lkml.org 
[lkml]   [1998]   [Aug]   [18]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [REPORT] Adaptec and SMP problems on 2.0.35
I received at leat very good comment of possible errors and workarounds
for these problems. It seems a lot of people is suffering lockups when SMP
and aicxxx are together (I was very surprised and feel much better now
;-).

So, my question is: Doug (Ledford...), are you receiving help for
developing the aicxxx driver? How can we help you?

My guess is that Doug and others are much better than me for kernel
hacking, but if I have to spend time learning it, I will do it. So if
anyone there has some good pointers, please send them to me. At least this
is better that just sending reports...

Doug, you received some of the recommendations, what do you think? What
the difference between the 5.0.14 driver (2.0.34 I guess) and 5.0.19 that
changed so much the behaviour?

Can I solve the timeout problem in 5.1.0 by changing hardware
configuration. I read in the docs that reset problems are solved when the
BIOS is enabled. Is it valid just for the boot reset?

Thanks to everyone.

PS: A VERY STRANGE reduction in response time without SMP...
I wrote a server that receives queries and update from radiusd and tacacs+
and read7write into a Empress relational database. The server does some
measurements (see below) of the time spent in different operations. Its
normal procedure is:

1. Receive a packet from a socket/port
2. Easy parsing.
3. Fork
4. The child accesses to the database...
5. It answers to the radius or tacacs daemon.
6. It finishes and exit.
7. Parent listen to the socket and go to 1.
8. if SIGCHLD is received, do wait3() and update stats

One of the measurements is the "average CPU time" consumed. When SMP was
enabled in the server, this average converged to 0.0349 seconds/operation,
now, whith UP, the current average is 0.0288 and is coming smaller (it
started with 0.029x after a few hundred operations).

Can the global spinlock increased so much the CPU time?

------ THE STATS
DBSERVER(862): CPU load (0-1): 0.001450 Sec/ops: 0.0288 PID: 9595
Elapsed: 0 days 08:43:55 Ops: 1582 Childr: 1 Max: 2 Overlds: 0
** Parent Times
User: 00:00:00.00 Sys: 00:00:00.84 Total: 00:00:00.84
** Total Times
User: 00:00:26.06 Sys: 00:00:19.52 Total: 00:00:45.58
------------------

--
Ricardo Galli
Member of ACM - Association for Computing Machinery
University of Balearic Islands
ATLAS Internet Access Provider
mailto:gallir@atlas-iap.es
mailto:rgalli@acm.org
http://www.atlas-iap.es



-
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

\
 
 \ /
  Last update: 2005-03-22 13:44    [W:0.071 / U:1.712 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site