August 2006 - coreboot - mail.coreboot.org

[LinuxBIOS] r2372 - trunk/LinuxBIOSv2/targets/olpc/rev_a
by svn＠openbios.org Aug. 10, 2006

Aug. 10, 2006

Author: rminnich Date: 2006-08-11 02:08:37 +0200 (Fri, 11 Aug 2006) New Revision: 2372 Modified: trunk/LinuxBIOSv2/targets/olpc/rev_a/Config.1M.lb Log: build 1024-128k binary as per requests. Modified: trunk/LinuxBIOSv2/targets/olpc/rev_a/Config.1M.lb =================================================================== --- trunk/LinuxBIOSv2/targets/olpc/rev_a/Config.1M.lb 2006-08-10 09:38:39 UTC (rev 2371) +++ trunk/LinuxBIOSv2/targets/olpc/rev_a/Config.1M.lb 2006-08-11 00:08:37 UTC (rev 2372) @@ -8,7 +8,7 @@ #option CONFIG_PRECOMPRESSED_ROM_STREAM=1 # leave 64k for vsa -option ROM_SIZE=1024*1024-64*1024 +option ROM_SIZE=1024*1024-64*1024-64*1024 option FALLBACK_SIZE=ROM_SIZE option DEFAULT_CONSOLE_LOGLEVEL = 11

1 0

[LinuxBIOS] boot windows, BSD and solaris on x86
by Lu, Yinghai Aug. 10, 2006

Aug. 10, 2006

Eric, Anyone try to use kexec to boot windows, BSD and solaris on x86? YH

1 0

[LinuxBIOS] AMD Turion - AGESA
by jf simon Aug. 10, 2006

Aug. 10, 2006

Hi, The company i worked for has designed a card that uses the AMD Turion. It uses the PHOENIX Bios. I am trying to port linuxbios to it. But among other issues, one hardware engineer told me to make sure that linuxbios was using the latest AGESA code from AMD (from what i understand that code changes the CPU processor state, frequency,...)as it seems we had trouble in the past getting the right BIOS with the latest AGESA code. How does linuxbios manage to do the same thing as the agesa code (and at what agesa version)? If so what is the file location that handles it? Thanks a lot, -jf simon ___________________________________________________________________________ D�couvrez un nouveau moyen de poser toutes vos questions quelque soit le sujet ! Yahoo! Questions/R�ponses pour partager vos connaissances, vos opinions et vos exp�riences. http://fr.answers.yahoo.com

2 5

[LinuxBIOS] LinuxBIOS r2371 build changes
by LinuxBIOS information Aug. 10, 2006

Aug. 10, 2006

Dear LinuxBIOS readers! This is the automated build check service of LinuxBIOS. The developer "stepan" checked in revision 2371 to the LinuxBIOS source repository and caused the following changes: Change Log: fix serial initialization (from Uwe Hermann) Build Log: Compilation of artecgroup:dbe61 is still broken. If something broke during this checkin please be a pain in stepan's neck until the issue is fixed. If this issue is not fixed within 24h the revision will be backed out. Yours truely, LinuxBIOS automatic build system

1 0

[LinuxBIOS] r2371 - trunk/LinuxBIOSv2/src/superio/ite/it8671f
by svn＠openbios.org Aug. 10, 2006

Aug. 10, 2006

Author: stepan Date: 2006-08-10 11:38:39 +0200 (Thu, 10 Aug 2006) New Revision: 2371 Modified: trunk/LinuxBIOSv2/src/superio/ite/it8671f/chip.h trunk/LinuxBIOSv2/src/superio/ite/it8671f/it8671f.h trunk/LinuxBIOSv2/src/superio/ite/it8671f/it8671f_early_serial.c trunk/LinuxBIOSv2/src/superio/ite/it8671f/superio.c Log: fix serial initialization (from Uwe Hermann) Modified: trunk/LinuxBIOSv2/src/superio/ite/it8671f/chip.h =================================================================== --- trunk/LinuxBIOSv2/src/superio/ite/it8671f/chip.h 2006-08-10 03:23:48 UTC (rev 2370) +++ trunk/LinuxBIOSv2/src/superio/ite/it8671f/chip.h 2006-08-10 09:38:39 UTC (rev 2371) @@ -19,14 +19,6 @@ #ifndef _SUPERIO_ITE_IT8671F #define _SUPERIO_ITE_IT8671F -// TODO: Unused? -// #ifndef SIO_COM1 -// #define SIO_COM1_BASE 0x3F8 -// #endif -// #ifndef SIO_COM2 -// #define SIO_COM2_BASE 0x2F8 -// #endif - #include <pc80/keyboard.h> #include <uart8250.h> Modified: trunk/LinuxBIOSv2/src/superio/ite/it8671f/it8671f.h =================================================================== --- trunk/LinuxBIOSv2/src/superio/ite/it8671f/it8671f.h 2006-08-10 03:23:48 UTC (rev 2370) +++ trunk/LinuxBIOSv2/src/superio/ite/it8671f/it8671f.h 2006-08-10 09:38:39 UTC (rev 2371) @@ -16,20 +16,11 @@ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ -// TODO: Untested, cut'n'pasted from some other file so it's probably wrong. -// #define IT8671F_FDC 0x00 /* Floppy */ -// #define IT8671F_PP 0x01 /* Parallel port */ +/* TODO: Find datasheet and check for correct values. */ +#define IT8671F_FDC 0x00 /* Floppy */ +/* #define IT8671F_PP 0x01 */ /* Parallel port */ #define IT8671F_SP2 0x02 /* Com2 */ #define IT8671F_SP1 0x03 /* Com1 */ -// #define IT8671F_SWC 0x04 -// #define IT8671F_KBCM 0x05 /* Mouse */ #define IT8671F_KBCK 0x06 /* Keyboard */ -// #define IT8671F_GPIO 0x07 -// #define IT8671F_ACB 0x08 -// #define IT8671F_FSCM 0x09 -// #define IT8671F_WDT 0x0A -// #define IT8671F_GMP 0x0B -// #define IT8671F_MIDI 0x0C -// #define IT8671F_VLM 0x0D -// #define IT8671F_TMS 0x0E +/* #define IT8671F_FAN 0x09 */ Modified: trunk/LinuxBIOSv2/src/superio/ite/it8671f/it8671f_early_serial.c =================================================================== --- trunk/LinuxBIOSv2/src/superio/ite/it8671f/it8671f_early_serial.c 2006-08-10 03:23:48 UTC (rev 2370) +++ trunk/LinuxBIOSv2/src/superio/ite/it8671f/it8671f_early_serial.c 2006-08-10 09:38:39 UTC (rev 2371) @@ -19,11 +19,73 @@ #include <arch/romcc_io.h> #include "it8671f.h" +/* The base address is 0x3f0, 0x3bd, or 0x370, depending on config bytes. */ +#define SIO_BASE 0x3f0 +#define SIO_INDEX SIO_BASE +#define SIO_DATA SIO_BASE+1 + +/* TODO: These values are actually from the IT8673F datasheet; check if + they're also valid for the IT8671F. */ +#define IT8671F_CONFIG_REG_CC 0x02 /* Configure Control. */ +#define IT8671F_CONFIG_REG_LDN 0x07 /* Logical Device Number. */ +#define IT8671F_CONFIG_REG_CLOCKSEL 0x23 /* Clock Selection. */ +#define IT8671F_CONFIG_REG_SWSUSP 0x24 /* Software Suspend. */ + +#define IT8671F_ADDRESS_PORT 0x279 + +/* Special values used for entering MB PnP mode. The first four bytes of + * each line determine the address port, the last four are data. */ +static const uint8_t init_values[] = { + 0x6a, 0xb5, 0xda, 0xed, /**/ 0xf6, 0xfb, 0x7d, 0xbe, + 0xdf, 0x6f, 0x37, 0x1b, /**/ 0x0d, 0x86, 0xc3, 0x61, + 0xb0, 0x58, 0x2c, 0x16, /**/ 0x8b, 0x45, 0xa2, 0xd1, + 0xe8, 0x74, 0x3a, 0x9d, /**/ 0xce, 0xe7, 0x73, 0x39, +}; + +/* The content of IT8671F_CONFIG_REG_LDN (index 07h) must be set to the + * LDN the register belongs to, before you can access the register. */ +static void it8671f_sio_write(uint8_t ldn, uint8_t index, uint8_t value) +{ + outb(IT8671F_CONFIG_REG_LDN, SIO_BASE); + outb(ldn, SIO_DATA); + outb(index, SIO_BASE); + outb(value, SIO_DATA); +} + +/* Enable the peripheral devices on the IT8671F Super IO chip. */ static void it8671f_enable_serial(device_t dev, unsigned iobase) { - pnp_set_logical_device(dev); - pnp_set_enable(dev, 1); - pnp_set_iobase(dev, PNP_IDX_IO0, iobase); - pnp_set_enable(dev, 1); + uint8_t i; + + /* (1) Enter the configuration state (MB PnP mode). */ + + /* Perform MB PnP setup to put the SIO chip at 0x3f0. */ + /* Base address 0x3f0: 0x86 0x80 0x55 0x55. */ + /* Base address 0x3bd: 0x86 0x80 0x55 0xaa. */ + /* Base address 0x370: 0x86 0x80 0xaa 0x55. */ + outb(0x86, IT8671F_ADDRESS_PORT); + outb(0x80, IT8671F_ADDRESS_PORT); + outb(0x55, IT8671F_ADDRESS_PORT); + outb(0x55, IT8671F_ADDRESS_PORT); + + /* Sequentially write the 32 special values. */ + for (i = 0; i < 32; i++) { + outb(init_values[i], SIO_BASE); + } + + /* (2) Modify the data of configuration registers. */ + + /* Enable parallel port, serial port 1, serial port 2, floppy. */ + it8671f_sio_write(0x00, 0x23, 0x0f); + + /* Activate serial port 1 and 2. */ + it8671f_sio_write(0x01, 0x30, 0x1); + it8671f_sio_write(0x02, 0x30, 0x1); + + /* Select 24MHz CLKIN and clear software suspend mode. */ + it8671f_sio_write(0x00, IT8671F_CONFIG_REG_SWSUSP, 0x00); + + /* (3) Exit the configuration state (MB PnP mode). */ + it8671f_sio_write(0x00, IT8671F_CONFIG_REG_CC, 0x02); } Modified: trunk/LinuxBIOSv2/src/superio/ite/it8671f/superio.c =================================================================== --- trunk/LinuxBIOSv2/src/superio/ite/it8671f/superio.c 2006-08-10 03:23:48 UTC (rev 2370) +++ trunk/LinuxBIOSv2/src/superio/ite/it8671f/superio.c 2006-08-10 09:38:39 UTC (rev 2371) @@ -16,12 +16,6 @@ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ -// #include <arch/io.h> -// #include <device/device.h> -// #include <device/pnp.h> -// #include <console/console.h> -// #include <string.h> -// #include <bitops.h> #include <uart8250.h> #include <pc80/keyboard.h> #include "chip.h" @@ -38,7 +32,10 @@ conf = dev->chip_info; - switch(dev->path.u.pnp.device) { + switch (dev->path.u.pnp.device) { + case IT8671F_FDC: + /* TODO. */ + break; case IT8671F_SP1: res0 = find_resource(dev, PNP_IDX_IO0); init_uart8250(res0->base, &conf->com1); @@ -63,18 +60,13 @@ .init = init, }; -// TODO. +/* TODO: Find and check datasheet. */ static struct pnp_info pnp_dev_info[] = { -// { &ops, IT8671F_FDC, PNP_IO0 | PNP_IRQ0 | PNP_DRQ0, { 0x07fa, 0}, }, -// { &ops, IT8671F_PP, PNP_IO0 | PNP_IRQ0 | PNP_DRQ0, { 0x04f8, 0}, }, + { &ops, IT8671F_FDC, PNP_IO0 | PNP_IRQ0 | PNP_DRQ0, { 0x07fa, 0}, }, + /* { &ops, IT8671F_PP, PNP_IO0 | PNP_IRQ0 | PNP_DRQ0, { 0x04f8, 0}, }, */ { &ops, IT8671F_SP2, PNP_IO0 | PNP_IRQ0 | PNP_DRQ0 | PNP_DRQ1, { 0x7f8, 0 }, }, { &ops, IT8671F_SP1, PNP_IO0 | PNP_IRQ0, { 0x7f8, 0 }, }, -// { &ops, IT8671F_SWC, PNP_IO0 | PNP_IRQ0, { 0xfff0, 0 }, }, -// { &ops, IT8671F_KBCM, PNP_IRQ0 }, -{ &ops, IT8671F_KBCK, PNP_IO0 | PNP_IO1 | PNP_IRQ0, { 0x7f8, 0 }, { 0x7f8, 0x4}, }, -// { &ops, IT8671F_GPIO, PNP_IO0 | PNP_IRQ0, { 0xfff8, 0 } }, -// { &ops, IT8671F_XBUS, PNP_IO0 | PNP_IRQ0, { 0xffe0, 0 } }, -// { &ops, IT8671F_RTC, PNP_IO0 | PNP_IO1, { 0xfffe, 0 }, {0xfffe, 0x4} }, + { &ops, IT8671F_KBCK, PNP_IO0 | PNP_IO1 | PNP_IRQ0, { 0x7f8, 0 }, { 0x7f8, 0x4}, }, }; static void enable_dev(struct device *dev)

1 0

[LinuxBIOS] LinuxBIOS r2370 build changes
by LinuxBIOS information Aug. 9, 2006

Aug. 9, 2006

Dear LinuxBIOS readers! This is the automated build check service of LinuxBIOS. The developer "rminnich" checked in revision 2370 to the LinuxBIOS source repository and caused the following changes: Change Log: change from AMD for the IRQ10 problem. Build Log: Compilation of artecgroup:dbe61 is still broken. If something broke during this checkin please be a pain in rminnich's neck until the issue is fixed. If this issue is not fixed within 24h the revision will be backed out. Yours truely, LinuxBIOS automatic build system

1 0

[LinuxBIOS] r2370 - in trunk/LinuxBIOSv2/src: mainboard/olpc/rev_a northbridge/amd/gx2
by svn＠openbios.org Aug. 9, 2006

Aug. 9, 2006

Author: rminnich Date: 2006-08-10 05:23:48 +0200 (Thu, 10 Aug 2006) New Revision: 2370 Modified: trunk/LinuxBIOSv2/src/mainboard/olpc/rev_a/auto.c trunk/LinuxBIOSv2/src/northbridge/amd/gx2/northbridge.c Log: change from AMD for the IRQ10 problem. Modified: trunk/LinuxBIOSv2/src/mainboard/olpc/rev_a/auto.c =================================================================== --- trunk/LinuxBIOSv2/src/mainboard/olpc/rev_a/auto.c 2006-08-09 02:21:49 UTC (rev 2369) +++ trunk/LinuxBIOSv2/src/mainboard/olpc/rev_a/auto.c 2006-08-10 03:23:48 UTC (rev 2370) @@ -149,7 +149,18 @@ __builtin_wrmsr(0x40000021, 0x80fffe0, 0x20000000); } - +static void gpio_init(void) +{ + unsigned long m; + + /* Make sure events enable for gpio 12 is off */ + + m = inl(GPIOL_EVENTS_ENABLE); + m &= ~GPIOL_12_SET; + m |= GPIOL_12_CLEAR; + outl(m, GPIOL_EVENTS_ENABLE); +} + static void main(unsigned long bist) { static const struct mem_controller memctrl [] = { @@ -166,6 +177,7 @@ * for cs5536 */ cs5536_setup_onchipuart(); + gpio_init(); uart_init(); console_init(); Modified: trunk/LinuxBIOSv2/src/northbridge/amd/gx2/northbridge.c =================================================================== --- trunk/LinuxBIOSv2/src/northbridge/amd/gx2/northbridge.c 2006-08-09 02:21:49 UTC (rev 2369) +++ trunk/LinuxBIOSv2/src/northbridge/amd/gx2/northbridge.c 2006-08-10 03:23:48 UTC (rev 2370) @@ -276,11 +276,20 @@ static void northbridge_init(device_t dev) { + unsigned long m; + struct northbridge_amd_gx2_config *nb = (struct northbridge_amd_gx2_config *)dev->chip_info; printk_debug("northbridge: %s()\n", __FUNCTION__); enable_shadow(dev); irq_init_steering(dev, nb->irqmap); + + /* HACK HACK HACK HACK */ + /* 0x1000 is where GPIO is being assigned */ + m = inl(0x1038); + m &= ~GPIOL_12_SET; + m |= GPIOL_12_CLEAR; + outl(m, 0x1038); } /* due to vsa interactions, we need not not touch the nb settings ... */

1 0

[LinuxBIOS] RD1-PL on EPIA-M10000 problem
by lists＠actweb.info Aug. 9, 2006

Aug. 9, 2006

hi can someone confirm that i have the following correct? i have an EPIA-M m/board with an SST39SF020A bios chip, along with an RD1-PL 2Mbit bios savour, are these both 2Mb chips? i have tried writing the RD1 with a image file, but it always fails :( anyone any pointers? have so far tried to write image aprox 8 times, still with no luck :( this is the output from flashrom :- Original BIOS # ./flashrom Calibrating delay loop... ok No LinuxBIOS table found. Enabling flash write on VT8235...tried to set 0x45 to 0x55 on VT8235 failed (WARNING ONLY) SST39SF020A found at physical address: 0xfffc0000 Flash part is SST39SF020A OK, only ENABLING flash write, but NOT FLASHING. RD1-PL BIOS # ./flashrom Calibrating delay loop... ok No LinuxBIOS table found. Enabling flash write on VT8235...OK W49F002U found at physical address: 0xfffc0000 Flash part is W49F002U OK, only ENABLING flash write, but NOT FLASHING. as far as i can tell both are 2M bios, is htere something that prevent writing to the differant chip number properly? PS. this bios savour worked fine on my EPIA-PD board, always writing corectly first time :( Many thanks in advance Matt

4 4

[LinuxBIOS] Linux & memory regions
by Indrek Kruusa Aug. 9, 2006

Aug. 9, 2006

Hi! I am looking for clarification about what and how (Linux)BIOS should provide to kernel about memory size/memory regions. I'm wondering about linuxbios table: there are lot of different variation with new_resource/ram_resource, where those calls are placed, what regions they are setting up, are those regions 0-TopOfSysRAM, or 0-640/768-TopOfSysRAM etc. And reportedly it all works with linux kernel. I'm not sure how to properly control linuxbios table content - and is this table mandatory for linux kernel? thanks, Indrek

1 0

[LinuxBIOS] LinuxBIOS r2369 build changes
by LinuxBIOS information Aug. 8, 2006

Aug. 8, 2006

Dear LinuxBIOS readers! This is the automated build check service of LinuxBIOS. The developer "rminnich" checked in revision 2369 to the LinuxBIOS source repository and caused the following changes: Change Log: mods for the ultra40 bringup. This now builds.amd gx2 north -- don't set anything in the north, it conflicts with vsasettings. So we have our own pci_set_resources that is essentially ano-op -- just calls the kids.olpc rev_a config -- DISABLE the compressed rom stream. This SHOULD NOThave been set -- it is untested and caused real trouble. Build Log: Compilation of artecgroup:dbe61 is still broken. Compilation of sunw:ultra40 has been fixed. If something broke during this checkin please be a pain in rminnich's neck until the issue is fixed. If this issue is not fixed within 24h the revision will be backed out. Yours truely, LinuxBIOS automatic build system

1 0