aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* sysdump: use the same serial receive threshold as the coresyslinux-3.85-pre9H. Peter Anvin2010-02-071-1/+1
| | | | | | Match the core with regards to the serial receive threshold. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: make ymodem work even on a serial consoleH. Peter Anvin2010-02-073-13/+47
| | | | | | | | | Make it possible to do a ymodem transfer even on the port used for a serial console. Furthermore, on a serial console, get the parameters from the serial console information, so port number and speed don't have to be specified. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: don't include final null byte in version fileH. Peter Anvin2010-02-071-1/+1
| | | | | | We don't want the null byte at the end as part of the version file. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: ymodem: actually retransmit a block if we get NAKH. Peter Anvin2010-02-071-2/+2
| | | | | | | Incorrectly nested loops, we need to actually retransmit if we get a NAK reply... Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: actually make the new backend system worksyslinux-3.85-pre8H. Peter Anvin2010-02-077-57/+72
| | | | | | With these changes, both TFTP and Ymodem seem to actually work. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: change to a two-phase generate/output model, buffer in memoryH. Peter Anvin2010-02-0710-193/+183
| | | | | | | | Instead of outputting data as it is generated, buffer it all in memory and then output it all at once. This not only gives us exact size and so on before we start, but turns out to be faster at least for TFTP. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: serial: add missing header fileH. Peter Anvin2010-02-071-0/+1
| | | | | | Missing header file -> warning. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: be_tftp.c: fix prototypeH. Peter Anvin2010-02-071-1/+3
| | | | | | The init function now takes an additional length argument. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: *actually* generate the proper version stringH. Peter Anvin2010-02-072-4/+10
| | | | | | Make the gen-id.sh script actually do the right thing... Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: handle the case of more than one DMI tableH. Peter Anvin2010-02-071-4/+11
| | | | | | | | On the offchance a system has more than one valid DMI table, record them all. Also record the address of the signature structure and of the table itself. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: record the sysdump version used to generate a bundleH. Peter Anvin2010-02-072-1/+15
| | | | | | Record the sysdump version used to actually generate a data set. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: dump CPUID informationsyslinux-3.85-pre7H. Peter Anvin2010-02-073-0/+115
| | | | | | | Dump CPUID information in as generic of a way as is possible, given the ugliness in certain places. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: remove unused variableH. Peter Anvin2010-02-071-1/+0
| | | | Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: be more explicit about what we are doingH. Peter Anvin2010-02-064-2/+9
| | | | | | | Print more information on the console about what is going on. That can help detect problems if we find a machine which hangs on sysdump. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: fix the dumping of the e820 memory informationH. Peter Anvin2010-02-061-0/+1
| | | | | | Need to advance the pointer... Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: fix hacking of the DMI structure addressH. Peter Anvin2010-02-061-7/+4
| | | | | | Fix the hacking of the DMI structure address, for dmidecode's sake. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: dump memory map informationH. Peter Anvin2010-02-063-0/+82
| | | | | | Dump the memory map information - e820 et al. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: add PCI config space dumpingH. Peter Anvin2010-02-064-1/+72
| | | | | | Dump PCI configuration space. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: add support for dumping DMI tablesH. Peter Anvin2010-02-066-91/+136
| | | | | | | Add support for dumping DMI tables; hopefully in a way that is compatible with dmidecode. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: add back ymodem support; clean up memory dump structureH. Peter Anvin2010-02-0615-275/+325
| | | | | | Now we just need more front ends (information providers)... Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: add missing return statement in cpio_writefile()H. Peter Anvin2010-02-061-0/+2
| | | | | | cpio_writefile() should return the actual return value... Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* pxe: add missing prototype for pxe_dns()H. Peter Anvin2010-02-061-0/+1
| | | | | | All global functions should have a prototype... Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: dump vesa modes, generate usable timestampsH. Peter Anvin2010-02-0612-15/+165
| | | | | | | Dump all the VESA modes; give functional timestamps for the cpio members. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: first working version (can dump lowmem over TFTP)H. Peter Anvin2010-02-069-6/+389
| | | | | | | | Most of the infrastructure for sysdump, and enough to make it useful. Sysdump will produce a gzipped cpio archive containing individual data members; for now only a memory dump of lowmem. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* com32: add library function for DNS lookupH. Peter Anvin2010-02-062-0/+71
| | | | | | Add a library function for DNS lookup Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* syslinux/pxe.h use the same type as <com32.h> for a far pointerH. Peter Anvin2010-02-061-4/+2
| | | | | | | | <syslinux/pxe.h> was using segoff16_t, <com32.h> was using far_ptr_t for the same thing -- use the same type so the macros are available for both. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* pxelinux: give our own IP address in the "derivate info" API callH. Peter Anvin2010-02-063-1/+4
| | | | | | Add our own IP address to the "get derivative-specific info" API call. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* sysdump: port memdump to com32 with the intent of enhancing itH. Peter Anvin2010-02-0610-0/+2652
| | | | | | Port memdump to com32, with the intent of modifying and enhancing it. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* memdump: srecs need a length, and max 32 data bytesH. Peter Anvin2010-02-051-7/+2
| | | | | | Length + 32 data bytes max... Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* memdump: allow outputting S-recordsH. Peter Anvin2010-02-058-26/+176
| | | | | | | | Allow outputting S-records, for users who only have the capability of passively monitoring a serial port as opposed to being able to capture the contents directly. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* pxelinux: call reset_pxe before local bootsyslinux-3.85-pre6H. Peter Anvin2010-02-051-0/+1
| | | | | | Reset the PXE configuration before we do a local boot. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* gPXE: change heap size to 512KH. Peter Anvin2010-02-052-2/+18
| | | | | | Change heap size from 128K to 512K to accommodate larger TCP windows. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* memdisk: update documentationH. Peter Anvin2010-02-041-4/+11
| | | | | | | Document the nopass option, eltorito.sys is now in the standard distro. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* gPXE: bump the window size to 64KH. Peter Anvin2010-02-042-2/+17
| | | | | | | | Change the window size from 4K to 64K to have at least a hope of sane WAN performance. Retransmission cost is mostly related to timeouts, not amount retransmitted, anyway. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* gPXE: suppress the Ctrl+B bannersyslinux-3.85-pre5H. Peter Anvin2010-02-033-1/+31
| | | | | | | | Suppress the Ctrl+B banner and zero out the timeout. It's not very useful for an integrated product, and very visually annoying on NBP chainload. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* Merge branch 'chainload'H. Peter Anvin2010-02-032-110/+210
|\
| * pxelinux: allow chainloading from gpxelinuxH. Peter Anvin2010-02-032-108/+208
| | | | | | | | | | | | | | Use the new "exit hook" functionality in gPXE to allow NBP chainloading from gpxelinux.0. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
| * pxelinux: stop using the PXE-provided stackH. Peter Anvin2010-02-031-2/+2
| | | | | | | | | | | | | | | | | | | | Using the PXE-provided stack complicates things tremendously for gPXE chainloading support, plus we are hostages to the stack size set up by the PXE stack (1.5K by spec.) Use a private stack instead, as the fsc branch already does. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* | menu: add MENU CLEAR optionH. Peter Anvin2010-02-034-1/+16
| | | | | | | | | | | | Add MENU CLEAR to clear the screen on menu exit. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* | menu: when running hidden, clear the screen before final bootH. Peter Anvin2010-02-031-5/+17
|/ | | | | | | Clear away the countdown message before actually running the final boot. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* gpxelinux: use the cached DHCP packetH. Peter Anvin2010-02-021-0/+1
| | | | | | | gPXE 1.0.0 has support for using the cached DHCP packet from the previous PXE stack, so do that rather than doing another DHCP. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* Update gPXE to version 1.0.0H. Peter Anvin2010-02-02683-9709/+64099
|
* mkdiskimage: add option -s to make a sparse imagesyslinux-3.85-pre4H. Peter Anvin2010-02-011-6/+10
| | | | | | | It is often fine to have a sparse file for the output image, so add an -s option to generate a sparse image. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* core: set QUIET_FLAG if we load Linux "quiet"H. Peter Anvin2010-02-012-4/+8
| | | | | | | | If we see the "quiet" flag for the Linux kernel, also suppress the early Linux kernel messages; for people who want a completely clean boot. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* chain.c32: Handle nonlinear logical partitionssyslinux-3.85-pre3H. Peter Anvin2010-01-271-5/+8
| | | | | | | | | | | Some partitioning program, including at least parted, apparently create the list of logical partition headers nonlinearly with the partition themselves. Thus, make it a requirement that then fit inside the overall extended partition, but not inside the corresponding sublogical partition. Reported-by: Gert Huselmans <gerth@zytor.com> Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* FAT: generate EOF depending on exceeding the number of clustersH. Peter Anvin2010-01-251-12/+20
| | | | | | | | Generate EOF by comparing to the actual number of clusters. We have to compute the number of clusters in the filesystem anyway during setup, so save the number and use it for the EOF computation. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* ldlinux: fix FAT EOF detectionH. Peter Anvin2010-01-251-3/+6
| | | | | | | We had the wrong highest possible cluster for the FAT; adjust it so that it matches the definition. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* runkernel: we only look at the first letter in vga=...H. Peter Anvin2010-01-251-4/+4
| | | | | | | | We only look at the first letter in the handling of the special vga= strings, make that manifest in the assembly code so we don't end up with unnecessary warnings. Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* Add support for "vga=current"H. Peter Anvin2010-01-222-4/+10
| | | | | | "vga=current" is a user-friendly synonym for "vga=0x0f04". Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* core: don't reset the video mode when invoking the kernelH. Peter Anvin2010-01-222-1/+10
| | | | | | | | Don't reset the video mode when invoking the kernel if we're booting quiet -- it breaks "vga=current" which some distros use for a smooth graphical transition all the way into the kernel. Signed-off-by: H. Peter Anvin <hpa@zytor.com>