Red Hat Bugzilla – Bug 162254
Unable to execute netbootimage created with mkzimage
Last modified: 2007-11-30 17:07:18 EST
From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.6) Gecko/20050302 Firefox/1.0.1 Fedora/1.0.1-1.3.2 Description of problem: Unable to execute the kernel from network boot image (zImage.initrd) created using /usr/bin/mkzimage script. Version-Release number of selected component (if applicable): vmlinuz-2.6.9-5.EL zImage.stub-2.6.9-5.EL How reproducible: Always Steps to Reproduce: 1. Create custom initrd image (works on other Linux platforms) 2. run the command: mkzimage vmlinuz-2.6.9-5.EL no no storix-initrd.img zImage.stub-2.6.9-5.EL zImage.initrd.storix 3. Copy image to bootp server and network boot IBM 7044-170 pSeries Actual Results: When attempting to boot, the error message is as follows: Load Addr=0x4000 MaxSize=0xbfc000 Final Packet Count=6317 Final File Size=3234300 bytes DEFAULT CATCH!, code fff00300 at %SRR0: 00c1d278 %SRR1: 00003030 ok> Expected Results: Should have at least seen the kernel executing and kernel load messages. The custom initrd is not the cause because the kernel should at least still execute. Additional info:
Can you also attach the ouput of .registers from the of prompt.
Created attachment 116250 [details] .registers from OF Prompt
Just in case... Clients Fix Pt Regs: 00 00000000 0121ffe0 00000000 00000000 00000000 00c1b320 00000000 00000000 08 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 10 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 18 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 Special Regs: %IV: 00000000 %SRR0: 00400000 %SRR1: 00003000 %MQ: 00000000 %CR: 00000000 %LR: 00000000 %CTR: 00000000 %XER: 00000000 %DAR: 00000000 %DSISR: 00000000 %SDRI: 013c0000 %ASR: 00000000 ok>
Dave - have you seen this?
I don't recall seeing this except when the image has been too large, no.
This looks like the image is to big. I'm NAKing this for 4.5 and closing it. If there's solid evidence found that this is not the problem, please re-open the bug and add that data.