File 0042-LU-14713-llite-tighten-condition-for-fault-not-drop-.patch of Package lustre_2_12
From 81aec05103558f57adb10fd319847304cdf44aa7 Mon Sep 17 00:00:00 2001
From: Bobi Jam <bobijam@whamcloud.com>
Date: Fri, 3 Sep 2021 00:38:43 +0800
Subject: [PATCH] LU-14713 llite: tighten condition for fault not drop mmap_sem
As __lock_page_or_retry() indicates, filemap_fault() will return
VM_FAULT_RETRY without releasing mmap_sem iff flags contains
FAULT_FLAG_ALLOW_RETRY and FAULT_FLAG_RETRY_NOWAIT.
So ll_fault0() should pass in FAULT_FLAG_ALLOW_RETRY |
FAULT_FLAG_RETRY_NOWAIT in ll_filemap_fault() so that when it
returns VM_FAULT_RETRY, we can pass on trying normal fault
under DLM lock as mmap_sem is still being held.
While in Linux 5.1 (6b4c9f4469819) FAULT_FLAG_RETRY_NOWAIT is enough
to not drop mmap_sem when failed to lock the page.
Fixes: 87865e4ae9 ("LU-13182 llite: Avoid eternel retry loops with MAP_POPULATE")
Signed-off-by: Bobi Jam <bobijam@whamcloud.com>
Change-Id: I9420c587301722b597155558657577349a8141e4
Reviewed-on: https://review.whamcloud.com/44715
Reviewed-by: Patrick Farrell <pfarrell@whamcloud.com>
Tested-by: jenkins <devops@whamcloud.com>
Reviewed-by: Neil Brown <neilb@suse.de>
Tested-by: Maloo <maloo@whamcloud.com>
Reviewed-by: Oleg Drokin <green@whamcloud.com>
---
lustre/llite/llite_mmap.c | 10 +++++++++-
1 file changed, 9 insertions(+), 1 deletion(-)
--- a/lustre/llite/llite_mmap.c
+++ b/lustre/llite/llite_mmap.c
@@ -280,15 +280,23 @@ static vm_fault_t ll_fault0(struct vm_ar
if (ll_sbi_has_fast_read(ll_i2sbi(file_inode(vma->vm_file)))) {
/* do fast fault */
+ bool allow_retry = vmf->flags & FAULT_FLAG_ALLOW_RETRY;
bool has_retry = vmf->flags & FAULT_FLAG_RETRY_NOWAIT;
/* To avoid loops, instruct downstream to not drop mmap_sem */
- vmf->flags |= FAULT_FLAG_RETRY_NOWAIT;
+ /**
+ * only need FAULT_FLAG_ALLOW_RETRY prior to Linux 5.1
+ * (6b4c9f4469819), where FAULT_FLAG_RETRY_NOWAIT is enough
+ * to not drop mmap_sem when failed to lock the page.
+ */
+ vmf->flags |= FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_RETRY_NOWAIT;
ll_cl_add(vma->vm_file, env, NULL, LCC_MMAP);
fault_ret = ll_filemap_fault(vma, vmf);
ll_cl_remove(vma->vm_file, env);
if (!has_retry)
vmf->flags &= ~FAULT_FLAG_RETRY_NOWAIT;
+ if (!allow_retry)
+ vmf->flags &= ~FAULT_FLAG_ALLOW_RETRY;
/* - If there is no error, then the page was found in cache and
* uptodate;
