Hi Vladimir,
Le 10/02/2012 21:14, Vladimir Zapolskiy a écrit :
Hi Albert,
On 10.02.2012 20:00, Albert ARIBAUD wrote:
Hi Vladimir,
Le 13/01/2012 01:42, Vladimir Zapolskiy a écrit :
diff --git a/arch/arm/cpu/arm926ejs/lpc32xx/cpu.c
b/arch/arm/cpu/arm926ejs/lpc32xx/cpu.c
new file mode 100644
index 0000000..e29e130
--- /dev/null
+++ b/arch/arm/cpu/arm926ejs/lpc32xx/cpu.c
@@ -0,0 +1,70 @@
+/*
+ * Copyright (C) 2011 by Vladimir Zapolskiy<v...@mleia.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#include<common.h>
+#include<asm/arch/cpu.h>
+#include<asm/arch/clk.h>
+#include<asm/arch/wdt.h>
+#include<asm/io.h>
+
+static struct clk_pm_regs *clk = (struct clk_pm_regs *)CLK_PM_BASE;
+static struct wdt_regs *wdt = (struct wdt_regs *)WDT_BASE;
+
+void reset_cpu(ulong addr)
+{
+ /* Enable watchdog clock */
+ setbits_le32(&clk->timclk_ctrl, CLK_TIMCLK_WATCHDOG);
+
+ /* Reset pulse length is 13005 peripheral clock frames */
+ writel(13000,&wdt->pulse);
I can't help but ask, if the comment says 13005, then why does the code
say 13000?
the provided comment is done according to the descriptions of
WDTIM_PULSE register from User Manual, for register value X it gives the
RESET pulse length of (X + 5) PERIPH_CLKs.
+ /* Force WDOG_RESET2 and RESOUT_N signal active */
+ writel(WDTIM_MCTRL_RESFRC2 | WDTIM_MCTRL_RESFRC1 | WDTIM_MCTRL_M_RES2,
+ &wdt->mctrl);
+
+ while (1)
+ /* NOP */;
+}
+
+#if defined(CONFIG_ARCH_CPU_INIT)
+int arch_cpu_init(void)
+{
+ /*
+ * It might be necessary to flush data cache, if U-boot is loaded
+ * from kickstart bootloader, e.g. from S1L loader
+ */
+ flush_dcache_all();
+
+ return 0;
+}
+#else
+#error "You have to select CONFIG_ARCH_CPU_INIT"
+#endif
No other code ever has this kind of #error on CONFIG_ARCH_CPU_INIT not
being defined. What's the point of it? People won't suddenly decide to
undefine CONFIG_ARCH_CPU_INIT for fun, I think.
Generally for all future boards it's a good idea to clean up caches
after a bootstrap/kickstart bootloader's work, which is expected to be
present mostly on all LPC32XX boards.
The macro CONFIG_ARCH_CPU_INIT is defined by default in header
include/asm/arch0lpc32xx/config.h, which is included in board
configuration files for shared definitions, but of course technically
for some of future boards it can be undefined after inclusion of the
header.
Originally I didn't have such an #else case, however Marek convinced me
to add the clause as a clear enough reminder about its importance for
board developers.
If you think this #else case is redundant, I won't argue and it could be
just removed.
diff --git a/arch/arm/include/asm/arch-lpc32xx/sys_proto.h
b/arch/arm/include/asm/arch-lpc32xx/sys_proto.h
new file mode 100644
index 0000000..b2cbfcd
--- /dev/null
+++ b/arch/arm/include/asm/arch-lpc32xx/sys_proto.h
@@ -0,0 +1,25 @@
+/*
+ * Copyright (C) 2011 Vladimir Zapolskiy<v...@mleia.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston,
+ * MA 02110-1301, USA.
+ */
+
+#ifndef _LPC32XX_SYS_PROTO_H
+#define _LPC32XX_SYS_PROTO_H
+
+void lpc32xx_uart_init(unsigned int uart_id);
+
+#endif /* _LPC32XX_SYS_PROTO_H */
Do we need this .h for a single function when we already have uart.h?
I think we need it, I found that's quite a common practice to have such
a header with SoC specific declarations for boards, and these headers
have the same name. I expect that later on this header will contain SoC
EMAC specific declarations and may be something else depending on U-boot
support of other LPC32XX peripherals.
Fair enough.
Thank you for review.
--
With best wishes,
Vladimir
Amicalement,
--
Albert.
_______________________________________________
U-Boot mailing list
U-Boot@lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot