Re: [PATCH v6 05/10] block: Add core atomic write support
John Garry <[email protected]>
| Newsgroups | gmane.linux.kernel.aio.general,gmane.linux.block,gmane.linux.kernel,gmane.linux.file-systems,gmane.linux.scsi,gmane.comp.file-systems.btrfs,gmane.linux.kernel.io-uring |
|---|---|
| Organization | Oracle Corporation |
| Message-ID | <[email protected]> |
On 11/04/2024 00:34, Luis Chamberlain wrote: >> On 3/26/24 06:38, John Garry wrote: >>> diff --git a/Documentation/ABI/stable/sysfs-block b/Documentation/ABI/stable/sysfs-block >>> index 1fe9a553c37b..4c775f4bdefe 100644 >>> --- a/Documentation/ABI/stable/sysfs-block >>> +++ b/Documentation/ABI/stable/sysfs-block >>> +What: /sys/block/<disk>/atomic_write_boundary_bytes >>> +Date: February 2024 >>> +Contact: Himanshu Madhani<[email protected]> >>> +Description: >>> + [RO] A device may need to internally split I/Os which >>> + straddle a given logical block address boundary. In that >>> + case a single atomic write operation will be processed as >>> + one of more sub-operations which each complete atomically. >> or > If*or* was meant, wouldn't it be better just to say one or more > operations may be processed as one atomically in this situation? "Or" was meant (thanks Randy). I think that we just need to say that the write operation will not complete atomically if it straddles the boundary. Whether the separate parts of the write operation which straddle the boundary complete atomically is undefined and irrelevant. Thanks, John -- To unsubscribe, send a message with 'unsubscribe linux-aio' in the body to [email protected]. For more info on Linux AIO, see: http://www.kvack.org/aio/ Don't email: <a href=mailto:"[email protected]">[email protected]</a>