AstroPulse for ATI GPUs released

[Raistmer]

ATI/AMD GPU-only and HYBRID GPU+CPU Optimized AstroPulse 5.06 (r449-r456) for Windows OS

Three editions available now at GPU Apps Public Downloads area

First of all: This release intended for ADVANCED USERS. If you not sure you can repair your BOINC installation if something will go wrong - wait when this app will be included in Lunatics installer.

There are 3 different builds suitable for different types of ATI/AMD hardware.
1) OpenCL based application that can be used on HD4xxx/5xxx series of ATI GPUs.
2) OpenCL/Brook+ combo app that can be used on same GPU classes (HD4xxx/HD5xxx) but maybe provide different (better) performance/experience for some host setups
3) Brook+ GPU/CPU "hybrid" build - it suitable for older ATI GPUs and on-board chips starting from HD2xxx seria. It's update release for released before Hybrid AP rev393.

Requirements:
1), 2):
ATI GPU 4xxx or 5xxx, ATI Stream SDK2.2 installed, CPU with SSE2 and better support, Catalyst 10.7b drivers.
[
 for some hosts with only HD4xxx GPU installed additional environment variable may be needed: GPU_MAX_HEAP_SIZE=256
To set it go My computer->Properties->Advanced->Environment variables (path can be slightly different in different windows versions). But before setting such variable run application and check stderr.
If you see line:
  Global memory size:            
with value >128MB - you don't need this environment variable
]
You can download SDK here: http://developer.amd.com/gpu/ATIStreamSDK/Pages/default.aspx

Release notes for 1):
There are 2 command line switches that can be used for app performance tuning:

-ffa_block 8192 (default value) - defines how many  different periods GPU will process per single kernel call
-ffa_block_fetch 2048 (default value) - defines how many threads will be used in FFA initial fetch kernel
Rules for using these values:
-ffa_block_fetch <number> can be used only if -ffa_block <number> already listed in command line
numbers should be even,better if they will be power of 2, ffa_block should be divisible by ffa_block_fetch.
If you experience lags during application execution try to decrease these values.

 -disable_slot <slot_number>
It should allow to exclude broken device even if BOINC assigns task to it. For better use it's worth to set count value in coproc section from 1 to 2 (or as appropriate) to limit number of tasks launched by BOINC. If all GPUs work OK no need to use this option.

Known issues::

1) App will not work (no progress on secondary chip at all or will produce invalid results) on second core of dual-core GPUs like 5970 and will produce invalid results(or no progress at all) with CrossFire enabled on non-primary GPUs in system.
It's ATI SDK2.2 related bug, ask ATI for fix.

2) It can blink with popup console windows at each new task start. Again, it's long standing bug with ATI SDK: http://forums.amd.com/devforum/messageview.cfm?catid=390&threadid=136672&enterthread=y
Ask AMD for fix.
EDIT: this bug was solved in Cat 10.12, update drivers.

Requirements for 3) :
ATI GPU 2xxx or better, Catalyst 9.x or better drivers


To get any AP tasks for GPU don't forget to enable "Send work from other apps" on web preferencies page
app_info.xml section for this app (for testing on main):

app_info.xml file sections for using these apps:
1), 2): ( for 2) don't forget to change executable file name in provided sample. Sample can be used as is only for 1) app )

Code: [Select]

<app>
<name>astropulse_v505</name>
</app>
<file_info>
        <name>ap_5.06_win_x86_SSE2_OpenCL_ATI_r456.exe</name>
        <executable/>
    </file_info>
<file_info>
    <name>AstroPulse_Kernels.cl</name>
    <executable/>
</file_info>
    <app_version>
        <app_name>astropulse_v505</app_name>
        <version_num>506</version_num>
<avg_ncpus>0.01</avg_ncpus>
<max_ncpus>0.01</max_ncpus>
                <plan_class>ati13ati</plan_class>
                <cmdline>-ffa_block 8192 -ffa_block_fetch 2048</cmdline>
          <coproc>
              <type>ATI</type>
        <count>1</count>
          </coproc>
        <flops>15987654321</flops>

        <file_ref>
            <file_name>ap_5.06_win_x86_SSE2_OpenCL_ATI_r456.exe</file_name>
            <main_program/>                           
        </file_ref>
<file_ref>
    <file_name>AstroPulse_Kernels.cl</file_name>
    <copy_file/>
</file_ref>
    </app_version>
3)

Code: [Select]

<app>
<name>astropulse_v505</name>
</app>
<file_info>
        <name>ap_5.06_x86_win_SSE_BROOK_NO_DOUBLE_r453.exe</name>
        <executable/>
    </file_info>
    <app_version>
        <app_name>astropulse_v505</app_name>
        <version_num>505</version_num>
<avg_ncpus>0.1</avg_ncpus>
<max_ncpus>0.1</max_ncpus>
             <plan_class>ati13ati</plan_class>
            <coproc>
                  <type>ATI</type>
              <count>0.5</count>
            </coproc>
           <flops>6000000000</flops>
        <file_ref>
            <file_name>ap_5.06_x86_win_SSE_BROOK_NO_DOUBLE_r453.exe</file_name>
            <main_program/>                           
        </file_ref>
    </app_version>
Please, edit <max/avg_ncpus> and <count> tags values as appropriate for your config.
For hosts participating in other ATI-based projects               <count>0.5</count> can be changed to               <count>0.1</count>

I would like to thank all Lunatics crew for continual and very helpfull support. W/o it this release could not be happen.

[Morten]

2xHD5970:
There has not been any response on AMD Forum, but I have established a "truce" so to speak, so that the percentage of validated task are now almost normal.

By disabling only slot 1, there is no longer a memory leak (both tasks @ around 90MB RAM), and the tasks are using only 2 of the 4 GPUs on the 2x5970.

The only problem remaining is that both tasks will after a random amount of time hang, and I have to suspend/activate tasks for them to progress again.

Is this something that you can look into, or is this SDK 2.2-bound?

M

[Raistmer]

Is this something that you can look into, or is this SDK 2.2-bound?

If you see no error messages in stderr I think nothing can be done for now until new SDK/driver release.

[timiman]

Well done and many thanks to everyone who made this real!!!

Despite the bad omens:
a. problems with 10.8/10.8b drivers on 4870x2 with games (stayed at 10.5a for stability on games) and
b. very difficult to get one AP unit nowadays,

I will give it a try on my 4870x2 very soon and will give any feedback to the crew.

Again, thank you guys!!!

[Raistmer]

a. problems with 10.8/10.8b drivers on 4870x2 with games (stayed at 10.5a for stability on games) and

Look first post carefully - Cat 10.7 listed, not Cat 10.8/10.8b.
I will not say it will not work with Cat 10.8, but recommended one is 10.7

[timiman]

Yes, I saw it.
But because only with 10.5a I have stability in games and possible with the 10.8b ones, I'll give 10.8b a try with games and then try the AP with them.
Unfortunately, the main role of the pc with the 4870x2 is for games, so I have to be careful with the catalysts I install.

[KarVi]

On my rig it works fine with 10.8.

[timiman]

Well, installed Catalysts 10.8b, AMD SDK 2.2, fixed app_info.xml and selected "OpenCL" for starters.

Until SETI is up again (stupid A/C), I have a question on:

To get any AP tasks for GPU don't forget to enable "Send work from other apps" on web preferences page

I already have enabled all three checks on preferences (Seti@Home Enhanced, Astropulse and Astropulse v5),
are you sure that I have to enable also this to get any AP task?
Does this check refer to other projects, out of the Seti@Home project family?

PS. I hope this whole release is not for Seti@Home beta only!

[Claggy]

Well, installed Catalysts 10.8b, AMD SDK 2.2, fixed app_info.xml and selected "OpenCL" for starters.

Until SETI is up again (stupid A/C), I have a question on:

To get any AP tasks for GPU don't forget to enable "Send work from other apps" on web preferences page

I already have enabled all three checks on preferences (Seti@Home Enhanced, Astropulse and Astropulse v5),
are you sure that I have to enable also this to get any AP task?
Does this check refer to other projects, out of the Seti@Home project family?

PS. I hope this whole release is not for Seti@Home beta only!

Yes, you need "Send work from other apps" ticked, as the Astropulse and Astropulse_v5 switches are for obsolete apps,
and there's no Astropulse_v505 switch yet, the switch is in Setiathome Project Preferences, so is only for the Main Setiathome Project.

Claggy

[timiman]

OK then. Waiting for seti servers to "wake up", as the preferences pages are offline right now.
Thank you!

[Fredericx51]

Due to the A/C failiar, it could be friday, before we see any activity from SETI, hoping I'm wrong with this......

But I've todo some 'maintenance' ,  too, computers, they can keep you busy 

Hope they'll look/repair the A/C unit, too and a "Back-Up A/C", comes to mind, too.

[zangetsu]

how long till its included in the lunatics installer? i can probably figure it out but if its an easier way not far away, let me know


ps what is meant by MB unit? can i make the ati gpus run regular seti units that hte cpu does? nvidia seems to crunch them pretty fast. let me know

[timiman]

I just can't wait for downloading a AP workunit to start crunching with my ATi4870x2 !!!

MB stands for MultiBeam, which is the setiathome enchanced, which CPU and Nvidia GPU can crunch.
AP stands for Astropulse, which its workunits can be crucnhed by Raistmer's application on a supported ATi GPU.

[Raistmer]

I just can't wait for downloading a AP workunit to start crunching with my ATi4870x2 !!!

Hope it will go OK. Our beta tester with two-core GPU had lot of issues before found a way to run app more or less smoothly.
Look first page - two-cores GPUs not quite supported (ATi SDK limitation). But worth to try anyway

[timiman]

Well, I have bad news. 
Just downloaded the first one, and after 5 seconds, the computation stops as completed without any errors.
Raistmer, could you guide me to return any logs back to you?

It just downloads AP units one after the other now.

Do these helps?

http://setiathome.berkeley.edu/result.php?resultid=1709396634
http://setiathome.berkeley.edu/result.php?resultid=1709395807