This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
Re: hppa build broken
- From: "John David Anglin" <dave at hiauly1 dot hia dot nrc dot ca>
- To: amodra at bigpond dot net dot au (Alan Modra)
- Cc: sje at cup dot hp dot com, binutils at sources dot redhat dot com, dave dot anglin at nrc-cnrc dot gc dot ca, fnf at specifixinc dot com
- Date: Tue, 14 Jun 2005 23:28:01 -0400 (EDT)
- Subject: Re: hppa build broken
> You might like to see whether increasing the buffer size at write.c:1146
> helps poor HPUX file I/O. 4k would probably be a better size.
Doesn't help:
4k:
real 16m13.524s
user 0m1.250s
sys 0m21.310s
64k:
real 16m18.884s
user 0m0.230s
sys 0m21.430s
vmstat indicates ~ 300 xfer/sec for the drive. At 4k per transfer,
that's about 1000s or 16m. Why linux would be so much faster isn't
clear although the drive and interface used is faster.
I'd also previously tried changing pa_block to output 8MB blocks
followed by the residual (i.e., using the old way but limiting
memory usage to 8MB). While the user and sys times went down, the
real time actually got worse. The old code suffers from the same
problem.
Dave
--
J. David Anglin dave.anglin@nrc-cnrc.gc.ca
National Research Council of Canada (613) 990-0752 (FAX: 952-6602)