Re: [PATCH v2 13/14] ARM: DTS: am33xx: Use the new DT bindings for the eDMA3

From: Peter Ujfalusi <peter.ujfalusi@ti.com>
To: Vinod Koul <vinod.koul@intel.com>, Olof Johansson <olof@lixom.net>
Cc: Sekhar Nori <nsekhar@ti.com>,
	"linux-arm-kernel@lists.infradead.org" 
	<linux-arm-kernel@lists.infradead.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	linux-omap <linux-omap@vger.kernel.org>,
	"dmaengine@vger.kernel.org" <dmaengine@vger.kernel.org>,
	"devicetree@vger.kernel.org" <devicetree@vger.kernel.org>,
	Tony Lindgren <tony@atomide.com>,
	Robert Schwebel <r.schwebel@pengutronix.de>,
	"arm@kernel.org" <arm@kernel.org>,
	"Kristo, Tero" <t-kristo@ti.com>
Subject: Re: [PATCH v2 13/14] ARM: DTS: am33xx: Use the new DT bindings for the eDMA3
Date: Mon, 2 Nov 2015 14:13:01 +0200	[thread overview]
Message-ID: <5637534D.2020304@ti.com> (raw)
In-Reply-To: <20151102100405.GJ21326@localhost>

Vinod,

On 11/02/2015 12:04 PM, Vinod Koul wrote:
> On Mon, Nov 02, 2015 at 01:21:19AM -0800, Olof Johansson wrote:
>> Hi,
>>
>> 1) This seems to have broken BBB in -next for me, bisected down to this patch.
>>
>> For bootlog:
>> http://arm-soc.lixom.net/bootlogs/next/next-20151102/bbb-arm-omap2plus_defconfig.html
>>
>> 2) Please avoid merging DT/platform code in your driver tree, Vinod,
>> at least without an ack from the platform maintainer. It can be a a
>> huge mess if they end up causing conflicts, so we always ask to merge
>> the DT changes through the platform maintainer (Tony in this case) by
>> default.
> 
> I did warn when applying that I am doing so without ACK on ARM code, noone
> said a thing!
> 
> I knew Tony was following the work by Peter so assumed he must have been okay
> with it otherwise would have spoken for ~couple of weeks these were in
> review
> 
> Anyway now that we have a regression, I can revert this patch if that fixes,
> please confirm, but might break edma... peter?

Can you revert or drop the last two DTS patches?

I think I will try a different route to get the split of the tpcc and tptc.
Without the DT patches the driver will fall back to the legacy mode so things
will work in a same way they did before.
Or I can send a followup patch for edma.c, with that there is no need to add
the HWMOD_INIT_NO_IDLE to hwmod and power management looks better.
Basically I'm registering a 'dummy' driver for the edma3-tptc so omap hwmod
code will not shut it down but we will keep the possibility to manage the
power state still.

-- 
Péter