Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Acovea-4.0.0 : Try out my ebuilds (and scripts)
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3 ... , 14, 15, 16  Next  
Reply to topic    Gentoo Forums Forum Index Gentoo Chat
View previous topic :: View next topic  
Author Message
taskara
Advocate
Advocate


Joined: 10 Apr 2002
Posts: 3763
Location: Australia

PostPosted: Fri Sep 24, 2004 4:07 am    Post subject: Reply with quote

Thanks for your help :)
_________________
Kororaa install method - have Gentoo up and running quickly and easily, fully automated with an installer!
Back to top
View user's profile Send private message
Crocodil
Apprentice
Apprentice


Joined: 06 May 2004
Posts: 163
Location: Poznan, Poland

PostPosted: Wed Sep 29, 2004 8:42 am    Post subject: Reply with quote

Hi :)

I've done the test through the last 2 days and here are my "Yes" answers:
Quote:
40.9 | Yes | -fforce-mem (-O2)
35.8 | Yes | -fno-if-conversion2 (! -O1)
34.6 | Yes | -finline-functions (-O3)
31.9 | Yes | -fno-thread-jumps (! -O1)
30.9 | Yes | -mno-align-stringops
30.9 | Yes | -falign-loops (-O2 GCC 3.3)
29.2 | Yes | -freorder-blocks (-O2)
29.1 | Yes | -fno-merge-constants (! -O1)
27.9 | Yes | -fdelete-null-pointer-checks (-O2)
27.8 | Yes | -malign-double
27.6 | Yes | -minline-all-stringops
27.2 | Yes | -fsched-spec (-O2 GCC 3.3)
25.3 | Yes | -fno-defer-pop (! -O1)
25.2 | Yes | -fno-delayed-branch (! -O1)
24.7 | Yes | -finline-limit
24.6 | Yes | -fno-cprop-registers (! -O1)


Are any of these too unsafe to use?


Thank you very much Hypnos for the scripts!! :)

I also have this idea / question to you...

Maybe you could edit your first post and add to it the scripts for running and getting results from Acovea (both on gcc -3.3.x and gcc-3.4.x)? This topic has gotten quite long and finding them isn't an easy task ;)

Also it would be great if you could add a small "database" of flags that you know are unsafe to use.... :)

This would make a small "HOWTO" out of the first post :)

I hope you think over my propositions :)

Best regards,

Croodil.
Back to top
View user's profile Send private message
Hypnos
Advocate
Advocate


Joined: 18 Jul 2002
Posts: 2867
Location: Omnipresent

PostPosted: Wed Sep 29, 2004 8:50 am    Post subject: Reply with quote

Crocodil wrote:
Are any of these too unsafe to use?

"-malign-double" can cause significant breakage, but the rest look okay.

Quote:
I hope you think over my propositions :)

When I have time ;)
_________________
Personal overlay | Simple backup scheme
Back to top
View user's profile Send private message
celtx
n00b
n00b


Joined: 28 Sep 2004
Posts: 3

PostPosted: Fri Oct 01, 2004 6:05 pm    Post subject: my results athlon-xp Reply with quote

I ran the acovea test suite over several days one or two tests at a time.
used GCC 3.3 on an athon-xp system.

My configuration :
copied the opteron files
modified the march flag
-march=athlon-xp

and added in changes to comment out known unsafe items.

<!-- Options specific to "fast math" -->
<!-- flag type="simple" value="-fno-math-errno" /-->
<!-- flag type="simple" value="-funsafe-math-optimizations" /-->
<!-- flag type="simple" value="-fno-trapping-math" /-->
<!-- flag type="simple" value="-ffinite-math-only" /-->
<!--flag type="simple" value="-fno-signaling-nans" /-->

<!-- flag type="simple" value="-mieee-fp" /-->
<!-- flag type="simple" value="-fnew-ra" /-->

here were the Yes answers

34.3 | Yes | -fsched-spec (-O2 GCC 3.3)
28.2 | Yes | -falign-loops (-O2 GCC 3.3)
27.7 | Yes | -fdelete-null-pointer-checks (-O2)
26.2 | Yes | -freorder-functions (-O2 GCC 3.3)
25.2 | Yes | -freorder-blocks (-O2)
24.1 | Yes | -fno-merge-constants (! -O1)

which translates to
-O1 -march=athlon=xp -fscehd-spec -falign-loops -fdelete-null-pointer-checks -freorder-functions -freorder-blocks -fno-merge-constants -pipe

by default I was already using -O2 which covers majority of the items above

a few flags contained in -O2 were listed as No
-frerun-loop-opt
-fforce-mem
-fschedule-insns
-fschedule-insns2
-fregmove
-fstrict-aliasing
-falign-labels

the rest were all listed as Maybe

the test showed that on my system adding -fno-merge-constants
to the default of -O2 may help and that a few -O2 flags may not help that much.

So the default -O2 -march=athlon-xp -pipe is a good choice already on this system.
Back to top
View user's profile Send private message
galay2
Apprentice
Apprentice


Joined: 02 Feb 2004
Posts: 208

PostPosted: Mon Oct 04, 2004 7:20 pm    Post subject: Reply with quote

I copied the script form page 2 and whenever I run the first script (named as start), I get "Could not find that benchmark, so here's some help.". How come? I have these files in

Code:
ls /usr/share/acovea/benchmarks/
almabench.c  fftbench.c   linbench.c   treebench.c evobench.c   huffbench.c  mat1bench.c


Heres the script I use
Code:
#!/bin/sh

BENCHES="alma evo fft huff lin mat1 tree"

for bench in $BENCHES; do
        echo ""
        echo "*** $bench ***"
        echo ${bench}bench.c
        time runacovea -config gcc33_pentium4.acovea -bench ${bench}bench.c\
                1> ${bench}.run 2> ${bench}.err
done


Also I can run the command itself
Code:

time runacovea -config gcc33_pentium4.acovea -bench fftbench.c                1> ${bench}.run 2> ${bench}.err
Back to top
View user's profile Send private message
Hypnos
Advocate
Advocate


Joined: 18 Jul 2002
Posts: 2867
Location: Omnipresent

PostPosted: Mon Oct 04, 2004 8:20 pm    Post subject: Reply with quote

galay2,

try putting the command in the script all on one line as you did standalone.

Also, make sure there are no trailing spaces after the "\" in the script.
_________________
Personal overlay | Simple backup scheme
Back to top
View user's profile Send private message
galay2
Apprentice
Apprentice


Joined: 02 Feb 2004
Posts: 208

PostPosted: Wed Oct 06, 2004 3:08 pm    Post subject: Reply with quote

Hypnos: I got it working thx but I got a question. Are you not supposed to run anything while it's running? And also turn of all services/X etc? Coz if its based on the time as a means of scoring, wouldnt other applications affect its benchmarks?

Also, how accurate is a GA in general? Its not exactly a exact answer is it?

Lastly is it possible to resume? I'm assuming each test can be run indivudialy? and the result script wont show anything until all tests are done? (coz I'm getting all 'no's after alma evo and fft)

Thanks
Back to top
View user's profile Send private message
stikboy
n00b
n00b


Joined: 14 Sep 2003
Posts: 68
Location: Colorado

PostPosted: Thu Oct 07, 2004 4:45 am    Post subject: Reply with quote

I am currently running the scripts to see what flags I could change to. My question is this:

After I finish running these tests, I am going to re-install Gentoo. This is so I can use reiser4, nptl and udev. I'm wondering, since everything else I am going to install after the new base install will be the same as I currently have, will the CFLAGS that are recommended by these scripts be usable on the next install? or will reiser4 and/or nptl skew the results a lot?
Back to top
View user's profile Send private message
Deranger
Veteran
Veteran


Joined: 26 Aug 2004
Posts: 1215

PostPosted: Thu Oct 07, 2004 6:34 am    Post subject: Reply with quote

stikboy wrote:
I am currently running the scripts to see what flags I could change to. My question is this:

After I finish running these tests, I am going to re-install Gentoo. This is so I can use reiser4, nptl and udev. I'm wondering, since everything else I am going to install after the new base install will be the same as I currently have, will the CFLAGS that are recommended by these scripts be usable on the next install? or will reiser4 and/or nptl skew the results a lot?

Those C(XX)FLAGS are very, very optimized, based on your current system. I would say if you run these tests under your new system, you'll get different results. But these tests give a good overview of C(XX)FLAGS you should use.
Back to top
View user's profile Send private message
TheKat
n00b
n00b


Joined: 24 Jan 2004
Posts: 49

PostPosted: Sat Oct 09, 2004 6:48 pm    Post subject: "Safe" to use system while acovea runs? Reply with quote

So here's my question, basic as it is.

I currently have Acovea running on my system and am using a roommate's.

If I use my system to actually do stuf while Acovea runs, will I skew the results at all? Does Acovea perform it's own, CPU time independent analysis, or is it best to just leave a system running until it's done, and not put any other stress at all on it?
Back to top
View user's profile Send private message
SirRichard
n00b
n00b


Joined: 10 Oct 2004
Posts: 13
Location: Stockelsdorf, Germany

PostPosted: Sun Oct 10, 2004 2:51 pm    Post subject: Reply with quote

And here are MY results - calculated on my system (Dual Athlon-MP 2800+, 2G RAM, gcc 3.4.2 with modified config for this system):

Code:

 Score |  So?  | Switch (annotation)
------------------------------------------------------------------------------
  35.5 |  Yes  | -finline-functions (-O3)
  31.9 |  Yes  | -ftracer
  31.7 |  Yes  | -fno-merge-constants (! -O1)
  31.3 | Maybe | -fgcse (-O2)
  31.1 |  Yes  | -fno-defer-pop (! -O1)
  30.6 |  Yes  | -fsched-interblock (-O2 GCC 3.3)
  29.4 | Maybe | -fweb
  29.4 | Maybe | -ffinite-math-only (fast math)
  28.5 |  Yes  | -fno-delayed-branch (! -O1)
  28.1 |  Yes  | -finline-limit
  27.2 |  Yes  | -frerun-cse-after-loop (-O2)
  27.0 | Maybe | -fno-math-errno (fast math)
  26.4 | Maybe | -fstrength-reduce (-O2)
  25.9 | Maybe | -funsafe-math-optimizations (fast math)
  25.8 |  Yes  | -fno-thread-jumps (! -O1)
  25.8 | Maybe | -fcaller-saves (-O2)
  25.5 |  Yes  | -falign-functions
  25.3 | Maybe | -mieee-fp
  25.3 | Maybe | -fpeel-loops
  25.1 | Maybe | -funswitch-loops
  24.9 |   No  | -fstrict-aliasing (-O2)
  24.3 |  Yes  | -fdelete-null-pointer-checks (-O2)
  24.2 |  Yes  | -mno-align-stringops
  24.2 |  Yes  | -fsched-spec (-O2 GCC 3.3)
  24.1 | Maybe | -fexpensive-optimizations (-O2)
  23.5 |  Yes  | -foptimize-sibling-calls (-O2)
  23.2 | Maybe | -malign-double
  23.2 |  Yes  | -fcse-skip-blocks (-O2)
  23.0 | Maybe | -fno-signaling-nans (fast math)
  22.9 | Maybe | -maccumulate-outgoing-args
  22.6 | Maybe | -falign-loops (-O2 GCC 3.3)
  22.3 | Maybe | -minline-all-stringops
  21.8 | Maybe | -fno-crossjumping (! -O1)
  21.8 | Maybe | -fschedule-insns2 (-O2)
  21.5 | Maybe | -falign-jumps (-O2 GCC 3.3)
  20.8 | Maybe | -frerun-loop-opt (-O2)
  20.8 |  Yes  | -fmove-all-movables
  20.7 | Maybe | -freorder-blocks (-O2)
  20.7 | Maybe | -fno-cprop-registers (! -O1)
  20.4 | Maybe | -fno-if-conversion2 (! -O1)
  20.2 | Maybe | -falign-labels (-O2 GCC 3.3)
  20.1 |   No  | -fprefetch-loop-arrays
  18.6 | Maybe | -mno-push-args
  18.1 |   No  | -fno-trapping-math (fast math)
  17.8 | Maybe | -fno-omit-frame-pointer (! -O1)
  17.5 |   No  | -fforce-mem (-O2)
  17.2 | Maybe | -fregmove (-O2)
  17.0 | Maybe | -fcse-follow-jumps (-O2)
  16.5 | Maybe | -frename-registers (-O3)
  16.2 | Maybe | -freduce-all-givs
  15.6 |   No  | -fomit-frame-pointer
  15.6 |   No  | -fno-if-conversion (! -O1)
  15.2 |   No  | -funit-at-a-time
  13.7 |   No  | -freorder-functions (-O2 GCC 3.3)
  13.5 |   No  | -fpeephole2 (-O2)
  12.2 | Maybe | -fbranch-target-load-optimize2
  11.6 |   No  | -fnew-ra
   9.7 |   No  | -fno-inline
   8.0 |   No  | -fschedule-insns (-O2)
   6.6 |   No  | -ffloat-store
   0.0 |   No  | -fno-guess-branch-probability (! -O1)
   0.0 |   No  | -fno-loop-optimize (! -O1)
   0.0 |   No  | -funroll-loops
   0.0 |   No  | -funroll-all-loops
   0.0 |   No  | -fbranch-target-load-optimize
   0.0 |   No  | -mfpmath=387
   0.0 |   No  | -mfpmath=sse
   0.0 |   No  | -mfpmath=sse,387
   0.0 |   No  | -momit-leaf-frame-pointer


I don't know how much it differs from single-CPU Athlon XP systems, but I wanted to try on myself. The test ran about 26 hours 48 mins and gave me the results above. Now I consider the flags that will be useful and boost up my system. -ftracer goes straight into my CFLAGS, but what about the others? I think and so "-O3 -pipe -march=athlon-mp -ftracer" is right for me (as "-mtune=athlon-mp" is implied by "-march=athlon-mp").

I dislike the thought of using either -O1 and activating some flags or -O3 and disabling some flags. I trust in -O3 creating generally good code (yes, I know, that may be naive). Do these flags like "-fno-merge-constants" really make such a difference?
Back to top
View user's profile Send private message
Tanisete
Guru
Guru


Joined: 12 Mar 2004
Posts: 312

PostPosted: Fri Nov 05, 2004 3:36 pm    Post subject: Reply with quote

I've run the acovea test on my amd 1700+ and afetr the perl script i have this "Yes":

Code:

 Score |  So?  | Switch (annotation)
------------------------------------------------------------------------------
  42.5 |  Yes  | -fstrict-aliasing (-O2)
  30.1 |  Yes  | -fno-omit-frame-pointer (! -O1)
  29.9 |  Yes  | -fno-defer-pop (! -O1)
  29.8 |  Yes  | -mno-push-args
  29.7 |  Yes  | -fno-trapping-math (fast math)
  29.7 |  Yes  | -ftracer
  29.4 |  Yes  | -fno-delayed-branch (! -O1)
  29.2 |  Yes  | -frerun-cse-after-loop (-O2)
  28.1 |  Yes  | -maccumulate-outgoing-args
  28.0 |  Yes  | -fno-cprop-registers (!
-O1)
  27.9 |  Yes  | -fno-merge-constants (! -O1)
  27.8 |  Yes  | -fno-crossjumping (! -O1)
  27.4 |  Yes  | -falign-jumps (-O2 GCC 3.3)
  26.7 |  Yes  | -malign-double
  25.9 |  Yes  | -fsched-interblock (-O2 GCC 3.3)
  25.9 |  Yes  | -freorder-functions (-O2 GCC 3.3)
  25.5 |  Yes  | -fno-thread-jumps (! -O1)
  25.3 |  Yes  | -mno-align-stringops
  25.2 |  Yes  | -fsched-spec (-O2 GCC 3.3)
  25.0 |  Yes  | -frerun-loop-opt (-O2)
  22.7 |  Yes  | -falign-labels (-O2 GCC 3.3)


Which of them could be dangerous or cause an inestable system? I Know that -malign-double is dangerous, but i don't know about the other ones...

Thaks a lot for all the scripts!!! They're great!!!!!!!!!! Very easy to use.
Back to top
View user's profile Send private message
tnt
Veteran
Veteran


Joined: 27 Feb 2004
Posts: 1127

PostPosted: Wed Nov 10, 2004 11:16 am    Post subject: Reply with quote

I've run tests and got this results on my Barton box:
Code:
 Score |  So?  | Switch (annotation)
------------------------------------------------------------------------------
  41.2 |  Yes  | -fgcse (-O2)
  37.4 |  Yes  | -finline-functions (-O3)
  31.9 |  Yes  | -mno-push-args
  30.1 | Maybe | -fstrict-aliasing (-O2)
  30.0 | Maybe | -fno-delayed-branch (! -O1)
  29.0 |  Yes  | -mno-align-stringops
  29.0 |  Yes  | -fno-signaling-nans (fast math)
  28.9 | Maybe | -fno-trapping-math (fast math)
  28.6 | Maybe | -fmove-all-movables
  28.6 | Maybe | -fno-math-errno (fast math)
  28.6 | Maybe | -ftracer
  27.6 | Maybe | -frename-registers (-O3)
  27.5 |  Yes  | -fno-cprop-registers (! -O1)
  27.3 |  Yes  | -fno-crossjumping (! -O1)
  27.3 |  Yes  | -fexpensive-optimizations (-O2)
  26.9 | Maybe | -frerun-cse-after-loop (-O2)
  26.6 |  Yes  | -funswitch-loops
  25.8 | Maybe | -fsched-interblock (-O2 GCC 3.3)
  24.9 | Maybe | -fweb (-O3 GCC 3.4)
  23.9 |   No  | -fstrength-reduce (-O2)
  23.9 | Maybe | -funsafe-math-optimizations (fast math)
  23.4 | Maybe | -fno-if-conversion2 (! -O1)
  22.9 | Maybe | -fdelete-null-pointer-checks (-O2)
  22.7 | Maybe | -fno-omit-frame-pointer (! -O1)
  22.2 |   No  | -fprefetch-loop-arrays
  22.2 | Maybe | -maccumulate-outgoing-args
  21.8 |  Yes  | -fno-merge-constants (! -O1)
  21.6 |   No  | -fschedule-insns2 (-O2)
  21.6 |   No  | -fforce-mem (-O2)
  21.5 | Maybe | -fno-thread-jumps (! -O1)
  21.5 | Maybe | -falign-functions (-O2 GCC 3.4)
  21.4 | Maybe | -falign-loops (-O2 GCC 3.3)
  21.2 |   No  | -ffinite-math-only (fast math)
  21.2 | Maybe | -finline-limit
  20.8 |   No  | -freduce-all-givs
  20.8 | Maybe | -fcse-skip-blocks (-O2)
  20.4 | Maybe | -freorder-blocks (-O2)
  19.7 | Maybe | -fno-defer-pop (! -O1)
  19.5 |   No  | -funit-at-a-time (-O2 GCC 3.4)
  19.1 | Maybe | -fcse-follow-jumps (-O2)
  18.8 | Maybe | -fpeel-loops
  18.4 | Maybe | -fcaller-saves (-O2)
  18.4 |   No  | -fno-if-conversion (! -O1)
  18.1 | Maybe | -fpeephole2 (-O2)
  17.8 |   No  | -foptimize-sibling-calls (-O2)
  17.7 |   No  | -frerun-loop-opt (-O2)
  17.5 |   No  | -mieee-fp
  17.1 |   No  | -falign-jumps (-O2 GCC 3.3)
  16.7 | Maybe | -falign-labels (-O2 GCC 3.3)
  15.6 |   No  | -freorder-functions (-O2 GCC 3.3)
  15.2 | Maybe | -fsched-spec (-O2 GCC 3.3)
  12.3 |   No  | -fregmove (-O2)
  11.5 |   No  | -fno-inline
  11.2 |   No  | -fschedule-insns (-O2)
  11.1 |   No  | -minline-all-stringops
   9.2 |   No  | -funroll-loops
   8.5 |   No  | -fnew-ra
   8.2 |   No  | -ffloat-store
   7.0 |   No  | -fbranch-target-load-optimize2
   6.8 |   No  | -fno-guess-branch-probability (! -O1)
   6.5 |   No  | -fbranch-target-load-optimize
   6.2 |   No  | -mfpmath=sse
   0.0 |   No  | -fno-loop-optimize (! -O1)
   0.0 |   No  | -funroll-all-loops
   0.0 |   No  | -mfpmath=387
   0.0 |   No  | -mfpmath=sse,387

But, I've noticed that some of my old flags are missing - they are nor Yes nor Maybe nor No. What happened with them.
Code:
-pipe
-fomit-frame-pointer
-fforce-addr
-fprefetch-loop-arrays-ffast-math

Maybe it's because I've changed Opteron's config file? Should I use some other config?

:roll:
Back to top
View user's profile Send private message
nmcsween
Guru
Guru


Joined: 12 Nov 2003
Posts: 381

PostPosted: Thu Nov 11, 2004 10:18 am    Post subject: Reply with quote

Hypnos wrote:

....
Well, in principle, if you run Acovea with nothing else running, you're optimizing towards a dedicated system. You are more likely to optimize for smaller code size in a "dirtier" testing environment.

Wouldn't this be best since most environments are dirty, meaning that theres multiple processes running. Or is acovea multithreaded? Which means the environment would already be dirty.
_________________
Great Resources
Back to top
View user's profile Send private message
TiE10
n00b
n00b


Joined: 23 Jun 2004
Posts: 59
Location: NA

PostPosted: Fri Nov 12, 2004 7:27 pm    Post subject: Reply with quote

Sorry if this was already addressed, but I skimmed and I didn't see, so here goes:
I copied the scripts posted earlier in the forum
Code:

#!/bin/sh

BENCHES="alma evo fft huff lin mat1 tree"

for bench in $BENCHES; do
        echo ""
        echo "*** $bench ***"
        time runacovea -config gcc33_pentium4.acovea -bench ${bench}benc$
                1> ${bench}.run 2> ${bench}.err
done

and i tried running them, but the output was this:
Code:

tie10@phoenix4188 ~/acovea $ sh ./acoveascript.sh

*** alma ***
./acoveascript.sh: line 8: alma.run: Permission denied

real    0m0.001s
user    0m0.000s
sys     0m0.001s

*** evo ***
./acoveascript.sh: line 8: evo.run: Permission denied

real    0m0.001s
user    0m0.000s
sys     0m0.001s

*** fft ***
./acoveascript.sh: line 8: fft.run: Permission denied

real    0m0.001s
user    0m0.001s
sys     0m0.000s

*** huff ***
./acoveascript.sh: line 8: huff.run: Permission denied

real    0m0.002s
user    0m0.000s
sys     0m0.000s

*** lin ***
./acoveascript.sh: line 8: lin.run: Permission denied

real    0m0.001s
user    0m0.000s
sys     0m0.001s

*** mat1 ***
./acoveascript.sh: line 8: mat1.run: Permission denied

real    0m0.001s
user    0m0.000s
sys     0m0.001s

*** tree ***
./acoveascript.sh: line 8: tree.run: Permission denied

real    0m0.001s
user    0m0.000s
sys     0m0.001s

How can I fix that..?
_________________
"Common sense is something earned by thinking."
Back to top
View user's profile Send private message
TiE10
n00b
n00b


Joined: 23 Jun 2004
Posts: 59
Location: NA

PostPosted: Sun Nov 14, 2004 5:56 pm    Post subject: Reply with quote

bump
_________________
"Common sense is something earned by thinking."
Back to top
View user's profile Send private message
TiE10
n00b
n00b


Joined: 23 Jun 2004
Posts: 59
Location: NA

PostPosted: Tue Nov 16, 2004 4:02 am    Post subject: Reply with quote

bump :(
_________________
"Common sense is something earned by thinking."


Last edited by TiE10 on Tue Nov 16, 2004 5:01 am; edited 1 time in total
Back to top
View user's profile Send private message
aethyr
Veteran
Veteran


Joined: 06 Apr 2003
Posts: 1085
Location: NYC

PostPosted: Tue Nov 16, 2004 4:17 am    Post subject: Reply with quote

I have no idea, do you have permission to run it in that directory? How did you install it?

Try this:
Code:
cd ~/
runacovea -config gcc33_pentium4.acovea -bench almabench.c 1> almabench.run 2> almabench.err


If that works, then make sure you're running it somewhere you can write to.

[edit] I'm not sure you if you just pasted it wrong into the forums, but the script you pasted is sligtly different than the one Hypnos wrote. Make sure you're running the proper script:
Code:
#!/bin/sh
 
BENCHES="alma evo fft huff lin mat1 tree"
 
for bench in $BENCHES; do
        echo ""
        echo "*** $bench ***"
        time runacovea -config gcc33_pentium4.acovea -bench ${bench}bench.c\
                1> ${bench}.run 2> ${bench}.err
done


Here's where what you pasted is wrong:
Code:
        time runacovea -config gcc33_pentium4.acovea -bench ${bench}benc$
                1> ${bench}.run 2> ${bench}.err
Back to top
View user's profile Send private message
TiE10
n00b
n00b


Joined: 23 Jun 2004
Posts: 59
Location: NA

PostPosted: Tue Nov 16, 2004 5:00 am    Post subject: Reply with quote

:( well i couldn't get the script to work still for some reason. When i did it before i upraded to nitro3, i just copied the scripts and did, for example:
Code:

tie10@phoenix4188 ~/acovea $ sh ./acoveascript.sh

and it ran the tests
I took your suggestion to run it the recommended way and emerged acovea and used your command:
Code:

runacovea -config gcc33_pentium4.acovea -bench almabench.c 1> almabench.run 2> almabench.err

and it seems to be running fine now. thx for the reply

oh yea, and also that was just a mistake in the copying.. the $ where the difference was located was where my terminal cut-off the text :) , so i was using the right script but what was wrong? which permissions does it need..?well whatever i've got it benchmarking now either way
_________________
"Common sense is something earned by thinking."
Back to top
View user's profile Send private message
Gergan Penkov
Veteran
Veteran


Joined: 17 Jul 2004
Posts: 1464
Location: das kleinste Kuhdorf Deutschlands :)

PostPosted: Wed Nov 17, 2004 11:44 pm    Post subject: this program is simply buggy Reply with quote

Ok, here it is - I've installed acovea, because I have had a problem in the past - my system crawled, that was after a fresh install - reiserfs4,udev and new cflags ::(( in the beginning i thought it was the reiserfs4 - yes, it slows down the kernel because of the 8k stack, but that was not the main problem. Then I have also tried without udev all at vain and i had no idea what the hell is going on - from log in to prompt 2 sec, gnome and xfce started so slow - the system was crawling at starting application, it was laggy not slow. Then i have changed the kernel from mm, which is too risky and not stable in the last time to nitro and dev-sources with and without preempt and so on - the result was absolutly nothing, probably a little bit faster with 4k stack. I have checked the harddisk absolutly normal I had a steady:
Code:
  # hdparm -Tt /dev/hda

/dev/hda:
 Timing cached reads:   608 MB in  2.00 seconds = 303.74 MB/sec
 Timing buffered disk reads:  164 MB in  3.03 seconds =  54.12 MB/sec

that was the crucial point I understood that the problem was in the cflags, i had march=athlon-xp -O3 -fomit-fram-pointer, nothing fancy - proclaimed to be fast for athlon and stable :::)))) and I wanted to optimize :::)) installed acovea, waited to the last test - ***glibc detected*** double free or something similar for all the runs in the tree test and at the end segmentation :::)) - i was in panic - i thought it was a hardware problem, although the system functions without problem slow but very stable it was not overclocked. Then was the second thought I have changed the config to work for athlon-xp (esp. -fomit-frame-pointer - which is off for all -O levels for athlon-xp and quite right) , than I saw that the test used -fomit-frame-pointer together with -fno-omit-frame-pointer - I thought relaxed that i had found the problem - alas ...
than I have recompiled gcc, glibc and acovea and its dependant libraries simply with -march=athlon-xp -O2 -pipe, this didn't help also, but the result was that the system was not so laggy the compilation was faster :::)) than I have read the annoncenment from red hat for the new glibc and here we are :
http://www.fedorazine.com/index.php?option=content&task=view&id=292
for your viewing pleasure. So the question is what will achieve a buggy program, which claims to optimize the binaries and instead is using code which is not secure and have bugs ...! And another question what tests the program in fact the speed of execution - ok but what is the connection between the speed of execution under heavy load (CPU, Disc for example). probably the program is fast but its loading is slow....
The conclusion ... i have learned my lesson and I will not more strive to achieve unreachable speeds.
It's enough that doom3 is faster on linux ....
Have fun
Back to top
View user's profile Send private message
mekong
Tux's lil' helper
Tux's lil' helper


Joined: 23 Apr 2004
Posts: 92
Location: Rdam - NL - EU

PostPosted: Thu Nov 18, 2004 5:07 am    Post subject: Acovea & Pentium2 Reply with quote

I don't see alot people running Acovea on Pentium2, I did it and it took a week :D

My system:

Code:

processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 3
model name      : Pentium II (Klamath)
stepping        : 3
cpu MHz         : 233.364
cache size      : 512 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 2
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov mmx
bogomips        : 459.77


Code:

gcc version 3.4.3  (Gentoo Linux 3.4.3, ssp-3.4.3-0, pie-8.7.6.6)


The result:
Code:

  Mean  | Std. Dev. | Conf. | Score |  So?  | Switch (annotation)
------------------------------------------------------------------------------
  0.322 |   0.080   | 1.000 |  32.2 |  Yes  | -fforce-mem (-O2)
  0.362 |   0.255   | 0.845 |  30.6 | Maybe | -fstrict-aliasing (-O2)
  0.304 |   0.111   | 0.994 |  30.2 |  Yes  | -foptimize-sibling-calls (-O2)
  0.342 |   0.239   | 0.848 |  29.1 | Maybe | -funsafe-math-optimizations (fast math)
  0.281 |   0.093   | 0.998 |  28.0 |  Yes  | -fdelete-null-pointer-checks (-O2)
  0.278 |   0.093   | 0.997 |  27.8 |  Yes  | -fno-defer-pop (! -O1)
  0.290 |   0.144   | 0.957 |  27.7 |  Yes  | -fno-math-errno (fast math)
  0.287 |   0.143   | 0.956 |  27.5 |  Yes  | -freorder-functions (-O2 GCC 3.3)
  0.283 |   0.134   | 0.965 |  27.3 |  Yes  | -ffinite-math-only (fast math)
  0.327 |   0.238   | 0.829 |  27.1 | Maybe | -freorder-blocks (-O2)
  0.267 |   0.102   | 0.991 |  26.4 |  Yes  | -fno-crossjumping (! -O1)
  0.267 |   0.112   | 0.984 |  26.3 |  Yes  | -finline-functions (-O3)
  0.262 |   0.053   | 1.000 |  26.2 |  Yes  | -fpeel-loops
  0.323 |   0.252   | 0.801 |  25.9 | Maybe | -fstrength-reduce (-O2)
  0.264 |   0.117   | 0.976 |  25.8 |  Yes  | -fno-merge-constants (! -O1)
  0.254 |   0.085   | 0.997 |  25.3 |  Yes  | -fno-cprop-registers (! -O1)
  0.293 |   0.198   | 0.862 |  25.3 | Maybe | -fexpensive-optimizations (-O2)
  0.268 |   0.144   | 0.938 |  25.2 |  Yes  | -fno-trapping-math (fast math)
  0.248 |   0.081   | 0.998 |  24.8 |  Yes  | -falign-labels (-O2 GCC 3.3)
  0.249 |   0.104   | 0.984 |  24.5 |  Yes  | -frerun-cse-after-loop (-O2)
  0.242 |   0.070   | 0.999 |  24.2 |  Yes  | -fno-omit-frame-pointer (! -O1)
  0.257 |   0.139   | 0.936 |  24.0 |  Yes  | -fcse-skip-blocks (-O2)
  0.264 |   0.157   | 0.908 |  24.0 |  Yes  | -maccumulate-outgoing-args
  0.244 |   0.104   | 0.982 |  24.0 |  Yes  | -fcse-follow-jumps (-O2)
  0.283 |   0.207   | 0.828 |  23.5 | Maybe | -fschedule-insns2 (-O2)
  0.249 |   0.133   | 0.939 |  23.4 |  Yes  | -fno-delayed-branch (! -O1)
  0.233 |   0.054   | 1.000 |  23.2 |  Yes  | -fmove-all-movables
  0.257 |   0.157   | 0.898 |  23.0 |  Yes  | -fno-signaling-nans (fast math)
  0.241 |   0.122   | 0.952 |  22.9 |  Yes  | -mfpmath=387
  0.262 |   0.175   | 0.864 |  22.6 | Maybe | -fsched-interblock (-O2 GCC 3.3)
  0.226 |   0.075   | 0.997 |  22.5 |  Yes  | -fno-if-conversion2 (! -O1)
  0.226 |   0.086   | 0.991 |  22.4 |  Yes  | -funit-at-a-time
  0.242 |   0.139   | 0.919 |  22.2 |  Yes  | -fpeephole2 (-O2)
  0.253 |   0.168   | 0.867 |  22.0 |  Yes  | -mieee-fp
  0.231 |   0.119   | 0.948 |  21.9 |  Yes  | -mno-align-stringops
  0.227 |   0.109   | 0.962 |  21.8 |  Yes  | -minline-all-stringops
  0.246 |   0.155   | 0.887 |  21.8 |  Yes  | -falign-functions
  0.280 |   0.229   | 0.778 |  21.8 | Maybe | -fweb
  0.228 |   0.115   | 0.953 |  21.8 |  Yes  | -fsched-spec (-O2 GCC 3.3)
  0.225 |   0.120   | 0.939 |  21.1 |  Yes  | -funswitch-loops
  0.253 |   0.185   | 0.828 |  21.0 | Maybe | -fcaller-saves (-O2)
  0.247 |   0.175   | 0.840 |  20.7 | Maybe | -fno-thread-jumps (! -O1)
  0.223 |   0.128   | 0.917 |  20.4 |  Yes  | -mno-push-args
  0.204 |   0.065   | 0.998 |  20.4 |  Yes  | -frerun-loop-opt (-O2)
  0.237 |   0.162   | 0.855 |  20.2 | Maybe | -malign-double
  0.291 |   0.298   | 0.671 |  19.5 |   No  | -fgcse (-O2)
  0.260 |   0.227   | 0.747 |  19.4 | Maybe | -fno-if-conversion (! -O1)
  0.214 |   0.136   | 0.885 |  19.0 |  Yes  | -fregmove (-O2)
  0.253 |   0.221   | 0.747 |  18.9 | Maybe | -finline-limit
  0.212 |   0.133   | 0.890 |  18.8 |  Yes  | -ftracer
  0.257 |   0.241   | 0.714 |  18.3 | Maybe | -falign-jumps (-O2 GCC 3.3)
  0.254 |   0.245   | 0.701 |  17.8 | Maybe | -falign-loops (-O2 GCC 3.3)
  0.240 |   0.213   | 0.740 |  17.8 | Maybe | -frename-registers (-O3)
  0.241 |   0.231   | 0.702 |  16.9 | Maybe | -funroll-all-loops
  0.216 |   0.191   | 0.741 |  16.0 | Maybe | -freduce-all-givs
  0.206 |   0.171   | 0.771 |  15.9 | Maybe | -fschedule-insns (-O2)
  0.226 |   0.235   | 0.662 |  15.0 |   No  | -fno-inline
  0.217 |   0.284   | 0.556 |  12.1 |   No  | -fnew-ra
  0.210 |   0.268   | 0.566 |  11.9 |   No  | -fomit-frame-pointer
  0.181 |   0.218   | 0.594 |  10.7 |   No  | -funroll-loops
  0.130 |   0.153   | 0.605 |   7.9 |   No  | -fbranch-target-load-optimize
  0.147 |   0.224   | 0.488 |   7.2 |   No  | -fno-loop-optimize (! -O1)
  0.057 |   0.158   | 0.000 |   0.0 |   No  | -fno-guess-branch-probability (! -O1)
  0.101 |   0.214   | 0.000 |   0.0 |   No  | -ffloat-store
  0.114 |   0.088   | 0.000 |   0.0 |   No  | -fbranch-target-load-optimize2
  0.098 |   0.115   | 0.000 |   0.0 |   No  | -momit-leaf-frame-pointer
Back to top
View user's profile Send private message
mrfree
Veteran
Veteran


Joined: 15 Mar 2003
Posts: 1303
Location: Europe.Italy.Sulmona

PostPosted: Mon Dec 20, 2004 9:06 am    Post subject: Reply with quote

I post my acovea experience here

Any ideas??? :?
_________________
Please EU, pimp my country!

ICE: /etc/init.d/iptables panic
Back to top
View user's profile Send private message
Twist
Guru
Guru


Joined: 03 Jan 2003
Posts: 414
Location: San Diego

PostPosted: Thu Dec 23, 2004 11:05 am    Post subject: Reply with quote

These were my results, using:

Athlon64 3400+ (used the gcc34_opteron.acovea profile for acovea)
gcc 3.4.3
Tested with Gnome running, no apps active
Code:

 Score |  So?  | Switch (annotation)
------------------------------------------------------------------------------
  35.8 |  Yes  | -minline-all-stringops
  32.6 |  Yes  | -mno-push-args
  31.8 | Maybe | -finline-functions (-O3)
  31.8 |  Yes  | -fexpensive-optimizations (-O2)
  30.4 | Maybe | -fschedule-insns (-O2)
  30.3 | Maybe | -fpeel-loops
  30.1 |  Yes  | -fno-if-conversion2 (! -O1)
  29.8 |  Yes  | -fno-defer-pop (! -O1)
  29.7 |  Yes  | -fcse-skip-blocks (-O2)
  29.1 | Maybe | -frerun-loop-opt (-O2)
  28.3 |  Yes  | -fsched-interblock (-O2 GCC 3.3)
  28.2 |  Yes  | -foptimize-sibling-calls (-O2)
  27.4 |  Yes  | -falign-jumps (-O2 GCC 3.3)
  27.4 | Maybe | -fstrict-aliasing (-O2)
  26.9 | Maybe | -fno-merge-constants (! -O1)
  26.5 | Maybe | -finline-limit
  26.1 | Maybe | -falign-functions
  25.7 | Maybe | -fno-delayed-branch (! -O1)
  25.4 | Maybe | -fpeephole2 (-O2)
  25.4 | Maybe | -freorder-functions (-O2 GCC 3.3)
  25.0 | Maybe | -fno-signaling-nans (fast math)
  25.0 | Maybe | -freorder-blocks (-O2)
  24.7 |   No  | -fstrength-reduce (-O2)
  24.4 | Maybe | -frerun-cse-after-loop (-O2)
  24.3 |  Yes  | -fmove-all-movables
  24.2 | Maybe | -fcse-follow-jumps (-O2)
  23.6 | Maybe | -fschedule-insns2 (-O2)
  23.2 | Maybe | -fno-math-errno (fast math)
  22.8 |  Yes  | -fsched-spec (-O2 GCC 3.3)
  22.8 | Maybe | -maccumulate-outgoing-args
  22.5 | Maybe | -fdelete-null-pointer-checks (-O2)
  22.5 | Maybe | -falign-labels (-O2 GCC 3.3)
  22.4 | Maybe | -fno-thread-jumps (! -O1)
  22.3 | Maybe | -mieee-fp
  22.2 | Maybe | -ftracer
  22.0 | Maybe | -mno-align-stringops
  21.4 | Maybe | -fno-crossjumping (! -O1)
  21.3 | Maybe | -fno-cprop-registers (! -O1)
  21.3 |  Yes  | -funit-at-a-time
  21.1 | Maybe | -frename-registers (-O3)
  20.9 | Maybe | -ffinite-math-only (fast math)
  20.8 | Maybe | -fno-trapping-math (fast math)
  20.6 | Maybe | -funswitch-loops
  20.4 |   No  | -fweb
  20.2 | Maybe | -fcaller-saves (-O2)
  20.1 |   No  | -falign-loops (-O2 GCC 3.3)
  19.9 |   No  | -fgcse (-O2)
  19.1 |   No  | -fno-omit-frame-pointer (! -O1)
  17.3 |   No  | -funsafe-math-optimizations (fast math)
  17.1 |   No  | -fno-if-conversion (! -O1)
  15.6 |   No  | -fregmove (-O2)
  15.4 | Maybe | -fbranch-target-load-optimize
  15.1 |   No  | -fprefetch-loop-arrays
  13.6 |   No  | -fnew-ra
  13.4 |   No  | -fno-inline
  12.2 |   No  | -freduce-all-givs
  12.2 |   No  | -funroll-all-loops
  11.5 |   No  | -fforce-mem (-O2)
   8.7 |   No  | -funroll-loops
   5.2 |   No  | -fno-loop-optimize (! -O1)
   4.6 |   No  | -ffloat-store
   0.0 |   No  | -fno-guess-branch-probability (! -O1)
   0.0 |   No  | -fbranch-target-load-optimize2
   0.0 |   No  | -mfpmath=387
   0.0 |   No  | -mfpmath=sse
   0.0 |   No  | -mfpmath=sse,387
Back to top
View user's profile Send private message
Mango Madness
n00b
n00b


Joined: 16 Jun 2003
Posts: 6

PostPosted: Sat Dec 25, 2004 9:18 pm    Post subject: Reply with quote

so has anyone done any real world testing with something like mencoder to see if running acovea by itself or in a "busy" environment would produce better results? AS soon as i get my Christmas hardware going, i'll do some of my own benchmarks...just trying to gather input from others. Merry Christmas!
Back to top
View user's profile Send private message
Twist
Guru
Guru


Joined: 03 Jan 2003
Posts: 414
Location: San Diego

PostPosted: Sun Dec 26, 2004 12:57 pm    Post subject: Reply with quote

I started a new topic to discuss my results using the flags indicated by these scripts, since it is a seperate discussion from this one (this should be more appropriately the usage and support of the aforementioned scripts).

You can see the results here. The very short form I'm afraid to say is that I did not get positive results with Acovea flags. Read the topic for details.

-Twist
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Gentoo Chat All times are GMT
Goto page Previous  1, 2, 3 ... , 14, 15, 16  Next
Page 15 of 16

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum