> Index: fs/buffer.c
> ===================================================================
> RCS file: /var/cvs/linux/fs/buffer.c,v
> retrieving revision 1.1.2.88
> diff -u -r1.1.2.88 buffer.c
> --- buffer.c 1999/04/29 18:54:03 1.1.2.88
> +++ buffer.c 1999/04/30 02:24:32
> @@ -764,7 +764,8 @@
> */
> if (too_many_dirty_buffers())
> {
> - flush_dirty_buffers();
> + if (nr_free_pages <= freepages.high)
> + flush_dirty_buffers();
> if (too_many_dirty_buffers())
> wakeup_bdflush();
> }
>
>
> I am not completly sure though. I removed the code above after
> 2.2.6_andrea2 in order to have a saner VM but it wasn't a bug at
> all. So now I reinserted it and I am waiting a feedback from Harvey
> who is doing a great work (as you all) in testing my code.
Sorry, but it doesn't make it to the legendary 2.2.6andrea2
performance. I called this one 2.2.7andrea2a. Here're the results
for 2.2.6andrea2, 2.2.7andrea2 & 2.2.7andrea2a (50,000 records):
Test 1 (noninteractive) Test 2 (interactive)
user sys elapsed worst 2nd user sys elapsed worst 2nd Feel
2.2.6a2** 28.04 7.24 0:42.24 2.82 2.10 29.89 7.33 1:30.70 6.09 4.34 A
28.44 7.06 0:42.37 2.82 2.10 30.45 7.73 1:25.50 5.09 4.27 A
28.02 6.96 0:41.94 2.79 2.09 30.19 8.10 1:20.83 4.57 3.22 A
27.88 7.23 0:42.39 2.80 2.09* 29.50 8.38 2:36.11 28.08 11.74 A
32.26 8.00 1:06.17 3.10 2.87 A
33.31 7.97 1:10.71 3.22 2.69 A
2.2.7a2 28.25 6.25 0:41.27 2.64 2.00 29.24 7.03 3:26.01 47.64 18.21 B+
28.82 5.75 0:41.00 2.66 1.98
28.40 6.19 0:41.01 2.68 1.99
2.2.7a2a 28.39 6.13 0:41.23 2.67 1.98 31.03 6.91 3:58.71 52.15 25.37 A
28.76 5.74 0:40.89 2.67 1.97 30.71 6.45 2:01.42 12.16 11.98 A
28.32 5.61 0:40.28 2.67 1.95 31.59 7.30 2:14.57 8.48 6.65 A
I just did 4 more 2.2.6a2 tests just to make sure. I managed to get
it up to 2:36 elapsed time when bringing up 4 netscape windows, but
then ran the test again & it made its best time.
-- Harvey J. Stein BFM Financial Research hjstein@bfr.co.il- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/