From: Logan Gunthorpe <logang@deltatee.com> To: linux-block@vger.kernel.org, linux-nvme@lists.infradead.org, Omar Sandoval <osandov@fb.com> Cc: Chaitanya Kulkarni <chaitanya.kulkarni@wdc.com>, Michael Moese <mmoese@suse.de>, Theodore Ts'o <tytso@mit.edu>, Johannes Thumshirn <jthumshirn@suse.de>, Stephen Bates <sbates@raithlin.com>, Logan Gunthorpe <logang@deltatee.com> Subject: [PATCH blktests v2 00/12] Fix nvme block test issues Date: Wed, 17 Jul 2019 11:12:47 -0600 [thread overview] Message-ID: <20190717171259.3311-1-logang@deltatee.com> (raw) Changes since v1: * Use second sed expression instead of another call to grep in _filter_discovery() for Patch 2 (per Omar) * Redirect error output to $FULL in for nvme/018 in Patch 7 (Per Johannes) * Rework _have_module_param() in Patch 11 so that it supports cases where the module is not inserted. -- This patchset cleans up a number of issues and pain points I've had with getting the nvme blktests to pass and run cleanly. The first three patches are meant to fix the Generation Counter issue that's been discussed before but hasn't been fixed in months. I primarily use a slightly fixed up patch posted by Michael Moese[1] but add another patch to properly test the Generation Counter that would no longer be tested otherwise. I've also taken it a step further to filter out more of the discovery information so that we are less fragile against churn and less dependant on the version of nvme-cli in use. This should also fix the issue discussed in [2]. Patches 4 through 7 fix a number of smaller issues I've found with individual tests. Patches 8 through 10 implement a system to ensure the nvme tests clean themselves up properly especially when ctrl-c is used to interrupt a test (working with the tests before this was a huge pain seeing, when you ctrl-c, you have to either manually clean up the nvmet configuration or reboot to get the system in a state where it's sane to run the tests again). Patches 11 and 12 make some minor changes that allow running the tests with the nvme modules built into the kernel. With these patches, plus a couple I've sent to the nvme list for the kernel, I can consistently pass the entire nvme suite with the exception of the lockdep false-positive with nvme/012 that still seems to be in a bit of contention[3]. [1] https://github.com/osandov/blktests/pull/34 [2] https://lore.kernel.org/linux-block/20190505150611.15776-4-minwoo.im.dev@gmail.com/ [3] https://lore.kernel.org/lkml/20190214230058.196511-22-bvanassche@acm.org/ -- Logan Gunthorpe (11): nvme: More agressively filter the discovery output nvme: Add new test to verify the generation counter nvme/003,004: Add missing calls to nvme disconnect nvme/005: Don't rely on modprobing to set the multipath paramater nvme/015: Ensure the namespace is flushed not the char device nvme/018: Ignore error message generated by nvme read check: Add the ability to call a cleanup function after a test ends nvme: Cleanup modprobe lines into helper functions nvme: Ensure all ports and subsystems are removed on cleanup common: Use sysfs instead of modinfo for _have_module_param() nvme: Ignore errors when removing modules Michael Moese (1): Add filter function for nvme discover check | 9 + common/rc | 24 + tests/nvme/002 | 10 +- tests/nvme/002.out | 6003 +------------------------------------------- tests/nvme/003 | 6 +- tests/nvme/003.out | 1 + tests/nvme/004 | 6 +- tests/nvme/004.out | 1 + tests/nvme/005 | 16 +- tests/nvme/006 | 6 +- tests/nvme/007 | 6 +- tests/nvme/008 | 6 +- tests/nvme/009 | 5 +- tests/nvme/010 | 6 +- tests/nvme/011 | 6 +- tests/nvme/012 | 6 +- tests/nvme/013 | 6 +- tests/nvme/014 | 6 +- tests/nvme/015 | 5 +- tests/nvme/016 | 6 +- tests/nvme/016.out | 9 +- tests/nvme/017 | 8 +- tests/nvme/017.out | 9 +- tests/nvme/018 | 8 +- tests/nvme/019 | 6 +- tests/nvme/020 | 5 +- tests/nvme/021 | 6 +- tests/nvme/022 | 6 +- tests/nvme/023 | 6 +- tests/nvme/024 | 6 +- tests/nvme/025 | 6 +- tests/nvme/026 | 6 +- tests/nvme/027 | 6 +- tests/nvme/028 | 6 +- tests/nvme/029 | 6 +- tests/nvme/030 | 72 + tests/nvme/030.out | 2 + tests/nvme/rc | 65 + 38 files changed, 216 insertions(+), 6162 deletions(-) create mode 100755 tests/nvme/030 create mode 100644 tests/nvme/030.out -- 2.17.1
WARNING: multiple messages have this Message-ID (diff)
From: logang@deltatee.com (Logan Gunthorpe) Subject: [PATCH blktests v2 00/12] Fix nvme block test issues Date: Wed, 17 Jul 2019 11:12:47 -0600 [thread overview] Message-ID: <20190717171259.3311-1-logang@deltatee.com> (raw) Changes since v1: * Use second sed expression instead of another call to grep in _filter_discovery() for Patch 2 (per Omar) * Redirect error output to $FULL in for nvme/018 in Patch 7 (Per Johannes) * Rework _have_module_param() in Patch 11 so that it supports cases where the module is not inserted. -- This patchset cleans up a number of issues and pain points I've had with getting the nvme blktests to pass and run cleanly. The first three patches are meant to fix the Generation Counter issue that's been discussed before but hasn't been fixed in months. I primarily use a slightly fixed up patch posted by Michael Moese[1] but add another patch to properly test the Generation Counter that would no longer be tested otherwise. I've also taken it a step further to filter out more of the discovery information so that we are less fragile against churn and less dependant on the version of nvme-cli in use. This should also fix the issue discussed in [2]. Patches 4 through 7 fix a number of smaller issues I've found with individual tests. Patches 8 through 10 implement a system to ensure the nvme tests clean themselves up properly especially when ctrl-c is used to interrupt a test (working with the tests before this was a huge pain seeing, when you ctrl-c, you have to either manually clean up the nvmet configuration or reboot to get the system in a state where it's sane to run the tests again). Patches 11 and 12 make some minor changes that allow running the tests with the nvme modules built into the kernel. With these patches, plus a couple I've sent to the nvme list for the kernel, I can consistently pass the entire nvme suite with the exception of the lockdep false-positive with nvme/012 that still seems to be in a bit of contention[3]. [1] https://github.com/osandov/blktests/pull/34 [2] https://lore.kernel.org/linux-block/20190505150611.15776-4-minwoo.im.dev at gmail.com/ [3] https://lore.kernel.org/lkml/20190214230058.196511-22-bvanassche at acm.org/ -- Logan Gunthorpe (11): nvme: More agressively filter the discovery output nvme: Add new test to verify the generation counter nvme/003,004: Add missing calls to nvme disconnect nvme/005: Don't rely on modprobing to set the multipath paramater nvme/015: Ensure the namespace is flushed not the char device nvme/018: Ignore error message generated by nvme read check: Add the ability to call a cleanup function after a test ends nvme: Cleanup modprobe lines into helper functions nvme: Ensure all ports and subsystems are removed on cleanup common: Use sysfs instead of modinfo for _have_module_param() nvme: Ignore errors when removing modules Michael Moese (1): Add filter function for nvme discover check | 9 + common/rc | 24 + tests/nvme/002 | 10 +- tests/nvme/002.out | 6003 +------------------------------------------- tests/nvme/003 | 6 +- tests/nvme/003.out | 1 + tests/nvme/004 | 6 +- tests/nvme/004.out | 1 + tests/nvme/005 | 16 +- tests/nvme/006 | 6 +- tests/nvme/007 | 6 +- tests/nvme/008 | 6 +- tests/nvme/009 | 5 +- tests/nvme/010 | 6 +- tests/nvme/011 | 6 +- tests/nvme/012 | 6 +- tests/nvme/013 | 6 +- tests/nvme/014 | 6 +- tests/nvme/015 | 5 +- tests/nvme/016 | 6 +- tests/nvme/016.out | 9 +- tests/nvme/017 | 8 +- tests/nvme/017.out | 9 +- tests/nvme/018 | 8 +- tests/nvme/019 | 6 +- tests/nvme/020 | 5 +- tests/nvme/021 | 6 +- tests/nvme/022 | 6 +- tests/nvme/023 | 6 +- tests/nvme/024 | 6 +- tests/nvme/025 | 6 +- tests/nvme/026 | 6 +- tests/nvme/027 | 6 +- tests/nvme/028 | 6 +- tests/nvme/029 | 6 +- tests/nvme/030 | 72 + tests/nvme/030.out | 2 + tests/nvme/rc | 65 + 38 files changed, 216 insertions(+), 6162 deletions(-) create mode 100755 tests/nvme/030 create mode 100644 tests/nvme/030.out -- 2.17.1
next reply other threads:[~2019-07-17 17:13 UTC|newest] Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top 2019-07-17 17:12 Logan Gunthorpe [this message] 2019-07-17 17:12 ` [PATCH blktests v2 00/12] Fix nvme block test issues Logan Gunthorpe 2019-07-17 17:12 ` [PATCH blktests v2 01/12] Add filter function for nvme discover Logan Gunthorpe 2019-07-17 17:12 ` Logan Gunthorpe 2019-07-18 6:58 ` Johannes Thumshirn 2019-07-18 6:58 ` Johannes Thumshirn 2019-07-17 17:12 ` [PATCH blktests v2 02/12] nvme: More agressively filter the discovery output Logan Gunthorpe 2019-07-18 6:59 ` Johannes Thumshirn 2019-07-18 6:59 ` Johannes Thumshirn 2019-07-17 17:12 ` [PATCH blktests v2 03/12] nvme: Add new test to verify the generation counter Logan Gunthorpe 2019-07-17 17:12 ` Logan Gunthorpe 2019-07-18 7:03 ` Johannes Thumshirn 2019-07-18 7:03 ` Johannes Thumshirn 2019-07-18 15:55 ` Logan Gunthorpe 2019-07-18 15:55 ` Logan Gunthorpe 2019-07-17 17:12 ` [PATCH blktests v2 04/12] nvme/003,004: Add missing calls to nvme disconnect Logan Gunthorpe 2019-07-17 17:12 ` [PATCH blktests v2 04/12] nvme/003, 004: " Logan Gunthorpe 2019-07-18 7:03 ` [PATCH blktests v2 04/12] nvme/003,004: " Johannes Thumshirn 2019-07-18 7:03 ` Johannes Thumshirn 2019-07-17 17:12 ` [PATCH blktests v2 05/12] nvme/005: Don't rely on modprobing to set the multipath paramater Logan Gunthorpe 2019-07-17 17:12 ` Logan Gunthorpe 2019-07-18 7:04 ` Johannes Thumshirn 2019-07-18 7:04 ` Johannes Thumshirn 2019-07-17 17:12 ` [PATCH blktests v2 06/12] nvme/015: Ensure the namespace is flushed not the char device Logan Gunthorpe 2019-07-17 17:12 ` Logan Gunthorpe 2019-07-18 7:04 ` Johannes Thumshirn 2019-07-18 7:04 ` Johannes Thumshirn 2019-07-17 17:12 ` [PATCH blktests v2 07/12] nvme/018: Ignore error message generated by nvme read Logan Gunthorpe 2019-07-17 17:12 ` Logan Gunthorpe 2019-07-18 7:05 ` Johannes Thumshirn 2019-07-18 7:05 ` Johannes Thumshirn 2019-07-17 17:12 ` [PATCH blktests v2 08/12] check: Add the ability to call a cleanup function after a test ends Logan Gunthorpe 2019-07-17 17:12 ` Logan Gunthorpe 2019-07-18 7:07 ` Johannes Thumshirn 2019-07-18 7:07 ` Johannes Thumshirn 2019-07-17 17:12 ` [PATCH blktests v2 09/12] nvme: Cleanup modprobe lines into helper functions Logan Gunthorpe 2019-07-17 17:12 ` Logan Gunthorpe 2019-07-18 7:07 ` Johannes Thumshirn 2019-07-18 7:07 ` Johannes Thumshirn 2019-07-17 17:12 ` [PATCH blktests v2 10/12] nvme: Ensure all ports and subsystems are removed on cleanup Logan Gunthorpe 2019-07-17 17:12 ` Logan Gunthorpe 2019-07-18 7:08 ` Johannes Thumshirn 2019-07-18 7:08 ` Johannes Thumshirn 2019-07-17 17:12 ` [PATCH blktests v2 11/12] common: Use sysfs instead of modinfo for _have_module_param() Logan Gunthorpe 2019-07-17 17:12 ` Logan Gunthorpe 2019-07-18 7:08 ` Johannes Thumshirn 2019-07-18 7:08 ` Johannes Thumshirn 2019-07-17 17:12 ` [PATCH blktests v2 12/12] nvme: Ignore errors when removing modules Logan Gunthorpe 2019-07-17 17:12 ` Logan Gunthorpe 2019-07-18 7:09 ` Johannes Thumshirn 2019-07-18 7:09 ` Johannes Thumshirn 2019-07-29 17:31 ` [PATCH blktests v2 00/12] Fix nvme block test issues Omar Sandoval 2019-07-29 17:31 ` Omar Sandoval
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20190717171259.3311-1-logang@deltatee.com \ --to=logang@deltatee.com \ --cc=chaitanya.kulkarni@wdc.com \ --cc=jthumshirn@suse.de \ --cc=linux-block@vger.kernel.org \ --cc=linux-nvme@lists.infradead.org \ --cc=mmoese@suse.de \ --cc=osandov@fb.com \ --cc=sbates@raithlin.com \ --cc=tytso@mit.edu \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.