=?GB2312?B?UmU6ILTwuLQ6ILTwuLQ6ILTwuLQ6ILTwuLQ6ILTwuLQ6IA==?= =?GB2312?B?W2Rpc3RjY10gUHJvYmxlbSBhYm91dCBEaQ==?= =?GB2312?B?c3RjYyBvbiBXaW5kb3dzIG5vdCBjeWd3?= =?GB2312?B?aW4gdG8gY29tcGlsZSBBUk0gcHJvamVjdA==?=

[Fergus Henderson]

2008/10/8 Feng Tan <feng.tan at revoview.com>

>  I have a question about the parallelism. How to judge this?
>
One way would be to replace "distcc" with "sleep 10; echo" in your Makefile,
and then run "make clean" (or equivalent) and then "make -j", and at the
same time as "make -j" is running, run "ps" from another terminal window to
see how many "sleep 10" processes are executing concurrently.

> By the way do you have tried to use pump mode for Distcc on Windows?
>
Craig Silverstein built distcc and ran its test suite ("make
maintainer-check") on Cygwin, which tests pump mode.
But that just tests that it's not broken, it doesn't test performance. I
don't know if anyone has tested performance of pump mode on Windows.
-- 
Fergus Henderson <fergus at google.com>
-------------- next part --------------
HTML attachment scrubbed and removed