Displaying 2 results from an estimated 2 matches for "lbbmain_10".
2004 Jun 19
0
[LLVMdev] benchmarking LLVM
...nstruction, so if this test
> will be boosted then it might influence very much other test too.
You're right, this one is pretty bad. Looking at the (most) inner loop in
the "benchmark":
for (f=0; f<n; f++)
x++;
It looks like we're generating this machine code:
.LBBmain_10: # no_exit.5
mov %EDX, %EBX
mov %EBP, %EDX
inc %EBP
mov %EBX, %EDX
inc %EBX
cmp %EBP, %ESI
jl .LBBmain_10 # no_exit.5
The code looks somewhat reasonable except for the three extra copies in
it. GCC destroys us on this loop though, produci...
2004 Jun 19
2
[LLVMdev] benchmarking LLVM
Hi all
i took a look into LLVM benchmarks from nightly tester and
ran Shootout tests on my own. Below go just few outlines.
1. results on my AMD AthlonXP and Xeon used by LLVM
team are different sometime. In particular, both Shootout
and Shootout-C++ show great speed up with LLVM (in
comparison to GCC) on ackerman test on my AthlonXP.
But here: