This is the mail archive of the gdb-patches@sourceware.org mailing list for the GDB project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Other format: | [Raw text] |
On Fri, Jul 17, 2009 at 10:19 AM, Paul Pluzhnikov<ppluzhnikov@google.com> wrote: > NNN before after > > 256 0m1.042s 0m0.337s > 512 0m6.837s 0m0.887s > 1024 0m31.406s 0m4.594s Looks like I managed to confuse myself even more (too many variants of gen.pl :-( Above results are for a simplified test case (gen-20090720.pl, attached). Here are the timings I get using unpatched vs. patched GDB using the original gen.pl from http://sourceware.org/ml/gdb-patches/2009-07/msg00417/gen.pl NNN before after 256 0m03.544s 0m02.724s 512 0m15.693s 0m10.025s 1024 1m06.844s 0m39.742s So not such a drastic improvement for *this* test case. Most of the time for this test is actually spent in dwarf2_frame_find_fde, for which I'll send another patch shortly. Thanks, -- Paul Pluzhnikov
Attachment:
gen-20090720.pl
Description: Perl program
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |