On Tue, 12 Dec 2023 11:49:20 -0500
Mathieu Desnoyers wrote:
> >> So the old "bottom" value is returned, which is wrong.
> >
> > Ah, OK that makes more sense. Yeah, if I had the three words from the
> > beginning, I would have tested to make sure they all match an not just the
> > two :-p
>
On 2023-12-11 23:38, Steven Rostedt wrote:
On Mon, 11 Dec 2023 22:51:04 -0500
Mathieu Desnoyers wrote:
[...]
For this first issue, here is the race:
rb_time_cmpxchg()
[...]
if (!rb_time_read_cmpxchg(&t->msb, msb, msb2))
return false;
if (!rb_time_read_cm
On Mon, 11 Dec 2023 22:51:04 -0500
Mathieu Desnoyers wrote:
> On 2023-12-11 17:59, Steven Rostedt wrote:
> > On Mon, 11 Dec 2023 15:13:24 -0500
> > Mathieu Desnoyers wrote:
> >
> >> Going through a review of the ring buffer rb_time functions for 32-bit
> >> architectures, I updated the commen
On 2023-12-11 17:59, Steven Rostedt wrote:
On Mon, 11 Dec 2023 15:13:24 -0500
Mathieu Desnoyers wrote:
Going through a review of the ring buffer rb_time functions for 32-bit
architectures, I updated the comments to match the code, and identified
the following issues:
Thanks Mathieu!
- rb_
On Mon, 11 Dec 2023 17:59:04 -0500
Steven Rostedt wrote:
> >
> > - A cmpxchg interrupted by 4 writes or cmpxchg overflows the counter
> > and produces corrupted time stamps. This is _not_ fixed by this patch.
>
> Except that it's not 4 bits that is compared, but 32 bits.
That should be "no
On Mon, 11 Dec 2023 15:13:24 -0500
Mathieu Desnoyers wrote:
> Going through a review of the ring buffer rb_time functions for 32-bit
> architectures, I updated the comments to match the code, and identified
> the following issues:
Thanks Mathieu!
>
> - rb_time_cmpxchg() needs to update the msb
Going through a review of the ring buffer rb_time functions for 32-bit
architectures, I updated the comments to match the code, and identified
the following issues:
- rb_time_cmpxchg() needs to update the msb last, so it matches
the validation of top and msb by __rb_time_read(). This is fixed by
7 matches
Mail list logo
