> Do you have any logs which hint at when exactly the MCE gets raised? And the values of the MCi_STATUS (+ADDR/MISC if appropriate) registers that were being logged. Did the kernel try to log in response to an INT#18 machine check? A CMCI? Or was the kernel just polling the banks (it does that early in boot to see if there are errors left over from a previous crash). -Tony ��.n��������+%������w��{.n�����{�����ǫ���ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f