From 63b0c8f28dbd8513774219b8802370a638668811 Mon Sep 17 00:00:00 2001 From: Mikhail Pershin Date: Wed, 15 Jul 2020 08:12:55 +0300 Subject: [PATCH] LU-13759 dom: lock cancel to drop pages Prevent stale pages after lock cancel by creating cl_page connection for read-on-open pages. Since VM pages are connected to cl_object they can be found and discarded by CLIO properly. Lustre-change: https://review.whamcloud.com/39401 Lustre-commit: e95eca236471cf23083ef281ef204a5920e4db9b Signed-off-by: Mikhail Pershin Change-Id: Iba8c87c934c442b4c0b45d7d3821ceede1a6e68f Reviewed-on: https://review.whamcloud.com/40302 Tested-by: Maloo Tested-by: jenkins Reviewed-by: Andreas Dilger Reviewed-by: Oleg Drokin --- lustre/llite/file.c | 68 ++++++++++++++++++++++++++++++++----------- lustre/llite/llite_internal.h | 3 +- lustre/llite/namei.c | 14 +++------ lustre/mdc/mdc_dev.c | 5 ++++ lustre/tests/sanity-dom.sh | 18 ++++++++++-- 5 files changed, 77 insertions(+), 31 deletions(-) diff --git a/lustre/llite/file.c b/lustre/llite/file.c index 69ce884..9344d26 100644 --- a/lustre/llite/file.c +++ b/lustre/llite/file.c @@ -407,9 +407,10 @@ static inline int ll_dom_readpage(void *data, struct page *page) return 0; } -void ll_dom_finish_open(struct inode *inode, struct ptlrpc_request *req, - struct lookup_intent *it) +void ll_dom_finish_open(struct inode *inode, struct ptlrpc_request *req) { + struct lu_env *env; + struct cl_io *io; struct ll_inode_info *lli = ll_i2info(inode); struct cl_object *obj = lli->lli_clob; struct address_space *mapping = inode->i_mapping; @@ -419,6 +420,8 @@ void ll_dom_finish_open(struct inode *inode, struct ptlrpc_request *req, char *data; unsigned long index, start; struct niobuf_local lnb; + __u16 refcheck; + int rc; ENTRY; @@ -452,6 +455,16 @@ void ll_dom_finish_open(struct inode *inode, struct ptlrpc_request *req, RETURN_EXIT; } + env = cl_env_get(&refcheck); + if (IS_ERR(env)) + RETURN_EXIT; + io = vvp_env_thread_io(env); + io->ci_obj = obj; + io->ci_ignore_layout = 1; + rc = cl_io_init(env, io, CIT_MISC, obj); + if (rc) + GOTO(out_io, rc); + CDEBUG(D_INFO, "Get data along with open at %llu len %i, size %llu\n", rnb->rnb_offset, rnb->rnb_len, body->mbo_dom_size); @@ -463,6 +476,8 @@ void ll_dom_finish_open(struct inode *inode, struct ptlrpc_request *req, LASSERT(lnb.lnb_file_offset % PAGE_SIZE == 0); lnb.lnb_page_offset = 0; do { + struct cl_page *page; + lnb.lnb_data = data + (index << PAGE_SHIFT); lnb.lnb_len = rnb->rnb_len - (index << PAGE_SHIFT); if (lnb.lnb_len > PAGE_SIZE) @@ -478,9 +493,33 @@ void ll_dom_finish_open(struct inode *inode, struct ptlrpc_request *req, PTR_ERR(vmpage)); break; } + lock_page(vmpage); + if (vmpage->mapping == NULL) { + unlock_page(vmpage); + put_page(vmpage); + /* page was truncated */ + break; + } + /* attach VM page to CL page cache */ + page = cl_page_find(env, obj, vmpage->index, vmpage, + CPT_CACHEABLE); + if (IS_ERR(page)) { + ClearPageUptodate(vmpage); + unlock_page(vmpage); + put_page(vmpage); + break; + } + cl_page_export(env, page, 1); + cl_page_put(env, page); + unlock_page(vmpage); put_page(vmpage); index++; } while (rnb->rnb_len > (index << PAGE_SHIFT)); + +out_io: + cl_io_fini(env, io); + cl_env_put(env, &refcheck); + EXIT; } @@ -562,27 +601,22 @@ retry: rc = ll_prep_inode(&de->d_inode, req, NULL, itp); if (!rc && itp->it_lock_mode) { - struct lustre_handle handle = {.cookie = itp->it_lock_handle}; - struct ldlm_lock *lock; - bool has_dom_bit = false; + __u64 bits = 0; /* If we got a lock back and it has a LOOKUP bit set, * make sure the dentry is marked as valid so we can find it. * We don't need to care about actual hashing since other bits * of kernel will deal with that later. */ - lock = ldlm_handle2lock(&handle); - if (lock) { - has_dom_bit = ldlm_has_dom(lock); - if (lock->l_policy_data.l_inodebits.bits & - MDS_INODELOCK_LOOKUP) - d_lustre_revalidate(de); - - LDLM_LOCK_PUT(lock); - } - ll_set_lock_data(sbi->ll_md_exp, de->d_inode, itp, NULL); - if (has_dom_bit) - ll_dom_finish_open(de->d_inode, req, itp); + ll_set_lock_data(sbi->ll_md_exp, de->d_inode, itp, &bits); + if (bits & MDS_INODELOCK_LOOKUP) + d_lustre_revalidate(de); + + /* if DoM bit returned along with LAYOUT bit then there + * can be read-on-open data returned. + */ + if (bits & MDS_INODELOCK_DOM && bits & MDS_INODELOCK_LAYOUT) + ll_dom_finish_open(de->d_inode, req); } out: diff --git a/lustre/llite/llite_internal.h b/lustre/llite/llite_internal.h index a254348..4bfa305 100644 --- a/lustre/llite/llite_internal.h +++ b/lustre/llite/llite_internal.h @@ -997,8 +997,7 @@ ssize_t ll_copy_user_md(const struct lov_user_md __user *md, struct lov_user_md **kbuf); void ll_open_cleanup(struct super_block *sb, struct ptlrpc_request *open_req); -void ll_dom_finish_open(struct inode *inode, struct ptlrpc_request *req, - struct lookup_intent *it); +void ll_dom_finish_open(struct inode *inode, struct ptlrpc_request *req); /* Compute expected user md size when passing in a md from user space */ static inline ssize_t ll_lov_user_md_size(const struct lov_user_md *lum) diff --git a/lustre/llite/namei.c b/lustre/llite/namei.c index 36a949a..d0cd5e9 100644 --- a/lustre/llite/namei.c +++ b/lustre/llite/namei.c @@ -192,13 +192,6 @@ static int ll_dom_lock_cancel(struct inode *inode, struct ldlm_lock *lock) int rc; ENTRY; - if (!lli->lli_clob) { - /* due to DoM read on open, there may exist pages for Lustre - * regular file even though cl_object is not set up yet. */ - truncate_inode_pages(inode->i_mapping, 0); - RETURN(0); - } - env = cl_env_get(&refcheck); if (IS_ERR(env)) RETURN(PTR_ERR(env)); @@ -641,10 +634,11 @@ static int ll_lookup_it_finish(struct ptlrpc_request *request, if (rc) RETURN(rc); - if (it->it_op & IT_OPEN) - ll_dom_finish_open(inode, request, it); - ll_set_lock_data(ll_i2sbi(parent)->ll_md_exp, inode, it, &bits); + /* OPEN can return data if lock has DoM+LAYOUT bits set */ + if (it->it_op & IT_OPEN && + bits & MDS_INODELOCK_DOM && bits & MDS_INODELOCK_LAYOUT) + ll_dom_finish_open(inode, request); /* We used to query real size from OSTs here, but actually * this is not needed. For stat() calls size would be updated diff --git a/lustre/mdc/mdc_dev.c b/lustre/mdc/mdc_dev.c index 726165c..54d8205 100644 --- a/lustre/mdc/mdc_dev.c +++ b/lustre/mdc/mdc_dev.c @@ -1405,6 +1405,11 @@ int mdc_object_prune(const struct lu_env *env, struct cl_object *obj) static int mdc_object_flush(const struct lu_env *env, struct cl_object *obj, struct ldlm_lock *lock) { + /* if lock cancel is initiated from llite then it is combined + * lock with DOM bit and it may have no l_ast_data initialized yet, + * so init it here with given osc_object. + */ + mdc_set_dom_lock_data(env, lock, cl2osc(obj)); RETURN(mdc_dlm_blocking_ast0(env, lock, LDLM_CB_CANCELING)); } diff --git a/lustre/tests/sanity-dom.sh b/lustre/tests/sanity-dom.sh index 3df0133..1befc20 100644 --- a/lustre/tests/sanity-dom.sh +++ b/lustre/tests/sanity-dom.sh @@ -170,17 +170,31 @@ test_6() { run_test 6 "Race two writes, check file size" test_7() { + dd if=/dev/zero of=$DIR1/$tfile bs=1000 count=1 + cancel_lru_locks + + $MULTIOP $DIR1/$tfile or1000c + dd if=/dev/urandom of=$DIR2/$tfile bs=1000 count=1 + local md5_1=$(md5sum $DIR/$tfile | awk '{ print $1 }') + local md5_2=$(md5sum $DIR2/$tfile | awk '{ print $1 }') + [[ $md5_1 == $md5_2 ]] || + error "Client reads stale page" +} +run_test 7 "Stale pages after read-on-open" + +test_8() { dd if=/dev/zero of=$DIR1/$tfile bs=1k count=1 || error "write 10k" $TRUNCATE $DIR1/$tfile 123456 || error "truncate up to 123456" $TRUNCATE $DIR2/$tfile 55555 || error "truncate down to 55555" $CHECKSTAT -t file -s 55555 $DIR1/$tfile || error "stat" $CHECKSTAT -t file -s 55555 $DIR2/$tfile || error "stat" - dd if=/dev/zero of=$DIR1/$tfile bs=10k count=1 seek=10 conv=notrunc || error "write 10k@1M" + dd if=/dev/zero of=$DIR1/$tfile bs=10k count=1 seek=10 conv=notrunc || + error "write 10k@1M" $CHECKSTAT -t file -s 112640 $DIR1/$tfile || error "stat" $CHECKSTAT -t file -s 112640 $DIR2/$tfile || error "stat" rm $DIR1/$tfile } -run_test 7 "truncate up, truncate down on another mount, write in between" +run_test 8 "truncate up, truncate down on another mount, write in between" test_fsx() { local file1=$DIR1/$tfile -- 1.8.3.1