19:24:51 oh how i wish the kernel's bzero() actually set up a stack frame 19:25:44 it's a leaf right? what do you need the frame for? 19:26:23 (I'm confused about this one, basically) 19:26:49 gwr: do you still need reviews from me for my bugs? it's _so hot_ here rn my brain doesn't work 19:29:41 well more the lack of a frame pointer 19:29:52 so the calling function doesn't show up in the stack 19:30:21 while you can usually infer it from the offset (the instruction offset will be after the function that was called) 19:30:36 but then you have to crawl the stack to find the arguments passed to that invisible function 19:30:44 which is tedious 19:31:17 and for various reasons, I'm rather brain fried and exhausted today, so it's not a mental exercise I really want to attempt today 19:35:23 are those reasons that it's a million fucking degrees there too? 19:35:31 because I sympethesize from the surface of the sun 19:35:48 oh that's pretty much summer here.. 19:37:29 no, I was up rather late fighting ESX's horrid NFS4 support 19:37:53 and then was woken up early this morning because people decided to set off fireworks (and down here you can get the 'big' fireworks) 19:37:59 a few days early 19:38:21 so I'm working off like 4-5 hours of sleep, so the focus needed isn't there today :) 19:39:32 Mayhap it is naptime. 19:39:37 We have New Hampshire only a couple miles from here, which some call "The South of the North". The let you have fireworks too. 19:40:34 Fireworks are illegal in Seattle. That didn't stop $idiotneighbor from setting off several loud explosions between 23:00 and midnightish last night. 19:42:05 I suppose it's at least refreshing they don't bother with the pretense... where I grew up, they could sell all sorts of stuff that was illegal to set off in the state, but they were fine with selling it as long as you signed a waiver you were taking them out of state *wink*wink* 19:42:31 (despite they not being legal in any state within a 4-5 hour drive) 19:44:14 We have plenty of tribes with different laws about fireworks so it's trivial to get them in .wa.us. 19:44:16 illinois troopers would often camp out on the state line this time of year just to bust people taking the fireworks into the state 19:47:09 [illumos-gate] 5048 panic on re-adding an interface to an aggr after removing it -- Dan McDonald 19:48:59 i mean it was fun since some family friends had moved out of town to the country and would have a big party for the 4th of july, and one of the people would usually drop a few hundred bucks on the big fireworks to set off... 19:50:11 and never got busted 19:50:25 and would do it on the 4th and not too late 19:51:43 jbk there is a proposal somewhere to teach `mdb` about some of these functions, rather than incur the cost of the frame setup. 19:52:28 illumos 17585 (fenix) 19:52:29 FEATURE 17585: MDB should look past mutex_enter() and find the real stack frame (New) 19:52:29 ↳ https://www.illumos.org/issues/17585 19:54:24 oh that would be sooooo nice 20:15:04 there was that code that someone at joyent wrote (but not sure anyone remembers who :P) that'll walk the dwarf/amd64 exception data... I had been trying to carefully fold it into libproc (as an alternative -- not to replace -- walking the frame pointers.. mostly for the lx stuff) 20:15:24 but never got around to finishing it (the code was just a proof of concept program to decode the data) 20:15:53 though I don't recall offhand if we strip that off of kernel modules or not 20:16:02 rmustacc has had a variety of dwarven prototypes 20:16:23 we strip all dwarf unless you do a build with source debugging 20:16:34 (I feel like a LOTR refrence would be appropriate now, but can't think of one :P) 20:16:45 when you do such a build, kernel stuff has dwarf too and it is 80% usable 20:17:06 (20% unusable because gdb insists on being told where in memory stuff got placed, and we don't have a strictly great answer for bss) 20:18:53 have any of you used `git cherry` (that creates a list of things you might `cherry-pick`) and been convinced that the patch-id equivalence etc, etc, actually worked? 20:19:13 my list of "cherries" includes everything I have just already picked 20:19:19 suspicious, no? 21:05:18 jbk: I'm having a strange issue with i40e and toolong_error kstat, anyway to dtrace to see what size we are rx'ing that bumps the counter? 21:07:39 (a kstat dtrace provider would be cool...) 21:08:48 is that the actual kstat name? 21:12:09 (it might be the aforementioned lack of sleep, but i'm not seeing a kstat with a name like that in the i40e driver) 21:12:10 https://github.com/TritonDataCenter/illumos-joyent/blob/master/usr/src/uts/common/io/i40e/i40e_stats.c#L865 21:12:44 oooh 21:13:38 Those are coming from the hardware. 21:13:44 I was just typing that 21:14:11 you wont' see it, because well, the NIC rejected it 21:14:23 hardware and then softmac for the stat to become a stat 21:14:40 I guess? 21:14:41 errrrr, so I'll need to get it from switchport I guess? 21:15:06 This register counts the number of received frames that are longer than maximum size as defined by the Set MAC config command 21:15:17 Smithx10: just tell people to shorten their packets. brevity is the soul of wit, or something 21:15:25 Receive Oversize Count - GLPRT_ROC in the datasheet. 21:17:52 'you using those sized packets? in this economy?' 21:18:07 jbk hahaha yes 21:19:00 Very strange.... When we tx 9k and rx 9k from another CN that are in the same vlan and never go over the gateway we are good 21:19:20 when the gateway is in the flow we get the err toolong and its getting dumped 21:19:37 oddly chelsio is working without issues 21:36:36 Oh, this is tickling an old memory of mine when I was working on a maxbw fix. I believe this could do with our setting of m_margin. Basically all our drivers set it to VLAN_TAGSZ, but I believe that's wrong. The only driver that doesn't set it to VLAN_TAGSZ...cxgbe. It sets it to 22. I would have to look more closely to give coherent story here, but basically this could be related to m_max_sdu/m_margin/etc. 21:37:12 Though I wonder if that chelsio is running an older PI, because I noticed Patrick changed cxgbe to set m_margin to VLAN_TAGSZ while he was doing his cleanup 21:37:33 There's some discussion in dlpi(7P) about the margin value. 21:37:55 Basically I ran into this while doing the maxbw work and it seemed like a huge quagmire and then I got pulled into seomthing else 21:38:11 best way out of a quagmire tbf 21:38:22 rip artax 21:38:25 richlowe: I'm really good at Irish goodbyes 21:39:41 both on SunOS 7fa0a8b5-da12-477b-a1f9-870151b56edb 5.11 joyent_20260522T154557Z i86pc i386 i86pc illumos 21:41:26 Patrick's change: 77983a892403763fb778c095cdc6d7c5388508c9 21:44:14 great thing about the network is that it will resent the packets 21:44:28 unless it's UDP, then whatever I guess 21:45:01 "left package at front door" delivery 21:52:15 Something very strange is the following: https://gist.github.com/Smithx10/56b702605e2c7517e8d63c299f2112c3 21:53:28 with 9k zones on src and destination anything over 1519 bytes is bumping the toolong_errors 21:54:43 1518 looks like VLAN-tagged ethernet.... 21:55:23 Your net0 has 9000, doe the underlying NICs in the aggr have that? There isn't a network hop between them... Same questions on the receiving side. 21:55:45 This stinks of some part of the many layers having kneecapped MTU at 1500. 21:56:16 Check the vnic you're dealing with (both sides), check any other layers of vnic between it and either an aggr or HW link. 21:56:35 If an aggr is involved, check the MTU on each link in the aggr... just to be sure. 21:56:43 @danmcd IP: Total length = 1501 bytes doesn't work IP: Total length = 1500 does. If I use a network on the same network where we don't traverse a gateway we work always 21:56:45 And finally, does your switch do stupid things? 21:57:16 Oh... if you're traversing a router then it isn't doing PathMTU discovery, or ICMP TOO BIG packets are getting dropped. 21:57:40 Not sure tho what is making our toolong_error climb? toolong_errors 9899 21:57:49 that's on the target? 21:57:54 Source 21:58:26 Then check all the way down, INCLUDING if your'e routing on-host... we should catch that. If ICMP TOO BIG messages aren't making it up?! 21:58:43 (Dammit, why do you show up this close to dinner?) 21:58:53 hahah go to dinner 21:59:17 I'll make the gist cleaner* 22:09:32 a little cleaner https://gist.github.com/Smithx10/56b702605e2c7517e8d63c299f2112c3 22:13:42 Not 100% I get how 'ping -D -s 10.82.131.32 1473' is resulting in toolong_errors ... would anything of the networking infrastructure forward that to source cn? 22:14:39 Yikes, missing the edit button from discord on that one. bah haahah 22:56:39 so any ATAPI experts around? :) 22:59:55 apparently if the device triggers an error during attach, we end up here in a taskq: https://github.com/illumos/illumos-gate/blob/master/usr/src/uts/common/io/sata/impl/sata.c#L10690-L10699 23:00:06 trying to get the error 23:00:31 except since we're in attach, satacmd_acdb_len is not yet set 23:00:47 so the bzero() length underflows (and saddness quickly follows) 23:01:36 the immediate fix is just bzero() the whole fixed sized buffer first 23:01:51 before the bcopy and don't bother with any math 23:02:30 but the 0 satacmd_acdb_len is still bothering me... it should be 12 or 16 (depending on what IDENTIFY PACKET DEVICE returns) 23:02:48 but since we need to just retry until we get a response 23:03:06 i'm wondering about just using 12 if we don't know 23:03:30 (I figure worst case it's going to ignore it if it really wants 16, but even that isn't clear)