Bug 16609

Summary: Marvell SATA controller not correctly recognized (regression since 3.15)
Product: Mageia Reporter: Julien Moragny <julien.moragny>
Component: RPM PackagesAssignee: Kernel and Drivers maintainers <kernel>
Status: RESOLVED FIXED QA Contact:
Severity: normal    
Priority: Normal CC: jfbachelet, marja11, thanos17997, tmb
Version: 5Keywords: NEEDINFO, UPSTREAM
Target Milestone: ---   
Hardware: x86_64   
OS: Linux   
URL: https://bugzilla.kernel.org/show_bug.cgi?id=98171
Whiteboard:
Source RPM: kernel-3.19.8-3.mga5.src.rpm CVE:
Status comment:

Description Julien Moragny 2015-08-18 22:27:20 CEST 
Hi,

Since the installation of mga5, at each boot, I get an error message seemingly related to SATA (the section in the middle is in red in the log).
An internet search point to a regression with some Marvell controller since linux 3.15, https://bugzilla.kernel.org/show_bug.cgi?id=98171 . I have the message only once per boot and as nothing is plugged on the controller on this machine, I don't have any issue associated with it.


août 18 17:56:55 tue-amour kernel: ata9: SATA link down (SStatus 0 SControl 300)
août 18 17:56:55 tue-amour kernel: ata14: SATA link up <unknown> (SStatus FFFFFF93 SControl 300)
août 18 17:56:55 tue-amour kernel: ata14.00: ATAPI: MARVELL VIRTUALL, 1.09, max UDMA/66
août 18 17:56:55 tue-amour kernel: ata14.00: configured for UDMA/66
août 18 17:56:55 tue-amour kernel: ata1: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
août 18 17:56:55 tue-amour kernel: ata4: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
août 18 17:56:55 tue-amour kernel: ata1.00: ATA-8: SAMSUNG HD502HJ, 1AJ10001, max UDMA/133
août 18 17:56:55 tue-amour kernel: ata1.00: 976773168 sectors, multi 0: LBA48 NCQ (depth 31/32), AA
août 18 17:56:55 tue-amour kernel: ata1.00: configured for UDMA/133
août 18 17:56:55 tue-amour kernel: scsi 0:0:0:0: Direct-Access     ATA      SAMSUNG HD502HJ  0001 PQ: 0 ANSI: 5
août 18 17:56:55 tue-amour kernel: sd 0:0:0:0: [sda] 976773168 512-byte logical blocks: (500 GB/465 GiB)
août 18 17:56:55 tue-amour kernel: sd 0:0:0:0: [sda] Write Protect is off
août 18 17:56:55 tue-amour kernel: sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00
août 18 17:56:55 tue-amour kernel: sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
août 18 17:56:55 tue-amour kernel: ata4.00: ATA-8: WDC WD5000AAKS-00V1A0, 05.01D05, max UDMA/133
août 18 17:56:55 tue-amour kernel: ata4.00: 976773168 sectors, multi 0: LBA48 NCQ (depth 31/32), AA
août 18 17:56:55 tue-amour kernel: ata4.00: configured for UDMA/133
août 18 17:56:55 tue-amour kernel: scsi 3:0:0:0: Direct-Access     ATA      WDC WD5000AAKS-0 1D05 PQ: 0 ANSI: 5
août 18 17:56:55 tue-amour kernel: sd 3:0:0:0: [sdb] 976773168 512-byte logical blocks: (500 GB/465 GiB)
août 18 17:56:55 tue-amour kernel: sd 3:0:0:0: [sdb] Write Protect is off
août 18 17:56:55 tue-amour kernel: sd 3:0:0:0: [sdb] Mode Sense: 00 3a 00 00
août 18 17:56:55 tue-amour kernel: sd 3:0:0:0: [sdb] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
août 18 17:56:55 tue-amour kernel: scsi 13:0:0:0: Processor         Marvell  91xx Config      1.01 PQ: 0 ANSI: 5

août 18 17:56:55 tue-amour kernel: ata14.00: exception Emask 0x1 SAct 0x0 SErr 0x0 action 0x6
août 18 17:56:55 tue-amour kernel: ata14.00: irq_stat 0x40000001
août 18 17:56:55 tue-amour kernel: scsi 13:0:0:0: CDB: 
août 18 17:56:55 tue-amour kernel: Inquiry: 12 01 00 00 ff 00
août 18 17:56:55 tue-amour kernel: ata14.00: cmd a0/01:00:00:00:01/00:00:00:00:00/a0 tag 2 dma 16640 in
                                             res 00/00:00:00:00:00/00:00:00:00:00/00 Emask 0x3 (HSM violation)
août 18 17:56:55 tue-amour kernel: ata14: hard resetting link

août 18 17:56:55 tue-amour kernel:  sdb: sdb1 sdb2 < sdb5 >
août 18 17:56:55 tue-amour kernel: sd 3:0:0:0: [sdb] Attached SCSI disk
août 18 17:56:55 tue-amour kernel:  sda: sda1 sda2 < sda5 sda6 sda7 >
août 18 17:56:55 tue-amour kernel: sd 0:0:0:0: [sda] Attached SCSI disk
août 18 17:56:55 tue-amour kernel: ata14: SATA link up <unknown> (SStatus FFFFFF93 SControl 300)
août 18 17:56:55 tue-amour kernel: ata14.00: configured for UDMA/66
août 18 17:56:55 tue-amour kernel: ata14: EH complete



Hardware :
Gigabyte GA-770TA-UD3 with a Marvell 9128 chip
Mageia 5 x86_64
kernel 3.19.8-desktop-3.mga5 


Same bug on archlinux :
https://bugs.archlinux.org/task/40776

thanks
Julien

Reproducible: 

Steps to Reproduce:
Julien Moragny 2015-08-18 22:28:02 CEST

See Also: (none) => https://bugzilla.kernel.org/show_bug.cgi?id=98171

Thierry Vignaud 2015-08-24 10:49:04 CEST

URL: (none) => https://bugzilla.kernel.org/show_bug.cgi?id=98171
CC: (none) => tmb
See Also: https://bugzilla.kernel.org/show_bug.cgi?id=98171 => (none)
Summary: Marvell SATA controller not correctly recognized => Marvell SATA controller not correctly recognized (regression since 3.15)

Comment 1 Jean-François Bachelet 2015-09-08 21:24:25 CEST 
get continuous "HSM Violation error" on a new install of Mageia5 32 bits from an Intel 82801EB (ICH5) SATA controler too.

There is two 250GB Seagate disk on the two SATA channels (as unit ata0 and ata1) and one Ricoh DVD Rom  (as IDE unit ata2)

that's the Ricoh that spit that error continuously  :

ata2.00: status: { DRDY DRQ ERR }
ata2.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen
ata2.00: ST_First: DRQ=1 with device error, dev_stat0x59
ata2.00: cmd a0/00:00:00:24:00/00:00:00:00:00/a0 tag 0 pio 36 in
         res 59/00:01:00:24:00/00:00:00:00:00/a0 Emask 0x2 (HSM violation)

but cease to spit errors when a disc is present in the drive.

CC: (none) => jfbachelet

Comment 2 Jean-François Bachelet 2015-09-09 13:46:02 CEST 
> but cease to spit errors when a disc is present in the drive.

Must add that if the inserted disk is blank the errors continue to appear :-( only a valid disk makes the error to stop.
Comment 3 Marja Van Waes 2016-11-03 11:45:12 CET 
The upstream report is still open.

@ Julien

Can you confirm it is still valid?

Keywords: (none) => NEEDINFO, UPSTREAM
CC: (none) => marja11
Assignee: bugsquad => kernel

Comment 4 Julien Moragny 2016-11-03 18:41:46 CET 
Hi,

Yes, it's still valid with kernel 4.4.26.
Linux tue-amour 4.4.26-desktop-1.mga5 #1 SMP Thu Oct 20 09:30:12 UTC 2016 x86_64 x86_64 x86_64 GNU/Linux

I will try to test with latest cauldron kernel.

regards
Julien
Comment 5 Julien Moragny 2016-11-03 18:54:34 CET 
Hi²,

Just tested with kernel-desktop-4.8.6-1 from cauldron (I'm still on mga5) and it seems to be fixed, no error message and the controller appears correctly detected in the logs :-)
(again, I can't test as I don't have anything plugged on it).

regards
Julien
Comment 6 Julien Moragny 2018-02-20 21:30:06 CET 
As I can't test on mga5 anymore and it's fixed on mga6, closing the bug.

Resolution: (none) => FIXED
Status: NEW => RESOLVED

play game 2019-03-28 07:03:37 CET

CC: (none) => thanos17997