[patch|rfc][0/3] fix aio+dio+O_SYNC writes

Jeff Moyer jmoyer at redhat.com
Fri Jan 27 15:15:46 CST 2012


Currently, if an AIO/DIO/O_SYNC write is issued, generic_write_sync is
called after the submission of the I/O instead of after the completion.
This can result in flushing the disk cache before the actual write even
gets there!  This patch set attempts to fix that for xfs and ext4.
Comments would be greatly appreciated.

Cheers,
Jeff

[PATCH 1/3] xfs: honor the O_SYNC flag for aysnchronous direct I/O requests
[PATCH 2/3] ext4: honor the O_SYNC flag for aysnchronous direct I/O requests
[PATCH 3/3] filemap: don't call generic_write_sync for -EIOCBQUEUED




More information about the xfs mailing list