diff mbox series

[azure,1/3] ext4: fix fault handling when mounted with -o dax, ro

Message ID 1505511880-19182-2-git-send-email-marcelo.cerri@canonical.com
State New
Headers show
Series linux-azure: persistent memory is not working | expand

Commit Message

Marcelo Henrique Cerri Sept. 15, 2017, 9:44 p.m. UTC 
From: Randy Dodgen <dodgen@google.com>

BugLink: http://bugs.launchpad.net/bugs/1715755

If an ext4 filesystem is mounted with both the DAX and read-only
options, executables on that filesystem will fail to start (claiming
'Segmentation fault') due to the fault handler returning
VM_FAULT_SIGBUS.

This is due to the DAX fault handler (see ext4_dax_huge_fault)
attempting to write to the journal when FAULT_FLAG_WRITE is set. This is
the wrong behavior for write faults which will lead to a COW page; in
particular, this fails for readonly mounts.

This change avoids journal writes for faults that are expected to COW.

It might be the case that this could be better handled in
ext4_iomap_begin / ext4_iomap_end (called via iomap_ops inside
dax_iomap_fault). These is some overlap already (e.g. grabbing journal
handles).

Signed-off-by: Randy Dodgen <dodgen@google.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ross Zwisler <ross.zwisler@linux.intel.com>
(cherry picked from commit fd96b8da68d32a9403726db09b229f4b5ac849c7)
Signed-off-by: Marcelo Henrique Cerri <marcelo.cerri@canonical.com>
---
 fs/ext4/file.c | 15 ++++++++++++++-
 1 file changed, 14 insertions(+), 1 deletion(-)
diff mbox series

Patch

diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index bbea2dccd584..2ac020b3cbf8 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -260,7 +260,20 @@  static int ext4_dax_huge_fault(struct vm_fault *vmf,
 	handle_t *handle = NULL;
 	struct inode *inode = file_inode(vmf->vma->vm_file);
 	struct super_block *sb = inode->i_sb;
-	bool write = vmf->flags & FAULT_FLAG_WRITE;
+
+	/*
+	 * We have to distinguish real writes from writes which will result in a
+	 * COW page; COW writes should *not* poke the journal (the file will not
+	 * be changed). Doing so would cause unintended failures when mounted
+	 * read-only.
+	 *
+	 * We check for VM_SHARED rather than vmf->cow_page since the latter is
+	 * unset for pe_size != PE_SIZE_PTE (i.e. only in do_cow_fault); for
+	 * other sizes, dax_iomap_fault will handle splitting / fallback so that
+	 * we eventually come back with a COW page.
+	 */
+	bool write = (vmf->flags & FAULT_FLAG_WRITE) &&
+		(vmf->vma->vm_flags & VM_SHARED);
 
 	if (write) {
 		sb_start_pagefault(sb);