March 2011 - U-Boot

[U-Boot] git merge summary
by Alexander Holler 11 Mar '11

11 Mar '11

Hello, may I ask if there is reason why merge commits don't have a summary? I find those summaries (enabled with git config [--global] merge.summary true) very hand to see what was changed with the merge commit. Regards, Alexander

5 11

[U-Boot] AT91 timer patch broke DataFlash driver
by Harald Kipp 10 Mar '11

10 Mar '11

Hi, Commit 5dca710a3d7703e41da0e9894f2d71f9e25bea6b http://lists.denx.de/pipermail/u-boot/2010-October/078600.html removed the two global routines reset_timer_masked() and get_timer_masked() from arch/arm/cpu/arm926ejs/at91/timer.c. Both routines are still referenced in drivers/spi/atmel_dataflash_spi.c. As a result SPI DataFlash support fails to compile for AT91 boards. This had been also reported by http://lists.denx.de/pipermail/u-boot/2011-January/085399.html http://lists.denx.de/pipermail/u-boot/2010-November/081287.html but the problem still exists in the master trunk and the at91 branch. For a first test I simply added void reset_timer_masked(void) { reset_timer(); } ulong get_timer_masked(void) { return tick_to_time(get_ticks() - gd->timer_reset_value); } to drivers/spi/atmel_dataflash_spi.c. All DataFlash functions seem to work with this. But I'm not sure if this is a proper replacement for the original code. Regards, Harald

1 1

[U-Boot] uboot for MIPS: need help to skip relocate uboot and start uboot from RAM
by Pandurang Kale 10 Mar '11

10 Mar '11

Hello Everyone, We have MIPS based development board and I am trying to get the uboot up and running on it. We have a primary bootloader which check for the valid mod-image stored on the flash. This mod-image consist of header, uboot and linux kernel. Depending on the recent, active and stable imagethe primary bootloader first copies the uboot image. and later on we would copy the kernel image from uboot. But when primary bootloader copies the uboot image to the RAM and passes the control to the uboot, uboot (MIPS version of start.S and arch/mips/lib/borad.c) tries to relocate the already copied image from RAM (the primary bootloader copied it to start of the RAM+1MB address) to top of the RAM (0x87fc0000) region thinking that the uboot image is stored in flash. All I need to do is skip the uboot relocate code in MIPS version of uboot startup as the primary bootloader has already relocated the uboot from Flash to RAM and set up the stack pointer and other global data appropriately, which it does after relocation. I can see there is a switch for ARM processor, CONFIG_SKIP_RELOCATE_UBOOT, which skips the relocation of uboot code and tries to run the uboot from RAM. I cannot see a similar switch implemented for MIPS and didnt find any related thread anywhere in mailing list or on net. Do we have similar ARM like switch to SKIP the RELOCATION? If not has anyone done this before? I would really appreciate if you can guide me to overcome this issue to run the uboot cleanly skipping the relocation. Thanks in advance, Pandu

4 8

[U-Boot] errno confusion for fw_env?
by Robert McCullough 10 Mar '11

10 Mar '11

Hi Detlev, Did you ever get a solution for this problem? Best regards, Robert McCullough Senior Software Engineer Promess Incorporated 11429 Grand River Road Brighton, Michigan 48116 Email: <mailto://rob@promessdev.com> mailto://rob@promessinc.com Web: <http://promessinc.com/> http://promessinc.com Tel: (810) 229-9334 Fax: (810) 229-8125

1 0

[U-Boot] MPC837x PCIe no link
by Andre Schwarz 10 Mar '11

10 Mar '11

All, sorry to bother you again. After my harddisk died I'm porting my latest 8377 work to TOT. Unfortunately I'm lacking some commits that haven't been pushed to backup. But everything's fine so far and the system is up and running, except PCIe. Using the "old" binary left on the server PCIe has been working fine : U-Boot 2010.12-rc1-00596-gcef659f-dirty (Nov 22 2010 - 17:22:20) MPC83XX ... PCIE1: link and the Wifi card is present and usable by Linux. => pci 2 Scanning PCI devices on bus 2 BusDevFun VendorId DeviceId Device Class Sub-Class _____________________________________________________________ 02.00.00 0x168c 0x001c Network controller 0x00 With latest code running on the *same piece of hardware* U-Boot 2011.03-rc1-00129-g1ce1ed1-dirty (Mar 04 2011 - 16:31:40) MPC83XX I get "PCIE1: No link" Dumping SerDes (e3000 + e3100) and PCIe core (400-7ff) registers shows now difference between both versions. Any ideas what might be responsible for the link not being established ? Regards, André MATRIX VISION GmbH, Talstrasse 16, DE-71570 Oppenweiler Registergericht: Amtsgericht Stuttgart, HRB 271090 Geschaeftsfuehrer: Gerhard Thullner, Werner Armingeon, Uwe Furtner

3 3

[U-Boot] errno confusion for fw_env?
by Detlev Zundel 10 Mar '11

10 Mar '11

Hi, with current master I tried to cross-compile a fw_env tool for the da850evm configuration. So I tried the "standard scheme" (means it worked a few years ago) make distclean make da850evm_config make env Doing this I noticed that commit abd315a1357ab05e82f0d61ebad666bc0d5378c3 disabled cross-compilation. So I reverted the commit locally on my tree to enable cross-compilation again, but now I get: [dzu@pollux u-boot-testing (master)]$ make env make -C tools/env all MTD_VERSION= make[1]: Entering directory `/work/dzu/git/u-boot-testing/tools/env' arm-linux-gcc -Wall -DUSE_HOSTCC -I/work/dzu/git/u-boot-testing/include crc32.c fw_env.c fw_env_main.c -o fw_printenv /opt/eldk-4.2-2008-11-24/usr/bin/../lib/gcc/arm-linux-gnueabi/4.2.2/../../../../arm-linux-gnueabi/bin/ld: errno: TLS definition in /opt/eldk-4.2-2008-11-24/usr/../arm/lib/libc.so.6 section .tbss mismatches non-TLS reference in /tmp/cc45c402.o /opt/eldk-4.2-2008-11-24/usr/../arm/lib/libc.so.6: could not read symbols: Bad value collect2: ld returned 1 exit status make[1]: *** [fw_printenv] Error 1 make[1]: Leaving directory `/work/dzu/git/u-boot-testing/tools/env' make: *** [env] Error 2 The cross-compilation and the "-DUSE_HOSTCC" doesn't look right somehow. And moreover we now pull in the "U-Boot errno" stuff for a regular user-space tool through "-I$(SRCTREE)/include". The latter comes from this commit: commit f3b6d528e4dd719640a4bfcd954f4e4c7f5db0d6 Author: Grant Erickson <gerickson(a)nuovations.com> Date: Tue May 6 16:18:00 2008 -0700 Fix Compilation Errors with 'tools/env/fw_printenv' In the current top-of-tree, 1.3.3.-rc2, the optional tool 'tools/env/fw_printenv' fails to compile for two reasons: 1) The header watchdog.h cannot be found. 2) The header zlib.h is picked up from the tool chain rather than the project causing a prototype conflict for crc32. This patch addresses both of these issues. Platforms Tested On: - AMCC "Kilauea" Signed-off-by: Grant Erickson <gerickson(a)nuovations.com> It was probably not anticipated that this now picks up errno.h from the project as well. Does anybody have a good idea on how to tackle this? Thanks Detlev -- These days, even the most pure and abstract mathematics is in danger to be applied. -- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-40 Fax: (+49)-8142-66989-80 Email: dzu(a)denx.de

2 1

[U-Boot] [RFC 0/3] mmc inizialization steps enhanced
by Raffaele Recalcati 10 Mar '11

10 Mar '11

I have added some better inizializations and status check. I have created following patchset for Davinci dm365 against git://arago-project.org/git/projects/u-boot-davinci.git git tree. There are two commits more in that tree, but they are not in conflict with my work. I have finally tested this series on basi board, based on Davinci dm365, that contains an eMMC device. Raffaele Recalcati (3): mmc: checking status after commands with R1b response mmc: SEND_OP_COND considers card capabilities (voltage) mmc: trace added drivers/mmc/mmc.c | 198 +++++++++++++++++++++++++++++++++++++++++++++-------- include/mmc.h | 6 ++ 2 files changed, 174 insertions(+), 30 deletions(-)

2 5

[U-Boot] [PATCH 4/4] S5P: universal: Enable the pwm driver
by Minkyu Kang 10 Mar '11

10 Mar '11

Signed-off-by: Minkyu Kang <mk7.kang(a)samsung.com> --- include/configs/s5pc210_universal.h | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/include/configs/s5pc210_universal.h b/include/configs/s5pc210_universal.h index c033a8d..5915984 100644 --- a/include/configs/s5pc210_universal.h +++ b/include/configs/s5pc210_universal.h @@ -70,6 +70,9 @@ #define CONFIG_MMC 1 #define CONFIG_S5P_MMC 1 +/* PWM */ +#define CONFIG_PWM 1 + /* It should define before config_cmd_default.h */ #define CONFIG_SYS_NO_FLASH 1 -- 1.7.1

1 0

[U-Boot] [PATCH 3/4] S5P: goni: Enable the pwm driver
by Minkyu Kang 10 Mar '11

10 Mar '11

Signed-off-by: Minkyu Kang <mk7.kang(a)samsung.com> --- include/configs/s5p_goni.h | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/include/configs/s5p_goni.h b/include/configs/s5p_goni.h index f53e798..597cf57 100644 --- a/include/configs/s5p_goni.h +++ b/include/configs/s5p_goni.h @@ -69,6 +69,9 @@ #define CONFIG_MMC 1 #define CONFIG_S5P_MMC 1 +/* PWM */ +#define CONFIG_PWM 1 + /* It should define before config_cmd_default.h */ #define CONFIG_SYS_NO_FLASH 1 -- 1.7.1

1 0

[U-Boot] [PATCH 2/4] S5P: smdkc100: Enable the pwm driver
by Minkyu Kang 10 Mar '11

10 Mar '11

Signed-off-by: Minkyu Kang <mk7.kang(a)samsung.com> --- include/configs/smdkc100.h | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/include/configs/smdkc100.h b/include/configs/smdkc100.h index fb8b43c..1ae1234 100644 --- a/include/configs/smdkc100.h +++ b/include/configs/smdkc100.h @@ -68,6 +68,9 @@ #define CONFIG_SERIAL0 1 /* use SERIAL 0 on SMDKC100 */ #define CONFIG_SERIAL_MULTI 1 +/* PWM */ +#define CONFIG_PWM 1 + /* allow to overwrite serial and ethaddr */ #define CONFIG_ENV_OVERWRITE #define CONFIG_BAUDRATE 115200 -- 1.7.1

1 0