[00:51:03] *** Quits: tmonjalo (~quassel@184.203.134.77.rev.sfr.net) (Read error: Connection reset by peer) [00:51:16] *** Joins: tmonjalo (~quassel@184.203.134.77.rev.sfr.net) [00:57:20] *** Quits: ziyeyang_ (ziyeyang@nat/intel/x-eqshrztdriodukou) (Quit: Leaving) [01:33:38] *** Joins: gila (~gila@5ED4D9C8.cm-7-5d.dynamic.ziggo.nl) [01:43:44] *** Quits: baruch (~baruch@141.226.177.138) (Read error: Connection reset by peer) [02:38:45] *** Quits: Shuhei (caf6fc61@gateway/web/freenode/ip.202.246.252.97) (Ping timeout: 260 seconds) [02:45:40] *** Joins: baruch (~baruch@bzq-82-81-85-138.red.bezeqint.net) [03:05:45] *** Joins: baruch_ (~baruch@bzq-82-81-85-138.red.bezeqint.net) [03:07:01] *** Quits: baruch (~baruch@bzq-82-81-85-138.red.bezeqint.net) (Ping timeout: 252 seconds) [05:59:02] *** Quits: baruch_ (~baruch@bzq-82-81-85-138.red.bezeqint.net) (Ping timeout: 268 seconds) [06:05:50] *** Quits: gila (~gila@5ED4D9C8.cm-7-5d.dynamic.ziggo.nl) (Ping timeout: 255 seconds) [06:08:41] *** Joins: gila (~gila@106.51.36.232) [06:16:20] *** Quits: edwinrodriguez (d8f01e17@gateway/web/freenode/ip.216.240.30.23) (Ping timeout: 260 seconds) [06:45:35] *** Quits: gila (~gila@106.51.36.232) (Ping timeout: 258 seconds) [06:51:49] *** Joins: gila (~gila@5ED4D9C8.cm-7-5d.dynamic.ziggo.nl) [07:52:03] bwalker: could we enable virtio initiator tests for a sec? [07:52:49] i'd like to test one patch: https://review.gerrithub.io/c/388424 [09:28:41] drv: push darsto's 17.10.x patches for dpdk 17.11 compatbility? [09:29:02] sounds good [09:30:59] mszwed suggested 385950 and 388402 for 17.10.x - can I just hit the cherry pick button and choose the 17.10.x branch and gerrit will do the rest? [09:37:25] I think that should work [09:37:41] hm, one of my review builds failed with an NBD crash I haven't seen before: https://ci.spdk.io/spdk/builds/review/3e72fd9d46f39aec3ebeaa14dc6d73af89987f3b.1511280686/fedora-02/build.log [09:40:19] i'll need to take another look at the nbd shutdown code [09:42:43] I enabled v17.10.x release builds (forgot to do that earlier), so there will be some queued up [10:04:02] drv: i've reproduced that nbd crash locally - i'll have a fix here shortly [10:06:13] cool [10:08:32] this blob_bdev message about not supporting unmap is pretty misleading - we print it before we even know if the device is a blobstore [10:09:07] so it gets dumped out every time the lvol examine function is called [10:23:33] one more patch that should fix virtio initiator issue: https://review.gerrithub.io/c/388505/ [10:24:37] it might silence the crashes, but there will be a data race somewhere in spdk [10:25:37] but still there is* [13:13:54] drv: nbd fix: https://review.gerrithub.io/#/c/388520/ [13:13:58] two patches actually [13:14:08] it's all related to how we were using fork() in the nbd library [13:14:34] oh, nice, so it works with just a new thread? [13:14:54] yeah - at least it works on my system with the few tests I ran [13:15:39] i was seeing weird double event calls in the log and couldn't figure out why [13:16:00] yeah, this looks much better [13:16:06] and setting breakpoints in spdk_event_call() with a debug print - was getting the debug print but no breakpoint [13:16:23] major *facepalm* when I remembered about the fork [14:00:15] *** Joins: baruch (~baruch@141.226.177.138) [14:06:21] drv, bwalker: can we get this series in, and then push darsto's big virtio series? [14:06:23] https://review.gerrithub.io/#/c/388307/ [14:06:40] patches #3 and #4 in that series still need bwalker's +2 [14:10:11] we still need to turn on the bdev/virtio tests and re-run at least the end of darsto's series before we merge it [14:17:54] drv: oh please do :) [14:21:12] jimharris: I say we go ahead and merge that series, so we can turn on the virtio tests [14:22:39] which series? [14:22:54] your series of 5? [14:23:11] yes [14:23:21] bwalker reviewed earlier revisions of those, so I don't think he'd object [14:39:10] pushed [14:42:29] ok, once the final patch is in, I'll re-enable virtio testing [14:43:14] how do we want to handle darsto_'s series? probably don't want to re-run all of the patches if we can avoid it [14:44:00] I agree - I'm sure darsto_'s patches are all perfect :) [14:44:26] we could just merge the series and deal with the fallout if there is any [14:44:33] sounds good to me [15:18:06] is virtio testing re-enabled? [15:18:54] waiting for this test to finish and then I'll turn it on [15:19:05] you can push the patches [16:02:59] *** Joins: Shuhei (caf6fc61@gateway/web/freenode/ip.202.246.252.97) [16:37:45] *** Joins: ziyeyang_ (~ziyeyang@192.55.54.44) [16:59:57] *** Joins: ziyeyang__ (~ziyeyang@134.134.139.83) [16:59:57] *** Quits: ziyeyang_ (~ziyeyang@192.55.54.44) (Remote host closed the connection) [17:58:42] Hi jim and Daniel, is this issue related with bdev fio plugin for vhost test [17:58:43] http://spdk.intel.com/public/spdk/builds/review/653c9d80bba77038773e7850f79eac227ecb8d54.1511311639/fedora-07/ [18:00:06] hi Ziye, it should be resolved if you rebase on latest master [18:02:26] OK, got it [18:02:30] Thanks, daniel [18:02:54] *** Quits: ziyeyang__ (~ziyeyang@134.134.139.83) (Remote host closed the connection) [18:46:45] Hi, after update, "modified: dpdk (new commits)" was displayed and it could not be removed even if "git checkout -- dpdk". Hence according to Google, I tried "git submodule update" and this looked working. Is this correct method? [18:49:46] after update is after git pull from GerritHub. [18:56:04] "git submodule update" did not fix the issue because make failed. I will find how to fix this wrong measure. [18:58:52] Shuhei: 'git submodule update' is the right thing to do - can you try 'make clean' and see if it fixes the build for you? [19:00:41] Daniel, thank you, I'll try make clean -> make. [19:12:44] hi Shuhei - did the make clean work? [20:13:51] *** Joins: ziyeyang_ (~ziyeyang@134.134.139.83) [20:41:16] Hi Jim, daniel: I rebased this patch due to the master changed without any code change, could you review it again and merge it? [20:41:17] Thanks [23:47:23] @shuhei, Updated the patch according to Jim's comments.