From mboxrd@z Thu Jan 1 00:00:00 1970 From: Joakim Tjernlund Date: Tue, 3 Nov 2009 14:30:17 +0100 Subject: [U-Boot] Does uboot EBS(erase block summary) to reduce JFFS2 scaning time? In-Reply-To: <20091103132108.E06963F6EC@gemini.denx.de> References: <20091103132108.E06963F6EC@gemini.denx.de> Message-ID: List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de Wolfgang Denk wrote on 03/11/2009 14:21:08: > > Dear Joakim Tjernlund, > > In message 003BFEAC at transmode.se> you wrote: > > > > > > HeLei wrote on 03/11/2009 09:21:04: > > > > > > > > Thank you, Jocke > > > > > > > - impl. a better crc32(use the one from linux) > > > > > > Attaching a very crude port of linux crc32. This boots a linux img > > > for me and handles the environment crc as well. Feel free > > > to clean it up and submit to u-boot. > > > > > > Jocke > > > > So I could not help myself, here is a better port of crc32 to u-boot. > > You will probably get at small conflict due to LINK_OFF, just remove > > the LINK_OFF stuff for now. > > Could you test and report? > > Do you have a little or big endian target? > > I understand you will resend this patch with a proper Subject: for > real review on the list? Yes, I just wanted some external testing but this seems not to happen. > > You also might want to explain in which way this patch is "more > efficient" - in terms of memory footprint, or execution time, or > both? And by how much? Tested in which envrionment(s) ? hmm, I did this optimization many years ago for JFFS2 in linux. I don't have any numbers but I can give you some hints w.r.t number of insn in the inner loop(later though). Footprint will be higher. Jocke