<!-- Some styling for better description lists --><style type='text/css'>dt { font-weight: bold;float: left;display:inline;margin-right: 1em} dd { display:block; margin-left: 2em}</style> hverkuil: <u>arnd</u>: will you look at the syzbot memset failure in v4l2-ioctl.c? I believe that memset fails for _IOC_NONE ioctls. <br> should be trivial to fix. arnd: <u>hverkuil</u>: sure, I'll have a look and send a fix dafna2: <u>hverkuil</u>: we discussed in the past a crash when opening a subdevice node while vimc is unregisterd, , this is the log http://ix.io/26Hk <br> ezequiel suggested this patch http://ix.io/247H it sets sd->v4l2_dev to NULL in the release cb instead of v4l2_device_unregister_subdev , this ensures that sd->v4l2_dev is non null when opening the subdevice node arnd: https://www.irccloud.com/pastebin/DMwaLVaZ/ <br> <u>hverkuil</u>: that would have been a bad bug to see in a release dafna2: does it looks like the right apporach to you? hverkuil: <u>dafna2</u>: I'll take a look this afternoon <br> <u>arnd</u>: I'll fast-track that fix once it is posted to the ML. arnd: <u>hverkuil</u>: sent now ezequielg: <u>bbrezillon</u>: pH5: just noticed commit 0fd7ada14eef merged a runtime_resume hook for imx8, but in the ended we didn't merge imx8... what are the plans for imx8? pH5: <u>ezequielg</u>: I wanted to make sure the power domain / vpumix bindings work for i.MX8MM, but the 8mm power domains are still up in the air. ezequielg: right. svarbanov: hverkuil, Hi, how you would like to proceed with this patch https://patchwork.linuxtv.org/patch/60999/ i.e. to be part of venus PULL request or part of your PULL requests? hverkuil: <u>svarbanov</u>: reviewed that patch, I think the documentation needs a bit more work. In any case, once I've acked it it can be merged via a venus PR. ***: benjiG has left koike: <u>hverkuil</u>: hi, I just submitted v13 of rockchip isp1 driver with what was being discussed regarding the bindings and also the other suggested modifications, I think it is good to go now.