PowerPC 64 x 32 bits performance

2005-03-04 Thread Edmar Wienskoski
I checked with several gcc versions. Some of the spec2k benchmarks has a considerable performance loss when the benchmark is compiled with a 64 bits tool. Ex: 175.vpr using: Reading specs from /usr/lib/gcc-lib/ppc64-yellowdog-linux/3.3.3/specs Configured with: ../configure --prefix=/usr --mandir=/u

[RFC] Unused variable in profile.c

2009-03-10 Thread Edmar Wienskoski
I accidentally found that the local variable num_never_executed in function compute_branch_probabilities, is initialized to zero and never gets modified after that. I suppose the statement in line 603: num_branches++, num_never_executed; was intended to be: num_branches++, num_never_executed++; R

Re: Vector types and type conversions

2006-04-19 Thread Edmar Wienskoski
There is the Spe PEM as well (Freescale 8540/8548). On the Spe we defined a vector type that has no definition of vector elements. (__ev64_opaque__). It was designed such that casting from this type to any other vector type of the same size can only be accomplished by bit-preserving conversion

powerpc targets, long double implementation, and c++ programs

2006-09-08 Thread Edmar Wienskoski
I am confused with gcc configuration, and I cannot determine if I have a bug or if I am misconfiguring the compiler. Here is the situation: gcc sources: 4.2 snapshot of 20060905 If compiler is configured with: --target=powerpc-*-linux-gnualtivec then I have the following libraries and everythin

Re: powerpc targets, long double implementation, and c++ programs

2006-09-08 Thread Edmar Wienskoski
Ok. I am starting to see the whole picture now. So the whole thing appears to work with --disable-shared, just because the way the linker loads symbols in presence of libgcc_s.so versus libgcc.a. Follow up question: The e500 abi actualy defines long double to be 128bits floats. On rs6000.c, rs6

Re: powerpc targets, long double implementation, and c++ programs

2006-09-11 Thread Edmar Wienskoski
Joseph S. Myers wrote: On Fri, 8 Sep 2006, Edmar Wienskoski wrote: Ok. I am starting to see the whole picture now. So the whole thing appears to work with --disable-shared, just because the way the linker loads symbols in presence of libgcc_s.so versus libgcc.a. Follow up question: The

Use of option -fprofile-arcs is not compatible with -fprofile-use

2008-04-24 Thread Edmar Wienskoski-RA8797
The test case gcc.misc/bprob-1.c is compiled with fprofile-arcs / fprofile-use. The option fprofile-arcs does not enable value profiling. At the second stage compilation, the option fprofile-use enables value profiling. Within tree_find_values_to_profile, if one of the value optimizations alg

Re: Use of option -fprofile-arcs is not compatible with -fprofile-use

2008-05-07 Thread Edmar Wienskoski-RA8797
is compiled with -fprofile-generate and -fprofile-use. so there won't be any warnings. 2008/4/25 Edmar Wienskoski-RA8797 <[EMAIL PROTECTED]>: The test case gcc.misc/bprob-1.c is compiled with fprofile-arcs / fprofile-use. The option fprofile-arcs does not enable value profilin

Re: Use of option -fprofile-arcs is not compatible with -fprofile-use

2008-05-08 Thread Edmar Wienskoski-RA8797
c-4.2 --with-gmp=/import/dr3/s10/gcc-4.2 Thread model: posix gcc version 4.2.0 20070228 (prerelease) so is gcc-4.3 on the platform. 2008/5/7 Edmar Wienskoski-RA8797 <[EMAIL PROTECTED]>: I said if you compile val-prof-1.c the same way bprob-1.c is compiled you get an warning. gcc -g -fp