Messages in this thread | | | Date | Tue, 29 Jul 2014 17:12:48 -0700 | From | Christoph Hellwig <> | Subject | Re: [Regression 3.16-rc kernel] -55% reduction in throughput for OLTP benchmark |
| |
On Tue, Jul 29, 2014 at 04:57:12PM -0700, Tim Chen wrote: > Christoph, we're glad to report that this patch does fix the problem. > Wonder if Al or someone can pick it up. It is a large regression > that should not be left unfixed for 3.16.
Al has been away for a bit, and I prepared the last VFS pull. I'll pick it up for now and either Al or I will send it on before next weekend.
Can I get a tested-by tag from you for the patch? Can anyone else please cross check it? The new total value might be different from the old size value when we do a short write, but I can't see how an async write would be a good thing if we had to finish the remainder using buffered I/O if the old code really allowed that.
| |