This is the mail archive of the
mailing list for the GCC project.
Re: [PATCH] Speed up for_each_rtx
- From: Paolo Bonzini <paolo dot bonzini at lu dot unisi dot ch>
- To: Steven Bosscher <stevenb at suse dot de>, GCC Patches <gcc-patches at gcc dot gnu dot org>
- Date: Thu, 20 Jan 2005 15:53:21 +0100
- Subject: Re: [PATCH] Speed up for_each_rtx
- References: <2280960.1106228233413.SLOX.WebMail.firstname.lastname@example.org>
But this is also interesting:
(I'll point out once more that a long time in bootstrap is spent
in ld -- about 8 minutes).
How did you measure this?
This is just a conservative approximation, I used a stopwatch and
measured the time spent in the final link for libstdc++ and libjava.
Also consider several minutes pass before stage2 starts running, and
that compiling libjava to libgcj.jar does not use for_each_rtx. A
better measurement of the speedup would be "make bootstrap2 && time make
bootstrap", or something akin to that, but still it would not account