[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#931507: kernel-wedge: HDA sound board detection takes 60s in d-i



On Sun, 2019-07-07 at 18:44 +0200, Samuel Thibault wrote:
> Ben Hutchings, le dim. 07 juil. 2019 17:36:03 +0100, a ecrit:
> > On Sun, 2019-07-07 at 17:16 +0200, Samuel Thibault wrote:
> > > Ben Hutchings, le dim. 07 juil. 2019 13:35:20 +0100, a ecrit:
> > > > i915 belongs in fb-modules.  I'm not sure that sound-modules should
> > > > depend on it, as it's not a hard dependency.
> > > 
> > > It is not a hard-hard dependency for the HDA driver, but without it
> > > there is a 60s delay for the detection of HDA-based devices...
> > 
> > I think that could (and should) be avoided by only waiting if
> > request_module() succeeds.
> 
> Ah, request_modules() is synchronous, I then wonder what this is waiting
> for, actually.

Device probing might still be partially asynchronous (though for PCI
it's synchronous by default).

> > And I think all installer configurations that include sound-modules
> > also include fb-modules already.
> 
> Probably indeed, but I don't find i915.ko in fb-modules.

Yes that's still to be done.

Ben.

-- 
Ben Hutchings
Time is nature's way of making sure that
everything doesn't happen at once.


Attachment: signature.asc
Description: This is a digitally signed message part


Reply to: