>>>>> "Izumi" == Izumi Tsutsui <tsutsui@ceres.dti.ne.jp> writes:

    >> Concerning the frozen system: I'm not able to enter ddb, no
    >> reaction to serial break signal (while 'ddb.fromconsole = 1').

    Izumi> Does the freeze also happen *without* my patch or not? If
    Izumi> my patch causes the freeze, I have to rethink what is
    Izumi> wrong.

I don't think your patch causes the freeze, because they happened
before, too! Before (== w/o your patch) data corruption was so
frequent, that I would blame the freezes on them, too, not explicitly
mentioning them ('find' in daily and security stuff was a typical
trigger for them). In my hope your patch solves everything I have
reported the freezes as a separate point, but they are nothing new to
me.

    >> So, it's much much better, but (as expected?) not perfect.

    Izumi> Yes, the patch doesn't handle all possible virtual aliases.
    Izumi> It's just a workaround, but I'll commit it as an interim
    Izumi> fix.

So, which version are you going to commit? The older, more cache
flushier one, or the latter?

Your cobalt's and mi's pci related patch checked in tobay... :-)

    >> First thanks for your effort, and anything else I can test?

    Izumi> If you have some R4000/R4400 with L2 machines, I'd like to
    Izumi> see how your test script works on such machines.

My qube2 is the only mips machine I own, sorry!

Markus.