From mboxrd@z Thu Jan 1 00:00:00 1970 From: Steven Scholz Date: Mon, 13 Dec 2004 10:52:45 +0100 Subject: [U-Boot-Users] PATCH cpu/at91rm9200/start.S In-Reply-To: <20041213091431.CB0E5C1430@atlas.denx.de> References: <20041213091431.CB0E5C1430@atlas.denx.de> Message-ID: <41BD666D.7060207@imc-berlin.de> 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: > In message <41BD53FA.1010002@imc-berlin.de> you wrote: > >>since enabling the "main oscillator" of Atmel's AT91RM9200 is very board >>specific and must not be done on some boards (which use external clock >>oscilators instead of crystals) I suggest >> >>* Patch by Steven Scholz, 13 Dec 2004: >> Move code for AT91 "main oscillator" into board specific >> memsetup.S files > > > I reject this patch as it causes a duplication of the same code into > several files. My understanding is that this code is poretty generic > for all boards that use it; #undef'fing it for boards which don;t use > it seems to be the better choice here. I did not start the duplication! memsetup.S for at91rm9200dk and cmc_pu2 look pretty much the same already! The clock setup for both boards is done in memsetup.S. I thought about a CFG_USE_MAIN_OSCILLATOR. But since we want to move cpu/at91rm9200 to cpu/arm920t/at91rm9200 someday we realy should not put such board and SoC specific code into start.S. I aggree that duplicating code is bad and a generic cpu/at91rm9200/clocks.S would be better. But why did you accept the other patch * Update for AT91RM9200DK and CMC_PU2 boards: - Enable booting directly from flash in the first place??? -- Steven Scholz