Re: Flood of RIPs in DRM driver starting with 6.16-rc5

2025-07-08 Thread Steve French
just tried the fix. Don't see the warnings in dmesg anymore. thx On Tue, Jul 8, 2025 at 2:31 AM Simona Vetter wrote: > > On Mon, Jul 07, 2025 at 05:21:00PM -0500, Steve French wrote: > > I also see it manually building 6.16-rc4 from Saturday (not just > > seeing it with U

Re: Flood of RIPs in DRM driver starting with 6.16-rc5

2025-07-07 Thread Steve French
0121281): [] __irq_exit_rcu+0x158/0x180 [ 520.390136] ---[ end trace 00000000 ]--- On Mon, Jul 7, 2025 at 4:51 PM Steve French wrote: > > I see these RIP logged every second or two with 6.16-rc5 on boot, and > they continue indefinitely. I don't remember seeing these on rc4

Flood of RIPs in DRM driver starting with 6.16-rc5

2025-07-07 Thread Steve French
I see these RIP logged every second or two with 6.16-rc5 on boot, and they continue indefinitely. I don't remember seeing these on rc4. This happens when booting up, even when doing nothing on laptop (Lenovo P52, Ubuntu 24 running normal build of mainline 6.16-rc5) and they keep occurring. 6.16.0

[PATCH 4/7] cifs: use workqueue instead of slow-work

2010-07-21 Thread Steve French
On Tue, Jul 20, 2010 at 3:34 PM, Tejun Heo wrote: > Workqueue can now handle high concurrency. ?Use system_nrt_wq > instead of slow-work. > > * Updated is_valid_oplock_break() to not call cifs_oplock_break_put() > ?as advised by Steve French. ?It might cause deadlock. ?Instead,

Re: [PATCH 4/7] cifs: use workqueue instead of slow-work

2010-07-21 Thread Steve French
On Tue, Jul 20, 2010 at 3:34 PM, Tejun Heo wrote: > Workqueue can now handle high concurrency.  Use system_nrt_wq > instead of slow-work. > > * Updated is_valid_oplock_break() to not call cifs_oplock_break_put() >  as advised by Steve French.  It might cause deadlock.  Instead,