Skip site navigation (1)Skip section navigation (2)
Date:      15 Oct 1999 09:05:45 -0000
From:      nnd@mail.nsk.ru
To:        FreeBSD-gnats-submit@freebsd.org
Subject:   kern/14345: Version 1.27 of the 'src/sys/dev/ata/ata-all.c' prevents HPT366 matching
Message-ID:  <19991015090545.29193.qmail@infoteka.nsk.ru>

next in thread | raw e-mail | index | archive | help

>Number:         14345
>Category:       kern
>Synopsis:       Version 1.27 of the 'src/sys/dev/ata/ata-all.c' prevents HPT366 matching
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    freebsd-bugs
>State:          open
>Quarter:        
>Keywords:       
>Date-Required:
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Fri Oct 15 02:10:00 PDT 1999
>Closed-Date:
>Last-Modified:
>Originator:     Nickolay N. Dudorov
>Release:        FreeBSD-current i386
>Organization:
STE Infoteka
>Environment:

	FreeBSD-current after 1999/10/13 11:56:50 PDT

>Description:

	Doug Rabson's commit to the file in question with the
  Log:
  Don't match non-IDE devices in probe.

prevents at least HighPoint HPT366 IDE controller from matching.
Kernel with version 1.26 file successfully recognizes and works
with the disk on this controller on the Abit's BP6 motherboard.
The reason for this is that motherbord (or PCI-BIOS ?) reports
this controller class as 'class=01-80-00' i.e. class=PCIC_STORGE
ans subclass=PCIS_OTHER. This does'nt prevent 'ata_pciattach'
from attaching HPT366 controller.

	Loking at the diff between -r1.26 and -r1.27 I dont
understand the purpose of the patch at all. Additional test
at the beginning of the 'ata_pcimatch' function can only return
NULL where previous version return non-NULL for one of the
"supported chipsets" or "unsupported but known chipsets".
But if there is such a chipset - is'nt it better to change
it's 'case' instead of the general (pre)condition ?

>How-To-Repeat:

	Take an FreeBSD-current system with HPT366 controller
and kernels with the -r1.26 and -r1.27 of the 'src/sys/dev/ata/ata_all.c'.
Then the first one will work with this controller and the second
not recognized it at all.


>Fix:
	
	Revert this file to the -r1.26


>Release-Note:
>Audit-Trail:
>Unformatted:


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-bugs" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19991015090545.29193.qmail>