AMD 890GX vs 6Gbps SATA?

AMD 890GX vs 6Gbps SATA?

[James Cloos]

All of the ad copy and doc I can find claim that the sata controller in
the 890gx supports 6Gbps sata3.  But I still get this:

ahci 0000:00:11.0: AHCI 0001.0200 32 slots 6 ports 3 Gbps 0x3f impl SATA mode
ata2: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
ata2.00: ATA-8: WDC WD1002FAEX-xxxxxx, 05.xxxxx, max UDMA/133

I hadn't noticed before because that box's first drive only supports 3 Gbps.

Does the 6Gpbs mode require that all of the disks are sata3?  (That would
be /most/ annoying, given that optical drives all seem to be 1.5Gbps.)

Did MSI screw something up on the board?

Or is the ad copy exaggerative?

I took a look though the bios config; I didn't find anything relevant.

-JimC
-- 
James Cloos <cloos@jhcloos.com>         OpenPGP: 1024D/ED7DAEA6

Re: AMD 890GX vs 6Gbps SATA?

[Jeff Garzik]

On 11/14/2010 12:50 PM, James Cloos wrote:
> All of the ad copy and doc I can find claim that the sata controller in
> the 890gx supports 6Gbps sata3.  But I still get this:
>
> ahci 0000:00:11.0: AHCI 0001.0200 32 slots 6 ports 3 Gbps 0x3f impl SATA mode
> ata2: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
> ata2.00: ATA-8: WDC WD1002FAEX-xxxxxx, 05.xxxxx, max UDMA/133
>
> I hadn't noticed before because that box's first drive only supports 3 Gbps.
>
> Does the 6Gpbs mode require that all of the disks are sata3?  (That would
> be /most/ annoying, given that optical drives all seem to be 1.5Gbps.)

No, on SATA each link is entirely independent of the other links 
(ignoring port multipliers).

The "3 Gbps" maximum you see on the first line comes directly from what 
capabilities the chip reports, so perhaps the firmware/BIOS is not 
initializing the chip properly.

AHCI 1.2 hardware, the AHCI driver and libata core should all support 
operation at 6 Gbps.

	Jeff

RE: AMD 890GX vs 6Gbps SATA?

[Huang, Shane]

Hi James,

Can you send us your full dmesg and lspci?

I submitted one patch to support SATA AHCI 6Gbps message
in 2008, does your kernel contain it?

http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commi
t;h=8522ee25f3a645577d41e71328cd4fcf8610dfeb


Thanks,
Shane

----- Forwarded message from James Cloos <cloos@jhcloos.com> -----

Date: Sun, 14 Nov 2010 12:50:41 -0500
From: James Cloos <cloos@jhcloos.com>
To: linux-ide@vger.kernel.org
Subject: AMD 890GX vs 6Gbps SATA?
User-Agent: Gnus/5.110011 (No Gnus v0.11) Emacs/24.0.50 (gnu/linux)

All of the ad copy and doc I can find claim that the sata controller in
the 890gx supports 6Gbps sata3.  But I still get this:

ahci 0000:00:11.0: AHCI 0001.0200 32 slots 6 ports 3 Gbps 0x3f impl SATA
mode
ata2: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
ata2.00: ATA-8: WDC WD1002FAEX-xxxxxx, 05.xxxxx, max UDMA/133

I hadn't noticed before because that box's first drive only supports 3
Gbps.

Does the 6Gpbs mode require that all of the disks are sata3?  (That
would
be /most/ annoying, given that optical drives all seem to be 1.5Gbps.)

Did MSI screw something up on the board?

Or is the ad copy exaggerative?

I took a look though the bios config; I didn't find anything relevant.

-JimC

Re: AMD 890GX vs 6Gbps SATA?

[James Cloos]

[-- Attachment #1: Type: text/plain, Size: 3792 bytes --]

>>>>> "HS" == Huang, Shane <Shane.Huang@amd.com> writes:

HS> Can you send us your full dmesg and lspci?

lspci is:

00:00.0 Host bridge: Advanced Micro Devices [AMD] RS880 Host Bridge
00:01.0 PCI bridge: Advanced Micro Devices [AMD] RS780 PCI to PCI bridge (int gfx)
00:05.0 PCI bridge: Advanced Micro Devices [AMD] RS780 PCI to PCI bridge (PCIE port 1)
00:06.0 PCI bridge: Advanced Micro Devices [AMD] RS780 PCI to PCI bridge (PCIE port 2)
00:11.0 SATA controller: ATI Technologies Inc SB700/SB800 SATA Controller [AHCI mode] (rev 40)
00:12.0 USB Controller: ATI Technologies Inc SB700/SB800 USB OHCI0 Controller
00:12.2 USB Controller: ATI Technologies Inc SB700/SB800 USB EHCI Controller
00:13.0 USB Controller: ATI Technologies Inc SB700/SB800 USB OHCI0 Controller
00:13.2 USB Controller: ATI Technologies Inc SB700/SB800 USB EHCI Controller
00:14.0 SMBus: ATI Technologies Inc SBx00 SMBus Controller (rev 41)
00:14.2 Audio device: ATI Technologies Inc SBx00 Azalia (Intel HDA) (rev 40)
00:14.3 ISA bridge: ATI Technologies Inc SB700/SB800 LPC host controller (rev 40)
00:14.4 PCI bridge: ATI Technologies Inc SBx00 PCI to PCI Bridge (rev 40)
00:14.5 USB Controller: ATI Technologies Inc SB700/SB800 USB OHCI2 Controller
00:15.0 PCI bridge: ATI Technologies Inc Device 43a0
00:16.0 USB Controller: ATI Technologies Inc SB700/SB800 USB OHCI0 Controller
00:16.2 USB Controller: ATI Technologies Inc SB700/SB800 USB EHCI Controller
00:18.0 Host bridge: Advanced Micro Devices [AMD] Family 10h Processor HyperTransport Configuration
00:18.1 Host bridge: Advanced Micro Devices [AMD] Family 10h Processor Address Map
00:18.2 Host bridge: Advanced Micro Devices [AMD] Family 10h Processor DRAM Controller
00:18.3 Host bridge: Advanced Micro Devices [AMD] Family 10h Processor Miscellaneous Control
00:18.4 Host bridge: Advanced Micro Devices [AMD] Family 10h Processor Link Control
01:05.0 VGA compatible controller: ATI Technologies Inc RS880 [Radeon HD 4290]
01:05.1 Audio device: ATI Technologies Inc RS880 Audio Device [Radeon HD 4200]
02:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 03)
03:00.0 USB Controller: NEC Corporation uPD720200 USB 3.0 Host Controller (rev 03)
04:05.0 PCI bridge: Digital Equipment Corporation DECchip 21152 (rev 03)
05:04.0 Ethernet controller: Intel Corporation 82557/8/9/0/1 Ethernet Pro 100 (rev 05)
05:05.0 Ethernet controller: Intel Corporation 82557/8/9/0/1 Ethernet Pro 100 (rev 05)
06:00.0 IDE interface: JMicron Technology Corp. JMB368 IDE controller

The verbose output for 00:11.0 is:

00:11.0 SATA controller: ATI Technologies Inc SB700/SB800 SATA Controller [AHCI mode] (rev 40) (prog-if 01 [AHCI 1.0])
        Subsystem: Micro-Star International Co., Ltd. Device 7642
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B- DisINTx+
        Status: Cap+ 66MHz+ UDF- FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 64, Cache Line Size: 64 bytes
        Interrupt: pin A routed to IRQ 40
        Region 0: I/O ports at 9000 [size=8]
        Region 1: I/O ports at 8000 [size=4]
        Region 2: I/O ports at 7000 [size=8]
        Region 3: I/O ports at 6000 [size=4]
        Region 4: I/O ports at 5000 [size=16]
        Region 5: Memory at fe2ffc00 (32-bit, non-prefetchable) [size=1K]
        Capabilities: [50] MSI: Enable+ Count=1/8 Maskable- 64bit+
                Address: 00000000fee0f00c  Data: 4159
        Capabilities: [70] SATA HBA v1.0 InCfgSpace
        Capabilities: [a4] PCI Advanced Features
                AFCap: TP+ FLR+
                AFCtrl: FLR-
                AFStatus: TP-
        Kernel driver in use: ahci

And the dmesg, compressed to fit, is attached:


[-- Attachment #2: Gzip(1)ed dmesg --]
[-- Type: application/octet-stream, Size: 12099 bytes --]

[-- Attachment #3: Type: text/plain, Size: 612 bytes --]


(Heh.  The gzip is smaller than the bzip2.)

(The carbon1 in the kernel version is a .config tracking localversion.)

HS> I submitted one patch to support SATA AHCI 6Gbps message
HS> in 2008, does your kernel contain it?

HS> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commi
HS> t;h=8522ee25f3a645577d41e71328cd4fcf8610dfeb

I currently run the v2.6.36 tag on that box; newer kernels have a pty
deadlock (in bz).  That commit is in the tree, but the function it
changed has since moved from ahci.c to libahci.c.

-JimC
-- 
James Cloos <cloos@jhcloos.com>         OpenPGP: 1024D/ED7DAEA6

Re: AMD 890GX vs 6Gbps SATA?

[James Cloos]

>>>>> "JG" == Jeff Garzik <jeff@garzik.org> writes:

JG> No, on SATA each link is entirely independent of the other links
JG> (ignoring port multipliers).

That is what I thought.

JG> The "3 Gbps" maximum you see on the first line comes directly from
JG> what capabilities the chip reports, so perhaps the firmware/BIOS is
JG> not initializing the chip properly.

And that is what I was afraid of. ☹

JG> AHCI 1.2 hardware, the AHCI driver and libata core should all support
JG> operation at 6 Gbps.

Again, what I thought.

-JimC
-- 
James Cloos <cloos@jhcloos.com>         OpenPGP: 1024D/ED7DAEA6

RE: AMD 890GX vs 6Gbps SATA?

[Huang, Shane]

Hi James:

> lspci is:
> ...
> And the dmesg, compressed to fit, is attached.

Are your SATA HDDs(WDC WD6401AALS-00J7B1,
WD1002FAEX-00Z3A0) also supporting 6Gbps?

Only the support from host controller is not enough...


Thanks,
Shane

Re: AMD 890GX vs 6Gbps SATA?

[James Cloos]

>>>>> "HS" == Huang, Shane <Shane.Huang@amd.com> writes:

HS> Are your SATA HDDs(WDC WD6401AALS-00J7B1,
HS> WD1002FAEX-00Z3A0) also supporting 6Gbps?

As I at least implied in the original post, while the 640 is only a
3 Gbps drive, WDC says the WD1002FAEX is a 6 Gbps.

(And, before anyone asks, no jumpers -- such as the one which limits
it to 3 Gbps -- are installed.)

-JimC
-- 
James Cloos <cloos@jhcloos.com>         OpenPGP: 1024D/ED7DAEA6

Re: AMD 890GX vs 6Gbps SATA?

[Stan Hoeppner]

James Cloos put forth on 11/15/2010 12:50 PM:
>>>>>> "SH" == Stan Hoeppner <stan@hardwarefreak.com> writes:
> 
> SH> What does "hdparm -I /dev/sdx" say?

>            *    Gen3 signaling speed (6.0Gb/s)

That's what I was looking for--confirms it's def a 6Gb drive.  Apologies
for my messed up reply chain.  I'm so used to TB's "reply-to-list"
option I frequently forget to "reply-to-all" to Linux kernel list mail...

-- 
Stan

Re: AMD 890GX vs 6Gbps SATA?

[James Cloos]

>>>>> "SH" == Stan Hoeppner <stan@hardwarefreak.com> writes:

>> *    Gen3 signaling speed (6.0Gb/s)

SH> That's what I was looking for--confirms it's def a 6Gb drive.

Yes.

And thanks for the reminder that hdparm -I is no longer a raw (and
wrong-endian) dump of the -i info.  I'd forgotten that.  It even
works on my usb-storage backup drive, where hdparm -i fails.

-JimC
-- 
James Cloos <cloos@jhcloos.com>         OpenPGP: 1024D/ED7DAEA6

RE: AMD 890GX vs 6Gbps SATA?

[Huang, Shane]

Hi James,

> As I at least implied in the original post, while the 640 is only a
> 3 Gbps drive, WDC says the WD1002FAEX is a 6 Gbps.
> 
> (And, before anyone asks, no jumpers -- such as the one which limits
> it to 3 Gbps -- are installed.)

Your south bridge chip is SB800 which should support SATA 6Gbps.

Quoting your dmesg:
> ahci 0000:00:11.0: AHCI 0001.0200 32 slots 6 ports 3 Gbps
> 0x3f impl SATA mode

Quoting AHCI spec, HBA CAP.ISS(RO):
> Interface Speed Support (ISS): Indicates the maximum speed
> the HBA can support on its ports. These encodings match the
> system software programmable PxSCTL.DET.SPD field. Values are:
> Bits Definition
> 0000 Reserved
> 0001 Gen 1 (1.5 Gbps)
> 0010 Gen 2 (3 Gbps)
> 0011 Gen 3 (6 Gbps)
> 0100 - 1111 Reserved

Since your CAP.ISS is set to 0010(3Gbps), most likely you'll
have to ask your BIOS vendor to enable 6Gbps in one release.
Some BIOS vendors may set the max speed to Gen2 to save power.


Thanks,
Shane

Re: AMD 890GX vs 6Gbps SATA?

[Mark Lord]

On 10-11-15 02:40 PM, James Cloos wrote:
>>>>>> "SH" == Stan Hoeppner<stan@hardwarefreak.com>  writes:
>
>>> *    Gen3 signaling speed (6.0Gb/s)
>
> SH>  That's what I was looking for--confirms it's def a 6Gb drive.
>
> Yes.
>
> And thanks for the reminder that hdparm -I is no longer a raw (and
> wrong-endian) dump of the -i info.  I'd forgotten that.  It even
> works on my usb-storage backup drive, where hdparm -i fails.

Heh heh.

Just for historical background, the reason they work(ed) that way,
is hdparm was originally written (by me) for testing the new Linux IDE
driver subsystem.  Well, okay, it *was* "new" back in the mid-90s.  :)

The -i reported what the device driver probed from the drive at boot time,
and -I always interrogated the drive itself, and displayed the raw data
exactly as the drive reported it.  Thus, for some early drives, the data
looked good, and for others it was byte-swapped.

Eventually ATA standardization kicked in, and all modern drives now agree
on the correct byte-ordering, so I changed -I to give human readable output.  :)

Cheers

Re: AMD 890GX vs 6Gbps SATA?

[Stan Hoeppner]

Huang, Shane put forth on 11/16/2010 4:04 AM:

> Since your CAP.ISS is set to 0010(3Gbps), most likely you'll
> have to ask your BIOS vendor to enable 6Gbps in one release.
> Some BIOS vendors may set the max speed to Gen2 to save power.

Unless you have a PMP on the other end of that SATA link with a bunch of
striped drives, or an expensive SSD, you won't notice a performance
difference between 3 Gb/s and 6 Gb/s link speed anyway:  no single mech
drive on the planet can come close to pushing 300 MB/s let alone 600.
The fastest 15k SAS drives are peaking at around 200 MB/s IIRC.  And
there are few [affordable] SSDs that will saturate a SATA II link, let
alone a SATA III link.

My advice, FWIW, is to simply ignore this, unless you plan on attaching
a device in the near future that can actually exceed 300 MB/s.
Currently the only way to do this is with striped mech drives on a PMP,
or with a $500+ SSD.

-- 
Stan

Re: AMD 890GX vs 6Gbps SATA?

[Stan Hoeppner]

Stan Hoeppner put forth on 11/16/2010 6:31 PM:

> Currently the only way to do this is with striped mech drives on a PMP,
> or with a $500+ SSD.

Let me correct that slightly.  There are also host (OS) independent
external RAID enclosures on the market with a real embedded RAID
controller, and host connection via single or multiple lane eSATA.  If
using a single 3 Gb/s eSATA link with such an enclosure, with enough
disks, you could easily saturate a 3 Gb/s link.  Some of these may even
saturate a single 6 Gb/s link depending on drive count and quality of
the RAID controller.

-- 
Stan

RE: AMD 890GX vs 6Gbps SATA?

[Huang, Shane]

James & Stan,


> > Since your CAP.ISS is set to 0010(3Gbps), most likely you'll
> > have to ask your BIOS vendor to enable 6Gbps in one release.
> > Some BIOS vendors may set the max speed to Gen2 to save power.
> 
> Unless you have a PMP on the other end of that SATA link with a bunch
> of
> striped drives, or an expensive SSD, you won't notice a performance
> difference between 3 Gb/s and 6 Gb/s link speed anyway:  no single
mech
> drive on the planet can come close to pushing 300 MB/s let alone 600.
> The fastest 15k SAS drives are peaking at around 200 MB/s IIRC.  And
> there are few [affordable] SSDs that will saturate a SATA II link, let
> alone a SATA III link.
> 
> My advice, FWIW, is to simply ignore this, unless you plan on
attaching
> a device in the near future that can actually exceed 300 MB/s.
> Currently the only way to do this is with striped mech drives on a
PMP,
> or with a $500+ SSD.

There is one RW bit in SB800 which makes SATA controller operate
in maximum Gen2 (3.0Gbps), this bit will limit CAP.ISS(RO) to
0010(3Gbps).

Some BIOS vendors may enable this bit "to save more power"(from spec,
driver needs not touch it). That's why I suggested James contact BIOS
vendor if you do want to get the 6Gbps dmesg.



Thanks,
Shane