[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1452567284.8331.12.camel@mtksdaap41>
Date: Tue, 12 Jan 2016 10:54:44 +0800
From: PC Liao <pc.liao@...iatek.com>
To: Takashi Iwai <tiwai@...e.de>
CC: "broonie@...nel.org" <broonie@...nel.org>,
srv_heupstream <srv_heupstream@...iatek.com>,
"linux-mediatek@...ts.infradead.org"
<linux-mediatek@...ts.infradead.org>,
"s.hauer@...gutronix.de" <s.hauer@...gutronix.de>,
"linux-arm-kernel@...ts.infradead.org"
<linux-arm-kernel@...ts.infradead.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"alsa-devel@...a-project.org" <alsa-devel@...a-project.org>,
"Koro Chen (陳思翰)"
<koro.chen@...iatek.com>,
"Hidalgo Huang (黃正和)"
<Hidalgo.Huang@...iatek.com>
Subject: Re: [PATCH] ASoC: mediatek: Enable 33bit memory address to support
4GB DRAM
On Mon, 2016-01-11 at 17:02 +0800, Takashi Iwai wrote:
> On Mon, 11 Jan 2016 09:00:50 +0100,
> PC Liao wrote:
> >
> > @@ -603,7 +606,10 @@ static int mtk_afe_dais_hw_params(struct snd_pcm_substream *substream,
> > if (ret < 0)
> > return ret;
> >
> > - memif->phys_buf_addr = substream->runtime->dma_addr;
> > + if (sizeof(dma_addr_t) > 4)
> > + msb_at_bit33 = (substream->runtime->dma_addr & 0x100000000) ? 1 : 0;
>
> Better to put a proper suffix for the constant over 32bit.
>
> Or use upper_32_bits(). Then sizeof() check can be omitted, as the
> compiler should be smart enough to know it beforehand.
>
>
> Takashi
Hi Takashi,
Thanks for your comment.
I change as below:
@@ -606,10 +606,8 @@ static int mtk_afe_dais_hw_params(struct
snd_pcm_substream
if (ret < 0)
return ret;
- if (sizeof(dma_addr_t) > 4)
- msb_at_bit33 = (substream->runtime->dma_addr &
0x100000000) ? 1
-
- memif->phys_buf_addr = substream->runtime->dma_addr &
0xffffffff;
+ msb_at_bit33 = upper_32_bits(substream->runtime->dma_addr) ? 1 :
0;
+ memif->phys_buf_addr =
lower_32_bits(substream->runtime->dma_addr);
Dose this change follow your idea?
Thanks!
PC Liao
Powered by blists - more mailing lists