3 ms·
> according to Intel, it happened only when the microcode was loaded by the kernel; they probably had tested only loading it from the firmware I followed the S
by theevilsharpie 6y ago
> according to Intel, it happened only when the microcode was loaded by the kernel; they probably had tested only loading it from the firmware
I followed the Spectre situation quite closely, and never heard anything along those lines. Do you have a source?
Many system vendors pulled the faulty firmware from their own sites, and Dell went so far as to recommend rolling back if you had installed it, so the partner ecosystem didn't want people running the firmware at all.
- cesarb 6y ago> > according to Intel, it happened only when the microcode was loaded by the kernel; they probably had tested only loading it from the firmware > I followed the Spectre situation quite closely, and never heard anything along those lines. Do you have a source? Sure, it was this one: https://github.com/intel/Intel-Linux-Processor-Microcode-Data-Files/issues/31#issuecomment-644885826 https://github.com/intel/Intel-Linux-Processor-Microcode-Dat... "Intel identified an issue when OS loading microcode update revision 0xDC for cpuid 406E3 and 506E3. The microcode update has been reverted to revision 0xD6. This issue does not affect the microcode update when loaded from BIOS."
- theevilsharpie 6y agoAh, I see. This is for a different firmware issue. They have so many these days, it's hard to keep track. :/ Thanks!