My 128Gb x-series support TRIM, possible?

[fvbarc]

Hi everyone, according what is written here

 

http://forum.corsair.com/v3/showthread.php?t=84677

 

my CMFSSD-128D1 support TRIM, but firmware rev. is 1

Is it possible?

How to be sure about firmware version?

Also other sofware say that my HDD support TRIM

[RAM GUY]

Please run Crystal Disk Info and post the results.

[fvbarc]

here it is ;):

[RAM GUY]

The drive you have is the first version of Firmware and does not support Windows 7 Trim, but the drives have supported TRIM, just not Windows version of TRIM.

[Psycho101]

When TRIm is highlighted in Crystal (ie not greyed out) it means that the controller is capable of supporting TRIM, but the FW may not necessarily currently support it.

 

As far as I know, there are only three types of TRIM, Linux, Mac OS-X and Windows and the X series supports none of those currently.

[RAM GUY]

I think you are maybe a little miss informed please do a search for TRIM and ATA, S-ATA. I think it will help to explain it more in detail if you really want to understand.

But in a nut Shell TRIM is a command that is defined by Wikipidia is a command or function that was defined in the S-ATA Spec. The page is down at the moment or I would post some quotes but Windows 7 and 2008 Server are the only O.S. that support that command ATM.

In Microsoft's case there version of TRIM is not necessarily compatible with the S-ATA TRIM command. However it does meet the spec as does Indolinx and Intel and Samsung and a few other controller manufacturers have all been working with Microsoft to get that ratified. But the solution looks to be simply have the firmware of said drives support Microsoft's version of TRIM

Chicken before the egg so to speak.

And in the case of Indolinx the new firmware version 2.0 on our drives supports Winodws7 version of TRIM and Garbage collection and is in production on new drives; the issue is and has been the update utility has caused some drives not to function after the update. Once that has been fixed the firmware will be released.

[Psycho101]

Yes I understand about the update.

 

I'm not 100% sure what you mean by being misinformed. I understand that in the purest sense, TRIM is an ATA command and that different OS's implement this in different ways for example Windows 7 uses what's called a "Volume Bitmap" to store and pass on data about which files are deleted, meaning the controller can now pre-erase those blocks.

 

From Wikkipedia (don't usually quote that site because one never knows who's been editing it): "TRIM has been prepared in Linux 2.6.28[7], but not yet fully implemented[8][9]. In Microsoft Windows, TRIM has been implemented in Windows 7[10] and Windows Server 2008 R2,[11] [12] and was released with the final versions of those operating systems in October 2009. Older solid state drives will need firmware updates, otherwise the new command will be ignored.

Where TRIM is not automatically supported by the operating system, there are utilities which can send TRIM commands manually. Usually they list all free blocks as specified by the operating system and then pass this list as a series of TRIM commands to the disk. These utilities are available from various manufacturers (Intel[13], XtremeSystems[14]) or as general utilities (hdparm since v9.17[15][16])."

 

The different implementations of TRIM , as I understand it, only refer to the way the OS handles the preparation needed to pass the data on to the controller. Windows uses volume bitmaps, Linux may use a different method. The end results are the same though, Linux and OS-X TRIM capable versions will also use the ATA TRIM command. Problems do happen though and currently only the default Win 7 IDE and AHCI drivers will pass the command through, proving that it's ATA TRIM that's the function instigating TRIM. The command can not be passed throughif using a third party driver, if running a RAID array and even if running SSD's in JBOD.

 

I can't be 100% sure but I'm fairly positive that crystal is either just detecting the SSD controller type and then displaying whether the Hardware itself is TRIM compatible using a list or it could even be detecting whether Win 7 TRIM is turned on via checking status of the fsutil flag.

[RAM GUY]

I think you have answered the question reading what you posted.

TRIM is just a command or function as stated previously and X-Series drives have supported TRIM but not Windows 7 TRIM until version 2.0 of the firmware also as previously stated.

I do not want to offend you; just do not want to confuse other users is the goal here. Some may not have as deep of an understanding as you with this.

But there are two aspects to this one is hardware and the other is software.

[Cadencia]

there are two aspects to this one is hardware and the other is software.

 

Actually, firmware probably makes a third aspect, unless you are counting that as software.

 

To summarise, we have:

• The X series drives ALL support the ATA command "TRIM"
  
• Only the Microsoft default Win 7 IDE and AHCI driver passes the ATA trim command to the SSD.
  
• Windows 7 passes the ATA trim command to the SSD via the MS default driver
  
• You need firmware 2.0 in the SSD for it to recognize the ATA trim command as issued by Win 7.
  

 

The question is, then: how can you make the SSD which has firmware prior to release 2.0 recognize ANY trim command, by ANY means at all? Is there ANY software in existence that can actually make a pre 2.0 X series SSD execute the trim function?

 

If not, it seems a bit sophisticated to claim that the device supports trim.

 

Please, just tell us how to get at it!

[RAM GUY]

Yes Firmware is considered software.

And yes I would agree with these statements.

 

The question is, then: how can you make the SSD which has firmware prior to release 2.0 recognize ANY trim command, by ANY means at all? Is there ANY software in existence that can actually make a pre 2.0 X series SSD execute the trim function?

 

A: Yes the wiper utility was only ever released as a Beta and we never released it for that reason and it was not reliable from our testing.

[Psycho101]

Without the TRIM function being physically programmed into FW, there is no way an OS can pass the command through on the fly unfortunately. This is why wiper.exe is a "run and process" kind of program. It not only goes through all blocks to sort out the good from the junk, but it also does the job of pre-erasing the blocks too. The actual process going on inside the drive is a bit more complicated. I'll not say too much as like RAM Guy says, sometimes too much info is confusing, sufficed to say that a drive capable of TRIM only uses that ATA command to flag blocks as "ready" for pre-erasing. It's then the job of the controllers Garbage Collection to immediately pre-erase those blocks or intelligently wait until the drive isn't transferring data.

 

Hopefully having TRIM programmed into the FW will mean that as long as the OS ends up passing the command, and the GC pre-erases the blocks, then what goes on behind the scenes on the OS side shouldn't make any difference to the SSD's ability to TRIM.

 

Note that when I say garbage Collection in this context I don't mean the soon to be added aggressive type used to maintain performance in RAID, or on XP/Vista. Every SSD has its own less aggressive Garbage Collection that does clear up at least some of the detritus left by usage.

 

Crystal reporting TRIM capabilities is like for instance a CPU being branded as x64 capable, it will process 64bit commands but only if running a 64bit OS. What Crystal is saying is "Yes your drive can support TRIM" The fact that it neglects to mention that the FW on the drive currently does not pass the command through would be an oversight on the part of CDI's author(s).

[RAM GUY]

OK OK I will let that slide, LOL! Great explanation by the way and in simple clear terms. Good job! Do you speak French or Russian Psycho101?

There are some pretty good explanations about TRIM here And on an article wrote in 2002 from this site. flash SSD Jargon Explained