All of lore.kernel.org
 help / color / mirror / Atom feed
From: Gilles Chanteperdrix <gilles.chanteperdrix@xenomai.org>
To: Stefan Roese <sr@denx.de>
Cc: xenomai@xenomai.org
Subject: Re: [Xenomai] [PATCH v2] hal/arm: Add Zynq v3.5.7 patches
Date: Wed, 15 Jan 2014 15:15:39 +0100	[thread overview]
Message-ID: <52D6980B.3080605@xenomai.org> (raw)
In-Reply-To: <1389793982-18114-1-git-send-email-sr@denx.de>

On 01/15/2014 02:53 PM, Stefan Roese wrote:
> Xilinx Zynq is already supported in the mainline Xenomai git repository.
> This patch adds support for the v3.5.7 Linux Kernel with the latest
> arm-6 i-pipe patch additionally to the already present v3.8 support.
>
> Signed-off-by: Stefan Roese <sr@denx.de>
> ---
> v2:
> - Base pre-patch on top of a different git reopsitory based on v3.5.7
> - Remove defconfig and other board specific stuff from post-patch
> - All this leads to minimal pre- and post- patches as requested
>    by Gilles

Hi Stefan,

this looks fine to me, except:


> +diff --git a/arch/arm/kernel/ipipe.c b/arch/arm/kernel/ipipe.c
> +index ff4ad1a..cf5f29b 100644
> +--- a/arch/arm/kernel/ipipe.c
> ++++ b/arch/arm/kernel/ipipe.c
> +@@ -574,8 +574,6 @@ EXPORT_SYMBOL_GPL(__ipipe_serial_debug);
> +
> + EXPORT_SYMBOL_GPL(do_munmap);
> + EXPORT_SYMBOL_GPL(show_stack);
> +-EXPORT_SYMBOL_GPL(init_mm);
> +-EXPORT_SYMBOL_GPL(cpu_architecture);
> + #ifndef MULTI_CPU
> + EXPORT_SYMBOL_GPL(cpu_do_switch_mm);
> + #endif

If you need to remove these two lines, they should be removed from the 
I-pipe patch itself, not in the Zynq-specific patch. But, why do you 
need to remove them?

> +diff --git a/drivers/gpio/gpio-xilinxps.c b/drivers/gpio/gpio-xilinxps.c

> +-	device_pin_num = irq_to_gpio(irq_data->irq); /* get pin num within the device */
> ++	device_pin_num = irq_data->hwirq; /* get pin num within the device */

Why do you need to change that? Is it related to the I-pipe patch?

> +-	device_pin_num = irq_to_gpio(irq_data->irq); /* get pin num within the device */
> ++	device_pin_num = irq_data->hwirq; /* get pin num within the device */

> +-	device_pin_num = irq_to_gpio(irq_data->irq); /* get pin num within the device */
> ++	device_pin_num = irq_data->hwirq; /* get pin num within the device */

> +-	device_pin_num = irq_to_gpio(irq_data->irq); /* get pin num within the device */
> ++	device_pin_num = irq_data->hwirq; /* get pin num within the device */


> ++static int xgpiops_set_wake(struct irq_data *data, unsigned int on)
> ++{
> ++	if (on)
> ++		xgpiops_irq_unmask(data);
> ++	else
> ++		xgpiops_irq_mask(data);
> ++
> ++	return 0;
> ++}
> ++
> + /* irq chip descriptor */
> + static struct irq_chip xgpiops_irqchip = {
> + 	.name		= DRIVER_NAME,
> +@@ -381,6 +388,7 @@ static struct irq_chip xgpiops_irqchip = {
> + 	.irq_mask	= xgpiops_irq_mask,
> + 	.irq_unmask	= xgpiops_irq_unmask,
> + 	.irq_set_type	= xgpiops_set_irq_type,
> ++	.irq_set_wake	= xgpiops_set_wake,

Same question

> +-	chip->irq_ack(irq_data);
> ++	chained_irq_enter(chip, desc);

Same question.

> +-	chip->irq_unmask(irq_data);
> ++
> ++	chip = irq_desc_get_chip(desc);
> ++	chained_irq_exit(chip, desc);

Same question, and chip is probably still valid, so, calling 
irq_desc_get_chip seems useless.

I do not really mind the changes in gpios, but I do not think removing 
the EXPORT_SYMBOL from ipipe.c makes sense. These should only be useful 
if you compile Xenomai completely as module, but should still be useful.

-- 
					    Gilles.


  reply	other threads:[~2014-01-15 14:15 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-01-15 13:53 [Xenomai] [PATCH v2] hal/arm: Add Zynq v3.5.7 patches Stefan Roese
2014-01-15 14:15 ` Gilles Chanteperdrix [this message]
2014-01-15 14:39   ` Stefan Roese
2014-01-15 15:53     ` Gilles Chanteperdrix
2014-01-15 15:57       ` Stefan Roese

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=52D6980B.3080605@xenomai.org \
    --to=gilles.chanteperdrix@xenomai.org \
    --cc=sr@denx.de \
    --cc=xenomai@xenomai.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.