On 23 May 2018 at 05:50, Yinbo Zhu <yinbo.zhu@xxxxxxx> wrote: > > > -----Original Message----- > From: Ulf Hansson [mailto:ulf.hansson@xxxxxxxxxx] > Sent: 2018年5月22日 20:16 > To: Yinbo Zhu <yinbo.zhu@xxxxxxx> > Cc: Y.b. Lu <yangbo.lu@xxxxxxx>; Adrian Hunter <adrian.hunter@xxxxxxxxx>; linux-mmc@xxxxxxxxxxxxxxx; Xiaobo Xie <xiaobo.xie@xxxxxxx> > Subject: Re: [PATCH v2] mmc: sdhci-of-esdhc: modify the sd clock in soc_device_match way > > On 22 May 2018 at 14:06, Yinbo Zhu <yinbo.zhu@xxxxxxx> wrote: >> From: yinbo.zhu <yinbo.zhu@xxxxxxx> >> >> Convert to use soc_device_match method to fix up eSDHC clock for >> ls1046a/ls1012a/p1010. Also add eSDHC clock fixup for ls1021a >> according to its datasheet. The maxmum speed for ls1021a eSDHC high >> speed mode is 46.5MHz. > >>Why not use mmc DT property "max-frequency" instead? > >>Why exactly do you need a different max frequency depending on the selected speed mode? > >>Kind regards >>Uffe > Accroding to datasheet that the different max frequency depending on the slected speed mode and SoC, That it depends on the SoC, that I am or course fine with. My point is rather that I think it depends on the actual slot connector (or board), rather that on the speed mode. For example, one port may have an eMMC and for some reason it can be run in a speed mode using full clock rate. Another one may be an SD-card slot, and perhaps because of some external logic, the signal quality doesn't survive a too high clock-rate. Etc. Other SoCs and platforms works like this, I am sure yours do as well. No? > In the current state that which only shows one of the max-frequency, so it can't use mmc DT property "max-frequency" instead According to my comments above, could you please investigate once more, whether really "max-frequency" can be used as the DT binding? If it can't then I think we should consider introducing new mmc DT bindings instead. [...] Kind regards Uffe -- To unsubscribe from this list: send the line "unsubscribe linux-mmc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
