Re: [PATCH v2] RISC-V: Fix /proc/cpuinfo cpumask warning

From: Andrew Jones
Date: Wed Oct 12 2022 - 09:13:48 EST


On Wed, Oct 12, 2022 at 05:55:29AM -0700, Yury Norov wrote:
> On Wed, Oct 12, 2022 at 10:29:49AM +0200, Andrew Jones wrote:
> > Commit 78e5a3399421 ("cpumask: fix checking valid cpu range") has
> > started issuing warnings[*] when cpu indices equal to nr_cpu_ids - 1
> > are passed to cpumask_next* functions. seq_read_iter() and cpuinfo's
> > start and next seq operations implement a pattern like
> >
> > n = cpumask_next(n - 1, mask);
> > show(n);
> > while (1) {
> > ++n;
> > n = cpumask_next(n - 1, mask);
> > if (n >= nr_cpu_ids)
> > break;
> > show(n);
> > }
>
> Can you instead of sudo-code print show the real control flow? What
> function hosts the infinite loop?

The function is seq_read_iter(), which is pointed out above. I'd rather
not reproduce / describe more than what I've done here, as the function
is large. I'd be happy for reviewers to double check my pseudocode to
make sure I got it and the analysis right, though.

>
> > which will issue the warning when reading /proc/cpuinfo. Ensure no
> > warning is generated by validating the cpu index before calling
> > cpumask_next().
> >
> > [*] Warnings will only appear with DEBUG_PER_CPU_MAPS enabled.
> >
> > Signed-off-by: Andrew Jones <ajones@xxxxxxxxxxxxxxxx>
> > Cc: Yury Norov <yury.norov@xxxxxxxxx>
> > ---
> > v2:
> > - Got comments on the x86 equivalent patch and made the same
> > changes to this one
> > - Added all the information I should have in the first place
> > to the commit message [Boris]
> > - Changed style of fix [Boris]
> >
> >
> > arch/riscv/kernel/cpu.c | 3 +++
> > 1 file changed, 3 insertions(+)
> >
> > diff --git a/arch/riscv/kernel/cpu.c b/arch/riscv/kernel/cpu.c
> > index 4aa8cd749441..63138b880b92 100644
> > --- a/arch/riscv/kernel/cpu.c
> > +++ b/arch/riscv/kernel/cpu.c
> > @@ -166,6 +166,9 @@ static void print_mmu(struct seq_file *f)
> >
> > static void *c_start(struct seq_file *m, loff_t *pos)
> > {
> > + if (*pos >= nr_cpu_ids)
> > + return NULL;
> > +
> > *pos = cpumask_next(*pos - 1, cpu_online_mask);
> > if ((*pos) < nr_cpu_ids)
> > return (void *)(uintptr_t)(1 + *pos);
>
> OK, as far as I understood your explanations, *pos == nr_cpu_ids
> is a valid index because it's used as stop-code for traversing.
>
> However, you're completely silencing cpumask_check(), including
> those cases where *pos > nr_cpu_ids. I suspect there's no valid
> cases for it. If so, the patch should look like:
>
> + if (*pos == nr_cpu_ids)
> + return NULL;
> +

That makes sense and it's probably worth a v3. I'll wait and see if more
comments roll in before sending though.

>
> The same for x86 patch.
>
> If it comes to v3, can you send both as a series?

OK. I'll write a cover letter trying to explain that I don't expect them
to both go through the same tree.

Thanks,
drew