|Summary:||[mfi] Logic issue in mfi_tbolt.c|
|Component:||kern||Assignee:||freebsd-bugs mailing list <bugs>|
|Severity:||Affects Only Me||CC:||ambrisko, dcb314, emaste|
Description swildner 2014-03-21 21:10:00 UTC
In sys/dev/mfi/mfi_tbolt.c, line #1111, the "if (cdb != 0x28 || cdb != 0x2A)" will be always true.
Comment 1 David Binderman 2016-02-08 09:08:20 UTC
Still broken, nearly two years later.
Comment 2 Ed Maste 2016-02-11 19:46:09 UTC
Introduced in r227068. On a quick inspection it's probably supposed to be if (cdb != 0x28 && cdb != 0x2A) but I'm not completely familiar with this driver and don't have hardware to test -- adding abrisko@ to CC as he may be able to take care of this.
Comment 3 Doug Ambrisko 2016-03-21 20:59:42 UTC
This is more complicated and apparently something that I wasn't sure of from the LSI code. It could help fix some other syspd type bugs with newer controllers. Adding code to deal with this didn't just work in my first attempt so I'm going to have to debug this more. My guess is that this needs to handle the various size SCSI I/O as well and not just READ10/WRITE10.
Comment 4 Eitan Adler 2018-05-20 23:53:45 UTC
For bugs matching the following conditions: - Status == In Progress - Assignee == "bugs@FreeBSD.org" - Last Modified Year <= 2017 Do - Set Status to "Open"
Comment 5 David Binderman 2018-07-05 20:11:36 UTC
Still broken a couple of years later.