profile/ivi/syslinux.git
14 years agosetadv: asprintf() doesn't return an error code
H. Peter Anvin [Thu, 10 Jun 2010 01:28:34 +0000 (18:28 -0700)]
setadv: asprintf() doesn't return an error code

asprintf() doesn't return an error number, but a count of characters
(which we don't need.)

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agosyslxopt: add missing break;
H. Peter Anvin [Thu, 10 Jun 2010 01:27:53 +0000 (18:27 -0700)]
syslxopt: add missing break;

Add missing break; in the command-line parser.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoloadconfig: fold FAT pathname search into generic_load_config
H. Peter Anvin [Wed, 9 Jun 2010 18:07:19 +0000 (11:07 -0700)]
loadconfig: fold FAT pathname search into generic_load_config

Fold the FAT pathname searching into generic_load_config; make it a
simple set of loops.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agosyslinux: provide for backwards compatibility
H. Peter Anvin [Thu, 10 Jun 2010 00:45:47 +0000 (17:45 -0700)]
syslinux: provide for backwards compatibility

Provide for backwards command-line compatibility, together with
warning messages where appropriate.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agosyslinux: don't break -o just yet; print warning and resume
H. Peter Anvin [Thu, 10 Jun 2010 00:20:10 +0000 (17:20 -0700)]
syslinux: don't break -o just yet; print warning and resume

-o has been supported for a very long time; recommend using -t
instead, but for now proceed with an error message.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoUnify instances of integer log2
H. Peter Anvin [Wed, 9 Jun 2010 18:21:52 +0000 (11:21 -0700)]
Unify instances of integer log2

Unify multiple open-coded instances of integer binary logarithm.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoClean up warnings in previous checkin
H. Peter Anvin [Wed, 9 Jun 2010 18:20:41 +0000 (11:20 -0700)]
Clean up warnings in previous checkin

Clean up warnings in checkin
146c34a2 Make syslinux installer real "pathbased"

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoMake syslinux installer real "pathbased"
Alek Du [Wed, 9 Jun 2010 05:28:33 +0000 (22:28 -0700)]
Make syslinux installer real "pathbased"

The previous merging extlinux/syslinux patch is not sufficient, the
syslinux is not real "patchbased", hence:

* fill currentdir in patcharea
* search "extlinux.conf" and "syslinux.cfg" for generic_load_config
* define boot_image and boot_sector macro for ldsyslinux
  this could make further merging extlinux and syslinux possible

Signed-off-by: Alek Du <alek.du@intel.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agomboot.c32: don't rely on possibly incomplete fields
H. Peter Anvin [Tue, 8 Jun 2010 21:37:14 +0000 (14:37 -0700)]
mboot.c32: don't rely on possibly incomplete fields

Since load_end_addr and bss_end_addr may be incomplete, we can't
actually use them for information.  An alternative would be to
actually modify these fields in-memory, but it is unclear if that is
permitted by the spec.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agomboot.c32: handle non-self-terminating images
H. Peter Anvin [Tue, 8 Jun 2010 17:58:48 +0000 (10:58 -0700)]
mboot.c32: handle non-self-terminating images

Handle non-ELF Multiboot images with 0 entries for data and/or bss
end, which means "until end of the file" and "nothing" for the two fields,
respectively.

Reported-and-tested-by: Thomas Bächler <thomas@archlinux.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agomenu docs: document MENU SAVE caveat with RAID-1
Ferenc Wagner [Mon, 31 May 2010 16:40:24 +0000 (18:40 +0200)]
menu docs: document MENU SAVE caveat with RAID-1

The usual RAID-1 ADV constraint...

Signed-off-by: Ferenc Wagner <wferi@niif.hu>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoextlinux: add a --menu-save option
Thomas Bächler [Mon, 7 Jun 2010 19:23:46 +0000 (21:23 +0200)]
extlinux: add a --menu-save option

Add the --menu-save option to set the MENU SAVE value from the running
system using extlinux.

From: Thomas Bächler <thomas@archlinux.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoextlinux docs: recommend the --raid option
Ferenc Wagner [Mon, 31 May 2010 15:14:41 +0000 (17:14 +0200)]
extlinux docs: recommend the --raid option

When booting from RAID, recommend --raid.

Signed-off-by: Ferenc Wagner <wferi@niif.hu>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agodocs: explain the danger of IPAPPEND 1
Ferenc Wagner [Mon, 31 May 2010 16:53:20 +0000 (18:53 +0200)]
docs: explain the danger of IPAPPEND 1

Explain what the actual issue with IPAPPEND 1 is, and what the
constraints on using it really are.

Signed-off-by: Ferenc Wagner <wferi@niif.hu>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agochain.c32: add support for loading GRUB stage2
Paul Bolle [Fri, 4 Jun 2010 12:14:49 +0000 (14:14 +0200)]
chain.c32: add support for loading GRUB stage2

This adds (basic) support for GRUB stage2 image files.

Loading a stage2 image will probably give you a GRUB prompt, with GRUB's
"root" set at "(hd0)" (ie, entire first disk). Maybe the "root" will
differ in less common setups. (One can of course select another disk
and/or partition with GRUB's "root" command.)

This has only been tested with version 3.2 stage2 images (as used by
GRUB 0.97). I'm not familiair with differences with other versions.

GRUB's loading mechanism allows to somehow provide stage2 with the
selected disk and partition, BSD slice, etc. (ie, to tell stage2 what
it's "root" is). I don't yet understand the notation used in that
mechanism. Besides, since stage2 images will not necessarily be loaded
from the disk (and partition, etc.) they were installed to, it seems
best to just use the first disk.

GRUB stage1_5 image files load quite similarly. However, for some
reason, a short test only got those images to print an error ("Error
17"). This could be related to the partition info these images are
provided with when they're loaded. I have never used stage1_5 images,
and do not know how to properly use and configure those, so my test
stopped there, and stage1_5 images are not supported.

Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoMENU IMMEDIATE: hotkeys which do not require Enter syslinux-4.00-pre47
H. Peter Anvin [Fri, 28 May 2010 01:22:21 +0000 (18:22 -0700)]
MENU IMMEDIATE: hotkeys which do not require Enter

The normal behavior for a hotkey is to jump to a specific menu entry.
With MENU IMMEDIATE, it activates the menu entry as well.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoMerge remote branch 'origin/master' into pathbased syslinux-4.00-pre46
H. Peter Anvin [Fri, 21 May 2010 03:36:16 +0000 (20:36 -0700)]
Merge remote branch 'origin/master' into pathbased

14 years agoisohybrid: use getopt_long_only()
H. Peter Anvin [Fri, 21 May 2010 03:35:24 +0000 (20:35 -0700)]
isohybrid: use getopt_long_only()

For compatibility with isohybrid.pl, use getopt_long_only() so long
options are accepted with a single dash.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agoMerge remote branch 'origin/master' into pathbased
H. Peter Anvin [Fri, 21 May 2010 03:32:21 +0000 (20:32 -0700)]
Merge remote branch 'origin/master' into pathbased

Resolved Conflicts:
com32/modules/cat.c

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agoisohybrid: create a C version
P. J. Pandit [Fri, 21 May 2010 03:29:52 +0000 (20:29 -0700)]
isohybrid: create a C version

Some Linux distributions have complained about Perl as a prerequisite
for isohybrid, so create a C version.

[ hpa: modified the array generator to put it all in one script ]

14 years agocat.c32: handle multiple files, use argv[0], copy 4K at a time
H. Peter Anvin [Fri, 21 May 2010 03:12:44 +0000 (20:12 -0700)]
cat.c32: handle multiple files, use argv[0], copy 4K at a time

Loop over multiple files, use argv[0] for the program name, and copy
4K at a time.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agounify common parts of extlinux and syslinux installer
Alek Du [Wed, 19 May 2010 01:39:57 +0000 (09:39 +0800)]
unify common parts of extlinux and syslinux installer

Thus we can share same command line options and reduce a lot of dup
code...

Seems like a big patch, but the changes are quite safe, no much logical
change.

Signed-off-by: Alek Du <alek.du@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agosyslinux: fix sector arrays
Alek Du [Thu, 20 May 2010 10:24:35 +0000 (18:24 +0800)]
syslinux: fix sector arrays

The first sector ptr is in bs->NextSector not in the array in
the patch_area. And actually the ADV sectors counting is wrong for a
while ...

Signed-off-by: Alek Du <alek.du@intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agoRestore skip CR and LF in the comboot API copies of the banner strings
H. Peter Anvin [Mon, 17 May 2010 21:14:21 +0000 (14:14 -0700)]
Restore skip CR and LF in the comboot API copies of the banner strings

The comboot API should not report leading CR and LF in the banner
strings.  This was done in 3.86, but apparently was lost due to
mismerge (52c9d5ba2af4e1f77aa2486a2af2bcab0b0af52c) in the 4.00 branch.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agocore: fix "sector size" confusion syslinux-4.00-pre45
H. Peter Anvin [Fri, 14 May 2010 22:22:00 +0000 (15:22 -0700)]
core: fix "sector size" confusion

Fix the case where the "sector size" used by the pm filesystem driver
isn't the same thing as the SECTOR_SIZE/SECTOR_SHIFT macros used in
the assembly code.  This is a per-device property, and in the
particular case of isolinux hybrid, they are not even currently the
same (for all others, they are the same for now, but not necessarily
in the future.)

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agodiskstart: print CHS/EDD instead of CBIOS/EBIOS
H. Peter Anvin [Fri, 14 May 2010 22:20:45 +0000 (15:20 -0700)]
diskstart: print CHS/EDD instead of CBIOS/EBIOS

The terms CHS or EDD are more well-known, so use them.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agocomboot: fix comapi_open
Sebastian Herbszt [Sat, 1 May 2010 13:28:49 +0000 (15:28 +0200)]
comboot: fix comapi_open

Fix breakage introduced by commit e375515ddc712f1f69ee21337db2a3267caa5d49
(Add 32-bit versions of open file/close file).

Signed-off-by: Sebastian Herbszt <herbszt@gmx.de>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agoFAT: need to track the current position past discontinuities syslinux-4.00-pre44
H. Peter Anvin [Thu, 13 May 2010 22:40:35 +0000 (15:40 -0700)]
FAT: need to track the current position past discontinuities

When we have a discontiguous file (broken extent), then we need to
keep track of where the next extent starts, not where the current
extent ends.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agodiskio: compare to the proper pointer
H. Peter Anvin [Thu, 13 May 2010 22:39:29 +0000 (15:39 -0700)]
diskio: compare to the proper pointer

We need to compare against the advanced pointer, not the original one.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agofat: fix confusion between byte and sector counts
H. Peter Anvin [Thu, 13 May 2010 21:38:04 +0000 (14:38 -0700)]
fat: fix confusion between byte and sector counts

clust_shift is in units of sectors, not in bytes; this was mixed up in
the extent finder.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agodiskio: make maxtransfer per-device, cap to 127, imported from head syslinux-4.00-pre43
H. Peter Anvin [Thu, 13 May 2010 04:23:13 +0000 (21:23 -0700)]
diskio: make maxtransfer per-device, cap to 127, imported from head

Make the maxtransfer per device, as it should be; properly imported
from the head loader (in case it is patched with -s).  Also enforce
capping to 127 for EBIOS and 63 for CBIOS.  This is structured so that
once EDD4 is approved we can remove the capping for that particular
subcase.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agosyslinux.ld: don't pad between .got and .data
H. Peter Anvin [Thu, 13 May 2010 04:22:14 +0000 (21:22 -0700)]
syslinux.ld: don't pad between .got and .data

Adding padding between .got and .data is unnecessary, and can
cause spurious decompression failures when .data is empty (which it
almost is already...)

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agodiskio: sanitize the reduced transfer sizes
H. Peter Anvin [Thu, 13 May 2010 00:28:51 +0000 (17:28 -0700)]
diskio: sanitize the reduced transfer sizes

Simply shift the size left by 1 when computing the transfer sizes.
This will always end with the values ..., 1, 0 as it should.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agocmenu: remove obsolete Makefile rules syslinux-4.00-pre42
H. Peter Anvin [Wed, 12 May 2010 23:14:19 +0000 (16:14 -0700)]
cmenu: remove obsolete Makefile rules

Remove Makefile rules which were both redundant (../MCONFIG contains
the rules in a central manner) and just plain wrong.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoMCONFIG: add MCONFIG.devel for use from git, removed in tarballs
H. Peter Anvin [Wed, 12 May 2010 23:09:01 +0000 (16:09 -0700)]
MCONFIG: add MCONFIG.devel for use from git, removed in tarballs

Add an MCONFIG.devel file which is present in git, but removed in
tarballs.  This lets us set -Werror there and still not break users
who are using a different compiler.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoinstallers: fix warnings
H. Peter Anvin [Wed, 12 May 2010 23:06:10 +0000 (16:06 -0700)]
installers: fix warnings

CLean up warnings in the installers.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agohdt-cli.c: clean up unused warnings
H. Peter Anvin [Wed, 12 May 2010 23:01:54 +0000 (16:01 -0700)]
hdt-cli.c: clean up unused warnings

-Werror cleanup.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoadv_menu.tpl: fix type of login_handler()
H. Peter Anvin [Wed, 12 May 2010 22:59:29 +0000 (15:59 -0700)]
adv_menu.tpl: fix type of login_handler()

Fix the type of the login_handler() function.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agocomplex.c: fix warnings
H. Peter Anvin [Wed, 12 May 2010 22:53:35 +0000 (15:53 -0700)]
complex.c: fix warnings

Clean up warnings, even though this is sample code.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agolibmenu/des.c: fix signedness error
H. Peter Anvin [Wed, 12 May 2010 22:52:00 +0000 (15:52 -0700)]
libmenu/des.c: fix signedness error

Fix a signedness error, which in turn was caused by needlessly
complicated code.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agolibmenu/passwords.h: function declaration isn't a prototype
H. Peter Anvin [Wed, 12 May 2010 22:49:30 +0000 (15:49 -0700)]
libmenu/passwords.h: function declaration isn't a prototype

Make it a prototype

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agolibmenu/syslnx.c: mark ipappend used
H. Peter Anvin [Wed, 12 May 2010 22:48:43 +0000 (15:48 -0700)]
libmenu/syslnx.c: mark ipappend used

Is this really correct?!

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agorosh.c: mark argument used
H. Peter Anvin [Wed, 12 May 2010 22:47:06 +0000 (15:47 -0700)]
rosh.c: mark argument used

-Werror cleanup.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoreadconfig.c: avoid uninitialized variable warnings
H. Peter Anvin [Wed, 12 May 2010 22:46:21 +0000 (15:46 -0700)]
readconfig.c: avoid uninitialized variable warnings

-Werror cleanup.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agocat.c32: remove unused variable
H. Peter Anvin [Wed, 12 May 2010 22:44:44 +0000 (15:44 -0700)]
cat.c32: remove unused variable

Remove unused variable as part of -Werror cleanup.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agovpdtest.c32: remove unused variable
H. Peter Anvin [Wed, 12 May 2010 22:44:12 +0000 (15:44 -0700)]
vpdtest.c32: remove unused variable

Remove unused variable as part of -Werror cleanup.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agosdi.c32: remove unused variable
H. Peter Anvin [Wed, 12 May 2010 22:43:30 +0000 (15:43 -0700)]
sdi.c32: remove unused variable

Remove unused variable as part of -Werror cleanup.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agopcitest.c32: mark argc/argv used
H. Peter Anvin [Wed, 12 May 2010 22:42:50 +0000 (15:42 -0700)]
pcitest.c32: mark argc/argv used

Part of -Werror cleanup.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agodisk.c32: remove unused variable
H. Peter Anvin [Wed, 12 May 2010 22:42:03 +0000 (15:42 -0700)]
disk.c32: remove unused variable

Remove unused variable as part of -Werror cleanup.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agosha256/512: fix signedness errors
H. Peter Anvin [Wed, 12 May 2010 22:40:46 +0000 (15:40 -0700)]
sha256/512: fix signedness errors

Fix signedness errors as part of -Werror cleanup.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoget_key.c: fix signedness error
H. Peter Anvin [Wed, 12 May 2010 22:36:49 +0000 (15:36 -0700)]
get_key.c: fix signedness error

Fix signedness error, part of -Werror cleanup.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agorunimage.c: remove unused variable
H. Peter Anvin [Wed, 12 May 2010 22:35:19 +0000 (15:35 -0700)]
runimage.c: remove unused variable

-Werror cleanup: remove unused variable.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoMerge branch 'master' into pathbased
H. Peter Anvin [Wed, 12 May 2010 21:57:39 +0000 (14:57 -0700)]
Merge branch 'master' into pathbased

14 years agoCMENU: Fixing login_handler prototype
Erwan Velu [Wed, 12 May 2010 21:21:25 +0000 (23:21 +0200)]
CMENU: Fixing login_handler prototype

login_handler have to follow the (t_menusystem * ms, t_menuitem * mi)
prototype defined by t_handler_return

14 years agocom32: Fixing cast in argv to avoid warning
Erwan Velu [Wed, 12 May 2010 20:54:03 +0000 (22:54 +0200)]
com32: Fixing cast in argv to avoid warning

const char * confuse the *argv++ assignation
Casting to char * removes this warning

14 years agohdt: casting to avoid useless warning
Erwan Velu [Wed, 12 May 2010 19:49:10 +0000 (21:49 +0200)]
hdt: casting to avoid useless warning

14 years agohdt: Fixing print_history() prototype
Erwan Velu [Wed, 12 May 2010 19:46:59 +0000 (21:46 +0200)]
hdt: Fixing print_history() prototype

print_history have to follow the exec prototype

14 years agocpuid: Typo between argv & argc
Erwan Velu [Wed, 12 May 2010 19:18:09 +0000 (21:18 +0200)]
cpuid: Typo between argv & argc

argv isn't a number ;)

14 years agocmd: Adding proper include to syslinux_run_command
Erwan Velu [Wed, 12 May 2010 19:12:34 +0000 (21:12 +0200)]
cmd: Adding proper include to syslinux_run_command

14 years agodisk: Fixing get_error() call
Erwan Velu [Wed, 12 May 2010 19:10:47 +0000 (21:10 +0200)]
disk: Fixing get_error() call

get_error prototype changed since this module was written

14 years agoMerge branch 'master' into pathbased
H. Peter Anvin [Tue, 11 May 2010 23:10:06 +0000 (16:10 -0700)]
Merge branch 'master' into pathbased

Resolved Conflicts:
com32/modules/Makefile

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agocat.c32: move from samples to modules; remove debug code
H. Peter Anvin [Tue, 11 May 2010 23:08:20 +0000 (16:08 -0700)]
cat.c32: move from samples to modules; remove debug code

Move cat.c32 from samples to modules, and remove debugging messages.
Real users may want to use this.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agoMerge branch 'master' into pathbased
H. Peter Anvin [Tue, 11 May 2010 23:07:07 +0000 (16:07 -0700)]
Merge branch 'master' into pathbased

14 years agocat.c32: check if filename exists
Gert Hulselmans [Tue, 4 May 2010 00:07:43 +0000 (17:07 -0700)]
cat.c32: check if filename exists

Don't crash if the file doesn't exist.

Signed-off-by: Gert Hulselmans <gerth@zytor.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agoMerge branch 'master' into pathbased
H. Peter Anvin [Tue, 11 May 2010 21:39:07 +0000 (14:39 -0700)]
Merge branch 'master' into pathbased

14 years agoMerge remote branch 'hdt-pierre/fixes-for-3.86'
H. Peter Anvin [Tue, 11 May 2010 21:38:57 +0000 (14:38 -0700)]
Merge remote branch 'hdt-pierre/fixes-for-3.86'

14 years agoMerge branch 'master' into pathbased
H. Peter Anvin [Tue, 11 May 2010 20:55:32 +0000 (13:55 -0700)]
Merge branch 'master' into pathbased

14 years agoMerge remote branch 'erwan/master'
H. Peter Anvin [Tue, 11 May 2010 20:55:19 +0000 (13:55 -0700)]
Merge remote branch 'erwan/master'

14 years agoVPD: Fixing base_address
Erwan Velu [Tue, 11 May 2010 20:52:27 +0000 (22:52 +0200)]
VPD: Fixing base_address

This commit now save properly VPD's address.

14 years agocpuid.c32: print unsigned values syslinux-4.00-pre41
H. Peter Anvin [Tue, 11 May 2010 20:00:17 +0000 (13:00 -0700)]
cpuid.c32: print unsigned values

Print unsigned values; formatting was wrong for unsigned values
anyway.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agosysdump: use <sys/cpu.h>
H. Peter Anvin [Tue, 11 May 2010 19:59:56 +0000 (12:59 -0700)]
sysdump: use <sys/cpu.h>

Use <sys/cpu.h> where it makes sense.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agocom32: Unbreak <sys/cpu.h>
H. Peter Anvin [Tue, 11 May 2010 19:59:06 +0000 (12:59 -0700)]
com32: Unbreak <sys/cpu.h>

CPUID functions were apparently broken when made PIC-safe; clean up
and fix.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agoMerge branch 'master' into pathbased
H. Peter Anvin [Tue, 11 May 2010 19:28:28 +0000 (12:28 -0700)]
Merge branch 'master' into pathbased

14 years agoMerge remote branch 'erwan/master'
H. Peter Anvin [Tue, 11 May 2010 19:12:01 +0000 (12:12 -0700)]
Merge remote branch 'erwan/master'

14 years agocore: remove "Hello, World!" debugging message
H. Peter Anvin [Mon, 10 May 2010 19:45:24 +0000 (12:45 -0700)]
core: remove "Hello, World!" debugging message

Remove "Hello, World!" debugging message -- no longer relevant.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoDMI: Increasing processor buffer size
Erwan Velu [Thu, 6 May 2010 19:24:35 +0000 (21:24 +0200)]
DMI: Increasing processor buffer size

Impact: avoid string truncation

The version field could be bigger than 32 char.

14 years agoDMI: Fixing memory corruption
Erwan Velu [Thu, 6 May 2010 19:23:22 +0000 (21:23 +0200)]
DMI: Fixing memory corruption

Impact: avoid memory corruption

Too much usage of strcpy & sprintf was leading to serious memory
corruption. Using snprintf & strncpy is prefered.

14 years agoDMI: removing useless variable
Erwan Velu [Thu, 6 May 2010 19:22:46 +0000 (21:22 +0200)]
DMI: removing useless variable

Impact: none

Removing useless variable

14 years agomboot.c32: autodetect Solaris
H. Peter Anvin [Mon, 3 May 2010 22:11:10 +0000 (15:11 -0700)]
mboot.c32: autodetect Solaris

Autodetect Solaris kernels (based on the ELF header OSABI field) and
use the Solaris workarounds in that case.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoMerge remote branch 'origin/master' into pathbased syslinux-4.00-pre40
H. Peter Anvin [Thu, 29 Apr 2010 04:56:25 +0000 (21:56 -0700)]
Merge remote branch 'origin/master' into pathbased

14 years agoextlinux: handle cases of a single level of directories syslinux-4.00-pre39
H. Peter Anvin [Wed, 28 Apr 2010 22:50:20 +0000 (15:50 -0700)]
extlinux: handle cases of a single level of directories

We had a boundary condition error where a single-level directory from
the global root (e.g. /boot) would be incorrectly truncated; fix that.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agogfxboot: fix menu label handling
Steffen Winterfeldt [Mon, 19 Apr 2010 13:17:57 +0000 (15:17 +0200)]
gfxboot: fix menu label handling

Signed-off-by: Steffen Winterfeldt <wfeldt@opensuse.org>
Signed-off-by: Sebastian Herbszt <herbszt@gmx.de>
14 years agogfxboot: better error handling
Steffen Winterfeldt [Fri, 16 Apr 2010 11:18:46 +0000 (13:18 +0200)]
gfxboot: better error handling

Signed-off-by: Steffen Winterfeldt <wfeldt@opensuse.org>
Signed-off-by: Sebastian Herbszt <herbszt@gmx.de>
14 years agogfxboot: fix localboot
Steffen Winterfeldt [Wed, 14 Apr 2010 15:43:16 +0000 (17:43 +0200)]
gfxboot: fix localboot

Signed-off-by: Steffen Winterfeldt <wfeldt@opensuse.org>
Signed-off-by: Sebastian Herbszt <herbszt@gmx.de>
14 years agogfxboot: turn off graphics before starting kernel
Steffen Winterfeldt [Tue, 13 Apr 2010 15:42:44 +0000 (17:42 +0200)]
gfxboot: turn off graphics before starting kernel

Signed-off-by: Steffen Winterfeldt <wfeldt@opensuse.org>
Signed-off-by: Sebastian Herbszt <herbszt@gmx.de>
14 years agogfxboot: handle IPAPPEND
Steffen Winterfeldt [Tue, 13 Apr 2010 15:42:12 +0000 (17:42 +0200)]
gfxboot: handle IPAPPEND

Signed-off-by: Steffen Winterfeldt <wfeldt@opensuse.org>
Signed-off-by: Sebastian Herbszt <herbszt@gmx.de>
14 years agogfxboot: make config statements case-insensitive
Steffen Winterfeldt [Tue, 13 Apr 2010 15:39:46 +0000 (17:39 +0200)]
gfxboot: make config statements case-insensitive

Signed-off-by: Steffen Winterfeldt <wfeldt@opensuse.org>
Signed-off-by: Sebastian Herbszt <herbszt@gmx.de>
14 years agogfxboot: support MENU LABEL statement
Steffen Winterfeldt [Tue, 13 Apr 2010 15:38:39 +0000 (17:38 +0200)]
gfxboot: support MENU LABEL statement

Signed-off-by: Steffen Winterfeldt <wfeldt@opensuse.org>
Signed-off-by: Sebastian Herbszt <herbszt@gmx.de>
14 years agochain.c32 sethidden option fix
Michal Soltys [Wed, 7 Apr 2010 08:37:25 +0000 (10:37 +0200)]
chain.c32 sethidden option fix

Hidden sectors value was applied at the wrong offset.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agocom32/tools: nothing to install here
H. Peter Anvin [Tue, 6 Apr 2010 23:29:03 +0000 (16:29 -0700)]
com32/tools: nothing to install here

This directory contains builttime tools, so they don't need to be
installed anywhere.

This fixes "make install-all".

Reported-by: Dag Wieers <dag@wieers.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agocore: fs: verify call to fs_ops->readdir
Sebastian Herbszt [Sun, 7 Mar 2010 15:02:50 +0000 (16:02 +0100)]
core: fs: verify call to fs_ops->readdir

Check if fs_ops->readdir is available before calling it.
At least PXELINUX doesn't implement it.

Signed-off-by: Sebastian Herbszt <herbszt@gmx.de>
LKML-Reference: <1267974170$3058@local>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoMerge commit 'syslinux-3.86' into pathbased syslinux-4.00-pre38
H. Peter Anvin [Thu, 1 Apr 2010 18:14:56 +0000 (11:14 -0700)]
Merge commit 'syslinux-3.86' into pathbased

Resolved Conflicts:
core/cpuinit.inc

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agocore: make sure we observe EFLAGS_ID only syslinux-3.86
H. Peter Anvin [Wed, 31 Mar 2010 16:24:25 +0000 (09:24 -0700)]
core: make sure we observe EFLAGS_ID only

When looking for CPUID, make sure we only look at EFLAGS_ID and no
other flags that could possibly have changed for extrinsic reasons.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agocpuid.c32: output the correct register names
H. Peter Anvin [Wed, 31 Mar 2010 15:56:12 +0000 (08:56 -0700)]
cpuid.c32: output the correct register names

Output the correct register names in cpuid.c32; fix copy-and-paste
error.

Reported-by: Andrew Jones <drjones@redhat.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agoNEWS: document mBFT fix syslinux-3.86-pre4
H. Peter Anvin [Wed, 31 Mar 2010 03:16:14 +0000 (20:16 -0700)]
NEWS: document mBFT fix

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agomemdisk: Calculate mBFT checksum later
Shao Miller [Tue, 9 Mar 2010 06:12:37 +0000 (01:12 -0500)]
memdisk: Calculate mBFT checksum later

We were calculating the mBFT checksum too early; dpt_ptr could
change after checksumming, so we now do it after that possibility.

Signed-off-by: Shao Miller <shao.miller@yrdsb.edu.on.ca>
14 years agocore: remove stray debugging statement syslinux-3.86-pre3
H. Peter Anvin [Wed, 31 Mar 2010 00:20:23 +0000 (17:20 -0700)]
core: remove stray debugging statement

Remove a stray debugging printout when running on Xen.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
14 years agocore: move __jiffies and _IdleTimer to .data16 syslinux-4.00-pre37
H. Peter Anvin [Tue, 30 Mar 2010 23:31:47 +0000 (16:31 -0700)]
core: move __jiffies and _IdleTimer to .data16

Move the __jiffies and _IdleTimer variables to the .data16 segment.
This ensures that if do_idle is invoked very early that it will not
jump off into a non-existent idle routine.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agoMerge branch 'master' into pathbased
H. Peter Anvin [Tue, 30 Mar 2010 23:17:59 +0000 (16:17 -0700)]
Merge branch 'master' into pathbased

Resolved Conflicts:
com32/modules/Makefile
core/comboot.inc
core/cpuinit.inc
core/idle.inc

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
14 years agocore: move idle handling into protected mode
H. Peter Anvin [Tue, 30 Mar 2010 23:06:39 +0000 (16:06 -0700)]
core: move idle handling into protected mode

Do the actual idling in protected mode.  This both allows PM code a
more efficient interface, but also handles bugs in HVM implementations
which don't handle HLT in real mode.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>