December 2008 - U-Boot

[U-Boot] [PATCH v2] Add secondary CPUs processor frequency for e500 core
by Haiying Wang 05 Dec '08

05 Dec '08

This patch updates e500 freqProcessor to array based on CONFIG_NUM_CPUS, and prints each CPU's frequency separately. It also fixes up each CPU's frequency in "clock-frequency" of fdt blob. Signed-off-by: James Yang <James.Yang(a)freescale.com> Signed-off-by: Haiying Wang <Haiying.Wang(a)freescale.com> --- This patch has been tested on MPC8572DS board. Two CPU's frequencies showed up correctly, "fdt print" showed clock-frequency was filled in fdt blob correctly. This v2 patch can be applied cleanly against 85xx-next branch. cpu/mpc85xx/cpu.c | 7 +++++-- cpu/mpc85xx/fdt.c | 15 +++++++++++++-- cpu/mpc85xx/speed.c | 10 ++++++---- include/e500.h | 6 +++++- 4 files changed, 29 insertions(+), 9 deletions(-) diff --git a/cpu/mpc85xx/cpu.c b/cpu/mpc85xx/cpu.c index 782cd71..38186e7 100644 --- a/cpu/mpc85xx/cpu.c +++ b/cpu/mpc85xx/cpu.c @@ -89,6 +89,7 @@ int checkcpu (void) #else u32 ddr_ratio = 0; #endif + int i; svr = get_svr(); ver = SVR_SOC_VER(svr); @@ -140,8 +141,10 @@ int checkcpu (void) get_sys_info(&sysinfo); - puts("Clock Configuration:\n"); - printf(" CPU:%-4s MHz, ", strmhz(buf1, sysinfo.freqProcessor)); + puts("Clock Configuration:\n "); + for (i = 0; i < CONFIG_NUM_CPUS; i++) + printf("CPU%d:%-4s MHz, ", + i,strmhz(buf1, sysinfo.freqProcessor[i])); printf("CCB:%-4s MHz,\n", strmhz(buf1, sysinfo.freqSystemBus)); switch (ddr_ratio) { diff --git a/cpu/mpc85xx/fdt.c b/cpu/mpc85xx/fdt.c index 69fab99..1fae47c 100644 --- a/cpu/mpc85xx/fdt.c +++ b/cpu/mpc85xx/fdt.c @@ -213,6 +213,10 @@ void fdt_add_enet_stashing(void *fdt) void ft_cpu_setup(void *blob, bd_t *bd) { + int off; + int val; + sys_info_t sysinfo; + /* delete crypto node if not on an E-processor */ if (!IS_E_PROCESSOR(get_svr())) fdt_fixup_crypto_node(blob, 0); @@ -228,8 +232,15 @@ void ft_cpu_setup(void *blob, bd_t *bd) "timebase-frequency", bd->bi_busfreq / 8, 1); do_fixup_by_prop_u32(blob, "device_type", "cpu", 4, "bus-frequency", bd->bi_busfreq, 1); - do_fixup_by_prop_u32(blob, "device_type", "cpu", 4, - "clock-frequency", bd->bi_intfreq, 1); + get_sys_info(&sysinfo); + off = fdt_node_offset_by_prop_value(blob, -1, "device_type", "cpu", 4); + while (off != -FDT_ERR_NOTFOUND) { + u32 *reg = (u32 *)fdt_getprop(blob, off, "reg", 0); + val = cpu_to_fdt32(sysinfo.freqProcessor[*reg]); + fdt_setprop(blob, off, "clock-frequency", &val, 4); + off = fdt_node_offset_by_prop_value(blob, off, "device_type", + "cpu", 4); + } do_fixup_by_prop_u32(blob, "device_type", "soc", 4, "bus-frequency", bd->bi_busfreq, 1); diff --git a/cpu/mpc85xx/speed.c b/cpu/mpc85xx/speed.c index 0d55228..b0f47e0 100644 --- a/cpu/mpc85xx/speed.c +++ b/cpu/mpc85xx/speed.c @@ -39,17 +39,19 @@ void get_sys_info (sys_info_t * sysInfo) volatile ccsr_gur_t *gur = (void *)(CONFIG_SYS_MPC85xx_GUTS_ADDR); uint plat_ratio,e500_ratio,half_freqSystemBus; uint lcrr_div; + int i; plat_ratio = (gur->porpllsr) & 0x0000003e; plat_ratio >>= 1; sysInfo->freqSystemBus = plat_ratio * CONFIG_SYS_CLK_FREQ; - e500_ratio = (gur->porpllsr) & 0x003f0000; - e500_ratio >>= 16; /* Divide before multiply to avoid integer * overflow for processor speeds above 2GHz */ half_freqSystemBus = sysInfo->freqSystemBus/2; - sysInfo->freqProcessor = e500_ratio*half_freqSystemBus; + for (i = 0; i < CONFIG_NUM_CPUS; i++) { + e500_ratio = ((gur->porpllsr) >> (i * 8 + 16)) & 0x3f; + sysInfo->freqProcessor[i] = e500_ratio * half_freqSystemBus; + } /* Note: freqDDRBus is the MCLK frequency, not the data rate. */ sysInfo->freqDDRBus = sysInfo->freqSystemBus; @@ -105,7 +107,7 @@ int get_clocks (void) dfbrg = (sccr & SCCR_DFBRG_MSK) >> SCCR_DFBRG_SHIFT; #endif get_sys_info (&sys_info); - gd->cpu_clk = sys_info.freqProcessor; + gd->cpu_clk = sys_info.freqProcessor[0]; gd->bus_clk = sys_info.freqSystemBus; gd->mem_clk = sys_info.freqDDRBus; gd->lbc_clk = sys_info.freqLocalBus; diff --git a/include/e500.h b/include/e500.h index 9d3c841..4c5eeb7 100644 --- a/include/e500.h +++ b/include/e500.h @@ -8,9 +8,13 @@ #ifndef __ASSEMBLY__ +#ifndef CONFIG_NUM_CPUS +#define CONFIG_NUM_CPUS 1 +#endif + typedef struct { - unsigned long freqProcessor; + unsigned long freqProcessor[CONFIG_NUM_CPUS]; unsigned long freqSystemBus; unsigned long freqDDRBus; unsigned long freqLocalBus; -- 1.6.0.2

1 0

Re: [U-Boot] u-boot-2008.10 nand boot not working for SMDK6400
by Alagan S 05 Dec '08

05 Dec '08

hi Guennadi thanks for ur immediate reply... i had flash u-boot-nand.bin into nand, after reseting nothing is coming in console but 4 LEDs(D11-D14) are glowing.i'm using SMDK6400 rev0.1 Thanks & Regards Alagan.S --- On Fri, 5/12/08, Guennadi Liakhovetski <g.liakhovetski(a)gmx.de> wrote: From: Guennadi Liakhovetski <g.liakhovetski(a)gmx.de> Subject: Re: u-boot-2008.10 nand boot not working for SMDK6400 To: "Alagan S" <contact_ssa82(a)yahoo.in> Date: Friday, 5 December, 2008, 8:07 AM Hi Alagan, On Fri, 5 Dec 2008, Alagan S wrote: > hi Guennadi, > Â Â Â Â Â Â Â Â Â Â Â i am working on smdk6400 board.i've downloaded > u-boot-2008.10.i've compiled successfully and loaded u-boot.bin image > into nand flash(say F2) using working u-boot-1.1.6(say F1). > step i followed.... Please, write to the u-boot mailing list with any u-boot questions. If you feel that one specific person can help better than the others you can add him or her to CC, but you _have_ to have the mailing list among recepients. Also if you reply to this mail, please, write to the list, otherwise, I do not think I will be able to reply to you. > 1.booting board with nand-boot using F1 nand flash (u-boot-1.1.6) > 2.removed F1 from nand socket > 3.placed F2(fresh nand) in nand socket > 4.downloaded u-boot.bin in memory(tftp c0008000 u-boot.bin) using tftp > 5.erased F2 (nand erase 0 40000) > 6.has written u-boot.bin into F2 (nand write c0008000 0 40000) > 7.then reset the board with F2 > > but board is not booting. > nand boot support is there?? > what is u-boot-nand.bin??i was trying to write into F2 but i has given error?? u-boot-nand.bin _IS_ the image you need. It is this image that you have to flash to NAND. Thanks Guennadi --- Guennadi Liakhovetski, Ph.D. Freelance Open-Source Software Developer Add more friends to your messenger and enjoy! Go to http://messenger.yahoo.com/invite/

2 1

[U-Boot] [PATCH] net: Fix download command parsing
by Peter Tyser 05 Dec '08

05 Dec '08

When CONFIG_SYS_HUSH_PARSER is defined network download commands with 1 argument in the format 'tftp "/path/file"' do not work as expected. The hush command parser strips the quotes from "/path/file" which causes the network commands to interpret "/path/file" as an address instead of the intended filename. The previous check for a leading quote in netboot_common() was replaced with a check which ensures only valid numbers are treated as addresses. Signed-off-by: Peter Tyser <ptyser(a)xes-inc.com> --- common/cmd_net.c | 20 ++++++++++++-------- 1 files changed, 12 insertions(+), 8 deletions(-) diff --git a/common/cmd_net.c b/common/cmd_net.c index af691a4..1d297ec 100644 --- a/common/cmd_net.c +++ b/common/cmd_net.c @@ -154,8 +154,10 @@ static int netboot_common (proto_t proto, cmd_tbl_t *cmdtp, int argc, char *argv[]) { char *s; + char *end; int rcode = 0; int size; + ulong addr; /* pre-set load_addr */ if ((s = getenv("loadaddr")) != NULL) { @@ -166,15 +168,17 @@ netboot_common (proto_t proto, cmd_tbl_t *cmdtp, int argc, char *argv[]) case 1: break; - case 2: /* only one arg - accept two forms: - * just load address, or just boot file name. - * The latter form must be written "filename" here. + case 2: /* + * Only one arg - accept two forms: + * Just load address, or just boot file name. The latter + * form must be written in a format which can not be + * mis-interpreted as a valid number. */ - if (argv[1][0] == '"') { /* just boot filename */ - copy_filename (BootFile, argv[1], sizeof(BootFile)); - } else { /* load address */ - load_addr = simple_strtoul(argv[1], NULL, 16); - } + addr = simple_strtoul(argv[1], &end, 16); + if (end == (argv[1] + strlen(argv[1]))) + load_addr = addr; + else + copy_filename(BootFile, argv[1], sizeof(BootFile)); break; case 3: load_addr = simple_strtoul(argv[1], NULL, 16); -- 1.6.0.2.GIT

2 1

[U-Boot] [PATCH] net: tsec: Fix Marvell 88E1121R phy init
by Anatolij Gustschin 05 Dec '08

05 Dec '08

This patch tries to ensure that phy interrupt pin won't be asserted after booting. We experienced following issues with current 88E1121R phy init: Marvell 88E1121R phy can be hardware-configured to share MDC/MDIO and interrupt pins for both ports P0 and P1 (e.g. as configured on socrates board). Port 0 interrupt pin will be shared by both ports in such configuration. After booting Linux and configuring eth0 interface, port 0 phy interrupts are enabled. After rebooting without proper eth0 interface shutdown port 0 phy interrupts remain enabled so any change on port 0 (link status, etc.) cause assertion of the interrupt. Now booting Linux and configuring eth1 interface will cause permanent phy interrupt storm as the registered phy 1 interrupt handler doesn't acknowledge phy 0 interrupts. This of course should be fixed in Linux driver too. Signed-off-by: Anatolij Gustschin <agust(a)denx.de> --- drivers/net/tsec.c | 3 +++ include/tsec.h | 4 ++++ 2 files changed, 7 insertions(+), 0 deletions(-) diff --git a/drivers/net/tsec.c b/drivers/net/tsec.c index d7da081..2ced585 100644 --- a/drivers/net/tsec.c +++ b/drivers/net/tsec.c @@ -1196,6 +1196,9 @@ struct phy_info phy_info_M88E1121R = { {MIIM_88E1121_PHY_LED_CTRL, miim_read, &mii_88E1121_set_led}, {MIIM_CONTROL, MIIM_CONTROL_INIT, &mii_cr_init}, + /* Disable IRQs and de-assert interrupt */ + {MIIM_88E1121_PHY_IRQ_EN, 0, NULL}, + {MIIM_88E1121_PHY_IRQ_STATUS, miim_read, NULL}, {miim_end,} }, (struct phy_cmd[]){ /* startup */ diff --git a/include/tsec.h b/include/tsec.h index d2951f6..7b52e06 100644 --- a/include/tsec.h +++ b/include/tsec.h @@ -226,6 +226,10 @@ #define MIIM_88E1121_PHY_LED_PAGE 3 #define MIIM_88E1121_PHY_LED_DEF 0x0030 +/* 88E1121 PHY IRQ Enable/Status Register */ +#define MIIM_88E1121_PHY_IRQ_EN 18 +#define MIIM_88E1121_PHY_IRQ_STATUS 19 + #define MIIM_88E1121_PHY_PAGE 22 /* 88E1145 Extended PHY Specific Control Register */ -- 1.5.3.3

3 2

[U-Boot] [PATCH v2] net: Fix TftpStart() ip:filename bug
by Peter Tyser 05 Dec '08

05 Dec '08

The TftpStart() function modifies the 'BootFile' string when 'BootFile' contains both an IP address and filename (eg 1.2.3.4:/path/file). This causes subsequent calls to TftpStart to incorrectly parse the TFTP filename and server IP address to use. For example: => tftp 0x100000 10.52.0.62:/home/ptyser/non_existant Speed: 100, half duplex Using eTSEC1 device TFTP from server 10.52.0.62; our IP address is 10.52.253.79 ^^^^^^^^^^ CORRECT Filename '/home/ptyser/non_existant'. ^^^^^^^^^^^^^^^^^^^^^^^^^ CORRECT Load address: 0x100000 Loading: * TFTP error: 'File not found' (1) Starting again eTSEC2: No link. Speed: 100, half duplex Using eTSEC1 device TFTP from server 10.52.0.33; our IP address is 10.52.253.79 ^^^^^^^^^^ WRONG Filename '10.52.0.62'. ^^^^^^^^^^ WRONG Load address: 0x100000 Loading: * TFTP error: 'File not found' (1) Starting again TftpStart() was modified to not modify the 'BootFile' string. Signed-off-by: Peter Tyser <ptyser(a)xes-inc.com> --- doah, forgot SOB net/tftp.c | 3 +-- 1 files changed, 1 insertions(+), 2 deletions(-) diff --git a/net/tftp.c b/net/tftp.c index ce6ea3d..3dac3d8 100644 --- a/net/tftp.c +++ b/net/tftp.c @@ -499,9 +499,8 @@ TftpStart (void) strncpy(tftp_filename, BootFile, MAX_LEN); tftp_filename[MAX_LEN-1] = 0; } else { - *p++ = '\0'; TftpServerIP = string_to_ip (BootFile); - strncpy(tftp_filename, p, MAX_LEN); + strncpy(tftp_filename, p + 1, MAX_LEN); tftp_filename[MAX_LEN-1] = 0; } } -- 1.6.0.2.GIT

2 1

[U-Boot] [PATCH 0/2] net: Additional IP fragmentation check
by Peter Tyser 05 Dec '08

05 Dec '08

It looks like U-Boot ignores fragmented IP packets with non-zero "fragment offset" fields, but doesn't ignore the initial fragmented IP packet which has a "fragment offset" field value of 0. An additional check was added to catch the initial fragmented packet which should have the "more fragments" bit set in its flags field. The bug initially resulted in TFTP transfers which appeared to work, but in reality failed as some fragmented packets were received, others were not. With these patches applied a TFTP download from a server with a low MTU results in a timeout as U-Boot drops all incominig fragmented TFTP data packets. I quickly tried the "Add simple IP/UDP fragmentation support" patch in the net/testing repository, but it did not work out of the box. It looks like U-Boot was assembling the fragmented packets correctly based on a memory display of &NetFragBuf, but the assembled packet was not making its way up the stack FWIW. Peter Tyser (2): net: Define IP flag field values net: Add additional IP fragmentation check include/net.h | 6 ++++++ net/net.c | 8 +++++--- 2 files changed, 11 insertions(+), 3 deletions(-)

2 3

Re: [U-Boot] Is arm-2008-11-24.iso the right image to compile U Boot for ARM ?
by Diptopal Basu 05 Dec '08

05 Dec '08

Hi Wolfgang , I am unable to download a linux kernel for ARM from the site under git.denx.de . Where am I going wrong ? I can see the tree but no zip file which has the source . Suppose if I downloaded uC Linux from some other source say uCLinux , would it be possible to compile it with the ELDK tool ? Please advise . Regards Diptopal Basu On Wed, Dec 3, 2008 at 6:33 PM, Wolfgang Denk <wd(a)denx.de> wrote: > Dear "Diptopal Basu", > > In message <bd97640b0812030356s7cc1f52chbcc60b9b634e74ef(a)mail.gmail.com> > you wrote: > > > > Thanks a great deal , I actually saw the readme . I could compile an > > existing image successfully (at91rm9200dk) . Though it is yet to be tried > > out ,( I am working on the skyeye simulator ) . But the compilation goes > > through fine . One question , if I want to know the board memory layout > of > > this image , where should I look ? > > In the board documentation nd the board config files. > > Please post U-Boot related questions on the mailing list - I don't > know much about the at91rm9200dk ... > > 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: wd(a)denx.de > Sometimes a feeling is all we humans have to go on. > -- Kirk, "A Taste of Armageddon", stardate 3193.9 >

1 0

[U-Boot] how compilor know where .text section starts from?
by hank.peng 05 Dec '08

05 Dec '08

Hi, everybody: I am new to uboot and reading source code, I have a question: how does compilor know where .text section starts from? For example, I am using MPC8548CDS, document says .text section begins from 0xfff8000, but I haven't find this value in u-boot.lds? So it must be specified in some other place, right? -- The simplest is not all best but the best is surely the simplest!

4 3

[U-Boot] AT91 Pull request
by Jean-Christophe PLAGNIOL-VILLARD 05 Dec '08

05 Dec '08

Hi Wolfgang, please pull The following changes since commit 2077e348c2a84901022ad95311b47b70361e6daa: Scott Wood (1): NAND: Fix misplaced return statement in nand_{read,write}_skip_bad(). are available in the git repository at: git://git.denx.de/u-boot-at91.git master Ilko Iliev (2): fix some coding style violations. mtd/dataflash.c: fix a problem with the last partition Jean-Christophe PLAGNIOL-VILLARD (2): AT91: remove non supported board AT91RM9200DF macro at91rm9200dk: Fix typo cpu/arm920t/start.S | 4 ++-- drivers/mtd/dataflash.c | 25 ++++++++++++------------- include/configs/at91rm9200dk.h | 2 +- 3 files changed, 15 insertions(+), 16 deletions(-) Best Regards, J.

2 1

[U-Boot] Pull Request: u-boot-usb.git
by Remy Bohmer 05 Dec '08

05 Dec '08

The following changes since commit 2077e348c2a84901022ad95311b47b70361e6daa: Scott Wood (1): NAND: Fix misplaced return statement in nand_{read,write}_skip_bad(). are available in the git repository at: git://git.denx.de/u-boot-usb.git master Jean-Christophe PLAGNIOL-VILLARD (1): drivers/usb: Move conditional compilation to Makefile Michael Trimarchi (2): USB style patch, 80 chars strict Remove obsolete command (apply afte USB style patch, 80 chars strict) Remy Bohmer (1): Remove non-ascii characters from fat code common/cmd_usb.c | 596 ++++++++++++++++++++------------------- common/usb.c | 40 ++- drivers/usb/Makefile | 19 +- drivers/usb/isp116x-hcd.c | 4 - drivers/usb/sl811_usb.c | 3 - drivers/usb/usb_ohci.c | 4 - drivers/usb/usbdcore_ep0.c | 4 - drivers/usb/usbdcore_mpc8xx.c | 4 - drivers/usb/usbdcore_omap1510.c | 4 - fs/fat/fat.c | 4 +- include/fat.h | 2 +- include/usb.h | 222 ++++++++------- 12 files changed, 477 insertions(+), 429 deletions(-)

2 1