This is the mail archive of the libstdc++@sources.redhat.com mailing list for the libstdc++ project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: slow V3 configures (was Re: ICE during bootstrap.)


On Tue, Dec 05, 2000 at 01:11:42PM -0600, Robert Lipe wrote:
> 
> If the cache hits were as quick as they are elsewhere in the GCC build
> process, they wouldn't stand out.

It's not the caching that's the problem; check my mail again.

The typical cached test, both in v3 and elsewhere, is basically

    test-for-cached-answer-oh-it's-there-never-mind-then

In the math tests, it's

    push-current-language-on-stack
    set-current-language-to-C++
    test-for-cached-answer-oh-it's-there-never-mind-then
    pop-language-from-stack

Those LANG_SAVE and LANG_RESTORE are what contribute the extra
three-quarters of a banana.  :-)


> Perhaps I'm oversensitized becuase I've been chasing problems in the
> v3 configure/build process and have thus had to sit through this
> many-minute process more times than I feel I've deserved to. :-)

You can imagine what it was like before we cached the answers.  Even a
caffeine addict like me was falling asleep.  (I find that a config.cache,
primed with just the math test results, and moved into the builddir/target
subdirectory between configuring and making, speeds up quick test builds
by a lot.  Every so often, of course, you want to do it fresh and get a
new config.cache to work with.)


-- 
pedwards at disaster dot jaj dot com  |  pme at sources dot redhat dot com
devphil at several other less interesting addresses in various dot domains
The gods do not protect fools.  Fools are protected by more capable fools.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]