| 1 |
On Saturday 30 Apr 2016 12:00:44 Peter Humphrey wrote: |
| 2 |
> On Saturday 30 Apr 2016 10:24:43 Mick wrote: |
| 3 |
> > On Saturday 30 Apr 2016 09:29:08 Neil Bothwick wrote: |
| 4 |
> > > On Sat, 30 Apr 2016 09:07:29 +0100, Mick wrote: |
| 5 |
> > > > I seem to have mislaid my microcode somewhere, in the latest stable |
| 6 |
> > > > gentoo kernel: |
| 7 |
> > > > |
| 8 |
> > > > # grep -i MICROCODE .config |
| 9 |
> > > > # |
| 10 |
> > > |
| 11 |
> > > Grepping .config is unreliable, and always has been. You should use the |
| 12 |
> > > search option in make menuconfig. If th eoption you want does not show |
| 13 |
> > > up |
| 14 |
> > > in .config, you are probably missing one of the options it depends on. |
| 15 |
> > |
| 16 |
> > Thanks Neil, I use both, especially if grepping does not reveal what I am |
| 17 |
> > looking for. |
| 18 |
> > |
| 19 |
> > > The search results show what they are and whether they are set. I get |
| 20 |
> > > this |
| 21 |
> > > |
| 22 |
> > > Symbol: MICROCODE [=y] |
| 23 |
> > > |
| 24 |
> > > Type : boolean |
| 25 |
> > > Prompt: CPU microcode loading support |
| 26 |
> > > Location: |
| 27 |
> > > (1) -> Processor type and features |
| 28 |
> > > |
| 29 |
> > > Defined at arch/x86/Kconfig:1159 |
| 30 |
> > > Depends on: CPU_SUP_AMD [=y] || CPU_SUP_INTEL [=y] |
| 31 |
> > > Selects: FW_LOADER [=y] |
| 32 |
> > |
| 33 |
> > Yes, I got (almost) the same and it shows that MICROCODE is a top level |
| 34 |
> > module, but I can't see it when I walk through the 'Processor type and |
| 35 |
> > features' contents. Fine, my eyesight is not what it used to be, but it |
| 36 |
> > can't be this bad! :p |
| 37 |
> > |
| 38 |
> > Symbol: MICROCODE [=n] |
| 39 |
> > Type : boolean |
| 40 |
> > Prompt: CPU microcode loading support |
| 41 |
> > |
| 42 |
> > Location: |
| 43 |
> > (1) -> Processor type and features |
| 44 |
> > |
| 45 |
> > Defined at arch/x86/Kconfig:1125 |
| 46 |
> > Depends on: (CPU_SUP_AMD [=y] || CPU_SUP_INTEL [=y]) && BLK_DEV_INITRD |
| 47 |
> > Selects: FW_LOADER [=y] |
| 48 |
> |
| 49 |
> It seems to require BLK_DEV_INITRD if my logic is still up to scratch. What |
| 50 |
> happens if you set this; does MICROCODE appear then? |
| 51 |
|
| 52 |
Yes! What the ... ? O_O |
| 53 |
|
| 54 |
< > Dell i8k legacy laptop support |
| 55 |
[*] CPU microcode loading support (NEW) |
| 56 |
[*] Intel microcode loading support (NEW) |
| 57 |
[ ] AMD microcode loading support (NEW) |
| 58 |
|
| 59 |
|
| 60 |
It appears and as you can see old settings still stand. This is really odd. |
| 61 |
I don't have INITRD set on most/all of my PCs, but on other PCs with the same |
| 62 |
kernel I can see MICROCODE. This disappearance act happens on a no-multilib |
| 63 |
system. Could this have anything to do with it? |
| 64 |
|
| 65 |
I will compile a couple more of my systems this weekend to see what gives. I |
| 66 |
can't recall ever coming across this problem in the past. |
| 67 |
|
| 68 |
-- |
| 69 |
Regards, |
| 70 |
Mick |
Archives