Re: Minor bug in SCSI definition

From: Alexander Motin <mav_at_FreeBSD.org>
Date: Thu, 13 Nov 2014 20:16:05 +0200
On 13.11.2014 05:44, NGie Cooper wrote:
> On Wed, Nov 12, 2014 at 7:30 PM, Rang, Anton <anton.rang_at_isilon.com> wrote:
>> Coverity found an issue in this area which I tracked down to the incorrect definition patched below.
>>
>> The SID_QUAL macro is (((inq_data)->device & 0xE0) >> 5) which extracts the peripheral qualifier.
>> Per SCSI-2 (draft 10L) table 46, the vendor-specific values are "1XXb".
>>
>> This probably affects almost nobody, but it will clear up a couple of Coverity warnings.
>>
>> Anton
>>
>> Index: sys/cam/scsi/scsi_all.h
>> ===================================================================
>> --- sys/cam/scsi/scsi_all.h     (revision 274352)
>> +++ sys/cam/scsi/scsi_all.h  (working copy)
>> _at__at_ -1817,7 +1817,7 _at__at_
>>                                                             * reserved for this peripheral
>>                                                             * qualifier.
>>                                                             */
>> -#define          SID_QUAL_IS_VENDOR_UNIQUE(inq_data) ((SID_QUAL(inq_data) & 0x08) != 0)
>> +#define         SID_QUAL_IS_VENDOR_UNIQUE(inq_data) ((SID_QUAL(inq_data) & 0x04) != 0)
>>            u_int8_t dev_qual2;
>> #define          SID_QUAL2  0x7F
>> #define          SID_LU_CONG        0x40
> 
> CCing ken_at_/mav_at_/scottl_at_ -- thanks!

Looks good to me. Committed it to head at r274477. Thank you!

-- 
Alexander Motin
Received on Thu Nov 13 2014 - 17:16:07 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:40:54 UTC