG(); + } + + if (IS_ERR_VALUE(transferred_or_error)) { + subreq->error = transferred_or_error; + if (subreq->error == -EAGAIN) + set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + else + set_bit(NETFS_SREQ_FAILED, &subreq->flags); + trace_netfs_failure(wreq, subreq, transferred_or_error, netfs_fail_write); + + switch (subreq->source) { + case NETFS_WRITE_TO_CACHE: + netfs_stat(&netfs_n_wh_write_failed); + break; + case NETFS_UPLOAD_TO_SERVER: + netfs_stat(&netfs_n_wh_upload_failed); + break; + default: + break; + } + trace_netfs_rreq(wreq, netfs_rreq_trace_set_pause); + set_bit(NETFS_RREQ_PAUSE, &wreq->flags); + } else { + if (WARN(transferred_or_error > subreq->len - subreq->transferred, + "Subreq excess write: R=%x[%x] %zd > %zu - %zu", + wreq->debug_id, subreq->debug_index, + transferred_or_error, subreq->len, subreq->transferred)) + transferred_or_error = subreq->len - subreq->transferred; + + subreq->error = 0; + subreq->transferred += transferred_or_error; + + if (subreq->transferred < subreq->len) + set_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + } + + trace_netfs_sreq(subreq, netfs_sreq_trace_terminated); + + clear_and_wake_up_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + + /* If we are at the head of the queue, wake up the collector, + * transferring a ref to it if we were the ones to do so. + */ + if (list_is_first(&subreq->rreq_link, &stream->subrequests)) + netfs_wake_write_collector(wreq, was_async); + + netfs_put_subrequest(subreq, was_async, netfs_sreq_trace_put_terminated); +} +EXPORT_SYMBOL(netfs_write_subrequest_terminated); diff --git a/fs/netfs/write_issue.c b/fs/netfs/write_issue.c new file mode 100644 index 000000000000..77279fc5b5a7 --- /dev/null +++ b/fs/netfs/write_issue.c @@ -0,0 +1,927 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Network filesystem high-level (buffered) writeback. + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + * + * + * To support network filesystems with local caching, we manage a situation + * that can be envisioned like the following: + * + * +---+---+-----+-----+---+----------+ + * Folios: | | | | | | | + * +---+---+-----+-----+---+----------+ + * + * +------+------+ +----+----+ + * Upload: | | |.....| | | + * (Stream 0) +------+------+ +----+----+ + * + * +------+------+------+------+------+ + * Cache: | | | | | | + * (Stream 1) +------+------+------+------+------+ + * + * Where we have a sequence of folios of varying sizes that we need to overlay + * with multiple parallel streams of I/O requests, where the I/O requests in a + * stream may also be of various sizes (in cifs, for example, the sizes are + * negotiated with the server; in something like ceph, they may represent the + * sizes of storage objects). + * + * The sequence in each stream may contain gaps and noncontiguous subrequests + * may be glued together into single vectored write RPCs. + */ + +#include <linux/export.h> +#include <linux/fs.h> +#include <linux/mm.h> +#include <linux/pagemap.h> +#include "internal.h" + +/* + * Kill all dirty folios in the event of an unrecoverable error, starting with + * a locked folio we've already obtained from writeback_iter(). + */ +static void netfs_kill_dirty_pages(struct address_space *mapping, + struct writeback_control *wbc, + struct folio *folio) +{ + int error = 0; + + do { + enum netfs_folio_trace why = netfs_folio_trace_kill; + struct netfs_group *group = NULL; + struct netfs_folio *finfo = NULL; + void *priv; + + priv = folio_detach_private(folio); + if (priv) { + finfo = __netfs_folio_info(priv); + if (finfo) { + /* Kill folio from streaming write. */ + group = finfo->netfs_group; + why = netfs_folio_trace_kill_s; + } else { + group = priv; + if (group == NETFS_FOLIO_COPY_TO_CACHE) { + /* Kill copy-to-cache folio */ + why = netfs_folio_trace_kill_cc; + group = NULL; + } else { + /* Kill folio with group */ + why = netfs_folio_trace_kill_g; + } + } + } + + trace_netfs_folio(folio, why); + + folio_start_writeback(folio); + folio_unlock(folio); + folio_end_writeback(folio); + + netfs_put_group(group); + kfree(finfo); + + } while ((folio = writeback_iter(mapping, wbc, folio, &error))); +} + +/* + * Create a write request and set it up appropriately for the origin type. + */ +struct netfs_io_request *netfs_create_write_req(struct address_space *mapping, + struct file *file, + loff_t start, + enum netfs_io_origin origin) +{ + struct netfs_io_request *wreq; + struct netfs_inode *ictx; + bool is_cacheable = (origin == NETFS_WRITEBACK || + origin == NETFS_WRITEBACK_SINGLE || + origin == NETFS_WRITETHROUGH || + origin == NETFS_PGPRIV2_COPY_TO_CACHE); + + wreq = netfs_alloc_request(mapping, file, start, 0, origin); + if (IS_ERR(wreq)) + return wreq; + + _enter("R=%x", wreq->debug_id); + + ictx = netfs_inode(wreq->inode); + if (is_cacheable && netfs_is_cache_enabled(ictx)) + fscache_begin_write_operation(&wreq->cache_resources, netfs_i_cookie(ictx)); + if (rolling_buffer_init(&wreq->buffer, wreq->debug_id, ITER_SOURCE) < 0) + goto nomem; + + wreq->cleaned_to = wreq->start; + + wreq->io_streams[0].stream_nr = 0; + wreq->io_streams[0].source = NETFS_UPLOAD_TO_SERVER; + wreq->io_streams[0].prepare_write = ictx->ops->prepare_write; + wreq->io_streams[0].issue_write = ictx->ops->issue_write; + wreq->io_streams[0].collected_to = start; + wreq->io_streams[0].transferred = LONG_MAX; + + wreq->io_streams[1].stream_nr = 1; + wreq->io_streams[1].source = NETFS_WRITE_TO_CACHE; + wreq->io_streams[1].collected_to = start; + wreq->io_streams[1].transferred = LONG_MAX; + if (fscache_resources_valid(&wreq->cache_resources)) { + wreq->io_streams[1].avail = true; + wreq->io_streams[1].active = true; + wreq->io_streams[1].prepare_write = wreq->cache_resources.ops->prepare_write_subreq; + wreq->io_streams[1].issue_write = wreq->cache_resources.ops->issue_write; + } + + return wreq; +nomem: + wreq->error = -ENOMEM; + netfs_put_request(wreq, false, netfs_rreq_trace_put_failed); + return ERR_PTR(-ENOMEM); +} + +/** + * netfs_prepare_write_failed - Note write preparation failed + * @subreq: The subrequest to mark + * + * Mark a subrequest to note that preparation for write failed. + */ +void netfs_prepare_write_failed(struct netfs_io_subrequest *subreq) +{ + __set_bit(NETFS_SREQ_FAILED, &subreq->flags); + trace_netfs_sreq(subreq, netfs_sreq_trace_prep_failed); +} +EXPORT_SYMBOL(netfs_prepare_write_failed); + +/* + * Prepare a write subrequest. We need to allocate a new subrequest + * if we don't have one. + */ +static void netfs_prepare_write(struct netfs_io_request *wreq, + struct netfs_io_stream *stream, + loff_t start) +{ + struct netfs_io_subrequest *subreq; + struct iov_iter *wreq_iter = &wreq->buffer.iter; + + /* Make sure we don't point the iterator at a used-up folio_queue + * struct being used as a placeholder to prevent the queue from + * collapsing. In such a case, extend the queue. + */ + if (iov_iter_is_folioq(wreq_iter) && + wreq_iter->folioq_slot >= folioq_nr_slots(wreq_iter->folioq)) + rolling_buffer_make_space(&wreq->buffer); + + subreq = netfs_alloc_subrequest(wreq); + subreq->source = stream->source; + subreq->start = start; + subreq->stream_nr = stream->stream_nr; + subreq->io_iter = *wreq_iter; + + _enter("R=%x[%x]", wreq->debug_id, subreq->debug_index); + + trace_netfs_sreq(subreq, netfs_sreq_trace_prepare); + + stream->sreq_max_len = UINT_MAX; + stream->sreq_max_segs = INT_MAX; + switch (stream->source) { + case NETFS_UPLOAD_TO_SERVER: + netfs_stat(&netfs_n_wh_upload); + stream->sreq_max_len = wreq->wsize; + break; + case NETFS_WRITE_TO_CACHE: + netfs_stat(&netfs_n_wh_write); + break; + default: + WARN_ON_ONCE(1); + break; + } + + if (stream->prepare_write) + stream->prepare_write(subreq); + + __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + + /* We add to the end of the list whilst the collector may be walking + * the list. The collector only goes nextwards and uses the lock to + * remove entries off of the front. + */ + spin_lock(&wreq->lock); + list_add_tail(&subreq->rreq_link, &stream->subrequests); + if (list_is_first(&subreq->rreq_link, &stream->subrequests)) { + stream->front = subreq; + if (!stream->active) { + stream->collected_to = stream->front->start; + /* Write list pointers before active flag */ + smp_store_release(&stream->active, true); + } + } + + spin_unlock(&wreq->lock); + + stream->construct = subreq; +} + +/* + * Set the I/O iterator for the filesystem/cache to use and dispatch the I/O + * operation. The operation may be asynchronous and should call + * netfs_write_subrequest_terminated() when complete. + */ +static void netfs_do_issue_write(struct netfs_io_stream *stream, + struct netfs_io_subrequest *subreq) +{ + struct netfs_io_request *wreq = subreq->rreq; + + _enter("R=%x[%x],%zx", wreq->debug_id, subreq->debug_index, subreq->len); + + if (test_bit(NETFS_SREQ_FAILED, &subreq->flags)) + return netfs_write_subrequest_terminated(subreq, subreq->error, false); + + trace_netfs_sreq(subreq, netfs_sreq_trace_submit); + stream->issue_write(subreq); +} + +void netfs_reissue_write(struct netfs_io_stream *stream, + struct netfs_io_subrequest *subreq, + struct iov_iter *source) +{ + size_t size = subreq->len - subreq->transferred; + + // TODO: Use encrypted buffer + subreq->io_iter = *source; + iov_iter_advance(source, size); + iov_iter_truncate(&subreq->io_iter, size); + + subreq->retry_count++; + __clear_bit(NETFS_SREQ_MADE_PROGRESS, &subreq->flags); + __set_bit(NETFS_SREQ_IN_PROGRESS, &subreq->flags); + netfs_stat(&netfs_n_wh_retry_write_subreq); + netfs_do_issue_write(stream, subreq); +} + +void netfs_issue_write(struct netfs_io_request *wreq, + struct netfs_io_stream *stream) +{ + struct netfs_io_subrequest *subreq = stream->construct; + + if (!subreq) + return; + stream->construct = NULL; + subreq->io_iter.count = subreq->len; + netfs_do_issue_write(stream, subreq); +} + +/* + * Add data to the write subrequest, dispatching each as we fill it up or if it + * is discontiguous with the previous. We only fill one part at a time so that + * we can avoid overrunning the credits obtained (cifs) and try to parallelise + * content-crypto preparation with network writes. + */ +size_t netfs_advance_write(struct netfs_io_request *wreq, + struct netfs_io_stream *stream, + loff_t start, size_t len, bool to_eof) +{ + struct netfs_io_subrequest *subreq = stream->construct; + size_t part; + + if (!stream->avail) { + _leave("no write"); + return len; + } + + _enter("R=%x[%x]", wreq->debug_id, subreq ? subreq->debug_index : 0); + + if (subreq && start != subreq->start + subreq->len) { + netfs_issue_write(wreq, stream); + subreq = NULL; + } + + if (!stream->construct) + netfs_prepare_write(wreq, stream, start); + subreq = stream->construct; + + part = umin(stream->sreq_max_len - subreq->len, len); + _debug("part %zx/%zx %zx/%zx", subreq->len, stream->sreq_max_len, part, len); + subreq->len += part; + subreq->nr_segs++; + stream->submit_extendable_to -= part; + + if (subreq->len >= stream->sreq_max_len || + subreq->nr_segs >= stream->sreq_max_segs || + to_eof) { + netfs_issue_write(wreq, stream); + subreq = NULL; + } + + return part; +} + +/* + * Write some of a pending folio data back to the server. + */ +static int netfs_write_folio(struct netfs_io_request *wreq, + struct writeback_control *wbc, + struct folio *folio) +{ + struct netfs_io_stream *upload = &wreq->io_streams[0]; + struct netfs_io_stream *cache = &wreq->io_streams[1]; + struct netfs_io_stream *stream; + struct netfs_group *fgroup; /* TODO: Use this with ceph */ + struct netfs_folio *finfo; + size_t iter_off = 0; + size_t fsize = folio_size(folio), flen = fsize, foff = 0; + loff_t fpos = folio_pos(folio), i_size; + bool to_eof = false, streamw = false; + bool debug = false; + + _enter(""); + + if (rolling_buffer_make_space(&wreq->buffer) < 0) + return -ENOMEM; + + /* netfs_perform_write() may shift i_size around the page or from out + * of the page to beyond it, but cannot move i_size into or through the + * page since we have it locked. + */ + i_size = i_size_read(wreq->inode); + + if (fpos >= i_size) { + /* mmap beyond eof. */ + _debug("beyond eof"); + folio_start_writeback(folio); + folio_unlock(folio); + wreq->nr_group_rel += netfs_folio_written_back(folio); + netfs_put_group_many(wreq->group, wreq->nr_group_rel); + wreq->nr_group_rel = 0; + return 0; + } + + if (fpos + fsize > wreq->i_size) + wreq->i_size = i_size; + + fgroup = netfs_folio_group(folio); + finfo = netfs_folio_info(folio); + if (finfo) { + foff = finfo->dirty_offset; + flen = foff + finfo->dirty_len; + streamw = true; + } + + if (wreq->origin == NETFS_WRITETHROUGH) { + to_eof = false; + if (flen > i_size - fpos) + flen = i_size - fpos; + } else if (flen > i_size - fpos) { + flen = i_size - fpos; + if (!streamw) + folio_zero_segment(folio, flen, fsize); + to_eof = true; + } else if (flen == i_size - fpos) { + to_eof = true; + } + flen -= foff; + + _debug("folio %zx %zx %zx", foff, flen, fsize); + + /* Deal with discontinuities in the stream of dirty pages. These can + * arise from a number of sources: + * + * (1) Intervening non-dirty pages from random-access writes, multiple + * flushers writing back different parts simultaneously and manual + * syncing. + * + * (2) Partially-written pages from write-streaming. + * + * (3) Pages that belong to a different write-back group (eg. Ceph + * snapshots). + * + * (4) Actually-clean pages that were marked for write to the cache + * when they were read. Note that these appear as a special + * write-back group. + */ + if (fgroup == NETFS_FOLIO_COPY_TO_CACHE) { + netfs_issue_write(wreq, upload); + } else if (fgroup != wreq->group) { + /* We can't write this page to the server yet. */ + kdebug("wrong group"); + folio_redirty_for_writepage(wbc, folio); + folio_unlock(folio); + netfs_issue_write(wreq, upload); + netfs_issue_write(wreq, cache); + return 0; + } + + if (foff > 0) + netfs_issue_write(wreq, upload); + if (streamw) + netfs_issue_write(wreq, cache); + + /* Flip the page to the writeback state and unlock. If we're called + * from write-through, then the page has already been put into the wb + * state. + */ + if (wreq->origin == NETFS_WRITEBACK) + folio_start_writeback(folio); + folio_unlock(folio); + + if (fgroup == NETFS_FOLIO_COPY_TO_CACHE) { + if (!cache->avail) { + trace_netfs_folio(folio, netfs_folio_trace_cancel_copy); + netfs_issue_write(wreq, upload); + netfs_folio_written_back(folio); + return 0; + } + trace_netfs_folio(folio, netfs_folio_trace_store_copy); + } else if (!upload->avail && !cache->avail) { + trace_netfs_folio(folio, netfs_folio_trace_cancel_store); + netfs_folio_written_back(folio); + return 0; + } else if (!upload->construct) { + trace_netfs_folio(folio, netfs_folio_trace_store); + } else { + trace_netfs_folio(folio, netfs_folio_trace_store_plus); + } + + /* Attach the folio to the rolling buffer. */ + rolling_buffer_append(&wreq->buffer, folio, 0); + + /* Move the submission point forward to allow for write-streaming data + * not starting at the front of the page. We don't do write-streaming + * with the cache as the cache requires DIO alignment. + * + * Also skip uploading for data that's been read and just needs copying + * to the cache. + */ + for (int s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + stream->submit_off = foff; + stream->submit_len = flen; + if (!stream->avail || + (stream->source == NETFS_WRITE_TO_CACHE && streamw) || + (stream->source == NETFS_UPLOAD_TO_SERVER && + fgroup == NETFS_FOLIO_COPY_TO_CACHE)) { + stream->submit_off = UINT_MAX; + stream->submit_len = 0; + } + } + + /* Attach the folio to one or more subrequests. For a big folio, we + * could end up with thousands of subrequests if the wsize is small - + * but we might need to wait during the creation of subrequests for + * network resources (eg. SMB credits). + */ + for (;;) { + ssize_t part; + size_t lowest_off = ULONG_MAX; + int choose_s = -1; + + /* Always add to the lowest-submitted stream first. */ + for (int s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (stream->submit_len > 0 && + stream->submit_off < lowest_off) { + lowest_off = stream->submit_off; + choose_s = s; + } + } + + if (choose_s < 0) + break; + stream = &wreq->io_streams[choose_s]; + + /* Advance the iterator(s). */ + if (stream->submit_off > iter_off) { + rolling_buffer_advance(&wreq->buffer, stream->submit_off - iter_off); + iter_off = stream->submit_off; + } + + atomic64_set(&wreq->issued_to, fpos + stream->submit_off); + stream->submit_extendable_to = fsize - stream->submit_off; + part = netfs_advance_write(wreq, stream, fpos + stream->submit_off, + stream->submit_len, to_eof); + stream->submit_off += part; + if (part > stream->submit_len) + stream->submit_len = 0; + else + stream->submit_len -= part; + if (part > 0) + debug = true; + } + + if (fsize > iter_off) + rolling_buffer_advance(&wreq->buffer, fsize - iter_off); + atomic64_set(&wreq->issued_to, fpos + fsize); + + if (!debug) + kdebug("R=%x: No submit", wreq->debug_id); + + if (foff + flen < fsize) + for (int s = 0; s < NR_IO_STREAMS; s++) + netfs_issue_write(wreq, &wreq->io_streams[s]); + + _leave(" = 0"); + return 0; +} + +/* + * End the issuing of writes, letting the collector know we're done. + */ +static void netfs_end_issue_write(struct netfs_io_request *wreq) +{ + bool needs_poke = true; + + smp_wmb(); /* Write subreq lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &wreq->flags); + + for (int s = 0; s < NR_IO_STREAMS; s++) { + struct netfs_io_stream *stream = &wreq->io_streams[s]; + + if (!stream->active) + continue; + if (!list_empty(&stream->subrequests)) + needs_poke = false; + netfs_issue_write(wreq, stream); + } + + if (needs_poke) + netfs_wake_write_collector(wreq, false); +} + +/* + * Write some of the pending data back to the server + */ +int netfs_writepages(struct address_space *mapping, + struct writeback_control *wbc) +{ + struct netfs_inode *ictx = netfs_inode(mapping->host); + struct netfs_io_request *wreq = NULL; + struct folio *folio; + int error = 0; + + if (!mutex_trylock(&ictx->wb_lock)) { + if (wbc->sync_mode == WB_SYNC_NONE) { + netfs_stat(&netfs_n_wb_lock_skip); + return 0; + } + netfs_stat(&netfs_n_wb_lock_wait); + mutex_lock(&ictx->wb_lock); + } + + /* Need the first folio to be able to set up the op. */ + folio = writeback_iter(mapping, wbc, NULL, &error); + if (!folio) + goto out; + + wreq = netfs_create_write_req(mapping, NULL, folio_pos(folio), NETFS_WRITEBACK); + if (IS_ERR(wreq)) { + error = PTR_ERR(wreq); + goto couldnt_start; + } + + trace_netfs_write(wreq, netfs_write_trace_writeback); + netfs_stat(&netfs_n_wh_writepages); + + do { + _debug("wbiter %lx %llx", folio->index, atomic64_read(&wreq->issued_to)); + + /* It appears we don't have to handle cyclic writeback wrapping. */ + WARN_ON_ONCE(wreq && folio_pos(folio) < atomic64_read(&wreq->issued_to)); + + if (netfs_folio_group(folio) != NETFS_FOLIO_COPY_TO_CACHE && + unlikely(!test_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags))) { + set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags); + wreq->netfs_ops->begin_writeback(wreq); + } + + error = netfs_write_folio(wreq, wbc, folio); + if (error < 0) + break; + } while ((folio = writeback_iter(mapping, wbc, folio, &error))); + + netfs_end_issue_write(wreq); + + mutex_unlock(&ictx->wb_lock); + + netfs_put_request(wreq, false, netfs_rreq_trace_put_return); + _leave(" = %d", error); + return error; + +couldnt_start: + netfs_kill_dirty_pages(mapping, wbc, folio); +out: + mutex_unlock(&ictx->wb_lock); + _leave(" = %d", error); + return error; +} +EXPORT_SYMBOL(netfs_writepages); + +/* + * Begin a write operation for writing through the pagecache. + */ +struct netfs_io_request *netfs_begin_writethrough(struct kiocb *iocb, size_t len) +{ + struct netfs_io_request *wreq = NULL; + struct netfs_inode *ictx = netfs_inode(file_inode(iocb->ki_filp)); + + mutex_lock(&ictx->wb_lock); + + wreq = netfs_create_write_req(iocb->ki_filp->f_mapping, iocb->ki_filp, + iocb->ki_pos, NETFS_WRITETHROUGH); + if (IS_ERR(wreq)) { + mutex_unlock(&ictx->wb_lock); + return wreq; + } + + wreq->io_streams[0].avail = true; + trace_netfs_write(wreq, netfs_write_trace_writethrough); + return wreq; +} + +/* + * Advance the state of the write operation used when writing through the + * pagecache. Data has been copied into the pagecache that we need to append + * to the request. If we've added more than wsize then we need to create a new + * subrequest. + */ +int netfs_advance_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, + struct folio *folio, size_t copied, bool to_page_end, + struct folio **writethrough_cache) +{ + _enter("R=%x ic=%zu ws=%u cp=%zu tp=%u", + wreq->debug_id, wreq->buffer.iter.count, wreq->wsize, copied, to_page_end); + + if (!*writethrough_cache) { + if (folio_test_dirty(folio)) + /* Sigh. mmap. */ + folio_clear_dirty_for_io(folio); + + /* We can make multiple writes to the folio... */ + folio_start_writeback(folio); + if (wreq->len == 0) + trace_netfs_folio(folio, netfs_folio_trace_wthru); + else + trace_netfs_folio(folio, netfs_folio_trace_wthru_plus); + *writethrough_cache = folio; + } + + wreq->len += copied; + if (!to_page_end) + return 0; + + *writethrough_cache = NULL; + return netfs_write_folio(wreq, wbc, folio); +} + +/* + * End a write operation used when writing through the pagecache. + */ +int netfs_end_writethrough(struct netfs_io_request *wreq, struct writeback_control *wbc, + struct folio *writethrough_cache) +{ + struct netfs_inode *ictx = netfs_inode(wreq->inode); + int ret; + + _enter("R=%x", wreq->debug_id); + + if (writethrough_cache) + netfs_write_folio(wreq, wbc, writethrough_cache); + + netfs_end_issue_write(wreq); + + mutex_unlock(&ictx->wb_lock); + + if (wreq->iocb) { + ret = -EIOCBQUEUED; + } else { + wait_on_bit(&wreq->flags, NETFS_RREQ_IN_PROGRESS, TASK_UNINTERRUPTIBLE); + ret = wreq->error; + } + netfs_put_request(wreq, false, netfs_rreq_trace_put_return); + return ret; +} + +/* + * Write data to the server without going through the pagecache and without + * writing it to the local cache. + */ +int netfs_unbuffered_write(struct netfs_io_request *wreq, bool may_wait, size_t len) +{ + struct netfs_io_stream *upload = &wreq->io_streams[0]; + ssize_t part; + loff_t start = wreq->start; + int error = 0; + + _enter("%zx", len); + + if (wreq->origin == NETFS_DIO_WRITE) + inode_dio_begin(wreq->inode); + + while (len) { + // TODO: Prepare content encryption + + _debug("unbuffered %zx", len); + part = netfs_advance_write(wreq, upload, start, len, false); + start += part; + len -= part; + rolling_buffer_advance(&wreq->buffer, part); + if (test_bit(NETFS_RREQ_PAUSE, &wreq->flags)) { + trace_netfs_rreq(wreq, netfs_rreq_trace_wait_pause); + wait_event(wreq->waitq, !test_bit(NETFS_RREQ_PAUSE, &wreq->flags)); + } + if (test_bit(NETFS_RREQ_FAILED, &wreq->flags)) + break; + } + + netfs_end_issue_write(wreq); + _leave(" = %d", error); + return error; +} + +/* + * Write some of a pending folio data back to the server and/or the cache. + */ +static int netfs_write_folio_single(struct netfs_io_request *wreq, + struct folio *folio) +{ + struct netfs_io_stream *upload = &wreq->io_streams[0]; + struct netfs_io_stream *cache = &wreq->io_streams[1]; + struct netfs_io_stream *stream; + size_t iter_off = 0; + size_t fsize = folio_size(folio), flen; + loff_t fpos = folio_pos(folio); + bool to_eof = false; + bool no_debug = false; + + _enter(""); + + flen = folio_size(folio); + if (flen > wreq->i_size - fpos) { + flen = wreq->i_size - fpos; + folio_zero_segment(folio, flen, fsize); + to_eof = true; + } else if (flen == wreq->i_size - fpos) { + to_eof = true; + } + + _debug("folio %zx/%zx", flen, fsize); + + if (!upload->avail && !cache->avail) { + trace_netfs_folio(folio, netfs_folio_trace_cancel_store); + return 0; + } + + if (!upload->construct) + trace_netfs_folio(folio, netfs_folio_trace_store); + else + trace_netfs_folio(folio, netfs_folio_trace_store_plus); + + /* Attach the folio to the rolling buffer. */ + folio_get(folio); + rolling_buffer_append(&wreq->buffer, folio, NETFS_ROLLBUF_PUT_MARK); + + /* Move the submission point forward to allow for write-streaming data + * not starting at the front of the page. We don't do write-streaming + * with the cache as the cache requires DIO alignment. + * + * Also skip uploading for data that's been read and just needs copying + * to the cache. + */ + for (int s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + stream->submit_off = 0; + stream->submit_len = flen; + if (!stream->avail) { + stream->submit_off = UINT_MAX; + stream->submit_len = 0; + } + } + + /* Attach the folio to one or more subrequests. For a big folio, we + * could end up with thousands of subrequests if the wsize is small - + * but we might need to wait during the creation of subrequests for + * network resources (eg. SMB credits). + */ + for (;;) { + ssize_t part; + size_t lowest_off = ULONG_MAX; + int choose_s = -1; + + /* Always add to the lowest-submitted stream first. */ + for (int s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (stream->submit_len > 0 && + stream->submit_off < lowest_off) { + lowest_off = stream->submit_off; + choose_s = s; + } + } + + if (choose_s < 0) + break; + stream = &wreq->io_streams[choose_s]; + + /* Advance the iterator(s). */ + if (stream->submit_off > iter_off) { + rolling_buffer_advance(&wreq->buffer, stream->submit_off - iter_off); + iter_off = stream->submit_off; + } + + atomic64_set(&wreq->issued_to, fpos + stream->submit_off); + stream->submit_extendable_to = fsize - stream->submit_off; + part = netfs_advance_write(wreq, stream, fpos + stream->submit_off, + stream->submit_len, to_eof); + stream->submit_off += part; + if (part > stream->submit_len) + stream->submit_len = 0; + else + stream->submit_len -= part; + if (part > 0) + no_debug = true; + } + + wreq->buffer.iter.iov_offset = 0; + if (fsize > iter_off) + rolling_buffer_advance(&wreq->buffer, fsize - iter_off); + atomic64_set(&wreq->issued_to, fpos + fsize); + + if (!no_debug) + kdebug("R=%x: No submit", wreq->debug_id); + _leave(" = 0"); + return 0; +} + +/** + * netfs_writeback_single - Write back a monolithic payload + * @mapping: The mapping to write from + * @wbc: Hints from the VM + * @iter: Data to write, must be ITER_FOLIOQ. + * + * Write a monolithic, non-pagecache object back to the server and/or + * the cache. + */ +int netfs_writeback_single(struct address_space *mapping, + struct writeback_control *wbc, + struct iov_iter *iter) +{ + struct netfs_io_request *wreq; + struct netfs_inode *ictx = netfs_inode(mapping->host); + struct folio_queue *fq; + size_t size = iov_iter_count(iter); + int ret; + + if (WARN_ON_ONCE(!iov_iter_is_folioq(iter))) + return -EIO; + + if (!mutex_trylock(&ictx->wb_lock)) { + if (wbc->sync_mode == WB_SYNC_NONE) { + netfs_stat(&netfs_n_wb_lock_skip); + return 0; + } + netfs_stat(&netfs_n_wb_lock_wait); + mutex_lock(&ictx->wb_lock); + } + + wreq = netfs_create_write_req(mapping, NULL, 0, NETFS_WRITEBACK_SINGLE); + if (IS_ERR(wreq)) { + ret = PTR_ERR(wreq); + goto couldnt_start; + } + + trace_netfs_write(wreq, netfs_write_trace_writeback); + netfs_stat(&netfs_n_wh_writepages); + + if (__test_and_set_bit(NETFS_RREQ_UPLOAD_TO_SERVER, &wreq->flags)) + wreq->netfs_ops->begin_writeback(wreq); + + for (fq = (struct folio_queue *)iter->folioq; fq; fq = fq->next) { + for (int slot = 0; slot < folioq_count(fq); slot++) { + struct folio *folio = folioq_folio(fq, slot); + size_t part = umin(folioq_folio_size(fq, slot), size); + + _debug("wbiter %lx %llx", folio->index, atomic64_read(&wreq->issued_to)); + + ret = netfs_write_folio_single(wreq, folio); + if (ret < 0) + goto stop; + size -= part; + if (size <= 0) + goto stop; + } + } + +stop: + for (int s = 0; s < NR_IO_STREAMS; s++) + netfs_issue_write(wreq, &wreq->io_streams[s]); + smp_wmb(); /* Write lists before ALL_QUEUED. */ + set_bit(NETFS_RREQ_ALL_QUEUED, &wreq->flags); + + mutex_unlock(&ictx->wb_lock); + + netfs_put_request(wreq, false, netfs_rreq_trace_put_return); + _leave(" = %d", ret); + return ret; + +couldnt_start: + mutex_unlock(&ictx->wb_lock); + _leave(" = %d", ret); + return ret; +} +EXPORT_SYMBOL(netfs_writeback_single); diff --git a/fs/netfs/write_retry.c b/fs/netfs/write_retry.c new file mode 100644 index 000000000000..545d33079a77 --- /dev/null +++ b/fs/netfs/write_retry.c @@ -0,0 +1,234 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Network filesystem write retrying. + * + * Copyright (C) 2024 Red Hat, Inc. All Rights Reserved. + * Written by David Howells (dhowells@redhat.com) + */ + +#include <linux/fs.h> +#include <linux/mm.h> +#include <linux/pagemap.h> +#include <linux/slab.h> +#include "internal.h" + +/* + * Perform retries on the streams that need it. + */ +static void netfs_retry_write_stream(struct netfs_io_request *wreq, + struct netfs_io_stream *stream) +{ + struct list_head *next; + + _enter("R=%x[%x:]", wreq->debug_id, stream->stream_nr); + + if (list_empty(&stream->subrequests)) + return; + + if (stream->source == NETFS_UPLOAD_TO_SERVER && + wreq->netfs_ops->retry_request) + wreq->netfs_ops->retry_request(wreq, stream); + + if (unlikely(stream->failed)) + return; + + /* If there's no renegotiation to do, just resend each failed subreq. */ + if (!stream->prepare_write) { + struct netfs_io_subrequest *subreq; + + list_for_each_entry(subreq, &stream->subrequests, rreq_link) { + if (test_bit(NETFS_SREQ_FAILED, &subreq->flags)) + break; + if (__test_and_clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) { + struct iov_iter source = subreq->io_iter; + + iov_iter_revert(&source, subreq->len - source.count); + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); + netfs_reissue_write(stream, subreq, &source); + } + } + return; + } + + next = stream->subrequests.next; + + do { + struct netfs_io_subrequest *subreq = NULL, *from, *to, *tmp; + struct iov_iter source; + unsigned long long start, len; + size_t part; + bool boundary = false; + + /* Go through the stream and find the next span of contiguous + * data that we then rejig (cifs, for example, needs the wsize + * renegotiating) and reissue. + */ + from = list_entry(next, struct netfs_io_subrequest, rreq_link); + to = from; + start = from->start + from->transferred; + len = from->len - from->transferred; + + if (test_bit(NETFS_SREQ_FAILED, &from->flags) || + !test_bit(NETFS_SREQ_NEED_RETRY, &from->flags)) + return; + + list_for_each_continue(next, &stream->subrequests) { + subreq = list_entry(next, struct netfs_io_subrequest, rreq_link); + if (subreq->start + subreq->transferred != start + len || + test_bit(NETFS_SREQ_BOUNDARY, &subreq->flags) || + !test_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags)) + break; + to = subreq; + len += to->len; + } + + /* Determine the set of buffers we're going to use. Each + * subreq gets a subset of a single overall contiguous buffer. + */ + netfs_reset_iter(from); + source = from->io_iter; + source.count = len; + + /* Work through the sublist. */ + subreq = from; + list_for_each_entry_from(subreq, &stream->subrequests, rreq_link) { + if (!len) + break; + + subreq->start = start; + subreq->len = len; + __clear_bit(NETFS_SREQ_NEED_RETRY, &subreq->flags); + subreq->retry_count++; + trace_netfs_sreq(subreq, netfs_sreq_trace_retry); + + /* Renegotiate max_len (wsize) */ + stream->sreq_max_len = len; + stream->prepare_write(subreq); + + part = umin(len, stream->sreq_max_len); + if (unlikely(stream->sreq_max_segs)) + part = netfs_limit_iter(&source, 0, part, stream->sreq_max_segs); + subreq->len = part; + subreq->transferred = 0; + len -= part; + start += part; + if (len && subreq == to && + __test_and_clear_bit(NETFS_SREQ_BOUNDARY, &to->flags)) + boundary = true; + + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); + netfs_reissue_write(stream, subreq, &source); + if (subreq == to) + break; + } + + /* If we managed to use fewer subreqs, we can discard the + * excess; if we used the same number, then we're done. + */ + if (!len) { + if (subreq == to) + continue; + list_for_each_entry_safe_from(subreq, tmp, + &stream->subrequests, rreq_link) { + trace_netfs_sreq(subreq, netfs_sreq_trace_discard); + list_del(&subreq->rreq_link); + netfs_put_subrequest(subreq, false, netfs_sreq_trace_put_done); + if (subreq == to) + break; + } + continue; + } + + /* We ran out of subrequests, so we need to allocate some more + * and insert them after. + */ + do { + subreq = netfs_alloc_subrequest(wreq); + subreq->source = to->source; + subreq->start = start; + subreq->debug_index = atomic_inc_return(&wreq->subreq_counter); + subreq->stream_nr = to->stream_nr; + subreq->retry_count = 1; + + trace_netfs_sreq_ref(wreq->debug_id, subreq->debug_index, + refcount_read(&subreq->ref), + netfs_sreq_trace_new); + netfs_get_subrequest(subreq, netfs_sreq_trace_get_resubmit); + + list_add(&subreq->rreq_link, &to->rreq_link); + to = list_next_entry(to, rreq_link); + trace_netfs_sreq(subreq, netfs_sreq_trace_retry); + + stream->sreq_max_len = len; + stream->sreq_max_segs = INT_MAX; + switch (stream->source) { + case NETFS_UPLOAD_TO_SERVER: + netfs_stat(&netfs_n_wh_upload); + stream->sreq_max_len = umin(len, wreq->wsize); + break; + case NETFS_WRITE_TO_CACHE: + netfs_stat(&netfs_n_wh_write); + break; + default: + WARN_ON_ONCE(1); + } + + stream->prepare_write(subreq); + + part = umin(len, stream->sreq_max_len); + subreq->len = subreq->transferred + part; + len -= part; + start += part; + if (!len && boundary) { + __set_bit(NETFS_SREQ_BOUNDARY, &to->flags); + boundary = false; + } + + netfs_reissue_write(stream, subreq, &source); + if (!len) + break; + + } while (len); + + } while (!list_is_head(next, &stream->subrequests)); +} + +/* + * Perform retries on the streams that need it. If we're doing content + * encryption and the server copy changed due to a third-party write, we may + * need to do an RMW cycle and also rewrite the data to the cache. + */ +void netfs_retry_writes(struct netfs_io_request *wreq) +{ + struct netfs_io_subrequest *subreq; + struct netfs_io_stream *stream; + int s; + + netfs_stat(&netfs_n_wh_retry_write_req); + + /* Wait for all outstanding I/O to quiesce before performing retries as + * we may need to renegotiate the I/O sizes. + */ + for (s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (!stream->active) + continue; + + list_for_each_entry(subreq, &stream->subrequests, rreq_link) { + wait_on_bit(&subreq->flags, NETFS_SREQ_IN_PROGRESS, + TASK_UNINTERRUPTIBLE); + } + } + + // TODO: Enc: Fetch changed partial pages + // TODO: Enc: Reencrypt content if needed. + // TODO: Enc: Wind back transferred point. + // TODO: Enc: Mark cache pages for retry. + + for (s = 0; s < NR_IO_STREAMS; s++) { + stream = &wreq->io_streams[s]; + if (stream->need_retry) { + stream->need_retry = false; + netfs_retry_write_stream(wreq, stream); + } + } +} |