linux-kernel - SCSI link power management patches

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:	Tue, 10 Jul 2007 21:42:39 -0400
From:	Valdis.Kletnieks@...edu
To:	Kristen Carlson Accardi <kristen.c.accardi@...el.com>
Cc:	jeff@...zik.org, linux-kernel@...r.kernel.org
Subject: SCSI link power management patches - a question..

I refitted the 4 patches you posted to a 22-rc6-mm1 kernel, and tried it on
my Dell Latitude D820 laptop, using the libata driver for the disk and DVD.
I got this:

[    0.702000] scsi0 : ata_piix
[    0.702000] scsi1 : ata_piix
[    0.702000] ata1: SATA max UDMA/133 cmd 0x00000000000101f0 ctl 0x00000000000103f6 bmdma 0x
000000000001bfa0 irq 14
[    0.702000] ata2: PATA max UDMA/100 cmd 0x0000000000010170 ctl 0x0000000000010376 bmdma 0x
000000000001bfa8 irq 15
[    0.857000] ata1.00: ATA-7: ST980825AS, 8.04, max UDMA/133
[    0.857000] ata1.00: 156301488 sectors, multi 8: LBA48 NCQ (depth 0/32)
[    0.864000] ata1.00: configured for UDMA/133
[    1.169000] ata2.00: ATAPI: PHILIPS DVD+/-RW SDVD8820, AD15, max UDMA/33
[    1.342000] ata2.00: configured for UDMA/33
....
[    1.342000]  sda: sda1 sda2
[    1.433000] sd 0:0:0:0: [sda] Attached SCSI disk
[    1.435000] scsi 1:0:0:0: CD-ROM            PHILIPS  DVD+-RW SDVD8820 AD15 PQ: 0 ANSI: 5
[    1.454000] sr0: scsi3-mmc drive: 24x/24x writer cd/rw xa/form2 cdda tray
[    1.454000] Uniform CD-ROM driver Revision: 3.20
[    1.454000] sr 1:0:0:0: Attached scsi CD-ROM sr0
[    1.454000] ata2.00: Unable to set Link PM policy

and /sys/class/scsi_host/host[01]/link_power_management_policy are both
stuck at 'max_performance'.  After reading the code,it *appears* that this
is because:

host0 - the 2 ports on the first ATA chain, and it's stuck because the DVD
doesn't support the required power management, and *all* devices on the chain
need to support it (as it didnt complain on ata1, the hard drive).

host1 - the second ATA chain, not connected to anything, so no PM available.

How do I double-check my analysis? Neither hdparm nor sdparm gave me anything
on /dev/dvd that told me either way, and I'd like to verify that the base
problem is "screwed by the hardware" before I give up....

Content of type "application/pgp-signature" skipped