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

[B Gates]

Thank You Sergio!

[radier]

NEW RST 11.5.2.1001 WHQL

http://www.station-drivers.com/telec...ivers.com).exe

P.S. no fix for memory leak, only driver is new.

[B Gates]

Is that memory leak the reason its kind of unstable? I'm running that driver with windows 8 and sometimes my rig fails to respond correctly.

[Marc HFR]

Hex dump is useless, you're looking at certain LBAs that do not correspond to any specific NAND address. It's just how to FW presents the logical volume to the OS.

For all you know, these new drivers just tell the FS to wipe it's arse better - there is no way to check whether or not TRIM is working past the FTL.

And how does the firmware know that certain LBAs previously used do not correspond to any specific NAND address anymore ? TRIM.

[andressergio]

Thank You Sergio!

[andressergio]

this are my ADATA SX900 256GB Sata 3 on RAID0 very fast drives

with latest RST 4K improved a lot



still using RAID ROM 11.5.0.1414 as the latest gives me low numbers, didnt try with the ADATAs but with the intel was hella low

[m.oreilly]

well heck. i feel a bit ripped off (socket 2011), if you folks with z77 systems are the only ones getting intel raid trim with win7
why would intel do this for this one socket iteration?
hehe, a bit rhetorical, but wtf

[Fernando]

well heck. i feel a bit ripped off (socket 2011), if you folks with z77 systems are the only ones getting intel raid trim with win7
why would intel do this for this one socket iteration?

I am pretty sure, that the complete Intel 7 series chipsets will be supported by the TRIM in RAID0 feature of the actual RST(e) v11.5 drivers/OROM versions. The only problem for users with an Intel X79 RAID0 system is to get the v11.5 OROM flashed into the BIOS and the v11.5 drivers installed. They have to add the missing HardwareIDs into the INF file or to use an already modded v11.5.0.1207 driver pack (look >here<).

[bluestang]

So are we guys with P67 gonna be SOL too?

[m.oreilly]

I am pretty sure, that the complete Intel 7 series chipsets will be supported by the TRIM in RAID0 feature of the actual RST(e) v11.5 drivers/OROM versions. The only problem for users with an Intel X79 RAID0 system is to get the v11.5 OROM flashed into the BIOS and the v11.5 drivers installed. They have to add the missing HardwareIDs into the INF file or to use an already modded v11.5.0.1207 driver pack (look >here<).

i'm hoping, and i was going to start asking around, if there is a corresponding 'e' 11.5 orom for the x79? for drivers, i was thinking of modding the inf. thanks for the link
i'm thinking intel would focus on the latest socket first, though i wonder if there is some hardware id in the orom that could be altered. i had been focusing on hardware raid and unmap. i can now see how intel could have passthrough with our intel 'fake' raid/ahci trim hints. someone just showed me some very interesting results, without using a hex editor. very interesting to the point i'm on board with you guys and win7 raid, atm

[andressergio]

NEW RST 11.5.2.1001 WHQL

http://www.station-drivers.com/telec...ivers.com).exe

P.S. no fix for memory leak, only driver is new.

i have a nice increase in perfomance with this drivers as the other ones, i dont know what mem leaks you refeer to but very good around here

cheers
Sergio

[Fernando]

i'm hoping, and i was going to start asking around, if there is a corresponding 'e' 11.5 orom for the x79?

There are no different Intel RAID OROMs for different chipsets available (only exception: v3 series for X79). That means, that the RAID ROM modules, which are released by Intel, either do work with a special chipset or not.
Intels v11.5 AHCI and RAID drivers are named RST, but they belong to the RSTe branch using the additional SCSI driver named iaStorF.sys. Furthermore the RST v11.5.0.1109 Alpha drivers, which were the very first of the v11.5 branch, natively did support X79 chipsets.
Since I have already the certification, that the modded RST drivers v11.5.0.1207 do work with X79 systems, I am very optimistic, that the Intel RAID ROM v11.5.0.1414 or v11.5.0.1582 will work with X79 SATA RAD Controllers too.

[Zaxx]

RWL have sunk their teeth into the 'trim for raid' topic. Makes for a great read too.

http://www.rwlabs.com/article.php?id=672

[m.oreilly]

yep. that's the article. i was hoping he'd have it up soon

[m.oreilly]

fernando, i've heard that the 11.5 oroms will not allow the x79 boards to boot. i was warned about this some time ago, so haven't toyed with modding a bios with any of these oroms. i would assume there would be a 3 series that had the necessary stuff to allow the same goodies that the 11.5 oroms bring? we have the 11.5 drivers as you mentioned based on the 'e' series, so i don't see why there would not be a 'e' series 3x orom in the near future, on that plays nice with socket 2011? vr zone had an article up recently which sated x79 versions of the intel 11.5 drivers/orom would be released. not sure where they got their info, but it would be nice

[m.oreilly]

RWL have sunk their teeth into the 'trim for raid' topic. Makes for a great read too.

http://www.rwlabs.com/article.php?id=672

oops...pulled!

[B Gates]

There are a couple of typo's being corrected. It should be back up soon.

[Zaxx]

Hmmm...he musta spotted a mistake or something, he's usually pretty thorough. He'll prolly repost it before long.

edit: heh...told ya...:P

[m.oreilly]

the devil

[B Gates]

RWL have sunk their teeth into the 'trim for raid' topic. Makes for a great read too.

http://www.rwlabs.com/article.php?id=672

Chris said he's going to send you an email tonight

[Fernando]

RWL have sunk their teeth into the 'trim for raid' topic. Makes for a great read too.
http://www.rwlabs.com/article.php?id=672

Thanks for the link.
Although the authors have used another and more time-consuming method to verify the TRIM function, the results are exactly the sames as ours (btester, andressergio and me) we have published already in July.

[Zaxx]

Good deal. Let him know I may not get back to him til tomorrow...was up all night, think I got a touch of food poisoning and it's NOT pretty. Gonna call it an early night tonight probably...stomach is still tied in a knot. Sux.

[B Gates]

Its back up http://www.rwlabs.com/article.php?id=672

[Zaxx]

Thanks for the link.
Although the authors have used another and more time-consuming method to verify the TRIM function, the results are exactly the sames as ours (btester, andressergio and me) we have published already in July.

I know...just figured some may find the article a good read with all the details in one write up vs spread thru out a 5 page, 170+ post thread.

[m.oreilly]

i'd like to see non SF drives tested also, but this is really cool! well done jon!