Hi all.
While working on LinuxBios i erased bios content in the
origional flashrom chip while doing this my system become strucked at
that time when i rebooted the system but system not booted.
this is hapened while building the coreboot for MSI board by using the
AsRock board ..at the same time my harddisk is also damaged right now i dont
have backup file(origional bios ),but in my office i dont have other AsRock
motherboard for taking the bios content from that ,but i downloaded the bios
content for AsRock board from the website and then i tried to flash that
content to the erased flashrom chip but flashing is not happening ..... is
there any idea to get my bios back please inform me im in big troble
................
With regards
Malatesh kamatad
This patch consolidates the payload .conf files to eliminate redundancy. A
couple of them were not consolidated because it appeared that they will be
more specialized in the future.
follow the patch with:
svn rm config/payloads/etherboot.conf
svn rm config/payloads/memtest.conf
svn rm config/payloads/tint.conf
svn rm config/payloads/grub2.conf
svn rm config/payloads/gpxe.conf
svn rm config/payloads/coreinfo.conf
svn rm config/payloads/filo.conf
since they are no longer needed.
file by file changes:
Makefile:
Eliminated a payload copy to simplify build.
Added empty targets for custom and custom-clean.
packages/coreboot-v3/coreboot-v3.mk:
Same payload copy change as Makefile.
packages/ofw/ofw.mk:
Moved the SVN information here from ofw.conf because it is here for every
other payload.
config/payloads/ofw.conf:
The only thing keeping this file here is a comment about crc32sum.
config/payloads/payloads.conf:
Change the way PAYLOAD and PAYLOAD-y are set. Both are needed for lab since
lab is the name of the payload, but isn't an actual payload.
config/payloads/lab.conf:
Removed redundant PAYLOAD assignment.
config/payloads/openbios.conf:
Include the generic and leave this file for possible fcode-utils dependency.
config/payloads/kernel.conf:
Removed redundant PAYLOAD and PAYLOAD-y assignment.
config/payloads/Config.in:
Removed extra menu for Custom Payload with only one option in it.
config/payloads/custom.conf:
Simplified this file since there was no CUSTOM_PAYLOAD variable set anywhere.
Included Makefile.generic so that lzma compression would just work.
Signed-off-by: Myles Watson <mylesgw(a)gmail.com>
Thanks,
Myles
Hi,
please be careful with any recent gcc if you expect it to compile
v3/lib/lar.c the way it is intended to. We are likely to be affected by
US-CERT Vulnerability Note VU#162289: C compilers may silently discard
some wraparound checks.
More info here:
http://lwn.net/Articles/278137/http://www.kb.cert.org/vuls/id/162289
Basically, it turned out that a long-time recommended C secure
programming practice depended on undefined behaviour and nobody figured
this out for years. Now that compilers optimize away undefined code all
those wraparound checks explode.
Most of the proposed fixes to existing code so far have been ugly
(casting pointers to unsigned long) or advocate changing the code
structure (calling an extra function to check for wraparound).
v3/lib/lar.c:find_file() has the following for loop:
> char *walk;
> [...]
> for (walk = archive->start;
> (walk < (char *)(archive->start + archive->len - sizeof(struct
> lar_header))) &&
> (walk >= (char *)archive->start); walk += 16) {
AFAICS the check (walk >= (char *)archive->start) can be optimized away.
Pointers to other possible affected code would be appreciated.
Statements about the correctness of that for loop would be appreciated
as well.
Regards,
Carl-Daniel
work with a customer-owned LB system today that is using LAB made me
realize -- a very useful thing to have would be a grub2 that ran under
linux, parsed the standard grub.cfg, and kexec'ed.
Really, really needed. Not because LAB is inflexible, but because most
sysadmins are.
ron