public inbox for u-boot@lists.denx.de
 help / color / mirror / Atom feed
From: Maxime Ripard <maxime.ripard@bootlin.com>
To: u-boot@lists.denx.de
Subject: [U-Boot] [PATCH] sunxi: H6: DRAM: avoid memcpy() on MMIO registers
Date: Thu, 6 Dec 2018 08:42:30 +0100	[thread overview]
Message-ID: <20181206074230.rcurmal3454kjucf@flea> (raw)
In-Reply-To: <20181206000220.2302-1-andre.przywara@arm.com>

On Thu, Dec 06, 2018 at 12:02:20AM +0000, Andre Przywara wrote:
> Using memcpy() for MMIO operations is, however tempting, not a good idea:
> It depends on the specific implementation of memcpy, also lacks barriers.
> In this particular case the first registers were written using 64-bit
> writes, and the last register using four separate single-byte writes.
> Neither is what we actually want.
> We get away with it in this case because of the particular details of
> the bus implementation, the DRAM controller IP and the values that we
> actually write, but we should not leave a bad example around.
> 
> Replace the memcpy with a proper loop using the writel() accessor.
> 
> Signed-off-by: Andre Przywara <andre.przywara@arm.com>
> ---
>  arch/arm/mach-sunxi/dram_sun50i_h6.c | 6 ++++--
>  1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/arm/mach-sunxi/dram_sun50i_h6.c b/arch/arm/mach-sunxi/dram_sun50i_h6.c
> index 5da90a2835..e2f141eb9b 100644
> --- a/arch/arm/mach-sunxi/dram_sun50i_h6.c
> +++ b/arch/arm/mach-sunxi/dram_sun50i_h6.c
> @@ -182,6 +182,7 @@ static void mctl_set_timing_lpddr3(struct dram_para *para)
>  			(struct sunxi_mctl_ctl_reg *)SUNXI_DRAM_CTL0_BASE;
>  	struct sunxi_mctl_phy_reg * const mctl_phy =
>  			(struct sunxi_mctl_phy_reg *)SUNXI_DRAM_PHY0_BASE;
> +	int i;
>  
>  	u8 tccd		= 2;
>  	u8 tfaw		= max(ns_to_t(50), 4);
> @@ -237,8 +238,9 @@ static void mctl_set_timing_lpddr3(struct dram_para *para)
>  	u8 twr2rd	= tcwl + 4 + 1 + twtr;
>  	u8 trd2wr	= tcl + 4 + (tcksrea >> 1) - tcwl + 1;
>  
> -	/* set mode register */
> -	memcpy(mctl_phy->mr, mr_lpddr3, sizeof(mr_lpddr3));
> +	/* set mode registers */
> +	for (i = 0; i < ARRAY_SIZE(mr_lpddr3); i++)
> +		writel(mr_lpddr3[i], &mctl_phy->mr[i]);

memcpy_toio is meant to do just that.

Maxime

-- 
Maxime Ripard, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 228 bytes
Desc: not available
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20181206/93f2f7c7/attachment.sig>

  reply	other threads:[~2018-12-06  7:42 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-12-06  0:02 [U-Boot] [PATCH] sunxi: H6: DRAM: avoid memcpy() on MMIO registers Andre Przywara
2018-12-06  7:42 ` Maxime Ripard [this message]
2018-12-07 14:17   ` Andre Przywara

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=20181206074230.rcurmal3454kjucf@flea \
    --to=maxime.ripard@bootlin.com \
    --cc=u-boot@lists.denx.de \
    /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