Stojsavljevic, Zoran wrote:
> > If you are distributing coreboot then it would be significantly
> > easier for you to work with the community with the goal to have
> > your needed functionality available in the public coreboot git.
>
> INTEL is a different animal in this "Open Source" world.
Yes, I know. :)
> I need to work it out with INTEL first...
> Please, stay Tuned.
Glad to hear that you're on it! Thanks for your efforts!
//Peter
Am 19.08.2013 14:21, schrieb Stojsavljevic, Zoran:
> The latest Coreboot I git cloned an hour ago (coreboot
> v4.0-4632-g4159a80) does compile and link with latest git clone FILO
> (with some .autogen.sh script changes and taking off USB drivers,
> since they produce errors, to adapt FILO positioning). Since I cannot
> use this one (I do not have hooks to use IVB FSP), I need to direct
> relevant people to make this adjustment and commit these changes for
> x86 to latest Coreboot git tree.
For the short term: Consider updating libpayload in your tree to the
latest version while keeping coreboot in your current state. This works
since these are really two independent projects.
However you will fare better long term by keeping your coreboot tree
close to upstream.
Regards,
Patrick
Hi Zoran,
Glad that you found the source of the problem!
Stojsavljevic, Zoran wrote:
> I need to direct relevant people to make this adjustment and commit
> these changes for x86 to latest Coreboot git tree.
If you are distributing coreboot then it would be significantly
easier for you to work with the community with the goal to have
your needed functionality available in the public coreboot git.
Like with other open source projects.
Kind regards
//Peter
Am 19.08.2013 09:37, schrieb Stojsavljevic, Zoran:
> CPPFLAGS := -nostdinc -imacros $(obj)/config.h
>
> ! ## CPPFLAGS += -I$(INCPAYLOAD) -I$(INCPAYLOAD)/$(ARCHDIR-y)
>
> ! CPPFLAGS += -I$(INCPAYLOAD) -I$(INCPAYLOAD)/i386
> -I$(INCPAYLOAD)/i386/arch
>
> CPPFLAGS += -I$(ARCHDIR-y)/include -Iinclude -I$(obj)
>
> CPPFLAGS += -I$(GCCINCDIR)
>
My guess is that the libpayload in your coreboot tree is older. FILO's
master branch was adapted to an i386->x86 rename in December 2012.
Please retry building FILO in a coreboot master checkout (so you get the
master libpayload as well). These are no ABI changes, so a FILO payload
built that way is still usable with an older coreboot tree.
Regards,
Patrick
Hi Zoran,
Stojsavljevic, Zoran wrote:
> And then I got the following errors (end of compilation):
> CC build/fs/fsys_reiserfs.o
> CC build/fs/fsys_iso9660.o
> CC build/fs/fsys_cbfs.o
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/fs/fsys_cbfs.c: In function 'cbfs_dir':
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/fs/fsys_cbfs.c:62:3: warning: implicit declaration of function 'be32toh' [-Wimplicit-function-declaration]
> CC build/drivers/ide_new.o
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c: In function 'ob_ide_error':
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:182:17: warning: variable 'old_cdb' set but not used [-Wunused-but-set-variable]
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:162:16: warning: variable 'err' set but not used [-Wunused-but-set-variable]
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c: In function 'ob_ide_atapi_packet':
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:563:17: error: 'struct request_sense' has no member named 'sense_key'
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:565:24: error: 'struct request_sense' has no member named 'sense_key'
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c: In function 'ob_ide_atapi_drive_ready':
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:679:2: warning: implicit declaration of function 'be32toh' [-Wimplicit-function-declaration]
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:625:6: warning: unused variable 'i' [-Wunused-variable]
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c: In function 'ob_ide_read_ata_chs':
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:742:27: error: 'struct ata_sector' has no member named 'low'
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:733:20: warning: variable 'ata_sector' set but not used [-Wunused-but-set-variable]
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c: In function 'ob_ide_read_ata_lba48':
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:795:27: error: 'struct ata_sector' has no member named 'low'
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:796:27: error: 'struct ata_sector' has no member named 'high'
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:784:20: warning: variable 'ata_sector' set but not used [-Wunused-but-set-variable]
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c: In function 'ob_ide_fixup_id':
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:887:2: warning: implicit declaration of function 'le16toh' [-Wimplicit-function-declaration]
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:888:2: warning: implicit declaration of function 'le32toh' [-Wimplicit-function-declaration]
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c: In function 'find_ide_controller':
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:1279:19: warning: variable 'device' set but not used [-Wunused-but-set-variable]
> /home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:1279:11: warning: variable 'vendor' set but not used [-Wunused-but-set-variable]
> make: *** [/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/build/drivers/ide_new.o] Error 1
>
> Any idea why this is happening?
I don't know about your environment but it seems to me that the FILO
code is buggy. It is not so popular anymore so it's possible that
there is some bitrot. Hopefully you can help look into what the
problem is.
> Did somehow I misconfigured coreboot/payloads/libpayload
> (it was compiled prior these steps)?
I'm not sure what you mean, but yes it has been possible to build
FILO in the past. If you can it would be great if you use git bisect
to find the commit which broke the build.
//Peter
Hello to Coreboot community,
Went to filo Coreboot page: http://www.coreboot.org/FILO
As written, executed the following:
cd coreboot/payloads
git clone http://review.coreboot.org/p/filo.git
cd filo
make menuconfig
make produces several errors, on the way, so I did temporarily fix it:
[zoran@localhost filo]$ diff -c Makefile Makefile.orig
*** Makefile 2013-08-16 14:54:09.314253711 +0200
--- Makefile.orig 2013-08-16 15:17:06.190593523 +0200
***************
*** 91,98 ****
ARCHDIR-$(CONFIG_TARGET_I386) := x86
CPPFLAGS := -nostdinc -imacros $(obj)/config.h
! ## CPPFLAGS += -I$(INCPAYLOAD) -I$(INCPAYLOAD)/$(ARCHDIR-y)
! CPPFLAGS += -I$(INCPAYLOAD) -I$(INCPAYLOAD)/i386 -I$(INCPAYLOAD)/i386/arch
CPPFLAGS += -I$(ARCHDIR-y)/include -Iinclude -I$(obj)
CPPFLAGS += -I$(GCCINCDIR)
--- 91,97 ----
ARCHDIR-$(CONFIG_TARGET_I386) := x86
CPPFLAGS := -nostdinc -imacros $(obj)/config.h
! CPPFLAGS += -I$(INCPAYLOAD) -I$(INCPAYLOAD)/$(ARCHDIR-y)
CPPFLAGS += -I$(ARCHDIR-y)/include -Iinclude -I$(obj)
CPPFLAGS += -I$(GCCINCDIR)
And then I got the following errors (end of compilation):
CC build/fs/fsys_reiserfs.o
CC build/fs/fsys_iso9660.o
CC build/fs/fsys_cbfs.o
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/fs/fsys_cbfs.c: In function 'cbfs_dir':
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/fs/fsys_cbfs.c:62:3: warning: implicit declaration of function 'be32toh' [-Wimplicit-function-declaration]
CC build/drivers/ide_new.o
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c: In function 'ob_ide_error':
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:182:17: warning: variable 'old_cdb' set but not used [-Wunused-but-set-variable]
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:162:16: warning: variable 'err' set but not used [-Wunused-but-set-variable]
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c: In function 'ob_ide_atapi_packet':
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:563:17: error: 'struct request_sense' has no member named 'sense_key'
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:565:24: error: 'struct request_sense' has no member named 'sense_key'
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c: In function 'ob_ide_atapi_drive_ready':
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:679:2: warning: implicit declaration of function 'be32toh' [-Wimplicit-function-declaration]
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:625:6: warning: unused variable 'i' [-Wunused-variable]
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c: In function 'ob_ide_read_ata_chs':
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:742:27: error: 'struct ata_sector' has no member named 'low'
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:733:20: warning: variable 'ata_sector' set but not used [-Wunused-but-set-variable]
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c: In function 'ob_ide_read_ata_lba48':
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:795:27: error: 'struct ata_sector' has no member named 'low'
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:796:27: error: 'struct ata_sector' has no member named 'high'
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:784:20: warning: variable 'ata_sector' set but not used [-Wunused-but-set-variable]
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c: In function 'ob_ide_fixup_id':
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:887:2: warning: implicit declaration of function 'le16toh' [-Wimplicit-function-declaration]
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:888:2: warning: implicit declaration of function 'le32toh' [-Wimplicit-function-declaration]
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c: In function 'find_ide_controller':
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:1279:19: warning: variable 'device' set but not used [-Wunused-but-set-variable]
/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/drivers/ide_new.c:1279:11: warning: variable 'vendor' set but not used [-Wunused-but-set-variable]
make: *** [/home/zoran/projects/cc2_dev_coreboot/coreboot/payloads/filo/build/drivers/ide_new.o] Error 1
Any idea why this is happening? Did somehow I misconfigured coreboot/payloads/libpayload (it was compiled prior these steps)?
Thank you,
Zoran
_______
Most of The Time you should be "intel inside" to be capable to think "out of the box".
Intel GmbH
Dornacher Strasse 1
85622 Feldkirchen/Muenchen, Deutschland
Sitz der Gesellschaft: Feldkirchen bei Muenchen
Geschaeftsfuehrer: Christian Lamprechter, Hannes Schwaderer, Douglas Lusk
Registergericht: Muenchen HRB 47456
Ust.-IdNr./VAT Registration No.: DE129385895
Citibank Frankfurt a.M. (BLZ 502 109 00) 600119052
Am 2013-08-15 14:01, schrieb Andrew Wu:
> is 256-byte aligned, otherwise it will failback to use slower MOV
> loop. MOV loop is much slower then
> REP MOVSL on Vortex86EX, because reading instructions from ROM is slow.
Can't you cache the ROM area?
Other than that: Optimizing the common code is better than a special
case. memmove won't change much, so that's a stable interface, but you
never know what will happen (and few people will be able to test their
changes on Vortex86).
Patrick
On Fri, Aug 16, 2013 at 2:19 AM, Luke Leighton
<luke.leighton(a)pathintel.com>wrote:
> On 16 August 2013 00:16, David Hendricks <dhendrix(a)google.com> wrote:
> > Hi Luke,
> > You may be in luck -- That chipset is similar to a couple others which
> are
> > already supported for the "Stumpy" and "Lumpy" ChromeOS products.
>
> awesome. hey i have one of those samsung chromebox's. where's the
> spec for the chipset on those?
>
They use the 6-series/C200/HM65 chipset. The public datasheet is here:
http://www.intel.com/content/www/us/en/chipsets/6-chipset-c200-chipset-data…
The newer products (Acer C7, Lenovo X131e, and HP 14-c050 Chromebooks) use
the 7-Series/C210 chipset. The public datasheet is here:
http://www.intel.com/content/www/us/en/chipsets/7-series-chipset-pch-datash…
It's been a while since I've looked at the huge matrix of codenames for
those generations of chipsets, but IIRC Cougar Point is in there
somewhere...
BTW -- For your goals of PXE booting you might want to look into SeaBIOS (
http://www.coreboot.org/SeaBIOS) and IPXE (http://www.coreboot.org/IPXE).
--
David Hendricks (dhendrix)
Systems Software Engineer, Google Inc.
On 16 August 2013 00:16, David Hendricks <dhendrix(a)google.com> wrote:
> Hi Luke,
> You may be in luck -- That chipset is similar to a couple others which are
> already supported for the "Stumpy" and "Lumpy" ChromeOS products.
awesome. hey i have one of those samsung chromebox's. where's the
spec for the chipset on those?
> Of course, the devils are always in the details and this being an
> Intel-based product will likely require proprietary bits, so there might be
> other business considerations to take into account. I recommend getting in
> touch with Sage Engineering (http://www.se-eng.com/).
ok thanks, that's a good lead.
On Thu, 2013-08-15 at 16:16 -0700, David Hendricks wrote:
> Hi Luke,
> You may be in luck -- That chipset is similar to a couple others which are
> already supported for the "Stumpy" and "Lumpy" ChromeOS products.
>
Funny you should mention Stumpy and Lumpy since the source of firmware
one can download at coreboot.org git for these devices has been broken
for several weeks now.
Where can one download the sources and build toolchain for the firmware
images Google ships these products with? This looks a lot like license
terms violation to me, when customer cannot recreate the binary.
I think I have asked at #coreboot a couple of times how to track the
revision tags or commit hashes from a production image to the sources.
And have gotten no answer.
And to be clear, I am _not_ referring to the sources of the systemagent
binary blob of these Intel chipsets here.
Kyösti
