Re: Samsung SSD TRIM [Was: Heads up]

[Andriy Gapon]

On 15/04/2016 14:20, Steven Hartland wrote:
> Prior to Warners commit there was no NCQ TRIM support in FreeBSD, so while it
> was working with standard non-NCQ TRIM (and I can corroborate that as we use 
> the
> 840's and 850's all over with ZFS with TRIM enabled) its possible that it 
> could
> cause issues when NCQ TRIM comes into play.

Sorry, I didn't realize those were different commands.
Thank you for clarifying.

-- 
Andriy Gapon
___
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: Samsung SSD TRIM [Was: Heads up]

[Warner Losh]

> On Apr 15, 2016, at 7:14 AM, Tommi Pernila  wrote:
> 
> Hi,
> 
> On Friday, 15 April 2016, Steven Hartland  wrote:
> 
>> On 15/04/2016 09:11, Kurt Jaeger wrote:
>> 
>>> Hi!
>>> 
>>> avg wrote:
>>> 
 For what it's worth, I have been using the following SSDs since
 September of 2015 :
 
 ada2:  ACS-2 ATA SATA 3.x device
 ada3:  ACS-2 ATA SATA 3.x device
 
>>> I have one in use with zfs and trim:
>>> 
>>> ada1:  ACS-2 ATA SATA 3.x device
>>> 
>>> Works as my ports build hosts, and is fine as far as I can see.
>>> 
>>> Prior to Warners commit there was no NCQ TRIM support in FreeBSD, so
>> while it was working with standard non-NCQ TRIM (and I can corroborate that
>> as we use the 840's and 850's all over with ZFS with TRIM enabled) its
>> possible that it could cause issues when NCQ TRIM comes into play.
>> 
>> From what I read when this issue first came to light, I believe the actual
>> issue was a Linux kernel bug not a FW bug in Samsung drives that caused the
>> corruption. This is the thread which details said issue:
>> http://www.spinics.net/lists/raid/msg49440.html
>> 
>> 
> Here is a link to the commit that fixed the Linux kernel bug.
> https://git.
> kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=f3f5da624e0a891c34d8cd513c57f1d9b0c7dadc
> 
> Mounting a Samsung SSD with a linux kernel with trim enabled without this
> commit can cause issues.

Right, that bug allocates space for the TRIM message to be queued. FreeBSD
already does that, so that’s not going to be an issue. This was the issue I 
referred
to in a prior message.

Both the Samsung 840 and 850 are black listed as far as NCQ trim are concerned.
I’ve had other people that have tested them w/o my blacklisted entries, and 
there
are issues at least with the 840 that are firmware related.

Warner


signature.asc
Description: Message signed with OpenPGP using GPGMail

Re: Samsung SSD TRIM [Was: Heads up]

[Tommi Pernila]

Hi,

On Friday, 15 April 2016, Steven Hartland  wrote:

> On 15/04/2016 09:11, Kurt Jaeger wrote:
>
>> Hi!
>>
>> avg wrote:
>>
>>> For what it's worth, I have been using the following SSDs since
>>> September of 2015 :
>>>
>>> ada2:  ACS-2 ATA SATA 3.x device
>>> ada3:  ACS-2 ATA SATA 3.x device
>>>
>> I have one in use with zfs and trim:
>>
>> ada1:  ACS-2 ATA SATA 3.x device
>>
>> Works as my ports build hosts, and is fine as far as I can see.
>>
>> Prior to Warners commit there was no NCQ TRIM support in FreeBSD, so
> while it was working with standard non-NCQ TRIM (and I can corroborate that
> as we use the 840's and 850's all over with ZFS with TRIM enabled) its
> possible that it could cause issues when NCQ TRIM comes into play.
>
> From what I read when this issue first came to light, I believe the actual
> issue was a Linux kernel bug not a FW bug in Samsung drives that caused the
> corruption. This is the thread which details said issue:
> http://www.spinics.net/lists/raid/msg49440.html
>
>
Here is a link to the commit that fixed the Linux kernel bug.
https://git.
kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=f3f5da624e0a891c34d8cd513c57f1d9b0c7dadc

Mounting a Samsung SSD with a linux kernel with trim enabled without this
commit can cause issues.

Br,

Tommi


> Regards
> Steve
> ___
> freebsd-current@freebsd.org mailing list
> https://lists.freebsd.org/mailman/listinfo/freebsd-current
> To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"
>
___
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: Samsung SSD TRIM [Was: Heads up]

[Steven Hartland]

On 15/04/2016 09:11, Kurt Jaeger wrote:

Hi!

avg wrote:

For what it's worth, I have been using the following SSDs since September of 
2015 :

ada2:  ACS-2 ATA SATA 3.x device
ada3:  ACS-2 ATA SATA 3.x device

I have one in use with zfs and trim:

ada1:  ACS-2 ATA SATA 3.x device

Works as my ports build hosts, and is fine as far as I can see.

Prior to Warners commit there was no NCQ TRIM support in FreeBSD, so 
while it was working with standard non-NCQ TRIM (and I can corroborate 
that as we use the 840's and 850's all over with ZFS with TRIM enabled) 
its possible that it could cause issues when NCQ TRIM comes into play.


From what I read when this issue first came to light, I believe the 
actual issue was a Linux kernel bug not a FW bug in Samsung drives that 
caused the corruption. This is the thread which details said issue: 
http://www.spinics.net/lists/raid/msg49440.html


Regards
Steve
___
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Re: Samsung SSD TRIM [Was: Heads up]

[Kurt Jaeger]

Hi!

avg wrote:
> For what it's worth, I have been using the following SSDs since September of 
> 2015 :
> 
> ada2:  ACS-2 ATA SATA 3.x device
> ada3:  ACS-2 ATA SATA 3.x device

I have one in use with zfs and trim:

ada1:  ACS-2 ATA SATA 3.x device

Works as my ports build hosts, and is fine as far as I can see.

-- 
p...@opsec.eu+49 171 3101372 4 years to go !
___
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"

Samsung SSD TRIM [Was: Heads up]

[Andriy Gapon]

On 15/04/2016 07:19, Warner Losh wrote:
> Samsung 830 all firmware
> Samsung 840 all firmware
> Samsung 850 all firmware
> 
> All of these are at least 18 months old (if not older). There's some
> confusing in Linux lists on
> the full impact of the Samsung drives (there was a bug in the Linux
> implementation (that can't
> be present in the FreeBSD implementation) that may have been the root cause
> for the Samsung
> black listing). Out of an abundance of caution, I've kept them in the list.

For what it's worth, I have been using the following SSDs since September of 
2015 :

ada2:  ACS-2 ATA SATA 3.x device
ada3:  ACS-2 ATA SATA 3.x device

with ZFS with TRIM enabled and I haven't seen any problems thus far:

$ sysctl kstat | fgrep trim
kstat.zfs.misc.zio_trim.failed: 0
kstat.zfs.misc.zio_trim.unsupported: 2268
kstat.zfs.misc.zio_trim.success: 36895209
kstat.zfs.misc.zio_trim.bytes: 976003989504

Those disks are partitioned and host a small ZFS root / system pool plus log and
cache vdevs for other HDD-based pool.
So, I think that if the SSDs had any problem with TRIM I would have noticed it
by now.

-- 
Andriy Gapon
___
freebsd-current@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"