Commit Graph

4605 Commits

Author SHA1 Message Date
Stefan Reinauer 39162f7b47 oops... this is a critical issue. Some boards in the tree don't compile with
romcc even though they don't have CAR either. We need to check all boards and
fix those that are broken like this.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5362 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-07 01:41:01 +00:00
Stefan Reinauer f8b1923848 - unify use of SB_HT_CHAIN_UNITID_OFFSET_ONLY
- cleanup reset
- some minor warning fixes.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5361 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-07 00:38:09 +00:00
Stefan Reinauer 0ff769baa5 fix CK804 boards.
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5360 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-06 23:55:17 +00:00
Stefan Reinauer 8f2c616dbc No warnings day, next round.
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5359 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-06 21:50:21 +00:00
Stefan Reinauerstepan 233f186e95 fam10 acpi fix
Signed-off-by: Stefan Reinauer<stepan@coresystems.de> 
Acked-by: Marc Jones <marcj303@gmail.com>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5358 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-06 21:49:31 +00:00
Rudolf Marek 133647aa86 Signed-off-by: Rudolf Marek <r.marek@assembler.cz>
Acked-by: Stefan Reinauer <stepan@coresystems.de>

Add Asrock 939a785gmh motherboard. The ACPI needs more cleanup, could be done when cleaning
the Mahagony board. The SidePort mode does not work because AMD hardcoded memory type in rs780_gfx.c
The UMA is enabled instead. The board boots, network and int VGA works, IDE too. 



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5357 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-05 19:47:34 +00:00
Rudolf Marek f932c2edad Add RS785G, looks like it works although it is RV620.
Signed-off-by: Rudolf Marek <r.marek@assembler.cz>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5356 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-05 19:21:18 +00:00
Stefan Reinauer c9250c6740 device_t wants device/device.h
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5355 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-03 15:41:27 +00:00
Stefan Reinauer 10dd8b4a08 no need to udelay()
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5354 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-03 14:19:22 +00:00
Stefan Reinauer 3c8ac786c8 remove more warnings.
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5353 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-03 13:33:01 +00:00
Stefan Reinauer c65666f70d remove more warnings
rename amd64_main to stage1_main.. 
copy src/mainboard/via/vt8454c/debug.c to src/lib/debug.c

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5352 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-03 12:41:41 +00:00
Stefan Reinauer bed872dedf Add noreturn attribute to die() in romcc.
This makes life a lot easier for static analyzers such as scan-build.
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5351 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-03 00:03:07 +00:00
Stefan Reinauer 26afd18e10 remove more warnings.
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5350 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-02 22:31:35 +00:00
Stefan Reinauer b4a3fcfb4b oops, sorry, overlapping work
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5349 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-02 22:13:27 +00:00
Stefan Reinauer 720297c3d4 remove some more warnings
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5348 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-02 22:11:20 +00:00
Myles Watson b54deb77f0 Fix console prototypes for non-romcc boards.
Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>


git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5347 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-02 21:39:12 +00:00
Myles Watson a4fa5ee88c Fix includes for showallroutes.
Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>


git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5346 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-02 20:14:21 +00:00
Stefan Reinauer 7355c7590b The error message is misleading, even for a SPEW, because
the slot is empty, it's NOT a bad ID (and the message is
more confusing than helpful even in SPEW)

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
Acked-by: Myles Watson <mylesgw@gmail.com>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5345 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-02 16:30:25 +00:00
Stefan Reinauer 0c781b2694 - get rid of ASM_CONSOLE_LOGLEVEL except in two assembler files.
- start naming all versions of post code output "post_code()"

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5344 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-04-01 09:50:32 +00:00
Myles Watson 84b685af5f Consolidate ifdefs in src/lib/cbmem.c for readability.
Signed-off-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>


git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5343 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-31 14:57:55 +00:00
Stefan Reinauer 64ed2b7345 Drop \r\n and \n\r as both print_XXX and printk now do this internally.
Only some assembler files still have \r\n ... Can we move that part to C
completely?

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5342 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-31 14:47:43 +00:00
Stefan Reinauer 5a1f597085 This patch drops arch/i386/lib/console.c and arch/i386/lib/console_print.c and
makes include/console/console.h and console/console.c usable both in
__PRE_RAM__ and coreboot_ram stages.

While debugging this, I removed an indirection from the e7520 ram init code
(same as we did on a couple of other chipsets, removes some register pressure
  from romcc)

Also, drop remainders of CONFIG_USE_INIT (except the one odd piece of dead code 
		in cache_as_ram.inc)

Then some ap_romstage.c fixes, at least the nvidia/l1_2pvv compiled for me with
CONFIG_AP_CODE_IN_CAR set in Kconfig which it did not before.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Myles Watson <mylesgw@gmail.com>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5341 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-31 14:34:40 +00:00
Stefan Reinauer b8ac05d187 mini part of the patch on the mailing list to fix the boards again
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5340 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-31 00:06:12 +00:00
Stefan Reinauer f2b7a43f9d no USE_INIT
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5339 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 22:21:06 +00:00
Stefan Reinauer 3ac400e6e5 drop USE_INIT should be USE_PRINTK_IN_CAR here.
uint32_t should be u32
DEBUG_RAM_SETUP was failing on some northbridges
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5338 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 22:12:59 +00:00
Stefan Reinauer b01fe696d4 that version of console_tx_byte is private.
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5337 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 22:08:48 +00:00
Stefan Reinauer 23a3e79405 fix some amd k8 warnings
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5336 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 22:03:10 +00:00
Stefan Reinauer cca626817d get coreboot_ap memory training in cache mechanism in place. Didn't work since
Kconfig (needs more patches to ap_romstage.c but this is a first step)

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5335 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 21:52:05 +00:00
Stefan Reinauer c269d237f9 fix some southbridge warnings (trivial)
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5334 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 21:48:23 +00:00
Stefan Reinauer 495b92b787 - drop unneeded Makefile.inc
- drop ap_romstage from Fam10 boards, the mechanism was never used on Fam10

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5333 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 21:47:10 +00:00
Stefan Reinauer 0b2f18523a indent mptable.c
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5332 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 21:43:15 +00:00
Stefan Reinauer fc2b8ec5ad drop USE_INIT from mainboard Kconfig files, it's already set in src/Kconfig
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5331 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 21:41:33 +00:00
Stefan Reinauer acdd52fa82 make crt0s and ldscripts evaluate late, so the chipset_* variables are there at
the time they are finally used. This should solve the Problem Myles was seeing
earlier today.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5330 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 20:32:01 +00:00
Stefan Reinauer 0ed0b7cfe8 abuild:
it seems make oldconfig is not safe for parallel compilation. The problem never
occured in my tests, but go back to non-parallel build again. Also pass on the 
return value of make oldconfig correctly.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5329 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 15:49:14 +00:00
Stefan Reinauer abdf84884e abuild:
- increase context to 6 lines in case of build error
- add update mechanism to automatically produce fallback+normal in one image.
- tighten up output
- in-coreboot-builds makefile main target is now "all" as coreboot.rom matches
  a file
- time measurement now includes "make config" step
- actually allow long-implemented long version of --remove|-r option.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5328 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 14:02:19 +00:00
Stefan Reinauer fd0bef825d build_opt_tbl: error instead of null pointer access if no checksum is defined.
xe7501devkit: If cmos.layout is used, there must be a checksum.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5327 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 09:59:23 +00:00
Stefan Reinauer c719f1a506 add CONFIG_NO_POST as it is used in the code, and move it together with CONFIG_SERIAL_POST
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5326 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 09:57:28 +00:00
Stefan Reinauer 8b547b1980 reduce warnings in MCP55 and Fam10 code
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5325 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 09:56:35 +00:00
Stefan Reinauer 9b70cb6243 fix ms9652_fam10
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5324 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 08:20:37 +00:00
Stefan Reinauer 9050b71802 unify cmos.layout wrt AMD extended configuration registers.
This removes double preprocessor define warnings from many boards.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5323 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 08:01:43 +00:00
Stefan Reinauer 7ac3df8ae2 this fixes 3 of 4 mainboards failed by the error on missing IRQ_SLOT_COUNT.
The third, MSI MS9652, creates an IRQ table on the fly, no kidding.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5322 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-30 00:22:29 +00:00
Stefan Reinauer 8e726b7363 Don't include option_table.h every time we include mc146818rtc.h, that was a
stupid idea. Instead include it where it is needed. And add some explicit
dependencies to it.

Also, error for missing IRQ_SLOT_COUNT for now, so we can fix up the boards.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5321 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-29 23:01:35 +00:00
Stefan Reinauer 798ef2893c This drops the ASSEMBLY define from romstage.c, too
(since it's not assembly code, this was a dirty hack anyways)
Also run 
    awk 1 RS= ORS="\n\n" < $FILE > $FILE.nonewlines
    mv $FILE.nonewlines $FILE
on romstage.c because my perl -pi -e 's,#define ASSEMBLY 1,,g' */*/romstage.c 
cut some holes into the source.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>




git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5320 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-29 22:08:01 +00:00
Stefan Reinauer 72bdfeb698 - drop GCC 2.x workaround, GCC 2.x is not supported anymore.
- warn if we didn't specify IRQ_SLOT_COUNT in Kconfig
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5319 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-29 22:05:26 +00:00
Stefan Reinauer 86b6dba7ea trivial warning cleanups
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5318 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-29 21:56:26 +00:00
Stefan Reinauer 8655412673 - include option_table.h when it is created, and that's HAVE_OPTION_TABLE
- add some __PRE_RAM__ guards where needed
- use OPTION_TABLE_H

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5317 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-29 21:50:53 +00:00
Stefan Reinauer 2c5dc65949 qemu: drop "northbridge.c" from src/cpu/...
It's not a real northbridge, so I just move it into the mainboard directory for
now (until we maybe have a qemu-q35 image some day?)

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5316 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-29 21:17:25 +00:00
Stefan Reinauer 5e32823a68 __PRE_RAM__ is now correctly specified in the Makefile. No need to hack it into
romstage.c anymore

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5315 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-29 19:19:16 +00:00
Stefan Reinauer 8013d5a568 cpu_driver structs need to be const, too...
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5314 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-29 19:04:13 +00:00
Stefan Reinauer b5828d7445 This patch drops the coreboot CMOS checksum ranges from Kconfig because
the information is already specified in cmos.layout. coreboot is changed
to use that version instead.

Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmai.com>



git-svn-id: svn://svn.coreboot.org/coreboot/trunk@5313 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
2010-03-29 17:14:28 +00:00