Dear Macpaul Lin, In message <[email protected]> you wrote: > --===============1276623933== > > Add nds32 architecture with generic cpu core support. > Add nds32 architecture with n1213s cpu core support.
Please do not send patches as attachments, send them inline. Try using "git format-patch" to create the patches and "git send-email" to send them. Don't indent the commit messages like that. > diff --git a/arch/nds32/config.mk b/arch/nds32/config.mk > new file mode 100644 > index 0000000..e88e516 > --- /dev/null > +++ b/arch/nds32/config.mk > @@ -0,0 +1,32 @@ > +# > +# (C) Copyright 2000-2002 > +# Wolfgang Denk, DENX Software Engineering, [email protected]. > +# > +# (C) Copyright 2006 > +# Shawn Lin, Andes Technology Corporation <[email protected]> > +# Macpaul Lin, Andes Technology Corporation <[email protected]> So no work has been done on this in the last 4 years? I guess you want to update your Copyright messages - all of them. ... > +START = start.o > +COBJS = interrupts.o cpu.o Please keep all such lists sorted. 'c' < 'i'. ... > +void do_reset (cmd_tbl_t *cmdtp, bd_t *bd, int flag, int argc, char *argv[]) > +{ > + extern void reset_cpu(ulong addr); > + > + disable_interrupts(); > +// reset_cpu(0); FIXME: -by Shawn, currently no ROM loader at addr 0 No C++ comments allowd. Please fix globally. > +void flush_cache (unsigned long dummy1, unsigned long dummy2) > +{ > +/* > + unsigned long u32IfRun = 0; > + > + if(u32IfRun) return; > + u32IfRun = 1; We do not allow CamelCase identifiers in U-Boot. Please fix globally. Also, pay attentiuon to indentation rules, the "return" belongs on a new line, indented by a TAB. > + reset_cpu((unsigned long)flush_cache); > + return; > +*/ Please do not add dead code. Remove it. Please fix globally. > --- /dev/null > +++ b/arch/nds32/cpu/interrupts.c ... > +//#include <board/AndesTech/include/porting.h> > +#include "../../../board/AndesTech/include/porting.h" This is fundeamentally flawed. Global code must never include any board specific header files. > +/* > + * TimerBase[0] is a NULL entry > + */ > +fLib_TimerReg *TimerBase[] ={0, (fLib_TimerReg *) NDS32_COMMON_TIMER1_BASE, > + (fLib_TimerReg *) NDS32_COMMON_TIMER2_BASE,(fLib_TimerReg > *)NDS32_COMMON_TIMER3_BASE}; Lines too long. Please fix globally. > +ulong get_timer_masked(void) > +{ > + ulong now = Read_Timer_Counter(1);; > + > + if (lastdec >= now) > + { > + /* normal mode */ > + timestamp += lastdec - now; > + } else { > + /* we have an overflow ... */ > + timestamp += lastdec + TIMER_LOAD_VAL - now; > + } Incorrect brace style. Please fix globally. Please re-read the CodingStyle document!!! ... > +uart_init: > +#! 1. fLib_SetSerialMode(DebugSerialPort, SERIAL_MDR_UART); > + li $r0, #0x99600000 !; CPE_UART4_BASE = 0x99600000 > + lwi $r1, [$r0+#0x20] !; mdr = mdr = > inw(CPE_UART4_BASE + SERIAL_MDR) > + li $r2, #0xfffffffc !; mdr &= ~(0x3) ; > SERIAL_MDR_MODE_SEL = 0x3 > + and $r1, $r1, $r2 > + swi $r1, [$r0+#0x20] !; outw(CPE_UART4_BASE + > SERIAL_MDR, mdr | mode); > +#! 2. fLib_SerialInit(DebugSerialPort, (int)DEFAULT_HOST_BAUD, PARITY_NONE, > 0, 8); > +#! dgbserialport = 0x99600000; baud = (UART_CLOCK / 614400) = 24; > PARITY_NONE = 0; num = 0; len = 8; > + li $r0, #0x99600000 !;lcr = inw(port + SERIAL_LCR) > & ~SERIAL_LCR_DLAB; SERIAL_LCR = 0x0c > + lwi $r1, [$r0+#0x0c] > + li $r2, #0xffffff7f !; SERIAL_LCR_DLAB = 0x80 > + and $r1, $r1, $r2 > + li $r2, #0x80 !; outw(port + > SERIAL_LCR,SERIAL_LCR_DLAB); > + swi $r2, [$r0+#0x0c] > + li $r2, #0 !; outw(port + SERIAL_DLM, > ((baudrate & 0xf00) >> 8)); > + swi $r2, [$r0+#0x4] > +#ifdef __NDS32_N1213_43U1H__ /* AG101 */ > + li $r2, #60 !; outw(port + SERIAL_DLL, > (baudrate & 0xff)); > +#else > + li $r2, #24 !; outw(port + SERIAL_DLL, > (baudrate & 0xff)); > +#endif > + swi $r2, [$r0+#0x0] > + andi $r1, $r1, #0xc0 !; lcr &= 0xc0; > + li $r2, #3 !; len-=5; > + or $r1, $r1, $r2 !; lcr|=len; > + swi $r1, [$r0+#0x0c] !; outw(port+SERIAL_LCR,lcr); > +#! 3. fLib_SetSerialFifoCtrl(DebugSerialPort, 0, 0, ENABLE(1), ENABLE(1)); > + li $r0, #0x99600000 > + li $r1, #0x7 !; fcr = 0x7 > + swi $r1, [$r0+#0x8] !; SERIAL_FCR = 0x08; > outw(port+SERIAL_FCR,fcr); > + > + ret Lines too long - why don't you implement uart_init() in C? > +void cleanup_before_linux(void) > +{ > + /* > + * this function is called just before we call linux > + * it prepares the processor for linux > + * > + * we disable interrupt and caches. > + */ > + > +#ifdef CONFIG_MMU > + unsigned long i; > +#endif Please move such block comments outside the functions. Functions should normally start with declarations. Please apply globally. > diff --git a/arch/nds32/cpu/n1213s/config.mk b/arch/nds32/cpu/n1213s/config.mk > new file mode 100644 > index 0000000..316332e > --- /dev/null > +++ b/arch/nds32/cpu/n1213s/config.mk ... > +# > + > +PLATFORM_CPPFLAGS += This is a noop - why do we need this file? > diff --git a/arch/nds32/cpu/n1213s/interrupts.c > b/arch/nds32/cpu/n1213s/interrupts.c > new file mode 100644 > index 0000000..a115a72 > --- /dev/null > +++ b/arch/nds32/cpu/n1213s/interrupts.c ... > + */ > + > +#include "../interrupts.c" Is this all? Then why do we need this file? > diff --git a/arch/nds32/cpu/n1213s/lowlevel_init.S > b/arch/nds32/cpu/n1213s/lowlevel_init.S > new file mode 100644 > index 0000000..26e0727 > --- /dev/null > +++ b/arch/nds32/cpu/n1213s/lowlevel_init.S ... > +#include "../lowlevel_init.S" Ditto? > diff --git a/arch/nds32/cpu/n1213s/start.S b/arch/nds32/cpu/n1213s/start.S > new file mode 100644 > index 0000000..e93d048 > --- /dev/null > +++ b/arch/nds32/cpu/n1213s/start.S ... > +#include "../start.S" And again. This makes no sense. Please find a better way to use common code. ... > +__start_andesboot: .word start_andesboot > + > + > + > +!========================================================================= Too many blank lines. absolute maximum is 2. Please fix globally. Best regards, Wolfgang Denk -- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: [email protected] The high cost of living hasn't affected its popularity. _______________________________________________ U-Boot mailing list [email protected] http://lists.denx.de/mailman/listinfo/u-boot
