Nouveau - Nov 2025 - [PATCH v2 08/12] nova-core: sequencer: Add register opcodes

On 11/5/25 8:30 AM, Timur Tabi wrote:
> On Wed, 2025-11-05 at 03:45 +0000, Joel Fernandes wrote:
>>>> +??????? dev_dbg!(
>>>> +??????????? sequencer.dev,
>>>> +??????????? "RegWrite: addr=0x{:x}, val=0x{:x}\n",
>>>
>>> Hi Joel,
>>>
>>> The RegRead, RegWrite, RegPoll prints generate over 400 lines
>>> per GPU, into the logs. This is too much, especially now that
>>> it's been working for a while.
>>>
>>> I'm thinking let's delete these entirely. If we somehow get
>>> into debugging this aspect of the sequencer, we can temporarily
>>> add whatever printing we need, but I think it's one notch too
>>> far for the final product, now that you have it working.
>>
>> Sure John, I am Ok with removing the prints. I will do so for the next
spin.
> 
> Or, you could do what Nouveau does, and define two more printk levels below
DBG specifically for
> stuff like this:
> 
> #define nvdev_trace(d,f,a...) nvdev_printk((d), TRACE,   info, f, ##a)
> #define nvdev_spam(d,f,a...)  nvdev_printk((d),  SPAM,    dbg, f, ##a)
...and those are unusable, unfortunately. I've tried.

ftrace/bpftrace, maybe those are the real way to "trace"...or
something
other than this.

thanks,
-- 
John Hubbard

On Wed, 2025-11-05 at 13:55 -0800, John Hubbard wrote:
> > #define nvdev_trace(d,f,a...) nvdev_printk((d), TRACE,?? info, f, ##a)
> > #define nvdev_spam(d,f,a...)? nvdev_printk((d),? SPAM,??? dbg, f, ##a)
> 
> ...and those are unusable, unfortunately. I've tried.
This works great for me:

modprobe nouveau dyndbg="+p" modeset=1 debug="gsp=spam"
config=NvGspRm=1

I get all sequencer messages when I boot with these options.
> ftrace/bpftrace, maybe those are the real way to "trace"...or
something
> other than this.
You could say the same thing about most dev_dbg() statements.

I agree that dev_dbg for sequencer commands is excessive, and that implementing
new debug levels
just to get sequencer prints is also excessive.  But Nouveau implement
nvkm_trace for a reason.  And
we all know that because of ? in Rust, NovaCore does a terrible job at telling
us where an error
actually occurred.  So there is a lot of room for improvement.