Messages in this thread | | | Date | Thu, 26 Aug 2004 00:40:30 +0100 | From | Paulo Marques <> | Subject | Re: [PATCH] kallsyms data size reduction / lookup speedup |
| |
viro@parcelfarce.linux.theplanet.co.uk wrote: > On Wed, Aug 25, 2004 at 08:09:33PM +0100, Paulo Marques wrote: > >>Matt Mackall wrote: >> >>>... >>> >>>FYI, killing the seq_file stuff will likely prove unpopular. So you'll >>>want to do that in a separate patch. If it doesn't affect the way >>>you're handling compression, please repost your compression patch. I >>>have a few comments, but otherwise I think we should move forward with it. >> >>I'm still not sure that the seq_file is the culprit, but doing >>a 10000 symbol decompression in a user space application takes >>about 340us, whereas doing a "time cat /proc/kallsyms > /dev/null" >>gives approx. 0.2s! (this is all on a Pentium4 2.8GHz) >> >>*If* the seq_file is the culprit, then I don't think removing >>it (or improving it) will be unpopular. > > > If it really spends that much in seq_file, I bet anything that it got > *very* dumb iterator. Which should be fixable...
That is why I kept a big *If* in that sentence. I'm quite new to all this, and I'm still reading a lot of source code.
If the culprit is in fact seq_file, and seq_file can be improved in a way that works for everyone (not only kallsyms), then I also agree that is is the way to go. But hunting this down might prove that the problem is somewhere else. It is just too soon to draw conclusions.
-- Paulo Marques - www.grupopie.com - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |