Raid0 and TRIM Support on Intel 11.5 OROM and Latest v11.5.0.1184 RST Drivers ?

[biassj]

Anyone with his problem?

I have the 11.5.0.1184 on my z77-UD5 and when I update the drivers to 11.5.0.1207 it's stuck at Windows loading animation?

edit:

Note: I was already on Raid 0 with my SSDs and using older Intel drivers, if that helps.

[andressergio]

i think you guys are relying on the hex 'results' a little too much, especially with multi drive lun and single/'shared' lba. both drives you mentioned are sandforce based, and with their inherent encryption/lba jiggery pokery, i think you would still be looking at programmed with 00 returned...still not buying it...

we are all looking for better performance mate...even if trim not working or maybe we dont know yet its good to see new RAID ROMS and drivers

[chispy]

[Fernando]

i think you guys are relying on the hex 'results' a little too much, especially with multi drive lun and single/'shared' lba. both drives you mentioned are sandforce based, and with their inherent encryption/lba jiggery pokery, i think you would still be looking at programmed with 00 returned...still not buying it...

1. Do you know a better method to verify, if Trim is working or not?
2. My Crucial M4 SSD's are not Sandforce based.
3. Today or tomorrow I will transfer my RAID0 array from a Z68 to a Z77 system, which is using the same Intel RAID ROM version. After having done the "Hex test" again, I probably can say more about the validity of it.

I have the 11.5.0.1184 on my z77-UD5 and when I update the drivers to 11.5.0.1207 it's stuck at Windows loading animation?

Which Intel RAID ROM version is in your mainboard BIOS chip?

[chispy]

Fernando did you estract the Bios OROM Raid Intel RAID ROM module v11.5.0.1582 from that Biostar Bios ? If so please can you post it here and share it with us to give it a try , im also in transition to upgrading from my old dying , and aging Asus Rampage 3 eXtreme x58 mobo i7 950 combo , to a newly fresh Asus Z77 Mobo and i5 3570k so i will have a lot of testing going on with Z77 Intel chipset during the next few days.

[chispy]

Nevermind i found it: http://www.xtremesystems.org/forums/...1582-available!!!

Thanks to canthearu : http://www.canthearu.com/pchraidr.bin

[Fernando]

Fernando did you estract the Bios OROM Raid Intel RAID ROM module v11.5.0.1582 from that Biostar Bios ? If so please can you post it here and share it with us to give it a try , im also in transition to upgrading from my old dying , and aging Asus Rampage 3 eXtreme x58 mobo i7 950 combo , to a newly fresh Asus Z77 Mobo and i5 3570k so i will have a lot of testing going on with Z77 Intel chipset during the next few days.

You can find this Intel RAID ROM v11.5.0.1582 and other RAID ROM modules >here<. They are already extracted and tested by me.

[biassj]

Which Intel RAID ROM version is in your mainboard BIOS chip?

I was on Raid Rom 11.5.0.1184 and now I'm on 11.5.0.1582. Tried using an older version of the drivers which you had on another website, same results. I updated my firmware on my two Agility 3s. I wonder if I need a fresh install?

[andressergio]

anyone saw any better perfomance with this new RAID ROM ???

[medeirosdez]

Fernando did you estract the Bios OROM Raid Intel RAID ROM module v11.5.0.1582 from that Biostar Bios ? If so please can you post it here and share it with us to give it a try , im also in transition to upgrading from my old dying , and aging Asus Rampage 3 eXtreme x58 mobo i7 950 combo , to a newly fresh Asus Z77 Mobo and i5 3570k so i will have a lot of testing going on with Z77 Intel chipset during the next few days.

Friend,

The OROM code has been extracted from the BIOS I provided and it is available at the thread I created myself.

Link for my thread at this very own forum is here.

Link for extracted OROM is provided by user canthearu (thank him for that!): http://www.canthearu.com/pchraidr.bin

User Fernando is also doing a GREAT job spreading news about this updated OROM (thank him too). Let's hope it's final and not beta.

[Fernando]

Fernando did you estract the Bios OROM Raid Intel RAID ROM module v11.5.0.1582 from that Biostar Bios ?

Just to clarify this: Yes, I have extracted the ROM module from the BIOS myself and that is the file I am offering >here<. I did not use the ROM, which was extracted by canthearu.

[chispy]

Thank you guys

[canthearu]

Well, in theory, the rom from my site should be exactly the same as Fernando's, since they were both extracted from the same ROM.

Fernando: TThe new 11.5.0.1582 ROM works fine with the older 11.1.0.1006 drivers on my P55 board.

[m.oreilly]

fernando, with non sandforce, i think you may be seeing GC rather than trim. what we are not privy to are the inner workings of controller FW and how they handle/don't handle metadata and how they utilize any sort of file system aware translation layering. for all we know, we could be seeing just a response to a simple delete, and the drive/FW responding GC algorithms. again, it would be ultra cool to have raid trim (i think) under win7/server 08r2. i don't think the hex thing is a be all end all when you include the drive's inner workings. proprietary inner workings. good luck.

[Fernando]

fernando, with non sandforce, i think you may be seeing GC rather than trim. what we are not privy to are the inner workings of controller FW and how they handle/don't handle metadata and how they utilize any sort of file system aware translation layering. for all we know, we could be seeing just a response to a simple delete, and the drive/FW responding GC algorithms. again, it would be ultra cool to have raid trim (i think) under win7/server 08r2. i don't think the hex thing is a be all end all when you include the drive's inner workings. proprietary inner workings. good luck.

Since the Garbage Collection is a feature of the SSD's and has nothing to do with the environment (chipset, OS, drivers, RAID ROM etc.), please explain:
How is it possible, that in my Z68 system the test file data have been still present within their RAID SSD cells hours after the delition. whereas they were completely erased within seconds with my Z77 system?
Please try to explain it.

[m.oreilly]

well, it does. unless you have personal access to proprietary goings on with FW creators, how do you know what you are 'seeing' is actually what is happening? the workings of translation layers and pointers within the FW/ssd themselves are pretty much guarded secrets. you may be surprised at how ssds these days can 'sync' with file system hints. i was.

[Fernando]

@ m.oreilly:
Thanks for your response, which did not persuade me really.

@ all:
Yesterday I have repeated the "Trim test" with an Intel Z77 mainboard while running Windows 7 x64.
Beside mainboard and CPU the system (hardware ansd software) was exactly the same as before with an Intel Z68 chipset mainboard, where the "Trim test" failed (data were still present after the delition).

Here are the test results:

1. Hex Offset data of the sector of the RAID partition C: with the JPG test file:


2. Hex Offset data of the same sector about 5 seconds after having deleted the file:


3. Just for a comparison here is the pic I took with my Z68 system 1 hour after having deleted the test file:


By the way: With the Z77 mainboard I got the same results while running Windows 8 x64. The related Offset data of the test file sector zeroed out within a few seconds after the delition.

Regards
Fernando

[andressergio]

@ m.oreilly:
Thanks for your response, which did not persuade me really.

@ all:
Yesterday I have repeated the "Trim test" with an Intel Z77 mainboard while running Windows 7 x64.
Beside mainboard and CPU the system (hardware ansd software) was exactly the same as before with an Intel Z68 chipset mainboard, where the "Trim test" failed (data were still present after the delition).

Here are the test results:

1. Hex Offset data of the sector of the RAID partition C: with the JPG test file:


2. Hex Offset data of the same sector about 5 seconds after having deleted the file:


3. Just for a comparison here is the pic I took with my Z68 system 1 hour after having deleted the test file:


By the way: With the Z77 mainboard I got the same results while running Windows 8 x64. The related Offset data of the test file sector zeroed out within a few seconds after the delition.

Regards
Fernando

Fernando but its on 00000 didnt that means ok ?

[Fernando]

Fernando but its on 00000 didnt that means ok ?

Yes, my test results do 100% confirm your earlier findings:
Trim is working within the RAID0 array under Win7 and Win8, if the mainboard has an Intel Z77 chipset and the user has flashed a BIOS with an Intel RAID ROM v11.5.0.1414 or higher and uses the Intel RST(e) RAID drivers v11.5.0.1184 or higher.
Additional finding:
Mainboards with an Intel Z68 chipset are not (yet) supported by this Trim into RAID feature.

[felix_w]

Well done, Fernando!

[chispy]

Thanks a lot Fernando for all the hard work , testing, help and findings. Well Done. This is what XS its all about to share and help each other

[andressergio]

Yes, my test results do 100% confirm your earlier findings:
Trim is working within the RAID0 array under Win7 and Win8, if the mainboard has an Intel Z77 chipset and the user has flashed a BIOS with an Intel RAID ROM v11.5.0.1414 or higher and uses the Intel RST(e) RAID drivers v11.5.0.1184 or higher.
Additional finding:
Mainboards with an Intel Z68 chipset are not (yet) supported by this Trim into RAID feature.

great !!!

[Zaxx]

great !!!

Yeah...unless your stuck with a Z68....

Plan on going R0 with another 520 in a month or two now that intel lowered their ssd prices...with any luck maybe trim (in whatever form or however implemented) will make it's way down to my platform before long.

[medeirosdez]

Release notes for RST 11.5.0.1207 drivers have been made available! Here is the link!

Also confirmed: OROM 11.5.0.1582 is FINAL, not beta!

[andressergio]

Release notes for RST 11.5.0.1207 drivers have been made available! Here is the link!

Also confirmed: OROM 11.5.0.1582 is FINAL, not beta!

thanks man !!!