On Sat, 2006-02-25 at 20:24 -0500, Chuck Martin wrote: > On Wed, Feb 08, 2006 at 11:20:38PM +0100, Florian Schmidt mista.tapas-at-gmx.net |LAU| wrote: > > > Any other suggestions? The original problem is that the system clock > > > is extremely slow, causing the time to be off and things like "sleep 1" > > > to take 10 to 30 times as long as they should. > > > > Hmm, > > > > try with or without high resolution timer support compiled in. Also if > > you can confirm it doesn't happen with vanilla, i'd suggest reporting > > this to Ingo. > > Nothing in the responses so far from this list has solved the problem. > I reported it to Ingo on Feb. 10, and received no reply. I've since > compiled and tested a number of patched kernels, and I've determined > that the problem began with patch-2.6.13-rt13, since the problem occurs > with patch-2.6.13-rt13, but not with patch-2.6.13-rt12. I then reported > this finding to Ingo on Feb. 22, and I still haven't received a reply. > > Is there anyone on this list familiar enough with the kernel and the > realtime patches to give me a clue as to where to start looking? I > can't find a changelog for the realtime patches anywhere, and looking > at a diff of the source trees for both patched kernels doesn't turn up > anything obvious. I know C, but am unfamiliar with the kernel code, > and trying to understand it enough to know where to begin looking will > probably take some time. Any help that might shorten the amount of > time I have to study the code would be much appreciated. Thanks. I didn't see a report of this on the kernel mailing list. A post to LKML with Ingo CC'ed is much more likely to get results than mailing him privately - I often send requests for tech support via private mail to /dev/null as I don't have the bandwidth to help people unless it's in a public forum where others can get the answer from Google next time. Lee
