"Darrick J. Wong" <djwong@xxxxxxxxxx> writes: > On Thu, Apr 03, 2025 at 11:52:27PM +0530, Ritesh Harjani (IBM) wrote: >> Let's document the use of these flags in iomap design doc where other >> flags are defined too - >> >> - IOMAP_F_BOUNDARY was added by XFS to prevent merging of ioends >> across RTG boundaries. >> - IOMAP_F_ATOMIC_BIO was added for supporting atomic I/O operations >> for filesystems to inform the iomap that it needs HW-offload based >> mechanism for torn-write protection >> >> Signed-off-by: Ritesh Harjani (IBM) <ritesh.list@xxxxxxxxx> >> --- >> Documentation/filesystems/iomap/design.rst | 10 ++++++++++ >> 1 file changed, 10 insertions(+) >> >> diff --git a/Documentation/filesystems/iomap/design.rst b/Documentation/filesystems/iomap/design.rst >> index e29651a42eec..b916e85bc930 100644 >> --- a/Documentation/filesystems/iomap/design.rst >> +++ b/Documentation/filesystems/iomap/design.rst >> @@ -243,6 +243,11 @@ The fields are as follows: >> regular file data. >> This is only useful for FIEMAP. >> >> + * **IOMAP_F_BOUNDARY**: This indicates that I/O and I/O completions >> + for this iomap must never be merged with the mapping before it. >> + Currently XFS uses this to prevent merging of ioends across RTG >> + (realtime group) boundaries. > > Hrm, ok. Based on hch's comment about not mentioning specific fs > behavior, I think I'll suggest something more like: > > IOMAP_F_BOUNDARY: This I/O and its completion must not be merged with > any other I/O or completion. Filesystems must use this when submitting > I/O to devices that cannot handle I/O crossing certain LBAs (e.g. ZNS > devices). This flag applies only to buffered I/O writeback; all other > functions ignore it. > Sure. >> * **IOMAP_F_PRIVATE**: Starting with this value, the upper bits can >> be set by the filesystem for its own purposes. >> >> @@ -250,6 +255,11 @@ The fields are as follows: >> block assigned to it yet and the file system will do that in the bio >> submission handler, splitting the I/O as needed. >> >> + * **IOMAP_F_ATOMIC_BIO**: Indicates that write I/O must be submitted >> + with the ``REQ_ATOMIC`` flag set in the bio. Filesystems need to set >> + this flag to inform iomap that the write I/O operation requires >> + torn-write protection based on HW-offload mechanism. > > They must also ensure that mapping updates upon the completion of the > I/O must be performed in a single metadata update. > Thanks Darrick, for the clarification and the review comments on this patch. Once the remaining tracing patch is reviewed, I will incorporate these comments in v2. -ritesh
