From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marcus Hall Date: Tue, 28 Feb 2006 11:46:59 -0700 Subject: [U-Boot-Users] [PATCH] CFI support for Versatile & Integrator/CP boards In-Reply-To: <200602281024.13958.sr@denx.de> References: <200602281024.13958.sr@denx.de> Message-ID: <44049AA3.2090800@lucent.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de Stefan Roese wrote: > On Monday, 27. February 2006 16:59, Peter Pearse wrote: >>- Changes flash_full_status_check() to test for timeout OR failure, >> rather than timeout AND failure. > > > I'm not sure here. Please take a look at the patch from Marcus Hall: > > http://article.gmane.org/gmane.comp.boot-loaders.u-boot/18530 > > Could you and others (Marcus, Tolunay) please comment on what patch should be > applied here. I tend to take the patch from Marcus right now. Well, there are two variables here. I will abstract it to be Timeout :: flash_status_check returns != ERR_OK Fail :: status != FLASH_STATUS_DONE The following table should correspond to what gets returned by either code patch: Timeout Fail Marcus Peter 0 0 ERR_OK ERR_OK 0 1 ERR_INVAL ERR_INVAL 1 0 ERR_TIMEOUT ERR_INVAL 1 1 ERR_TIMEOUT ERR_INVAL Additionally, Peter's patch may output an additional error message after a timeout if it appears that some error flags are set (but they are not necessarily valid if the flash has timed out) So, I believe that either would work to ensure that if there is an error it does get reported, but I believe that my patch returns a more useful return code and doesn't output potentially confusing error messages. Marcus Hall