| 1 |
On Fri, Jan 12, 2018 at 11:23 AM, Corbin Bird <corbinbird@×××××××.net> wrote: |
| 2 |
> |
| 3 |
> On 01/11/2018 05:02 PM, Rich Freeman wrote: |
| 4 |
>> |
| 5 |
>> IMO Spectre is going to drive some microcode updates for relatively |
| 6 |
>> recent CPUs, compiler improvements, and some hand-tuning of |
| 7 |
>> particularly critical code. |
| 8 |
>> |
| 9 |
> |
| 10 |
> The microcode updates pushed out for AMD by Gentoo seem to be only for : |
| 11 |
> Fam16h, Fam17h CPUs. |
| 12 |
> |
| 13 |
|
| 14 |
FWIW even the 17h microcode doesn't seem to be updating on my Ryzen: |
| 15 |
|
| 16 |
dmesg | grep microco |
| 17 |
[ 0.989279] microcode: CPU0: patch_level=0x08001129 |
| 18 |
[ 0.989421] microcode: CPU1: patch_level=0x08001129 |
| 19 |
[ 0.989565] microcode: CPU2: patch_level=0x08001129 |
| 20 |
[ 0.989708] microcode: CPU3: patch_level=0x08001129 |
| 21 |
[ 0.989857] microcode: CPU4: patch_level=0x08001129 |
| 22 |
[ 0.990001] microcode: CPU5: patch_level=0x08001129 |
| 23 |
[ 0.990183] microcode: CPU6: patch_level=0x08001129 |
| 24 |
[ 0.990332] microcode: CPU7: patch_level=0x08001129 |
| 25 |
[ 0.990475] microcode: CPU8: patch_level=0x08001129 |
| 26 |
[ 0.990619] microcode: CPU9: patch_level=0x08001129 |
| 27 |
[ 0.990764] microcode: CPU10: patch_level=0x08001129 |
| 28 |
[ 0.990905] microcode: CPU11: patch_level=0x08001129 |
| 29 |
[ 0.991095] microcode: Microcode Update Driver: v2.2. |
| 30 |
|
| 31 |
That said, there still isn't any AMD documentation around the |
| 32 |
microcode updates that I've been able to find, so I have no idea what |
| 33 |
the correct patch level is even supposed to be. I just know that I'm |
| 34 |
not getting a message about early updates. I do have linux 4.4.13 |
| 35 |
which includes the family 17h patch. |
| 36 |
|
| 37 |
The other odd thing is that a firmware update was released for my |
| 38 |
motherboard (ASRock AB350 Pro4) on the 10th, and if I flash it grub |
| 39 |
will no longer boot the linux kernel, and it is pretty slow overall, |
| 40 |
but it will still boot memtestx86 just fine. I figured I'd wait a few |
| 41 |
days and see if there is any further info on it. |
| 42 |
|
| 43 |
-- |
| 44 |
Rich |
Archives