/srv/irclogs.linaro.org/2018/06/27/#linaro-coresight.txt

ototo#startmeeting OpenCSD Weekly16:00
linarobotMeeting started Wed Jun 27 16:00:32 2018 UTC.  The chair is ototo. Information about MeetBot at http://wiki.debian.org/MeetBot.16:00
linarobotUseful Commands: #action #agreed #help #info #idea #link #topic.16:00
* ototo greets mpoirier and broonie16:00
mpoirier hello16:00
broonieHi16:00
* ototo is checking if Mike is supposed to be here16:01
broonieHe just sent a status report so I expect so.16:01
* ototo nods16:01
ototomikelbbn: thanks for joining, we've just started16:01
mikelbbnok16:01
ototompoirier: let's start from you, Mathieu16:01
mpoirierOk16:02
mpoirierI16:02
mpoirierI've been spending a lot of time reviewing patches and talking to people on the mailing lists.16:02
mpoirierOn the development side supporting more than one CPU in CPU-wide trace scenarios is going well.16:03
mpoirierUser space is done and I'm working on the kernel portion at this time.16:03
mpoirierThat will lead to a preparatory set that will need to be mainlined before implementation for CPU-wide.16:04
mpoirierIn short, I expect the work to be split in 4 sets, each needing to go in before the other.16:05
mpoirierI've also been working with Wookey on package maintenance and delivery.16:05
mpoirierDebian is pretty much done, now I need to see how we can get Ubuntu going.16:06
mpoirierThey are on a different schedule than Debian and their per-tools package doesn't yet include a reference to the openCSD library.16:06
mpoirierEOF16:06
mikelbbnmpoirier: these 4 sets - do they replace the 1-CPU cpu wide set you published to the coresight ml last week?16:07
ototompoirier: perhaps you've noticed that I've promoted the res[ective jira Story into an Epic - it's way bigger than what Story could be.16:08
mpoirierototo, Yes I'16:08
mpoirierYes I've seen this and I agree.16:08
mpoirierI also updated the kernel release as you requested.16:08
mpoiriermikelbbn, It will be a super set of that.16:08
ototo\o/16:08
mpoirierAnd because that is a lot of code do move around the whole thing can't go in together.16:09
mpoirierI'll break it in 4 part, that will be easier for Arnaldo to digest.16:09
mikelbbnunderstood16:09
mpoirierIt also included somewhat controversial changes to the perf core, which is what I'm working on.16:10
mpoirierThat will be part 1.16:10
mpoirier*includes16:10
mpoirierEOF16:10
ototomikelbbn: it's your turn, Mike16:11
mikelbbnOK, I've released v0.9.0 - this has the new programmers guide + performance improvements and API enhancements that support work in perf done by mpoirier and in ARM16:13
mikelbbnAlso been looking at patches, plus trying to get 4.18-rc1 that mporier put into the opencsd-perf github up and running with the CTI work that has been on the backburner.16:14
mikelbbnHit some snags here - my Juno broke as I was downloading the new kernel into flash and I have not been able to recover it correctly - may need a complete wipe and restore.16:14
mikelbbnPlus the 4.18 kernel does not boot reliably on the DB410c - there appears to be a nasty race condition which makes it random which of the mmc drivers is regarded as mmc0 - if it chooses wrong then the boot hangs.16:16
mikelbbnfinally, on the occasions that it does boot, it does not seem to be seeing the CTI components - which Juno was before I broke it.16:16
mikelbbnAll in all, a bit of work to sort this mess out.16:16
mikelbbnEOF16:16
ototoThanks16:17
ototoIs there anything anybody would like to discuss today? Any questions/concerns?16:17
mpoirierVacations perhaps...16:18
mpoirierIt's that time of year.16:18
mikelbbnI've nothing longer than 3 days before september16:19
ototoI'm taking a vertical vacation this year - all Mondays from now till the end of September. WIll also have to take some "horizontal" piece as otherwise I'm unable to use up all my remaining days by the ned of FY.16:19
mpoirierI'm in the same situation as ototo... Taking all Fridays off between now and September.16:20
mpoirierThat and the first week of August.16:21
mpoirierEOF16:21
ototomikelbbn: what will be the next release - 1.0 or 0.10? :)16:22
ototomikelbbn: also - do you have any idea about the date for that next release?16:23
mikelbbnExcellent question - kind of depends what issues get raised in the meantime. I'll aim for 1.0 that includes the library testing code, but if I need to tweak stuff to support other work, such as the perf stuff I did for 0.9, then that might result in 0.1016:23
mikelbbnOf if someone finds a horrible bug!16:24
* ototo nods16:25
mikelbbnNo date in mind, again input dependent - I'll focus on CTI for the next few weeks then go back to library testing with a testing release around connect time16:25
ototomikelbbn: that CTI work - it's not touching OpenCSD, or is it?16:26
mikelbbncompletely separate. Only really controls trace triggers / trace halt which are incidental in decoding16:27
ototoThanks16:27
ototoAnything else from anybody? If not - finishing in 30...16:28
ototo20...16:28
mpoirierAll good here.16:28
ototo10...16:28
mikelbbni'm fine16:28
ototoThanks for attending - the meeting is over.16:28
ototo#endmeeting16:28
linarobotMeeting ended Wed Jun 27 16:28:51 2018 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)16:28
linarobotMinutes:        https://irclogs.linaro.org/meeting-logs/linaro-coresight/2018/linaro-coresight.2018-06-27-16.00.html16:28
linarobotMinutes (text): https://irclogs.linaro.org/meeting-logs/linaro-coresight/2018/linaro-coresight.2018-06-27-16.00.txt16:28
linarobotLog:            https://irclogs.linaro.org/meeting-logs/linaro-coresight/2018/linaro-coresight.2018-06-27-16.00.log.html16:28

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!