Alan Hourihane
Fri, 23 Oct 2009 06:51:33 -0700
On Fri, 2009-10-23 at 14:39 +0100, Alan Hourihane wrote: > On Fri, 2009-10-23 at 15:30 +0200, Fabian Groffen wrote: > > On 23-10-2009 14:28:17 +0100, Alan Hourihane wrote: > > > On Fri, 2009-10-23 at 15:14 +0200, Fabian Groffen wrote: > > > > On 23-10-2009 11:00:14 +0200, Fabian Groffen wrote: > > > > > On 23-10-2009 10:15:11 +0200, Markus Duft wrote: > > > > > > it seems that the code using os.pipe() seems either broken or the > > > > > > interix implementation of that python code is broken... whichever > > > > > > it is > > > > > > - it's bad ... :( > > > > > > > > > > > > Could you (grobian), please add "Interix" to the existing white > > > > > > list for > > > > > > openpty platforms? > > > > > > > > > > geez... Interix really sucks. Anyway, if you add it to 701, does it > > > > > work then, or does it hang in that case? > > > > > > > > hmm that check in portage looks totally broken to me, and makes > > > > sparc/solaris hang too... > > > > > > Which area is this as I've noticed FreeMiNT's emerging process takes > > > much longer at the moment. > > > > usr/lib/portage/pym/portage/__init__.py > > > > look for openpty > > > > there is a bit of code to test whether this can work or not also > > involving python 3.1 > > O.k. I'll take a look.
This looks fine on FreeMiNT. I've still got two problems with the current portage though. 1. Constantly updating mtimedb on every invocation of emerge. 2. Very slow. Argh! Maybe I need to go back and find out which one it broke on. But I feel it's probably before 14487 which is the last in the tree. Alan.