[00:51:07] *** Joins: tkulasek (~tkulasek@134.134.139.76) [01:12:14] *** Joins: pwodkowx (pwodkowx@nat/intel/x-leeciyvtjvdfyddt) [02:01:29] *** Joins: johnmeneghini (~johnmeneg@pool-100-0-53-181.bstnma.fios.verizon.net) [03:38:57] *** Quits: dlw (~Thunderbi@114.255.44.143) (Ping timeout: 240 seconds) [04:10:18] *** Joins: lyan (~lyan@2605:a000:160e:2124:4a4d:7eff:fef2:eea3) [04:10:41] *** lyan is now known as Guest87841 [04:46:56] *** Quits: johnmeneghini (~johnmeneg@pool-100-0-53-181.bstnma.fios.verizon.net) (Quit: Leaving.) [04:55:46] *** Joins: johnmeneghini (~johnmeneg@pool-100-0-53-181.bstnma.fios.verizon.net) [05:25:06] *** Quits: johnmeneghini (~johnmeneg@pool-100-0-53-181.bstnma.fios.verizon.net) (Quit: Leaving.) [05:50:11] *** Joins: philipp-sk (~Philipp@ktnron0916w-lp130-02-76-66-162-159.dsl.bell.ca) [07:51:05] *** Joins: dlw (~Thunderbi@123.123.106.20) [08:02:14] *** Quits: dlw (~Thunderbi@123.123.106.20) (Ping timeout: 260 seconds) [08:08:34] *** Joins: johnmeneghini (~johnmeneg@216.240.30.5) [08:49:59] jimharris, please take another quick look at https://review.gerrithub.io/c/spdk/spdk/+/418549 when you get a chance [08:58:43] peluse: what do you want me to look at on that patch? [08:59:33] I added a note that it turns out I do need it. Otherwise ASAN complains about freeing something I never allocated because the alloc is mocked also. So my prev comment (and drvs) aren't valid after testing w/o it [08:59:58] ok - i didn't see anything about that in gerrithub [09:00:06] crap, I thought I put that in there [09:00:49] could you take a look at ben's patch that simplifies the mocking API? [09:00:51] you're right its not, maybe I just made a note in IRC, ugh. Or possibly just in my own head :) [09:01:01] yup [09:01:15] I'm looking at the latest zcopy stuff now, will do so after that [09:11:44] Hi, is spdk_dma_free(NULL) safe, like standard free(NULL) in C? [09:12:53] tkulasek: yes [09:13:11] Ok, thanks [09:46:35] *** Joins: Ben__ (6cab87bc@gateway/web/freenode/ip.108.171.135.188) [10:02:16] Seth/Paul: Is there some reason why my patches are not automatically running CITs? [10:02:17] Did the cutover to the new Jenkins CIT stuff somehow drop my name from the whitelist? [10:05:48] johnmeneghini - that's certainly possible - i'll send an e-mail ASAP [10:06:49] johnmeneghini, you've been "blue listed" LOL [10:16:42] johnmeneghini - sethhowe_ and klateck (Karol) are on it [10:31:12] johnmeneghini, is Madhu on IRC? [10:31:24] or Sriram? [10:31:44] that would be great if they were on IRC - we'd knock these questions out in no time [10:32:14] yeah, I have a few small ones that I hesitate to put in GH as they may just confuse things even more, LOL [10:32:20] jimharris: thats Jim. No Madhu and Sriram are not on IRC. We can talk later this morning about this. [10:36:51] jonmeneghini, jimharris: sorry about the mixup there. Karol and I have been maintaining our config files semi-independently, so there are some changes to the Jenkins list I will need to make. I will need just a little bit. [10:38:00] thanks sethhowe_ [10:38:01] sethhowe: no problem Seth… [10:38:11] Thanks [10:38:54] *** Quits: Ben__ (6cab87bc@gateway/web/freenode/ip.108.171.135.188) (Ping timeout: 252 seconds) [10:57:59] johnmeneghini: I have added you to the whitelist in Jenkins now. All patches you push going forward should be picked up, but let me know if you hit any issues. [11:06:05] *** Joins: gila (~gila@5ED74129.cm-7-8b.dynamic.ziggo.nl) [11:21:10] *** Quits: gila (~gila@5ED74129.cm-7-8b.dynamic.ziggo.nl) (Ping timeout: 268 seconds) [11:32:23] *** Joins: gila (~gila@5ED74129.cm-7-8b.dynamic.ziggo.nl) [11:33:20] is a lvol on an AIO backend supposed to work? Or is that exclusively for NVMe based bdevs? [11:38:05] *** Quits: johnmeneghini (~johnmeneg@216.240.30.5) (Quit: Leaving.) [11:46:12] I get all kinds of IO errors (ndb as well as iSCSI) when using a lvol ontop of vbdev, i recall something about atomic updates that blobstore was using underneath [11:52:51] *ontop of an AIO bdev [12:01:33] *** Quits: tkulasek (~tkulasek@134.134.139.76) (Ping timeout: 264 seconds) [12:06:05] *** Joins: johnmeneghini (~johnmeneg@216.240.30.5) [12:06:19] there are certainly concerns about atomicity that need to be accounted for, but unless you're really hammering the same blob from multiple threads at the same time those won't pop up [12:06:37] so I think it should be working gila [12:07:01] Hm.. then something else is wrong with my setup.. [12:07:28] I get (for iscsi) errors like these: sd 34:0:0:0: [sdc] tag#0 FAILED Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE [12:07:47] and hdparm also fails [12:08:02] for nbd - things are no better: blk_update_request: I/O error, dev nbd0, sector 0 [12:08:14] what are you using to send I/O? [12:08:33] trying to lay down a FS on top of it (ext4) [12:09:07] is the logical volume exposed over iscsi, and then you're using the kernel initiator? [12:09:14] yes sir. [12:09:33] does everything work fine without the logical volume in the picture? If you just expose the AIO disk over iSCSI [12:09:58] It did this afternoon, but let me double check that to make sure [12:12:45] bwalker same errors... strange [12:49:32] bwalker it seems that reads are failing for reasons beyond me [14:08:52] jimharris are you online? [14:12:04] he's not at the moment [14:13:04] I don't think he'll be away that long [14:21:27] ok, let's talk on monday, no hurry. thanks, have a nice weekend :P [14:43:40] *** Quits: johnmeneghini (~johnmeneg@216.240.30.5) (Quit: Leaving.) [15:28:40] pwodkowx - sorry - I'm back [15:28:50] i'm actually on vacation next week [16:51:01] *** Joins: johnmeneghini (~johnmeneg@pool-100-0-53-181.bstnma.fios.verizon.net) [17:39:20] *** Quits: johnmeneghini (~johnmeneg@pool-100-0-53-181.bstnma.fios.verizon.net) (Quit: Leaving.) [17:46:42] *** Joins: johnmeneghini (~johnmeneg@pool-100-0-53-181.bstnma.fios.verizon.net) [18:12:09] *** Quits: Guest87841 (~lyan@2605:a000:160e:2124:4a4d:7eff:fef2:eea3) (Quit: Leaving) [19:26:31] *** Quits: johnmeneghini (~johnmeneg@pool-100-0-53-181.bstnma.fios.verizon.net) (Quit: Leaving.)