Compiling qemu X64 for OS4.1 (Ryzen) and also other CPU (MAC)

I tried the script
"pegasos2-lame-test"

here is the .log with the OPTIMIZATIONS 03 etc. etc.


*** Running test 1
Loading, please wait...
*** Running lame ***
LAME 3.100 32bits (http://lame.sf.net)
Using polyphase lowpass filter, transition band: 18671 Hz - 19205 Hz
Encoding /file_example_WAV_5MG.wav to <stdout>
Encoding as 44.1 kHz j-stereo MPEG-1 Layer III VBR(q=2)
Frame | CPU time/estim | REAL time/estim | play/CPU | ETA
1149/1149 (100%)| 0:23/ 0:23| 0:23/ 0:23| 1.2691x| 0:00
QEMU 9.0.50 monitor - type 'help' for more information
Accelerator settings:
one-insn-per-tb: off

Translation buffer state:
gen code size 29895987/1073736704
TB count 52713
TB avg target size 24 max=2048 bytes
TB avg host size 329 bytes (expansion ratio: 13.5)
cross page TB count 0 (0%)
direct jump count 31907 (60%) (2 jumps=25822 48%)
TB hash buckets 24463/32768 (74.66% head buckets used)
TB hash occupancy 33.37% avg chain occ. Histogram: [0,10)%|▆ █ ▅▁▃▁▁|[90,100]%
TB hash avg chain 1.018 buckets. Histogram: 1|█▁|2

Statistics:
TB flush count 0
TB invalidate count 7841
TLB full flushes 0
TLB partial flushes 13275
TLB elided flushes 100197
[TCG profiler not compiled]
*** Done
*** Running test 2
Loading, please wait...
*** Running lame ***
LAME 3.100 32bits (http://lame.sf.net)
Using polyphase lowpass filter, transition band: 18671 Hz - 19205 Hz
Encoding /file_example_WAV_5MG.wav to <stdout>
Encoding as 44.1 kHz j-stereo MPEG-1 Layer III VBR(q=2)
Frame | CPU time/estim | REAL time/estim | play/CPU | ETA
1149/1149 (100%)| 0:23/ 0:23| 0:23/ 0:23| 1.2760x| 0:00
QEMU 9.0.50 monitor - type 'help' for more information
Accelerator settings:
one-insn-per-tb: off

Translation buffer state:
gen code size 29916323/1073736704
TB count 52755
TB avg target size 24 max=2048 bytes
TB avg host size 328 bytes (expansion ratio: 13.5)
cross page TB count 0 (0%)
direct jump count 31930 (60%) (2 jumps=25838 48%)
TB hash buckets 24484/32768 (74.72% head buckets used)
TB hash occupancy 33.41% avg chain occ. Histogram: [0,10)%|▆ █ ▅▁▃▁▁|[90,100]%
TB hash avg chain 1.018 buckets. Histogram: 1|█▁▁|3

Statistics:
TB flush count 0
TB invalidate count 7841
TLB full flushes 0
TLB partial flushes 13247
TLB elided flushes 99985
[TCG profiler not compiled]
*** Done
*** Running test 3
Loading, please wait...
*** Running lame ***
LAME 3.100 32bits (http://lame.sf.net)
Using polyphase lowpass filter, transition band: 18671 Hz - 19205 Hz
Encoding /file_example_WAV_5MG.wav to <stdout>
Encoding as 44.1 kHz j-stereo MPEG-1 Layer III VBR(q=2)
Frame | CPU time/estim | REAL time/estim | play/CPU | ETA
1149/1149 (100%)| 0:24/ 0:24| 0:24/ 0:24| 1.2111x| 0:00
QEMU 9.0.50 monitor - type 'help' for more information
Accelerator settings:
one-insn-per-tb: off

Translation buffer state:
gen code size 29895283/1073736704
TB count 52714
TB avg target size 24 max=2048 bytes
TB avg host size 329 bytes (expansion ratio: 13.5)
cross page TB count 0 (0%)
direct jump count 31909 (60%) (2 jumps=25824 48%)
TB hash buckets 24493/32768 (74.75% head buckets used)
TB hash occupancy 33.35% avg chain occ. Histogram: [0,10)%|▆ █ ▅▁▃▁▁|[90,100]%
TB hash avg chain 1.018 buckets. Histogram: 1|█▁|2

Statistics:
TB flush count 0
TB invalidate count 7841
TLB full flushes 0
TLB partial flushes 13184
TLB elided flushes 99536
[TCG profiler not compiled]
*** Done
white0@qemu:~/qemu/build$

Now I'll try without the optimizations

Even if I honestly didn't understand anything, what values should I look at to understand the differences ?

@white

I don't test it but use it like someone with real hardware

Since the Pegasos2 machine was the first one available under Qemu, I stayed with it and set up a nice AmigaOs4.1 system that I use every day. In addition, AmigaOs4.1 is on a real SSD that I have given to Qemu. Everything that is worthwhile for me I have installed (os4depot) or bought software myself.

I also did some tests with the AmigaOneXe and Sam460 machine. Mainly because of SDL1 problems that do not exist on these machines.

With lame test compiled as I have always done without 03 etc etc.

./configure --target-list=ppc-softmmu --disable-dbus-display --enable-slirp --enable-lto --enable-sdl --enable-gtk



*** Running test 1
Loading, please wait...
*** Running lame ***
LAME 3.100 32bits (http://lame.sf.net)
Using polyphase lowpass filter, transition band: 18671 Hz - 19205 Hz
Encoding /file_example_WAV_5MG.wav to <stdout>
Encoding as 44.1 kHz j-stereo MPEG-1 Layer III VBR(q=2)
Frame | CPU time/estim | REAL time/estim | play/CPU | ETA
1149/1149 (100%)| 0:26/ 0:26| 0:26/ 0:26| 1.1335x| 0:00
QEMU 9.0.50 monitor - type 'help' for more information
Accelerator settings:
one-insn-per-tb: off

Translation buffer state:
gen code size 29896995/1073736704
TB count 52715
TB avg target size 24 max=2048 bytes
TB avg host size 329 bytes (expansion ratio: 13.5)
cross page TB count 0 (0%)
direct jump count 31909 (60%) (2 jumps=25822 48%)
TB hash buckets 24492/32768 (74.74% head buckets used)
TB hash occupancy 33.37% avg chain occ. Histogram: [0,10)%|▆ █ ▅▁▃▁▁|[90,100]%
TB hash avg chain 1.018 buckets. Histogram: 1|█▁|2

Statistics:
TB flush count 0
TB invalidate count 7841
TLB full flushes 0
TLB partial flushes 13223
TLB elided flushes 99801
[TCG profiler not compiled]
*** Done
*** Running test 2
Loading, please wait...
*** Running lame ***
LAME 3.100 32bits (http://lame.sf.net)
Using polyphase lowpass filter, transition band: 18671 Hz - 19205 Hz
Encoding /file_example_WAV_5MG.wav to <stdout>
Encoding as 44.1 kHz j-stereo MPEG-1 Layer III VBR(q=2)
Frame | CPU time/estim | REAL time/estim | play/CPU | ETA
1149/1149 (100%)| 0:25/ 0:25| 0:25/ 0:25| 1.1983x| 0:00
QEMU 9.0.50 monitor - type 'help' for more information
Accelerator settings:
one-insn-per-tb: off

Translation buffer state:
gen code size 29907955/1073736704
TB count 52736
TB avg target size 24 max=2048 bytes
TB avg host size 329 bytes (expansion ratio: 13.5)
cross page TB count 0 (0%)
direct jump count 31917 (60%) (2 jumps=25828 48%)
TB hash buckets 24466/32768 (74.66% head buckets used)
TB hash occupancy 33.33% avg chain occ. Histogram: [0,10)%|▆ █ ▅▁▃▁▁|[90,100]%
TB hash avg chain 1.019 buckets. Histogram: 1|█▁|2

Statistics:
TB flush count 0
TB invalidate count 7841
TLB full flushes 0
TLB partial flushes 13265
TLB elided flushes 100159
[TCG profiler not compiled]
*** Done
*** Running test 3
Loading, please wait...
*** Running lame ***
LAME 3.100 32bits (http://lame.sf.net)
Using polyphase lowpass filter, transition band: 18671 Hz - 19205 Hz
Encoding /file_example_WAV_5MG.wav to <stdout>
Encoding as 44.1 kHz j-stereo MPEG-1 Layer III VBR(q=2)
Frame | CPU time/estim | REAL time/estim | play/CPU | ETA
1149/1149 (100%)| 0:25/ 0:25| 0:25/ 0:25| 1.1679x| 0:00
QEMU 9.0.50 monitor - type 'help' for more information
Accelerator settings:
one-insn-per-tb: off

Translation buffer state:
gen code size 29901411/1073736704
TB count 52716
TB avg target size 24 max=2048 bytes
TB avg host size 329 bytes (expansion ratio: 13.5)
cross page TB count 0 (0%)
direct jump count 31911 (60%) (2 jumps=25826 48%)
TB hash buckets 24474/32768 (74.69% head buckets used)
TB hash occupancy 33.37% avg chain occ. Histogram: [0,10)%|▆ █ ▅▁▃▁▁|[90,100]%
TB hash avg chain 1.018 buckets. Histogram: 1|█▁▁|3

Statistics:
TB flush count 0
TB invalidate count 7841
TLB full flushes 0
TLB partial flushes 13244
TLB elided flushes 99940
[TCG profiler not compiled]
*** Done
white0@qemu:~/qemu/build$

I'd like to know if there are any particular differences out of curiosity.

I guess this refers to audio decoding broadly speaking

@Maijestro

It's been a while since I used qemu
The other day I found a 500GB Samsung 860 SSD for 33 euros in a shopping center.
I need to find the command line to use real partitions to dedicate to AmigaOS.

At the moment I'm using Pegasos2 maybe then I'll try the others.

@white
Quote:

Qemu9 with the ‘-O3’ compilation is about 2 seconds faster than the standard one
My time is about 21 seconds with a good session on macOS M1, 50 seconds with a bad session
This tool was mainly created to check Qemu problems under macOS.


But it is also interesting to compare QEMU versions.
For me in this test, QEMU8 is about 1-2 seconds faster than the new Qemu9

You have a fast QEMU compilation and I guess you don't need to check anymore

@smarkusg

I'm using your version of E-UAE with the -7457 cpu
and it works well. The display is now perfect.
Without the smudged graphics.

I wanted to ask you if you could take a look at Amicygnix as soon as you had time (not for the ssh session).
But the graphical smudge that is still present using the -7457 cpu
Maybe it is possible to adopt the same solution you found for E-UAE.

Thank you.

Just seen this thread.

Regarding GPU passthrough, yes, I got it "working", but it was not properly working. And I might be wrong but I don't think we ever truely fixed it (IRQ issues from memory maybe?). In small part because I just went and bought an X5000 system instead which obviously works much better - or as best as any OS4 system can :D

I think the system may still be setup (it's my workbench system for when I'm working on PCB repair), but that's a guess. I might have reformatted it when I took it out to the workshop.

From memory, it "worked" but was pretty slow. At the time it was enough for me to test some software I was building and that was that. I wouldn't recommend the setup for anyone on a day-to-day basis or for anyone to spend money on hardware unless they are doing so in order to progress the issues. And by that I mean actually develop and / or debug the issues. I don't feel it's fair to just push everything onto Baleton all the time, and probably also Hans.

@MartinW

verified that @gennaam, it gave fast just need the right machine and configuration.

"Take a Ryzen 7700x and you'll definitely leave the X5020 in the dust..."

https://www.amigans.net/modules/newbb/ ... id=147757#forumpost147757

@white
I'm very glad that everything works for you
Currently in Poland we have a long weekend and I am busy remoneting my flat... but as I will try to check later Amicygnix and ssh
Have fun

@MartinW
Quote:IRQ issues on Pegasos2 (real and emulated) are fixed with BBoot version 0.7

https://www.amigans.net/modules/newbb/ ... id=147748#forumpost147748
Quote:

@MartinW

Quote:

The passing of real graphics cards has already been tested by @gennaam and was successful @Balaton had to solve a few IRQ problems in BBoot.

So technically it is feasible and gennaam described that the Qemu/Pegasos2 system with R9 270 behaved like a real system, based on the CPU display.

https://www.amigans.net/modules/newbb/ ... id=147759#forumpost147759

Of course this is not a solution for Mac users like me and the setup is currently only for power users, but it is impressive that it works.

Nice to see it's moved on then. I'll have to have a look over the weekend and see if my system is in any state to test it or not. I can't remember if I reformatted it or not. I had 3 or 4 cards I could test. None were 7000 series though.

I also might not get too much time as it's my birthday at the weekend which tends to mean family, despite everyone claiming it's up to me what I do

@MartinW
@all
the card is this ATI Radeon HD 7000 with VGA is this one in the link ?

https://www.techpowerup.com/gpu-specs/radeon-7000.c648

or

https://www.techpowerup.com/gpu-specs/radeon-7000-pci.c649

@white

I don't know, you'd have to ask Geennaam. As far as I was aware the 'HD7000' is a series of cards so I don't know which specific card he tested, what memory config and so on.

@MartinW
Thank you .

@Geennaam

the card is this ATI Radeon HD 7000 with VGA is this one in the link ?

https://www.techpowerup.com/gpu-specs/radeon-7000.c648

or

https://www.techpowerup.com/gpu-specs/radeon-7000-pci.c649

( I'm guessing it's PCI )

@white
@MartinW

This topic is only two pages long

see post

https://www.amigans.net/modules/newbb/ ... id=148064#forumpost148064

If someone is asking about purchasing hardware it would be better not to guess based on screenshots since I’ve had cards that are meant to be supported for stuff not work because they were the same in name only but different specs. Obviously feel free to guess. I’ll just go back to what I was doing.

@white

The card is much too old and can only be used via a PCIe to PCI bridge. You would probably have Warp3d (MiniGL) available under amigaOs4.1. Warp3d Nova (GLes2) does not work with these old cards.

As I understand it, an AMD Radeon R9 270/x is the best card to use for full 3d acceleration including Warp3D Nova under Qemu/Pegasos2. This card has 2 GB memory.

Here you have an example of this card:

https://www.techpowerup.com/gpu-specs/radeon-r9-270x.c2466

Unfortunately @geennaam didn't give enough information about which R9 270x he used exactly.

Here again the thread about it:

https://www.amigans.net/modules/newbb/ ... id=147761#forumpost147761

However, AmigaOs4.1 does not provide a driver for this card and you would have to use additional software (Enhancer Software) which provides current graphics card drivers among others. And you probably need a board with 2 PCIe slots.

So we have no proof that it actually works as it has only been tested by one user and there is no real useful information about it.

@MartinW

Quote:

I didn't give this as a guess. I was basing it on material about AOS4 and which cards work best.
Unfortunately, only @geennaam probably will no longer visit this forum and will not comment.
It's a shame, because he was probably the only person who was very open to new solutions and full of enthusiasm in what he was doing.
His software has been treated as unlicensed, his violation fully accepted and his point of view as a code writer not accepted.

@white
No, the cards on your techpowerup.com links are old Radeon 7000 cards with only 32 MB RAM, not Radeon HD cards.
Should be the same as the Micro-A1 onboard gfx, works with ATIRadeon.chip and would be better than nothing at all (SM50x emulation), but an ancient, 32 MB gfx card isn't much fun either.

@geennaam posted GfxBench2D results with all data on https://ftp.hdrlab.org.nz/benchmark/gf ... 2d/OS/AmigaOS/Result/2773
Quote:3.7 is included in Enhancer Software, i.e. you don't have to additionally buy the RadeonHD V5 driver for it.

@joerg
@all

Thanks for the info.
I bought two USB-Audio years ago, spending now I don't remember, let's say 15-20 euros to make them work with Morphos through qemu.

A video card like this one for example with the exact specifications:
EBAY
https://tinyurl.com/2c699sf6

clearly I have to have the characteristics sent to me as per specifications.
so preferably I should see in the Italian ads.

The part that interested me was this:

Apart from the Drivers.

Is the card recognized even without drivers?
I mean on real AmigaOS hardware it is recognized regardless of the drivers.


And then later you install the necessary drivers.
A bit like in the old Windows XP it was initially recognized as Hardware and then enabled 100% with the proprietary Drivers?

Anyone who has real hardware could answer this

Thank you,
As I already specified in another post some time ago
My motherboard with the Ryzen 5800 is this:

https://www.asus.com/motherboards-comp ... ing/tuf-gaming-b550-plus/

It seems clear that it is simply my choice to make this purchase.

If it doesn't work, that's my problem

Radeon HD 7800 Series
Product ID: 0x6810, Vendor ID: 0x1002
Sub-Product ID: 0x3030, Sub-Vendor ID: 0x1462
2 GB RAM
RadeonHD.chip version 3.7