Re: [PATCH] x86/insn_decoder_test: allow longer symbol-names

[Danilo Krummrich]

On 2/20/24 18:07, Miguel Ojeda wrote:
> On Fri, Jan 27, 2023 at 11:39 AM David Rheinsberg <david@xxxxxxxxxxxx> wrote:
> > The current workaround is to just disable CONFIG_X86_DECODER_SELFTEST, which I thought is a sad state. I can gladly use `256 + KSYM_NAME_LEN` and add a comment ala "room for insn-encoding and a symbol name". Would that be acceptable? The alternative would be to try to dyn-alloc a buffer and increase it to the actual line-length?
>
> John independently hit this issue again. Could we fix this? Going for
> the `256 + KSYM_NAME_LEN` sounds good enough for the moment since it
> would be a clear improvement, though I agree this could be cleaned up
> further.

I hit this independently as well. Miguel pointed me on this mail thread
when I sent another fix for this in [1].

> Sergio took the approach David suggested in a related patch [1], but
> perhaps it is best to submit the fix on its own so that it is easier
> to put it in. David, would you be so kind as to submit a v2 with that?
> Hopefully x86 can pick it up, otherwise with an Acked-by I am happy to
> take it too; and then Sergio can submit his patch on top again.

Sergio, David: Do you intend to follow up on this? Otherwise, I can also
pick this up and re-submit.

- Danilo

> Thanks!
>
> (Cc'ing also Masami who wrote this originally)
>
> [1] https://lore.kernel.org/rust-for-linux/20231119180145.157455-1-sergio.collado@xxxxxxxxx/
>
> Cheers,
> Miguel

[1] https://lore.kernel.org/rust-for-linux/20240325174924.95899-2-dakr@xxxxxxxxxx/