Re: commit 7ffb791423c7 breaks steam game

[Balbir Singh <balbirs@nvidia.com>]

On 3/15/25 01:18, Bert Karwatzki wrote:
> Am Samstag, dem 15.03.2025 um 00:34 +1100 schrieb Balbir Singh:
>> On 3/14/25 17:14, Balbir Singh wrote:
>>> On 3/14/25 09:22, Bert Karwatzki wrote:
>>>> Am Freitag, dem 14.03.2025 um 08:54 +1100 schrieb Balbir Singh:
>>>>> On 3/14/25 05:12, Bert Karwatzki wrote:
>>>>>> Am Donnerstag, dem 13.03.2025 um 22:47 +1100 schrieb Balbir Singh:
>>>>>>>
>>>>>>>
>>>>>>> Anyway, I think the nokaslr result is interesting, it seems like with nokaslr
>>>>>>> even the older kernels have problems with the game
>>>>>>>
>>>>>>> Could you confirm if with nokaslr
>>>>>>>
>>>>>> Now I've tested kernel 6.8.12 with nokaslr
>>>>>>
>>>>>>> 1. Only one single game stellaris is not working?
>>>>>>> 2. The entire laptop does not work?
>>>>>>> 3. Laptop works and other games work? Just one game is not working as
>>>>>> expected?
>>>>>>
>>>>>>
>>>>>> Stellaris is showing the input lag and the entire graphical user interface shows
>>>>>> the same input lag as long as stellaris is running.
>>>>>> Civilization 6 shows the same input lag as stellaris, probably even worse.
>>>>>> Magic the Gathering: Arena (with wine) works normally.
>>>>>> Valheim also works normally.
>>>>>> Crusader Kings 2 works normally
>>>>>> Rogue Heroes: Ruins of Tasos (a Zelda lookalike) works normally.
>>>>>> Baldur's Gate I & II and Icewind Dale work normally.
>>>>>>
>>>>>> Also the input lag is only in the GUI, if I switch to a text console (ctrl + alt
>>>>>> + Fn), input works normally even while the affected games are running.
>>>>>>
>>>>>> Games aside everything else (e.g. compiling kernels) seems to work with nokaslr.
>>>>>>
>>>>>
>>>>> Would it be fair to assume that anything Xorg/Wayland is working fine
>>>>> when the game is not running, even with nokaslr?
>>>>>
>>>> Yes, Xorg (I'm normally using xfce4 as desktop) works fine. I also tested with
>>>> gnome using Xwayland, here the buggy behaviour also exists, with the addtion
>>>> that mouse position is off, i.e. to click a button in the game you have to click
>>>> somewhat above it.
>>>
>>> So the issue is narrowed down to just the games you've mentioned with nokaslr/patch?
>>>
>>>>
>>>>> +amd-gfx@lists.freedesktop.org to see if there are known issues with
>>>>> nokaslr and the games you mentioned.
>>>>>
>>>>>
>>>>> Balbir Singh
>>>>>
>>>>> PS: I came across an interesting link
>>>>> https://www.alex-ionescu.com/behind-windows-x64s-44-bit-memory-addressing-limit/
>>>>>
>>>>> I think SLIST_HEADER is used by wine as well for user space and I am not sure
>>>>> if in this situation the game is hitting this scenario, but surprisingly the other
>>>>> games are not. This is assuming the game uses wine. I am not sure it's related,
>>>>> but the 44 bits caught my attention.
>>>>
>>>> Stellaris is a native linux game (x86_64), the one game (MTGA) I tested with
>>>> wine worked fine.
>>>>
>>>
>>> Thanks for the update! I wonder if there are any game logs. If you can look for any
>>> warnings/errors it might be interesting to see where the difference is coming from?
>>>
>>
>> In addition to the above, does radeontop give you any info about what might be
>> going on? I am also curious if
> 
> This got me to test stellaris and Civ6 using the discrete Graphics card:
> 03:00.0 Display controller: Advanced Micro Devices, Inc. [AMD/ATI] Navi 23
> [Radeon RX 6600/6600 XT/6600M] (rev c3)
>  (with DRI_PRIME=1) and here the problems do not occur.
> 
> This is the CPU-builtin GPU, which I normally use to play stellaris as graphics
> aren't very demanding, here the problems occur when using nokaslr
> 08:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI]
> Cezanne [Radeon Vega Series / Radeon Vega Mobile Series] (rev c5)
> 

Aaah.. this is strange, because the only movement in iomem was for the discrete GPU

From your eariler message

 fee00000-fee00fff : pnp 00:04
 ff000000-ffffffff : pnp 00:04
 100000000-fee2fffff : System RAM
-  3a7e00000-3a89d2f56 : Kernel code
-  3a8a00000-3a8e31fff : Kernel rodata
-  3a9000000-3a912a5ff : Kernel data
-  3a969c000-3a97fffff : Kernel bss
+  d32200000-d32dd0f56 : Kernel code
+  d32e00000-d33231fff : Kernel rodata
+  d33400000-d3352a5ff : Kernel data
+  d33a9c000-d33bfffff : Kernel bss
 fee300000-100fffffff : Reserved
 1010000000-ffffffffff : PCI Bus 0000:00
   fc00000000-fe0fffffff : PCI Bus 0000:01
@@ -104,4 +104,4 @@
       fe30300000-fe303fffff : 0000:04:00.0
     fe30400000-fe30403fff : 0000:04:00.0
     fe30404000-fe30404fff : 0000:04:00.0
-3ffe00000000-3fffffffffff : 0000:03:00.0
+afe00000000-affffffffff : 0000:03:00.0

I am hoping someone from amd-gfx to chime in with known issues of the driver
and nokaslr or help debug further.

Balbir Singh