Hi, Currently I am building a new PXE boot environment and during that I noticed something odd while attempting to boot using one of those universal netbootdisks based on the '3Com Universal NDIS driver v1.00' (aka undis3c) driver. Using pxelinux.0 this works like a charm, but when I switch to gpxelinux.0 the 3Com driver bails out on me... all other configuration is the same. Have reproduced these results using syslinux 3.72 and also 3.74 (latest version at the time of writing) Tia, Nico
Nico de Haer wrote:> Hi, > > Currently I am building a new PXE boot environment and during that I noticed > something odd while attempting to boot using one of those universal > netbootdisks based on the '3Com Universal NDIS driver v1.00' (aka undis3c) > driver. > > Using pxelinux.0 this works like a charm, but when I switch to gpxelinux.0 > the 3Com driver bails out on me... all other configuration is the same. Have > reproduced these results using syslinux 3.72 and also 3.74 (latest version > at the time of writing) >Got a pointer to said disc and how it is booted? This class of problems isn't really unexpected, unfortunately. The cleanup phase in gpxelinux.0 is still not quite there yet. -hpa
Nope - tried that, and the wiki states that it *should* be on the kernel line. On Fri, Apr 10, 2009 at 9:41 PM, Kevin Landreth <crackerjackmack at gmail.com>wrote:> On Fri, Apr 10, 2009 at 2:22 PM, Nico de Haer <nico.dehaer at gmail.com> > wrote: > > > Peter, > > > > Done some more tests, it also seems that the keeppxe parameter is being > > ignored by 3.74. > > > > Results below are done using 3.72: > > > > 1) booting the disk using pxelinux.0: works > > 2) booting the disk using gpxelinux.0 fails > > > > Output for test 2: (hope that this includes the pointer info you > requested) > > > > Disk is floppy 0, 1440 K, C/H/S = 80/2/18, EDD off > > Using raw access to high memory > > Total size needed = 2507 bytes, allocating 3K > > Old dos memory at 0x8f000 (map says 0x8f000), loading at 0x8e400 > > 1588: 0xffff 15E801: 0x3c00 0x0ed2 > > INT 13 08: Success, count 1, BPT = f000:81e9 > > old: int13 = f000ec59 int15 = 8f28061b int1e = f00081e9 > > new: int13 = 8e40000a int15 = 8e40037c int1e = f00081e9 > > Loading boot sector... booting... > > > > small chunk from my config file: > > kernel memdisk raw keeppxe > > append initrd=bootdisk/xpboot-144.ima > > > > > does moving the additional options from the kernel line down to append help > with the issue? > > label testing > kernel memdisk > append initrd=bootdisk/xpboot-144.ima raw keeppxe > > -- > Kevin Landreth > _______________________________________________ > Syslinux mailing list > Submissions to Syslinux at zytor.com > Unsubscribe or set options at: > http://www.zytor.com/mailman/listinfo/syslinux > Please do not send private replies to mailing list traffic. > >
--On Friday, April 10, 2009 10:01:24 PM +0200 Nico de Haer <nico.dehaer at gmail.com> wrote:> Nope - tried that, and the wiki states that it *should* be on the kernel > line.>> > small chunk from my config file: >> > kernel memdisk raw keeppxe >> > append initrd=bootdisk/xpboot-144.ima >> > >> > >> does moving the additional options from the kernel line down to append >> help with the issue? >> >> label testing >> kernel memdisk >> append initrd=bootdisk/xpboot-144.ima raw keeppxe"keeppxe" is an option to the "kernel" config command, and so should be on the kernel line. Putting it elsewhere will have no effect. "raw" is a parameter to memdisk, and so should be on the "append" line. Putting it on the kernel line instead may sometimes work in the menu system, but will not work in the command line, and in any case the command "kernel memdisk raw keeppxe" prevents the parser from seeing the "keeppxe" option. label testing kernel memdisk keeppxe append initrd=bootdisk/xpboot-144.ima raw -- Jeffrey T. Hutzelman (N3NHS) <jhutz+ at cmu.edu> Carnegie Mellon University - Pittsburgh, PA