linux-kernel - Re: [char-misc-next 07/15] mei: use wrietl/readl instead of io wrappers

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20121204143151.1cce530a@pyramind.ukuu.org.uk>
Date:	Tue, 4 Dec 2012 14:31:51 +0000
From:	Alan Cox <alan@...rguk.ukuu.org.uk>
To:	Tomas Winkler <tomas.winkler@...el.com>
Cc:	gregkh@...uxfoundation.org, arnd@...db.de, alan@...ux.intel.com,
	linux-kernel@...r.kernel.org
Subject: Re: [char-misc-next 07/15] mei: use wrietl/readl instead of io
 wrappers

On Tue,  4 Dec 2012 16:04:36 +0200
Tomas Winkler <tomas.winkler@...el.com> wrote:

> on intel registers are alwasy memory mapped so drop
> the overhead of iowrite32 and ioread32
> 
> Signed-off-by: Tomas Winkler <tomas.winkler@...el.com>
> ---
>  drivers/misc/mei/mei_dev.h |    4 ++--
>  1 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/misc/mei/mei_dev.h b/drivers/misc/mei/mei_dev.h
> index 92a5d9a..37eb041 100644
> --- a/drivers/misc/mei/mei_dev.h
> +++ b/drivers/misc/mei/mei_dev.h
> @@ -438,7 +438,7 @@ int mei_amthif_irq_read(struct mei_device *dev, s32 *slots);
>  static inline u32 mei_reg_read(const struct mei_device *dev,
>  			       unsigned long offset)
>  {
> -	return ioread32(dev->mem_addr + offset);
> +	return readl(dev->mem_addr + offset);
>  }
>  
>  /**
> @@ -451,7 +451,7 @@ static inline u32 mei_reg_read(const struct mei_device *dev,
>  static inline void mei_reg_write(const struct mei_device *dev,
>  				 unsigned long offset, u32 value)
>  {
> -	iowrite32(value, dev->mem_addr + offset);
> +	writel(value, dev->mem_addr + offset);

ioremap should match readl/writel
iomap iowrite32/ioread32 

so you need to tweak the map/unmap if you do this.

As to overhead. I'd love to see a measurement that can detect the
difference. Do you have a benchmark that shows it ?
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/