[ipxe-devel] Hard Lock on Chain Load from undionly.kpxe -> ipxe.pxe

[Matthew Helton]

I have observed the following on commit 4e03a: When chainloading from
undionly.ipxe to ipxe.pxe there is a hard lockup:

Client system PXEs into undionly.kpxe normally...reaches out for the
${uuid}.ipxe file as normal:

[Error code]
http://mywebserver.mydomain.net/NetBoot/iPXE/bin/ipxe.pxe... ok
PXE->EB: !PXE at 94D4:07A0, entry point at 94D4:033D
                 UNDI Code segment 94D4:083C, data segment 955A:@e60
(595-609kB)
                 UNDI device is PCI 02:01.0, type DIX+802.3
                 609 kB free bade memory after PXE unload
_
[Error code/]

The actual scripted invocation from ${uuid}.ipxe is:

[code]
#!ipxe
iseq ${chip} undionly && chain -ar ${17}/NetBoot/iPXE/bin/ipxe.pxe ||
[/code]

Can anyone else confirm?

Best,

Matt


-- 
There is never time enough to do it right, but there always seems to be
enough time to do it again.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ipxe.org/pipermail/ipxe-devel/attachments/20150821/7c3a42c3/attachment.htm>