the following patch was just integrated into master:
commit c1ef740d88ea88c4ec7ef5449bd8d348fc15a9a2
Author: Duncan Laurie <dlaurie(a)chromium.org>
Date: Tue Apr 23 08:39:19 2013 -0700
haswell: Fix VGA option rom INT15 handler
The format of this function changed but was not updated in
all mainboards. This fixes BaskingRidge and WTM2.
The int15 handler no longer takes a regs structure as an
argument and instead uses global variables. The yabel interface
is now similar enough that we can drop the duplicate handler.
Change-Id: Ia717ae14f99cee6d83ccdb1e26b9d7defe1638c4
Signed-off-by: Duncan Laurie <dlaurie(a)chromium.org>
Reviewed-on: https://gerrit.chromium.org/gerrit/48896
Reviewed-by: Aaron Durbin <adurbin(a)chromium.org>
Reviewed-by: Stefan Reinauer <reinauer(a)google.com>
Reviewed-on: http://review.coreboot.org/3230
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich(a)gmail.com>
See http://review.coreboot.org/3230 for details.
-gerrit
Ronald G. Minnich (rminnich(a)gmail.com) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/3238
-gerrit
commit 35bac6837e90dd98083ababc57e36c5963c79f5f
Author: Ronald G. Minnich <rminnich(a)gmail.com>
Date: Fri May 10 22:18:09 2013 +0200
EXYNOS5250/SNOW: fix the build script. Add a script to get the bl1.
build-snow got broken when the snow makefile improved. So fix it.
While we're at it, create a script like the update-microcode
scripts that gets the bl1. I thought about making this a common
script but the various names and paths always evolve, leaving
me thinking it's not worth it. This script is just a
piece of the snow build script.
Change-Id: I65c0f8697a978c62fe12533c4f0152d14dbaefda
Signed-off-by: Ronald G. Minnich <rminnich(a)gmail.com>
---
src/cpu/samsung/exynos5250/update-bl1.sh | 26 ++++++++++++++++++++++++++
src/vendorcode/google/chromeos/build-snow | 21 ++++-----------------
2 files changed, 30 insertions(+), 17 deletions(-)
diff --git a/src/cpu/samsung/exynos5250/update-bl1.sh b/src/cpu/samsung/exynos5250/update-bl1.sh
new file mode 100644
index 0000000..e47b25d
--- /dev/null
+++ b/src/cpu/samsung/exynos5250/update-bl1.sh
@@ -0,0 +1,26 @@
+#!/bin/sh
+
+BL1_NAME="E5250.nbl1.bin"
+BL1_PATH="3rdparty/cpu/samsung/exynos5250/"
+BL1_URL="http://commondatastorage.googleapis.com/chromeos-localmirror/distfiles/exyn…"
+
+get_bl1() {
+ mkdir -p "${BL1_PATH}"
+ cd "${BL1_PATH}"
+ wget "${BL1_URL}" -O bl1.tbz2
+ tar jxvf bl1.tbz2
+ mv "exynos-pre-boot/firmware/${BL1_NAME}" .
+ rm -rf exynos-pre-boot
+ if [ ! -e "${BL1_NAME}" ]; then
+ echo "Error getting BL1"
+ fi
+}
+
+main() {
+ if [ ! -e ${BL1_PATH}/${BL1_NAME} ]; then
+ get_bl1
+ fi
+}
+
+set -e
+main "$@"
diff --git a/src/vendorcode/google/chromeos/build-snow b/src/vendorcode/google/chromeos/build-snow
index a749ba5..da1566b 100755
--- a/src/vendorcode/google/chromeos/build-snow
+++ b/src/vendorcode/google/chromeos/build-snow
@@ -44,29 +44,16 @@ get_bl1() {
fi
}
-merge_bl1() {
- local outfile="$1"
- local bl1="${BL1_PATH}/${BL1_NAME}"
-
- if [ ! -e "$bl1" ]; then
- get_bl1
- fi
-
- # use the new BL1 which supports 30KB BL2/SPL/Coreboot
- local size="$(stat -c "%s" "$outfile")"
- local bl1_size="$(stat -c "%s" "$bl1")"
-
- [ "$bl1_size" = "$((0x2000))" ] || die "Incorrect BL1 input file."
- dd if="$bl1" of=${outfile} conv=notrunc
-}
-
is_servod_ready() {
ps -C servod >/dev/null 2>&1
}
main() {
+ if [ ! -e "$bl1" ]; then
+ get_bl1
+ fi
+
make
- merge_bl1 "$OUTPUT"
create_diff_192k "$OUTPUT" "$TMP_DIFF"
echo "OK: Generated image (with BL1) in $OUTPUT"
if is_servod_ready; then
Martin Roth (martin.roth(a)se-eng.com) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/3237
-gerrit
commit 9ead000e548f4ddac8b00f7389b52b5644df1f00
Author: Martin Roth <martin.roth(a)se-eng.com>
Date: Fri May 10 13:40:29 2013 -0600
Make early x86 POST codes written to IO port optional
This continues the work done in patch 6b908d08ab
http://review.coreboot.org/#/c/1685/
and makes the early x86 post codes follow the same options.
Change-Id: Idf0c17b27b3516e79a9a53048bc203245f7c18ff
Signed-off-by: Martin Roth <martin.roth(a)se-eng.com>
---
src/include/cpu/x86/post_code.h | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/src/include/cpu/x86/post_code.h b/src/include/cpu/x86/post_code.h
index 20b87c6..7465f62 100644
--- a/src/include/cpu/x86/post_code.h
+++ b/src/include/cpu/x86/post_code.h
@@ -1,7 +1,12 @@
#include <console/post_codes.h>
+
+#if CONFIG_IO_POST
#define post_code(value) \
movb $value, %al; \
- outb %al, $0x80
+ outb %al, $CONFIG_IO_POST_PORT
+#else
+#define post_code(value)
+#endif
the following patch was just integrated into master:
commit ec664bcfa4ac3151f58853565b930b20b61b1bdd
Author: Stefan Reinauer <reinauer(a)chromium.org>
Date: Thu May 9 14:06:04 2013 -0700
romcc: support attribute((packed))
right now this is just a fake option to get rid of ifdefs in
coreboot's code.
Change-Id: I59233f3c1d266b4e716a5921e9db298c7f96751d
Signed-off-by: Stefan Reinauer <reinauer(a)google.com>
Reviewed-on: http://review.coreboot.org/3225
Reviewed-by: Ronald G. Minnich <rminnich(a)gmail.com>
Tested-by: build bot (Jenkins)
See http://review.coreboot.org/3225 for details.
-gerrit
the following patch was just integrated into master:
commit 711a6fde0db2f7a0c388fc97faec1013caaa2789
Author: Ronald G. Minnich <rminnich(a)gmail.com>
Date: Fri May 3 18:25:27 2013 +0200
Get rid of MAXIMUM_CONSOLE_LOGLEVEL; compile all messages into the coreboot binary
This option has never had much if any use. It solved a problem over 10
years ago that resulted from an argument over the value or lack thereof
of including all the debug strings in a coreboot image. The answer is
in: it's a good idea to maintain the capability to print all messages,
for many reasons.
This option is also misleading people, as in a recent discussion, to
believe that log messges are controlled at build time in a way they are
not. For the record, from this day forward, we can print messages at all
log levels and the default log level is set at boot time, as directed by
DEFAULT_CONSOLE_LOGLEVEL. You can set the default to 0 at build time and
if you are having trouble override it in CMOS and get more messages.
Besides, a quick glance shows it's always set to max (9 in this case) in
the very few cases (1) in which it is set.
Change-Id: I60c4cdaf4dcd318b841a6d6c70546417c5626f21
Signed-off-by: Ronald G. Minnich <rminnich(a)gmail.com>
Reviewed-on: http://review.coreboot.org/3188
Tested-by: build bot (Jenkins)
See http://review.coreboot.org/3188 for details.
-gerrit
the following patch was just integrated into master:
commit c5e036a04368186fe73925c6fb101c594513391c
Author: Stefan Reinauer <reinauer(a)chromium.org>
Date: Thu May 9 14:33:14 2013 -0700
Get rid of a number of __GNUC__ checks
In the process of streamlining coreboot code and getting
rid of unneeded ifdefs, drop a number of unneeded checks
for the GNU C compiler. This also cleans up x86emu/types.h
significantly by dropping all the duplicate types in there.
Change-Id: I0bf289e149ed02e5170751c101adc335b849a410
Signed-off-by: Stefan Reinauer <reinauer(a)google.com>
Reviewed-on: http://review.coreboot.org/3226
Reviewed-by: Ronald G. Minnich <rminnich(a)gmail.com>
Tested-by: build bot (Jenkins)
See http://review.coreboot.org/3226 for details.
-gerrit
the following patch was just integrated into master:
commit b8b3e8bff32ee7dddcacec11e015f6683783eb2f
Author: Denis 'GNUtoo' Carikli <GNUtoo(a)no-log.org>
Date: Thu May 9 16:14:59 2013 +0200
Asus M4A785T-M: Add CMOS defaults.
After removing power and the CMOS Battery, putting it back
and booting coreboot we have:
# ./nvramtool -a
boot_option = Fallback
last_boot = Fallback
ECC_memory = Enable
baud_rate = 115200
hw_scrubber = Enable
interleave_chip_selects = Enable
max_mem_clock = 400Mhz
multi_core = Enable
power_on_after_fail = Disable
debug_level = Spew
boot_first = HDD
boot_second = Fallback_Floppy
boot_third = Fallback_Network
boot_index = 0xf
boot_countdown = 0xc
slow_cpu = off
nmi = Enable
iommu = Enable
nvramtool: Can not read coreboot parameter user_data because layout info specifies CMOS area that is too wide.
nvramtool: Warning: Coreboot CMOS checksum is bad.
Change-Id: Ifa09c7a468e3e0713b426763266ae633e67d8397
Signed-off-by: Denis 'GNUtoo' Carikli <GNUtoo(a)no-log.org>
Reviewed-on: http://review.coreboot.org/3224
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich(a)gmail.com>
See http://review.coreboot.org/3224 for details.
-gerrit
the following patch was just integrated into master:
commit e04dc7529fdeb2730b826b687fc365e9720d6647
Author: Siyuan Wang <wangsiyuanbuaa(a)gmail.com>
Date: Mon Apr 15 17:11:19 2013 +0800
AMD Thatcher: remove unused macros in PlatformGnbPcieComplex.h
The macros GNB_GPP_PORTx_PORT_PRESENT, GNB_GPP_PORTx_SPEED_MODE,
GNB_GPP_PORTx_LINK_ASPM and GNB_GPP_PORTx_CHANNEL_TYPE are not used.
Change-Id: I5c7b7d45880367dba452ebcd4f01fbd0c15aac22
Signed-off-by: Siyuan Wang <SiYuan.Wang(a)amd.com>
Signed-off-by: Siyuan Wang <wangsiyuanbuaa(a)gmail.com>
Reviewed-on: http://review.coreboot.org/3087
Reviewed-by: Paul Menzel <paulepanter(a)users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martin.roth(a)se-eng.com>
Reviewed-by: Dave Frodin <dave.frodin(a)se-eng.com>
Reviewed-by: Bruce Griffith <Bruce.Griffith(a)se-eng.com>
See http://review.coreboot.org/3087 for details.
-gerrit