H. Peter Anvin [Wed, 1 Apr 2009 06:04:08 +0000 (23:04 -0700)]
shuffler: minor cleanups in bcopyxx.inc
Minor cleanups: calling convention, documentation, etc.
H. Peter Anvin [Wed, 1 Apr 2009 06:03:09 +0000 (23:03 -0700)]
shuffler: reload EDI, ESI, ECX before calling shuffle_and_boot_raw
comapi_cleanup can clobber registers, so reload EDI, ESI, ECX before
calling shuffle_and_boot_raw.
H. Peter Anvin [Wed, 1 Apr 2009 04:25:18 +0000 (21:25 -0700)]
shuffler: fix infinite loop in syslinux_memmap_find()
Fix an infinite loop in syslinux_memmap_find().
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Wed, 1 Apr 2009 00:30:24 +0000 (17:30 -0700)]
chain.c32: remove misleading comment
Remove a never-implemented, misleading comment
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
H. Peter Anvin [Wed, 1 Apr 2009 00:29:20 +0000 (17:29 -0700)]
shuffle: use SMT_RESERVED for the forbidden region
If we're going to mark memory unusable, it's cleaner to mark it
SMT_RESERVED.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
H. Peter Anvin [Tue, 31 Mar 2009 23:37:56 +0000 (16:37 -0700)]
chain.c32: remove loadbase < 0x7c00 hack
Remove the special hack for loadbase < 0x7c00. We no longer need it,
since the shuffler will relocate appropriately by magic.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
H. Peter Anvin [Tue, 31 Mar 2009 23:31:18 +0000 (16:31 -0700)]
shuffler: make the new shuffler actually work
Make the new shuffler actually work. This includes changing
rllpack.inc to run in 32-bit mode (since simple_pm_call now switches
to 32-bit mode) and changing the new shuffler interface to move the
shuffle list before actually doing any work.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
H. Peter Anvin [Tue, 31 Mar 2009 16:54:31 +0000 (09:54 -0700)]
shuffler: comboot interface to the new shuffler
Impact: new feature
COMBOOT interface to the new shuffler; resurrect comapi_chainboot as
that functionality is still potentially relevant.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Tue, 31 Mar 2009 16:25:27 +0000 (09:25 -0700)]
shuffer: make the new shuffler not pollute unrelated memory
Keep the shuffler from polluting memory outside its own "safe area".
This means being more clever about the relocation code, but it should
make it a lot easier to use for our own uses.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Tue, 31 Mar 2009 15:59:00 +0000 (08:59 -0700)]
shuffler: first cut of a simpler shuffle routine
Impact: major restructuring
New shuffler routine which runs entirely in protected mode, unlike the
old one which would enter PM for the bcopy phase only. This is not
only faster, but it greatly reduces the size of the shuffler "critical
area". The interfaces to this new shuffler are not yet implemented,
so this is a non-functional checkpoint.
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
H. Peter Anvin [Mon, 30 Mar 2009 23:03:30 +0000 (16:03 -0700)]
Fix the new kernel command-line parser
Impact: bug fix
We were incorrectly looking at [ds:di] instead of [es:di] for
verifying that a command was properly terminated. Fix.
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
H. Peter Anvin [Mon, 30 Mar 2009 20:37:09 +0000 (13:37 -0700)]
memdisk: Int13FuncsCnt is a constant, not a memory location
Impact: bug fix
Int13FuncsCnt is a compile-time constant, not a memory location. This
is a rather important distinction.
Erwan Velu [Sun, 29 Mar 2009 19:28:23 +0000 (21:28 +0200)]
pci: Rename get_module_name_from_pci_ids
Impact: none
Rename get_module_name_from_pci_ids to get_module_name_from_pcimap since
the module names are read from the modules.pcimap and not pci.ids file.
Signed-off-by: Sebastian Herbszt (herbszt@gmx.de)
Erwan Velu [Fri, 27 Mar 2009 21:03:24 +0000 (22:03 +0100)]
hdt: prevent unless disk's menu
Impact: none
No need to compute disk's menu if no disk are detected
Erwan Velu [Fri, 27 Mar 2009 18:04:00 +0000 (19:04 +0100)]
hdt: Adding memtest= boot parameter
Impact: memtest=<label> allow users to give another label name for the
memtest call
Prior to that version, memtest label was hardcoded to "memtest".
Now, "memtest" is the default label but could be overrided by users.
Erwan Velu [Fri, 27 Mar 2009 17:40:59 +0000 (18:40 +0100)]
hdt: Adding PXE information on the booted pci device
Impact: Display which pci device is the one used for PXE
Sync with cli's behavior
Erwan Velu [Fri, 27 Mar 2009 15:21:21 +0000 (16:21 +0100)]
doc: Adding more initrd details
Signed-off-by: Gert Hulselmans (hulselmansgert@gmail.com)
Erwan Velu [Fri, 27 Mar 2009 15:14:43 +0000 (16:14 +0100)]
doc: Fixing typo
Signed-off-by: Gert Hulselmans (hulselmansgert@gmail.com)
Erwan Velu [Thu, 26 Mar 2009 19:17:32 +0000 (20:17 +0100)]
hdt: Updating author name
Impact: none
Made for testing purpose
Erwan Velu [Thu, 26 Mar 2009 17:47:57 +0000 (18:47 +0100)]
hdt: bump version to 0.2.7
Impact: None
Bump to 0.2.7
Erwan Velu [Thu, 26 Mar 2009 17:36:50 +0000 (18:36 +0100)]
Merge commit 'origin/repair'
Erwan Velu [Wed, 25 Mar 2009 22:25:38 +0000 (23:25 +0100)]
hdt: in the cli, we have to Skip empty lines
Impact: empty lines were confusing history
History works better when we ignore empty lines
Erwan Velu [Wed, 25 Mar 2009 19:50:39 +0000 (20:50 +0100)]
hdt: Workaround for PXE detection
Impact: The PCI device matching could fails in some cases
In some cases, trying to find the exact bus/dev/fun/vendor/product
fails reported by the pxe rom fails. In such case, that workaround will
only look at the vendor/product id
Erwan Velu [Wed, 25 Mar 2009 17:51:48 +0000 (18:51 +0100)]
hdt: Fixing output of vesa software revision
Impact: Vesa's software revision was misdisplayed and could lead to
corruption
Displaying an int with %s isn't a good idea ;o)
Pierre-Alexandre Meyer [Wed, 25 Mar 2009 15:53:51 +0000 (08:53 -0700)]
hdt: Fix typo in vesa initialization
Impact: Bug fixing.
This caused some garbage to be displayed when detecting vesa.
Signed-off-by: Pierre-Alexandre Meyer <pierre@mouraf.org>
Erwan Velu [Wed, 25 Mar 2009 11:59:00 +0000 (12:59 +0100)]
hdt: Pxe mode might generate corruption
Impact: Prevent some hard lock due to pxe detection
In some cases, we can't find the associated PCI device.
In such situation, we shouldn't try to display its properties as
pci_device is set to NULL.
Erwan Velu [Tue, 24 Mar 2009 21:43:07 +0000 (22:43 +0100)]
hdt: Fixing crash on some optical drives
Impact: Prevent some hard crashes
On some optical drives, requesting int 13, func 0x48 make the system
crash. In fact, if the geometry isn't valid (optical drives), no need to
check for extended attributes
Erwan Velu [Tue, 24 Mar 2009 18:50:04 +0000 (19:50 +0100)]
menu: Increasing the maximum number of menus
Impact: some hosts can generate many menus when used with hdt
In some cases, we can run out of menus
Erwan Velu [Mon, 23 Mar 2009 17:20:38 +0000 (18:20 +0100)]
pci: Check if configuration type is PCI_CFG_NONE
Impact: prevent from grabbing configuration when PCI is PCI_CFG_NONE
Signed-off-by: Sebastian Herbszt (herbszt@gmx.de)
Erwan Velu [Mon, 23 Mar 2009 17:17:55 +0000 (18:17 +0100)]
pci: Remove "(void)cfgtype"
Impact: none
Remove useless "(void)cfgtype"
Signed-off-by: Sebastian Herbszt (herbszt@gmx.de)
Erwan Velu [Mon, 23 Mar 2009 09:15:11 +0000 (10:15 +0100)]
pci modules: Check for pci_scan() == NULL.
Impact: we don't have to grab more details if pci_scan returns NULL
Signed-off-by: Sebastian Herbszt (herbszt@gmx.de)
Erwan Velu [Mon, 23 Mar 2009 09:05:58 +0000 (10:05 +0100)]
hdt: Check for pci_scan() == NULL.
Impact: we don't have to grab more details if pci_scan returns NULL
Signed-off-by: Sebastian Herbszt (herbszt@gmx.de)
Erwan Velu [Sun, 22 Mar 2009 08:24:23 +0000 (09:24 +0100)]
hdt: Adding IPMI detection
Impact: HDT can now reports if an IPMI base board exists
By using DMI, it's now possible to detect IPMI base boards
Bump to 0.2.6
Erwan Velu [Sun, 22 Mar 2009 08:23:15 +0000 (09:23 +0100)]
dmi: Adding IPMI detection
Impact: IPMI base board can now be detected
Detecting IPMI via DMI structures
Erwan Velu [Wed, 25 Mar 2009 22:38:29 +0000 (23:38 +0100)]
Merge commit 'origin/erwan'
Erwan Velu [Wed, 25 Mar 2009 22:25:38 +0000 (23:25 +0100)]
hdt: in the cli, we have to Skip empty lines
Impact: empty lines were confusing history
History works better when we ignore empty lines
Erwan Velu [Wed, 25 Mar 2009 21:06:18 +0000 (22:06 +0100)]
Merge branch 'erwan'
Erwan Velu [Wed, 25 Mar 2009 19:50:39 +0000 (20:50 +0100)]
hdt: Workaround for PXE detection
Impact: The PCI device matching could fails in some cases
In some cases, trying to find the exact bus/dev/fun/vendor/product
fails reported by the pxe rom fails. In such case, that workaround will
only look at the vendor/product id
Erwan Velu [Wed, 25 Mar 2009 17:51:48 +0000 (18:51 +0100)]
hdt: Fixing output of vesa software revision
Impact: Vesa's software revision was misdisplayed and could lead to
corruption
Displaying an int with %s isn't a good idea ;o)
Pierre-Alexandre Meyer [Wed, 25 Mar 2009 15:53:51 +0000 (08:53 -0700)]
hdt: Fix typo in vesa initialization
Impact: Bug fixing.
This caused some garbage to be displayed when detecting vesa.
Signed-off-by: Pierre-Alexandre Meyer <pierre@mouraf.org>
Pierre-Alexandre Meyer [Wed, 25 Mar 2009 15:53:51 +0000 (08:53 -0700)]
hdt: Fix typo in vesa initialization
Impact: Bug fixing.
This caused some garbage to be displayed when detecting vesa.
Signed-off-by: Pierre-Alexandre Meyer <pierre@mouraf.org>
Erwan Velu [Wed, 25 Mar 2009 11:59:00 +0000 (12:59 +0100)]
hdt: Pxe mode might generate corruption
Impact: Prevent some hard lock due to pxe detection
In some cases, we can't find the associated PCI device.
In such situation, we shouldn't try to display its properties as
pci_device is set to NULL.
Erwan Velu [Tue, 24 Mar 2009 21:43:07 +0000 (22:43 +0100)]
hdt: Fixing crash on some optical drives
Impact: Prevent some hard crashes
On some optical drives, requesting int 13, func 0x48 make the system
crash. In fact, if the geometry isn't valid (optical drives), no need to
check for extended attributes
Erwan Velu [Tue, 24 Mar 2009 18:50:04 +0000 (19:50 +0100)]
menu: Increasing the maximum number of menus
Impact: some hosts can generate many menus when used with hdt
In some cases, we can run out of menus
Erwan Velu [Mon, 23 Mar 2009 17:20:38 +0000 (18:20 +0100)]
pci: Check if configuration type is PCI_CFG_NONE
Impact: prevent from grabbing configuration when PCI is PCI_CFG_NONE
Signed-off-by: Sebastian Herbszt (herbszt@gmx.de)
Erwan Velu [Mon, 23 Mar 2009 17:17:55 +0000 (18:17 +0100)]
pci: Remove "(void)cfgtype"
Impact: none
Remove useless "(void)cfgtype"
Signed-off-by: Sebastian Herbszt (herbszt@gmx.de)
Erwan Velu [Mon, 23 Mar 2009 09:15:11 +0000 (10:15 +0100)]
pci modules: Check for pci_scan() == NULL.
Impact: we don't have to grab more details if pci_scan returns NULL
Signed-off-by: Sebastian Herbszt (herbszt@gmx.de)
Erwan Velu [Mon, 23 Mar 2009 09:05:58 +0000 (10:05 +0100)]
hdt: Check for pci_scan() == NULL.
Impact: we don't have to grab more details if pci_scan returns NULL
Signed-off-by: Sebastian Herbszt (herbszt@gmx.de)
Erwan Velu [Sun, 22 Mar 2009 08:24:23 +0000 (09:24 +0100)]
hdt: Adding IPMI detection
Impact: HDT can now reports if an IPMI base board exists
By using DMI, it's now possible to detect IPMI base boards
Bump to 0.2.6
Erwan Velu [Sun, 22 Mar 2009 08:23:15 +0000 (09:23 +0100)]
dmi: Adding IPMI detection
Impact: IPMI base board can now be detected
Detecting IPMI via DMI structures
H. Peter Anvin [Sun, 22 Mar 2009 05:29:09 +0000 (22:29 -0700)]
Add "and contributors" to the copyright banner
Clarify that there are nowadays more copyright holders than just
myself.
Shao Miller [Sun, 22 Mar 2009 04:27:56 +0000 (00:27 -0400)]
memdisk: additional comments in memdisk.inc and postprocess.pl
Just a sprinkling of comments adding further description to the code.
H. Peter Anvin [Fri, 20 Mar 2009 23:34:19 +0000 (16:34 -0700)]
vesacon: remove unused com32sys_t variables
No more direct __intcall()'s in this code, so remove unused register
set variables.
H. Peter Anvin [Fri, 20 Mar 2009 23:31:11 +0000 (16:31 -0700)]
Merge branch 'memdisk-cd'
H. Peter Anvin [Fri, 20 Mar 2009 22:45:55 +0000 (15:45 -0700)]
Merge commit 'hdt/master'
Erwan Velu [Fri, 20 Mar 2009 22:37:05 +0000 (23:37 +0100)]
hdt: Adding contributors
Impact: A new cli command : hdt> show hdt
Bump to 0.2.5
Adding contributor's list
Erwan Velu [Fri, 20 Mar 2009 22:03:28 +0000 (23:03 +0100)]
hdt: commenting some devel messages
Impact: remove useless error messages
Some error messages were displayed even if commands were valid
Erwan Velu [Fri, 20 Mar 2009 22:02:05 +0000 (23:02 +0100)]
hdt: Fixing more_printf to display a proper \n
Impact: Improve the visibility of the message
more_printf could have printed that message without any \n
Erwan Velu [Fri, 20 Mar 2009 21:50:55 +0000 (22:50 +0100)]
hdt: Adding an history
Impact: UP & Down keys can be used to reach the command history
UP & Down keys can be used to parse the history
H. Peter Anvin [Fri, 20 Mar 2009 20:54:54 +0000 (13:54 -0700)]
vesacon: migrate disable fixes from ansicon
Copy the disabled fixes already done in ansicon from checkin
22cd4e378e3aafd402f43030941d6f6b300d0cea.
H. Peter Anvin [Fri, 20 Mar 2009 20:44:39 +0000 (13:44 -0700)]
Document menu.c32 + console 0 fix.
H. Peter Anvin [Fri, 20 Mar 2009 20:43:11 +0000 (13:43 -0700)]
ansicon: handle the disabled case properly
Even when disabled, we need to set the screen size, because the
downstream code depends on it. Set it to 25x80 (maybe we should
swtich this is 24x80) in this case.
Furthermore, when disabled, we should return success == bytes
"written", not zero.
H. Peter Anvin [Fri, 20 Mar 2009 20:41:58 +0000 (13:41 -0700)]
getc: move parseint_esdi out of the getint control flow
parseint_esdi was located in the middle of the getint -> parseint fall
through, with expectedly disastrous results.
Erwan Velu [Thu, 19 Mar 2009 22:17:09 +0000 (23:17 +0100)]
hdt: Fixing backspace
Impact: Backspace now have the proper behaviour
Since we can move in the string, backspace have to behave differently.
Fixing style.
Erwan Velu [Thu, 19 Mar 2009 20:40:43 +0000 (21:40 +0100)]
hdt: Adding append mode to cli
Impact: Users can now edit a line
Append mode is now the default mode
This features moving using LEFT,RIGHT,HOME & END keys.
Erwan Velu [Thu, 19 Mar 2009 18:01:43 +0000 (19:01 +0100)]
hdt: ctrl+a or HOME key return to the beginning of line
Impact: Users can now uses ctrl+a or HOME to return at the beginning of
a line
Ctrl+a or home key is now implemented
Erwan Velu [Thu, 19 Mar 2009 17:27:11 +0000 (18:27 +0100)]
hdt: Skip heading spaces
Impact: Allow user to put some space char before the real command
Before that patch, it was required not having any space before the
command. That's not so user friendly ;o)
Erwan Velu [Thu, 19 Mar 2009 17:21:55 +0000 (18:21 +0100)]
hdt: Using a single cli structure to manage the input line
Impact: Makes the code easier to write
Using a single s_cli structure for managing the input line, the cursor
position and the prompt
Erwan Velu [Thu, 19 Mar 2009 16:32:00 +0000 (17:32 +0100)]
hdt: Missing include for getkey
Impact: remove a warning
more_printf now uses the proper includes
Erwan Velu [Thu, 19 Mar 2009 11:08:34 +0000 (12:08 +0100)]
hdt: more_printf must use getkey
Impact: Users can now press a key to continue
Since we switch to an ansi raw console, the fgets() used
to catch the return when we display to much lines didn't worked.
We now use the getkey function to grab __any__ key, which is better
then.
Erwan Velu [Thu, 19 Mar 2009 10:34:49 +0000 (11:34 +0100)]
hdt: Rewrite DMI module using a new syntax for the CLI
Impact: Start refactoring the CLI.
In order to have an extensible CLI, the commands handling needs to be
refactored. This patch starts the migration by refactoring the DMI
module.
Note that the core of the new commands parser has been implemented.
Adapting other modules to use it simply requires to declare
new structures that maps the command string to the callback.
Signed-off-by: Pierre-Alexandre Meyer <pierre@mouraf.org>
Erwan Velu [Thu, 19 Mar 2009 10:03:29 +0000 (11:03 +0100)]
hdt: fixing one remaining display_prompt
Impact: make is now successfull ;o)
A display_prompt didn't get updated
Erwan Velu [Thu, 19 Mar 2009 09:57:46 +0000 (10:57 +0100)]
hdt: Renaming display_prompt to reset_prompt
Impact: Fix buffer corruption in cli.
The rename occurs to match the reality.
We actually do more than displaying the prompt as we reset some vars.
One of this parameters where wrong too and could occurs corruption.
Signed-off-by: Pierre-Alexandre Meyer <pierre@mouraf.org>
Erwan Velu [Thu, 19 Mar 2009 09:56:43 +0000 (10:56 +0100)]
Merge branch 'master' of git://git./boot/syslinux/syslinux
H. Peter Anvin [Thu, 19 Mar 2009 04:06:30 +0000 (21:06 -0700)]
runkernel: further simplify the command-line parser
Further simplifications to the command-line parser, both from a code
size and a usability perspective.
H. Peter Anvin [Thu, 19 Mar 2009 03:40:26 +0000 (20:40 -0700)]
Refactor command line parsing; support "quiet" option
Refactor the command line parser to be more of a general parser
instead of treating each option as an ad hoc feature.
Suppress the Loading... prompt if "quiet" is specified on the command
line. Some messed-up people want it this way.
Erwan Velu [Wed, 18 Mar 2009 19:01:39 +0000 (20:01 +0100)]
hdt: Handling CTRL+C
Impact: User can use the CTRL+C
Basic Ctrl+C implementation
Erwan Velu [Tue, 17 Mar 2009 21:56:53 +0000 (22:56 +0100)]
hdt: cli now uses the ansi raw interface
Impact: cli is now managed char by char instead of line by line
This patch is a first step toward a new version of the cli.
Current code isn't scalable to manage a complex cli code.
First step is to manage input char by char.
It will now possible to implement :
- completion using tab,
- using special keys (F1-F12)
This patch implements F1 as "show help"
Erwan Velu [Tue, 17 Mar 2009 19:10:48 +0000 (20:10 +0100)]
Merge branch 'master' of git://git./boot/syslinux/syslinux
Erwan Velu [Tue, 17 Mar 2009 11:39:20 +0000 (12:39 +0100)]
hdt: Adding vesa in available modes & help message
Impact: Fixes CLI behavior. show vesa is a valid command
"help" command didn't reported vesa as possible mode
"show vesa" command was also missing even if already implemented
Erwan Velu [Tue, 17 Mar 2009 09:31:00 +0000 (10:31 +0100)]
hdt: putting some vesa function static
Impact: none
Signed-off-by: Pierre-Alexandre Meyer <pierre@mouraf.org>
Erwan Velu [Tue, 17 Mar 2009 09:29:44 +0000 (10:29 +0100)]
hdt: Making the build silent
Impact: Silence the build.
It is easier to detect warnings and errors if make doesn't print out the
commands before executing them.
Signed-off-by: Pierre-Alexandre Meyer <pierre@mouraf.org>
H. Peter Anvin [Tue, 17 Mar 2009 04:42:01 +0000 (21:42 -0700)]
Merge branch 'master' into memdisk-cd
H. Peter Anvin [Tue, 17 Mar 2009 04:32:32 +0000 (21:32 -0700)]
*/Makefile: fixes for older versions of NASM
Deal with issues that made older versions of nasm:
a) prevent "make spotless" from running
b) generate bad dependencies (no final newline)
H. Peter Anvin [Tue, 17 Mar 2009 04:19:37 +0000 (21:19 -0700)]
*/Makefile: use -include when including dependencies
This seems to have better behavior for "make spotless".
H. Peter Anvin [Tue, 17 Mar 2009 04:18:42 +0000 (21:18 -0700)]
Revert "memdisk: memdisk16.asm needs %ifndef DEPEND around %include"
This reverts commit
8f3919c8f5dd37ccd3b7565d32ef1dac8b2522d5.
Wrong fix.
H. Peter Anvin [Tue, 17 Mar 2009 04:11:52 +0000 (21:11 -0700)]
memdisk: memdisk16.asm needs %ifndef DEPEND around %include
Older versions of NASM puke if there is an %include of a non-existent
file during dependency generation. Make it more forgiving.
H. Peter Anvin [Tue, 17 Mar 2009 02:20:02 +0000 (19:20 -0700)]
memdisk: align .bss to a 16-byte boundary for better zeroing perf
Some CPUs really want a reasonable alignment for a rep;stosl, so give
it to them...
H. Peter Anvin [Tue, 17 Mar 2009 02:17:57 +0000 (19:17 -0700)]
memdisk: zero the .bss section
It is just too confusing to have a .bss section that isn't zeroed, to
explicitly zero it.
H. Peter Anvin [Tue, 17 Mar 2009 02:05:03 +0000 (19:05 -0700)]
memdisk: move the floppy disk geometry code in parallel with the MBR code
Move the floppy disk geometry generation code into parallel with the
MBR code, which pretty much mimics the previous behavior.
H. Peter Anvin [Tue, 17 Mar 2009 01:50:36 +0000 (18:50 -0700)]
memdisk: don't key FAT vs MBR on the type of disk
We can't rely on the disk type to say if this is FAT vs MBR, since we
haven't determined that yet. Thus, move the FAT test up before the
MBR test; both are heuristic, but the FAT test has much less
probability of false positives, and can be further tightened if
needed.
H. Peter Anvin [Mon, 16 Mar 2009 23:16:39 +0000 (16:16 -0700)]
SYSLINUX: fix handling of .bss files for FAT12/FAT16
The size of the superblock is different for FAT12/FAT16 and FAT32. We
should only copy the relevant portions of the superblock over.
Technically speaking, it is possible for FAT12/FAT16 to have an older,
smaller superblock as well (without the 0x29 boot signature); we don't
handle that at the moment.
H. Peter Anvin [Mon, 16 Mar 2009 20:21:04 +0000 (13:21 -0700)]
Merge commit 'hdt/master'
H. Peter Anvin [Mon, 16 Mar 2009 20:18:49 +0000 (13:18 -0700)]
memdisk: auto-detect large floppy geometry if it is FAT
If a "large floppy" image is formatted with a FAT filesystem, we can
use the headers in the FAT image to derive the geometry. This is nice
and user-friendly, so do it that way.
H. Peter Anvin [Mon, 16 Mar 2009 16:24:02 +0000 (09:24 -0700)]
isolinux: split MaxTransfer for HD and CD modes
The MaxTransfer for CD modes has to be smaller than for HD modes to
avoid 64K overrun (even though in theory it's permitted, we don't
trust it.) Thus have separate variables for HD and CD modes.
Erwan Velu [Mon, 16 Mar 2009 15:23:47 +0000 (16:23 +0100)]
hdt: Improving vesa output
Impact: Use proper words to be better understood by users
Cosmetic fix.
Erwan Velu [Mon, 16 Mar 2009 15:20:28 +0000 (16:20 +0100)]
hdt: Adding missing install target
Impact: Allow the build process to complete properly
The install: target was missing and prevented the global make install to
end properly
Erwan Velu [Sun, 15 Mar 2009 19:34:48 +0000 (20:34 +0100)]
hdt: Uploading missing vesa files
The vesa files were missing
Erwan Velu [Sun, 15 Mar 2009 19:28:23 +0000 (20:28 +0100)]
Merge branch 'master' of git://git./boot/syslinux/syslinux
Pierre-Alexandre Meyer [Sat, 14 Mar 2009 23:51:56 +0000 (16:51 -0700)]
hdt: Command-line arguments should be const
Impact: Fix compilation warnings, signatures changes.
argc and argv should be const.
Signed-off-by: Pierre-Alexandre Meyer <pierre@mouraf.org>