On Tue, 18 May 2021 08:53:58 +0900, Hyeonggon Yoo said: > I recently was tracking syzbot report that is occured in > sound driver. I don't have the device, but was curious about > what the test method is. Depends what the bug was. > well, build tests are useful but it is not enough, we should run the > code on actual device. is it possible to make the bot test the device > (which I don't have) even if there is no syzbot reproducer? Bugs that are obvious logic errors (use after free, locking issues, failure to build because a previous commit didn't cover all cases, and so on) can often be either tested or proven correct without the actual device being available. I've lost count how many times I've reported that some device driver or other simply doesn't even *compile* because an API was changed but incomplete - stuff that a 'make allmodconfig' will find.
Attachment:
pgp_6_34oJ1L9.pgp
Description: PGP signature
_______________________________________________ Kernelnewbies mailing list Kernelnewbies@xxxxxxxxxxxxxxxxx https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
