Messages in this thread | | | Date | Fri, 17 Jul 1998 15:33:03 +0200 (CEST) | From | Andrea Arcangeli <> | Subject | Re: Linux-2.1.109.. preliminary code freeze. |
| |
On Thu, 16 Jul 1998, Linus Torvalds wrote:
>In addition, I've spent a large part of my day running with a 12MB >machine, and low-memory behaviour seems to be reasonable. People who have >been unhappy with low-memory behaviour should check out 109 and comment on >it - the heuristics are fairly different, and seem to be better.
andrea@dragon:~$ free total used free shared buffers cached Mem: 31216 30556 660 776 1044 27920 -/+ buffers/cache: 1592 29624 Swap: 52412 1696 50716 andrea@dragon:~$ free total used free shared buffers cached Mem: 31216 30480 736 584 1044 27888 -/+ buffers/cache: 1548 29668 Swap: 52412 1732 50680
Between these two `free' invocation are been swapped out only some kbyte of ram and this little swapout is enough to stall `free' for at least one second (against the istantaneous free while kswapd is not swapping out).
I am running cp file /dev/zero at the same time of free (that is pratically the same of open linux-kernel in pine at the end of the month, with the addin that pine alloc the read data in memory and so it cause again more swapout).
Stephen C. Tweedie has perfectly understood the problem though and I hope he could make the aging more clever.
I can see a big improvement in 109 with 32 Mbyte when I launch a lot of application at the same time and in general (I hope that it isn' t my immagination since I have not benchmarked ;-). _Sure_ 109 doesn' t perform worse than 108 ;-). So thanks Linus!
Andrea[s] Arcangeli
- 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.altern.org/andrebalsa/doc/lkml-faq.html
| |