From thomas.freeman at hgst.com Mon Feb 9 08:21:28 2015 From: thomas.freeman at hgst.com (Thomas Freeman) Date: Mon, 9 Feb 2015 16:21:28 +0000 Subject: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes In-Reply-To: <49158E750348AA499168FD41D88983606B5D2AC7@fmsmsx117.amr.corp.intel.com> References: <49158E750348AA499168FD41D88983606B5D2AC7@fmsmsx117.amr.corp.intel.com> Message-ID: HGST will commit to being a required reviewer. Unfortunately, we cannot commit to the required testing. Tom Freeman Software Engineer, Device Manager and Driver Development HGST, a Western Digital company Thomas.Freeman at hgst.com 507-322-2311 From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Robles, Raymond C Sent: Saturday, January 31, 2015 12:58 AM To: nvmewin at lists.openfabrics.org Cc: Gianopulos, Mark P; Olsson, Claes Subject: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes Meeting Minutes (Friday 1/23/15) - OFA NVMe Windows Driver Q1`2015 Meeting Agenda: - Review 1.4 Release (Kwok - PMC) - Confirmation of Ray and Carolyn as the chairperson and source maintainer (All) - Planning for 2016 and beyond (Ray) - AOB Opens: - PMC: Kwok nominated Intel reps Raymond Robles and Carolyn Foster to take over OFA chair and source maintainer role, respectively, for 2015. No objections from working group. Ray and Carolyn will take over roles effective immediately. A huge thanks to Kwok and Alex for all their hard work over the last year. - Dell: How will future releases be WHQL'd? 1.4 release was WHQL'd with PMC hardware. Identify a WHQL strategy moving forward for OFA driver (which hardware, who owns the WHQL certification process, price, etc.). - Intel: Confirm with OFA working group on any changes to driver releases, planning meetings, and cadence. Working group agreed to keep all planning items "as is". - Intel: Requested a replacement for LSI/Avago/Seagate as a mandatory reviewing member company. A vote was taken on whether a 3rd mandatory company was needed (for reviewing and testing). Approved. o Dell: Abstain o Intel: Yes o Samsung: Yes o PMC: Yes o HGST: Abstain - Intel: With the working group approving the need for a 3rd mandatory reviewing member company, a request for volunteers was made by Intel. Kwok suggested Samsung. HGST was suggested by Judy. o Samsung: Will follow up with management and notify OFA working group within 2 weeks (AR: Judy - 2 weeks) o HGST: Will follow up with management and notify OFA working group within 2 weeks (AR: Tom - 2 weeks) - Intel: Open discussion on features for 2015 OFA releases: o Intel: ? NVMe 1.1/1.2 features. Discussion on mandatory and optional features. Follow up discussion needed to close. ? Start I/O Concurrent Channels (potential performance increase on high core counts). o Samsung: ? 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. ? Storport Notification usage. Hot plug and IOCTLs, this does not work, for Win8.1... StorportRequestTimer() function needed. ? CFS bit monitoring and handling. Look into adding additional handling code just to monitor CSTS.CFS. ? Read only support for devices at EOL. Detection at init, hot plug, or run-time. AR's: - Dell (Austin): Identify initial WHQL strategy for 2015 (mid-February... next meeting to discuss). - Samsung (Judy): Follow up with management on becoming 3rd mandatory reviewing member company for OFA (2 weeks). o UPDATE (1/30/15): Samsung has agreed to become 3rd mandatory reviewing member company. Thanks Judy! - HGST (Ken): Follow up with management on becoming 3rd mandatory reviewing member company for OFA (2 weeks). - Intel (Ray): Provide preliminary list of 1.1/1.2 features, both mandatory and optional, to start discussion on what to include in 2015 releases. - Intel (Ray): Communicate with Ken Strandberg at OFA website for logistical changes (mid-February... next meeting). Next Meeting: - Mid February 2015... to plan for next release and finalize feature list for 2015 OFA driver releases. Thanks, Ray [cid:image001.png at 01CB3870.4BB88E70] Raymond C. Robles Host Storage System Software Non-Volatile Memory Solutions Group Intel Corporation Office: 480-554-2600 Mobile: 480-399-0645 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1756 bytes Desc: image001.png URL: From raymond.c.robles at intel.com Mon Feb 9 16:51:17 2015 From: raymond.c.robles at intel.com (Robles, Raymond C) Date: Tue, 10 Feb 2015 00:51:17 +0000 Subject: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes In-Reply-To: References: <49158E750348AA499168FD41D88983606B5D2AC7@fmsmsx117.amr.corp.intel.com> Message-ID: <49158E750348AA499168FD41D88983606B5D6DF7@fmsmsx117.amr.corp.intel.com> Hi Tom, Thank you for responding and offering to be a required reviewer. It is perfectly understandable on the required testing component. NVMe is still very much in the early stages. As noted in my meeting minutes from our initial 2015 meeting, Samsung has agreed to be a reviewer and tester, if approved upon by the working group. I will send out a meeting invite in the next day or two and we can put forth the vote on Samsung. Again, thank you for following up. Thanks, Ray From: Thomas Freeman [mailto:thomas.freeman at hgst.com] Sent: Monday, February 09, 2015 9:21 AM To: Robles, Raymond C; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: RE: OFA NVMe Windows Driver Q12015 - Meeting Minutes HGST will commit to being a required reviewer. Unfortunately, we cannot commit to the required testing. Tom Freeman Software Engineer, Device Manager and Driver Development HGST, a Western Digital company Thomas.Freeman at hgst.com 507-322-2311 From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Robles, Raymond C Sent: Saturday, January 31, 2015 12:58 AM To: nvmewin at lists.openfabrics.org Cc: Gianopulos, Mark P; Olsson, Claes Subject: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes Meeting Minutes (Friday 1/23/15) - OFA NVMe Windows Driver Q1`2015 Meeting Agenda: - Review 1.4 Release (Kwok - PMC) - Confirmation of Ray and Carolyn as the chairperson and source maintainer (All) - Planning for 2016 and beyond (Ray) - AOB Opens: - PMC: Kwok nominated Intel reps Raymond Robles and Carolyn Foster to take over OFA chair and source maintainer role, respectively, for 2015. No objections from working group. Ray and Carolyn will take over roles effective immediately. A huge thanks to Kwok and Alex for all their hard work over the last year. - Dell: How will future releases be WHQL'd? 1.4 release was WHQL'd with PMC hardware. Identify a WHQL strategy moving forward for OFA driver (which hardware, who owns the WHQL certification process, price, etc.). - Intel: Confirm with OFA working group on any changes to driver releases, planning meetings, and cadence. Working group agreed to keep all planning items "as is". - Intel: Requested a replacement for LSI/Avago/Seagate as a mandatory reviewing member company. A vote was taken on whether a 3rd mandatory company was needed (for reviewing and testing). Approved. o Dell: Abstain o Intel: Yes o Samsung: Yes o PMC: Yes o HGST: Abstain - Intel: With the working group approving the need for a 3rd mandatory reviewing member company, a request for volunteers was made by Intel. Kwok suggested Samsung. HGST was suggested by Judy. o Samsung: Will follow up with management and notify OFA working group within 2 weeks (AR: Judy - 2 weeks) o HGST: Will follow up with management and notify OFA working group within 2 weeks (AR: Tom - 2 weeks) - Intel: Open discussion on features for 2015 OFA releases: o Intel: ? NVMe 1.1/1.2 features. Discussion on mandatory and optional features. Follow up discussion needed to close. ? Start I/O Concurrent Channels (potential performance increase on high core counts). o Samsung: ? 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. ? Storport Notification usage. Hot plug and IOCTLs, this does not work, for Win8.1... StorportRequestTimer() function needed. ? CFS bit monitoring and handling. Look into adding additional handling code just to monitor CSTS.CFS. ? Read only support for devices at EOL. Detection at init, hot plug, or run-time. AR's: - Dell (Austin): Identify initial WHQL strategy for 2015 (mid-February... next meeting to discuss). - Samsung (Judy): Follow up with management on becoming 3rd mandatory reviewing member company for OFA (2 weeks). o UPDATE (1/30/15): Samsung has agreed to become 3rd mandatory reviewing member company. Thanks Judy! - HGST (Ken): Follow up with management on becoming 3rd mandatory reviewing member company for OFA (2 weeks). - Intel (Ray): Provide preliminary list of 1.1/1.2 features, both mandatory and optional, to start discussion on what to include in 2015 releases. - Intel (Ray): Communicate with Ken Strandberg at OFA website for logistical changes (mid-February... next meeting). Next Meeting: - Mid February 2015... to plan for next release and finalize feature list for 2015 OFA driver releases. Thanks, Ray [cid:image001.png at 01CB3870.4BB88E70] Raymond C. Robles Host Storage System Software Non-Volatile Memory Solutions Group Intel Corporation Office: 480-554-2600 Mobile: 480-399-0645 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1756 bytes Desc: image001.png URL: From raymond.c.robles at intel.com Mon Feb 9 17:09:51 2015 From: raymond.c.robles at intel.com (Robles, Raymond C) Date: Tue, 10 Feb 2015 01:09:51 +0000 Subject: [nvmewin] NVMe 1.1 and 1.2 Feature Breakdown (with M/O definition) Message-ID: <49158E750348AA499168FD41D88983606B5D6E5C@fmsmsx117.amr.corp.intel.com> Hello, Here is the NVMe 1.1/1.2 feature breakdown that I had previously done. The list is high level (taken straight from an NVMe working group press release) and I've added descriptions, potential work required, and where the feature is mandatory or optional. We can use this as a starting point for our next OFA meeting. Note there might be additional sub-features or smaller additions, but I chose to leave those out. Please feel free to discuss, add/subtract, or correct any mistakes I may have made in this document (either on this email distribution list or directly with me). I will send out a meeting notice in the next day or two. My hope is that will start the proper discussion around which features to include in the OFA releases for 2015. Thanks, Ray [cid:image001.png at 01CB3870.4BB88E70] Raymond C. Robles Host Storage System Software Non-Volatile Memory Solutions Group Intel Corporation Office: 480-554-2600 Mobile: 480-399-0645 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1756 bytes Desc: image001.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: NVMe 1.1 and 1.2 Feature Support - OFA Driver.xlsx Type: application/vnd.openxmlformats-officedocument.spreadsheetml.sheet Size: 16477 bytes Desc: NVMe 1.1 and 1.2 Feature Support - OFA Driver.xlsx URL: From judy.brock at ssi.samsung.com Mon Feb 9 17:11:20 2015 From: judy.brock at ssi.samsung.com (Judy Brock-SSI) Date: Tue, 10 Feb 2015 01:11:20 +0000 Subject: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes In-Reply-To: <49158E750348AA499168FD41D88983606B5D6DF7@fmsmsx117.amr.corp.intel.com> References: <49158E750348AA499168FD41D88983606B5D2AC7@fmsmsx117.amr.corp.intel.com> <49158E750348AA499168FD41D88983606B5D6DF7@fmsmsx117.amr.corp.intel.com> Message-ID: <36E8D38D6B771A4BBDB1C0D800158A516B69B5D5@SSIEXCH-MB3.ssi.samsung.com> It might not hurt to have a fourth required reviewer - the more eyes the better. Perhaps we could vote on that as well? Thanks, Judy From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Robles, Raymond C Sent: Monday, February 09, 2015 4:51 PM To: Thomas Freeman; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: Re: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes Hi Tom, Thank you for responding and offering to be a required reviewer. It is perfectly understandable on the required testing component. NVMe is still very much in the early stages. As noted in my meeting minutes from our initial 2015 meeting, Samsung has agreed to be a reviewer and tester, if approved upon by the working group. I will send out a meeting invite in the next day or two and we can put forth the vote on Samsung. Again, thank you for following up. Thanks, Ray From: Thomas Freeman [mailto:thomas.freeman at hgst.com] Sent: Monday, February 09, 2015 9:21 AM To: Robles, Raymond C; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: RE: OFA NVMe Windows Driver Q12015 - Meeting Minutes HGST will commit to being a required reviewer. Unfortunately, we cannot commit to the required testing. Tom Freeman Software Engineer, Device Manager and Driver Development HGST, a Western Digital company Thomas.Freeman at hgst.com 507-322-2311 From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Robles, Raymond C Sent: Saturday, January 31, 2015 12:58 AM To: nvmewin at lists.openfabrics.org Cc: Gianopulos, Mark P; Olsson, Claes Subject: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes Meeting Minutes (Friday 1/23/15) - OFA NVMe Windows Driver Q1`2015 Meeting Agenda: - Review 1.4 Release (Kwok - PMC) - Confirmation of Ray and Carolyn as the chairperson and source maintainer (All) - Planning for 2016 and beyond (Ray) - AOB Opens: - PMC: Kwok nominated Intel reps Raymond Robles and Carolyn Foster to take over OFA chair and source maintainer role, respectively, for 2015. No objections from working group. Ray and Carolyn will take over roles effective immediately. A huge thanks to Kwok and Alex for all their hard work over the last year. - Dell: How will future releases be WHQL'd? 1.4 release was WHQL'd with PMC hardware. Identify a WHQL strategy moving forward for OFA driver (which hardware, who owns the WHQL certification process, price, etc.). - Intel: Confirm with OFA working group on any changes to driver releases, planning meetings, and cadence. Working group agreed to keep all planning items "as is". - Intel: Requested a replacement for LSI/Avago/Seagate as a mandatory reviewing member company. A vote was taken on whether a 3rd mandatory company was needed (for reviewing and testing). Approved. o Dell: Abstain o Intel: Yes o Samsung: Yes o PMC: Yes o HGST: Abstain - Intel: With the working group approving the need for a 3rd mandatory reviewing member company, a request for volunteers was made by Intel. Kwok suggested Samsung. HGST was suggested by Judy. o Samsung: Will follow up with management and notify OFA working group within 2 weeks (AR: Judy - 2 weeks) o HGST: Will follow up with management and notify OFA working group within 2 weeks (AR: Tom - 2 weeks) - Intel: Open discussion on features for 2015 OFA releases: o Intel: * NVMe 1.1/1.2 features. Discussion on mandatory and optional features. Follow up discussion needed to close. * Start I/O Concurrent Channels (potential performance increase on high core counts). o Samsung: * 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. * Storport Notification usage. Hot plug and IOCTLs, this does not work, for Win8.1... StorportRequestTimer() function needed. * CFS bit monitoring and handling. Look into adding additional handling code just to monitor CSTS.CFS. * Read only support for devices at EOL. Detection at init, hot plug, or run-time. AR's: - Dell (Austin): Identify initial WHQL strategy for 2015 (mid-February... next meeting to discuss). - Samsung (Judy): Follow up with management on becoming 3rd mandatory reviewing member company for OFA (2 weeks). o UPDATE (1/30/15): Samsung has agreed to become 3rd mandatory reviewing member company. Thanks Judy! - HGST (Ken): Follow up with management on becoming 3rd mandatory reviewing member company for OFA (2 weeks). - Intel (Ray): Provide preliminary list of 1.1/1.2 features, both mandatory and optional, to start discussion on what to include in 2015 releases. - Intel (Ray): Communicate with Ken Strandberg at OFA website for logistical changes (mid-February... next meeting). Next Meeting: - Mid February 2015... to plan for next release and finalize feature list for 2015 OFA driver releases. Thanks, Ray [cid:image001.png at 01CB3870.4BB88E70] Raymond C. Robles Host Storage System Software Non-Volatile Memory Solutions Group Intel Corporation Office: 480-554-2600 Mobile: 480-399-0645 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1756 bytes Desc: image001.png URL: From raymond.c.robles at intel.com Mon Feb 9 17:13:03 2015 From: raymond.c.robles at intel.com (Robles, Raymond C) Date: Tue, 10 Feb 2015 01:13:03 +0000 Subject: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes In-Reply-To: <36E8D38D6B771A4BBDB1C0D800158A516B69B5D5@SSIEXCH-MB3.ssi.samsung.com> References: <49158E750348AA499168FD41D88983606B5D2AC7@fmsmsx117.amr.corp.intel.com> <49158E750348AA499168FD41D88983606B5D6DF7@fmsmsx117.amr.corp.intel.com> <36E8D38D6B771A4BBDB1C0D800158A516B69B5D5@SSIEXCH-MB3.ssi.samsung.com> Message-ID: <49158E750348AA499168FD41D88983606B5D6E84@fmsmsx117.amr.corp.intel.com> Great idea! Let's definitely discuss and/or vote. Thanks, Ray From: Judy Brock-SSI [mailto:judy.brock at ssi.samsung.com] Sent: Monday, February 09, 2015 6:11 PM To: Robles, Raymond C; Thomas Freeman; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: RE: OFA NVMe Windows Driver Q12015 - Meeting Minutes It might not hurt to have a fourth required reviewer - the more eyes the better. Perhaps we could vote on that as well? Thanks, Judy From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Robles, Raymond C Sent: Monday, February 09, 2015 4:51 PM To: Thomas Freeman; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: Re: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes Hi Tom, Thank you for responding and offering to be a required reviewer. It is perfectly understandable on the required testing component. NVMe is still very much in the early stages. As noted in my meeting minutes from our initial 2015 meeting, Samsung has agreed to be a reviewer and tester, if approved upon by the working group. I will send out a meeting invite in the next day or two and we can put forth the vote on Samsung. Again, thank you for following up. Thanks, Ray From: Thomas Freeman [mailto:thomas.freeman at hgst.com] Sent: Monday, February 09, 2015 9:21 AM To: Robles, Raymond C; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: RE: OFA NVMe Windows Driver Q12015 - Meeting Minutes HGST will commit to being a required reviewer. Unfortunately, we cannot commit to the required testing. Tom Freeman Software Engineer, Device Manager and Driver Development HGST, a Western Digital company Thomas.Freeman at hgst.com 507-322-2311 From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Robles, Raymond C Sent: Saturday, January 31, 2015 12:58 AM To: nvmewin at lists.openfabrics.org Cc: Gianopulos, Mark P; Olsson, Claes Subject: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes Meeting Minutes (Friday 1/23/15) - OFA NVMe Windows Driver Q1`2015 Meeting Agenda: - Review 1.4 Release (Kwok - PMC) - Confirmation of Ray and Carolyn as the chairperson and source maintainer (All) - Planning for 2016 and beyond (Ray) - AOB Opens: - PMC: Kwok nominated Intel reps Raymond Robles and Carolyn Foster to take over OFA chair and source maintainer role, respectively, for 2015. No objections from working group. Ray and Carolyn will take over roles effective immediately. A huge thanks to Kwok and Alex for all their hard work over the last year. - Dell: How will future releases be WHQL'd? 1.4 release was WHQL'd with PMC hardware. Identify a WHQL strategy moving forward for OFA driver (which hardware, who owns the WHQL certification process, price, etc.). - Intel: Confirm with OFA working group on any changes to driver releases, planning meetings, and cadence. Working group agreed to keep all planning items "as is". - Intel: Requested a replacement for LSI/Avago/Seagate as a mandatory reviewing member company. A vote was taken on whether a 3rd mandatory company was needed (for reviewing and testing). Approved. o Dell: Abstain o Intel: Yes o Samsung: Yes o PMC: Yes o HGST: Abstain - Intel: With the working group approving the need for a 3rd mandatory reviewing member company, a request for volunteers was made by Intel. Kwok suggested Samsung. HGST was suggested by Judy. o Samsung: Will follow up with management and notify OFA working group within 2 weeks (AR: Judy - 2 weeks) o HGST: Will follow up with management and notify OFA working group within 2 weeks (AR: Tom - 2 weeks) - Intel: Open discussion on features for 2015 OFA releases: o Intel: * NVMe 1.1/1.2 features. Discussion on mandatory and optional features. Follow up discussion needed to close. * Start I/O Concurrent Channels (potential performance increase on high core counts). o Samsung: * 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. * Storport Notification usage. Hot plug and IOCTLs, this does not work, for Win8.1... StorportRequestTimer() function needed. * CFS bit monitoring and handling. Look into adding additional handling code just to monitor CSTS.CFS. * Read only support for devices at EOL. Detection at init, hot plug, or run-time. AR's: - Dell (Austin): Identify initial WHQL strategy for 2015 (mid-February... next meeting to discuss). - Samsung (Judy): Follow up with management on becoming 3rd mandatory reviewing member company for OFA (2 weeks). o UPDATE (1/30/15): Samsung has agreed to become 3rd mandatory reviewing member company. Thanks Judy! - HGST (Ken): Follow up with management on becoming 3rd mandatory reviewing member company for OFA (2 weeks). - Intel (Ray): Provide preliminary list of 1.1/1.2 features, both mandatory and optional, to start discussion on what to include in 2015 releases. - Intel (Ray): Communicate with Ken Strandberg at OFA website for logistical changes (mid-February... next meeting). Next Meeting: - Mid February 2015... to plan for next release and finalize feature list for 2015 OFA driver releases. Thanks, Ray [cid:image001.png at 01CB3870.4BB88E70] Raymond C. Robles Host Storage System Software Non-Volatile Memory Solutions Group Intel Corporation Office: 480-554-2600 Mobile: 480-399-0645 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1756 bytes Desc: image001.png URL: From Uma.Parepalli at skhms.com Mon Feb 9 18:04:11 2015 From: Uma.Parepalli at skhms.com (Uma Parepalli) Date: Tue, 10 Feb 2015 02:04:11 +0000 Subject: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes In-Reply-To: <49158E750348AA499168FD41D88983606B5D6E84@fmsmsx117.amr.corp.intel.com> References: <49158E750348AA499168FD41D88983606B5D2AC7@fmsmsx117.amr.corp.intel.com> <49158E750348AA499168FD41D88983606B5D6DF7@fmsmsx117.amr.corp.intel.com> <36E8D38D6B771A4BBDB1C0D800158A516B69B5D5@SSIEXCH-MB3.ssi.samsung.com> <49158E750348AA499168FD41D88983606B5D6E84@fmsmsx117.amr.corp.intel.com> Message-ID: <5420a8bf4f9241a6aacef25c28feb6bc@N111XMB0240.skhms.com> I may or may not attend the next Windows driver meeting. However, I have observed the following with the last release: Disk format using Windows 7 driver on both Windows 7 and Windows 8 is faster than using Windows 8 driver on Windows 8 and it is slower than Windows 8.1/2012 R2 inbox drivers. I am not sure whether anyone is using Windows 8 at this time, since Windows 8.1 and Windows Server 2012 R2 are the current versions. Please test and fix the issue with the Windows 8 driver on Windows 8 (not 8.1) before next release and the community driver should be on par with the inbox drivers. Thank you, Uma Uma Parepalli ఉమా పారేపల్లి 우마 파레팔리 uma.parepalli at skhms.com Cell: 408 805 9260 SK Hynix Memory Solutions, 3103 N 1st St, San Jose, CA 95134 [Description: http://www.skhms.com/wp-content/uploads/2013/05/web-logo.png] From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Robles, Raymond C Sent: Monday, February 09, 2015 5:13 PM To: Judy Brock-SSI; Thomas Freeman; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: Re: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes Great idea! Let’s definitely discuss and/or vote. Thanks, Ray From: Judy Brock-SSI [mailto:judy.brock at ssi.samsung.com] Sent: Monday, February 09, 2015 6:11 PM To: Robles, Raymond C; Thomas Freeman; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: RE: OFA NVMe Windows Driver Q12015 - Meeting Minutes It might not hurt to have a fourth required reviewer – the more eyes the better. Perhaps we could vote on that as well? Thanks, Judy From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Robles, Raymond C Sent: Monday, February 09, 2015 4:51 PM To: Thomas Freeman; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: Re: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes Hi Tom, Thank you for responding and offering to be a required reviewer. It is perfectly understandable on the required testing component. NVMe is still very much in the early stages. As noted in my meeting minutes from our initial 2015 meeting, Samsung has agreed to be a reviewer and tester, if approved upon by the working group. I will send out a meeting invite in the next day or two and we can put forth the vote on Samsung. Again, thank you for following up. Thanks, Ray From: Thomas Freeman [mailto:thomas.freeman at hgst.com] Sent: Monday, February 09, 2015 9:21 AM To: Robles, Raymond C; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: RE: OFA NVMe Windows Driver Q12015 - Meeting Minutes HGST will commit to being a required reviewer. Unfortunately, we cannot commit to the required testing. Tom Freeman Software Engineer, Device Manager and Driver Development HGST, a Western Digital company Thomas.Freeman at hgst.com 507-322-2311 From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Robles, Raymond C Sent: Saturday, January 31, 2015 12:58 AM To: nvmewin at lists.openfabrics.org Cc: Gianopulos, Mark P; Olsson, Claes Subject: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes Meeting Minutes (Friday 1/23/15) – OFA NVMe Windows Driver Q1`2015 Meeting Agenda: - Review 1.4 Release (Kwok – PMC) - Confirmation of Ray and Carolyn as the chairperson and source maintainer (All) - Planning for 2016 and beyond (Ray) - AOB Opens: - PMC: Kwok nominated Intel reps Raymond Robles and Carolyn Foster to take over OFA chair and source maintainer role, respectively, for 2015. No objections from working group. Ray and Carolyn will take over roles effective immediately. A huge thanks to Kwok and Alex for all their hard work over the last year. - Dell: How will future releases be WHQL’d? 1.4 release was WHQL’d with PMC hardware. Identify a WHQL strategy moving forward for OFA driver (which hardware, who owns the WHQL certification process, price, etc.). - Intel: Confirm with OFA working group on any changes to driver releases, planning meetings, and cadence. Working group agreed to keep all planning items “as is”. - Intel: Requested a replacement for LSI/Avago/Seagate as a mandatory reviewing member company. A vote was taken on whether a 3rd mandatory company was needed (for reviewing and testing). Approved. o Dell: Abstain o Intel: Yes o Samsung: Yes o PMC: Yes o HGST: Abstain - Intel: With the working group approving the need for a 3rd mandatory reviewing member company, a request for volunteers was made by Intel. Kwok suggested Samsung. HGST was suggested by Judy. o Samsung: Will follow up with management and notify OFA working group within 2 weeks (AR: Judy - 2 weeks) o HGST: Will follow up with management and notify OFA working group within 2 weeks (AR: Tom - 2 weeks) - Intel: Open discussion on features for 2015 OFA releases: o Intel: • NVMe 1.1/1.2 features. Discussion on mandatory and optional features. Follow up discussion needed to close. • Start I/O Concurrent Channels (potential performance increase on high core counts). o Samsung: • 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. • Storport Notification usage. Hot plug and IOCTLs, this does not work, for Win8.1… StorportRequestTimer() function needed. • CFS bit monitoring and handling. Look into adding additional handling code just to monitor CSTS.CFS. • Read only support for devices at EOL. Detection at init, hot plug, or run-time. AR’s: - Dell (Austin): Identify initial WHQL strategy for 2015 (mid-February… next meeting to discuss). - Samsung (Judy): Follow up with management on becoming 3rd mandatory reviewing member company for OFA (2 weeks). o UPDATE (1/30/15): Samsung has agreed to become 3rd mandatory reviewing member company. Thanks Judy! - HGST (Ken): Follow up with management on becoming 3rd mandatory reviewing member company for OFA (2 weeks). - Intel (Ray): Provide preliminary list of 1.1/1.2 features, both mandatory and optional, to start discussion on what to include in 2015 releases. - Intel (Ray): Communicate with Ken Strandberg at OFA website for logistical changes (mid-February… next meeting). Next Meeting: - Mid February 2015… to plan for next release and finalize feature list for 2015 OFA driver releases. Thanks, Ray [cid:image001.png at 01CB3870.4BB88E70] Raymond C. Robles Host Storage System Software Non-Volatile Memory Solutions Group Intel Corporation Office: 480-554-2600 Mobile: 480-399-0645 The information contained in this e-mail is considered confidential of SK hynix memory solutions Inc. and intended only for the persons addressed or copied in this e-mail. Any unauthorized use, dissemination of the information, or copying of this message is strictly prohibited. If you are not the intended recipient, please contact the sender immediately and permanently delete the original and any copies of this email. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image002.png Type: image/png Size: 6439 bytes Desc: image002.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image003.png Type: image/png Size: 1756 bytes Desc: image003.png URL: From Uma.Parepalli at skhms.com Mon Feb 9 18:13:16 2015 From: Uma.Parepalli at skhms.com (Uma Parepalli) Date: Tue, 10 Feb 2015 02:13:16 +0000 Subject: [nvmewin] NVMe 1.1 and 1.2 Feature Breakdown (with M/O definition) In-Reply-To: <49158E750348AA499168FD41D88983606B5D6E5C@fmsmsx117.amr.corp.intel.com> References: <49158E750348AA499168FD41D88983606B5D6E5C@fmsmsx117.amr.corp.intel.com> Message-ID: <006590bc82254112a2c9bacf90c0d531@N111XMB0240.skhms.com> Thank you Raymond, We can use this as template for future releases. Uma Uma Parepalli ఉమా పారేపల్లి 우마 파레팔리 uma.parepalli at skhms.com Cell: 408 805 9260 SK Hynix Memory Solutions, 3103 N 1st St, San Jose, CA 95134 [Description: http://www.skhms.com/wp-content/uploads/2013/05/web-logo.png] From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Robles, Raymond C Sent: Monday, February 09, 2015 5:10 PM To: nvmewin at lists.openfabrics.org Subject: [nvmewin] NVMe 1.1 and 1.2 Feature Breakdown (with M/O definition) Hello, Here is the NVMe 1.1/1.2 feature breakdown that I had previously done. The list is high level (taken straight from an NVMe working group press release) and I’ve added descriptions, potential work required, and where the feature is mandatory or optional. We can use this as a starting point for our next OFA meeting. Note there might be additional sub-features or smaller additions, but I chose to leave those out. Please feel free to discuss, add/subtract, or correct any mistakes I may have made in this document (either on this email distribution list or directly with me). I will send out a meeting notice in the next day or two. My hope is that will start the proper discussion around which features to include in the OFA releases for 2015. Thanks, Ray [cid:image001.png at 01CB3870.4BB88E70] Raymond C. Robles Host Storage System Software Non-Volatile Memory Solutions Group Intel Corporation Office: 480-554-2600 Mobile: 480-399-0645 The information contained in this e-mail is considered confidential of SK hynix memory solutions Inc. and intended only for the persons addressed or copied in this e-mail. Any unauthorized use, dissemination of the information, or copying of this message is strictly prohibited. If you are not the intended recipient, please contact the sender immediately and permanently delete the original and any copies of this email. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image002.png Type: image/png Size: 6439 bytes Desc: image002.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image003.png Type: image/png Size: 1756 bytes Desc: image003.png URL: From wuwq85 at gmail.com Wed Feb 11 15:58:27 2015 From: wuwq85 at gmail.com (Wenqian Wu) Date: Wed, 11 Feb 2015 15:58:27 -0800 Subject: [nvmewin] NVMe Queue entry number Message-ID: Hi OFA driver member, I have one question regarding the Queue entry size. The driver will allocate number of entries aligned to memory page (line877, nvmeInit.c), instead of the actual queue size controller supports (CAP.MQES). Controller can return error if host request more entries while ignoring controller's capability. I think as long as the base address is page aligned, there is no reason to make the entries number aligned to page boundary. Can this be considered a driver bug or is there any particular consideration? Thanks, Wenqian -------------- next part -------------- An HTML attachment was scrubbed... URL: From raymond.c.robles at intel.com Thu Feb 12 11:56:26 2015 From: raymond.c.robles at intel.com (Robles, Raymond C) Date: Thu, 12 Feb 2015 19:56:26 +0000 Subject: [nvmewin] OFA NVMe Release Planning Meeting #2 (Placeholder) Message-ID: <49158E750348AA499168FD41D88983606B5D97ED@fmsmsx117.amr.corp.intel.com> Agenda: TBD ......................................................................................................................................... --> Join Lync Meeting Join by phone +1(916)356-2663 (or your local bridge access #) Choose bridge 5. (Global) English (United States) Find a local number Conference ID: 823200407 Forgot your dial-in PIN? |Help [!OC([1033])!] ......................................................................................................................................... -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/calendar Size: 2286 bytes Desc: not available URL: From raymond.c.robles at intel.com Thu Feb 12 20:42:37 2015 From: raymond.c.robles at intel.com (Robles, Raymond C) Date: Fri, 13 Feb 2015 04:42:37 +0000 Subject: [nvmewin] NVMe Queue entry number In-Reply-To: References: Message-ID: <49158E750348AA499168FD41D88983606B5DA191@fmsmsx117.amr.corp.intel.com> Hi Wenqian, Thank you for your inquiry. In the FindAdapter routine (in nvmeStd.c), the driver checks CAP.MQES and saves the value in pAE->InitInfo.IoQEntries (if the default or register override is smaller than CAP.MQES, it is saved instead). Within the function NVMeAllocIoQueues, there is loop that will iterate to create queue pairs based on the number of cpu cores. In this loop, a call to NVMeAllocQueues is made. Just prior to this call, the value saved in pAE->InitInfo.IoQEntires is retrieved (stack variable “QEntres”) and passed in as the 3rd parameter. So, once the allocation is taking place where you mention below, the 3rd parameter of the function has already been checked against CAP.MQES. Also, per the NVMe spec sections 5.3 and 5.4 – Figures 33 and 37 (create completion queue and submission queue PRP 1), all queue memory must be “physically contiguous and memory page aligned”. Thanks Ray From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Wenqian Wu Sent: Wednesday, February 11, 2015 4:58 PM To: nvmewin at lists.openfabrics.org Subject: [nvmewin] NVMe Queue entry number Hi OFA driver member, I have one question regarding the Queue entry size. The driver will allocate number of entries aligned to memory page (line877, nvmeInit.c), instead of the actual queue size controller supports (CAP.MQES). Controller can return error if host request more entries while ignoring controller's capability. I think as long as the base address is page aligned, there is no reason to make the entries number aligned to page boundary. Can this be considered a driver bug or is there any particular consideration? Thanks, Wenqian -------------- next part -------------- An HTML attachment was scrubbed... URL: From raymond.c.robles at intel.com Fri Feb 13 12:41:21 2015 From: raymond.c.robles at intel.com (Robles, Raymond C) Date: Fri, 13 Feb 2015 20:41:21 +0000 Subject: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes In-Reply-To: <5420a8bf4f9241a6aacef25c28feb6bc@N111XMB0240.skhms.com> References: <49158E750348AA499168FD41D88983606B5D2AC7@fmsmsx117.amr.corp.intel.com> <49158E750348AA499168FD41D88983606B5D6DF7@fmsmsx117.amr.corp.intel.com> <36E8D38D6B771A4BBDB1C0D800158A516B69B5D5@SSIEXCH-MB3.ssi.samsung.com> <49158E750348AA499168FD41D88983606B5D6E84@fmsmsx117.amr.corp.intel.com> <5420a8bf4f9241a6aacef25c28feb6bc@N111XMB0240.skhms.com> Message-ID: <49158E750348AA499168FD41D88983606B5DA5FF@fmsmsx117.amr.corp.intel.com> Hello Uma, Thank you for your email. We always encourage others to submit patches for issues they have discovered. We would welcome a patch from you to help resolve any issues you’ve discovered. Thanks, Ray From: Uma Parepalli [mailto:Uma.Parepalli at skhms.com] Sent: Monday, February 09, 2015 7:04 PM To: Robles, Raymond C; Judy Brock-SSI; Thomas Freeman; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: RE: OFA NVMe Windows Driver Q12015 - Meeting Minutes I may or may not attend the next Windows driver meeting. However, I have observed the following with the last release: Disk format using Windows 7 driver on both Windows 7 and Windows 8 is faster than using Windows 8 driver on Windows 8 and it is slower than Windows 8.1/2012 R2 inbox drivers. I am not sure whether anyone is using Windows 8 at this time, since Windows 8.1 and Windows Server 2012 R2 are the current versions. Please test and fix the issue with the Windows 8 driver on Windows 8 (not 8.1) before next release and the community driver should be on par with the inbox drivers. Thank you, Uma Uma Parepalli ఉమా పారేపల్లి 우마 파레팔리 uma.parepalli at skhms.com Cell: 408 805 9260 SK Hynix Memory Solutions, 3103 N 1st St, San Jose, CA 95134 [Description: http://www.skhms.com/wp-content/uploads/2013/05/web-logo.png] From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Robles, Raymond C Sent: Monday, February 09, 2015 5:13 PM To: Judy Brock-SSI; Thomas Freeman; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: Re: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes Great idea! Let’s definitely discuss and/or vote. Thanks, Ray From: Judy Brock-SSI [mailto:judy.brock at ssi.samsung.com] Sent: Monday, February 09, 2015 6:11 PM To: Robles, Raymond C; Thomas Freeman; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: RE: OFA NVMe Windows Driver Q12015 - Meeting Minutes It might not hurt to have a fourth required reviewer – the more eyes the better. Perhaps we could vote on that as well? Thanks, Judy From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Robles, Raymond C Sent: Monday, February 09, 2015 4:51 PM To: Thomas Freeman; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: Re: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes Hi Tom, Thank you for responding and offering to be a required reviewer. It is perfectly understandable on the required testing component. NVMe is still very much in the early stages. As noted in my meeting minutes from our initial 2015 meeting, Samsung has agreed to be a reviewer and tester, if approved upon by the working group. I will send out a meeting invite in the next day or two and we can put forth the vote on Samsung. Again, thank you for following up. Thanks, Ray From: Thomas Freeman [mailto:thomas.freeman at hgst.com] Sent: Monday, February 09, 2015 9:21 AM To: Robles, Raymond C; nvmewin at lists.openfabrics.org Cc: Sean Brownlow Subject: RE: OFA NVMe Windows Driver Q12015 - Meeting Minutes HGST will commit to being a required reviewer. Unfortunately, we cannot commit to the required testing. Tom Freeman Software Engineer, Device Manager and Driver Development HGST, a Western Digital company Thomas.Freeman at hgst.com 507-322-2311 From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Robles, Raymond C Sent: Saturday, January 31, 2015 12:58 AM To: nvmewin at lists.openfabrics.org Cc: Gianopulos, Mark P; Olsson, Claes Subject: [nvmewin] OFA NVMe Windows Driver Q12015 - Meeting Minutes Meeting Minutes (Friday 1/23/15) – OFA NVMe Windows Driver Q1`2015 Meeting Agenda: - Review 1.4 Release (Kwok – PMC) - Confirmation of Ray and Carolyn as the chairperson and source maintainer (All) - Planning for 2016 and beyond (Ray) - AOB Opens: - PMC: Kwok nominated Intel reps Raymond Robles and Carolyn Foster to take over OFA chair and source maintainer role, respectively, for 2015. No objections from working group. Ray and Carolyn will take over roles effective immediately. A huge thanks to Kwok and Alex for all their hard work over the last year. - Dell: How will future releases be WHQL’d? 1.4 release was WHQL’d with PMC hardware. Identify a WHQL strategy moving forward for OFA driver (which hardware, who owns the WHQL certification process, price, etc.). - Intel: Confirm with OFA working group on any changes to driver releases, planning meetings, and cadence. Working group agreed to keep all planning items “as is”. - Intel: Requested a replacement for LSI/Avago/Seagate as a mandatory reviewing member company. A vote was taken on whether a 3rd mandatory company was needed (for reviewing and testing). Approved. o Dell: Abstain o Intel: Yes o Samsung: Yes o PMC: Yes o HGST: Abstain - Intel: With the working group approving the need for a 3rd mandatory reviewing member company, a request for volunteers was made by Intel. Kwok suggested Samsung. HGST was suggested by Judy. o Samsung: Will follow up with management and notify OFA working group within 2 weeks (AR: Judy - 2 weeks) o HGST: Will follow up with management and notify OFA working group within 2 weeks (AR: Tom - 2 weeks) - Intel: Open discussion on features for 2015 OFA releases: o Intel: • NVMe 1.1/1.2 features. Discussion on mandatory and optional features. Follow up discussion needed to close. • Start I/O Concurrent Channels (potential performance increase on high core counts). o Samsung: • 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. • Storport Notification usage. Hot plug and IOCTLs, this does not work, for Win8.1… StorportRequestTimer() function needed. • CFS bit monitoring and handling. Look into adding additional handling code just to monitor CSTS.CFS. • Read only support for devices at EOL. Detection at init, hot plug, or run-time. AR’s: - Dell (Austin): Identify initial WHQL strategy for 2015 (mid-February… next meeting to discuss). - Samsung (Judy): Follow up with management on becoming 3rd mandatory reviewing member company for OFA (2 weeks). o UPDATE (1/30/15): Samsung has agreed to become 3rd mandatory reviewing member company. Thanks Judy! - HGST (Ken): Follow up with management on becoming 3rd mandatory reviewing member company for OFA (2 weeks). - Intel (Ray): Provide preliminary list of 1.1/1.2 features, both mandatory and optional, to start discussion on what to include in 2015 releases. - Intel (Ray): Communicate with Ken Strandberg at OFA website for logistical changes (mid-February… next meeting). Next Meeting: - Mid February 2015… to plan for next release and finalize feature list for 2015 OFA driver releases. Thanks, Ray [cid:image001.png at 01CB3870.4BB88E70] Raymond C. Robles Host Storage System Software Non-Volatile Memory Solutions Group Intel Corporation Office: 480-554-2600 Mobile: 480-399-0645 The information contained in this e-mail is considered confidential of SK hynix memory solutions Inc. and intended only for the persons addressed or copied in this e-mail. Any unauthorized use, dissemination of the information, or copying of this message is strictly prohibited. If you are not the intended recipient, please contact the sender immediately and permanently delete the original and any copies of this email. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 6439 bytes Desc: image001.png URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image002.png Type: image/png Size: 1756 bytes Desc: image002.png URL: From wuwq85 at gmail.com Sun Feb 15 14:43:26 2015 From: wuwq85 at gmail.com (Wenqian Wu) Date: Sun, 15 Feb 2015 14:43:26 -0800 Subject: [nvmewin] nvmewin Digest, Vol 38, Issue 9 In-Reply-To: References: Message-ID: Hi Ray, Thanks for your reply. I understand your saying that before calling NVMeAllocQueues, the CAP.MQES has been checked and passed in as 3rd parameter(QEntries). But if you look further in NVMeAllocQueues function, QEntires will be modified. if ((QEntries % SysPageSizeInSubEntries) != 0) QEntries = (QEntries + SysPageSizeInSubEntries) & ~(SysPageSizeInSubEntries - 1); And finally QEntires will be used to create the Queue. But QEntries may already bigger the CAP.MQES after the above round up. [image: Inline image 1] A simple example is if controller reports CAP.MQES is 32, host memory page size is 4K. The round up procedure will change QEntries to 64 (4 * 1024 / 64), and used to create the queue. Controller will returen Invalid Queue Size and failed the command. In summary, I think the alignment requirement is for host memory only. Driver can still allocate 4K(page aligned) for 32 entries for the above example, just keeping the second half unused. But when creating the queue, the entries should be 32, instead of 64. Please let me know if this makes sense. Thanks, Wenqian On Fri, Feb 13, 2015 at 12:00 PM, 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: NVMe Queue entry number (Robles, Raymond C) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Fri, 13 Feb 2015 04:42:37 +0000 > From: "Robles, Raymond C" > To: 'Wenqian Wu' , "nvmewin at lists.openfabrics.org" > > Subject: Re: [nvmewin] NVMe Queue entry number > Message-ID: > < > 49158E750348AA499168FD41D88983606B5DA191 at fmsmsx117.amr.corp.intel.com> > > Content-Type: text/plain; charset="utf-8" > > Hi Wenqian, > > Thank you for your inquiry. In the FindAdapter routine (in nvmeStd.c), the > driver checks CAP.MQES and saves the value in pAE->InitInfo.IoQEntries (if > the default or register override is smaller than CAP.MQES, it is saved > instead). Within the function NVMeAllocIoQueues, there is loop that will > iterate to create queue pairs based on the number of cpu cores. > > In this loop, a call to NVMeAllocQueues is made. Just prior to this call, > the value saved in pAE->InitInfo.IoQEntires is retrieved (stack variable > ?QEntres?) and passed in as the 3rd parameter. So, once the allocation is > taking place where you mention below, the 3rd parameter of the function has > already been checked against CAP.MQES. Also, per the NVMe spec sections 5.3 > and 5.4 ? Figures 33 and 37 (create completion queue and submission queue > PRP 1), all queue memory must be ?physically contiguous and memory page > aligned?. > > Thanks > Ray > > From: nvmewin-bounces at lists.openfabrics.org [mailto: > nvmewin-bounces at lists.openfabrics.org] On Behalf Of Wenqian Wu > Sent: Wednesday, February 11, 2015 4:58 PM > To: nvmewin at lists.openfabrics.org > Subject: [nvmewin] NVMe Queue entry number > > Hi OFA driver member, > > I have one question regarding the Queue entry size. The driver will > allocate number of entries aligned to memory page (line877, nvmeInit.c), > instead of the actual queue size controller supports (CAP.MQES). Controller > can return error if host request more entries while ignoring controller's > capability. I think as long as the base address is page aligned, there is > no reason to make the entries number aligned to page boundary. Can this be > considered a driver bug or is there any particular consideration? > > Thanks, > Wenqian > -------------- next part -------------- > An HTML attachment was scrubbed... > URL: < > http://lists.openfabrics.org/pipermail/nvmewin/attachments/20150213/edd26ac2/attachment-0001.html > > > > ------------------------------ > > _______________________________________________ > nvmewin mailing list > nvmewin at lists.openfabrics.org > http://lists.openfabrics.org/mailman/listinfo/nvmewin > > > End of nvmewin Digest, Vol 38, Issue 9 > ************************************** > -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image.png Type: image/png Size: 95560 bytes Desc: not available URL: From susobhan.dey at gmail.com Mon Feb 16 21:34:03 2015 From: susobhan.dey at gmail.com (Susobhan Dey) Date: Tue, 17 Feb 2015 11:04:03 +0530 Subject: [nvmewin] I want to install OFA driver (1.3 release) Message-ID: Hi, I want to install OFA driver (1.3 release) in windows 8 Could you please help me out in this regards. -Susobhan -------------- next part -------------- An HTML attachment was scrubbed... URL: From raymond.c.robles at intel.com Tue Feb 17 12:58:20 2015 From: raymond.c.robles at intel.com (Robles, Raymond C) Date: Tue, 17 Feb 2015 20:58:20 +0000 Subject: [nvmewin] I want to install OFA driver (1.3 release) In-Reply-To: References: Message-ID: <49158E750348AA499168FD41D88983606B5EB2CE@fmsmsx117.amr.corp.intel.com> Hi Susobhan, There is documentation on MSDN on how to install Windows drivers. The following is a MSDN article that should help. https://msdn.microsoft.com/en-us/library/windows/hardware/ff549455(v=vs.85).aspx Thanks, Ray From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Susobhan Dey Sent: Monday, February 16, 2015 10:34 PM To: nvmewin at lists.openfabrics.org Subject: [nvmewin] I want to install OFA driver (1.3 release) Hi, I want to install OFA driver (1.3 release) in windows 8 Could you please help me out in this regards. -Susobhan -------------- next part -------------- An HTML attachment was scrubbed... URL: From raymond.c.robles at intel.com Tue Feb 17 16:17:23 2015 From: raymond.c.robles at intel.com (Robles, Raymond C) Date: Wed, 18 Feb 2015 00:17:23 +0000 Subject: [nvmewin] nvmewin Digest, Vol 38, Issue 9 In-Reply-To: References: Message-ID: <49158E750348AA499168FD41D88983606B5EC770@fmsmsx117.amr.corp.intel.com> Hi, The code below simply rounds up the memory allocated to the next page boundary. The NVMe controller expects this memory (queue size) to be host page aligned. Queue size will still be within the limits of what the controller communicates. Thanks, Ray From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Wenqian Wu Sent: Sunday, February 15, 2015 3:43 PM To: nvmewin at lists.openfabrics.org Subject: Re: [nvmewin] nvmewin Digest, Vol 38, Issue 9 Hi Ray, Thanks for your reply. I understand your saying that before calling NVMeAllocQueues, the CAP.MQES has been checked and passed in as 3rd parameter(QEntries). But if you look further in NVMeAllocQueues function, QEntires will be modified. if ((QEntries % SysPageSizeInSubEntries) != 0) QEntries = (QEntries + SysPageSizeInSubEntries) & ~(SysPageSizeInSubEntries - 1); And finally QEntires will be used to create the Queue. But QEntries may already bigger the CAP.MQES after the above round up. [Inline image 1] A simple example is if controller reports CAP.MQES is 32, host memory page size is 4K. The round up procedure will change QEntries to 64 (4 * 1024 / 64), and used to create the queue. Controller will returen Invalid Queue Size and failed the command. In summary, I think the alignment requirement is for host memory only. Driver can still allocate 4K(page aligned) for 32 entries for the above example, just keeping the second half unused. But when creating the queue, the entries should be 32, instead of 64. Please let me know if this makes sense. Thanks, Wenqian On Fri, Feb 13, 2015 at 12:00 PM, > 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: NVMe Queue entry number (Robles, Raymond C) ---------------------------------------------------------------------- Message: 1 Date: Fri, 13 Feb 2015 04:42:37 +0000 From: "Robles, Raymond C" > To: 'Wenqian Wu' >, "nvmewin at lists.openfabrics.org" > Subject: Re: [nvmewin] NVMe Queue entry number Message-ID: <49158E750348AA499168FD41D88983606B5DA191 at fmsmsx117.amr.corp.intel.com> Content-Type: text/plain; charset="utf-8" Hi Wenqian, Thank you for your inquiry. In the FindAdapter routine (in nvmeStd.c), the driver checks CAP.MQES and saves the value in pAE->InitInfo.IoQEntries (if the default or register override is smaller than CAP.MQES, it is saved instead). Within the function NVMeAllocIoQueues, there is loop that will iterate to create queue pairs based on the number of cpu cores. In this loop, a call to NVMeAllocQueues is made. Just prior to this call, the value saved in pAE->InitInfo.IoQEntires is retrieved (stack variable ?QEntres?) and passed in as the 3rd parameter. So, once the allocation is taking place where you mention below, the 3rd parameter of the function has already been checked against CAP.MQES. Also, per the NVMe spec sections 5.3 and 5.4 ? Figures 33 and 37 (create completion queue and submission queue PRP 1), all queue memory must be ?physically contiguous and memory page aligned?. Thanks Ray From: nvmewin-bounces at lists.openfabrics.org [mailto:nvmewin-bounces at lists.openfabrics.org] On Behalf Of Wenqian Wu Sent: Wednesday, February 11, 2015 4:58 PM To: nvmewin at lists.openfabrics.org Subject: [nvmewin] NVMe Queue entry number Hi OFA driver member, I have one question regarding the Queue entry size. The driver will allocate number of entries aligned to memory page (line877, nvmeInit.c), instead of the actual queue size controller supports (CAP.MQES). Controller can return error if host request more entries while ignoring controller's capability. I think as long as the base address is page aligned, there is no reason to make the entries number aligned to page boundary. Can this be considered a driver bug or is there any particular consideration? Thanks, Wenqian -------------- next part -------------- An HTML attachment was scrubbed... URL: ------------------------------ _______________________________________________ nvmewin mailing list nvmewin at lists.openfabrics.org http://lists.openfabrics.org/mailman/listinfo/nvmewin End of nvmewin Digest, Vol 38, Issue 9 ************************************** -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 94158 bytes Desc: image001.png URL: