linux-m68k.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Stefan Agner <stefan@agner.ch>
To: Krzysztof Kozlowski <krzk@kernel.org>,
	Angelo Dureghello <angelo@sysam.it>
Cc: dmaengine@vger.kernel.org, linux-m68k@vger.kernel.org,
	vinod.koul@linaro.org, linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v8 1/3] dmaengine: fsl-edma: extract common fsl-edma code (no changes in behavior intended)
Date: Tue, 07 Aug 2018 19:47:19 +0200	[thread overview]
Message-ID: <fbf2faba97ca1d6f5887e0192139c6fa@agner.ch> (raw)
In-Reply-To: <CAJKOXPchX3H9JVf1VyA+m9waLz74DnWRb2xLBg4yp=YbrEGu0Q@mail.gmail.com>

On 07.08.2018 14:14, Krzysztof Kozlowski wrote:
> On 7 August 2018 at 10:08, Stefan Agner <stefan@agner.ch> wrote:
>> On 03.08.2018 21:32, Angelo Dureghello wrote:
>>> This patch adds a new fsl-edma-common module to allow new
>>> mcf-edma module code to use most of the fsl-edma code.
>>>
>>> Signed-off-by: Angelo Dureghello <angelo@sysam.it>
>>> ---
>>> Changes for v2:
>>> - patch splitted into 4
>>> - add mcf-edma as minimal different parts from fsl-edma
>>>
>>> Changes for v3:
>>> none
>>>
>>> Changes for v4:
>>> - patch simplified from 4/4 into 2/2
>>> - collecting all the mcf-edma-related changes
>>>
>>> Changes for v5:
>>> none
>>>
>>> Changes for v6:
>>> - adjusted comment header
>>> - fixed bit shift with BIT()
>>> - we need to free the interrupts at remove(), so removed all devm_
>>>   interrupt related calls
>>>
>>> Changes for v7:
>>> none
>>>
>>> Changes for v8:
>>> - patch rewritten from scratch, splitted into 3, common code isolated,
>>>   minimal changes from the original Freescale code have been done.
>>>   The patch has been tested with both Iris + Colibri Vybrid VF50 and
>>>   stmark2/mcf54415 Coldfire boards.
>>> ---
>>>  drivers/dma/Makefile          |   2 +-
>>>  drivers/dma/fsl-edma-common.c | 576 ++++++++++++++++++++++++++++
>>>  drivers/dma/fsl-edma-common.h | 196 ++++++++++
>>>  drivers/dma/fsl-edma.c        | 697 +---------------------------------
>>>  4 files changed, 774 insertions(+), 697 deletions(-)
>>>  create mode 100644 drivers/dma/fsl-edma-common.c
>>>  create mode 100644 drivers/dma/fsl-edma-common.h
>>>
>>> diff --git a/drivers/dma/Makefile b/drivers/dma/Makefile
>>> index 203a99d68315..66022f59fca4 100644
>>> --- a/drivers/dma/Makefile
>>> +++ b/drivers/dma/Makefile
>>> @@ -31,7 +31,7 @@ obj-$(CONFIG_DW_AXI_DMAC) += dw-axi-dmac/
>>>  obj-$(CONFIG_DW_DMAC_CORE) += dw/
>>>  obj-$(CONFIG_EP93XX_DMA) += ep93xx_dma.o
>>>  obj-$(CONFIG_FSL_DMA) += fsldma.o
>>> -obj-$(CONFIG_FSL_EDMA) += fsl-edma.o
>>> +obj-$(CONFIG_FSL_EDMA) += fsl-edma.o fsl-edma-common.o
>>>  obj-$(CONFIG_FSL_RAID) += fsl_raid.o
>>>  obj-$(CONFIG_HSU_DMA) += hsu/
>>>  obj-$(CONFIG_IMG_MDC_DMA) += img-mdc-dma.o
>>> diff --git a/drivers/dma/fsl-edma-common.c b/drivers/dma/fsl-edma-common.c
>>> new file mode 100644
>>> index 000000000000..0ae7094f477a
>>> --- /dev/null
>>> +++ b/drivers/dma/fsl-edma-common.c
>>> @@ -0,0 +1,576 @@
>>> +// SPDX-License-Identifier: GPL-2.0
>>> +//
>>> +// Copyright (c) 2013-2014 Freescale Semiconductor, Inc
>>> +// Copyright (c) 2017 Sysam, Angelo Dureghello  <angelo@sysam.it>
>>> +
>>> +#include <linux/dmapool.h>
>>> +#include <linux/module.h>
>>> +#include <linux/slab.h>
>>> +
>>> +#include "fsl-edma-common.h"
>>> +
>>> +/*
>>> + * R/W functions for big- or little-endian registers:
>>> + * The eDMA controller's endian is independent of the CPU core's endian.
>>> + * For the big-endian IP module, the offset for 8-bit or 16-bit registers
>>> + * should also be swapped opposite to that in little-endian IP.
>>> + */
>>> +u32 edma_readl(struct fsl_edma_engine *edma, void __iomem *addr)
>>> +{
>>> +     if (edma->big_endian)
>>> +             return ioread32be(addr);
>>> +     else
>>> +             return ioread32(addr);
>>> +}
>>> +EXPORT_SYMBOL_GPL(edma_readl);
>>
>> In 3/3 you link the common object into the two modules individually:
>>
>> obj-$(CONFIG_FSL_EDMA) += fsl-edma.o fsl-edma-common.o
>> obj-$(CONFIG_MCF_EDMA) += mcf-edma.o fsl-edma-common.o
>>
>> Therefor you do not access those functions from another module (they are
>> within the module). No exporting should be necessary. Drop all those
>> exports.
> 
> The fsl-edma-common will be its own module so the exports are
> necessary for proper linking/modpost.

Hm, oh I see, I got that wrong.

We could use
fsl-edma-all-y = fsl-edma.o fsl-edma-common.o
obj-$(CONFIG_FSL_EDMA) += fsl-edma-all.o
mcf-edma-all-y = mcf-edma.o fsl-edma-common.o
obj-$(CONFIG_MCF_EDMA) += mcf-edma-all.o

to create two modules but that duplicates some code. It probably
wouldn't matter in practise since the two IPs are used on different
architecture...

However, if we stay with the three modules approach, we should introduce
a hidden config symbol, e.g.

config FSL_EDMA_COMMON
	tristate

In FSL_EDMA/MCF_EDMA use
	select FSL_EDMA_COMMON

And in the Makefile
obj-$(CONFIG_FSL_EDMA_COMMON) += fsl-edma-common.o
obj-$(CONFIG_FSL_EDMA) += fsl-edma.o
obj-$(CONFIG_MCF_EDMA) += mcf-edma.o


However, I do not like that the compiler can no longer inline simple
acesors like edma_(readl|writel). So if we stay with the three modules
approach then move at least the accesors and
to_fsl_edma_chan/to_fsl_edma_desc as inline functions to the header
file.

--
Stefan

> 
> Best regards,
> Krzysztof
> 
>> If possible I would prefer if you start with cleanup/conversions, then
>> split-up and finally add functionality.
>>
>> So ideally:
>> 1. Use macros for preprocessor defines (where you move to BIT/GENMASK)
>> 2. Split
>> 3. Add EDMA macros etc.
>> 4. Add ColdFire mcf5441x edma support
>>

  reply	other threads:[~2018-08-07 17:47 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-08-03 19:32 [PATCH v8 1/3] dmaengine: fsl-edma: extract common fsl-edma code (no changes in behavior intended) Angelo Dureghello
2018-08-03 19:32 ` [PATCH v8 2/3] dmaengine: fsl-edma: add edma version and configurable registers Angelo Dureghello
2018-08-06  7:59   ` Krzysztof Kozlowski
2018-08-06 20:31     ` Angelo Dureghello
2018-08-07  8:01   ` Stefan Agner
2018-08-03 19:32 ` [PATCH v8 3/3] dmaengine: fsl-edma: add ColdFire mcf5441x edma support Angelo Dureghello
2018-08-06  8:09   ` Krzysztof Kozlowski
2018-08-06 12:21     ` Stefan Agner
2018-08-06 20:23       ` Angelo Dureghello
2018-08-06 21:01     ` Angelo Dureghello
2018-08-06  7:43 ` [PATCH v8 1/3] dmaengine: fsl-edma: extract common fsl-edma code (no changes in behavior intended) Krzysztof Kozlowski
2018-08-06 20:40   ` Angelo Dureghello
2018-08-07  6:02     ` Krzysztof Kozlowski
2018-08-07  8:08 ` Stefan Agner
2018-08-07 12:14   ` Krzysztof Kozlowski
2018-08-07 17:47     ` Stefan Agner [this message]
2018-08-09 22:37       ` Angelo Dureghello
2018-08-10  8:05         ` Stefan Agner
2018-08-10  8:43           ` Angelo Dureghello

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=fbf2faba97ca1d6f5887e0192139c6fa@agner.ch \
    --to=stefan@agner.ch \
    --cc=angelo@sysam.it \
    --cc=dmaengine@vger.kernel.org \
    --cc=krzk@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-m68k@vger.kernel.org \
    --cc=vinod.koul@linaro.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).