profile/ivi/syslinux.git
12 years agoextlinux: main.c should not be executable
H. Peter Anvin [Wed, 8 Feb 2012 17:57:27 +0000 (09:57 -0800)]
extlinux: main.c should not be executable

Remove stray x bit.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
12 years agobootsect.inc: try to catch obvious .0 problems
H. Peter Anvin [Mon, 6 Feb 2012 20:12:26 +0000 (12:12 -0800)]
bootsect.inc: try to catch obvious .0 problems

A typical Linux kernel will be way, way too big for the low memory
bootstrap area, so throw an error message if we end up in the
bootstrap path with something that cannot possibly fit below 640K.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
12 years agostrreplace: additional cleanups
H. Peter Anvin [Mon, 6 Feb 2012 19:14:39 +0000 (11:14 -0800)]
strreplace: additional cleanups

Additional formatting cleanups.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
12 years agostrreplace: clean up and simplify
H. Peter Anvin [Mon, 6 Feb 2012 19:08:45 +0000 (11:08 -0800)]
strreplace: clean up and simplify

Clean up and simplify strreplace(); in particular there is no need to
call strlen() repeatedly on the same strings.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
12 years agohdt: Adding postexec= option
Erwan Velu [Sat, 17 Dec 2011 21:27:07 +0000 (22:27 +0100)]
hdt: Adding postexec= option

When HDT is exiting, you might need executing something else.
This could be used in the following scenario :

You start HDT, do an automatic command like 'dump; exit', but then after
you might need to launch something else from syslinux.

The postexec option will allow you to define what label you'd love
running one HDT got terminated.

Syntaxt is like the following:

postexec='menu_label_to_run_once_hdt_got_exited'

Note the quotes (') after the equal sign (=)

This could looks like :
APPEND auto='dump; exit' postexec='memtest'

12 years agohdt: On the road to 0.5.2
Erwan Velu [Sat, 17 Dec 2011 16:02:43 +0000 (17:02 +0100)]
hdt: On the road to 0.5.2

12 years agoMerge remote-tracking branch 'hpa/master'
Erwan Velu [Sat, 17 Dec 2011 15:23:37 +0000 (16:23 +0100)]
Merge remote-tracking branch 'hpa/master'

12 years agoClarify the license for the Syslinux logo syslinux-4.06-pre1
H. Peter Anvin [Wed, 14 Dec 2011 18:46:23 +0000 (10:46 -0800)]
Clarify the license for the Syslinux logo

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
12 years agoversion: now working on version 4.06
H. Peter Anvin [Wed, 14 Dec 2011 18:39:29 +0000 (10:39 -0800)]
version: now working on version 4.06

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
12 years agoNEWS: document NTFS support
H. Peter Anvin [Wed, 14 Dec 2011 18:39:10 +0000 (10:39 -0800)]
NEWS: document NTFS support

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
12 years agofind-mingw64: yet another set of prefixes to test...
H. Peter Anvin [Tue, 13 Dec 2011 23:52:28 +0000 (15:52 -0800)]
find-mingw64: yet another set of prefixes to test...

This seems to be the current set of prefixes used by the Fedora-cross
project.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
12 years agoMerge remote-tracking branch 'pcacjr/ntfs-for-hpa'
H. Peter Anvin [Tue, 13 Dec 2011 23:34:59 +0000 (15:34 -0800)]
Merge remote-tracking branch 'pcacjr/ntfs-for-hpa'

12 years agoAdd logo by Abi Rasheed syslinux-4.05
H. Peter Anvin [Fri, 9 Dec 2011 18:28:17 +0000 (10:28 -0800)]
Add logo by Abi Rasheed

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
12 years agoMakefile.private: remove obsolete target syslinux-4.05-pre7
H. Peter Anvin [Wed, 7 Dec 2011 18:03:23 +0000 (10:03 -0800)]
Makefile.private: remove obsolete target

Remove target made obsolete by upstream changes.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
12 years agoNEWS: document mk-lba-img removal
H. Peter Anvin [Tue, 6 Dec 2011 05:46:34 +0000 (21:46 -0800)]
NEWS: document mk-lba-img removal

12 years agodiag/geodsp: don't distribute mk-lba-img
H. Peter Anvin [Mon, 5 Dec 2011 23:31:46 +0000 (15:31 -0800)]
diag/geodsp: don't distribute mk-lba-img

Do not distribute the mk-lba-img host tool binary.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
12 years agoMerge remote-tracking branch 'hpa/master'
Erwan Velu [Wed, 2 Nov 2011 19:15:22 +0000 (20:15 +0100)]
Merge remote-tracking branch 'hpa/master'

12 years agohdt: Don't search HDDs when no HDD got detected
Erwan Velu [Thu, 22 Sep 2011 18:52:49 +0000 (20:52 +0200)]
hdt: Don't search HDDs when no HDD got detected

12 years agohdt: Avoid memory corruption on menu summary
Erwan Velu [Thu, 22 Sep 2011 20:44:53 +0000 (22:44 +0200)]
hdt: Avoid memory corruption on menu summary

During CPU's capabilites computation, let's sure the buffer we manage is
big enought....

12 years agohdt: Dumping disk failed when disk are present
Erwan Velu [Wed, 21 Sep 2011 19:44:47 +0000 (21:44 +0200)]
hdt: Dumping disk failed when disk are present

When physical disks were present, the computed items were not flushed
and so not present on the disk file.

Thanks to pscheie for reporting this.

12 years agontfs: keep a state structure for ntfs_readdir() callers
Paulo Alcantara [Tue, 20 Sep 2011 02:45:19 +0000 (02:45 +0000)]
ntfs: keep a state structure for ntfs_readdir() callers

Latetly, we kept a state information within the inode structure, that
was actually a mistake. Now, a ntfs_readdir_state structure is allocated
to keep a state that'll help on listing directory entries from sucessive
ntfs_readdir() calls.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: cleanups
Paulo Alcantara [Wed, 7 Sep 2011 22:02:56 +0000 (22:02 +0000)]
ntfs: cleanups

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: Check for disk read error during FS probing
Shao Miller [Sat, 3 Sep 2011 19:40:45 +0000 (15:40 -0400)]
ntfs: Check for disk read error during FS probing

If reading the NTFS BIOS parameter block fails for any reason,
we will not attempt to continue to work towards finding an
NTFS filesystem on the disk/partition.

Signed-off-by: Shao Miller <shao.miller@yrdsb.edu.on.ca>
Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: Deal with NTFS versions for MFT record lookups
Shao Miller [Thu, 1 Sep 2011 04:44:18 +0000 (04:44 +0000)]
ntfs: Deal with NTFS versions for MFT record lookups

The MFT record lookup strategies are different for NTFS
versions 3.0 and 3.1, so we use a function pointer for
the appropriate function.

We start off using 3.0 by default, then use that to read
the $Volume meta-file and find out the actual version.
Then we adjust the function pointer, as needed.

Signed-off-by: Shao Miller <shao.miller@yrdsb.edu.on.ca>
Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: use a runlist for handling multiple data runs
Paulo Alcantara [Wed, 7 Sep 2011 06:25:03 +0000 (06:25 +0000)]
ntfs: use a runlist for handling multiple data runs

NTFS files can contain multiple data runs, so use a runlist for handling
multiple data runs of a single file, and also modify ntfs_next_extent()
in order to support this new feature.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agowin: Fix an out-of-bounds array access
Shao Miller [Tue, 30 Aug 2011 00:21:09 +0000 (20:21 -0400)]
win: Fix an out-of-bounds array access

We should not attempt to make a map of more LDLINUX.SYS
sectors than we have.  Simply multiplying a cluster count
times the number of sectors per cluster does not account
for unused sectors in the last used cluster.

Signed-off-by: Shao Miller <shao.miller@yrdsb.edu.on.ca>
Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: don't start lookup from the same block in ntfs_mft_record_lookup()
Paulo Alcantara [Mon, 29 Aug 2011 05:49:32 +0000 (05:49 +0000)]
ntfs: don't start lookup from the same block in ntfs_mft_record_lookup()

ntfs_mft_record_lookup() always started its lookup from the same block,
which was though a mistake. So the lookup became very slow. Now
determine its starting block number from the calculated MFT record's LCN
number.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: fix ntfs_read()
Paulo Alcantara [Sun, 28 Aug 2011 21:40:46 +0000 (21:40 +0000)]
ntfs: fix ntfs_read()

lbytes variable wasn't set correctly.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: and more fixes...
Paulo Alcantara [Sat, 27 Aug 2011 04:31:34 +0000 (04:31 +0000)]
ntfs: and more fixes...

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agowin: Add further NTFS support to Windows installers
Shao Miller [Sun, 14 Aug 2011 23:41:24 +0000 (19:41 -0400)]
win: Add further NTFS support to Windows installers

We have a way of building a map of the LDLINUX.SYS sectors
on an NTFS filesystem, now.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
Signed-off-by: Shao Miller <shao.miller@yrdsb.edu.on.ca>
12 years agontfs: fix VCN calculation and the ntfs_readdir() function
Paulo Alcantara [Thu, 11 Aug 2011 03:46:06 +0000 (03:46 +0000)]
ntfs: fix VCN calculation and the ntfs_readdir() function

Now I'd recommend to test it :-)

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: more bugfixes
Paulo Alcantara [Tue, 9 Aug 2011 03:37:25 +0000 (03:37 +0000)]
ntfs: more bugfixes

Need more tests, hence I don't recommend to rely on it right now.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: fix ntfs_fs_init()
Paulo Alcantara [Thu, 4 Aug 2011 23:40:51 +0000 (23:40 +0000)]
ntfs: fix ntfs_fs_init()

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: I think ntfs_readdir() is now fixed
Paulo Alcantara [Tue, 2 Aug 2011 00:54:54 +0000 (00:54 +0000)]
ntfs: I think ntfs_readdir() is now fixed

gerth and sha0, you need to test it! :-)

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: fix ntfs_readdir()
Paulo Alcantara [Sun, 31 Jul 2011 21:00:13 +0000 (21:00 +0000)]
ntfs: fix ntfs_readdir()

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: mapping_chunk's length field must not be an uint8_t type
Paulo Alcantara [Fri, 29 Jul 2011 04:45:53 +0000 (04:45 +0000)]
ntfs: mapping_chunk's length field must not be an uint8_t type

Also change the other length field accordingly.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: better way to check whether a MFT record is file or directory
Paulo Alcantara [Fri, 29 Jul 2011 04:16:18 +0000 (04:16 +0000)]
ntfs: better way to check whether a MFT record is file or directory

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: and more speed improvements
Paulo Alcantara [Fri, 29 Jul 2011 04:03:29 +0000 (04:03 +0000)]
ntfs: and more speed improvements

gerth, you HAVE to test it! :-)

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: fix ntfs_match_longname() and make index_lookup() faster
Paulo Alcantara [Thu, 28 Jul 2011 22:56:31 +0000 (22:56 +0000)]
ntfs: fix ntfs_match_longname() and make index_lookup() faster

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: make index_inode_setup() faster
Paulo Alcantara [Thu, 28 Jul 2011 22:19:40 +0000 (22:19 +0000)]
ntfs: make index_inode_setup() faster

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: make ntfs_match_longname() faster
Paulo Alcantara [Thu, 28 Jul 2011 21:35:07 +0000 (21:35 +0000)]
ntfs: make ntfs_match_longname() faster

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: more cleanups
Paulo Alcantara [Thu, 28 Jul 2011 21:20:16 +0000 (21:20 +0000)]
ntfs: more cleanups

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: cleanups
Paulo Alcantara [Thu, 28 Jul 2011 03:49:34 +0000 (03:49 +0000)]
ntfs: cleanups

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: fix bug in parse_data_run()
Paulo Alcantara [Thu, 28 Jul 2011 00:11:28 +0000 (00:11 +0000)]
ntfs: fix bug in parse_data_run()

parse_data_run() used to get the wrong value of VCNs starting at LCN,
so it was only getting the value 1 and the NTFS driver didn't look at
the other VCNs.

Now, everything seems to work, except the readdir() function :-P

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: it's working, partially... :-/
Paulo Alcantara [Wed, 27 Jul 2011 01:35:36 +0000 (01:35 +0000)]
ntfs: it's working, partially... :-/

My tests were: menu.c32 (worked), cat.32 (worked, ls.c32 (not worked).
Looks like we have a bug in ntfs_readdir(). It will check it out later.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: continue on its implementation
Paulo Alcantara [Tue, 26 Jul 2011 04:19:51 +0000 (04:19 +0000)]
ntfs: continue on its implementation

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: heh :-)
Paulo Alcantara [Mon, 25 Jul 2011 00:24:24 +0000 (00:24 +0000)]
ntfs: heh :-)

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: set block shift and size to their correct values
Paulo Alcantara [Sun, 24 Jul 2011 20:36:25 +0000 (20:36 +0000)]
ntfs: set block shift and size to their correct values

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: implement fixups_realloc()
Paulo Alcantara [Sun, 24 Jul 2011 08:04:58 +0000 (08:04 +0000)]
ntfs: implement fixups_realloc()

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: implement ntfs_next_extent() and ntfs_getfssec()
Paulo Alcantara [Sat, 23 Jul 2011 16:52:20 +0000 (16:52 +0000)]
ntfs: implement ntfs_next_extent() and ntfs_getfssec()

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: implement ntfs_readdir()
Paulo Alcantara [Wed, 20 Jul 2011 23:27:13 +0000 (23:27 +0000)]
ntfs: implement ntfs_readdir()

Read one directory entry at one time.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: implement get_inode_mode()
Paulo Alcantara [Wed, 20 Jul 2011 01:59:23 +0000 (01:59 +0000)]
ntfs: implement get_inode_mode()

Determine the mode of an inode from a given file record.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: implement ntfs_iget()
Paulo Alcantara [Mon, 11 Jul 2011 17:17:26 +0000 (17:17 +0000)]
ntfs: implement ntfs_iget()

Return an allocated inode of a given UTF-16LE LFN passed as 2nd
argument.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: implement ntfs_iget_root()
Paulo Alcantara [Fri, 8 Jul 2011 01:59:33 +0000 (01:59 +0000)]
ntfs: implement ntfs_iget_root()

Return an allocated inode of the respective root directory along with
its information.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: implement for_each_mft_record() macro
Paulo Alcantara [Fri, 8 Jul 2011 00:57:26 +0000 (00:57 +0000)]
ntfs: implement for_each_mft_record() macro

Walk through the MFT records from a given block number. Also add more
NTFS-specific structures that'll be used in turn.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agontfs: implement new_ntfs_inode()
Paulo Alcantara [Thu, 7 Jul 2011 22:17:10 +0000 (22:17 +0000)]
ntfs: implement new_ntfs_inode()

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agoInitial NTFS readonly support code to Syslinux
Paulo Alcantara [Thu, 7 Jul 2011 21:28:05 +0000 (21:28 +0000)]
Initial NTFS readonly support code to Syslinux

Basically add the NTFS filesystem structures and also implement
ntfs_fs_init() to only initialize the filesystem metadata and return the
block size in bits.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agoAdd NTFS filesystem support to Linux and Windows installers
Paulo Alcantara [Tue, 5 Jul 2011 21:32:51 +0000 (21:32 +0000)]
Add NTFS filesystem support to Linux and Windows installers

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
12 years agoifmemdsk: remove spurious +x bit syslinux-4.05-pre6
H. Peter Anvin [Thu, 25 Aug 2011 17:58:44 +0000 (10:58 -0700)]
ifmemdsk: remove spurious +x bit

This is not an executable...

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
12 years agoMerge remote-tracking branch 'sha0/ifmemdsk'
H. Peter Anvin [Wed, 24 Aug 2011 22:40:05 +0000 (15:40 -0700)]
Merge remote-tracking branch 'sha0/ifmemdsk'

Conflicts:
NEWS

12 years agoNEWS: document isohybrid -m
H. Peter Anvin [Wed, 24 Aug 2011 22:39:35 +0000 (15:39 -0700)]
NEWS: document isohybrid -m

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
12 years agoisohybrid: Generate GPT and Mac bootable images
Matthew Garrett [Thu, 11 Aug 2011 18:58:09 +0000 (19:58 +0100)]
isohybrid: Generate GPT and Mac bootable images

EFI systems typically don't support booting off ISO 9660 filesystems,
even if written to USB sticks. This patch adds support for generating a
GPT that covers the stick as well, with an additional partition entry
pointing purely at the secondary El Torito image. When burned to CD the
secondary El Torito will be used as an EFI boot image, and when written
to a USB stick the GPT partition will be found and may be booted from.

However, some earlier EFI Macs don't support booting from El Torito
images via EFI. To cater for them this also supports generating an Apple
partition table, allowing a third El Torito image in HFS+ format to be
made available to the firmware. This requires padding the MBR images
slightly in order to leave space for the Apple header, but should have
no functional impact.

Sadly, this breaks the workaround for Acer BIOSes (magic xor
instruction) when Mac support is enabled via -m... not much that can
be done about that.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
12 years agoifmemdsk.c32: Allow boot options based on presence of MEMDISK
Shao Miller [Sat, 6 Aug 2011 09:24:46 +0000 (05:24 -0400)]
ifmemdsk.c32: Allow boot options based on presence of MEMDISK

Like 'ifcpu.c32' and 'ifplop.c32', this COMBOOT32 module will
check for the presence of MEMDISK(s) and choose one of two
(optionally) specified commands.

Signed-off-by: Shao Miller <shao.miller@yrdsb.edu.on.ca>
12 years agoextlinux: remove dead code in installer syslinux-4.05-pre5
H. Peter Anvin [Thu, 28 Jul 2011 21:13:31 +0000 (14:13 -0700)]
extlinux: remove dead code in installer

Remove some dead code in the installer, which caused build error with
gcc 4.6 and -Werror.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
12 years agobtrfs: Correctly determine the installation subvolume
Yi Yang [Tue, 12 Jul 2011 06:53:50 +0000 (14:53 +0800)]
btrfs: Correctly determine the installation subvolume

There are multiple ways to set up subvolumes in btrfs.  Use a general
determination method which works for all schemes.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
13 years agohdt: Fixing array size
Erwan Velu [Mon, 11 Jul 2011 20:10:34 +0000 (22:10 +0200)]
hdt: Fixing array size

Bootable flags shall be 6 char to let one NULL terminating char.
Thx Gerth for reporting this.

13 years agoMerge remote-tracking branch 'hpa/master'
Erwan Velu [Mon, 11 Jul 2011 18:20:59 +0000 (20:20 +0200)]
Merge remote-tracking branch 'hpa/master'

13 years agoextlinux: remove wrong use of strstr() syslinux-4.05-pre4
Paulo Alcantara [Tue, 5 Jul 2011 22:12:24 +0000 (22:12 +0000)]
extlinux: remove wrong use of strstr()

strstr() is only used for zero-terminated strings, so OEM Name is not
guaranteed to be zero-terminated. Instead, use fat_check_sb_fields() to
sanity check FAT superblock.

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
13 years agoimplement fat_check_sb_fields()
Paulo Alcantara [Tue, 5 Jul 2011 22:08:04 +0000 (22:08 +0000)]
implement fat_check_sb_fields()

Sanity check for FAT volumes

Signed-off-by: Paulo Alcantara <pcacjr@gmail.com>
13 years agoisohybrid: unbreak hex/octal arguments in C version syslinux-4.05-pre3
P J P [Tue, 28 Jun 2011 23:31:38 +0000 (16:31 -0700)]
isohybrid: unbreak hex/octal arguments in C version

Allow hex/octal arguments in the C version of isohybrid, since they
were accepted by the Perl version.

13 years agoMerge remote-tracking branch 'genec/extlinux-geo-for-hpa'
H. Peter Anvin [Tue, 28 Jun 2011 23:20:48 +0000 (16:20 -0700)]
Merge remote-tracking branch 'genec/extlinux-geo-for-hpa'

13 years agocom32: make syslinux_dump_*() pure debugging functions
H. Peter Anvin [Tue, 28 Jun 2011 02:06:43 +0000 (19:06 -0700)]
com32: make syslinux_dump_*() pure debugging functions

Make the syslinux_dump_*() functions pure debugging functions; that is
the way that they are used anyway, and this way they log to the same
place as dprintf.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
13 years agoUse <dprintf.h> instead of an ad hoc definition
H. Peter Anvin [Tue, 28 Jun 2011 01:51:11 +0000 (18:51 -0700)]
Use <dprintf.h> instead of an ad hoc definition

Instead of using ad hoc definitions of dprintf, dprintf2, vdprintf and
vdprintf2, use <dprintf.h> everywhere possible.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
13 years agodprintf: always define dprintf2/vdprintf2 syslinux-4.05-pre2
H. Peter Anvin [Tue, 21 Jun 2011 04:33:38 +0000 (21:33 -0700)]
dprintf: always define dprintf2/vdprintf2

Always define dprintf2/vdprintf2 even with no debugging at all
enabled.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
13 years agolibupload: use %zx to print a size_t argument
H. Peter Anvin [Tue, 21 Jun 2011 04:31:51 +0000 (21:31 -0700)]
libupload: use %zx to print a size_t argument

To print an argument of type size_t we should use the %z size
modifier.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
13 years agolib: sync <stdint.h> and <stddef.h> from klibc
H. Peter Anvin [Tue, 21 Jun 2011 04:30:39 +0000 (21:30 -0700)]
lib: sync <stdint.h> and <stddef.h> from klibc

The stdint.h definition appears to be a total mess.  Clean it up with
the current version from klibc, and revert change of size_t being long
rather than int (it is not on i386).

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
13 years agolib/shuffle: use <dprintf.h>
H. Peter Anvin [Tue, 21 Jun 2011 04:19:46 +0000 (21:19 -0700)]
lib/shuffle: use <dprintf.h>

Another straggler not yet using <dprintf.h> for debugging.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
13 years agodprintf: add the ability to log to stdio
H. Peter Anvin [Tue, 21 Jun 2011 04:19:17 +0000 (21:19 -0700)]
dprintf: add the ability to log to stdio

Add the ability to redirect dprintf to stdio when there is no other
choice.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
13 years agochain.c32: fix types in dprintf()
H. Peter Anvin [Tue, 21 Jun 2011 04:18:49 +0000 (21:18 -0700)]
chain.c32: fix types in dprintf()

Correct invalid types in dprintf().

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
13 years agolib: Add missing bits to stddef.h, stdint.h; pointers are "long"
H. Peter Anvin [Tue, 21 Jun 2011 04:17:23 +0000 (21:17 -0700)]
lib: Add missing bits to stddef.h, stdint.h; pointers are "long"

Add missing bits to stddef.h and stdint.h and correct the basic
pointer type to "long".  We really should use the gcc macros for
this...

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
13 years agohdt: New release 0.5.1
Erwan Velu [Mon, 6 Jun 2011 19:27:22 +0000 (21:27 +0200)]
hdt: New release 0.5.1

New release 0.5.1 to manage dynamic filenames based on "%" variables.

13 years agohdt: Ignoring checksums file
Erwan Velu [Mon, 6 Jun 2011 19:26:48 +0000 (21:26 +0200)]
hdt: Ignoring checksums file

Checksums file are generated by release target. No need to track it.

13 years agostrreplace: Fixing includes
Erwan Velu [Mon, 6 Jun 2011 19:23:38 +0000 (21:23 +0200)]
strreplace: Fixing includes

Stdlib was missing for malloc definition

13 years agoMerge remote-tracking branch 'hpa/master'
Erwan Velu [Mon, 6 Jun 2011 19:18:46 +0000 (21:18 +0200)]
Merge remote-tracking branch 'hpa/master'

13 years agocore: extend conventional memory warning with amount seen syslinux-4.05-pre1
H. Peter Anvin [Tue, 24 May 2011 16:40:12 +0000 (09:40 -0700)]
core: extend conventional memory warning with amount seen

Extend the conventional memory warning with the amount actually seen.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
13 years agoextlinux/main.c: Fix geometry handling
Gene Cumm [Wed, 11 May 2011 16:03:30 +0000 (12:03 -0400)]
extlinux/main.c: Fix geometry handling

1) ioctl HDIO_GETGEO expects a pointer to a struct hd_geometry
2) struct stat's st_dev is the parent file used; st_rdev is what we want

Reported-by: Michael Tokarev <mjt@tls.msk.ru>
Tested-By: Michael Tokarev <mjt@tls.msk.ru>
Signed-off-by: Gene Cumm <gene.cumm@gmail.com>
13 years agoAdd "make strip" target
H. Peter Anvin [Tue, 10 May 2011 03:02:27 +0000 (20:02 -0700)]
Add "make strip" target

Add a "make strip" target, to make doing the official build easier.
We want the official build to have stripped binaries for size reasons,
so do it right and make it an actual build target.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
13 years agoRemove -s for host binaries
H. Peter Anvin [Mon, 9 May 2011 22:04:29 +0000 (15:04 -0700)]
Remove -s for host binaries

Remove -s for host binaries; current practice is to let the distro
packaging systems do that themselves.  For the official binaries, we
should probably strip them, but via an external tool.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
13 years agolinux: remove debugging OPTFLAGS
H. Peter Anvin [Mon, 9 May 2011 22:01:45 +0000 (15:01 -0700)]
linux: remove debugging OPTFLAGS

Remove debugging OPTFLAGS which really should have made it in.

Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
13 years agosyslinux.mk: $(MAKEDIR), not $(makefiledir)
H. Peter Anvin [Mon, 2 May 2011 01:01:22 +0000 (18:01 -0700)]
syslinux.mk: $(MAKEDIR), not $(makefiledir)

Fix the path for the optional makefile segments.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
13 years agochain.c32: allow specifying arbitrary seg/offs/entry
H. Peter Anvin [Mon, 2 May 2011 01:00:19 +0000 (18:00 -0700)]
chain.c32: allow specifying arbitrary seg/offs/entry

Augment the seg= option to also allow for the offset and entry point
to be specified.

Signed-off-by: H. Peter Anvin <hpa@zytor.com>
13 years agochain: Support loading ReactOS' FreeLdr
Shao Miller [Sun, 1 May 2011 17:49:43 +0000 (13:49 -0400)]
chain: Support loading ReactOS' FreeLdr

Use something like:

  LABEL freeldr
    COM32 chain.c32
    APPEND freeldr=freeldr.sys

Signed-off-by: Shao Miller <shao.miller@yrdsb.edu.on.ca>
13 years agochain: Run Nindent on chain.c
Shao Miller [Sun, 1 May 2011 16:43:09 +0000 (12:43 -0400)]
chain: Run Nindent on chain.c

Minor formatting fix-ups

Signed-off-by: Shao Miller <shao.miller@yrdsb.edu.on.ca>
13 years agodiag/geodsp/Makefile: Use make variables
Gene Cumm [Fri, 29 Apr 2011 21:16:57 +0000 (17:16 -0400)]
diag/geodsp/Makefile: Use make variables

Use the appropriate variables but for now the flags/options are
overridden in this Makefile.  Also add XZ to mk/syslinux.mk

Signed-off-by: Gene Cumm <gene.cumm@gmail.com>
13 years agohdt: Adding dump_filename= option
Erwan Velu [Thu, 28 Apr 2011 18:26:24 +0000 (20:26 +0200)]
hdt: Adding dump_filename= option

This option allow the user to redefine the way filename is made-of.
By default that's "mac_address+dmi_vendor_name+dmi_product_name"

This can be overidded by the following syntax:

dump_filename=your_filename

Note that it's mandatory not to add "" or '' around your_filename.

your_filename could be a regular filename but can also includes %{}
directives as defined by :
%{m} = mac address
%{v} = vendor name of the machine
%{p} = product name of the machine
%{ba} = Asset tag of the base board
%{bs} = Serial number of the base board
%{ca} = Asset tag of the chassis
%{cs} = Serial number of the chassis
%{sk} = SKU number of the system
%{ss} = Serial number of the system

Here come a possible command line :
APPEND nomenu auto='dump;' dump_path=hdt2 tftp_ip=192.168.1.254 dump_filename=%{m}+%{bs}+%{ba}+%{cs}+%{ca}+%{sk}+%{ss}+%{p}+%{v}

13 years agostring: Adding strreplace
Erwan Velu [Wed, 27 Apr 2011 20:14:06 +0000 (22:14 +0200)]
string: Adding strreplace

13 years agodiag/geodsp: Save the .bin files
Gene Cumm [Tue, 26 Apr 2011 23:47:09 +0000 (19:47 -0400)]
diag/geodsp: Save the .bin files

They are a BTARGET file

Signed-off-by: Gene Cumm <gene.cumm@gmail.com>
13 years agodiag/geodsp: Makefile cleanup
Gene Cumm [Tue, 26 Apr 2011 23:43:44 +0000 (19:43 -0400)]
diag/geodsp: Makefile cleanup

Unneeded statements; delete *.lst like other Makefiles

13 years agodiag/geodsp: Fix .img.gz prerequisite
Gene Cumm [Tue, 26 Apr 2011 23:10:52 +0000 (19:10 -0400)]
diag/geodsp: Fix .img.gz prerequisite

13 years agohdt: Adding hybrid iso image during release
Erwan Velu [Tue, 26 Apr 2011 17:52:25 +0000 (19:52 +0200)]
hdt: Adding hybrid iso image during release

When we release hdt, generating an usb hybrid image could be useful.