Hi all I am still waiting for your response..... 2013/7/5 James Lebron <leileishangchina@xxxxxxxxx>: > Hi guys, > > Any comments? > > 2013/7/2 James Lebron <leileishangchina@xxxxxxxxx>: >> Hi Wolfram >> >> Have you finished reading datasheet? >> >> 2013/6/15 Wolfram Sang <wsa@xxxxxxxxxxxxx>: >>> Hi, >>> >>> On Fri, Jun 07, 2013 at 02:38:17PM +0800, Leilei Shang wrote: >>>> To enter high speed mode, following steps should be done: >>>> 1. When running in high speed mode, i2c clock rate is different >>>> from standard mode. Clock rate must be set according to >>>> specification first. >>>> 2. When i2c controller sends a master code and wins arbitration, >>>> high speed mode is entered. >>>> >>>> If you want to enable high speed mode, the following members of >>>> platform data should be set to proper value: >>> >>> I couldn't find any support for hs-mode in pxa2xx and pxa3xx. So it >>> shouldn't be set in platform data, but depending on checks if this is >>> supported on this device at all. And why don't simply switch to it if >>> rate is bigger than what is supported without hs-mode? >>> >> >> 1. High speed mode is hardware feature, you can't read it from >> software. It's similar to fast mode, and fast mode is set in platform >> data. >> >> The only method I can think of is to add a platform, like pxa910. >> And if you enable hs mode and platform is not pxa910, we'll set it to >> normal mode. Is this OK? >> >> 2. About hs mode clock setting. We don't know whether currently hs >> mode clock will be changed in future. Do you think below code is OK? >> >> if (i2c->rate) >> clk_set_rate(i2c->clk, i2c->rate); >> else >> clk_set_rate(i2c->clk, 62400000); >> >> pr_info("i2c: <%s> set rate to %ld\n", >> i2c->adap.name, clk_get_rate(i2c->clk)); >> >>>> 1. "high_mode" should be set to "1". >>>> 2. "master_code" should be set to "8'b 0000_1xxx"(x is 0 or 1). >>>> If no master_code is set, set to default value 0xe. >>>> 3. "rate" should be set according to specification. >>> >>> Note that you should fall back to full-speed mode after sending the stop >>> bit in high-speed mode. I can't find this in the code? >>> >> >> 3. Will fall back to normal mode when stop bit is sent. >> >>>> >>>> Signed-off-by: Leilei Shang <shangll@xxxxxxxxxxx> >>> >>> Which SoC supports this? I'd like to have a look at the datasheet. >>> >>> Thanks, >>> >>> Wolfram >>> >> >> Hope you can reply soon! >> >> Best regards >> Leilei -- To unsubscribe from this list: send the line "unsubscribe linux-i2c" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html