05:28:58 Hey look, I got ::bcmp working, I think... https://gist.github.com/danmcd/9945ae95e4a3db60c87a53fe86c88f02 05:40:51 Nice! I enjoy writing dcmds. 05:41:40 What does it print in a pipe context, the addr+offset? 05:42:20 test_string_1::bcmp test_string_2 -n a | /s 13:14:49 danmcd: and thank you for documenting the flags!! 13:15:03 (i've fixed several dcmds that didn't and just assumed you knew what they did) 14:58:45 @andyf ==> probably '/s' which doesn't help. 14:59:10 There's a WIP gerrit attached to the bug id now. 14:59:20 Well it would require an ::eval - I think it's useful to document the pipe behaviour whichever one you choose. 14:59:28 @jbk YW. Least I can do. 15:00:25 In the spirit of "keep it simple, no bikeshedding" I think people should prep for using ::eval, I guess. 15:03:01 Yep, I don't want to stop it being simple. It's trivial to implement either way (you can choose to print something different if (flags & DCMD_PIPE_OUT)) -- I would just like to see it in the help so we know. 15:04:03 I'm happy to take a review pass whenever you like - promise not to focus on the colour too much. 15:06:01 * danmcd notes needs-better-documentation. 15:06:37 Also I'm willing to wiggle on pointer vs. offset for the output, but I'll allow that explicitly on Gerrit when I'm happy with it going from WIP to review-me-please. 17:44:55 Advice on how `addr2` can be parsed as either a name OR an actual address? 17:52:15 Ah, is it always MDB_TYPE_STRING ? 17:56:22 ::dis and ::findbuf have examples, neither of them pretty.. 17:56:29 er, ::findsym even 17:56:47 https://github.com/illumos/illumos-gate/blob/master/usr/src/cmd/mdb/common/mdb/mdb_cmds.c#L1955-L1970 18:00:00 Yeah, look at dis_str2addr() 19:09:52 Thank you @andyf. Need to push it to gerrit again. 21:39:52 fenix illumos#17506 21:39:53 FEATURE 17506: Create ::bcmp dcmd (In Progress) 21:39:53 ↳ https://www.illumos.org/issues/17506 | https://code.illumos.org/c/illumos-gate/+/4326