Ich7-m

New to the forum and to SSD technology in general.

Have an older Toshiba Satellite A105-S4094 that has a 5400 rpm SATA drive that is starting to show signs of wearing out. I figure that I should go ahead and make the leap to an SSD boot drive while I have an opportunity.

However, the southbridge chipset on this mobo is part of the ICH7-M family and does not support SATA II while its desktop counterpart does. I was pretty much set on dropping an Intel X25-V 40gb in there as a boot drive, but now I am wondering if the SATA I bottleneck is going to take a noticeable chunk of performance away from this drive. Any input on how much of a hit I will take if I drop a SATA II drive in this laptop? Am I looking at the wrong SSD considering this bottleneck? Any advice would be great.

I use this laptop for light audio production via Ableton Live. I intend to use the SSD as an OS drive for WinXP, Live, some VST software perhaps and a few system utilities only.

Another dilemma I have is dealing with external storage. To minimize wear on the SSD and separate audio from the OS drive, I want to utilize some sort of external drive. I use an Echo Audiofire8 soundcard, which has a secondary firewire port designed for daisy chaining an external firewire drive. This was my initial targeted storage option, but I wanted to see what you guys think about secondary mass storage.

I have a single ExpressCard slot on this laptop, perhaps for an EC SSD, eSATA card, etc. Haven't found any real options for larger EC SSD storage options. Could go USB, but I also use a Novation keyboard and Kaoss Pad 3 over USB as well. Don't know about conflicts caused by using multiple devices over USB like that. Seems like going with the daisy chained external firewire drive is the most logical, but going EC eSATA--->external eSATA drive is a strong candidate, no? Also been looking at replacing the DVD burner with a shell for linking a hard drive through it, but the interface is just PATA/IDE. Looking for stability in steaming audio here. Any input?

Apologies for the long post; TIA.

 

Suggest try a SSD+HDD setup using an optical bay caddy. See here, converting optical drive to a USB external unit. ICH7M's PATA I/O controller can provide up to 100MB/s read, 88.9MB/s write which only the latest HDDs are now achieving.

For an idea of comparative SATA-I versus SATA-II performance of your Intel X25M see here and here. Big performance gains can be had if your bios can set AHCI sata mode mode.

 

Okay, after reading some of the DIY optical bay HDD thread here, the possibility of going this route has peaked my interest. A few questions:

1) Ordering an optical HDD caddy from Hong Kong eBay sellers. Obviously cheaper, but how does this impact the transit time?

2) What is the realistic possibility that my Toshiba will whitelist say, a Hitachi 7k200 drive if I were to use a SATA to PATA optical bay caddy?

3) This laptop uses a Pioneer DVR-K16a DVD burner, so I would assume I would need this optical caddy:

http://cgi.ebay.com/Laptop-IDE-50-p...ge_Internal?hash=item56398048a4#ht_1361wt_978

4) According to what was listed via PC Wizard, the DVD drive is running in UDMA-2 (PIO 4) mode at ATA-33 speed.

PIO Mode: PIO4
UDMA Mode max: 2 (ATA-33)
UDMA Mode Enabled:2 (ATA-33)

Also, this thing uses a Phoenix BIOS. Safe to say I am going to have to adhere to the section "Quirks applicable to PATA optical drive interfaces (ICH8M or older)" in the DIY optical bay HDD thread? Is this too much of a gamble to be worthwhile?

I am also curious to know how to go about enabling AHCI for the SSD OS drive if there are no settings for this in the BIOS. Any help would be appreciated.

Was responding to my own post before seeing your post. Any idea on how to enable AHCI if there are no settings for this in the BIOS? The BIOS on this machine is virtually bland; no advanced settings it would appear.

Oh yeah, and the SSD in question was the Intel X25-V, not the M version. I figured with the SATA I bottleneck it wasn't worth the extra cost considering the age of this laptop. You agree? Trying to keep the budget down if at all possible.

 

For ~US$20, I'd try the caddy since if it works, it would be the cleanest solution. Worth reading this: RE bootup , setting AHCI mode. Some scripts are included to switch SATA controller into AHCI mode if the bios sets it to run in IDE mode. I'd suggest check your Device Manager to see the Device ID that the controller is reporting as.

27c48086 = IDE (ICH7M)
27c58086 = AHCI (ICH7M)

 

The only performance metric that a SATA I bus would hamper would be the sequential read speeds. That being said, 150MB/s is a heck of a lot faster than most hard drives anyway.

All other performance characteristics, including the all important random read/write metrics, should be minimally impacted if at all.

 

Under Matching Device Id:

pci\ven_8086&dev_27c4

I am assuming the scripts you are talking about are the sata.bat and pciex.bat files. How do I utilize these files in order to change over to ACHI mode safely? Treat me as the noob that I am; I don't want to trash the existing system, but I would like to ensure I can implement ACHI now before I get the OS SSD and optical bay HDD.

Intel X25-V 40 gb a good budget candidate for a simplified OS drive? The only other candidates I have read about that rival this SSD in price and performance would be the OCZ Vertex 30 gb and perhaps the Kingston SSDNow V-Series 40gb.

By the way, the OS for the SSD will be WinXP. Still the king for audio rendering right now.

 

If not comfortable delving with the scripts, suggest instead search for a bios fix. If one isn't available, suggest contact Toshiba support and ask them for a bios fix to enable AHCI. Without AHCI you won't see multi-queued read/writes (NCQ), which give very good results with an Intel SSD. A Intel SSD will still be fast, but with NCQ you'll get excellent multi-threaded performance.

If can't enable AHCI, then the Indilinx SSDs are faster. There's a $129AR deal for an 60GB OCZ Agility at newegg atm.
________________________________________________________________________________________
[/TR]

Switching IDE -> AHCI on systems without a bios option Some systems' bios does not have a SATA Native or AHCI option to enable the SATA controller to work in AHCI mode. This means the sata controller is running in IDE mode, with no NCQ. NCQ allows queuing of read/writes to improve performance. In the benchmarks on the right, we see AHCI improves HDD 4kb multithreaded reads (4kb Q=32) by double. The operating system does many such reads in parallel during bootup. For a X25M G2 SSD, AHCI improves these reads by 9 times(!!). Indilinx SSDs see improvement too, but they tend to go from 20MB/s to 60MB/s, nowhere near the Intel's 162MB/s. A sata-I system equipped system will see this X25M G2 performance capped to ~130MB/s. Good user commentary about the real-life performance difference b/w AHCI and IDE mode on a X25M G2 is here. Solution Requires a pre-boot PCI fixup to change sata controller into AHCI mode.

Drive IDE mode​ AHCI mode​ Fujitsu 500GB 5400rpm HDD Intel X25M G2 SSD from here

Create the DIY ViDOck USB bootdisk. Save the autoexec.bat file below onto the USB bootdisk's root directory, commenting/uncommenting the ICH7M/ICH8M specific details depending on your I/O chipset. Upon booting from the USB drive, it will set AHCI mode and chainload into your OS.

ahcipci.bat - setup I/O and memory space based on AHCI mode from ICH7M DV2000 here

Code:

:: Eg for Toshiba A105 - need to ID free I/O and memory space
echo Setting up AHCI PCI Configuration space...
setpci -s 0:1f.2 10.w=17D1
setpci -s 0:1f.2 14.w=17C5
setpci -s 0:1f.2 18.w=17C9
setpci -s 0:1f.2 1c.w=17C1
setpci -s 0:1f.2 20.w=17B1

:: Add BAR5 address
setpci -s 0:1f.2 24.l=F0C44400

ahci.bat

Code:

:: Based off Coreboot i82801gx AHCI SATA code http://tiny.cc/cbahci
::
:: Enable BARs
setpci -s 0:1f.2 4.w=0007

:: Allow both Legacy and Native mode
setpci -s 0:1f.2 9.b=8f

:: set interrupt line.Interrupt Pin is set by D31IP.PIP
setpci -s 0:1f.2 3c.b=a

:: set timings
setpci -s 0:1f.2 40.w=A307
setpci -s 0:1f.2 42.w=A303

:: sync DMA
setpci -s 0:1f.2 48.b=1
setpci -s 0:1f.2 4a.b=1

:: Set IDE I/O Configuration
setpci -s 0:1f.2 54.l=3002

:: Set Sata Controller Mode.
setpci -s 0:1f.2 90.b=40

:: Port 0 and 4 enable
setpci -s 0:1f.2 92.b=1011

:: SATA initialisation register
setpci -s 0:1f.2 94.l=4a000180

:: SATA initialisation sequence
setpci -s 0:1f.2 a0.b=40
setpci -s 0:1f.2 a6.b=22
setpci -s 0:1f.2 a0.b=78
setpci -s 0:1f.2 a6.b=22
setpci -s 0:1f.2 a0.b=88
setpci -s 0:1f.2 a4.l=1b1b1212
setpci -s 0:1f.2 a0.b=8c
setpci -s 0:1f.2 a4.l=121212AA
setpci -s 0:1f.2 a0.b=0
setpci -s 0:1f.2 3c=0
setpci -s 0:1f.2 94.l=c4a000180

autoexec.bat (ICH7M or ICH8M, change code for either) converted to DOS setpci calls from this code

Code:

echo off
echo ICH7M/ICH8M: 27c48086/28288086 = IDE, 27c58086/27298086 = AHCI
echo ICH7M/ICH8M: original sata PCI ID. Visually confirm it is 27c48086/28288086 (IDE)
setpci -s 0:1f.2 0.l

:: setup PCI Configuration space. You'll need to find free space in PCI Configuration Space to allocate to
:: Use PCI Scope Trial to easily identify free space 

:: register 4, bit 1: Memory space enable (for AHCI)
:: Must be done while in IDE mode (register 90=0)
setpci -s 0:1f.2 4.b=2:2

:: Register 90: 0x00=IDE, 0x40=AHCI, 0x80=RAID. Set it to AHCI + non-combined.
setpci -s 0:1f.2 90.b=40

call ahcipci.bat

:: Register 94: set the SCRAE bit 9
:: setpci -s 0:1f.2 94.w=200:200

:: Register 9: set pci_class_prog=0x01
setpci -s 0:1f.2 9.b=1

:: Register a: set pci_class_device=0x06
setpci -s 0:1f.2 a.b=6

:: Switch off, then on the sata port
setpci -s 0:1f.2 92.w=0000
setpci -s 0:1f.2 92.w=1011

echo Visually inspect output below. Want a "1" on the left side to indicate a device was found
setpci -s 0:1f.2 92.w

echo ICH7M/ICH8M AHCI FIX APPLIED
echo ICH7M/ICH8M new sata PCI ID. Visually confirm it is 27c58086/27298086 (AHCI)
setpci -s 0:1f.2 0.l

:: quirk: Required a hardset to enable AHCI memory space
setpci -s 0:1f.2 4.b=7

:: rescan pci-bus
lspci > nul

:: Chainload operating system
:: grub --config-file="menu.xp"
:: Or for Vista/Win7
:: grub --config-file="menu.w7"

Once booted into your OS, it will have been as if you'd set your bios to AHCI mode. So now you'll need the AHCI driver to take advantage of the new speed. XP has some convoluted process to load it without causing a BSOD (search google).

Testing

On a ICH7M equipped Toshiba with ONLY a IDE compatibility mode, this did not work. The problem is the bios doesn't provide an INT13 disk read with AHCI enabled, a necessity to start the Window's bootloader. If there was a DOS based TSR to add AHCI INT13 support, integrating UIDE driver and this AHCI DOS code, then it might work, otherwise grub will hang trying to load Windows.

Could also use this method to create a XP/W7 image, use hmload to load the image into memory, switch into AHCI mode, run the image from memory to initiate the boot process. Would then ideally switch to the proper disk files *after* the AHCI protected mode driver has been loaded.

A ICH7M equipped HP DV2000 was switched from AHCI->IDE mode in the bios, then ran the above process and it successfully changed to AHCI mode, getting the higher benchmark numbers. The XP installation had been pre-built with the AHCI driver, so I didn't have to load the driver (phew!). Note too, the Linux documentation says the PATA devices won't work. Well, the DV2000's PATA optical drive reads CDs without any problem. This worked, while the Toshiba didn't because the DV2000 INT13 handler was designed for both AHCI and IDE modes.

 

Well, I am at the latest BIOS version, 6.0, which was released in 2007. So I doubt I can get anything done with Toshiba regarding the BIOS at this point. I am comfortable with running the scripts as long as I can reverse any negative reaction to them; I just directions on how to run the scripts.

Do I add this file to the startup folder? Run once via command prompt? Just give me some type of direction if this is not entirely frustrating for you. I am not completely oblivious, just a relative amateur compared to someone like you.

 

If you are going to be using XP, I highly recommend using an OCZ drive. Current Intel drives only use TRIM to maintain SSD performance levels, and since XP does not support TRIM you will find that over time your write speeds will be reduced. OCZ's drives, and Intel's earlier G1 drives, support Gargabe Collection which will maintain performance over time just as well - if not better than - TRIM would.

OCZ's drives running FW 1.41 or FW 1.5 both support GC.

 

You don't expect this to change for the current gen Intel drives? Just curious.

 

Nope. Considering they dropped GC support when they came out with TRIM support, I don't think they have any intention of adding it back in.

 

Noted. Will look into going the OCZ route then.

Bringing this over to the second page for nando4.

Scripts aside, when installing WinXP on the new SSD, is it possible to use the floppy F6 method described in the OP of this thread:

http://forums.pcper.com/showthread.php?t=444831

 

You can't load the AHCI driver until the PCI Configuration space is setup for AHCI mode. One way to do that is to convert the ICH7M IDE->AHCI code from here to use DOS based tools to do the memory and PCI writes (setpci/pt), then run 'grub' or 'grub --config-file="menu.xp"' to chainload into XP. These tools are provided on the DIY ViDock bootdisk.

If successful, you'd still need to start your boot process with either a USB stick, or image the USB stick files and use something like grub to boot from the image to greatly improve boot file. It would also be possible to write the same code as a grub2 module, if using grub2 bootloader, which would make it *very* easy.

 

So basically what you are saying here is that the BIOS for this laptop is going to force Compatibility mode short of the workaround you posted above. If I am following what you posted correctly, I would have to boot via this USB image indefinitely. Not ideal if so. Pretty clear to me that your technical savvy far exceeds my own, as I just don't think I can wrap my head around getting this done.

I am assuming it would not be possible to enable AHCI when using my original Toshiba Restore DVD. I was thinking that when I drop the new SSD in the system, I could use the F6 floppy method outlined here:

http://www.intel.com/support/chipsets/imsm/sb/CS-021736.htm

The restore DVD is essentially a custom WinXP disc that does allow loading drivers before the factory restoration via F6. But if I am understanding you correctly, even this would not work, as my system would revert to Compatibility mode regardless. Is this indeed what will happen in this case?

According to both you and Greg Ross, assuming I can't enable AHCI, my best bet would be going OCZ for SSD. If by some miracle AHCI does turn out to be in the cards, is the Intel X25-V a better candidate than an Agility/Vertex drive?

Also, since I didn't get a response initially, I will ask again: How does ordering an optical bay HDD caddy from Hong Kong affect transit to the US? Never ordered internationally before. I am sure some of you guys have some experience in how the process works.

 

Hello
I'm also tring to enable AHCI on my Inspiron 9400.
I try to do a custom pci option rom to replace one inside bios.
Until now I do some tries but without luck.
I'm using flat assembler.
Now I get stuck at offset 90,can't make it 40
In my config,offset 90-93 is 00150002 I try to write 10150040 but result is 10150042.

Code:

	mov eax,8000fa90h	      
	mov ebx,10150040h	      
	mov dx,0CF8h		      
	out dx,eax		     
	mov dx,0CFCh		     
	in eax,dx
	and eax,0FFFFFFFFh	   
	or eax,ebx		     
	out dx,eax		     

Can somebody help me?

 

Try the DOS setpci/pt commands here to switch to AHCI mode, then see if 'grub' works. If it doesn't, then your bios doesn't have INT13 AHCI read/write support, something that XP/Win7 needs to start the boot process. However, Linux would allow this switch to AHCI and work.

 

Which ones?
ahcipci.bat I see is included in autoexec.bat
ahci.bat have something what isn't in autoexec.bat
offset 94 is different in ahci.bat and ahcipci.bat.

I can change offset 90 with 40 value in windows,then dev id become 27c5 and if I let them windows freeze,if I change back to 02 or 42 in short time windows come back.
Now I'm trying one new pci rom inspired by your autoexec.bat.
Problem is I have always 02 value at offset 90 I think.

 

Tried.
Grub left me with just a blinking cursor.
Other way to test INT13 AHCI read/write support?

 

Exact same behavior on my Satellite A105.

 

No INT13 AHCI bios routines for you, so can't chainload using grub4dos. Not sure if Ontrack Disk Manager would help? Only other possibility then would be:

Could also use this method to create a XP/W7 image, use hmload to load the image into memory, switch into AHCI mode, run the image from memory to initiate the boot process. Would then ideally switch to the proper disk files *after* the AHCI protected mode driver has been loaded.

If you can figure it out, please post as I'm sure others would love to have AHCI speed.

 

Any info is very appreciated.
And since I don't have required skills on programming this will be very helpfull.

 

@ nando4
Can you explain how to do or where to find more info about

:: setup PCI Configuration space. You'll need to find free space in PCI Configuration Space to allocate to
:: Use PCI Scope Trial to easily identify free space

 

I've linked a screenshot of a DV2000 in AHCI mode in my original post. Suggest browse your memory map to allocate free I/O and memory in free PCI Config Space to match that image.

 

I build my pci option rom which write pci registers and seems correct but I can't boot from hdd neither grub can't see my hdd.
But if I boot into some linux distro I can see and get access to my hdd and is in sata mode.
Maybe is need for some chipset refresh command in my rom.

 

While you have switched to AHCI mode, the second component that is needed to boot XP/Win7 is INT13 bios (or TSR) AHCI disk read/write support. Without that you will not be able to boot XP/Win7.

Could try to use a kicker image to start the process which then dynamically swaps to your HDD *after* the protected mode drivers as loaded. See here.

 

I think I'm not at that point.
If I boot from a usb stick and enter grub to edit lines then I must have 2 hdd visible.
(when edit line,in grub editor or how is that named,TAB give you completion of what commands are available so if hdd was reported by bios or visible
"root (hdTAB" must list available like hd0 hd1 ...
but visible is just hd0 which is usb stick)
Also when booting from DIYVDOCK usb stick ,OS (i forget what is ,free dos or whatever) complain about read partition error on disk 1 or something like that.

 

I can get this,but bios still report my hdd as ide and now is switched to sata.
Maybe is need somehow to redetect drives in bios.

Code:

 
Detection utility for Int 13h extensions  v1.3   (c) 1997-2000 Jan Steunebrink

Harddisk 2: Int 13h extensions version 2.1 detected.
 Subset supported: Extended disk access support (for > 8GB) : Yes
                   Drive locking and ejecting support       : No
                   Enhanced Disk Drive (EDD) support        : Yes
 Device features : DMA boundary errors handled transparently: Yes
                   Device supports write with verify        : No
                   Media is removable                       : No
 Addressable geometry (CHS) is not available (drive is probably > 7560 MB)
 A total of 195371568 Sectors of 512 Bytes (95396 MB) are addressable in LBA.
 Enhanced Disk Drive (EDD) configuration parameters:  (DPTE revision 1.1)
                   ATA I/O base and Control ports are 01F0h and 03F6h
                   BIOS uses IRQ 14 for this disk
                   Master/Slave bit for this disk is set to : Master
                   BIOS to Device interface uses LBA        : Yes
                   Configuration of Hardware specific Options:
                   - Multi-word DMA transfers  : No
                   - Fast PIO accessing        : Mode 4
                   - PIO multi-sector transfers: 8 Sectors per transfer
                   - 32-bit transfer mode      : Yes
                   - Disk is ATAPI device      : No

 

Kingston SSDNow 40gb benchmarks in IDE compatibility mode:





Expected, but still would perform better if AHCI was possible I would bet.

 

ICH7M SATA-I IDE vs ICH9M SATA-II IDE vs ICH9M SATA-II AHCI

Kingston SSDNow 40GB (Intel) - IDE ICH7M SATA-I

Intel X25M G2 - IDE ICH9M SATA-II

Intel X25M G2 - AHCI ICH9M SATA-II

Are you happy with the improvement in performance over your previous lethargic HDD?

As to your current limitations, we can see that SATA-I is limiting your sequential reads to 130MB/s, or half it's potential 260MB/s and we see that IDE mode is limiting your multithreaded 4KB reads to 19MB/s, down from a potential to hit the 132MB/s SATA-I cap.

Unlocking AHCI via a DOS bootdisk, PCI fixup, then chainload Windows, would correct that. Only dilemma is you'd need a DOS INT13 driver to give you AHCI reads to allow chainloading Windows using grub4dos since your BIOS doesn't provide that. That AHCI driver doesn't exist as yet. Or alternatively some other way to get Windows loaded to the point where it's protected mode AHCI driver can start doing disk reads so bypass the need for INT13 support.

Or could just enjoy the current performance by several factors and know a system upgrade would see further bootup time and application response improvements.

 

It is clear that the SSD is a much stronger performer than the previous Toshiba HDD. I have no idea how long this system stays afloat, and I hope I will be able to make use of this SSD long into the future.

In the end, it was a worthwhile experiment with this older system.

 

Could somebody with ICH7 or ICH7-M working in AHCI mode provide me a dump of some areas?
For that install pciscope launch it,select sata controller (must be 8086 27c5) go to PCI register dump select all hex values then paste it into some hex editor or even text file. (since isn't much info will be good even a print screen snapshot).
Then go to Information tab,for same sata controller,take a note about something like
BAR Type Base Limit Size

needed values are base and size
Install HWDirect then launch it go to memory dump at Physical Address enter value readed from Base,without the h in the end and at size value from size.
Ex.

Code:

BAR Type  Base      Limit      Size     Comments 
0    Mem EF9FD800h EF9FDFFFh 00000800h   2 KB. 32 bit. Non-prefetchable. Locate anywhere in 32 bit address space 

in that case Physical Address is EF9FD800 and size is 800

Select all hex values and paste them into some hexeditor or even text file and upload them somewhere and send me link or any other method you preffer.
Thanks.