September 2012 - coreboot - mail.coreboot.org

New patch to review for coreboot: 1bb5f1d buildgcc: Dont get misled if sysctl gets nothing
by Zheng Bao Sept. 28, 2012

Sept. 28, 2012

Zheng Bao (zheng.bao(a)amd.com) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/1548 -gerrit commit 1bb5f1d81602781aecfa807da17fd872ec3fc89b Author: Zheng Bao <fishbaozi(a)gmail.com> Date: Fri Sep 28 16:16:40 2012 +0800 buildgcc: Dont get misled if sysctl gets nothing Change-Id: I7cd63248eb8abb711cecce41e3f8a282b34aa126 Signed-off-by: Zheng Bao <zheng.bao(a)amd.com> Signed-off-by: Zheng Bao <fishbaozi(a)gmail.com> --- util/crossgcc/buildgcc | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/util/crossgcc/buildgcc b/util/crossgcc/buildgcc index b79c5f6..67477a5 100755 --- a/util/crossgcc/buildgcc +++ b/util/crossgcc/buildgcc @@ -315,7 +315,7 @@ if [ `uname` = "Darwin" ]; then # binaries in 10.6 (even if the kernel is 32bit) # For some weird reason, 10.5 autodetects an ABI=64 though # so we're setting the ABI explicitly here. - if [ `sysctl -n hw.optional.x86_64` -eq 1 ]; then + if [ -n "`sysctl -n hw.optional.x86_64 2>/dev/null`" ] && [ `sysctl -n hw.optional.x86_64 2>/dev/null` -eq 1 ]; then OPTIONS="ABI=64" else OPTIONS="ABI=32"

1 0

Patch set updated for coreboot: 1518063 crossgcc: Allow the non-gnu tar and patch work on XxxBSD
by Zheng Bao Sept. 28, 2012

Sept. 28, 2012

Zheng Bao (zheng.bao(a)amd.com) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/1502 -gerrit commit 1518063a5f96afeed1ae19b93011c37e13b2ce44 Author: Zheng Bao <fishbaozi(a)gmail.com> Date: Fri Sep 28 16:18:58 2012 +0800 crossgcc: Allow the non-gnu tar and patch work on XxxBSD For BSD, patch and tar are not default GNU. Add a work around to let the non-gun patch and tar work. Change-Id: I0a9d0bb0e535aa5e0dde146db330c3c8d7b4d8cb Signed-off-by: Zheng Bao <zheng.bao(a)amd.com> Signed-off-by: Zheng Bao <fishbaozi(a)gmail.com> --- util/crossgcc/buildgcc | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/util/crossgcc/buildgcc b/util/crossgcc/buildgcc index 67477a5..c2003d2 100755 --- a/util/crossgcc/buildgcc +++ b/util/crossgcc/buildgcc @@ -99,6 +99,15 @@ searchgnu() fi fi done + # A workaround for XxxBSD, whose nongnu patch and tar also work. + if [ `uname` = "FreeBSD" -o `uname` = "NetBSD" ]; then + if [ $1 != "make" ]; then + if test -x "`which $1 2>/dev/null`"; then + echo $1 + return + fi + fi + fi printf "${RED}ERROR:${red} Missing toolchain: $1${NC}\n" >&2 exit 1 }

1 0

New patch to review for coreboot: 078f22e nvramtool: Require no hw access for integrated Makefile
by Zheng Bao Sept. 28, 2012

Sept. 28, 2012

Zheng Bao (zheng.bao(a)amd.com) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/1544 -gerrit commit 078f22ea8bd365c3152a32e72052cccfa60c155b Author: Zheng Bao <fishbaozi(a)gmail.com> Date: Fri Sep 28 15:58:05 2012 +0800 nvramtool: Require no hw access for integrated Makefile The Makefile.inc is integraged into coreboot Makefile. It doesn't need to access to HW like cmos. It doesn't include cmos-hw-unix.c, which is only for individual tools running seperatedly. Change-Id: Ib00b5c3da63acb4120cb23eb7d661c5bc75d7c86 Signed-off-by: Zheng Bao <zheng.bao(a)amd.com> Signed-off-by: Zheng Bao <fishbaozi(a)gmail.com> --- util/nvramtool/Makefile.inc | 4 ---- 1 file changed, 4 deletions(-) diff --git a/util/nvramtool/Makefile.inc b/util/nvramtool/Makefile.inc index 0acda04..67710da 100644 --- a/util/nvramtool/Makefile.inc +++ b/util/nvramtool/Makefile.inc @@ -21,15 +21,11 @@ OS_ARCH = $(shell uname) NVRAMTOOLFLAGS := -I$(top)/util/nvramtool -ifeq ($(OS_ARCH), Darwin) -NVRAMTOOLFLAGS += -framework DirectHW -endif ifeq ($(OS_ARCH), NetBSD) NVRAMTOOLLDLFLAGS = -l$(shell uname -p) endif ifeq ($(shell uname -o), Cygwin) NVRAMTOOLFLAGS += -O2 -g -Wall -W -D__GLIBC__ -NVRAMTOOLLDFLAGS = -lioperm HOSTCFLAGS = endif

1 0

New patch to review for coreboot: 454ab9e nvramtool: uname in NetBSD doesnt take "-o"
by Zheng Bao Sept. 28, 2012

Sept. 28, 2012

Zheng Bao (zheng.bao(a)amd.com) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/1545 -gerrit commit 454ab9e057e49d4031bbe478b3cdcc1847edfda0 Author: Zheng Bao <fishbaozi(a)gmail.com> Date: Fri Sep 28 16:02:35 2012 +0800 nvramtool: uname in NetBSD doesnt take "-o" see the Netbsd manual: http://netbsd.gw.com/cgi-bin/man-cgi?uname++NetBSD-current Error output needs to be redirected. Change-Id: I1853a0162e14be0ee9d7971466499af6c72b2427 Signed-off-by: Zheng Bao <zheng.bao(a)amd.com> Signed-off-by: Zheng Bao <fishbaozi(a)gmail.com> --- util/nvramtool/Makefile | 2 +- util/nvramtool/Makefile.inc | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/util/nvramtool/Makefile b/util/nvramtool/Makefile index 6723a48..fe2efa0 100644 --- a/util/nvramtool/Makefile +++ b/util/nvramtool/Makefile @@ -42,7 +42,7 @@ endif ifeq ($(OS_ARCH), NetBSD) LDFLAGS = -l$(shell uname -p) endif -ifeq ($(shell uname -o), Cygwin) +ifeq ($(shell uname -o 2>/dev/null), Cygwin) LDFLAGS = -lioperm CFLAGS += -D__GLIBC__ endif diff --git a/util/nvramtool/Makefile.inc b/util/nvramtool/Makefile.inc index 67710da..8471b66 100644 --- a/util/nvramtool/Makefile.inc +++ b/util/nvramtool/Makefile.inc @@ -24,7 +24,7 @@ NVRAMTOOLFLAGS := -I$(top)/util/nvramtool ifeq ($(OS_ARCH), NetBSD) NVRAMTOOLLDLFLAGS = -l$(shell uname -p) endif -ifeq ($(shell uname -o), Cygwin) +ifeq ($(shell uname -o 2>/dev/null), Cygwin) NVRAMTOOLFLAGS += -O2 -g -Wall -W -D__GLIBC__ HOSTCFLAGS = endif

1 0

New patch to review for coreboot: f11e706 build.h: Re-run hostname if it doesn't take '-s' option.
by Zheng Bao Sept. 28, 2012

Sept. 28, 2012

Zheng Bao (zheng.bao(a)amd.com) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/1546 -gerrit commit f11e706dcf93e0f8c87191ab99622fa55959786e Author: Zheng Bao <fishbaozi(a)gmail.com> Date: Fri Sep 28 16:06:44 2012 +0800 build.h: Re-run hostname if it doesn't take '-s' option. Cygwin's hostname comes from coreutils, which does not support all the options that some other hostname implementations provide. Change-Id: Ia6bd9157c351f440ad225046638a6bf3f9cfba11 Signed-off-by: Zheng Bao <zheng.bao(a)amd.com> Signed-off-by: Zheng Bao <fishbaozi(a)gmail.com> --- Makefile.inc | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile.inc b/Makefile.inc index bcf0e6e..24b3c34 100644 --- a/Makefile.inc +++ b/Makefile.inc @@ -186,7 +186,7 @@ $(obj)/build.h: .xcompile printf "#define COREBOOT_LINKER \"$(shell LANG= $(LD) --version | head -n1)\"\n" >> $(obj)/build.ht printf "#define COREBOOT_COMPILE_TIME \"`LANG= date +%T`\"\n" >> $(obj)/build.ht printf "#define COREBOOT_COMPILE_BY \"$(subst \,@,$(shell PATH=$$PATH:/usr/ucb whoami))\"\n" >> $(obj)/build.ht - printf "#define COREBOOT_COMPILE_HOST \"$(shell hostname -s 2>/dev/null)\"\n" >> $(obj)/build.ht + printf "#define COREBOOT_COMPILE_HOST \"$(shell hostname -s 2>/dev/null || hostname 2>/dev/null)\"\n" >> $(obj)/build.ht printf "#define COREBOOT_COMPILE_DOMAIN \"$(shell test `uname -s` = "Linux" && dnsdomainname || domainname 2>/dev/null)\"\n" >> $(obj)/build.ht printf "#endif\n" >> $(obj)/build.ht mv $(obj)/build.ht $(obj)/build.h

1 0

Re: [coreboot] good cheap basic board for workshop
by Carl-Daniel Hailfinger Sept. 28, 2012

Sept. 28, 2012

Hi Ron, Am 27.09.2012 23:26 schrieb ron minnich: > Hi, any one have good ideas on a cheap basic board I could use as > tutorial at a workshop? Needs to available in brazil and argentina. > Chromebooks are not. I don't know any price comparison websites in .br or .ar, but I think some AMD E350 based boards (e.g. Asrock E350M1) may be a good target: cheap, readily available pretty much all over the world, and supported. Sorry for not having any verified suggestions. Regards, Carl-Daniel

1 0

good cheap basic board for workshop
by ron minnich Sept. 27, 2012

Sept. 27, 2012

Hi, any one have good ideas on a cheap basic board I could use as tutorial at a workshop? Needs to available in brazil and argentina. Chromebooks are not. thanks ron

1 0

Re: [coreboot] Loading large ROMs?
by Steve Goodrich Sept. 27, 2012

Sept. 27, 2012

Ah, I think I'd been at it too long. SeaBIOS is loading the ROM, not coreboot. Heh... Just needed a few hours of sleep! Thanks, Ron. I think I was looking in the wrong haystack for my needle. -- Steve G. ----- Original Message ----- From: "ron minnich" <rminnich(a)gmail.com> To: "Steve Goodrich" <steve.goodrich(a)se-eng.com> Cc: "coreboot" <coreboot(a)coreboot.org> Sent: Wednesday, September 26, 2012 4:58:03 PM Subject: Re: [coreboot] Loading large ROMs? I can't see a problem with this. we've loaded 128KiB seabios. Where did you want this ipxe rom to go? ron

2 1

New patch to review for coreboot: b5b93c5 libpayload: Don't leave temporary files behind
by Patrick Georgi Sept. 27, 2012

Sept. 27, 2012

Patrick Georgi (patrick(a)georgi-clan.de) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/1543 -gerrit commit b5b93c582db9fca1e893d5d582a3366ac8cf5ff4 Author: Patrick Georgi <patrick.georgi(a)secunet.com> Date: Wed Sep 26 15:26:05 2012 +0200 libpayload: Don't leave temporary files behind For some reason the rm -f didn't quite work on my system, but sending gcc output to /dev/null does. Change-Id: I7ece9aa9abe564bbc646ae53df1d3cd0c5aa84a2 Signed-off-by: Patrick Georgi <patrick.georgi(a)secunet.com> --- payloads/libpayload/bin/lpgcc | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/payloads/libpayload/bin/lpgcc b/payloads/libpayload/bin/lpgcc index 80c2266..8ee0dcf 100755 --- a/payloads/libpayload/bin/lpgcc +++ b/payloads/libpayload/bin/lpgcc @@ -58,10 +58,8 @@ BASE=`dirname $0` _LDSCRIPT="-Wl,-T,$_LIBDIR/libpayload.ldscript" trygccoption() { - $DEFAULT_CC $1 -S -xc /dev/null -o .$$.tmp &> /dev/null - RET=$? - rm -f .$$.tmp - return $RET + $DEFAULT_CC $1 -S -xc /dev/null -o /dev/null &> /dev/null + return $? } DEBUGME=0

1 0

New patch to review for coreboot: 855bb95 libpayload: fix fetching integers from CMOS as string
by Patrick Georgi Sept. 27, 2012

Sept. 27, 2012

Patrick Georgi (patrick(a)georgi-clan.de) just uploaded a new patch set to gerrit, which you can find at http://review.coreboot.org/1542 -gerrit commit 855bb95ca1ca00957de044f9f4362c6cd054cf9a Author: Patrick Georgi <patrick.georgi(a)secunet.com> Date: Wed Sep 26 15:11:48 2012 +0200 libpayload: fix fetching integers from CMOS as string %ull -> %llu Change-Id: I330f681d713be7eb444870f81330cf6e9869a4fa Signed-off-by: Patrick Georgi <patrick.georgi(a)secunet.com> --- payloads/libpayload/drivers/options.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/payloads/libpayload/drivers/options.c b/payloads/libpayload/drivers/options.c index 03b6d36..e2fd663 100644 --- a/payloads/libpayload/drivers/options.c +++ b/payloads/libpayload/drivers/options.c @@ -282,7 +282,7 @@ int get_option_as_string(const struct nvram_accessor *nvram, struct cb_cmos_opti /* only works on little endian. 26 bytes is enough for a 64bit value in decimal */ *dest = malloc(26); - sprintf(*dest, "%ull", *(u64*)raw); + sprintf(*dest, "%llu", *(u64*)raw); break; case 's': *dest = strdup(raw);

1 0