Re: [PATCH][5/?] count writeback pages in nr_scanned

[Andrew Morton]

Andrea Arcangeli <andrea@xxxxxxx> wrote:
>
> But I dislike code that works by luck,

yup, it is code which works by luck.  And it's a bit inaccurate - there is
potential for (small) further latency improvement in there.

But we do need to watch the CPU consumption in there - end_page_writeback()
already figures fairly high sometimes.

> and sure I could demonstrate it
>  if I bothered to write an artificial testcase on simulated hardware.

Could well be so.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/