Re: [igt-dev] [PATCH i-g-t] Revert "lib/igt_device: Move intel_get_pci_device under igt_device"

From: Jani Nikula <jani.nikula@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>,
	Ville Syrjala <ville.syrjala@linux.intel.com>,
	igt-dev@lists.freedesktop.org
Subject: Re: [igt-dev] [PATCH i-g-t] Revert "lib/igt_device: Move intel_get_pci_device under igt_device"
Date: Thu, 28 Mar 2019 10:52:03 +0200	[thread overview]
Message-ID: <87o95v2xfw.fsf@intel.com> (raw)
In-Reply-To: <155376125792.24691.13846060111589600892@skylake-alporthouse-com>

On Thu, 28 Mar 2019, Chris Wilson <chris@chris-wilson.co.uk> wrote:
> Quoting Jani Nikula (2019-03-28 08:14:02)
>> On Wed, 27 Mar 2019, Ville Syrjala <ville.syrjala@linux.intel.com> wrote:
>> > From: Ville Syrjälä <ville.syrjala@linux.intel.com>
>> >
>> > One significant usecase for intel_reg/etc. is to be able to examine
>> > the hardware state *before* loading the driver. If the tool forces
>> > the driver to load we've totally lost that capability.
>> 
>> Agreed.
>
> Not quite, I see the problem of tools/ linking against the libigt.la
> smorgasbord remaining.

Well, agreed on the revert as stopgap because it breaks an important use
case.

I wrote and deleted some musings on how I think tools/ is increasingly
out of place in igt. The common igt code evolves from tests/
perspective, and I suppose rightly so. I think I've suggested in the
past tools/ should be split out to be more independent, either within
igt repo or in a separate project/repo of its own.

If tools/ remain in igt, perhaps we need a lower level no-nonsense lib
that doesn't, for example, do stack unwind on errors which is rather
silly with the general purpose tools. And then have a tests lib on top
for tests.

BR,
Jani.

-- 
Jani Nikula, Intel Open Source Graphics Center
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev