Today I com­mit­ted the updated linux aio stuff to p4. It’s a mod­ule now, so it can be loaded as needed. I also updated the p4 diff in the wiki, so any­one can down­load and test it (please do it!).

Major fea­tures in the p4 diff are futexes and TLS for amd64. So if you own an amd64 sys­tem please down­load the patch and give it a try. I’m inter­ested in reports for linux–fire­fox, skype, realplayer and acroread on amd64.

There’s one known bug with the mmap behav­ior on amd64 (even in cur­rent). I don’t know if this only affects the LTP tests or real world appli­ca­tions, but because there’s much time between the mmap com­mits and the report, I think it’s only a prob­lem for the regres­sion tests and not a prob­lem which shows up in typ­i­cal usage sce­nar­ios.

After the call for pub­lic test­ing we got some reports about LORs and a panic. This hap­pened after an exten­sive par­al­lel com­pil­ing ses­sion of linux stuff on a SMP sys­tem for sev­eral hours. Roman redid some of the lock­ing and the fixes are in the tree. Inter­me­di­ate patches already showed promis­ing results. We’re wait­ing for the test of the com­mit­ted solu­tion now.

Yes­ter­day we also got reports about seg­faults with linux Java 5 and 6. This is under inves­ti­ga­tion (any debug­ging help would be appre­ci­ated, Java is a large beast).

I also noticed that we don’t have the results of a LTP run of a native Linux sys­tem in the wiki. It would be very nice to have this, as it would allow us to see bro­ken test cases (Roman thinks the ope­nat tests are flawed) or at least we see where it doesn’t mat­ter much that we don’t PASS.

Apart from Kon­stan­tin (envy24* dri­ver author) I also got a p4 account for Yuriy (emu10kx dri­ver author). Let’s cross fin­gers for a lot of col­lab­o­ra­tion in //depot/projects/soundsys­tem/ now…