OS2 World Community Forum

OS/2, eCS & ArcaOS - Technical => Hardware => Topic started by: Dariusz Piatkowski on October 24, 2017, 09:31:45 pm

Title: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Dariusz Piatkowski on October 24, 2017, 09:31:45 pm
Folks,

I just installed the AN release of the SNAP video drivers here. What I found was both exhilarating and somewhat dissapointing (well, just a tad bit, but read on), so I thought I'd share my experience.

My findings relate to the differences between the IBM/SciTech SNAP drivers (the latest publicly available build being 3.1.8 I believe) and the current SNAP drivers that are shipping with the ANOS, that being 'Build No 506'.

Alright, so ATI X850 XT PE video card here, driving my Samsung 245T panel at 1900x1200 resolution.

OK, good, heck GREAT stuff first. I used to suffer from that annoying OS/2 version of the "Blue Screen of Death"...LOL, meaning, as the GUI comes up and WPS is starting to initialize my system would simply hang. Not a hard hang, CTL-ALT-DEL would allow a re-boot w/o a chkdsk. This would only happen if I ran all 6 cores of my AMD Phenom CPU enabled. Since that made my box un-usable I disabled the 6th core in BIOS and simply used a crippled system with 5 cores only. But hey, at least it worked!!!

Well, I am thrilled to report this issue is now GONE...no longer a problem, I can routinely re-boot using the full 6 cores, no WPS start-up hang, all good! This is a great result. My understanding is that the SNAP drivers had some SMP related issues addressed, so if this was one of them it certainly did the trick.

Alright...so what's the "problematic" piece to report on?

Hmm, the video performance is down. Not sure why that is the case, I will log a ticket for the SNAP team and point to a few things I'm seeing, but in short, take a look at the before & after Sysbench 0.9.5d results:

1) old IBM/SciTech SNAP drivers

Graphics
   BitBlt S->S copy      :     9175.950    Million pixels/second
   BitBlt M->S copy      :      293.762    Million pixels/second
   Filled Rectangle      :    22804.327    Million pixels/second
   Pattern Fill          :     3514.758    Million pixels/second
   Vertical Lines        :      173.654    Million pixels/second
   Horizontal Lines      :     1342.758    Million pixels/second
   Diagonal Lines        :       69.059    Million pixels/second
   Text Render           :      853.040    Million pixels/second
   -----------------------------------------------------------------------
   Total                 :     1145.656    PM-Graphics-marks

 Direct Interface to video extensions - DIVE
   Video bus bandwidth   :     1123.420    Megabytes/second
   DIVE fun              :     3834.609    fps normalised to 640x480x256
   M->S, DD,   1.00:1    :     3831.759    fps normalised to 640x480x256
   -----------------------------------------------------------------------
   Total                 :     1432.589    DIVE-marks



2) new AN SNAP drivers

Graphics
   BitBlt S->S copy      :    12437.376    Million pixels/second
   BitBlt M->S copy      :       58.650    Million pixels/second
   Filled Rectangle      :    46704.301    Million pixels/second
   Pattern Fill          :     3507.515    Million pixels/second
   Vertical Lines        :      173.718    Million pixels/second
   Horizontal Lines      :     1267.768    Million pixels/second
   Diagonal Lines        :       57.530    Million pixels/second
   Text Render           :      859.766    Million pixels/second
   -----------------------------------------------------------------------
   Total                 :     1033.508    PM-Graphics-marks

 Direct Interface to video extensions - DIVE
   Video bus bandwidth   :      223.803    Megabytes/second
   DIVE fun              :      763.788    fps normalised to 640x480x256
   M->S, DD,   1.00:1    :      763.845    fps normalised to 640x480x256
   -----------------------------------------------------------------------
   Total                 :      285.501    DIVE-marks



In particular, the 'BitBlt M->S copy' and DIVE performance is severely down. In terms of application impact that is making my FF so slow that I can literaly see the page re-draw...OK, we are not talking here line at a time, but you can see the refresh whereas before it was lighting fast.

The drivers claim that write combining is enabled but in the GRAPHICS.LOG I see there is a failure report of some type, so there is something here that probably needs further review.

That's all for now. I would love to hear from anyone else who's using the AN SNAP drivers on their system and what you are seeing?
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Roderick Klein on October 24, 2017, 10:33:31 pm
What is the error in graphics.log ?
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Dariusz Piatkowski on October 24, 2017, 10:48:37 pm
Hi Roderick,

What is the error in graphics.log ?

Here is a quick snippet, but I've attached the full log to the post as well:

GA_enumerateDevices: Found 1 PCI/AGP display devices

Global options for all devices:
  Force VBE Fallback ...... Off
  Force VGA Fallback ...... Off
  Allow non-certified ..... On
  Disable write combining . Off
  Use BIOS for LCD panel... Auto
  Shared AGP memory size... 4096 KB
  Use system memory driver. Off
  Disable DDC detection.... Off
  Virtual Display.......... Off

Loading driver for device 0 (radeon.drv)
---------------------------------------------------------

Partially mapped graphics memory: 0x08000000 instead of 0x10000000 bytes
You may not see all available memory due to memory mapping limitations

Attempting to enable write combining, base = 0xD0000000, length = 0x08000000 ... Success!
Attempting to disable caching for MMIO, base = 0xFE9F0000, length = 0x00004000 ...
Failed: UNKNOWN ERROR!
Loading chipset filters...done.
Loading splash screen...done.

Graphics device configuration:
  Manufacturer......... ATI
  Chipset.............. Radeon X850 Series
  Bus Type............. PCI Express
  Memory............... 131072 KB
  DAC.................. ATI Internal 24 bit DAC
  Clock................ ATI Internal Clock
  Memory Clock......... 400 MHz
  Default Memory Clock. 400 MHz
  Maximum Memory Clock. 400 MHz
  Driver Revision...... 3.2, Build 29
  Driver Build......... May  2 2017
  Certified Version.... Not Certified

Graphics device options:
  Invert .................. Off
  Rotation ................ Off
  Flipped ................. Off
  Prefer 16 bit per pixel.. On
  Prefer 32 bit per pixel.. On
  PCI bus mastering........ On
  Hardware acceleration.... Full


Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Dariusz Piatkowski on October 24, 2017, 10:51:32 pm
...and here is the matching section from the IBM_SNAP driver initialization:

GA_enumerateDevices: Found 1 PCI/AGP display devices

Global options for all devices:
  Force VBE Fallback ...... Off
  Force VGA Fallback ...... Off
  Allow non-certified ..... On
  Disable write combining . Off
  Use BIOS for LCD panel... Auto
  Shared AGP memory size... 4096 KB
  Use system memory driver. Off
  Disable DDC detection.... Off
  Virtual Display.......... Off

Loading driver for device 0 (radeon.drv.Sep.25.2006.13.01.08)
---------------------------------------------------------

Partially mapped graphics memory: 0x08000000 instead of 0x10000000 bytes
You may not see all available memory due to memory mapping limitations

Attempting to enable write combining, base = 0xD0000000, length = 0x08000000 ... Success!
Attempting to disable caching for MMIO, base = 0xFE9F0000, length = 0x00004000 ... Success!
Loading chipset filters...done.
Loading splash screen...done.

Graphics device configuration:
  Manufacturer......... ATI
  Chipset.............. Radeon X850 Series
  Bus Type............. PCI Express
  Memory............... 131072 KB
  DAC.................. ATI Internal 24 bit DAC
  Clock................ ATI Internal Clock
  Memory Clock......... 400 MHz
  Default Memory Clock. 400 MHz
  Maximum Memory Clock. 400 MHz
  Driver Revision...... 3.2, Build 29
  Driver Build......... Sep 25 2006
  Certified Version.... 1.60
  Certified Date....... Sep 25 2006

Graphics device options:
  Invert .................. Off
  Rotation ................ Off
  Flipped ................. Off
  Prefer 16 bit per pixel.. On
  Prefer 32 bit per pixel.. On
  PCI bus mastering........ On
  Hardware acceleration.... Full

Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Dave Yeo on October 25, 2017, 05:58:05 am
Out of curiosity, I tested on my old T42 with ATI Mobility Radeon 7500 and Pentium M 1.6Ghz
eCS with Snap build for T42, Ver 2.3, build 419.

  Graphics
   BitBlt S->S copy      :      204.195    Million pixels/second
   BitBlt M->S copy      :       30.545    Million pixels/second
   Filled Rectangle      :      927.106    Million pixels/second
   Pattern Fill          :      927.278    Million pixels/second
   Vertical Lines        :       12.047    Million pixels/second
   Horizontal Lines      :      451.019    Million pixels/second
   Diagonal Lines        :       14.393    Million pixels/second
   Text Render           :      242.816    Million pixels/second
   ---------------------------------------------------------------
   Total                 :      133.654    PM-Graphics-marks
 Direct Interface to video extensions - DIVE
   Video bus bandwidth   :      167.531    Megabytes/second
   DIVE fun              :      572.116    fps normalised to 640x480x256
   M->S, DD,   1.00:1    :      571.994    fps normalised to 640x480x256
   -----------------------------------------------------------------------
   Total                 :      213.790    DIVE-marks


ArcaOS with Snap v3.19, build local (506)

Graphics
   BitBlt S->S copy      :      205.285    Million pixels/second
   BitBlt M->S copy      :       42.508    Million pixels/second
   Filled Rectangle      :      933.017    Million pixels/second
   Pattern Fill          :      931.745    Million pixels/second
   Vertical Lines        :       12.027    Million pixels/second
   Horizontal Lines      :      481.276    Million pixels/second
   Diagonal Lines        :       15.136    Million pixels/second
   Text Render           :      238.063    Million pixels/second
   ---------------------------------------------------------------
   Total                 :      141.224    PM-Graphics-marks

 Direct Interface to video extensions - DIVE
   Video bus bandwidth   :      166.894    Megabytes/second
   DIVE fun              :      569.617    fps normalised to 640x480x256
   M->S, DD,   1.00:1    :      569.564    fps normalised to 640x480x256
   -----------------------------------------------------------------------
   Total                 :      212.895    DIVE-marks


For reference, ArcaOS with Panorama, 16 bit depth (others at 32 bit),

Graphics
   BitBlt S->S copy      :       69.600    Million pixels/second
   BitBlt M->S copy      :       70.938    Million pixels/second
   Filled Rectangle      :      145.558    Million pixels/second
   Pattern Fill          :      102.522    Million pixels/second
   Vertical Lines        :       10.434    Million pixels/second
   Horizontal Lines      :       27.892    Million pixels/second
   Diagonal Lines        :        1.011    Million pixels/second
   Text Render           :       42.929    Million pixels/second
   ---------------------------------------------------------------
   Total                 :       31.340    PM-Graphics-marks

 Direct Interface to video extensions - DIVE
   Video bus bandwidth   :      167.064    Megabytes/second
   DIVE fun              :      570.280    fps normalised to 640x480x256
   M->S, DD,   1.00:1    :      570.440    fps normalised to 640x480x256
   -----------------------------------------------------------------------
   Total                 :      213.187    DIVE-marks


As you can see, the ArcaOS supplied Snap was slightly faster, though this was basically a virgin install. DIVE seems consistent, eve with Panorama.

I also accidentally ran the Snap test on ArcaOS at 16 bit depth, pretty well doubled the score excepting DIVE.

Graphics
   BitBlt S->S copy      :      536.218    Million pixels/second
   BitBlt M->S copy      :       81.518    Million pixels/second
   Filled Rectangle      :     2308.661    Million pixels/second
   Pattern Fill          :     2306.782    Million pixels/second
   Vertical Lines        :       24.461    Million pixels/second
   Horizontal Lines      :      948.665    Million pixels/second
   Diagonal Lines        :       18.745    Million pixels/second
   Text Render           :      242.202    Million pixels/second
   ---------------------------------------------------------------
   Total                 :      265.277    PM-Graphics-marks

Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: ivan on October 25, 2017, 05:55:28 pm
A little bit off topic but has anyone tried to use snap or panorama to drive a hdmi monitor? 

Several video cards now have DVI-D and HDMI outputs and no VGA I was wondering if it was possible to have a hdmi  monitor driven by one of those cards.
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Pete on October 25, 2017, 08:04:54 pm
Hi Dariusz

Using an ATI X550 with the AN SNAP I see the same error in the graphics.log

Loading driver for device 0 (radeon.drv)
---------------------------------------------------------

Attempting to enable write combining, base = 0xD0000000, length = 0x08000000 ... Success!
Attempting to disable caching for MMIO, base = 0xFEA30000, length = 0x00004000 ...
Failed: UNKNOWN ERROR!
Loading chipset filters...done.
Loading splash screen...done.


Compared with the eCS SNAP 3.1.8 build 505 (log section pulled from an old backup) on eCS2.2beta2

Loading driver for device 0 (radeon.drv.Sep.25.2006.13.01.08)
---------------------------------------------------------

Attempting to enable write combining, base = 0xD0000000, length = 0x08000000 ... Success!
Attempting to disable caching for MMIO, base = 0xFEA30000, length = 0x00004000 ... Success!
Loading chipset filters...done.
Loading splash screen...done.


I can only guess that whatever has been changed within SNAP causes the error.

Just for fun I ran the sysbench tests:-

 Graphics
   BitBlt S->S copy      :      430.345    Million pixels/second
   BitBlt M->S copy      :      305.949    Million pixels/second
   Filled Rectangle      :     7580.586    Million pixels/second
   Pattern Fill          :     1778.032    Million pixels/second
   Vertical Lines        :       56.506    Million pixels/second
   Horizontal Lines      :      223.633    Million pixels/second
   Diagonal Lines        :       51.134    Million pixels/second
   Text Render           :      602.398    Million pixels/second
   -----------------------------------------------------------------------
   Total                 :      402.639    PM-Graphics-marks


 Direct Interface to video extensions - DIVE
   Video bus bandwidth   :     1175.934    Megabytes/second
   DIVE fun              :     4016.503    fps normalised to 640x480x256
   M->S, DD,   1.00:1    :     4016.782    fps normalised to 640x480x256
   -----------------------------------------------------------------------
   Total                 :     1501.114    DIVE-marks


Not overly impressive - especially the PM-Graphics-marks

Sorry, I do not plan on installing the older SNAP with SMP problems to see what the differences are. ArcaOS with AN SNAP seems a little slower than eCS2.2beta2 with eCS SNAP *but* the AN SNAP has, so far, always booted to a working Desktop unlike previous versions of SNAP which are very prone to hanging when the Desktop is starting.



Regards

Pete
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Dave Yeo on October 26, 2017, 03:56:12 am
Looking further, I see I also get the same error about disabling caching for mmio with the ArcaOS supplied Snap, but not the older one. Though it doesn't seem to affect performance for me.
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Dariusz Piatkowski on October 26, 2017, 04:35:01 am
Yeah, I am not sure what the MMIO is about, but here at least there is a very real performance impact.

Sure, when comparing across different configurations what I am seeing in my 'slow' setup may not seen all that bad...however, the old IBM SNAP driver was pretty much able to keep up with my Win7Pro box which happens to have the very same Phenom II X6 cpu and ATI X850 XT PE video card in there (but in a different motherboard), so a fair comparison, and I for one was always impressed that my old OS/2 box was holding up to the Win stuff.

In my troubleshooting I did investigate the potential aspect of having 6 CPU cores enabled vs the old 5 core setup, this however had no impact on driver loading, same LOG entry, same performance results (+/- the CPU extra power contribution).

Anyways, I did log a ticket for the AN/SNAP folks, see => https://mantis.arcanoae.com/view.php?id=1505 (https://mantis.arcanoae.com/view.php?id=1505), the Mantis system appears to be down at the moment (I can not reach it) but I do not suspect they will be able to look at this anytime soon, I think Rich is busy on some type of dynamic resolution changing process (I do not know this first-hand, just picking up the stuff from the "ether"...LOL).

Alright, thanks for the feedback you guys. The next step on my search for the "Holy Grail" was to re-try the dual-head configuration. The AN SNAP drivers though still are running into the same problem as the IBM ones, which is declaring the monitor EDID to be broken, subsequently, as best as I can tell only the 'Head 0' drivers get loaded. This will have to wait.

ivan,
My Samsung 245T panel has HDMI connection on it, but I have not tried it. Will give it a go tomorrow and report-out. As it happens, playing with the SNAP drivers hasn't been the most fun event, I did manage to mess up my INI files badly enough that I needed to recover from a backup to get back to a working PMMAIL, FIREFOX and PSI install. Not sure what it was, however, PSI would produce a trap entry showing a SYS3175 in REGISTRY.DLL.

Ah, one more thing, given that we are speaking of other video driver/hardware related stuff: who has a working dual-head configuration? What are you using, can you post your GRAPHICS.LOG contents?

I am asking this last question because the SNAP readme clearly states that even the ATI hardware will only support DVI for a single display and anything else needs to use RGB. Indeed, my ATI card has two DVI connectors, I ran DVI to my primary but used a DVI=>RGB adapter to my secondary. I did end up seeing the following in my GRAPHICS.LOG file, however the 'head 1' panel never produced anything on-screen.

Configuration for head 0:

Monitor configuration:
  Manufacturer... Samsung
  Model.......... SyncMaster
  Max Resolution. 1920x1440
  Max HScan...... 81 KHz
  Max VScan...... 75 Hz
  Features....... DPMS Wide

LCD panel detected:
  Width:  1920
  Height: 1200

Graphics mode information (DFP):


...and the matching entry for head 1:

Configuration for head 1:

Monitor configuration:
  Manufacturer... Samsung
  Model.......... SyncMaster
  Max Resolution. 1920x1440
  Max HScan...... 81 KHz
  Max VScan...... 75 Hz
  Features....... DPMS Wide

Graphics mode information (CRT):
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Olafur Gunnlaugsson on October 26, 2017, 12:59:50 pm
A little bit off topic but has anyone tried to use snap or panorama to drive a hdmi monitor? 

Several video cards now have DVI-D and HDMI outputs and no VGA I was wondering if it was possible to have a hdmi  monitor driven by one of those cards.

I have a monitor that uses a DVI to HDMI converter cable, using an old Dell X600PRO card via DVI output and Snap 3.18.

Panorama should work with the HDMI output if the VESA BIOS on the card or motherboard supports it, most should.
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Neil Waldhauer on October 26, 2017, 06:20:05 pm
A little bit off topic but has anyone tried to use snap or panorama to drive a hdmi monitor? 

Several video cards now have DVI-D and HDMI outputs and no VGA I was wondering if it was possible to have a hdmi  monitor driven by one of those cards.

I have a monitor that uses a DVI to HDMI converter cable, using an old Dell X600PRO card via DVI output and Snap 3.18.

Panorama should work with the HDMI output if the VESA BIOS on the card or motherboard supports it, most should.

I was going to post this. It depends on the computer (or graphics card) BIOS.

Also, if you have already thrown your HDMI in the trash to use DisplayPort, then that works on my Lenovo M93p tiny with eComStation and Panorama graphics.
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Pete on October 26, 2017, 08:00:05 pm
Hi Dariusz

Where is sysbench 0.9.5d available? - I only see 0.9.5c on http://www.os2warp.org/Sysbench/


Regards

Pete

Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Dariusz Piatkowski on October 26, 2017, 11:08:37 pm
Hey Pete!

...Where is sysbench 0.9.5d available? - I only see 0.9.5c on http://www.os2warp.org/Sysbench/...
Hmm...good question...I remember working with Trevor a few years back on some fixes, may have had something to do with either my HPFS386 fs, or maybe it was CPU speed related, so I'm thinking it may have been a test release version? Although I thought he did eventually make that available...I could have sworn I've see it out in the "wild"...LOL!

I looked for the ZIP on my system (in case I saved it) but I did not find it. I can ceratinly ZIP-up the contents of the directory and send it your way...I do not think there was a specific install program...
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Martin Iturbide on October 27, 2017, 03:11:50 pm
Hi

I have no records of sysbench 0.9.5d on the OS2World news archive, last one I have news from was 0.9.5c on May 2008 . If someone has it please upload it to hobbes.

Regards
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: DR34264 on October 27, 2017, 04:01:06 pm
SYSBENCH095D:
http://www.os2warp.org/sysbench/sysb095d.zip
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Neil Waldhauer on October 28, 2017, 03:30:49 pm
Looking at the DIVE results, is there something wrong with the measurement? I measured the M93p tiny

Code: [Select]
Direct Interface to video extensions - DIVE
   Video bus bandwidth   :     7282.561    Megabytes/second
   DIVE fun              :    27175.091    fps normalised to 640x480x256
   M->S, DD,   1.00:1    :    25132.707    fps normalised to 640x480x256
   -----------------------------------------------------------------------
   Total                 :     9516.640    DIVE-marks

Is it really that much faster than an accelerated ATI card?
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Dave Yeo on October 28, 2017, 06:23:39 pm
Probably, here I get the same DIVE speed with Panorama and accelerated Snap, so I guess it is just memory speed.
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Dariusz Piatkowski on October 29, 2017, 02:44:19 am
Probably, here I get the same DIVE speed with Panorama and accelerated Snap, so I guess it is just memory speed.

...I did a whole wack more testing, consistently though I found that even though the graphics.log entries claim that write combining is turned on the performance tells a different story.

So just to be sure I disabled write combine with "gaoptions", re-booted, re-ran the test. The Sysbench results are the same, so on my system at least there is no difference...and that implies that whatever the drivers think they are doing to successfully enable write combining, it is not actually getting done.

Here is something else to consider, I am having a hell of a time with the CONFIG.SYS VIRTUALADDRESSLIMIT entry. Previously with the SciTech drivers I used to run VIRTUALADDRESSLIMIT=3072, with this same setting I can not boot to a workable system, meaning, I continue to get the following error message whenever I attempt to start most of my programs:

LIBC Error: Couldn't register process in shared memory!

In particular, my networking is down entirely, like it doesn't even start up, "ping" goes nowhere, etc, etc.

Lowering the value to VIRTUALADDRESSLIMIT=2048 finally resulting in a working system. Now the graphics.log actually shows the full 256MB of video card ram, previously it only recognizes 128MB and complains that all of memory may not be available.

I'm curious, what are you guys setting your VIRTUALADDRESSLIMIT to?

Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Dariusz Piatkowski on October 29, 2017, 02:49:33 am
Hey Neil!

Looking at the DIVE results, is there something wrong with the measurement? I measured the M93p tiny

Code: [Select]
Direct Interface to video extensions - DIVE
   Video bus bandwidth   :     7282.561    Megabytes/second
   DIVE fun              :    27175.091    fps normalised to 640x480x256
   M->S, DD,   1.00:1    :    25132.707    fps normalised to 640x480x256
   -----------------------------------------------------------------------
   Total                 :     9516.640    DIVE-marks

Is it really that much faster than an accelerated ATI card?

OMG!!!...seriously, those M->S values are "to die for"...that's smoking fast, what are the non-DIVE test results?

My big complaint is that with such poor M->S values, on my 1920x1200 display I am literally seeing the Firefox (as well as other apps) frame window draw first and I can actually catch the inside of the window being drawn...sure it goes by fast, but the fact that I can actually preceive it is very distracting...so can not be faster then 1/30 secs. given typical human visual perception.

The old SciTech drives produced instant udpates, in my cases, the matching M->S scores were nearly 5-6x faster.

Neil, what video hardware are you running? Maybe it is time for me to drop my trusty ATI X850 XT PE card and move on? LOL
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Dave Yeo on October 29, 2017, 03:10:38 am
Quote
I'm curious, what are you guys setting your VIRTUALADDRESSLIMIT to?

While under eCS 2.x, I always used 3072, I found that made ArcaOS unstable. Using 2816 the OS is stable but both SeaMonkey and Thunderbird are very crashy, often just vanishing, sometimes writing a trp file with the crashes in the JavaScript engine, rather allocating memory or doing garbage collection. I'm going to decrease it to 2560 and keep decreasing by 0x100H perhaps to your 2048.
Today I decided to reboot to eCS 2.1, where the VIRTUALADDRESSLIMIT is still set to 3072 and SeaMonkey and Thunderbird have been stable.
The main difference seems to be the kernel, so I guess I'll also try downgrading the kernel later.
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: OS4User on October 29, 2017, 07:38:38 am
To Neil:


Code: [Select]
Direct Interface to video extensions - DIVE
   Video bus bandwidth   :     7282.561    Megabytes/second

Is it really that much faster than an accelerated ATI card?

This VBB value is typical for latest Intel built-in video with a proper set of caching mode for LFB.
We need somebody with such processor and ATI card  to know what is  faster (I am curios too).
And it will be necessary to test this issue on real  applications not only Sysbench. I am using SNAP with ATI x850 and when I switch my comp to Panorama for tests, I recognize  dramatic decreasing of scrolling  speed (for example, any help file in NewView). Probably, on latest Intel built-in video, such decrease will not happen.

To Dariusz:


LIBC Error: Couldn't register process in shared memory!

I'm curious, what are you guys setting your VIRTUALADDRESSLIMIT to?

Pre Arca SNAP has a bad habit - it is mapping ALL memory available on a video card into address space (if  I am not wrong, even twice).

It looks absolutely senseless - just a waste of address space.
Fortunately, there is a way to "fix" this. "gaoption vidmem 8" reduces size of mem to be mapped by SNAP to 8M (it is sufficient for 1280x1024x16M)

I always use  VIRTUALADDRESSLIMIT = 3072 (/CACHE:262144  for JFS.IFS)
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Doug Bissett on October 29, 2017, 07:45:36 am
Quote
I'm curious, what are you guys setting your VIRTUALADDRESSLIMIT to?

I have been using 2560, but 3072 seems to work just as well. With less than 2560 (2048), I start to run out of shared memory space, after a couple of days. I do use the CLAMD.EXE, supplied by RPM/YUM, from netlabs-rel, and it uses close to 1 GB of upper shared memory space. Not using ClamD.EXE doesn't change anything, except I can use a smaller number for VIRTUALADDRESSLIMIT.

Quote
LIBC Error: Couldn't register process in shared memory!

This error seems to have nothing to do with VIRTUALADDRESSLIMIT. It happens when I load DLLs high (I can't even run my system, if I don't do that with AOO, and FF). Take either one of those programs, open, and close, any combination, twice, and I start to get that message with anything that uses LIBC. Other things keep on working properly, but it takes a re-boot to get the problem solved (temporarily).

FWIW, I use the Quick Start feature with AOO, and that helps because it keeps the DLLs loaded, rather than unloading them, witch causes the problem. Firefox is another story. It will either crash, or I need to close it, at least once per day, so the longest that I can go, without a re-boot for the LIBC problem, is 2 days.

FWIW, I had to quit using eCS, because simply closing a program, with DLLs loaded high, would crash the system, about every second try, and trying to run AOO, and FF, at the same time, using DLLs loaded low,  was impossible. At least the new ArcaOS kernel doesn't crash the system, when the problem happens.
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Pete on October 29, 2017, 04:10:21 pm
Hi All

Looking in my config.sys I see the ArcaOS installation set VIRTUALADDRESSLIMIT=1536 which seems to work fine here - however, I do not leave Seamonkey running when I am not using it and I never use the AOO QuickStart feature.

On my eCS2.1 and 2.2b2 installations on the same hardware I see that I have VIRTUALADDRESSLIMIT=3072. I think it may be something to do with the AOO Draw app - will have to test how that works in ArcaOS.


Regards

Pete
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Dariusz Piatkowski on October 29, 2017, 05:04:56 pm
Hey OS4User,

To Dariusz:


LIBC Error: Couldn't register process in shared memory!

I'm curious, what are you guys setting your VIRTUALADDRESSLIMIT to?

Pre Arca SNAP has a bad habit - it is mapping ALL memory available on a video card into address space (if  I am not wrong, even twice).

It looks absolutely senseless - just a waste of address space.
Fortunately, there is a way to "fix" this. "gaoption vidmem 8" reduces size of mem to be mapped by SNAP to 8M (it is sufficient for 1280x1024x16M)

I always use  VIRTUALADDRESSLIMIT = 3072 (/CACHE:262144  for JFS.IFS)

Alright, happy to give this a go, let me explain my calculation, hopefully no mistakes:

So (for your 1280x1024 resolution):
1) 1280x1024 = 1310720 pixels
2) 16M colours means 32bit colour depth, that's 4 bytes/pixel
3) 1310720 pixels * 4 bytes/pixel = 5242880 bytes, which is approx 5 MB

If you assume 24bit colour depth, thus 3 bytes/pixel, the above result changes to 4 MB, am I understanding this correctly?

If yes, than the appropriate sizing for my 1920x1200 @ 32bit colour depth is:
1) 1920x1200 = 2304000 pixles
2) 32 bit colour depth, that's 4 bytes/pixel
3) 2304000 pixels * 4 bytes/pixel = 9216000 bytes, which is approx 9 MB

Since I ultimately want to have this setup as a dual-head display I will double up the 9MB allocation and give it a little extra 'room'...therefore, the required SNAP command is: 'gaoption vidmem 24', which is what I executed...so let's see what a re-boot brings??? lol
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Dariusz Piatkowski on October 29, 2017, 05:15:32 pm
Hey Doug!

Quote
I'm curious, what are you guys setting your VIRTUALADDRESSLIMIT to?

I have been using 2560, but 3072 seems to work just as well. With less than 2560 (2048), I start to run out of shared memory space, after a couple of days. I do use the CLAMD.EXE, supplied by RPM/YUM, from netlabs-rel, and it uses close to 1 GB of upper shared memory space. Not using ClamD.EXE doesn't change anything, except I can use a smaller number for VIRTUALADDRESSLIMIT...

Yeah, similar to how I got to my 3072 setting. I used to run FF full-time, multiple windows, multiple tabs...and I never shut down my machine, up 24x7, or at least as long as OS/2 itself is willing to stay up and running. I found that the higher I cranked this value up the more stable the system would be.

Quote
LIBC Error: Couldn't register process in shared memory!
...This error seems to have nothing to do with VIRTUALADDRESSLIMIT. It happens when I load DLLs high (I can't even run my system, if I don't do that with AOO, and FF). Take either one of those programs, open, and close, any combination, twice, and I start to get that message with anything that uses LIBC. Other things keep on working properly, but it takes a re-boot to get the problem solved (temporarily).

FWIW, I use the Quick Start feature with AOO, and that helps because it keeps the DLLs loaded, rather than unloading them, witch causes the problem....

My only experience with the impact this value has had is due to AN SNAP driver change. I can absolutely confirm that the only way for me to get a working environment (other than WPS coming up fine and some (most?) of the core OS/2 stuff working) is to lower this value. Subsequently, for now, the only way I can have a working system is by setting this to 2048.

Worth noting is the following:

1) I run HPFS386 with a 64M cache
2) my ATI X850 XT card comes equipped with 256M ram, which apparently is causing the drivers to consume that much of the actual system memory - making the change to limit this to 24M, re-boot coming up next
3) AOO Quick Start feature is executed upon start-up
4) I have about 19 apps in my XWP start-up folder, most of these are utility type things, such as: CPUMonitor, lSwitcher, dSync, Win 95 Key, SetTime, Xit, CAD-Popup, UPS Monitor, DskMon, Web/2, PSI, Privoxy, etc

Not sure if that makes for a 'heavy load' on a system, but I would hope not since most of these are small memory foot-print type apps.
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Andi B. on October 29, 2017, 05:52:54 pm
There's
Code: [Select]
SET SNAP_MAXVRAM_32MB=Y
in my config.sys. Pretty sure the SNAP documentation explains this.
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: OS4User on October 29, 2017, 06:17:33 pm
There's
Code: [Select]
SET SNAP_MAXVRAM_32MB=Y
in my config.sys. Pretty sure the SNAP documentation explains this.

according to SNAP documentation  SET SNAP_MAXVRAM_32MB=Y is not the same with "gaoption vidmem 32"

what is mentioned in your C:\OS2\DRIVERS\SNAP\CONFIG\GRAPHICS\graphics.log ?
Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: Dariusz Piatkowski on October 30, 2017, 01:55:08 am
Andi,

There's
Code: [Select]
SET SNAP_MAXVRAM_32MB=Y
in my config.sys. Pretty sure the SNAP documentation explains this.

...but as OS4User pointed out, and I agree, I think these two ways of controlling the video memory allocation are different.

I did both, the result is that I see the following in graphics.log:

Video memory limited to user supplied value of 32MB

Attempting to enable write combining, base = 0xD0000000, length = 0x02000000 ... Success!
Attempting to disable caching for MMIO, base = 0xFE9F0000, length = 0x00004000 ...
Failed: UNKNOWN ERROR!
Loading chipset filters...done.
Loading splash screen...done.


The Screen dialog now only shows 32Meg as the total memory for the video card, and CLI 'gaoption show' gives this:

Global options for all devices:

  Force VBE Fallback ...... Off
  Force VGA Fallback ...... Off
  Allow non-certified ..... On
  Disable write combining . Off
  Use BIOS for LCD panel... Auto
  Video Memory Limit....... 32 Mb
  Shared AGP memory size... 4096 Kb
  Use system memory driver. Off
  Disable DDC detection.... Off
  Enable AGP FastWrite..... Off
  Maximum AGP data rate.... 8X
  Virtual Display.......... Off


Bottom line here is that:
1) I can successfully boot with VIRTUALADDRESSLIMIT = 3072

2) ...but that made me greedy, because I miss my bigger HPFS386 cache, so I went back to playing with the sizing and was able to increase the cache size from 64M to 128M however I needed to drop to VIRTUALADDRESSLIMIT=2048 to make that happen

At this point in time I think I got the hang of the SNAP driver's memory allocation scheme and how it impacts the remainder of the system memory.

Title: Re: SNAP drivers and video speed - IBM vs AN release differences...
Post by: OS4User on October 30, 2017, 07:47:01 am
I did both, the result is that I see the following in graphics.log:

from snapos2.pdf:
Code: [Select]
SET SNAP_MAXVRAM_32MB=Y
...
Alternatively, you can use the gaoption “vidmem” option to not only limit the amount
of video memory reported, but also the amount mapped into the physical memory space.

Here is the mistake  - it has to be written into the virtual memory space but not into physical memory space, but the rest looks like true.

Indirect confirmation is in your log :
Code: [Select]
Attempting to enable write combining, base = 0xD0000000, length = 0x02000000 ... Success!So the set of LFB aperture is indeed 0x02000000 bytes =  32M.

I believe you will have the same even without SET SNAP_MAXVRAM_32MB=Y, but with “vidmem” only.

2) ...but that made me greedy, because I miss my bigger HPFS386 cache

You can use Theseus to analyze free System Memory to set HPFS386 cache as big as possible.
(System->Kernel Information->System Object Summary)

I have about 180M free system memory after system booted   - it looks sufficient.