[lfs-support] Final GCC Compile Weirdness - 1 Time? No, 2 nd Time? Perfect!

Firerat firer4t at googlemail.com
Tue Jan 10 04:33:58 PST 2012


On Tue, Jan 10, 2012 at 10:50 AM, Andrew Benton <b3nton at gmail.com> wrote:
> On Tue, 10 Jan 2012 09:21:01 +0000
> Firerat <firer4t at googlemail.com> wrote:
>
>> I find that if a make fails first time, but completes (or gets further ) on
>> second run it is due to some race condition where you have multiple jobs
>> and a new job thread is dependent on something that hasn't completed.
>> The simplest way around this is to flag the make with -j1
>
> FWIW I've never had a problem compiling gcc with make -j4
> Also, if you read his script he doesn't use multiple jobs when he runs:
> make -k check
>
true, however we have no idea if MAKEFLAGS has been set
actually, read the fist post, he is using MAKEFLAGS="-j8"

and I too have no issues compiling gcc with -j14 ( yes 14 )

in any case
-j1 removes the doubt from the equation, is it a make race condition or not?
also -j1 makes it much easier to read what went wrong, you are reading
just one job and not 4 , 5 or 6 jobs at the same time.

And, yes I'f I recall correctly one of his earlier posts last week
mentioned use of a VM ( windows hosting slackware )



More information about the lfs-support mailing list