Re: request review: branch "wingo"

[Andy Wingo]

Hi,

On Tue 31 Mar 2009 09:38, address@hidden (Ludovic Courtès) writes:

> Andy Wingo <address@hidden> writes:
>
>> So, I really intended to wait for review, but it's irritating having
>> `master' broken, so I went ahead and merged this in.
>
> You waited for 31 hours and I still don't know how it was "broken",
> which I find irritating as well.

    .scm.go:
        $(MKDIR_P) `dirname address@hidden
        $(top_builddir)/pre-inst-guile                                  \
          -l $(top_builddir)/libguile/stack-limit-calibration.scm       \
          $(top_srcdir)/scripts/compile -o "$@" "$<"

This will run the compile script, but since there is no entry point
(-e), you just load the compile script then exit. No compilation
happens.

>
>> I think the stack calibration stuff is correct,
>
> Again, this all boils down to an arbitrary choice: 1 MiB instead of
> 40 KiB.  Surely someday this won't be enough.

I can remove the 1 MiB limit -- perhaps that's the right thing to do,
then. Just use 80% of the rlimit.

> Besides, there's the thread about cross-compilation where we mention
> building the compiler with an already installed Guile that may have an
> inappropriate stack limit.

I don't think that is relevant. Since the Guile that is running would
choose a stack size appropriate for it, based on the host getrlimit,
there would be no problem.

>> but perhaps more jarring
>> in this commit is a move from ./pre-inst-guile to ./meta/guile, and
>> ./pre-inst-guile-env to ./meta/uninstalled-env. I describe the rationale
>> in 0b6d8fdc28ed8af56e93157179c305fef037e0a0.
>
> I think the rationale ("[...] I
> want to be able to build external packages against uninstalled Guile,
> [....]") is questionable.  Things are not
> meant to work this way (Libtool's `.la' and executable scripts, `.pc'
> files, etc.), so it looks quite hackish to me.

Linking against uninstalled libtool libraries works fine, as long as you
don't install. Pkg-config is designed for uninstalled operation, from
pkg-config(1):

  --uninstalled

         Normally if you request the package "foo" and the package
         "foo-uninstalled" exists, pkg-config will prefer the
         "-uninstalled" variant. This allows compilation/linking against
         uninstalled packages. If you specify the "--uninstalled"
         option, pkg-config will return successfully if any
         "-uninstalled" packages are being used, and return failure
         (false) otherwise. (The "PKG_CONFIG_DISABLE_UNINSTALLED"
         environment variable keeps pkg-config from implicitly choosing
         "-uninstalled" packages, so if that variable is set, they will
         only have been used if you pass a name like "foo-uninstalled"
         on the command line explicitly.)

> I'm probably biased because I've got used to installing Guile when I
> want to test apps against it (I have 1.8 and HEAD under a different
> prefix so I can test against both).

I have that too, but it adds a step to the debugging cycle. I don't
think there's any harm in supporting this additional mode of hacking,
which is only for hackers in any case.

Cheers,

Andy
-- 
http://wingolog.org/