[00:06:32] I think i have installed my os on nvme drive https://pastebin.com/xi1dmZsS [00:06:50] so i guess unbinding it wont be possible [00:15:29] *** Quits: Shuhei (caf6fc61@gateway/web/freenode/ip.202.246.252.97) (Ping timeout: 260 seconds) [00:56:21] *** Joins: tkulasek (~tkulasek@192.55.54.42) [04:47:19] *** Joins: tomzawadzki (~tomzawadz@192.55.54.39) [04:48:27] *** Quits: dlw (~Thunderbi@114.255.44.143) (Ping timeout: 240 seconds) [06:03:22] *** Quits: tomzawadzki (~tomzawadz@192.55.54.39) (Ping timeout: 264 seconds) [06:30:15] *** Joins: darsto_ (~darsto@89-68-115-185.dynamic.chello.pl) [06:31:15] *** Quits: darsto (~darsto@89-68-115-185.dynamic.chello.pl) (Ping timeout: 256 seconds) [06:46:28] *** Joins: dlw (~Thunderbi@123.122.66.130) [06:52:13] *** Joins: tomzawadzki (~tomzawadz@192.55.54.39) [07:32:00] *** Quits: dlw (~Thunderbi@123.122.66.130) (Ping timeout: 260 seconds) [08:37:03] *** Quits: tomzawadzki (~tomzawadz@192.55.54.39) (Ping timeout: 256 seconds) [09:02:14] *** darsto_ is now known as darsto [09:04:50] jimharris: replied on that ENOMEM unit test patch, https://review.gerrithub.io/c/spdk/spdk/+/414976 [09:15:41] thanks - good catch - I just replied [09:16:16] bwalker, drv: you should take a look at darsto's patch too - it's somewhat related to the ENOMEM spdk_bdev_io pool exhaustion stuff we were discussing yesterday [09:16:29] ok, I'll take a look shortly [09:17:15] there are paths in blobstore that do calloc's in the I/O path that could result in ENOMEM getting returned even if its the first I/O sent on the channel [09:24:25] yeah, it would be nice to get rid of those calloc's somehow, but I'm not sure what the alternative is [09:24:40] since we presumably need some context to maintain state across asynchronous operations [09:25:10] jimharris, bwalker: reminder to please review helloway's rpc patch: https://review.gerrithub.io/#/c/spdk/spdk/+/414296/ [09:29:04] set_nvmf_target_options - maybe this should be done in a separate patch, but can we resolve the discrepency between max_qpairs_per_session and max_qpairs_per_ctrlr? [09:30:00] yeah it should all be using the word "controller" [09:30:00] yes, it should be ctrlr, not sure where session came from [09:30:19] rpc.py still takes max_qpairs_per_session [09:30:47] https://review.gerrithub.io/#/c/spdk/spdk/+/414957/ [09:31:01] we can merge that first if you prefer [09:31:43] i'm fine merging helloway's patch first [09:32:59] for nvmf_subsystem_add/remove_listener - is adrfam supposed to be optional? [09:33:12] it's not specified with =None, but the code is checking if adrfam is specified [09:33:26] we can fix that up later [09:34:35] it should really be required, but I think we have some autodetection (default to IPv4) for backwards compatibility [09:48:10] *** Joins: travis-ci (~travis-ci@ec2-23-22-24-17.compute-1.amazonaws.com) [09:48:11] (spdk/master) nvmf/rpc: add max_namespaces to get_nvmf_subsystems (Daniel Verkamp) [09:48:11] Diff URL: https://github.com/spdk/spdk/compare/34de1542eb41...6546fd10f826 [09:48:11] *** Parts: travis-ci (~travis-ci@ec2-23-22-24-17.compute-1.amazonaws.com) () [10:09:58] *** Joins: travis-ci (~travis-ci@ec2-23-22-24-17.compute-1.amazonaws.com) [10:09:59] (spdk/master) scripts/rpc.py: pass named args to nvmf.py (heluwei) [10:09:59] Diff URL: https://github.com/spdk/spdk/compare/6546fd10f826...aceeb8cda3bc [10:09:59] *** Parts: travis-ci (~travis-ci@ec2-23-22-24-17.compute-1.amazonaws.com) () [10:53:07] *** Quits: tkulasek (~tkulasek@192.55.54.42) (Ping timeout: 245 seconds) [11:41:32] ugh, updated to fedora 28 [11:41:43] now soft roce doesn't work (crashes in rxe driver) [13:56:17] *** Quits: pohly (~pohly@p54BD5098.dip0.t-ipconnect.de) (Quit: Leaving.) [14:35:22] jimharris: any opinions on https://review.gerrithub.io/#/c/spdk/spdk.github.io/+/413315/ before I push it? [14:38:49] looking... [14:40:54] couple of small edits - otherwise looks good [14:42:03] first rev of spdk_bdev_queue_io_wait() posted: https://review.gerrithub.io/#/c/spdk/spdk/+/415074/ [14:46:13] jimharris: ok, I fixed up your edits on the blog post [14:46:27] the text is all from Anu, so I claim no blame and/or credit for any mistakes :) [14:47:40] yeah - I was pretty sure you didn't write it :-) [14:48:16] it exceeds my exclamation point quota [14:49:07] ha - yeah, my wife and i always laugh at text messages and e-mails we get from my mom - she sent one recently where every sentence ended with an exclamation point [14:55:31] *** Joins: travis-ci (~travis-ci@ec2-54-167-11-65.compute-1.amazonaws.com) [14:55:32] (spdk/master) nvmf: don't change NN while ctrlrs exist (Daniel Verkamp) [14:55:32] Diff URL: https://github.com/spdk/spdk/compare/aceeb8cda3bc...eb387189c2eb [14:55:32] *** Parts: travis-ci (~travis-ci@ec2-54-167-11-65.compute-1.amazonaws.com) () [14:59:33] jimharris: Looks good - I think this is the right solution [15:00:34] only made one comment [16:20:43] ugh, the intel-ipsec-mb library has a bunch of scan-build warnings for dead stores [16:20:52] don't really want to have to maintain our own fork of it just to fix that [16:24:07] *** Joins: Shuhei (caf6fc61@gateway/web/freenode/ip.202.246.252.97) [16:28:45] fun [16:28:52] Doxygen subpage changes: https://review.gerrithub.io/#/c/spdk/spdk/+/415085/ [16:29:00] really cleans up the left nav pane [16:35:36] *** Joins: travis-ci (~travis-ci@ec2-54-167-11-65.compute-1.amazonaws.com) [16:35:37] (spdk/master) nvmf: add JSON config dump (Pawel Wodkowski) [16:35:37] Diff URL: https://github.com/spdk/spdk/compare/42b6dd9399c6...7efc7b830266 [16:35:37] *** Parts: travis-ci (~travis-ci@ec2-54-167-11-65.compute-1.amazonaws.com) () [16:38:43] jimharris: I think the order of the files in INPUT in Doxyfile is also the order they show up in the nav pane [16:38:53] so we might want to relax our alphabetical order requirements just this once :) [16:39:13] (that's why index.md is first) [16:39:50] but otherwise looks much cleaner [16:39:56] ah - ok, I missed that detail, but i did notice that they weren't exactly alphabetized (i.e. blob.md is out of place) [16:39:59] i'll fix that up [16:41:08] it would be great if it could match the order in index.md [16:44:32] yeah - that's exactly what i'll do [17:07:15] hey do we have a per IO size limit anywhere up/down the stack? [17:26:48] the bdev layer should theoretically work with any IO size (modules are responsible for splitting if they have limits) [17:29:58] OK, thanks. Was hoping for a limit :) [18:30:52] *** Joins: dlw (~Thunderbi@114.255.44.143) [19:35:24] *** Joins: guerby_ (~guerby@ip165.tetaneutral.net) [19:36:43] *** Quits: guerby (~guerby@april/board/guerby) (Remote host closed the connection) [19:57:17] *** Joins: travis-ci (~travis-ci@ec2-54-226-66-126.compute-1.amazonaws.com) [19:57:18] (spdk/master) iscsi initiator: remove the deleted variable. (Ziye Yang) [19:57:18] Diff URL: https://github.com/spdk/spdk/compare/7efc7b830266...4f47c066c993 [19:57:19] *** Parts: travis-ci (~travis-ci@ec2-54-226-66-126.compute-1.amazonaws.com) () [20:38:09] *** Quits: Shuhei (caf6fc61@gateway/web/freenode/ip.202.246.252.97) (Ping timeout: 260 seconds) [20:38:43] *** Joins: travis-ci (~travis-ci@ec2-54-160-255-28.compute-1.amazonaws.com) [20:38:45] (spdk/master) iscsi: Refactor a few of loop and branch and remove unnecessary initialization (Shuhei Matsumoto) [20:38:45] Diff URL: https://github.com/spdk/spdk/compare/4f47c066c993...9e21207a7c23 [20:38:45] *** Parts: travis-ci (~travis-ci@ec2-54-160-255-28.compute-1.amazonaws.com) () [20:55:41] *** Joins: Shuhei (caf6fc61@gateway/web/freenode/ip.202.246.252.97) [20:57:21] *** Joins: travis-ci (~travis-ci@ec2-54-160-255-28.compute-1.amazonaws.com) [20:57:22] (spdk/master) copy/ioat: Add scan_ioat_copy_engine RPC (Shuhei Matsumoto) [20:57:22] Diff URL: https://github.com/spdk/spdk/compare/6f46e272c3c0...ce2e68289b13 [20:57:22] *** Parts: travis-ci (~travis-ci@ec2-54-160-255-28.compute-1.amazonaws.com) () [20:58:39] *** Joins: travis-ci (~travis-ci@ec2-54-226-66-126.compute-1.amazonaws.com) [20:58:40] (spdk/master) nvme/rdma: add timeout support to RDMA transport (Daniel Verkamp) [20:58:40] Diff URL: https://github.com/spdk/spdk/compare/ce2e68289b13...6b504fda5df6 [20:58:40] *** Parts: travis-ci (~travis-ci@ec2-54-226-66-126.compute-1.amazonaws.com) () [21:16:39] *** Quits: Shuhei (caf6fc61@gateway/web/freenode/ip.202.246.252.97) (Ping timeout: 260 seconds) [21:45:28] *** Joins: travis-ci (~travis-ci@ec2-54-226-66-126.compute-1.amazonaws.com) [21:45:29] (spdk/master) include/blobfs.h: add comments for callback functions (Yanbo Zhou) [21:45:29] Diff URL: https://github.com/spdk/spdk/compare/6b504fda5df6...d7541156b34e [21:45:29] *** Parts: travis-ci (~travis-ci@ec2-54-226-66-126.compute-1.amazonaws.com) () [22:10:33] *** guerby_ is now known as guerby [22:10:38] *** Quits: guerby (~guerby@ip165.tetaneutral.net) (Changing host) [22:10:38] *** Joins: guerby (~guerby@april/board/guerby) [22:58:03] *** Joins: pohly (~pohly@p54BD5098.dip0.t-ipconnect.de)