RE: [PATCH 3/7] i2c-eg20t: delete 10bit access processing

[Jeffrey (Sheng-Hui) Chu]

http://www.engadget.com/2011/09/26/broadcom-unveils-smaller-more-power-efficient-40nm-nfc-chips/ 

-----Original Message-----
From: Tomoya MORINAGA [mailto:tomoya-linux@xxxxxxxxxxxxxxx] 
Sent: Monday, September 26, 2011 8:05 PM
To: Jeffrey (Sheng-Hui) Chu
Cc: Jean Delvare; Ben Dooks; linux-i2c@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; Qi Wang; yong.y.wang@xxxxxxxxx; joel.clark@xxxxxxxxx; kok.howg.ewe@xxxxxxxxx; toshiharu-linux@xxxxxxxxxxxxxxx; Aaron Smith
Subject: Re: [PATCH 3/7] i2c-eg20t: delete 10bit access processing

(2011/09/23 2:49), Jeffrey (Sheng-Hui) Chu wrote:
> The device is BCM20791.  Please contact Aaron Smith (aarons@xxxxxxxxxxxx) for evaluation.
Though googling the device, I can't find the device.
Broadcom website too.

>
> By the way, the existing handling of 10-bit address in the patch you intended to delete does not work.  You'll need to fix that before you can test with the device successfully.
>
> For the write operation, it's a easy fix:
>
> @@ -410,25 +408,16 @@ static s32 pch_i2c_writebytes(struct i2c_adapter *i2c_adap,
>   	}
>
>   	if (msgs->flags&  I2C_M_TEN) {
> -		addr_2_msb = ((addr&  I2C_MSB_2B_MSK)>>  7);
> +		addr_2_msb = ((addr&  I2C_MSB_2B_MSK)>>  7)&  0x06;
>
>
> For the read operation, static s32 pch_i2c_readbytes(), you'll need to do a write followed by a restart and a read...  please refer to the i2c spec...
>
> Regards,

Thank you for your indication.
We will update this after supporting 10bit access.

Thanks,
tomoya

N‹§²æìr¸›yúèšØb²X¬¶ÇvØ^–)Þ{.nÇ+‰·¥Š{±‘êçzX§¶›¡Ü}©ž²ÆzÚ&j:+v‰¨¾«‘êçzZ+€Ê+zf£¢·hšˆ§~†­†Ûiÿûàz¹®w¥¢¸?™¨è­Ú&¢)ßf”ù^jÇy§m…á@A«a¶Úÿ0¶ìh®å’i