coreboot May 2007

coreboot@coreboot.org

98 participants
436 discussions

[LinuxBIOS] patch: add support for msm800sev to buildrom
by ron minnich 10 May '07

10 May '07

This patch adds support for the msm800sev to buildrom. It will need further work, but I am trying to get something going. Signed-off-by: Ronald G. Minnich <rminnich(a)gmail.com>

5 7

[LinuxBIOS] asus/p2b/auto.c bug with smbus_read_byte? (440bx)
by roger 10 May '07

10 May '07

Using the default asus/p2b/autoc spd_read_byte customized function, DRAM TEST hangs(right after DRAM Verify). Comparing to tyan/s1846/auto.c, I've simply replaced with "return smbus_read_byte(device, address);" and this gets me further. Following is a sniper of auto.c: --- Begin Snip auto.c --- /* TODO: fix raminit.c to use smbus_read_byte */ static inline int spd_read_byte(unsigned int device, unsigned int address) { /* uint8_t c; c = smbus_read_byte(device, address); return c; */ return smbus_read_byte(device, address); } --- End Snip auto.c --- Now I can get as far as: DRAM range verified. Done. Copying LinuxBIOS to RAM. Jumping to LinuxBIOS. LinuxBIOS-2.0.0.0Fallback Thu May 10 00:25:28 PDT 2007 booting... -- Roger http://www.eskimo.com/~roger/index.html Key fingerprint = 8977 A252 2623 F567 70CD 1261 640F C963 1005 1D61 Thu May 10 00:27:31 PDT 2007

3 2

[LinuxBIOS] How to parameterise SDRAM init on Geode GX1?
by Juergen Beisert 10 May '07

10 May '07

Hi, the routine sdram_init() in LinuxBIOSv2/src/northbridge/amd/gx1/raminit.c does a very slow SDRAM timing setup. On a 266MHz CPU the SDRAM will run at 76MHz with a CAS latency of 3. You can't run a SXGA reolution with 64k colours on such a slow system. On my mainboard I can't read the EEPROM on the SODIMM, as I have no schematics and do not know if the manufacturer has routed some of the GPIOs as a I2C bus to this EEPROM. But I know the SDRAM chips on the SODIMM in my system! Is there a "regular" way in LinuxBIOS to parameterise the SDRAM setup with some predefined macros in a (maybe) headerfile, so each GX1 based mainboard could overwrite the slowest timing with its own faster timing (if the programmer knows his SDRAM timings). In my case I can run some of my SODIMMs at 120MHz@CL2, some other with 100MHz@CL2 (300MHz Geode GX1). Maybe the trick with a searchpath for headerfiles. There will be a generic file with the slowest timing in a directory at the end of the searchpath and the prgrammer can provide his own timing headerfile that would be found first if present? Juergen

1 0

[LinuxBIOS] [PATCH] Fix Asus p2b and bitworks ims targets
by Corey Osgood 10 May '07

10 May '07

This patch uses auto.c from Uwe's tyan s1846 for both boards, with some minor changes, to bring them up to par. It also should remove (but might just clean out) the irq_tables.c from both boards, because they were just copied from Via Epia to begin with, and weren't usable. As far as I can tell, these are the only changes needed to the targets for now, aside from fixups to reset.c when the time comes. Both have been build tested, but not checked on hardware since I don't have it. I have left Uwe as the copyright holder since the only changes I've made are trivial. Signed-off-by: Corey Osgood <corey_osgood(a)verizon.net>

5 9

Re: [LinuxBIOS] Is that mainboard supported? - DTK PRM-0080I E1
by Simon Comeau Martel 10 May '07

10 May '07

roger wrote: > If all goes well & it's not just a syntax error on my part, patch > committing takes a couple of days. > > console_init is only available and there are still a few bugs for this > chipset, but patches are regularly occurring on the mailing list. > > FYI: You can find specific flash parts at digikey.com > > Atmel At29C020 (256 KB) is what i use here. > Thanks for the info; I received my BIOS savior today, so I can now start playing with Linux Bios :) -- Simon Comeau Martel simon(a)comeau.info

1 0

[LinuxBIOS] r2645 build service
by LinuxBIOS information 09 May '07

09 May '07

Dear LinuxBIOS readers! This is the automated build check service of LinuxBIOS. The developer "uwe" checked in revision 2645 to the LinuxBIOS source repository and caused the following changes: Change Log: The superio needs 0x87 sent twice to 0x3f0 to enable extended function (power-on strapping). Although this is already done in superio.c, it's not being done when w83977tf_early_serial.c is executed. As such, no console_init() without it. Signed-off-by: Roger Zauner <roger(a)eskimo.com> Acked-by: Uwe Hermann <uwe(a)hermann-uwe.de> Build Log: Compilation of arima:hdama is still broken Compilation of artecgroup:dbe61 is still broken Compilation of ibm:e325 is still broken Compilation of ibm:e326 is still broken Compilation of iwill:dk8s2 is still broken Compilation of iwill:dk8x is still broken If something broke during this checkin please be a pain in uwe'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] r2641 build service
by LinuxBIOS information 09 May '07

09 May '07

Dear LinuxBIOS readers! This is the automated build check service of LinuxBIOS. The developer "uwe" checked in revision 2641 to the LinuxBIOS source repository and caused the following changes: Change Log: Add initial support for the ASI/BCom MB-5BLMP mainboard, as used in the IGEL Winnet III thin client. It boots a Linux kernel, but there are some problems. The login prompt is never reached, it simply hangs at some point. One possible reason is the IRQ table, which needs fixing. Signed-off-by: Uwe Hermann <uwe(a)hermann-uwe.de> Acked-by: Stefan Reinauer <stepan(a)coresystems.de> Build Log: Compilation of arima:hdama is still broken Compilation of artecgroup:dbe61 is still broken Configuration of asi:mb_5blmp is still broken Compilation of ibm:e325 is still broken Compilation of ibm:e326 is still broken Compilation of iwill:dk8s2 is still broken Compilation of iwill:dk8x is still broken If something broke during this checkin please be a pain in uwe'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

3 2

[LinuxBIOS] Patch for w83977tf to enable early_serial
by roger 09 May '07

09 May '07

The superio needs 0x87 sent twice to 0x3f0 to enable extended function (power-on strapping). Although this is already done in superio.c, it's not being done when w83977tf_early_serial.c is executed. As such, no console_init -- atleast in my case. (Next on the agenda, check the other winbond superio chips to see if this is needed as well as I have the chip spec pdf files here as well. I noticed the other winbond chips use different addresses and values.) -- Roger http://www.eskimo.com/~roger/index.html Key fingerprint = 8977 A252 2623 F567 70CD 1261 640F C963 1005 1D61 Wed May 9 07:25:26 PDT 2007

2 3

[LinuxBIOS] r2645 - trunk/LinuxBIOSv2/src/superio/winbond/w83977tf
by svn＠openbios.org 09 May '07

09 May '07

Author: uwe Date: 2007-05-09 19:54:25 +0200 (Wed, 09 May 2007) New Revision: 2645 Modified: trunk/LinuxBIOSv2/src/superio/winbond/w83977tf/w83977tf_early_serial.c Log: The superio needs 0x87 sent twice to 0x3f0 to enable extended function (power-on strapping). Although this is already done in superio.c, it's not being done when w83977tf_early_serial.c is executed. As such, no console_init() without it. Signed-off-by: Roger Zauner <roger(a)eskimo.com> Acked-by: Uwe Hermann <uwe(a)hermann-uwe.de> Modified: trunk/LinuxBIOSv2/src/superio/winbond/w83977tf/w83977tf_early_serial.c =================================================================== --- trunk/LinuxBIOSv2/src/superio/winbond/w83977tf/w83977tf_early_serial.c 2007-05-09 15:11:03 UTC (rev 2644) +++ trunk/LinuxBIOSv2/src/superio/winbond/w83977tf/w83977tf_early_serial.c 2007-05-09 17:54:25 UTC (rev 2645) @@ -1,11 +1,25 @@ #include <arch/romcc_io.h> #include "w83977tf.h" +static inline void pnp_enter_ext_func_mode(device_t dev) +{ + unsigned int port = dev >> 8; + outb(0x87, port); + outb(0x87, port); +} -static void w83977tf_enable_serial(device_t dev, unsigned iobase) +static void pnp_exit_ext_func_mode(device_t dev) { + unsigned int port = dev >> 8; + outb(0xaa, port); +} + +static void w83977tf_enable_serial(device_t dev, unsigned int iobase) +{ + pnp_enter_ext_func_mode(dev); pnp_set_logical_device(dev); pnp_set_enable(dev, 0); pnp_set_iobase(dev, PNP_IDX_IO0, iobase); pnp_set_enable(dev, 1); + pnp_exit_ext_func_mode(dev); }

1 0

[LinuxBIOS] r2644 build service
by LinuxBIOS information 09 May '07

09 May '07

Dear LinuxBIOS readers! This is the automated build check service of LinuxBIOS. The developer "uwe" checked in revision 2644 to the LinuxBIOS source repository and caused the following changes: Change Log: Uncomment compression config variables. This should fix the abuild problems with the asi/mb_5bmlp (trivial). Signed-off-by: Uwe Hermann <uwe(a)hermann-uwe.de> Acked-by: Uwe Hermann <uwe(a)hermann-uwe.de> Build Log: Compilation of arima:hdama is still broken Compilation of artecgroup:dbe61 is still broken Configuration of asi:mb_5blmp has been fixed Compilation of ibm:e325 is still broken Compilation of ibm:e326 is still broken Compilation of iwill:dk8s2 is still broken Compilation of iwill:dk8x is still broken If something broke during this checkin please be a pain in uwe'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

← Newer
1
...
29
30
31
32
33
34
35
...
44
Older →

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

coreboot May 2007