Back on 22 Sep 2015, Paul Burton added the following patch to the Kernel: https://lore.kernel.org/linux-mips/1442945538-26141-7-git-send-email-paul.burton@xxxxxxxxxx/ I am trying to kexec a mips32 kernel on a MT7621 target (1004Kc Cores). The target has two cores/4 threads (It is NOT the single core MT7621 variant). The Kexeced OS is locking up hard in this code, randomly, about 30% of the time. So the boot sequence is uboot -> first linux kernel -> kexec -> second linux kernel. Its the second one locking up. The first never does. This is the output: [ 1.891609] Waiting for core 1 to start... STAT_CONF=0x4a0504 I can't find any documentation on the CPC Sequencer, so what state this is, and why its stuck is a mystery. I have tried analysis of various GCR and VPE/TC state registers but i can't find anything obviously different between when it doesn't lock up, and when it does. Can anyone assist with an explanation of what that state means, and where i should be looking to find why its locked up? My company would also be willing to pay for assistance in solving this issue, as it is a high threat to our product we are developing. Steven Johnson
