April 2009 - U-Boot

[U-Boot] i2c: rework multibus/multiadapter functionality
by Heiko Schocher 18 Jul '09

18 Jul '09

Hello, I want now, because the merge window is open again, restart the the multibus/multiadapter discussion. To have a base for this discusion, I made in the i2c git tree (git://git.denx.de/u-boot-i2c.git) the following new 2 branches: "multibus": Patches from Sergey Kubushyn <ksi(a)koi8.net> posted in the last merge window. see, http://lists.denx.de/pipermail/u-boot/2009-February/047465.html They were discussed, but not accepted ... also some CodingStyle comments, changes in patch hierarchic are not done ... I only synced them to actuall code to have a base for the discussion. "multibus_v2": (I couldn;t post the patches because one of them is bigger than 100k :-( and I cannot split it in 2 patches, because this would break git bisection compatibility. But the patches can be found in git://git.denx.de/u-boot-i2c.git): Based on the i2c_core from Sergeys patches, added the following suggestions from Wolfgang and me: - v2 is bisection compatible - commits are in (hopefully) logical blocks - "cur_adap" added in gd_t: This points allways to the actuall used i2c adapter: - because gd_t is writeable when running in flash, complete multiadapter/multibus functionality is usable, when running in flash - using a pointer brings faster accesses to the i2c_adapter_t struct and saves some bytes here and there (see later). - init from a i2c controller: In the "multibus" branch (also in actual code) all i2c controllers, as a precaution, getting initialized. In the "multibus_v2" branch, a i2c controller gets only initialized if it is used. This is done in i2c_set_bus_num(). Actually, I let the i2c_init_all() in code, but just call in this function i2c_set_bus_num(CONFIG_SYS_SPD_BUS_NUM). This can be dropped in a second step completely. Also, with this approach, we can easy add in a second step, a i2c_deinit() function in i2c_set_bus_num(), so we can deactivate a no longer used controller. - added "hw_adapnr" in i2c_adapter_t struct: when for example a CPU has more then one i2c controller we can use this variable to differentiate which controller the actual i2c adapter uses. This results in lower codesize and lower sourcecode changes. Example: fsl_i2c driver: (MPC8548CDS only with fsl driver) multibus: [hs@pollux u-boot-i2c]$ ./MAKEALL MPC8548CDS Configuring for MPC8548CDS board... fsl_i2c.c: In function '__i2c_init': fsl_i2c.c:173: warning: assignment discards qualifiers from pointer target type text data bss dec hex filename 222168 17344 27256 266768 41210 ./u-boot multibus_v2: [hs@pollux u-boot-i2c]$ ./MAKEALL MPC8548CDS Configuring for MPC8548CDS board... text data bss dec hex filename 222080 17344 27256 266680 411b8 ./u-boot also the bitbang driver gets better maintainable: Sourcecode changes: multibus: drivers/i2c/soft_i2c.c | 704 ++++++++++++++++++++--------------- multibus_v2: drivers/i2c/soft_i2c.c | 155 +++++--- Codesize bitbang driver: (MPC8548CDS with the following bus topology: * Busses: * * 0: Direct off of FSL_I2C[0] * 1: FSL_I2C[1]->PCA9542(0)->PCA9542(0) * 2: FSL_I2C[1]->PCA9542(0)->PCA9542(1) * 3: FSL_I2C[1]->PCA9542(1) * 4: Direct off of SOFT_I2C[0] * 5: Direct off of SOFT_I2C[1] ) multibus: [hs@pollux u-boot-i2c]$ ./MAKEALL MPC8548CDS Configuring for MPC8548CDS board... fsl_i2c.c: In function '__i2c_init': fsl_i2c.c:173: warning: assignment discards qualifiers from pointer target type text data bss dec hex filename 227092 17864 27256 272212 42754 ./u-boot [hs@pollux u-boot-i2c]$ multibus_v2: [hs@pollux u-boot-i2c]$ ./MAKEALL MPC8548CDS Configuring for MPC8548CDS board... text data bss dec hex filename 225048 17868 27256 270172 41f5c ./u-boot [hs@pollux u-boot-i2c]$ For this I call for all I2C_SDA, I2C_SCL, ... defines, which holds the board specific code, now functions (This is not a must, but I did this as an example see MPC8548CDS). In this functions I use "hw_adapnr" to switch to do the adapter specific gpio settings ... If we look at the "multibus" approach, there, this is not needed but Sourcecode size and Codesize grows with each bitbang adapter ... and if I think for a board with n bitbang adapter (with n > 5) this results in nearly unmaintable source code ... I just write this here because this has to be discussed (and this was a point, which drove the last discussion in chaos). We can now easy compare such a board ... see MPC8548CDS) and discuss on facts, which version we want to have in mainline. - adding a base_addr to i2c_adap_t struct (Did this not yet, but would also a good thing to have, because when I ported for example the ppc_4xx i2c hardware adapter, I saw, that some CPUs have more than one controller, and they just differ in the base addr, so this variable would be a "good to have"). I solved this actually in adding a function in this driver, which returns the base addr depending on hw_adapnr, which is a suboptimal way ... see for an example new ppc_4xx i2c driver: in drivers/i2c/ppc4xx_i2c.c: ppc4xx_get_base () Codesize: for a board with one soft_i2c driver: [hs@pollux u-boot-i2c]$ git checkout master Switched to branch "master" Your branch is ahead of 'origin/master' by 143 commits. [hs@pollux u-boot-i2c]$ ./MAKEALL CPU86 Configuring for CPU86 board... ... booting from 64-bit flash text data bss dec hex filename 146572 22068 24144 192784 2f110 ./u-boot [hs@pollux u-boot-i2c]$ git checkout multibus Switched to branch "multibus" [hs@pollux u-boot-i2c]$ ./MAKEALL CPU86 Configuring for CPU86 board... ... booting from 64-bit flash text data bss dec hex filename 147816 22124 24144 194084 2f624 ./u-boot [hs@pollux u-boot-i2c]$ git checkout multibus_v2 Switched to branch "multibus_v2" [hs@pollux u-boot-i2c]$ ./MAKEALL CPU86 Configuring for CPU86 board... ... booting from 64-bit flash text data bss dec hex filename 147688 22128 24144 193960 2f5a8 ./u-boot [hs@pollux u-boot-i2c]$ for a board with the fsl driver: [hs@pollux u-boot-i2c]$ git checkout master ./MA Switched to branch "master" Your branch is ahead of 'origin/master' by 143 commits. [hs@pollux u-boot-i2c]$ ./MAKEALL MPC8360EMDS Configuring for MPC8360EMDS board... text data bss dec hex filename 203504 11588 26900 241992 3b148 ./u-boot [hs@pollux u-boot-i2c]$ git checkout multibus Switched to branch "multibus" [hs@pollux u-boot-i2c]$ ./MAKEALL MPC8360EMDS Configuring for MPC8360EMDS board... fsl_i2c.c: In function '__i2c_init': fsl_i2c.c:173: warning: assignment discards qualifiers from pointer target type text data bss dec hex filename 204288 11628 26900 242816 3b480 ./u-boot [hs@pollux u-boot-i2c]$ git checkout multibus_v2 Switched to branch "multibus_v2" [hs@pollux u-boot-i2c]$ ./MAKEALL MPC8360EMDS Configuring for MPC8360EMDS board... text data bss dec hex filename 204256 11632 26900 242788 3b464 ./u-boot [hs@pollux u-boot-i2c]$ at last here comes a TODO list what should be done: - add README for the multibus functionality - list of drivers to port: drivbers/i2c: bfin-twi_i2c.c mxc_i2c.c omap1510_i2c.c omap24xx_i2c.c tsi108_i2c.c grep -lr HARD_I2C cpu/ cpu/arm920t/at91rm9200/i2c.c cpu/arm920t/s3c24x0/i2c.c cpu/mpc512x/i2c.c cpu/mpc5xxx/i2c.c cpu/mpc8220/i2c.c cpu/mpc824x/drivers/i2c/i2c.c cpu/mpc8xx/i2c.c cpu/pxa/i2c.c Hope didn;t forget some more ... - test, test, test ... - i2c devices must ported to new multibus functionality. Ideas here are welcome ;-) (They must "know", on which bus they are ...) - In actuall code it is possible to add new i2c busses with the "i2c bus" command (and from code with i2c_mux_ident_muxstring()) This actual is not included in the new code, but that could be done the following way (for the multibus_v2 version, for the approach in "multibus" branch I have no idea how to make this): When running in flash, we just can use the i2c busses defined in CONFIG_SYS_I2C_ADAPTERS (should be okay in the first step. When relocating code, the i2c busses in CONFIG_SYS_I2C_ADAPTERS are converted in a dynamcial list, so we can then easy add new busses to this list. Also we define a "i2c_bus" environment variable, which contains i2c busses, that gets added when relocating. This is needed for the keymile boards. There are a lot of board versions which differ only in the i2c bus topology. With this approach, we can use one u-boot binary for all board versions, just defining the i2c bus topology in the environment ... BTW: This is also a point for using "cur_adap", because the rest of the multibus functionality doesn;t have to be changed ... So, I hope I didn;t forget something ... lets start with the discussion ... bye heiko -- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany

3 8

[U-Boot] [PATCH 00/30] standardize mac address handling in u-boot (v2)
by Mike Frysinger 16 Jul '09

16 Jul '09

this set of patches aims to standardize the mac address handling in all of u-boot by doing the following: - add helper funcs for working with mac addresses (binary & string) - add format strings from Linux for printf-ing mac addresses - convert all duplicated code to use aforementioned helper funcs - convert all places to get the mac address from the environment - drop the global data instances of bi_enet*addr - document the whole damned thing ive taken pains to make sure the patchset doesnt break along the way (in terms of doing bisection and such). but in general, the interdependency looks something like: - add util funcs - convert net/boards/drivers/cpu/commands from global data to env - convert arch/board.c init to not touch global data - remove global data from all arch headers the ppc steps with load_sernum_ethaddr() are a little hairy due to the spread of code and trying to make sure things don't break bisection. but the other option is to squash all these commits (the common ppc board code as well as each individual board) into one change. doesn't matter to me either way. v2 of the patchset brings these changes: - import updated printf modifiers from the kernel - convert str_enetaddr() to %pM - convert print_IPaddr() to %pI4 - convert board_get_enetaddr() to board specific code - misc fixes/changes based on feedback from people while the new printf code is bigger by about 900bytes, i think we make up most of that by converting str_enetaddr/print_IPaddr users over to it. currently this patch series is maintained in my tree: git://www.denx.de/git/u-boot-blackfin.git macaddr Mike Frysinger (30): vsprintf: pull updates from Linux kernel convert print_IPaddr() to %pI4 net: new utility functions for working with enetaddr's doc/README.enetaddr: document proper MAC usage Blackfin: bfin_mac: force boards to setup the MAC themselves net: get mac address from environment and use eth util funcs bdinfo: get mac address from environment bootvx: get mac address from environment lynxkdi: get mac address from environment nvedit: do not update global bi_enetaddr and do not call eth_set_enetaddr() AmigaOneG3SE/enet: get mac address from environment boards: get mac address from environment drivers/net/: get mac address from environment bcm570x: get mac address from environment cs8900: get mac address from environment sh_eth: get mac address from environment lan91c96/smc91111/smc911x: get mac address from environment cpu/: get mac address from environment npe: get mac address from environment lib_*/board.c: do not initialize bi_enet*addr in global data nx823: get mac address from environment arm: get mac address from environment boards: move board_get_enetaddr() into board-specific init cmc_pu2: get mac address from environment pcs440ep: get mac address from environment and move load_sernum_ethaddr() to board init kup4k/kup4x: rename load_sernum_ethaddr() to kup_load_sernum_ethaddr() tqm8xx: rename load_sernum_ethaddr() to tqc_load_sernum_ethaddr() ppc: mark global bi_enet*addr as legacy drop now unused load_sernum_ethaddr() function remove bi_enet*addr from global data for all arches board/MAI/AmigaOneG3SE/enet.c | 42 +--- board/RPXClassic/RPXClassic.c | 16 +- board/cmc_pu2/load_sernum_ethaddr.c | 18 +- board/etin/debris/debris.c | 10 +- board/keymile/mgcoge/mgcoge.c | 4 +- board/keymile/mgsuvd/mgsuvd.c | 4 +- board/kup/common/load_sernum_ethaddr.c | 2 +- board/kup/kup4k/kup4k.c | 3 + board/kup/kup4x/kup4x.c | 4 +- board/m501sk/m501sk.c | 5 - board/mbx8xx/mbx8xx.c | 14 +- board/mpl/vcma9/cmd_vcma9.c | 11 +- board/muas3001/muas3001.c | 4 +- board/netstal/common/nm_bsp.c | 40 +-- board/nx823/flash.c | 5 +- board/nx823/nx823.c | 41 +-- board/pcs440ep/pcs440ep.c | 31 +-- board/pn62/pn62.c | 24 +- board/sandburst/common/sb_common.c | 4 +- board/sandburst/common/sb_common.h | 1 + board/sandburst/karef/karef.c | 29 ++ board/sandburst/metrobox/metrobox.c | 29 ++ board/siemens/IAD210/IAD210.c | 14 +- board/sixnet/sixnet.c | 11 +- board/tqc/tqm8xx/load_sernum_ethaddr.c | 2 +- board/tqc/tqm8xx/tqm8xx.c | 3 + board/v38b/v38b.c | 12 + board/xilinx/xilinx_enet/emac_adapter.c | 8 +- board/xpedite1k/xpedite1k.c | 51 +++- common/cmd_bdinfo.c | 166 ++++-------- common/cmd_elf.c | 6 +- common/cmd_ide.c | 10 +- common/cmd_nvedit.c | 24 +-- common/lynxkdi.c | 2 +- cpu/arm920t/at91rm9200/ether.c | 8 +- cpu/ixp/npe/npe.c | 36 +-- cpu/mpc512x/cpu.c | 6 +- cpu/mpc5xxx/cpu.c | 6 +- cpu/mpc8260/ether_fcc.c | 4 +- cpu/mpc8260/ether_scc.c | 4 +- cpu/ppc4xx/cpu_init.c | 14 +- disk/part.c | 2 +- doc/README.enetaddr | 97 +++++++ drivers/net/3c589.c | 7 +- drivers/net/4xx_enet.c | 13 +- drivers/net/bcm570x.c | 4 +- drivers/net/bcm570x_lm.h | 2 +- drivers/net/bfin_mac.c | 16 +- drivers/net/bfin_mac.h | 2 +- drivers/net/cs8900.c | 56 +--- drivers/net/dc2114x.c | 9 +- drivers/net/dm9000x.c | 26 +-- drivers/net/enc28j60.c | 4 +- drivers/net/fsl_mcdmafec.c | 11 +- drivers/net/ks8695eth.c | 8 +- drivers/net/lan91c96.c | 75 +---- drivers/net/mcffec.c | 10 +- drivers/net/rtl8019.c | 14 +- drivers/net/rtl8169.c | 2 +- drivers/net/s3c4510b_eth.c | 2 +- drivers/net/s3c4510b_eth.h | 2 +- drivers/net/sh_eth.c | 28 +-- drivers/net/smc91111.c | 72 +---- drivers/net/smc911x.c | 14 +- drivers/net/tigon3.c | 7 +- drivers/net/xilinx_emac.c | 12 +- drivers/net/xilinx_emaclite.c | 11 +- include/asm-arm/u-boot.h | 5 - include/asm-avr32/u-boot.h | 1 - include/asm-blackfin/u-boot.h | 1 - include/asm-i386/u-boot.h | 1 - include/asm-m68k/u-boot.h | 13 - include/asm-microblaze/u-boot.h | 1 - include/asm-mips/u-boot.h | 1 - include/asm-nios/u-boot.h | 1 - include/asm-nios2/u-boot.h | 1 - include/asm-ppc/u-boot.h | 14 +- include/asm-sh/u-boot.h | 1 - include/asm-sparc/u-boot.h | 12 - include/common.h | 9 - include/configs/IAD210.h | 1 + include/configs/MBX860T.h | 2 + include/configs/RPXClassic.h | 1 + include/configs/XPEDITE1K.h | 1 + include/configs/cmc_pu2.h | 1 + include/configs/v38b.h | 1 + include/net.h | 6 +- lib_arm/board.c | 47 +--- lib_blackfin/board.c | 48 +--- lib_generic/vsprintf.c | 479 ++++++++++++++++++++++++------- lib_i386/board.c | 17 -- lib_m68k/board.c | 38 --- lib_microblaze/board.c | 8 - lib_mips/board.c | 8 - lib_nios/board.c | 5 - lib_nios2/board.c | 5 - lib_ppc/board.c | 92 +------ lib_sh/board.c | 10 - lib_sparc/board.c | 19 -- net/bootp.c | 29 +-- net/eth.c | 76 +++-- net/net.c | 30 +-- net/nfs.c | 10 +- net/tftp.c | 10 +- post/cpu/mpc8xx/ether.c | 4 +- 105 files changed, 1055 insertions(+), 1188 deletions(-) create mode 100644 doc/README.enetaddr

5 43

[U-Boot] [PATCH 1/1] SMC911x driver fixed for NFS boot
by Manikandan Pillai 10 Jul '09

10 Jul '09

eth_halt() function in the smc911x drivers used to call the smc911x_reset() function. eth_halt() used to be called after tftp transfers. This used to put the ethernet chip in reset while the linux boots up resulting in the ethernet driver not coming up. NFS boot used to fail as a result. This patch calls smc911x_shutdown() instead of smc911x_reset(). Some comments received has also been fixed. Signed-off-by: Manikandan Pillai <mani.pillai(a)ti.com> --- drivers/net/smc911x.c | 23 ++++++++++++++++++++++- 1 files changed, 22 insertions(+), 1 deletions(-) diff --git a/drivers/net/smc911x.c b/drivers/net/smc911x.c index 1ded8f0..5bc3914 100644 --- a/drivers/net/smc911x.c +++ b/drivers/net/smc911x.c @@ -116,6 +116,27 @@ static int smc911x_phy_reset(void) return 0; } +static void smc911x_shutdown(void) +{ + unsigned int cr; + + /* Turn of Rx and TX */ + cr = smc911x_get_mac_csr(MAC_CR); + cr &= ~(MAC_CR_TXEN | MAC_CR_RXEN | MAC_CR_HBDIS); + smc911x_set_mac_csr(MAC_CR, cr); + + /* Stop Transmission */ + cr = smc911x_get_mac_csr(TX_CFG); + cr &= ~(TX_CFG_STOP_TX); + smc911x_set_mac_csr(TX_CFG, cr); + /* Stop receiving packets */ + cr = smc911x_get_mac_csr(RX_CFG); + cr &= ~(RX_CFG_RXDOFF); + smc911x_set_mac_csr(RX_CFG, cr); + +} + + static void smc911x_phy_configure(void) { int timeout; @@ -224,7 +245,7 @@ int eth_send(volatile void *packet, int length) void eth_halt(void) { - smc911x_reset(); + smc911x_shutdown(); } int eth_rx(void) -- 1.6.0.3

5 9

[U-Boot] show_boot_progess @ ppc not working
by Andre Schwarz 09 Jul '09

09 Jul '09

All, can someone tell me why the board specific function "void show_boot_progress(int arg)" is no longer called (at least on MPC5200). Of course this line is present in the board config : #define CONFIG_SHOW_BOOT_PROGRESS 1 What have I missed ? regards, André MATRIX VISION GmbH, Talstraße 16, DE-71570 Oppenweiler - Registergericht: Amtsgericht Stuttgart, HRB 271090 Geschäftsführer: Gerhard Thullner, Werner Armingeon, Uwe Furtner

6 14

[U-Boot] [U-BOOT] cfi_flash.c fails to program NOR Flash SST39LF040
by Po-Yu Chuang 06 Jul '09

06 Jul '09

Hello, I have a board on which there is a NOR Flash SST39LF040. Previously, I copied flash.c from other boards. Now I am trying to migrate NOR Flash driver to use CFI, but it fails to program sectors which are not on the 0x10000 boundaries. I found that in function flash_write_cfiword() @drivers/mtd/cfi_flash.c it issues command with address relative to sector address. However, according to SST39 series data sheet, it seems that we should issue command with address relative to flash base address? After I modified flash_write_cfiword() to issue command to sector 0, everything works fine. Is it a bug? Index: cfi_flash.c =================================================================== RCS file: /usr/local/cvsroot/ctd/FA5A320LINUX26_u-boot/drivers/mtd/cfi_flash.c,v retrieving revision 1.1.1.1 diff -u -r1.1.1.1 cfi_flash.c --- cfi_flash.c 23 Mar 2009 02:54:57 -0000 1.1.1.1 +++ cfi_flash.c 24 Apr 2009 02:28:32 -0000 @@ -839,8 +839,8 @@ case CFI_CMDSET_AMD_LEGACY: #endif sect = find_sector(info, dest); - flash_unlock_seq (info, sect); - flash_write_cmd (info, sect, info->addr_unlock1, AMD_CMD_WRITE); + flash_unlock_seq (info, 0); + flash_write_cmd (info, 0, info->addr_unlock1, AMD_CMD_WRITE); sect_found = 1; break; } regards Po-Yu Chuang

2 6

[U-Boot] "nand biterr" ... got an implementation?
by David Brownell 19 Jun '09

19 Jun '09

I was hoping that I could use that to test some NAND code, but then I noticed it wasn't implemented. I would have expected that U-Boot command wouldn't exist until they're implemented... I'm hoping someone has an implementation. With at least these characteristics: - Doesn't toggle the same bit each time - Repeating N times on a block introduces N distinct 1-bit errors - Normally affects only the data area, not spare/OOB * although forcing bit errors in ECC would be valuable too * ditto the rest of the OOB ... though likely they'd be ignored Am I hoping in vain? ;) - Dave

2 2

[U-Boot] [PATCH 1/2] at91sam9: add watchdog support
by Jean-Christophe PLAGNIOL-VILLARD 18 Jun '09

18 Jun '09

Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj(a)jcrosoft.com> --- Makefile | 1 + doc/README.at91 | 9 ++++ drivers/serial/atmel_usart.c | 4 +- drivers/watchdog/Makefile | 46 ++++++++++++++++++++ drivers/watchdog/at91sam9_wdt.c | 79 ++++++++++++++++++++++++++++++++++ include/asm-arm/arch-at91/at91_wdt.h | 38 ++++++++++++++++ 6 files changed, 176 insertions(+), 1 deletions(-) create mode 100644 drivers/watchdog/Makefile create mode 100644 drivers/watchdog/at91sam9_wdt.c create mode 100644 include/asm-arm/arch-at91/at91_wdt.h diff --git a/Makefile b/Makefile index fcf57f3..3e61417 100644 --- a/Makefile +++ b/Makefile @@ -262,6 +262,7 @@ LIBS += drivers/rtc/librtc.a LIBS += drivers/serial/libserial.a LIBS += drivers/usb/libusb.a LIBS += drivers/video/libvideo.a +LIBS += drivers/watchdog/libwatchdog.a LIBS += common/libcommon.a LIBS += libfdt/libfdt.a LIBS += api/libapi.a diff --git a/doc/README.at91 b/doc/README.at91 index 4e3928a..c883c7c 100644 --- a/doc/README.at91 +++ b/doc/README.at91 @@ -2,6 +2,7 @@ Atmel AT91 Evaluation kits http://atmel.com/dyn/products/tools.asp?family_id=605#1443 +I. Board mapping & boot media ------------------------------------------------------------------------------ AT91SAM9260EK & AT91SAM9XEEK ------------------------------------------------------------------------------ @@ -86,3 +87,11 @@ Environment variables make at91sam9263ek_config - use data flash (spi cs0) (default) make at91sam9263ek_nandflash_config - use nand flash make at91sam9263ek_dataflash_cs0_config - use data flash (spi cs0) + +II. Watchdog support + + The watchdog wan only be activate once so you need to activate in the + bootloader and in the bootstrap if you use it + + The you can activate with + CONFIG_AT91SAM9_WATCHDOG and CONFIG_HW_WATCHDOG diff --git a/drivers/serial/atmel_usart.c b/drivers/serial/atmel_usart.c index f3b146c..f50552a 100644 --- a/drivers/serial/atmel_usart.c +++ b/drivers/serial/atmel_usart.c @@ -16,6 +16,7 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ #include <common.h> +#include <watchdog.h> #include <asm/io.h> #include <asm/arch/clk.h> @@ -87,7 +88,8 @@ void serial_puts(const char *s) int serial_getc(void) { - while (!(usart3_readl(CSR) & USART3_BIT(RXRDY))) ; + while (!(usart3_readl(CSR) & USART3_BIT(RXRDY))) + WATCHDOG_RESET(); return usart3_readl(RHR); } diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile new file mode 100644 index 0000000..200968d --- /dev/null +++ b/drivers/watchdog/Makefile @@ -0,0 +1,46 @@ +# +# (C) Copyright 2008 +# Wolfgang Denk, DENX Software Engineering, wd(a)denx.de. +# +# See file CREDITS for list of people who contributed to this +# project. +# +# 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., 59 Temple Place, Suite 330, Boston, +# MA 02111-1307 USA +# + +include $(TOPDIR)/config.mk + +LIB := $(obj)libwatchdog.a + +COBJS-$(CONFIG_AT91SAM9_WATCHDOG) += at91sam9_wdt.o + +COBJS := $(COBJS-y) +SRCS := $(COBJS:.o=.c) +OBJS := $(addprefix $(obj),$(COBJS)) + +all: $(LIB) + +$(LIB): $(obj).depend $(OBJS) + $(AR) $(ARFLAGS) $@ $(OBJS) + +######################################################################### + +# defines $(obj).depend target +include $(SRCTREE)/rules.mk + +sinclude $(obj).depend + +######################################################################### diff --git a/drivers/watchdog/at91sam9_wdt.c b/drivers/watchdog/at91sam9_wdt.c new file mode 100644 index 0000000..5bb8b77 --- /dev/null +++ b/drivers/watchdog/at91sam9_wdt.c @@ -0,0 +1,79 @@ +/* + * [origin: Linux kernel drivers/watchdog/at91sam9_wdt.c] + * + * Watchdog driver for Atmel AT91SAM9x processors. + * + * Copyright (C) 2008 Jean-Christophe PLAGNIOL-VILLARD <plagnioj(a)jcrosoft.com> + * Copyright (C) 2008 Renaud CERRATO r.cerrato(a)til-technologies.fr + * + * 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. + */ + +/* + * The Watchdog Timer Mode Register can be only written to once. If the + * timeout need to be set from U-Boot, be sure that the bootstrap doesn't + * write to this register. Inform Linux to it too + */ + +#include <common.h> +#include <watchdog.h> +#include <asm/arch/hardware.h> +#include <asm/arch/io.h> +#include <asm/arch/at91_wdt.h> + +/* + * AT91SAM9 watchdog runs a 12bit counter @ 256Hz, + * use this to convert a watchdog + * value from/to milliseconds. + */ +#define ms_to_ticks(t) (((t << 8) / 1000) - 1) +#define ticks_to_ms(t) (((t + 1) * 1000) >> 8) + +/* Hardware timeout in seconds */ +#define WDT_HW_TIMEOUT 2 + +/* + * Set the watchdog time interval in 1/256Hz (write-once) + * Counter is 12 bit. + */ +static int at91_wdt_settimeout(unsigned int timeout) +{ + unsigned int reg; + unsigned int mr; + + /* Check if disabled */ + mr = at91_sys_read(AT91_WDT_MR); + if (mr & AT91_WDT_WDDIS) { + printf("sorry, watchdog is disabled\n"); + return -1; + } + + /* + * All counting occurs at SLOW_CLOCK / 128 = 256 Hz + * + * Since WDV is a 12-bit counter, the maximum period is + * 4096 / 256 = 16 seconds. + */ + reg = AT91_WDT_WDRSTEN /* causes watchdog reset */ + /* | AT91_WDT_WDRPROC causes processor reset only */ + | AT91_WDT_WDDBGHLT /* disabled in debug mode */ + | AT91_WDT_WDD /* restart at any time */ + | (timeout & AT91_WDT_WDV); /* timer value */ + at91_sys_write(AT91_WDT_MR, reg); + + return 0; +} + +void hw_watchdog_reset(void) +{ + at91_sys_write(AT91_WDT_CR, AT91_WDT_KEY | AT91_WDT_WDRSTT); +} + +void hw_watchdog_init(void) +{ + /* 16 seconds timer, resets enabled */ + at91_wdt_settimeout(ms_to_ticks(WDT_HW_TIMEOUT * 1000)); +} diff --git a/include/asm-arm/arch-at91/at91_wdt.h b/include/asm-arm/arch-at91/at91_wdt.h new file mode 100644 index 0000000..7e18537 --- /dev/null +++ b/include/asm-arm/arch-at91/at91_wdt.h @@ -0,0 +1,38 @@ +/* + * [origin: Linux kernel arch/arm/mach-at91/include/mach/at91_wdt.h] + * + * Copyright (C) 2008 Jean-Christophe PLAGNIOL-VILLARD <plagnioj(a)jcrosoft.com> + * Copyright (C) 2007 Andrew Victor + * Copyright (C) 2007 Atmel Corporation. + * + * Watchdog Timer (WDT) - System peripherals regsters. + * Based on AT91SAM9261 datasheet revision D. + * + * 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. + */ + +#ifndef AT91_WDT_H +#define AT91_WDT_H + +#define AT91_WDT_CR (AT91_WDT + 0x00) /* Watchdog Control Register */ +#define AT91_WDT_WDRSTT (1 << 0) /* Restart */ +#define AT91_WDT_KEY (0xa5 << 24) /* KEY Password */ + +#define AT91_WDT_MR (AT91_WDT + 0x04) /* Watchdog Mode Register */ +#define AT91_WDT_WDV (0xfff << 0) /* Counter Value */ +#define AT91_WDT_WDFIEN (1 << 12) /* Fault Interrupt Enable */ +#define AT91_WDT_WDRSTEN (1 << 13) /* Reset Processor */ +#define AT91_WDT_WDRPROC (1 << 14) /* Timer Restart */ +#define AT91_WDT_WDDIS (1 << 15) /* Watchdog Disable */ +#define AT91_WDT_WDD (0xfff << 16) /* Delta Value */ +#define AT91_WDT_WDDBGHLT (1 << 28) /* Debug Halt */ +#define AT91_WDT_WDIDLEHLT (1 << 29) /* Idle Halt */ + +#define AT91_WDT_SR (AT91_WDT + 0x08) /* Watchdog Status Register */ +#define AT91_WDT_WDUNF (1 << 0) /* Watchdog Underflow */ +#define AT91_WDT_WDERR (1 << 1) /* Watchdog Error */ + +#endif -- 1.5.6.5

4 9

[U-Boot] Out-of-tree build of API code fails
by Wolfgang Denk 15 Jun '09

15 Jun '09

Dear Rafal, it seems the api_examples/ code fails when building out of tree: -> make O=/work/wd/tmp-ppc PMC440_config Configuring for PMC440 board... -> make O=/work/wd/tmp-ppc all Generating /work/wd/tmp-ppc/include/autoconf.mk.dep Generating /work/wd/tmp-ppc/include/autoconf.mk for dir in tools examples api_examples ; do make -C $dir _depend ; done make[1]: Entering directory `/home/wd/git/u-boot/work/tools' ln -s /home/wd/git/u-boot/work/tools/../common/env_embedded.c /work/wd/tmp-ppc/tools/env_embedded.c ln -s /home/wd/git/u-boot/work/tools/../include/zlib.h /work/wd/tmp-ppc/tools/zlib.h ln -s /home/wd/git/u-boot/work/tools/../lib_generic/crc32.c /work/wd/tmp-ppc/tools/crc32.c ln -s /home/wd/git/u-boot/work/tools/../lib_generic/md5.c /work/wd/tmp-ppc/tools/md5.c ln -s /home/wd/git/u-boot/work/tools/../lib_generic/sha1.c /work/wd/tmp-ppc/tools/sha1.c ln -s /home/wd/git/u-boot/work/tools/../common/image.c /work/wd/tmp-ppc/tools/image.c if [ ! -f /work/wd/tmp-ppc/tools/mkimage.h ] ; then \ ln -s /home/wd/git/u-boot/work/tools/../tools/mkimage.h /work/wd/tmp-ppc/tools/mkimage.h; \ fi if [ ! -f /work/wd/tmp-ppc/tools/fdt_host.h ] ; then \ ln -s /home/wd/git/u-boot/work/tools/../tools/fdt_host.h /work/wd/tmp-ppc/tools/fdt_host.h; \ fi make[1]: Leaving directory `/home/wd/git/u-boot/work/tools' make[1]: Entering directory `/home/wd/git/u-boot/work/tools' make[1]: Nothing to be done for `_depend'. make[1]: Leaving directory `/home/wd/git/u-boot/work/tools' make[1]: Entering directory `/home/wd/git/u-boot/work/examples' make[1]: Leaving directory `/home/wd/git/u-boot/work/examples' make[1]: Entering directory `/home/wd/git/u-boot/work/examples' make[1]: Nothing to be done for `_depend'. make[1]: Leaving directory `/home/wd/git/u-boot/work/examples' make[1]: Entering directory `/home/wd/git/u-boot/work/api_examples' make[1]: *** No rule to make target `/work/wd/tmp-ppc/api_examples/.depend', needed by `_depend'. Stop. make[1]: Leaving directory `/home/wd/git/u-boot/work/api_examples' make: *** [depend] Error 2 Could you please have a look? Thanks in advance. 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 A Puritan is someone who is deathly afraid that someone, somewhere, is having fun.

2 5

[U-Boot] [PATCH] s3c64xx: support the soft reset at s3c6410
by Minkyu Kang 09 Jun '09

09 Jun '09

This patch support the soft reset at s3c6410 Signed-off-by: Minkyu Kang <mk7.kang(a)samsung.com> --- cpu/arm1176/cpu.c | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-) diff --git a/cpu/arm1176/cpu.c b/cpu/arm1176/cpu.c index 1e94f7d..cb13d9c 100644 --- a/cpu/arm1176/cpu.c +++ b/cpu/arm1176/cpu.c @@ -112,7 +112,11 @@ int cleanup_before_linux (void) void reset_cpu (ulong ignored) { printf("reset... \n\n\n"); +#if defined(CONFIG_S3C6410) + SW_RST_REG = 0x6410; +#else SW_RST_REG = 0x6400; +#endif /* loop forever and wait for reset to happen */ while (1) { if (serial_tstc()) {

2 2

[U-Boot] [PATCH 1/3] smc911x: write back the manually set MAC address
by Daniel Mack 05 Jun '09

05 Jun '09

If the MAX address is given by the environment, write it back to the hardware. Signed-off-by: Daniel Mack <daniel(a)caiaq.de> Cc: Sascha Hauer <s.hauer(a)pengutronix.de> --- drivers/net/smc911x.c | 9 +++++++-- 1 files changed, 7 insertions(+), 2 deletions(-) diff --git a/drivers/net/smc911x.c b/drivers/net/smc911x.c index 30f2dc2..8c9a2a8 100644 --- a/drivers/net/smc911x.c +++ b/drivers/net/smc911x.c @@ -41,8 +41,13 @@ static int smx911x_handle_mac_address(bd_t *bd) unsigned long addrh, addrl; uchar m[6]; - /* if the environment has a valid mac address then use it */ - if (!eth_getenv_enetaddr("ethaddr", m)) { + if (eth_getenv_enetaddr("ethaddr", m)) { + /* if the environment has a valid mac address then use it */ + addrl = m[0] | (m[1] << 8) | (m[2] << 16) | (m[3] << 24); + addrh = m[4] | (m[5] << 8); + smc911x_set_mac_csr(ADDRL, addrl); + smc911x_set_mac_csr(ADDRH, addrh); + } else { /* if not, try to get one from the eeprom */ addrh = smc911x_get_mac_csr(ADDRH); addrl = smc911x_get_mac_csr(ADDRL); -- 1.6.2.1

7 20