On Thu, 26 Jun 2025 18:04:59 +0200 Nam Cao <nam...@linutronix.de> wrote:
> > The above three interact with each other. Without the barriers, the > > tr->buffer_disabled = 0, can be set on one CPU, and the other CPU can think > > the buffer is still enabled and do work that will end up doing nothing. Or > > it can be set to 1, and the other CPU still sees it disabled and will not > > do work when it can. > > (I'm not that experienced with memory barrier, so I may be writing nonsense > here) So I did some git archeology and realized that the buffer_disabled is just a cache to make irqsoff tracing faster. The commit even said it's not going to be accurate. OK, I'll just take this patch. Thanks, -- Steve
