From mboxrd@z Thu Jan 1 00:00:00 1970 From: Wolfgang Wegner Date: Fri, 23 Apr 2010 17:49:10 +0200 Subject: [U-Boot] convention for SPI dummy data? In-Reply-To: <201004231134.54565.vapier@gentoo.org> References: <20100423084307.GN20047@leila.ping.de> <201004231108.11427.vapier@gentoo.org> <20100423151829.GO20047@leila.ping.de> <201004231134.54565.vapier@gentoo.org> Message-ID: <20100423154910.GQ20047@leila.ping.de> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de On Fri, Apr 23, 2010 at 11:34:53AM -0400, Mike Frysinger wrote: [...] > > the problem exists in the (coldfire) SPI driver, not in the MMC/SD code. > > sorry, i thought you were proposing to fix it in the SPI/MMC driver no, as far as I can see it is always set to 0xFF there if set explicitly, i.e. except for the read-only transfers. > what i've seen so far is that the idle val doesnt matter to most spi devices. > so on the Blackfin side, we've been defaulting to 0xff so that things do work > "out of the box" for everyone (at least so far). Sounds good - however I had just sent a patch implementing CONFIG_SPI_IDLE_VAL. (I am quite hesitating to simply change such a basic default value because I fear there might be other devices around relying on current behaviour. On the other hand, my implementation adds 7 lines of preprocessor stuff, which also does not look really good.) Regards, Wolfgang