Messages in this thread | | | Date | Fri, 08 Feb 2002 01:57:34 -0800 | From | Andrew Morton <> | Subject | Re: [patch] get_request starvation fix |
| |
Jens Axboe wrote: > > On Fri, Feb 08 2002, Andrew Morton wrote: > > Here's a patch which addresses the get_request starvation problem. > > [snip] > > Agrh, if only I knew you were working on this too :/. Oh well, from a > first-read the patch looks good.
Seems that with FIFO fairness, /bin/sync now also livelocks. And it's pretty easy to see why. There's nothing to make write_unlocked_buffers() terminate.
We'll worry about that tomorrow. I may just make it give up after writing (2 * nr_buffers_type[BUF_DIRTY]) buffers.
The patch works well with read-latency2 (and it didn't throw rejects). Smooth and fast. It's going to take some time and testing to settle everything in.
- - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |