From: Arnd Bergmann <arnd@arndb.de>
To: Anurup M <anurupvasu@gmail.com>
Cc: linux-arm-kernel@lists.infradead.org,
        Tan Xiaojun <tanxiaojun@huawei.com>, anurup.m@huawei.com,
        linux-kernel@vger.kernel.org, mark.rutland@arm.com,
        shyju.pv@huawei.com, gabriele.paoloni@huawei.com,
        john.garry@huawei.com, will.deacon@arm.com, linuxarm@huawei.com,
        xuwei5@hisilicon.com, zhangshaokun@hisilicon.com,
        sanil.kumar@hisilicon.com, shiju.jose@huawei.com
Subject: Re: [PATCH v1 03/11] drivers: soc: hisi: Add support for Hisilicon Djtag driver
Date: Tue, 08 Nov 2016 16:08:58 +0100
Message-ID: <4657586.c5MJoh65Ux@wuerfel>
User-Agent: KMail/5.1.3 (Linux/4.4.0-34-generic; KDE/5.18.0; x86_64; ; )
In-Reply-To: <5821D736.7080803@gmail.com>
References: <1478101374-18778-1-git-send-email-anurup.m@huawei.com> <22120038.dbaYpsQoUH@wuerfel> <5821D736.7080803@gmail.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 7Bit
Content-Type: text/plain; charset="us-ascii"
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2154
Lines: 48

On Tuesday, November 8, 2016 7:16:30 PM CET Anurup M wrote:
> >>>> If these are backwards compatible, just mark them as compatible in DT,
> >>>> e.g. hip06 can use
> >>>>
> >>>>    compatible = "hisilicon,hip06-cpu-djtag-v1", "hisilicon,hip05-cpu-djtag-v1";
> >>>>
> >>>> so you can tell the difference if you need to, but the driver only has to
> >>>> list the oldest one here.
> >>>>
> >>>> What is the difference between the cpu and io djtag interfaces?
> >> On some chips like hip06, the djtag version is different for IO die.
> > In what way? The driver doesn't seem to care about the difference.
> There is  a difference in djtag version of CPU and IO die (in some chips).
> For ex: in hip06 djtag for CPU is v1 and for IO is v2.
> so they use different readwrite handlers djtag_readwrite_(v1/2).
> 
> +       /* for hip06(D03) cpu die */
> +       { .compatible = "hisilicon,hip06-cpu-djtag-v1",
> +               .data = (void *)djtag_readwrite_v1 },
> +       /* for hip06(D03) io die */
> +       { .compatible = "hisilicon,hip06-io-djtag-v2",
> +               .data = (void *)djtag_readwrite_v2 },
> 
> 
> For the same djtag version, there is no difference in handling in the 
> driver.

Right, but my point was about the compatibility with the older chips
using the same IP block, marking the ones as compatible that actually
use the same interface.

I also see that the compatible strings have the version included in
them, and you can probably drop them by requiring them only in the
fallback:

	compatible = "hisilicon,hip05-cpu-djtag", "hisilicon,djtag-v1";
	compatible = "hisilicon,hip05-io-djtag", "hisilicon,djtag-v1";
	compatible = "hisilicon,hip06-cpu-djtag", "hisilicon,djtag-v1";
	compatible = "hisilicon,hip06-io-djtag", "hisilicon,djtag-v2";
	compatible = "hisilicon,hip07-cpu-djtag", "hisilicon,djtag-v2";
	compatible = "hisilicon,hip07-io-djtag", "hisilicon,djtag-v2";

We want to have the first entry be as specific as possible, but
the last (second) entry is the one that can be used by the driver
for matching. When a future hip08/hip09/... chip uses an existing
interface, you then don't have to update the driver.

	Arnd