* use relative paths in ldscript.ld and crt0_includes.h
* avoid use of dd(1) in xcompile
* build libregex for kconfig, if necessary
* work around missing utsname on win32
* unlink targets before rename on win32
* implement (crude) mkstemp for win32
* avoid open/read/close, use fopen/fread/fclose instead
* don't free certain data structures in romcc on win32 to
avoid crashes (likely use-after-free())
* handle "\CRLF" and win32 style absolute paths (X:/ or X:\)
in romcc
* make lzma (part of cbfstool) build on XP
* implement ntohl/htonl on win32
* handle CRLF in awk script
* set larger stack for romcc on win32
Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Peter Stuge <peter@stuge.se>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4952 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
- Add static and const where possible.
- Turn some #warning entries into TODO comments.
- Add missing prototypes.
- Remove unused variables.
- Fix printf arguments or cast them as needed.
- Make sconfig output look better. Drop useless "PARSED THE TREE" output.
- Print "(this may take a while)" while building romcc. Add missing "\n".
Signed-off-by: Uwe Hermann <uwe@hermann-uwe.de>
Acked-by: Myles Watosn <mylesgw@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4874 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
CBFS uses sprintf, which requires vtxprintf, which requires (in the
current design) a nested function. That works on x86, but on PPC this
requires a trampoline. In the ROM stage, this is not available, so
remove the single use of sprintf and replace it with a direct string
handler - it's only used to fill in fixed-length hex values.
20090819-3-more-noreturns-in-romcc:
Mark two more functions in romcc as noreturn. Helps clang's scan-build a
bit
20090819-4-cbfsify-ppc:
Make PPC use CBFS. Support big endian ELF in cbfs-mkstage. Untested and
not complete yet.
20090819-5-fix-ppc-build:
The CBFS build system requires ROM_IMAGE_SIZE to have a somewhat
plausible value.
With fixes to tohex* functions as discussed on the list, and correct
function names.
Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Myles Watson <mylesgw@gmail.com>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4558 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
It's basically done with the following script and some manual fixup:
VARS=`grep ^define src/config/Options.lb | cut -f2 -d\ | grep -v ^CONFIG | grep -v ^COREBOOT |grep -v ^CC`
for VAR in $VARS; do
find . -name .svn -prune -o -type f -exec perl -pi -e "s/(^|[^0-9a-zA-Z_]+)$VAR($|[^0-9a-zA-Z_]+)/\1CONFIG_$VAR\2/g" {} \;
done
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4381 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
- Replace $(PWD) with $(CURDIR) in Makefiles. I don't know why
the Solaris version behaves differently, but CURDIR is a safe
choice on gnu make (and we require gnu make already)
- Use tail -1 instead of tail -n1 in a file that already relies on
tail -1 support in another place
- Use tail -1 as alternative to tail -n1 in another place
- Use #define for ulong_t in romcc, as that name is used on Solaris
- Avoid fprinting a null pointer. The standard doesn't mandate that
this is a special case, and Solaris doesn't implement it that way.
Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Ronald G. Minnich <rminnich@gmail.com>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4305 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
With this change, romcc knows -include and the build system uses it.
Also use a full path to settings.h because scan-build has trouble
finding it otherwise.
Signed-off-by: Patrick Georgi <patrick.georgi@coresystems.de>
Acked-by: Joseph Smith <joe@settoplinux.org>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4151 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
The compile_file calls seem to be in the wrong order, but
romcc actually requires it that (probably some stack-like
file processing)
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Patrick Georgi <patrick.georgi@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@4051 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
Due to the automatic nature of this update, I am self-acking. It worked in
abuild.
Signed-off-by: Stefan Reinauer <stepan@coresystems.de>
Acked-by: Stefan Reinauer <stepan@coresystems.de>
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@3053 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
making the output of "make -s" finally usable.. (still trivial, doesn't
change any logic or remove any 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@2894 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
- Special version for HDAMA rev G with 33Mhz test and reboot out.
- Support for CPU rev E, dual core, memory hoisting,
- corrected an SST flashing problem. Kernel bug work around (NUMA)
- added a Kernel bug work around for assigning CPU's to memory.
r2@gog: svnadmin | 2005-08-03 08:47:54 -0600
Create local LNXI branch
r1110@gog: jschildt | 2005-08-09 10:35:51 -0600
- Merge from Tom Zimmerman's additions to the hdama code for dual core
and 33Mhz fix.
r1111@gog: jschildt | 2005-08-09 11:07:11 -0600
Stable Release tag for HDAMA-1.1.8.10 and HDAMA-1.1.8.10LANL
r1112@gog: jschildt | 2005-08-09 15:09:32 -0600
- temporarily removing hdama tag to update to public repository. Will
reset tag after update.
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@2004 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
1. x86_setup_mtrr take address bit.
2. generic ht, pcix, pcie beidge...
3. scan bus and reset_bus
4. ht read ctrl to decide if the ht chain
is ready
5. Intel e7520 and e7525 support
6. new ich5r support
7. intel sb 6300 support.
yhlu patch
1. split x86_setup_mtrrs to fixed and var
2. if (resource->flags & IORESOURCE_FIXED ) return; in device.c pick_largest_resource
3. in_conherent.c K8_SCAN_PCI_BUS
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1982 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
- optimize_link_read_pointers compiles now on the solo so don't disable it.
- Start sorting out the confusion between and object and an initobject on the ppc ports
- Major bugfix release of romcc to support to remove preprocessor deficiencies.
The line and column numbers are computed are now correct. But watch out
the error messages sometimes report the location of the next token so things
are still a little skewed.
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1784 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
- Preprocessor constant expression evaluation is no long a special case so
unsigned long values can not be used.
- Undefined macros are not converted to 0. But a big warning is printed.
- Garbage at the of an #include directive is now done in tokens instead of
in characters.
This allows comments after an #include directive.
- Repaired a previously unnoticed regression in constant expression
evaluation. Logical expressions can now be evaluated again.
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1767 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
ram linuxbios_ram instead of linuxbios_c and linuxbios_payload...
- Reordered the linker sections so the LinuxBIOS fallback image can take more the 64KiB on x86
- ROM_IMAGE_SIZE now will work when it is specified as larger than 64KiB.
- Tweaked the reset16.inc and reset16.lds to move the sanity check to see if everything will work.
- Start using romcc's built in preprocessor (This will simplify header compiler checks)
- Add helper functions for examining all of the resources
- Remove debug strings from chip.h
- Add llshell to src/arch/i386/llshell (Sometime later I can try it...)
- Add the ability to catch exceptions on x86
- Add gdb_stub support to x86
- Removed old cpu options
- Added an option so we can detect movnti support
- Remove some duplicate definitions from pci_ids.h
- Remove the 64bit resource code in amdk8/northbridge.c in preparation for making it generic
- Minor romcc bug fixes
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1727 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
This includes more test cases
Lots of small bug fixes
A built in C preprocessor
Initial support for not inlining everything
__attribute__((noinline)) works
Better command line options and help
Constants arrays can be read at compile time
Asm statements that are not volatile will now be removed when their outputs go unused
Loads and stores that are not volatile will be removed when their values go unused
The number of FIXMES in the code is finally starting to go down.
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1582 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
- Reworked pnp superio device support. Now complete superio support is less than 100 lines.
- Added support for hard coding resource assignments in Config.lb
- Minor bug fixes to romcc
- Initial support for catching the x86 processor BIST error codes. I've only seen
this trigger once in production during a very suspcious reset but...
- added raminit_test to test the code paths in raminit.c for the Opteron
- Removed the IORESOURCE_SET bit and added IORESOURCE_ASSIGNED and IORESOURCE_STORED
so we can tell what we have really done.
- Added generic AGP/IOMMU setting code to x86
- Added an implementation of memmove and removed reserved identifiers from memcpy
- Added minimal support for booting on pre b3 stepping K8 cores
- Moved the checksum on amd8111 boards because our default location was on top of
extended RTC registers
- On the Hdama added support for enabling i2c hub so we can get at the temperature
sensors. Not that i2c bus was implemented well enough to make that useful.
- Redid the Opteron port so we should only need one reset and most of memory initialization
is done in cpu_fixup. This is much, much faster.
- Attempted to make the VGA IO region assigment work. The code seems to work now...
- Redid the error handling in amdk8/raminit.c to distinguish between a bad value
and a smbus error, and moved memory clearing out to cpufixup.
- Removed CONFIG_KEYBOARD as it was useless. See pc87360/superio.c for how to
setup a legacy keyboard properly.
- Reworked the register values for standard hardware, moving the defintions from
chip.h into the headers of the initialization routines. This is much saner
and is actually implemented.
- Made the hdama port an under clockers BIOS. I debuged so many interesting problems.
- On amd8111_lpc added setup of architectural/legacy hardware
- Enabled PCI error reporting as much as possible.
- Enhanded build_opt_tbl to generate a header of the cmos option locations so
that romcc compiled code can query the cmos options.
- In romcc gracefully handle function names that degenerate into function pointers
- Bumped the version to 1.1.6 as we are getting closer to 2.0
TODO finish optimizing the HT links of non dual boards
TODO make all Opteron board work again
TODO convert all superio devices to use the new helpers
TODO convert the via/epia to freebios2 conventions
TODO cpu fixup/setup by cpu type
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1390 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
- Support for compiling romcc on non x86 platforms
- new romc options -msse and -mmmx for specifying extra registers to use
- Bug fixes to device the device disable/enable framework and an amd8111 implementation
- Move the link specification to the chip specification instead of the path
- Allow specifying devices with internal bridges.
- Initial via epia support
- Opteron errata fixes
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@1200 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
- SMP cleanups (remove SMP only use CONFIG_SMP)
- Minor tweaks to romcc to keep it from taking forever compiling
- failover fixes
- Get a good implementation of k8_cpufixup and sizeram for the opteron
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@998 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
- Reworked the transformation into ssa form and now I catch all unitialized
variable uses.
- Several more test cases
- Bumped the version to 0.34
- Verified that -O2 the scc_transform now works.
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@934 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
properly. And a few long standing bugs have been rooted out and removed.
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@931 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
- Add several more test cases.
- Bump the version number to .32
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@919 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
- Register allocator bug fixes.
* coalesce_live_ranges now also updates the interference graph of live instructions
* resolve_tangle now avoids copies to phi
* correct_tangles is now called in a loop so that all tangles get fixed
- Bug the version of romcc to 0.30
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@886 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
- Add an additional consistency check to romcc and fix the more obvious problems it has uncovered
With this update there are no known silent failures in romcc.
- Update the memory initialization code to setup all 3 of the memory sizing registers properly
- In auto.c test our dynamic maximum amount of ram.
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@885 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
- Minor romcc fixes
- In smbus_wail_until_done a romcc glitch with || in romcc where it likes
to run out of registers. Use | to be explicit that I don't need the short
circuiting behavior.
- Remove unused #defines from coherent_ht.c
- Update the test in auto.c to 512M
- Add definition of log2 to romcc_io.h
- Implement SPD memory sizing in raminit.c
- Reduce the number of memory devices back 2 to for the SOLO board.
git-svn-id: svn://svn.coreboot.org/coreboot/trunk@883 2b7e53f0-3cfb-0310-b3e9-8179ed1497e1
Patrick Georgi