Messages in this thread | | | From | Thomas Gleixner <> | Subject | Re: [PATCH 2/3] watchdog/softlockup: report the most time-consuming hardirq | Date | Thu, 15 Feb 2024 02:02:45 +0100 |
| |
On Wed, Feb 14 2024 at 15:39, Doug Anderson wrote: > On Wed, Feb 14, 2024 at 3:36 PM Thomas Gleixner <tglx@linutronix.de> wrote: >> On Tue, Jan 23 2024 at 20:12, Bitao Hu wrote: >> As a side note: While C does not allow proper encapsulation it's a non >> starter to fiddle with the interrupt descriptor internals in random code >> just because the compiler allows you to do so. While not enforced there >> are clear boundaries and we went a long way to encapsulate this. > > I think you must have gotten dropped from all the future versions of > this patch series when Bitao took my advice and started using > interrupt counts instead of tracing. For what it's worth, the latest > version can be found at: > > https://lore.kernel.org/r/20240214021430.87471-1-yaoma@linux.alibaba.com
Yes. I'm not on CC and for some stupid reason it evaded my LKML filters.
Let me find that thread and stare at it.
Thanks,
tglx
| |