| 1 |
> The 4.9.77 announcement upstream did introduce a number of objtool |
| 2 |
> changes. See below. |
| 3 |
> https://lkml.org/lkml/2018/1/17/198 |
| 4 |
> Josh Poimboeuf (4): |
| 5 |
> objtool, modules: Discard objtool annotation sections for modules |
| 6 |
> objtool: Detect jumps to retpoline thunks |
| 7 |
> objtool: Allow alternatives to be ignored |
| 8 |
> objtool: Fix retpoline support for pre-ORC objtool |
| 9 |
> |
| 10 |
> I can't seem to remember seeing reports of the issue you're having on |
| 11 |
> the LMKL, since 4.9.77 was announced. |
| 12 |
> As you said, try re-emerging the gentoo kernel, make mrproper, no need |
| 13 |
> for distclean, which does run mrproper as part of its operation, and |
| 14 |
> see if that helps. |
| 15 |
> Alternatively, you could try building the upstream version of 4.9.77, |
| 16 |
> and see if that works. Let me know if you need instructions on how to |
| 17 |
> proceed with that. |
| 18 |
> |
| 19 |
|
| 20 |
FYI : |
| 21 |
Just tried a different compile. |
| 22 |
Switched off expert and flipped the CPU to "Generic x86_64" ( no other |
| 23 |
changes ). |
| 24 |
|
| 25 |
It compiled. ?????? |
| 26 |
|
| 27 |
So ... what you choose for CPU in the kernel determines if the |
| 28 |
ext2/ext3/ext4 fs driver will compile. |
| 29 |
|
| 30 |
Been using Gentoo / "sys-kernel/gentoo-sources" w/experimental for 3+ |
| 31 |
years now. |
| 32 |
First time the CPU choice has caused a problem. |
| 33 |
|
| 34 |
Suspect that a mismatch of the "Spectre / Meltdown" patches that have |
| 35 |
somehow gotten into the ext4 fs driver and the code for specific CPUs. |
| 36 |
|
| 37 |
Wonderful ... just finished a complete reload of Gentoo. Now have to |
| 38 |
redo it again ... |
| 39 |
... the mistake? I used ext2/ext3 for the fs. |
| 40 |
|
| 41 |
Corbin |
Archives