I guess I really have two questions: 1. Does switching from unbuffered output to buffered output affect the correctness of any operations that we're performing?
2. The pkgsubprocess module used pspawn to reduce the as_dup overhead when forking in low-memory situations. I'm assuming that in most cases it's not a problem to allocate 128k. However, I would be curious if this is likely to cause trouble if we're performing an install in tight memory conditions. Do we actually care about the output that gets returned from the driver actions or the indexer? I had thought we were simply checking the return code in these cases? -j On Tue, Dec 09, 2008 at 02:48:23PM -0800, Danek Duvall wrote: > On Tue, Dec 09, 2008 at 02:41:42PM -0800, [EMAIL PROTECTED] wrote: > > > but does it have any adverse effects elsewhere? > > I'm not sure how to test for that. I can do an install of slim_install and > see that nothing goes awry, if that would assuage your doubts. > > > Do we really need a 128k buffer for every subprocess that we invoke? > > Certainly not. 8k should be sufficient for almost everything else, but > Bart suggested that cranking it up across the board should be maximally > performant. > > Danek > _______________________________________________ > pkg-discuss mailing list > [email protected] > http://mail.opensolaris.org/mailman/listinfo/pkg-discuss _______________________________________________ pkg-discuss mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
