This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC 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]

Re: [PATCH, i386]: Do not emit "cld" instructions

On 12/5/06, Jan Hubicka <hubicka@ucw.cz> wrote:

This is pretty high.  Would be possible for you to rerun the
test_stringops script on P4 machine after removing the CLD?  If it
really is 48 cycles, it should show difference in the preffered memcpy
codegen.


Yes, I got different results (attached) on p4 3.2 GHz. This was
measured on otherwise idle machine, and it is somehow faster than my
previous results.

I suspect that timings are faster due to the fact that this is HT
machine and some other compute intesive task was running in parallel
during the benchmark. But in any case, relative differences should be
the same, but they are not.

Uros.

Attachment: stringop.res_1
Description: Binary data

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