Re: [PATCH v6 3/4] atomic: Add alignment check to instrumented atomic operations

[Finn Thain <fthain@linux-m68k.org>]

On Mon, 5 Jan 2026, Peter Zijlstra wrote:

> On Wed, Dec 31, 2025 at 07:25:42PM +1100, Finn Thain wrote:
> > From: Peter Zijlstra <peterz@infradead.org>
> > 
> > Add a Kconfig option for debug builds which logs a warning when an
> > instrumented atomic operation takes place that's misaligned.
> > Some platforms don't trap for this.
> > 
> > [fthain: added __DISABLE_BUG_TABLE macro.]
> > 
> > Cc: Sasha Levin <sashal@kernel.org>
> > Cc: Thomas Gleixner <tglx@linutronix.de>
> > Cc: Ingo Molnar <mingo@redhat.com>
> > Cc: Borislav Petkov <bp@alien8.de>
> > Cc: Dave Hansen <dave.hansen@linux.intel.com>
> > Cc: x86@kernel.org
> > Cc: Ard Biesheuvel <ardb@kernel.org>
> > Cc: "H. Peter Anvin" <hpa@zytor.com>
> > Link: https://lore.kernel.org/lkml/20250901093600.GF4067720@noisy.programming.kicks-ass.net/
> > Link: https://lore.kernel.org/linux-next/df9fbd22-a648-ada4-fee0-68fe4325ff82@linux-m68k.org/
> > Signed-off-by: Finn Thain <fthain@linux-m68k.org>
> > ---
> > Checkpatch.pl says...
> > ERROR: Missing Signed-off-by: line by nominal patch author 'Peter Ziljstra <peterz@infradead.org>'
> > ---
> > Changed since v5:
> >  - Add new __DISABLE_BUG_TABLE macro to prevent a build failure on those
> > architectures which use atomics in pre-boot code like the EFI stub loader:
> > 
> > x86_64-linux-gnu-ld: error: unplaced orphan section `__bug_table' from `arch/x86/boot/compressed/sev-handle-vc.o'
> 
> Urgh, so why not simply use __DISABLE_EXPORTS, that's typically (ab)used
> for these things?
> 

OK, I'll change it back to __DISABLE_EXPORTS.

> Also, unless __DISABLE_BUG_TABLE goes live inside asm/bug.h and kills
> all __bug_table emissions, its a misnomer.
> 

Yes, __DISABLE_BUG_TABLE is certainly a misnomer, since what it actually 
does is to elide code that would emit bug table entries. I would argue 
that this distinction is splitting hairs but it's moot.

> Furthermore, that SEV thing is broken and needs to be fixed anyway, this 
> isn't helping it much. noinstr code should not be using instrumented 
> things to begin with.
> 

The problem is not confined to x86. I needed something that would work for 
loongarch, arm, riscv etc.

Anyway, thanks for your review. I will make the necessary changes.