2006/2/3, Nick Pilon <[EMAIL PROTECTED]>:
> On 2/2/06, Axel Liljencrantz <[EMAIL PROTECTED]> wrote:
> > Can't win them all, I guess. I've been thinking about what might be
> > causing this - and I have a theory. It might be that the code that
> > puts the child process into it's own process group is the culprit -
> > the child loops over sleep(0) until the parent gives it it's own
> > group. To test this theory, one simply has to turn of job control,
> > using 'status --job-control=none'. Could you test if this fixes your
> > problem?
>
> That does seem to fix the prompt problem.
Yes, but I'll have to figure out a way to actually actually do forking
without using sleep.
>
> But doing some more testing, I get all sorts of weird problems with
> this tarball. Things that worked fine connected to the machine over
> ssh now cause problems when running fish in Terminal. It turns out
> that fish_pager isn't fixed after all. I still get SIGBUS when I hit
> space in any fish_pager at 100%. And fish_pager goes nuts trying to
> list the contents of a directory - we're talking 65-80% CPU. This last
> problem seems to be new in this tarball - 1.20.1 has no such problem.
Is this always, or just when using 'status --job-control=none'.
Looking at the code, I can see that disabling job control should make
the pager go nuts. I've fixed that now.
>
> The version of fish from this tarball also goes nuts when it's
> executed as the first process in an empty terminal (IE, as the shell).
> Again, fish 1.20.1 and 1.20.0 didn't do this.
Nuts as in really slow startup, I take it?
>
> Not sure what's going on here.
Nor I.
>
> --
> -Nick Pilon
>
--
Axel
N�HY隊X���'���u���[�������
ަ�k��!�� �W�~�鮆�zk��C� [EMAIL PROTECTED],���� a{�
��,�H��4�m�����Z�� jY�w��ǥrg
