* [PATCH 00/15] Various writeback related fixes @ 2020-01-06 20:24 Trond Myklebust 2020-01-06 20:25 ` [PATCH 01/15] NFS: Revalidate the file size on a fatal write error Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:24 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs A combination of writeback fixes for both regular NFS and pNFS. Trond Myklebust (15): NFS: Revalidate the file size on a fatal write error NFS: Revalidate the file mapping on all fatal writeback errors SUNRPC: Remove broken gss_mech_list_pseudoflavors() NFS: Fix up fsync() when the server rebooted NFS/pnfs: Fix pnfs_generic_prepare_to_resend_writes() NFSv4: Improve read/write/commit tracing NFS: Fix fix of show_nfs_errors pNFS/flexfiles: Record resend attempts on I/O failure NFS: Clean up generic file read tracepoints NFS: Clean up generic writeback tracepoints NFS: Clean up generic file commit tracepoint pNFS/flexfiles: Add tracing for layout errors NFS: Improve tracing of permission calls NFS: When resending after a short write, reset the reply count to zero NFS: Fix nfs_direct_write_reschedule_io() fs/nfs/dir.c | 4 +- fs/nfs/direct.c | 7 +- fs/nfs/file.c | 37 ++-- fs/nfs/flexfilelayout/flexfilelayout.c | 34 ++-- fs/nfs/nfs3xdr.c | 5 +- fs/nfs/nfs4trace.c | 4 + fs/nfs/nfs4trace.h | 202 +++++++++++++++++---- fs/nfs/nfs4xdr.c | 5 +- fs/nfs/nfstrace.h | 234 +++++++++++++++++-------- fs/nfs/pnfs.h | 8 +- fs/nfs/pnfs_nfs.c | 7 +- fs/nfs/read.c | 7 +- fs/nfs/write.c | 29 ++- include/linux/sunrpc/auth.h | 2 - include/linux/sunrpc/gss_api.h | 3 - net/sunrpc/auth.c | 49 ------ net/sunrpc/auth_gss/auth_gss.c | 1 - net/sunrpc/auth_gss/gss_mech_switch.c | 29 --- 18 files changed, 415 insertions(+), 252 deletions(-) -- 2.24.1 ^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH 01/15] NFS: Revalidate the file size on a fatal write error 2020-01-06 20:24 [PATCH 00/15] Various writeback related fixes Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 2020-01-06 20:25 ` [PATCH 02/15] NFS: Revalidate the file mapping on all fatal writeback errors Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs If we suffer a fatal error upon writing a file, which causes us to need to revalidate the entire mapping, then we should also revalidate the file size. Fixes: d2ceb7e57086 ("NFS: Don't use page_file_mapping after removing the page") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- fs/nfs/write.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 52cab65f91cf..f5170bc839aa 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -243,7 +243,15 @@ static void nfs_grow_file(struct page *page, unsigned int offset, unsigned int c /* A writeback failed: mark the page as bad, and invalidate the page cache */ static void nfs_set_pageerror(struct address_space *mapping) { + struct inode *inode = mapping->host; + nfs_zap_mapping(mapping->host, mapping); + /* Force file size revalidation */ + spin_lock(&inode->i_lock); + NFS_I(inode)->cache_validity |= NFS_INO_REVAL_FORCED | + NFS_INO_REVAL_PAGECACHE | + NFS_INO_INVALID_SIZE; + spin_unlock(&inode->i_lock); } static void nfs_mapping_set_error(struct page *page, int error) -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 02/15] NFS: Revalidate the file mapping on all fatal writeback errors 2020-01-06 20:25 ` [PATCH 01/15] NFS: Revalidate the file size on a fatal write error Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 2020-01-06 20:25 ` [PATCH 03/15] SUNRPC: Remove broken gss_mech_list_pseudoflavors() Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs If a write or commit failed, and the mapping sees a fatal error, we need to revalidate the contents of that mapping. Fixes: 06c9fdf3b9f1 ("NFS: On fatal writeback errors, we need to call nfs_inode_remove_request()") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- fs/nfs/write.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/fs/nfs/write.c b/fs/nfs/write.c index f5170bc839aa..83e6f691368c 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -256,8 +256,11 @@ static void nfs_set_pageerror(struct address_space *mapping) static void nfs_mapping_set_error(struct page *page, int error) { + struct address_space *mapping = page_file_mapping(page); + SetPageError(page); - mapping_set_error(page_file_mapping(page), error); + mapping_set_error(mapping, error); + nfs_set_pageerror(mapping); } /* @@ -600,7 +603,6 @@ nfs_lock_and_join_requests(struct page *page) static void nfs_write_error(struct nfs_page *req, int error) { - nfs_set_pageerror(page_file_mapping(req->wb_page)); nfs_mapping_set_error(req->wb_page, error); nfs_inode_remove_request(req); nfs_end_page_writeback(req); @@ -1006,7 +1008,6 @@ static void nfs_write_completion(struct nfs_pgio_header *hdr) nfs_list_remove_request(req); if (test_bit(NFS_IOHDR_ERROR, &hdr->flags) && (hdr->good_bytes < bytes)) { - nfs_set_pageerror(page_file_mapping(req->wb_page)); nfs_mapping_set_error(req->wb_page, hdr->error); goto remove_req; } -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 03/15] SUNRPC: Remove broken gss_mech_list_pseudoflavors() 2020-01-06 20:25 ` [PATCH 02/15] NFS: Revalidate the file mapping on all fatal writeback errors Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 2020-01-06 20:25 ` [PATCH 04/15] NFS: Fix up fsync() when the server rebooted Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs Remove gss_mech_list_pseudoflavors() and its callers. This is part of an unused API, and could leak an RCU reference if it were ever called. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- include/linux/sunrpc/auth.h | 2 -- include/linux/sunrpc/gss_api.h | 3 -- net/sunrpc/auth.c | 49 --------------------------- net/sunrpc/auth_gss/auth_gss.c | 1 - net/sunrpc/auth_gss/gss_mech_switch.c | 29 ---------------- 5 files changed, 84 deletions(-) diff --git a/include/linux/sunrpc/auth.h b/include/linux/sunrpc/auth.h index e9ec742796e7..4f6b28487f28 100644 --- a/include/linux/sunrpc/auth.h +++ b/include/linux/sunrpc/auth.h @@ -113,7 +113,6 @@ struct rpc_authops { int (*hash_cred)(struct auth_cred *, unsigned int); struct rpc_cred * (*lookup_cred)(struct rpc_auth *, struct auth_cred *, int); struct rpc_cred * (*crcreate)(struct rpc_auth*, struct auth_cred *, int, gfp_t); - int (*list_pseudoflavors)(rpc_authflavor_t *, int); rpc_authflavor_t (*info2flavor)(struct rpcsec_gss_info *); int (*flavor2info)(rpc_authflavor_t, struct rpcsec_gss_info *); @@ -158,7 +157,6 @@ rpc_authflavor_t rpcauth_get_pseudoflavor(rpc_authflavor_t, struct rpcsec_gss_info *); int rpcauth_get_gssinfo(rpc_authflavor_t, struct rpcsec_gss_info *); -int rpcauth_list_flavors(rpc_authflavor_t *, int); struct rpc_cred * rpcauth_lookup_credcache(struct rpc_auth *, struct auth_cred *, int, gfp_t); void rpcauth_init_cred(struct rpc_cred *, const struct auth_cred *, struct rpc_auth *, const struct rpc_credops *); struct rpc_cred * rpcauth_lookupcred(struct rpc_auth *, int); diff --git a/include/linux/sunrpc/gss_api.h b/include/linux/sunrpc/gss_api.h index bd691e08be3b..4c4cb087f3e8 100644 --- a/include/linux/sunrpc/gss_api.h +++ b/include/linux/sunrpc/gss_api.h @@ -150,9 +150,6 @@ struct gss_api_mech *gss_mech_get_by_name(const char *); /* Similar, but get by pseudoflavor. */ struct gss_api_mech *gss_mech_get_by_pseudoflavor(u32); -/* Fill in an array with a list of supported pseudoflavors */ -int gss_mech_list_pseudoflavors(rpc_authflavor_t *, int); - struct gss_api_mech * gss_mech_get(struct gss_api_mech *); /* For every successful gss_mech_get or gss_mech_get_by_* call there must be a diff --git a/net/sunrpc/auth.c b/net/sunrpc/auth.c index cdb05b48de44..5748ad0ba1bd 100644 --- a/net/sunrpc/auth.c +++ b/net/sunrpc/auth.c @@ -221,55 +221,6 @@ rpcauth_get_gssinfo(rpc_authflavor_t pseudoflavor, struct rpcsec_gss_info *info) } EXPORT_SYMBOL_GPL(rpcauth_get_gssinfo); -/** - * rpcauth_list_flavors - discover registered flavors and pseudoflavors - * @array: array to fill in - * @size: size of "array" - * - * Returns the number of array items filled in, or a negative errno. - * - * The returned array is not sorted by any policy. Callers should not - * rely on the order of the items in the returned array. - */ -int -rpcauth_list_flavors(rpc_authflavor_t *array, int size) -{ - const struct rpc_authops *ops; - rpc_authflavor_t flavor, pseudos[4]; - int i, len, result = 0; - - rcu_read_lock(); - for (flavor = 0; flavor < RPC_AUTH_MAXFLAVOR; flavor++) { - ops = rcu_dereference(auth_flavors[flavor]); - if (result >= size) { - result = -ENOMEM; - break; - } - - if (ops == NULL) - continue; - if (ops->list_pseudoflavors == NULL) { - array[result++] = ops->au_flavor; - continue; - } - len = ops->list_pseudoflavors(pseudos, ARRAY_SIZE(pseudos)); - if (len < 0) { - result = len; - break; - } - for (i = 0; i < len; i++) { - if (result >= size) { - result = -ENOMEM; - break; - } - array[result++] = pseudos[i]; - } - } - rcu_read_unlock(); - return result; -} -EXPORT_SYMBOL_GPL(rpcauth_list_flavors); - struct rpc_auth * rpcauth_create(const struct rpc_auth_create_args *args, struct rpc_clnt *clnt) { diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c index d75fddca44c9..24ca861815b1 100644 --- a/net/sunrpc/auth_gss/auth_gss.c +++ b/net/sunrpc/auth_gss/auth_gss.c @@ -2118,7 +2118,6 @@ static const struct rpc_authops authgss_ops = { .hash_cred = gss_hash_cred, .lookup_cred = gss_lookup_cred, .crcreate = gss_create_cred, - .list_pseudoflavors = gss_mech_list_pseudoflavors, .info2flavor = gss_mech_info2flavor, .flavor2info = gss_mech_flavor2info, }; diff --git a/net/sunrpc/auth_gss/gss_mech_switch.c b/net/sunrpc/auth_gss/gss_mech_switch.c index 30b7de6f3d76..22d4be8e38c8 100644 --- a/net/sunrpc/auth_gss/gss_mech_switch.c +++ b/net/sunrpc/auth_gss/gss_mech_switch.c @@ -219,35 +219,6 @@ gss_mech_get_by_pseudoflavor(u32 pseudoflavor) return gm; } -/** - * gss_mech_list_pseudoflavors - Discover registered GSS pseudoflavors - * @array_ptr: array to fill in - * @size: size of "array" - * - * Returns the number of array items filled in, or a negative errno. - * - * The returned array is not sorted by any policy. Callers should not - * rely on the order of the items in the returned array. - */ -int gss_mech_list_pseudoflavors(rpc_authflavor_t *array_ptr, int size) -{ - struct gss_api_mech *pos = NULL; - int j, i = 0; - - rcu_read_lock(); - list_for_each_entry_rcu(pos, ®istered_mechs, gm_list) { - for (j = 0; j < pos->gm_pf_num; j++) { - if (i >= size) { - spin_unlock(®istered_mechs_lock); - return -ENOMEM; - } - array_ptr[i++] = pos->gm_pfs[j].pseudoflavor; - } - } - rcu_read_unlock(); - return i; -} - /** * gss_svc_to_pseudoflavor - map a GSS service number to a pseudoflavor * @gm: GSS mechanism handle -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 04/15] NFS: Fix up fsync() when the server rebooted 2020-01-06 20:25 ` [PATCH 03/15] SUNRPC: Remove broken gss_mech_list_pseudoflavors() Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 2020-01-06 20:25 ` [PATCH 05/15] NFS/pnfs: Fix pnfs_generic_prepare_to_resend_writes() Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs Don't clear the NFS_CONTEXT_RESEND_WRITES flag until after calling nfs_commit_inode(). Otherwise, if nfs_commit_inode() returns an error, we end up with dirty pages in the page cache, but no tag to tell us that those pages need resending. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- fs/nfs/file.c | 37 ++++++++++++++++--------------------- 1 file changed, 16 insertions(+), 21 deletions(-) diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 8eb731d9be3e..95a3445c8926 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -204,44 +204,39 @@ EXPORT_SYMBOL_GPL(nfs_file_mmap); static int nfs_file_fsync_commit(struct file *file, int datasync) { - struct nfs_open_context *ctx = nfs_file_open_context(file); struct inode *inode = file_inode(file); - int do_resend, status; - int ret = 0; + int ret; dprintk("NFS: fsync file(%pD2) datasync %d\n", file, datasync); nfs_inc_stats(inode, NFSIOS_VFSFSYNC); - do_resend = test_and_clear_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags); - status = nfs_commit_inode(inode, FLUSH_SYNC); - if (status == 0) - status = file_check_and_advance_wb_err(file); - if (status < 0) { - ret = status; - goto out; - } - do_resend |= test_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags); - if (do_resend) - ret = -EAGAIN; -out: - return ret; + ret = nfs_commit_inode(inode, FLUSH_SYNC); + if (ret < 0) + return ret; + return file_check_and_advance_wb_err(file); } int nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync) { - int ret; + struct nfs_open_context *ctx = nfs_file_open_context(file); struct inode *inode = file_inode(file); + int ret; trace_nfs_fsync_enter(inode); - do { + for (;;) { ret = file_write_and_wait_range(file, start, end); if (ret != 0) break; ret = nfs_file_fsync_commit(file, datasync); - if (!ret) - ret = pnfs_sync_inode(inode, !!datasync); + if (ret != 0) + break; + ret = pnfs_sync_inode(inode, !!datasync); + if (ret != 0) + break; + if (!test_and_clear_bit(NFS_CONTEXT_RESEND_WRITES, &ctx->flags)) + break; /* * If nfs_file_fsync_commit detected a server reboot, then * resend all dirty pages that might have been covered by @@ -249,7 +244,7 @@ nfs_file_fsync(struct file *file, loff_t start, loff_t end, int datasync) */ start = 0; end = LLONG_MAX; - } while (ret == -EAGAIN); + } trace_nfs_fsync_exit(inode, ret); return ret; -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 05/15] NFS/pnfs: Fix pnfs_generic_prepare_to_resend_writes() 2020-01-06 20:25 ` [PATCH 04/15] NFS: Fix up fsync() when the server rebooted Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 2020-01-06 20:25 ` [PATCH 06/15] NFSv4: Improve read/write/commit tracing Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs Instead of making assumptions about the commit verifier contents, change the commit code to ensure we always check that the verifier was set by the XDR code. Fixes: f54bcf2ecee9 ("pnfs: Prepare for flexfiles by pulling out common code") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- fs/nfs/direct.c | 4 ++-- fs/nfs/nfs3xdr.c | 5 ++++- fs/nfs/nfs4xdr.c | 5 ++++- fs/nfs/pnfs_nfs.c | 7 +++---- fs/nfs/write.c | 4 +++- 5 files changed, 16 insertions(+), 9 deletions(-) diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 040a50fd9bf3..29f00da8a0b7 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -245,10 +245,10 @@ static int nfs_direct_cmp_commit_data_verf(struct nfs_direct_req *dreq, data->ds_commit_index); /* verifier not set so always fail */ - if (verfp->committed < 0) + if (verfp->committed < 0 || data->res.verf->committed <= NFS_UNSTABLE) return 1; - return nfs_direct_cmp_verf(verfp, &data->verf); + return nfs_direct_cmp_verf(verfp, data->res.verf); } /** diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index 927eb680f161..69971f6c840d 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -2334,6 +2334,7 @@ static int nfs3_xdr_dec_commit3res(struct rpc_rqst *req, void *data) { struct nfs_commitres *result = data; + struct nfs_writeverf *verf = result->verf; enum nfs_stat status; int error; @@ -2346,7 +2347,9 @@ static int nfs3_xdr_dec_commit3res(struct rpc_rqst *req, result->op_status = status; if (status != NFS3_OK) goto out_status; - error = decode_writeverf3(xdr, &result->verf->verifier); + error = decode_writeverf3(xdr, &verf->verifier); + if (!error) + verf->committed = NFS_FILE_SYNC; out: return error; out_status: diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 936c57779ff4..d0feef17db50 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -4313,11 +4313,14 @@ static int decode_write_verifier(struct xdr_stream *xdr, struct nfs_write_verifi static int decode_commit(struct xdr_stream *xdr, struct nfs_commitres *res) { + struct nfs_writeverf *verf = res->verf; int status; status = decode_op_hdr(xdr, OP_COMMIT); if (!status) - status = decode_write_verifier(xdr, &res->verf->verifier); + status = decode_write_verifier(xdr, &verf->verifier); + if (!status) + verf->committed = NFS_FILE_SYNC; return status; } diff --git a/fs/nfs/pnfs_nfs.c b/fs/nfs/pnfs_nfs.c index 82af4809b869..8b37e7f8e789 100644 --- a/fs/nfs/pnfs_nfs.c +++ b/fs/nfs/pnfs_nfs.c @@ -31,12 +31,11 @@ EXPORT_SYMBOL_GPL(pnfs_generic_rw_release); /* Fake up some data that will cause nfs_commit_release to retry the writes. */ void pnfs_generic_prepare_to_resend_writes(struct nfs_commit_data *data) { - struct nfs_page *first = nfs_list_entry(data->pages.next); + struct nfs_writeverf *verf = data->res.verf; data->task.tk_status = 0; - memcpy(&data->verf.verifier, &first->wb_verf, - sizeof(data->verf.verifier)); - data->verf.verifier.data[0]++; /* ensure verifier mismatch */ + memset(&verf->verifier, 0, sizeof(verf->verifier)); + verf->committed = NFS_UNSTABLE; } EXPORT_SYMBOL_GPL(pnfs_generic_prepare_to_resend_writes); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 83e6f691368c..ab3d4611f0aa 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1838,6 +1838,7 @@ static void nfs_commit_done(struct rpc_task *task, void *calldata) static void nfs_commit_release_pages(struct nfs_commit_data *data) { + const struct nfs_writeverf *verf = data->res.verf; struct nfs_page *req; int status = data->task.tk_status; struct nfs_commit_info cinfo; @@ -1865,7 +1866,8 @@ static void nfs_commit_release_pages(struct nfs_commit_data *data) /* Okay, COMMIT succeeded, apparently. Check the verifier * returned by the server against all stored verfs. */ - if (!nfs_write_verifier_cmp(&req->wb_verf, &data->verf.verifier)) { + if (verf->committed > NFS_UNSTABLE && + !nfs_write_verifier_cmp(&req->wb_verf, &verf->verifier)) { /* We have a match */ if (req->wb_page) nfs_inode_remove_request(req); -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 06/15] NFSv4: Improve read/write/commit tracing 2020-01-06 20:25 ` [PATCH 05/15] NFS/pnfs: Fix pnfs_generic_prepare_to_resend_writes() Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 2020-01-06 20:25 ` [PATCH 07/15] NFS: Fix fix of show_nfs_errors Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs Ensure we always return the number of bytes read/written. Also display the pnfs filehandle if it is in use. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- fs/nfs/nfs4trace.h | 52 ++++++++++++++++++++++++++++++---------------- 1 file changed, 34 insertions(+), 18 deletions(-) diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index e60b6fbd5ada..a3155b7f8063 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -1694,7 +1694,8 @@ DECLARE_EVENT_CLASS(nfs4_read_event, __field(u32, fhandle) __field(u64, fileid) __field(loff_t, offset) - __field(size_t, count) + __field(u32, arg_count) + __field(u32, res_count) __field(unsigned long, error) __field(int, stateid_seq) __field(u32, stateid_hash) @@ -1702,13 +1703,18 @@ DECLARE_EVENT_CLASS(nfs4_read_event, TP_fast_assign( const struct inode *inode = hdr->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = hdr->args.fh ? + hdr->args.fh : &nfsi->fh; const struct nfs4_state *state = hdr->args.context->state; + __entry->dev = inode->i_sb->s_dev; - __entry->fileid = NFS_FILEID(inode); - __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->fileid = nfsi->fileid; + __entry->fhandle = nfs_fhandle_hash(fh); __entry->offset = hdr->args.offset; - __entry->count = hdr->args.count; + __entry->arg_count = hdr->args.count; + __entry->res_count = hdr->res.count; __entry->error = error < 0 ? -error : 0; __entry->stateid_seq = be32_to_cpu(state->stateid.seqid); @@ -1718,14 +1724,14 @@ DECLARE_EVENT_CLASS(nfs4_read_event, TP_printk( "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " - "offset=%lld count=%zu stateid=%d:0x%08x", + "offset=%lld count=%u res=%u stateid=%d:0x%08x", -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, (long long)__entry->offset, - __entry->count, + __entry->arg_count, __entry->res_count, __entry->stateid_seq, __entry->stateid_hash ) ); @@ -1754,7 +1760,8 @@ DECLARE_EVENT_CLASS(nfs4_write_event, __field(u32, fhandle) __field(u64, fileid) __field(loff_t, offset) - __field(size_t, count) + __field(u32, arg_count) + __field(u32, res_count) __field(unsigned long, error) __field(int, stateid_seq) __field(u32, stateid_hash) @@ -1762,13 +1769,18 @@ DECLARE_EVENT_CLASS(nfs4_write_event, TP_fast_assign( const struct inode *inode = hdr->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = hdr->args.fh ? + hdr->args.fh : &nfsi->fh; const struct nfs4_state *state = hdr->args.context->state; + __entry->dev = inode->i_sb->s_dev; - __entry->fileid = NFS_FILEID(inode); - __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->fileid = nfsi->fileid; + __entry->fhandle = nfs_fhandle_hash(fh); __entry->offset = hdr->args.offset; - __entry->count = hdr->args.count; + __entry->arg_count = hdr->args.count; + __entry->res_count = hdr->res.count; __entry->error = error < 0 ? -error : 0; __entry->stateid_seq = be32_to_cpu(state->stateid.seqid); @@ -1778,14 +1790,14 @@ DECLARE_EVENT_CLASS(nfs4_write_event, TP_printk( "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " - "offset=%lld count=%zu stateid=%d:0x%08x", + "offset=%lld count=%u res=%u stateid=%d:0x%08x", -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, (long long)__entry->offset, - __entry->count, + __entry->arg_count, __entry->res_count, __entry->stateid_seq, __entry->stateid_hash ) ); @@ -1814,24 +1826,28 @@ DECLARE_EVENT_CLASS(nfs4_commit_event, __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) - __field(loff_t, offset) - __field(size_t, count) __field(unsigned long, error) + __field(loff_t, offset) + __field(u32, count) ), TP_fast_assign( const struct inode *inode = data->inode; + const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = data->args.fh ? + data->args.fh : &nfsi->fh; + __entry->dev = inode->i_sb->s_dev; - __entry->fileid = NFS_FILEID(inode); - __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); + __entry->fileid = nfsi->fileid; + __entry->fhandle = nfs_fhandle_hash(fh); __entry->offset = data->args.offset; __entry->count = data->args.count; - __entry->error = error; + __entry->error = error < 0 ? -error : 0; ), TP_printk( "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " - "offset=%lld count=%zu", + "offset=%lld count=%u", -__entry->error, show_nfsv4_errors(__entry->error), MAJOR(__entry->dev), MINOR(__entry->dev), -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 07/15] NFS: Fix fix of show_nfs_errors 2020-01-06 20:25 ` [PATCH 06/15] NFSv4: Improve read/write/commit tracing Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 2020-01-06 20:25 ` [PATCH 08/15] pNFS/flexfiles: Record resend attempts on I/O failure Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs Casting a negative value to an unsigned long is not the same as converting it to its absolute value. Fixes: 96650e2effa2 ("NFS: Fix show_nfs_errors macros again") Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- fs/nfs/nfs4trace.h | 33 +++++++++++++++++---------------- 1 file changed, 17 insertions(+), 16 deletions(-) diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index a3155b7f8063..a291877c0c32 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -352,7 +352,7 @@ DECLARE_EVENT_CLASS(nfs4_clientid_event, ), TP_fast_assign( - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __assign_str(dstaddr, clp->cl_hostname); ), @@ -432,7 +432,8 @@ TRACE_EVENT(nfs4_sequence_done, __entry->target_highest_slotid = res->sr_target_highest_slotid; __entry->status_flags = res->sr_status_flags; - __entry->error = res->sr_status; + __entry->error = res->sr_status < 0 ? + -res->sr_status : 0; ), TP_printk( "error=%ld (%s) session=0x%08x slot_nr=%u seq_nr=%u " @@ -640,7 +641,7 @@ TRACE_EVENT(nfs4_state_mgr_failed, ), TP_fast_assign( - __entry->error = status; + __entry->error = status < 0 ? -status : 0; __entry->state = clp->cl_state; __assign_str(hostname, clp->cl_hostname); __assign_str(section, section); @@ -659,7 +660,7 @@ TRACE_EVENT(nfs4_xdr_status, TP_PROTO( const struct xdr_stream *xdr, u32 op, - int error + u32 error ), TP_ARGS(xdr, op, error), @@ -849,7 +850,7 @@ TRACE_EVENT(nfs4_close, __entry->fileid = NFS_FILEID(inode); __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); __entry->fmode = (__force unsigned int)state->state; - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __entry->stateid_seq = be32_to_cpu(args->stateid.seqid); __entry->stateid_hash = @@ -914,7 +915,7 @@ DECLARE_EVENT_CLASS(nfs4_lock_event, TP_fast_assign( const struct inode *inode = state->inode; - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __entry->cmd = cmd; __entry->type = request->fl_type; __entry->start = request->fl_start; @@ -986,7 +987,7 @@ TRACE_EVENT(nfs4_set_lock, TP_fast_assign( const struct inode *inode = state->inode; - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __entry->cmd = cmd; __entry->type = request->fl_type; __entry->start = request->fl_start; @@ -1164,7 +1165,7 @@ TRACE_EVENT(nfs4_delegreturn_exit, TP_fast_assign( __entry->dev = res->server->s_dev; __entry->fhandle = nfs_fhandle_hash(args->fhandle); - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __entry->stateid_seq = be32_to_cpu(args->stateid->seqid); __entry->stateid_hash = @@ -1204,7 +1205,7 @@ DECLARE_EVENT_CLASS(nfs4_test_stateid_event, TP_fast_assign( const struct inode *inode = state->inode; - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __entry->dev = inode->i_sb->s_dev; __entry->fileid = NFS_FILEID(inode); __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); @@ -1306,7 +1307,7 @@ TRACE_EVENT(nfs4_lookupp, TP_fast_assign( __entry->dev = inode->i_sb->s_dev; __entry->ino = NFS_FILEID(inode); - __entry->error = error; + __entry->error = error < 0 ? -error : 0; ), TP_printk( @@ -1342,7 +1343,7 @@ TRACE_EVENT(nfs4_rename, __entry->dev = olddir->i_sb->s_dev; __entry->olddir = NFS_FILEID(olddir); __entry->newdir = NFS_FILEID(newdir); - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __assign_str(oldname, oldname->name); __assign_str(newname, newname->name); ), @@ -1433,7 +1434,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_stateid_event, __entry->dev = inode->i_sb->s_dev; __entry->fileid = NFS_FILEID(inode); __entry->fhandle = nfs_fhandle_hash(NFS_FH(inode)); - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __entry->stateid_seq = be32_to_cpu(stateid->seqid); __entry->stateid_hash = @@ -1489,7 +1490,7 @@ DECLARE_EVENT_CLASS(nfs4_getattr_event, __entry->valid = fattr->valid; __entry->fhandle = nfs_fhandle_hash(fhandle); __entry->fileid = (fattr->valid & NFS_ATTR_FATTR_FILEID) ? fattr->fileid : 0; - __entry->error = error; + __entry->error = error < 0 ? -error : 0; ), TP_printk( @@ -1536,7 +1537,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_callback_event, ), TP_fast_assign( - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __entry->fhandle = nfs_fhandle_hash(fhandle); if (!IS_ERR_OR_NULL(inode)) { __entry->fileid = NFS_FILEID(inode); @@ -1593,7 +1594,7 @@ DECLARE_EVENT_CLASS(nfs4_inode_stateid_callback_event, ), TP_fast_assign( - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __entry->fhandle = nfs_fhandle_hash(fhandle); if (!IS_ERR_OR_NULL(inode)) { __entry->fileid = NFS_FILEID(inode); @@ -1912,7 +1913,7 @@ TRACE_EVENT(nfs4_layoutget, __entry->iomode = args->iomode; __entry->offset = args->offset; __entry->count = args->length; - __entry->error = error; + __entry->error = error < 0 ? -error : 0; __entry->stateid_seq = be32_to_cpu(state->stateid.seqid); __entry->stateid_hash = -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 08/15] pNFS/flexfiles: Record resend attempts on I/O failure 2020-01-06 20:25 ` [PATCH 07/15] NFS: Fix fix of show_nfs_errors Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 2020-01-06 20:25 ` [PATCH 09/15] NFS: Clean up generic file read tracepoints Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs If the attempt to do pNFS fails, then record what action we take to recover (resend, reset to pnfs or reset to mds). Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- fs/nfs/flexfilelayout/flexfilelayout.c | 6 +++--- fs/nfs/nfs4trace.h | 8 +++++++- fs/nfs/pnfs.h | 8 ++++---- 3 files changed, 14 insertions(+), 8 deletions(-) diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c index 5657b7f2611f..3163b78b1d2c 100644 --- a/fs/nfs/flexfilelayout/flexfilelayout.c +++ b/fs/nfs/flexfilelayout/flexfilelayout.c @@ -1321,7 +1321,6 @@ static int ff_layout_read_done_cb(struct rpc_task *task, int new_idx = hdr->pgio_mirror_idx; int err; - trace_nfs4_pnfs_read(hdr, task->tk_status); if (task->tk_status < 0) ff_layout_io_track_ds_error(hdr->lseg, hdr->pgio_mirror_idx, hdr->args.offset, hdr->args.count, @@ -1331,6 +1330,7 @@ static int ff_layout_read_done_cb(struct rpc_task *task, hdr->ds_clp, hdr->lseg, hdr->pgio_mirror_idx); + trace_nfs4_pnfs_read(hdr, err); clear_bit(NFS_IOHDR_RESEND_PNFS, &hdr->flags); clear_bit(NFS_IOHDR_RESEND_MDS, &hdr->flags); switch (err) { @@ -1494,7 +1494,6 @@ static int ff_layout_write_done_cb(struct rpc_task *task, loff_t end_offs = 0; int err; - trace_nfs4_pnfs_write(hdr, task->tk_status); if (task->tk_status < 0) ff_layout_io_track_ds_error(hdr->lseg, hdr->pgio_mirror_idx, hdr->args.offset, hdr->args.count, @@ -1504,6 +1503,7 @@ static int ff_layout_write_done_cb(struct rpc_task *task, hdr->ds_clp, hdr->lseg, hdr->pgio_mirror_idx); + trace_nfs4_pnfs_write(hdr, err); clear_bit(NFS_IOHDR_RESEND_PNFS, &hdr->flags); clear_bit(NFS_IOHDR_RESEND_MDS, &hdr->flags); switch (err) { @@ -1537,7 +1537,6 @@ static int ff_layout_commit_done_cb(struct rpc_task *task, { int err; - trace_nfs4_pnfs_commit_ds(data, task->tk_status); if (task->tk_status < 0) ff_layout_io_track_ds_error(data->lseg, data->ds_commit_index, data->args.offset, data->args.count, @@ -1546,6 +1545,7 @@ static int ff_layout_commit_done_cb(struct rpc_task *task, err = ff_layout_async_handle_error(task, NULL, data->ds_clp, data->lseg, data->ds_commit_index); + trace_nfs4_pnfs_commit_ds(data, err); switch (err) { case -NFS4ERR_RESET_TO_PNFS: pnfs_generic_prepare_to_resend_writes(data); diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index a291877c0c32..a467f49989f2 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -155,6 +155,9 @@ TRACE_DEFINE_ENUM(NFS4ERR_WRONG_CRED); TRACE_DEFINE_ENUM(NFS4ERR_WRONG_TYPE); TRACE_DEFINE_ENUM(NFS4ERR_XDEV); +TRACE_DEFINE_ENUM(NFS4ERR_RESET_TO_MDS); +TRACE_DEFINE_ENUM(NFS4ERR_RESET_TO_PNFS); + #define show_nfsv4_errors(error) \ __print_symbolic(error, \ { NFS4_OK, "OK" }, \ @@ -305,7 +308,10 @@ TRACE_DEFINE_ENUM(NFS4ERR_XDEV); { NFS4ERR_WRONGSEC, "WRONGSEC" }, \ { NFS4ERR_WRONG_CRED, "WRONG_CRED" }, \ { NFS4ERR_WRONG_TYPE, "WRONG_TYPE" }, \ - { NFS4ERR_XDEV, "XDEV" }) + { NFS4ERR_XDEV, "XDEV" }, \ + /* ***** Internal to Linux NFS client ***** */ \ + { NFS4ERR_RESET_TO_MDS, "RESET_TO_MDS" }, \ + { NFS4ERR_RESET_TO_PNFS, "RESET_TO_PNFS" }) #define show_open_flags(flags) \ __print_flags(flags, "|", \ diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index f8a38065c7e4..0fafdadc9c8d 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -79,6 +79,10 @@ enum pnfs_try_status { PNFS_TRY_AGAIN = 2, }; +/* error codes for internal use */ +#define NFS4ERR_RESET_TO_MDS 12001 +#define NFS4ERR_RESET_TO_PNFS 12002 + #ifdef CONFIG_NFS_V4_1 #define LAYOUT_NFSV4_1_MODULE_PREFIX "nfs-layouttype4" @@ -91,10 +95,6 @@ enum pnfs_try_status { #define NFS4_DEF_DS_RETRANS 5 #define PNFS_DEVICE_RETRY_TIMEOUT (120*HZ) -/* error codes for internal use */ -#define NFS4ERR_RESET_TO_MDS 12001 -#define NFS4ERR_RESET_TO_PNFS 12002 - enum { NFS_LAYOUT_RO_FAILED = 0, /* get ro layout failed stop trying */ NFS_LAYOUT_RW_FAILED, /* get rw layout failed stop trying */ -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 09/15] NFS: Clean up generic file read tracepoints 2020-01-06 20:25 ` [PATCH 08/15] pNFS/flexfiles: Record resend attempts on I/O failure Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 2020-01-06 20:25 ` [PATCH 10/15] NFS: Clean up generic writeback tracepoints Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs Clean up the generic file read tracepoints so they do pass the full structures as arguments. Also ensure we report the number of bytes actually read. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- fs/nfs/nfstrace.h | 56 ++++++++++++++++++++++++++++------------------- fs/nfs/read.c | 5 ++--- 2 files changed, 35 insertions(+), 26 deletions(-) diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index f64a33d2a1d1..f0e869d21368 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -820,75 +820,85 @@ TRACE_EVENT(nfs_sillyrename_unlink, TRACE_EVENT(nfs_initiate_read, TP_PROTO( - const struct inode *inode, - loff_t offset, unsigned long count + const struct nfs_pgio_header *hdr ), - TP_ARGS(inode, offset, count), + TP_ARGS(hdr), TP_STRUCT__entry( - __field(loff_t, offset) - __field(unsigned long, count) __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) + __field(loff_t, offset) + __field(u32, count) ), TP_fast_assign( + const struct inode *inode = hdr->inode; const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = hdr->args.fh ? + hdr->args.fh : &nfsi->fh; - __entry->offset = offset; - __entry->count = count; + __entry->offset = hdr->args.offset; + __entry->count = hdr->args.count; __entry->dev = inode->i_sb->s_dev; __entry->fileid = nfsi->fileid; - __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->fhandle = nfs_fhandle_hash(fh); ), TP_printk( "fileid=%02x:%02x:%llu fhandle=0x%08x " - "offset=%lld count=%lu", + "offset=%lld count=%u", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, - __entry->offset, __entry->count + (long long)__entry->offset, __entry->count ) ); TRACE_EVENT(nfs_readpage_done, TP_PROTO( - const struct inode *inode, - int status, loff_t offset, bool eof + const struct rpc_task *task, + const struct nfs_pgio_header *hdr ), - TP_ARGS(inode, status, offset, eof), + TP_ARGS(task, hdr), TP_STRUCT__entry( - __field(int, status) - __field(loff_t, offset) - __field(bool, eof) __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) + __field(loff_t, offset) + __field(u32, arg_count) + __field(u32, res_count) + __field(bool, eof) + __field(int, status) ), TP_fast_assign( + const struct inode *inode = hdr->inode; const struct nfs_inode *nfsi = NFS_I(inode); - - __entry->status = status; - __entry->offset = offset; - __entry->eof = eof; + const struct nfs_fh *fh = hdr->args.fh ? + hdr->args.fh : &nfsi->fh; + + __entry->status = task->tk_status; + __entry->offset = hdr->args.offset; + __entry->arg_count = hdr->args.count; + __entry->res_count = hdr->res.count; + __entry->eof = hdr->res.eof; __entry->dev = inode->i_sb->s_dev; __entry->fileid = nfsi->fileid; - __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->fhandle = nfs_fhandle_hash(fh); ), TP_printk( "fileid=%02x:%02x:%llu fhandle=0x%08x " - "offset=%lld status=%d%s", + "offset=%lld count=%u res=%u status=%d%s", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, - __entry->offset, __entry->status, + (long long)__entry->offset, __entry->arg_count, + __entry->res_count, __entry->status, __entry->eof ? " eof" : "" ) ); diff --git a/fs/nfs/read.c b/fs/nfs/read.c index cfe0b586eadd..12deb3bdb2a0 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -214,7 +214,7 @@ static void nfs_initiate_read(struct nfs_pgio_header *hdr, task_setup_data->flags |= swap_flags; rpc_ops->read_setup(hdr, msg); - trace_nfs_initiate_read(inode, hdr->io_start, hdr->good_bytes); + trace_nfs_initiate_read(hdr); } static void @@ -247,8 +247,7 @@ static int nfs_readpage_done(struct rpc_task *task, return status; nfs_add_stats(inode, NFSIOS_SERVERREADBYTES, hdr->res.count); - trace_nfs_readpage_done(inode, task->tk_status, - hdr->args.offset, hdr->res.eof); + trace_nfs_readpage_done(task, hdr); if (task->tk_status == -ESTALE) { set_bit(NFS_INO_STALE, &NFS_I(inode)->flags); -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 10/15] NFS: Clean up generic writeback tracepoints 2020-01-06 20:25 ` [PATCH 09/15] NFS: Clean up generic file read tracepoints Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 2020-01-06 20:25 ` [PATCH 11/15] NFS: Clean up generic file commit tracepoint Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs Clean up the generic writeback tracepoints so they do pass the full structures as arguments. Also ensure we report the number of bytes actually written. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- fs/nfs/nfstrace.h | 73 ++++++++++++++++++++++++++--------------------- fs/nfs/write.c | 6 ++-- 2 files changed, 43 insertions(+), 36 deletions(-) diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index f0e869d21368..3d8d36fe7b6f 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -915,87 +915,96 @@ TRACE_DEFINE_ENUM(NFS_FILE_SYNC); TRACE_EVENT(nfs_initiate_write, TP_PROTO( - const struct inode *inode, - loff_t offset, unsigned long count, - enum nfs3_stable_how stable + const struct nfs_pgio_header *hdr ), - TP_ARGS(inode, offset, count, stable), + TP_ARGS(hdr), TP_STRUCT__entry( - __field(loff_t, offset) - __field(unsigned long, count) - __field(enum nfs3_stable_how, stable) __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) + __field(loff_t, offset) + __field(u32, count) + __field(enum nfs3_stable_how, stable) ), TP_fast_assign( + const struct inode *inode = hdr->inode; const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = hdr->args.fh ? + hdr->args.fh : &nfsi->fh; - __entry->offset = offset; - __entry->count = count; - __entry->stable = stable; + __entry->offset = hdr->args.offset; + __entry->count = hdr->args.count; + __entry->stable = hdr->args.stable; __entry->dev = inode->i_sb->s_dev; __entry->fileid = nfsi->fileid; - __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->fhandle = nfs_fhandle_hash(fh); ), TP_printk( "fileid=%02x:%02x:%llu fhandle=0x%08x " - "offset=%lld count=%lu stable=%s", + "offset=%lld count=%u stable=%s", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, - __entry->offset, __entry->count, + (long long)__entry->offset, __entry->count, nfs_show_stable(__entry->stable) ) ); TRACE_EVENT(nfs_writeback_done, TP_PROTO( - const struct inode *inode, - int status, - loff_t offset, - struct nfs_writeverf *writeverf + const struct rpc_task *task, + const struct nfs_pgio_header *hdr ), - TP_ARGS(inode, status, offset, writeverf), + TP_ARGS(task, hdr), TP_STRUCT__entry( - __field(int, status) - __field(loff_t, offset) - __field(enum nfs3_stable_how, stable) - __field(unsigned long long, verifier) __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) + __field(loff_t, offset) + __field(u32, arg_count) + __field(u32, res_count) + __field(int, status) + __field(enum nfs3_stable_how, stable) + __array(char, verifier, NFS4_VERIFIER_SIZE) ), TP_fast_assign( + const struct inode *inode = hdr->inode; const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = hdr->args.fh ? + hdr->args.fh : &nfsi->fh; + const struct nfs_writeverf *verf = hdr->res.verf; - __entry->status = status; - __entry->offset = offset; - __entry->stable = writeverf->committed; - memcpy(&__entry->verifier, &writeverf->verifier, - sizeof(__entry->verifier)); + __entry->status = task->tk_status; + __entry->offset = hdr->args.offset; + __entry->arg_count = hdr->args.count; + __entry->res_count = hdr->res.count; + __entry->stable = verf->committed; + memcpy(__entry->verifier, + &verf->verifier, + NFS4_VERIFIER_SIZE); __entry->dev = inode->i_sb->s_dev; __entry->fileid = nfsi->fileid; - __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->fhandle = nfs_fhandle_hash(fh); ), TP_printk( "fileid=%02x:%02x:%llu fhandle=0x%08x " - "offset=%lld status=%d stable=%s " - "verifier 0x%016llx", + "offset=%lld count=%u res=%u status=%d stable=%s " + "verifier=%s", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, - __entry->offset, __entry->status, + (long long)__entry->offset, __entry->arg_count, + __entry->res_count, __entry->status, nfs_show_stable(__entry->stable), - __entry->verifier + __print_hex_str(__entry->verifier, NFS4_VERIFIER_SIZE) ) ); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index ab3d4611f0aa..d0f62aef3489 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1412,8 +1412,7 @@ static void nfs_initiate_write(struct nfs_pgio_header *hdr, task_setup_data->priority = priority; rpc_ops->write_setup(hdr, msg, &task_setup_data->rpc_client); - trace_nfs_initiate_write(hdr->inode, hdr->io_start, hdr->good_bytes, - hdr->args.stable); + trace_nfs_initiate_write(hdr); } /* If a nfs_flush_* function fails, it should remove reqs from @head and @@ -1577,8 +1576,7 @@ static int nfs_writeback_done(struct rpc_task *task, return status; nfs_add_stats(inode, NFSIOS_SERVERWRITTENBYTES, hdr->res.count); - trace_nfs_writeback_done(inode, task->tk_status, - hdr->args.offset, hdr->res.verf); + trace_nfs_writeback_done(task, hdr); if (hdr->res.verf->committed < hdr->args.stable && task->tk_status >= 0) { -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 11/15] NFS: Clean up generic file commit tracepoint 2020-01-06 20:25 ` [PATCH 10/15] NFS: Clean up generic writeback tracepoints Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 2020-01-06 20:25 ` [PATCH 12/15] pNFS/flexfiles: Add tracing for layout errors Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs Clean up the generic file commit tracepoints to use a 64-bit value for the verifier, and to display the pNFS filehandle, if it exists. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- fs/nfs/nfstrace.h | 42 ++++++++++++++++++++++++++---------------- fs/nfs/write.c | 2 +- 2 files changed, 27 insertions(+), 17 deletions(-) diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index 3d8d36fe7b6f..51043f02e86f 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -1016,71 +1016,81 @@ TRACE_EVENT(nfs_initiate_commit, TP_ARGS(data), TP_STRUCT__entry( - __field(loff_t, offset) - __field(unsigned long, count) __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) + __field(loff_t, offset) + __field(u32, count) ), TP_fast_assign( const struct inode *inode = data->inode; const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = data->args.fh ? + data->args.fh : &nfsi->fh; __entry->offset = data->args.offset; __entry->count = data->args.count; __entry->dev = inode->i_sb->s_dev; __entry->fileid = nfsi->fileid; - __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->fhandle = nfs_fhandle_hash(fh); ), TP_printk( "fileid=%02x:%02x:%llu fhandle=0x%08x " - "offset=%lld count=%lu", + "offset=%lld count=%u", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, - __entry->offset, __entry->count + (long long)__entry->offset, __entry->count ) ); TRACE_EVENT(nfs_commit_done, TP_PROTO( + const struct rpc_task *task, const struct nfs_commit_data *data ), - TP_ARGS(data), + TP_ARGS(task, data), TP_STRUCT__entry( - __field(int, status) - __field(loff_t, offset) - __field(unsigned long long, verifier) __field(dev_t, dev) __field(u32, fhandle) __field(u64, fileid) + __field(loff_t, offset) + __field(int, status) + __field(enum nfs3_stable_how, stable) + __array(char, verifier, NFS4_VERIFIER_SIZE) ), TP_fast_assign( const struct inode *inode = data->inode; const struct nfs_inode *nfsi = NFS_I(inode); + const struct nfs_fh *fh = data->args.fh ? + data->args.fh : &nfsi->fh; + const struct nfs_writeverf *verf = data->res.verf; - __entry->status = data->res.op_status; + __entry->status = task->tk_status; __entry->offset = data->args.offset; - memcpy(&__entry->verifier, &data->verf.verifier, - sizeof(__entry->verifier)); + __entry->stable = verf->committed; + memcpy(__entry->verifier, + &verf->verifier, + NFS4_VERIFIER_SIZE); __entry->dev = inode->i_sb->s_dev; __entry->fileid = nfsi->fileid; - __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->fhandle = nfs_fhandle_hash(fh); ), TP_printk( "fileid=%02x:%02x:%llu fhandle=0x%08x " - "offset=%lld status=%d verifier 0x%016llx", + "offset=%lld status=%d stable=%s verifier=%s", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->fileid, __entry->fhandle, - __entry->offset, __entry->status, - __entry->verifier + (long long)__entry->offset, __entry->status, + nfs_show_stable(__entry->stable), + __print_hex_str(__entry->verifier, NFS4_VERIFIER_SIZE) ) ); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index d0f62aef3489..985ddff46051 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1831,7 +1831,7 @@ static void nfs_commit_done(struct rpc_task *task, void *calldata) /* Call the NFS version-specific code */ NFS_PROTO(data->inode)->commit_done(task, data); - trace_nfs_commit_done(data); + trace_nfs_commit_done(task, data); } static void nfs_commit_release_pages(struct nfs_commit_data *data) -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 12/15] pNFS/flexfiles: Add tracing for layout errors 2020-01-06 20:25 ` [PATCH 11/15] NFS: Clean up generic file commit tracepoint Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 2020-01-06 20:25 ` [PATCH 13/15] NFS: Improve tracing of permission calls Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs Trace layout errors for pNFS/flexfiles on read/write/commit operations. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- fs/nfs/flexfilelayout/flexfilelayout.c | 28 +++++-- fs/nfs/nfs4trace.c | 4 + fs/nfs/nfs4trace.h | 109 +++++++++++++++++++++++++ 3 files changed, 132 insertions(+), 9 deletions(-) diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c index 3163b78b1d2c..bb9148b83166 100644 --- a/fs/nfs/flexfilelayout/flexfilelayout.c +++ b/fs/nfs/flexfilelayout/flexfilelayout.c @@ -1266,9 +1266,10 @@ static int ff_layout_async_handle_error(struct rpc_task *task, static void ff_layout_io_track_ds_error(struct pnfs_layout_segment *lseg, int idx, u64 offset, u64 length, - u32 status, int opnum, int error) + u32 *op_status, int opnum, int error) { struct nfs4_ff_layout_mirror *mirror; + u32 status = *op_status; int err; if (status == 0) { @@ -1286,10 +1287,10 @@ static void ff_layout_io_track_ds_error(struct pnfs_layout_segment *lseg, case -ENOBUFS: case -EPIPE: case -EPERM: - status = NFS4ERR_NXIO; + *op_status = status = NFS4ERR_NXIO; break; case -EACCES: - status = NFS4ERR_ACCESS; + *op_status = status = NFS4ERR_ACCESS; break; default: return; @@ -1321,11 +1322,14 @@ static int ff_layout_read_done_cb(struct rpc_task *task, int new_idx = hdr->pgio_mirror_idx; int err; - if (task->tk_status < 0) + if (task->tk_status < 0) { ff_layout_io_track_ds_error(hdr->lseg, hdr->pgio_mirror_idx, hdr->args.offset, hdr->args.count, - hdr->res.op_status, OP_READ, + &hdr->res.op_status, OP_READ, task->tk_status); + trace_ff_layout_read_error(hdr); + } + err = ff_layout_async_handle_error(task, hdr->args.context->state, hdr->ds_clp, hdr->lseg, hdr->pgio_mirror_idx); @@ -1494,11 +1498,14 @@ static int ff_layout_write_done_cb(struct rpc_task *task, loff_t end_offs = 0; int err; - if (task->tk_status < 0) + if (task->tk_status < 0) { ff_layout_io_track_ds_error(hdr->lseg, hdr->pgio_mirror_idx, hdr->args.offset, hdr->args.count, - hdr->res.op_status, OP_WRITE, + &hdr->res.op_status, OP_WRITE, task->tk_status); + trace_ff_layout_write_error(hdr); + } + err = ff_layout_async_handle_error(task, hdr->args.context->state, hdr->ds_clp, hdr->lseg, hdr->pgio_mirror_idx); @@ -1537,11 +1544,14 @@ static int ff_layout_commit_done_cb(struct rpc_task *task, { int err; - if (task->tk_status < 0) + if (task->tk_status < 0) { ff_layout_io_track_ds_error(data->lseg, data->ds_commit_index, data->args.offset, data->args.count, - data->res.op_status, OP_COMMIT, + &data->res.op_status, OP_COMMIT, task->tk_status); + trace_ff_layout_commit_error(data); + } + err = ff_layout_async_handle_error(task, NULL, data->ds_clp, data->lseg, data->ds_commit_index); diff --git a/fs/nfs/nfs4trace.c b/fs/nfs/nfs4trace.c index 1a8f376b3f73..d9ac556bebcf 100644 --- a/fs/nfs/nfs4trace.c +++ b/fs/nfs/nfs4trace.c @@ -24,4 +24,8 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_read_done); EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_write_done); EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_read_pagelist); EXPORT_TRACEPOINT_SYMBOL_GPL(pnfs_mds_fallback_write_pagelist); + +EXPORT_TRACEPOINT_SYMBOL_GPL(ff_layout_read_error); +EXPORT_TRACEPOINT_SYMBOL_GPL(ff_layout_write_error); +EXPORT_TRACEPOINT_SYMBOL_GPL(ff_layout_commit_error); #endif diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index a467f49989f2..d39897daa284 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -2117,6 +2117,115 @@ DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_write_done); DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_read_pagelist); DEFINE_PNFS_LAYOUT_EVENT(pnfs_mds_fallback_write_pagelist); +DECLARE_EVENT_CLASS(nfs4_flexfiles_io_event, + TP_PROTO( + const struct nfs_pgio_header *hdr + ), + + TP_ARGS(hdr), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(u32, count) + __field(int, stateid_seq) + __field(u32, stateid_hash) + __string(dstaddr, hdr->ds_clp ? + rpc_peeraddr2str(hdr->ds_clp->cl_rpcclient, + RPC_DISPLAY_ADDR) : "unknown") + ), + + TP_fast_assign( + const struct inode *inode = hdr->inode; + + __entry->error = hdr->res.op_status; + __entry->fhandle = nfs_fhandle_hash(hdr->args.fh); + __entry->fileid = NFS_FILEID(inode); + __entry->dev = inode->i_sb->s_dev; + __entry->offset = hdr->args.offset; + __entry->count = hdr->args.count; + __entry->stateid_seq = + be32_to_cpu(hdr->args.stateid.seqid); + __entry->stateid_hash = + nfs_stateid_hash(&hdr->args.stateid); + __assign_str(dstaddr, hdr->ds_clp ? + rpc_peeraddr2str(hdr->ds_clp->cl_rpcclient, + RPC_DISPLAY_ADDR) : "unknown"); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%llu count=%u stateid=%d:0x%08x dstaddr=%s", + -__entry->error, + show_nfsv4_errors(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->offset, __entry->count, + __entry->stateid_seq, __entry->stateid_hash, + __get_str(dstaddr) + ) +); + +#define DEFINE_NFS4_FLEXFILES_IO_EVENT(name) \ + DEFINE_EVENT(nfs4_flexfiles_io_event, name, \ + TP_PROTO( \ + const struct nfs_pgio_header *hdr \ + ), \ + TP_ARGS(hdr)) +DEFINE_NFS4_FLEXFILES_IO_EVENT(ff_layout_read_error); +DEFINE_NFS4_FLEXFILES_IO_EVENT(ff_layout_write_error); + +TRACE_EVENT(ff_layout_commit_error, + TP_PROTO( + const struct nfs_commit_data *data + ), + + TP_ARGS(data), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(dev_t, dev) + __field(u32, fhandle) + __field(u64, fileid) + __field(loff_t, offset) + __field(u32, count) + __string(dstaddr, data->ds_clp ? + rpc_peeraddr2str(data->ds_clp->cl_rpcclient, + RPC_DISPLAY_ADDR) : "unknown") + ), + + TP_fast_assign( + const struct inode *inode = data->inode; + + __entry->error = data->res.op_status; + __entry->fhandle = nfs_fhandle_hash(data->args.fh); + __entry->fileid = NFS_FILEID(inode); + __entry->dev = inode->i_sb->s_dev; + __entry->offset = data->args.offset; + __entry->count = data->args.count; + __assign_str(dstaddr, data->ds_clp ? + rpc_peeraddr2str(data->ds_clp->cl_rpcclient, + RPC_DISPLAY_ADDR) : "unknown"); + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "offset=%llu count=%u dstaddr=%s", + -__entry->error, + show_nfsv4_errors(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->offset, __entry->count, + __get_str(dstaddr) + ) +); + + #endif /* CONFIG_NFS_V4_1 */ #endif /* _TRACE_NFS4_H */ -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 13/15] NFS: Improve tracing of permission calls 2020-01-06 20:25 ` [PATCH 12/15] pNFS/flexfiles: Add tracing for layout errors Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 2020-01-06 20:25 ` [PATCH 14/15] NFS: When resending after a short write, reset the reply count to zero Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs On exit from nfs_do_access(), record the mask representing the requested permissions, as well as the server-supplied set of access rights for this user. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- fs/nfs/dir.c | 4 ++-- fs/nfs/nfstrace.h | 61 ++++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 62 insertions(+), 3 deletions(-) diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index e180033e35cf..372c16b3042c 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -2476,7 +2476,7 @@ static int nfs_do_access(struct inode *inode, const struct cred *cred, int mask) { struct nfs_access_entry cache; bool may_block = (mask & MAY_NOT_BLOCK) == 0; - int cache_mask; + int cache_mask = -1; int status; trace_nfs_access_enter(inode); @@ -2515,7 +2515,7 @@ static int nfs_do_access(struct inode *inode, const struct cred *cred, int mask) if ((mask & ~cache_mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) != 0) status = -EACCES; out: - trace_nfs_access_exit(inode, status); + trace_nfs_access_exit(inode, mask, cache_mask, status); return status; } diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index 51043f02e86f..b43f1b2501db 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -198,7 +198,66 @@ DEFINE_NFS_INODE_EVENT_DONE(nfs_writeback_inode_exit); DEFINE_NFS_INODE_EVENT(nfs_fsync_enter); DEFINE_NFS_INODE_EVENT_DONE(nfs_fsync_exit); DEFINE_NFS_INODE_EVENT(nfs_access_enter); -DEFINE_NFS_INODE_EVENT_DONE(nfs_access_exit); + +TRACE_EVENT(nfs_access_exit, + TP_PROTO( + const struct inode *inode, + unsigned int mask, + unsigned int permitted, + int error + ), + + TP_ARGS(inode, mask, permitted, error), + + TP_STRUCT__entry( + __field(unsigned long, error) + __field(dev_t, dev) + __field(u32, fhandle) + __field(unsigned char, type) + __field(u64, fileid) + __field(u64, version) + __field(loff_t, size) + __field(unsigned long, nfsi_flags) + __field(unsigned long, cache_validity) + __field(unsigned int, mask) + __field(unsigned int, permitted) + ), + + TP_fast_assign( + const struct nfs_inode *nfsi = NFS_I(inode); + __entry->error = error < 0 ? -error : 0; + __entry->dev = inode->i_sb->s_dev; + __entry->fileid = nfsi->fileid; + __entry->fhandle = nfs_fhandle_hash(&nfsi->fh); + __entry->type = nfs_umode_to_dtype(inode->i_mode); + __entry->version = inode_peek_iversion_raw(inode); + __entry->size = i_size_read(inode); + __entry->nfsi_flags = nfsi->flags; + __entry->cache_validity = nfsi->cache_validity; + __entry->mask = mask; + __entry->permitted = permitted; + ), + + TP_printk( + "error=%ld (%s) fileid=%02x:%02x:%llu fhandle=0x%08x " + "type=%u (%s) version=%llu size=%lld " + "cache_validity=0x%lx (%s) nfs_flags=0x%lx (%s) " + "mask=0x%x permitted=0x%x", + -__entry->error, nfs_show_status(__entry->error), + MAJOR(__entry->dev), MINOR(__entry->dev), + (unsigned long long)__entry->fileid, + __entry->fhandle, + __entry->type, + nfs_show_file_type(__entry->type), + (unsigned long long)__entry->version, + (long long)__entry->size, + __entry->cache_validity, + nfs_show_cache_validity(__entry->cache_validity), + __entry->nfsi_flags, + nfs_show_nfsi_flags(__entry->nfsi_flags), + __entry->mask, __entry->permitted + ) +); TRACE_DEFINE_ENUM(LOOKUP_FOLLOW); TRACE_DEFINE_ENUM(LOOKUP_DIRECTORY); -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 14/15] NFS: When resending after a short write, reset the reply count to zero 2020-01-06 20:25 ` [PATCH 13/15] NFS: Improve tracing of permission calls Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 2020-01-06 20:25 ` [PATCH 15/15] NFS: Fix nfs_direct_write_reschedule_io() Trond Myklebust 0 siblings, 1 reply; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs If we're resending a write due to a short read or write, ensure we reset the reply count to zero. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- fs/nfs/read.c | 2 ++ fs/nfs/write.c | 2 ++ 2 files changed, 4 insertions(+) diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 12deb3bdb2a0..34bb9add2302 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -281,6 +281,8 @@ static void nfs_readpage_retry(struct rpc_task *task, argp->offset += resp->count; argp->pgbase += resp->count; argp->count -= resp->count; + resp->count = 0; + resp->eof = 0; rpc_restart_call_prepare(task); } diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 985ddff46051..2c2020546e24 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1656,6 +1656,8 @@ static void nfs_writeback_result(struct rpc_task *task, */ argp->stable = NFS_FILE_SYNC; } + resp->count = 0; + resp->verf->committed = 0; rpc_restart_call_prepare(task); } } -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
* [PATCH 15/15] NFS: Fix nfs_direct_write_reschedule_io() 2020-01-06 20:25 ` [PATCH 14/15] NFS: When resending after a short write, reset the reply count to zero Trond Myklebust @ 2020-01-06 20:25 ` Trond Myklebust 0 siblings, 0 replies; 16+ messages in thread From: Trond Myklebust @ 2020-01-06 20:25 UTC (permalink / raw) To: Anna Schumaker; +Cc: linux-nfs The 'hdr->good_bytes' is defined as the number of bytes we expect to read or write starting at offset hdr->io_start. In the case of a partial read/write we may end up adjusting hdr->args.offset and hdr->args.count to skip I/O for data that was already read/written, and so we must ensure the calculation takes that into account. Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com> --- fs/nfs/direct.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 29f00da8a0b7..b768a0b42e82 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -824,7 +824,8 @@ static void nfs_direct_write_reschedule_io(struct nfs_pgio_header *hdr) dreq->flags = NFS_ODIRECT_RESCHED_WRITES; /* fake unstable write to let common nfs resend pages */ hdr->verf.committed = NFS_UNSTABLE; - hdr->good_bytes = hdr->args.count; + hdr->good_bytes = hdr->args.offset + hdr->args.count - + hdr->io_start; } spin_unlock(&dreq->lock); } -- 2.24.1 ^ permalink raw reply related [flat|nested] 16+ messages in thread
end of thread, other threads:[~2020-01-06 20:27 UTC | newest] Thread overview: 16+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2020-01-06 20:24 [PATCH 00/15] Various writeback related fixes Trond Myklebust 2020-01-06 20:25 ` [PATCH 01/15] NFS: Revalidate the file size on a fatal write error Trond Myklebust 2020-01-06 20:25 ` [PATCH 02/15] NFS: Revalidate the file mapping on all fatal writeback errors Trond Myklebust 2020-01-06 20:25 ` [PATCH 03/15] SUNRPC: Remove broken gss_mech_list_pseudoflavors() Trond Myklebust 2020-01-06 20:25 ` [PATCH 04/15] NFS: Fix up fsync() when the server rebooted Trond Myklebust 2020-01-06 20:25 ` [PATCH 05/15] NFS/pnfs: Fix pnfs_generic_prepare_to_resend_writes() Trond Myklebust 2020-01-06 20:25 ` [PATCH 06/15] NFSv4: Improve read/write/commit tracing Trond Myklebust 2020-01-06 20:25 ` [PATCH 07/15] NFS: Fix fix of show_nfs_errors Trond Myklebust 2020-01-06 20:25 ` [PATCH 08/15] pNFS/flexfiles: Record resend attempts on I/O failure Trond Myklebust 2020-01-06 20:25 ` [PATCH 09/15] NFS: Clean up generic file read tracepoints Trond Myklebust 2020-01-06 20:25 ` [PATCH 10/15] NFS: Clean up generic writeback tracepoints Trond Myklebust 2020-01-06 20:25 ` [PATCH 11/15] NFS: Clean up generic file commit tracepoint Trond Myklebust 2020-01-06 20:25 ` [PATCH 12/15] pNFS/flexfiles: Add tracing for layout errors Trond Myklebust 2020-01-06 20:25 ` [PATCH 13/15] NFS: Improve tracing of permission calls Trond Myklebust 2020-01-06 20:25 ` [PATCH 14/15] NFS: When resending after a short write, reset the reply count to zero Trond Myklebust 2020-01-06 20:25 ` [PATCH 15/15] NFS: Fix nfs_direct_write_reschedule_io() Trond Myklebust
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).