Hi Zhangfei, > -----Original Message----- > From: zhangfei gao [mailto:zhangfei.gao@xxxxxxxxx] > Sent: Friday, June 17, 2011 12:15 PM > To: subhashj@xxxxxxxxxxxxxx; Nath, Arindam > Cc: Philip Rakity; cjb@xxxxxxxxxx; linux-mmc@xxxxxxxxxxxxxxx; Lu, Aaron > Subject: Re: UHS-I bus speed mode should be set last in UHS > initialization > > On Wed, Jun 15, 2011 at 7:57 AM, <subhashj@xxxxxxxxxxxxxx> wrote: > > Hi Arindam, > > > >> -----Original Message----- > >> From: linux-mmc-owner@xxxxxxxxxxxxxxx [mailto:linux-mmc- > >> owner@xxxxxxxxxxxxxxx] On Behalf Of Nath, Arindam > >> Sent: Tuesday, June 14, 2011 12:00 AM > >> To: subhashj@xxxxxxxxxxxxxx > >> Cc: Philip Rakity; zhangfei gao; cjb@xxxxxxxxxx; linux- > >> mmc@xxxxxxxxxxxxxxx; Lu, Aaron > >> Subject: RE: UHS-I bus speed mode should be set last in UHS > >> initialization > >> > >> Adding Philip and Zhangfei. > >> > >> Hi Subhash, > >> > >> Does your controller follow some non-standard steps for > initialization? > >> I referred to Figure 3-14, section 3.9.4 of Physical Layer Spec > v3.01, > >> and it shows the order, setting Driver Strength Select, Bus Speed > Mode, > >> and then Current Limit. So please confirm back. > > > > Sorry, I don't know what do you mean by non standard steps for > > initialization. Nowhere in v3.01 spec, it is mentioned that Bus speed > mode > > **must** be set before current limit. Figure 3-14 also doesn't > indicate > > this. > > > > If you set the bus speed mode in sd_set_bus_speed_mode(), you are > changing > > the host controller timings as well and if it's SDR50/SDR104 mode, > then > > immediately after setting the host controller timing mode, we have to > > execute tuning sequence but here before executing tuning sequence, > you are > > sending the CMD6 which involves the read data transfer from card > which is > > bound to fail without tuning and it does fail with our host > controller. > > > > Do let me know if this explanation make sense. > > I think the concern does make sense. > > Though Figure 3-14 and 4-6 describe Bus Speed Mode is selected ahead > of Current limit, the controller timing is changed in > sd_set_bus_speed_mode, which may cause following cmd6 fail. > > However, the test here shows mmp2 does not care the sequence of > speed_mode and current_limit. > > Thanks We also verified that the changes suggested by Subhash's patch does not break things for our controller. But I am still waiting for Philip's confirmation. Thanks, Arindam -- 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
