[nvmewin] nvmewin Digest, Vol 40, Issue 5
Suman Prakash
bsuman.prakash at gmail.com
Tue Apr 14 12:50:03 PDT 2015
Hi Brandon,
Please find my answer below -
a. EoL: Detection during device hot plug -
During driver initialization, the SMART/Health Information, Critical
Warning field, bit 3 which mentions "If set to '1', then the media has been
placed in read only mode" is the trigger to enter Read Only mode.
In driver, we have introduced a new state in NVMeRunning(), which sends a
Get Log Page for SMART Health information, and save the critical warning
field. During disk initialization, when kernel sends the SCSI mode sense
command with Page code 0x3f(Return all pages), in the mode parameter
header, in Device specific parameter, the WP(write protect) bit is set to
1. Once the mode sense response is returned, the disk in the disk manager
is shown as 'Read Only' and no more write commands are sent to the device.
b. EoL: Detection during run time -
During run time, for write command, the completion entry with SCT = 1h
(Command Specific Status) and SC = 82h (Attempted Write to Read Only Range)
is the trigger to enter Read Only mode. Please note that this is currently
not part of the NVM Express: SCSI Translation Reference specification and
is in the process of being included.
In driver, for any write command, in the completion entry, if SCT = 1h and
SC = 82h, the following sense data is returned to the kernel:
SrbStatus = SRB_STATUS_ERROR
StatusCode = SCSISTAT_CHECK_CONDITION
SenseKey = SCSI_SENSE_DATA_PROTECT
ASC = SCSI_ADSENSE_WRITE_PROTECT
ASCQ = SCSI_ADSENSE_NO_SENSE
For which the kernel sends the mode sense with Page code 0x3f, which is
translated to Get log page for SMART Health information, which returns
critical warning field, with bit 3 set to 1. In response to the mode sense
command, in mode parameter header, in Device specific parameter, the
WP(write protect) bit is set. After the mode sense response is returned,
the disk in the disk manager is shown as 'Read Only' and no more write
commands are sent to the device.
We have tested this implementation in following ways -
1. Insert a RO device and start the PC. After PC is booted, the disk is
shown as Read Only in disk management.
2. Hot inserted the RO device and disk management shows disk are Read Only.
3. Executed IOmeter on the NVMe disk. Once the disk become Read Only, The
'Error Count' field in IOmeter start increases and IOPs starts dropping.
Disk managements shows disk as Read Only.
4. Tried copying files to Read Only drive and got pop up messages as Drive
is write protected.
5. When the Available Spare field in the SMART information log is 10%,
started copying files to the disk. And when disk becomes Read Only, got pop
up message as Driver has become write protected.
Thanks,
Suman
On Wed, Apr 15, 2015 at 12:30 AM, <nvmewin-request at lists.openfabrics.org>
wrote:
> Send nvmewin mailing list submissions to
> nvmewin at lists.openfabrics.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://lists.openfabrics.org/mailman/listinfo/nvmewin
> or, via email, send a message with subject or body 'help' to
> nvmewin-request at lists.openfabrics.org
>
> You can reach the person managing the list at
> nvmewin-owner at lists.openfabrics.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of nvmewin digest..."
>
>
> Today's Topics:
>
> 1. Re: nvmewin Digest, Vol 40, Issue 3 (Brandon Schulz)
> 2. Re: nvmewin (Tom Fones-SSI)
> 3. Re: nvmewin (Robles, Raymond C)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Mon, 13 Apr 2015 21:01:16 +0000
> From: Brandon Schulz <brandon.schulz at hgst.com>
> To: suman prakash <suman.p at samsung.com>,
> "nvmewin at lists.openfabrics.org" <nvmewin at lists.openfabrics.org>
> Subject: Re: [nvmewin] nvmewin Digest, Vol 40, Issue 3
> Message-ID:
> <
> BN1PR0401MB09316D3E00BDAD9B0F65B9D3F2E70 at BN1PR0401MB0931.namprd04.prod.outlook.com
> >
>
> Content-Type: text/plain; charset="us-ascii"
>
> Suman -
>
> Do you know what you are going to use as a trigger for entering read-only
> mode at drive EOL?
>
> Brandon
>
> -----Original Message-----
> From: nvmewin-bounces at lists.openfabrics.org [mailto:
> nvmewin-bounces at lists.openfabrics.org] On Behalf Of suman prakash
> Sent: Friday, April 10, 2015 7:39 AM
> To: nvmewin at lists.openfabrics.org
> Subject: Re: [nvmewin] nvmewin Digest, Vol 40, Issue 3
>
> All,
>
> Samsung would like to take up the following 3 patches -
>
> 1. Feature - Optimizing driver initialization performance.
> Description - Instead of learning cores, use StorPortInitializePerfOpts()
> to get the array of MSI numbers corresponding to each logical processor and
> make the Llogical processor-Queue-MSIx mapping.
> Submitter - Samsung (Suman)
> Date - End of June
>
> 2. Feature - Handling device surprise removal in IOCTL path using Win 8.1
> Timers.
> Description - The existing implementation of detecting surprise removal
> during IO using the StorPortNotification(RequestTimerCall,..) does not work
> when IOCTL commands are outstanding. Starting from Windows 8 kernel,
> StorPortRequestTimer() is supported which handles the surprise removal in
> IOCTL path too. For Windows 7 kernel, this will be a limitation.
> Submitter - Samsung (Suman)
> Date - End of July
>
> 3. Feature - Device EOF support (Read Only) Description - Read only
> support for devices at EOL. Detection at init, hot plug, or run-time.
> Submitter - Samsung (Suman)
> Date - End of August
>
> Thanks,
> Suman
>
>
> -----Original Message-----
> From: nvmewin-bounces at lists.openfabrics.org
> [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of
> nvmewin-request at lists.openfabrics.org
> Sent: Friday, April 10, 2015 6:00 AM
> To: nvmewin at lists.openfabrics.org
> Subject: nvmewin Digest, Vol 40, Issue 3
>
> Send nvmewin mailing list submissions to
> nvmewin at lists.openfabrics.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://lists.openfabrics.org/mailman/listinfo/nvmewin
> or, via email, send a message with subject or body 'help' to
> nvmewin-request at lists.openfabrics.org
>
> You can reach the person managing the list at
> nvmewin-owner at lists.openfabrics.org
>
> When replying, please edit your Subject line so it is more specific than
> "Re: Contents of nvmewin digest..."
>
>
> Today's Topics:
>
> 1. OFA Release Plan: Release #1 2015 -- Call for patch
> volunteers (Robles, Raymond C)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Fri, 10 Apr 2015 00:29:00 +0000
> From: "Robles, Raymond C" <raymond.c.robles at intel.com>
> To: "nvmewin at lists.openfabrics.org" <nvmewin at lists.openfabrics.org>
> Subject: [nvmewin] OFA Release Plan: Release #1 2015 -- Call for patch
> volunteers
> Message-ID:
>
> <49158E750348AA499168FD41D88983606B63477F at fmsmsx117.amr.corp.intel.com>
>
> Content-Type: text/plain; charset="us-ascii"
>
> All,
>
> As we enter Q2, I'm happy to announce that we a have a very good feature
> plan in place for both Q2 and Q4 releases in 2015 of the OFA driver. The
> list of requested features is included below for final review.
>
> I would like to formally ask for volunteers for patch submissions. As of
> today, there are no volunteers for any of the below patches.
>
> Intel would like to start with the volunteering... we will volunteer to
> provide the namespace management patch.
>
> 2015 Release #1 (Q2/Q3 - 2015)
>
> Feature
>
> Description
>
> Submitter
>
> Date
>
> Concurrent Channels
>
> Start I/O Concurrent Channels allows StartIo to be called in parallel
> resulting in improved IOPS.
>
> <TBD>
>
> <TBD>
>
> Perf Opts
>
> Instead of learning cores, use perf opts. Shave off 2 seconds in init
> code... using Storport initialize perf opts. No need for source core
> interrupt steering code in driver.
>
> <TBD>
>
> <TBD>
>
> Win 8.1 Timers
>
> Storport Notification usage. Hot plug and IOCTLs, this does not work, for
> Win8.1... StorportRequestTimer() function needed.
>
> <TBD>
>
> <TBD>
>
> CFS Bit Monitoring
>
> CFS bit monitoring and handling. Look into adding additional handling code
> just to monitor CSTS.CFS.
>
> <TBD>
>
> <TBD>
>
> EOL - Read Only
>
> Read only support for devices at EOL. Detection at init, hot plug, or
> run-time.
>
> <TBD>
>
> <TBD>
>
> Namespace Mgmt
>
> Namespace management - creation and deletion of namespaces.
>
> Intel (Carolyn)
>
> End of May
>
> WHQL
>
> WHQL test suite run (any bug fixes).
>
> <TBD>
>
> <TBD>
>
>
>
> 2015 Release #2 (Q4 - 2015)
>
> Feature
>
> Description
>
> Submitter
>
> Date
>
> Multi-Path
>
> Multipath feature set from 1.1 spec (Multi-path I/O, Namespace Sharing,
> Reservations)
>
> <TBD>
>
> <TBD>
>
> RTD3
>
> Active/idle RTD3
>
> <TBD>
>
> <TBD>
>
> Temp Thresholds
>
> Temperature thresholds (Get/Set Feature and AER trigger)
>
> <TBD>
>
> <TBD>
>
> Live FW Update
>
> Live firmware update
>
> <TBD>
>
> <TBD>
>
> Atomicity
>
> Atomicity Enhancements
>
> <TBD>
>
> <TBD>
>
> Win 10
>
> Win 10 support (push to Q4 release)
>
> <TBD>
>
> <TBD>
>
> SNTL
>
> SNTL 1.5 doc updates
>
> <TBD>
>
> <TBD>
>
> WHQL
>
> WHQL test suite run (any bug fixes) - for each release of OFA
>
> <TBD>
>
> <TBD>
>
>
>
> Thanks,
> Ray
>
> [cid:image001.png at 01CB3870.4BB88E70]
> Raymond C. Robles
> NSG ISE Host Storage Software
> Intel Corporation
> Office: 480-554-2600
> Mobile: 480-399-0645
>
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL:
> <
> http://lists.openfabrics.org/pipermail/nvmewin/attachments/20150410/3fb7f51
> 9/attachment.html>
> -------------- next part --------------
> A non-text attachment was scrubbed...
> Name: image001.png
> Type: image/png
> Size: 1756 bytes
> Desc: image001.png
> URL:
> <
> http://lists.openfabrics.org/pipermail/nvmewin/attachments/20150410/3fb7f51
> 9/attachment.png>
>
> ------------------------------
>
> _______________________________________________
> nvmewin mailing list
> nvmewin at lists.openfabrics.org
> http://lists.openfabrics.org/mailman/listinfo/nvmewin
>
>
> End of nvmewin Digest, Vol 40, Issue 3
> **************************************
>
> _______________________________________________
> nvmewin mailing list
> nvmewin at lists.openfabrics.org
> http://lists.openfabrics.org/mailman/listinfo/nvmewin
>
>
> ------------------------------
>
> Message: 2
> Date: Mon, 13 Apr 2015 21:36:35 +0000
> From: Tom Fones-SSI <tom.fones at ssi.samsung.com>
> To: suman prakash <suman.p at samsung.com>,
> "nvmewin at lists.openfabrics.org" <nvmewin at lists.openfabrics.org>
> Subject: Re: [nvmewin] nvmewin
> Message-ID:
> <
> EEAA00CCA2889D429E22F360CC0B0DC105C868 at SSIEXCH-MB3.ssi.samsung.com>
> Content-Type: text/plain; charset="us-ascii"
>
>
> May I ask the community if anyone has tried to install and use a custom
> storage class driver
> (replacement for Disk.sys in Windows) on an NVMe disk, and if so have they
> had any success ?
>
> Thanks in advance,
> Tom Fones @ Samsung
>
>
> ------------------------------
>
> Message: 3
> Date: Mon, 13 Apr 2015 23:01:03 +0000
> From: "Robles, Raymond C" <raymond.c.robles at intel.com>
> To: 'Tom Fones-SSI' <tom.fones at ssi.samsung.com>, suman prakash
> <suman.p at samsung.com>, "nvmewin at lists.openfabrics.org"
> <nvmewin at lists.openfabrics.org>
> Subject: Re: [nvmewin] nvmewin
> Message-ID:
> <
> 49158E750348AA499168FD41D88983606B636AC9 at fmsmsx117.amr.corp.intel.com>
>
> Content-Type: text/plain; charset="us-ascii"
>
> I have not heard of any such experiments.
>
> Thanks,
> Ray
>
> -----Original Message-----
> From: nvmewin-bounces at lists.openfabrics.org [mailto:
> nvmewin-bounces at lists.openfabrics.org] On Behalf Of Tom Fones-SSI
> Sent: Monday, April 13, 2015 2:37 PM
> To: suman prakash; nvmewin at lists.openfabrics.org
> Subject: Re: [nvmewin] nvmewin
>
>
> May I ask the community if anyone has tried to install and use a custom
> storage class driver (replacement for Disk.sys in Windows) on an NVMe disk,
> and if so have they had any success ?
>
> Thanks in advance,
> Tom Fones @ Samsung
> _______________________________________________
> nvmewin mailing list
> nvmewin at lists.openfabrics.org
> http://lists.openfabrics.org/mailman/listinfo/nvmewin
>
>
> ------------------------------
>
> _______________________________________________
> nvmewin mailing list
> nvmewin at lists.openfabrics.org
> http://lists.openfabrics.org/mailman/listinfo/nvmewin
>
>
> End of nvmewin Digest, Vol 40, Issue 5
> **************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/nvmewin/attachments/20150415/53c8d484/attachment.html>
More information about the nvmewin
mailing list