Hi Arnd, > -----Original Message----- > From: Arnd Bergmann [mailto:arnd@xxxxxxxx] > Sent: Friday, February 18, 2011 6:39 PM > To: Nath, Arindam > Cc: zhangfei gao; cjb@xxxxxxxxxx; linux-mmc@xxxxxxxxxxxxxxx; Su, Henry; > Lu, Aaron; anath.amd@xxxxxxxxx > Subject: Re: [PATCH 01/12] mmc: sdhci: add support for auto CMD23 > > On Friday 18 February 2011, Nath, Arindam wrote: > > > AFAICT, whether CMD23 can be used or not is primarily a decision > based > > > on > > > the card, and the block driver might want issue it for a longer > range > > > if it needs to split a block request into multiple MMC requests. > > > > Thanks for the comments. Yes, whether a card supports CMD23 depends > on > > the card, but Auto CMD23 is a feature added to the Host Controller > v3.00 > > just like Auto CMD12. I am already doing a check inside > > sdhci_host_auto_cmd23_supported() whether the card supports CMD23, > and > > whether the request is for multiple block transfers. Also I tried to > make > > my implementation consistent with Auto CMD12 implementation already > > there in the sdhc code. > > Ok, I see. > > But, assuming that the block driver was changed to always to CMD23 > where > appropriate, what is the advantage auf Auto CMD23? Quoting verbatim from the Host Controller Spec v3.00, section 1.15 -- Auto CMD23 is a feature that automatically issues a CMD23 before a CMD18 or CMD25 is sent. Objective of this function is to avoid performance deterioration during memory access by removing interrupt service of CMD23. Regards, 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
