<!-- 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> sailus: <u>hverkuil</u>: Just an idea --- the pad check could be embedded to v4l2_subdec_call for particular ops. <br> That would make the macro largish but it'd improve robustness of the driver implementation. <br> We only validate the arguments that come from user space, not from other drivers. <br> And the implementation of that would need no driver changes at all. :-) hverkuil: you'd also have to modify the macros v4l2-device.h and I'd rather not touch those. <br> I think such checks would get messy very quickly. <br> I'll repost a v2 soon. <br> repost -> post sailus: Wouldn't you call the subdev_do_ioctl() function messy as well? <br> The macro could be split into multiple smaller macros. <br> That's entirely unrelated to the patchset at hand, I have to say. <br> Just an idea. jmondi: <u>hverkuil</u>: sorry, I've missed your email from this morning and commented on the old implementation of v4l2_g/s_param :) mchehab: <u>sailus</u>: moikka! -: headless wonders where sailus lived in .fi headless: <u>sailus</u>: sorry about being overly friendly in Prague :-) <br> *live <br> *lives, dammit ***: awalls1 has left