<borneoa_>
NishanthMenon: sorry for late reply. Patches reviewed. Minor points to check.
jancoow has quit [Ping timeout: 256 seconds]
c4017w__ has joined #openocd
c4017w_ has quit [Ping timeout: 240 seconds]
<zapb_>
karlp, what are you trying to RE / what are you looking for?
Hawk777 has quit [Quit: Leaving.]
c4017w_ has joined #openocd
c4017w__ has quit [Ping timeout: 272 seconds]
thinkfat has quit [Ping timeout: 256 seconds]
thinkfat_ has joined #openocd
c4017w__ has joined #openocd
c4017w_ has quit [Ping timeout: 240 seconds]
jancoow has joined #openocd
Hawk777 has joined #openocd
c4017w_ has joined #openocd
c4017w__ has quit [Ping timeout: 256 seconds]
jancoow has quit [Ping timeout: 272 seconds]
c4017w__ has joined #openocd
c4017w_ has quit [Ping timeout: 256 seconds]
crabbedhaloablut has quit [Write error: Connection reset by peer]
crabbedhaloablut has joined #openocd
sbach has quit [Read error: Connection reset by peer]
sbach has joined #openocd
c4017w_ has joined #openocd
c4017w__ has quit [Ping timeout: 240 seconds]
crabbedhaloablut has quit [Read error: Connection reset by peer]
crabbedhaloablut has joined #openocd
Hawk777 has quit [Quit: Leaving.]
jancoow has joined #openocd
jancoow has quit [Client Quit]
jancoow has joined #openocd
c4017w__ has joined #openocd
c4017w_ has quit [Ping timeout: 240 seconds]
lukasz has joined #openocd
lukasz has quit [Client Quit]
Krazubu has joined #openocd
c4017w_ has joined #openocd
Haohmaru has joined #openocd
c4017w__ has quit [Ping timeout: 256 seconds]
nerozero has joined #openocd
Krazubu has quit [Quit: My Mac Pro has gone to sleep. ZZZzzz…]
<borneoa_>
PaulFertser: hi Paul, commit https://review.openocd.org/6771/ has been rebased but Jenkins did not rebuild it. Without Jenkins vote I think it could not be merged later. I can easily submit it again to trigger the rebuild, but could be interesting to understand why it didn't triggered Jenkins.
<PaulFertser>
borneoa_: I'm not thinking exactly straight these days, I can just log in to jenkins and press "retrigger" for it.
<borneoa_>
PaulFertser: that would fix the current issue, but I hope it will left the logs so one day you could look for the reason, if needed. Anyway no reason to rush there, it's not high priority
<borneoa_>
PaulFertser: thanks you
* olerem
has trouble to concentrate as well. at least not in a shoc state anymore
<PaulFertser>
I can only hope your relatives are safe. This is a nightmare but we're not sleeping.
<olerem>
suddenly not safe :( my mother is there and freands are in kiew
<karlp>
zapb_: I wanted to do "not work" and started poking at wireshark before you got back and said, "just use emucom blah directly" :9
<olerem>
PaulFertser: but they are loughing. means probably nothing more to loose.
<karlp>
and then I wanted to finish getting it to a functional stub, as there's been some wireshark changes since I last did it, but libjaylink or oocd direct usage is perfectly suitable.
c4017w__ has joined #openocd
c4017w_ has quit [Ping timeout: 245 seconds]
<zapb_>
karlp, I see, let me know if you need for information / help
<karlp>
thanks.
<karlp>
well, emucom channel 0 is correctly giving me ms since power on of the jlink at least :)
<karlp>
super professional single main(), no functions, pure c aesomeness....
<karlp>
thanks for the help zapb! that was much faster than I expected
<bencoh>
hm, neat / interesting
<bencoh>
(finally a real use for that builtin 5V supply! :) )
<zapb_>
karlp, you're welcome
<zapb_>
There are Python bindings for libjaylink somewhere, again no time for releasing :D
c4017w__ has joined #openocd
c4017w_ has quit [Ping timeout: 256 seconds]
<karlp>
simplicitystudio is only running on xorg, and it's still got lots of borked UI, so eventually I'd like to not be using it too much, but rebuilding all of the energy profiler is farrrr to big or right now.
<karlp>
still, AEM seems to have a higher resolution mode somewhere too, using that hcannel 0x10001 I get at best 1ms updates, but simplicitiy studio is going much finer than that.
<karlp>
looks like it can do about 80usec resolution in ssv5
<karlp>
don't suppose you know any other emucom channels used?
<karlp>
might need to get into the wireshark again afterall one day...
c4017w_ has joined #openocd
c4017w__ has quit [Ping timeout: 268 seconds]
urja has quit [Ping timeout: 256 seconds]
urja has joined #openocd
michalkotyla has joined #openocd
Bertl_zZ is now known as Bertl
c4017w__ has joined #openocd
c4017w_ has quit [Ping timeout: 272 seconds]
<karlp>
well, looks like it is using emucom chan 0x10003, requesting 32k reads, getting 84 byte chunks back...
<karlp>
some decoder updates if anyone ever wants to go down that rabbit hole...
<zapb_>
karlp, there is an extened ARM channel IIRC
<zapb_>
s/ARM/AEM
<karlp>
yeah, it's getting rapidly "interesting" but kinda out of scope of what I should really be spending time on :)
<zapb_>
:D
<zapb_>
What's you use case?
<zapb_>
Just for fun or "real work"?
<karlp>
just minimizing power consumption on a project.
<karlp>
doing it like this is hoenstly just a diversion
<karlp>
I could and will just go back to using the one built into simplicity studio
<zapb_>
:D
<zapb_>
Would be awesome to have something like that in OpenOCD
<karlp>
sure, but it gets largish quickly.
<karlp>
taht aem-dump at ~1ms resolution is already pretty neat, but having the code line colouring by correlating with SWO pc sampling is kinda cute...
<karlp>
I mean, SSv5 has mojibake, and the colour markings for code lines is busted, I can't resize that column at all, but it's still doing things... https://bin.jvnv.net/file/kTMeB.png
<karlp>
would like to be using this to explore some corner cases in the emlib stuff, explore how to save a few more uA..
<zapb_>
karlp, how to correlate SWO with AEM?
<karlp>
well, ssv5 jsut seems to enable pc sampling
c4017w_ has joined #openocd
<karlp>
and uses that, but without decoding the new 0x10003 channel, it probably won't get very far..
<karlp>
I'm not an ssv5 expert by any means...
c4017w__ has quit [Ping timeout: 256 seconds]
<zapb_>
karlp, That's the extended AEM channel?
<zapb_>
I'm still wondering how PC and AEM power traces can be correlated since AEM has no information from the core
<karlp>
yeah, according to my wireshark traces, at least,
<karlp>
I'm guessing theyre just saying "timing is close enough" as pc sampling is only sampling anyway
<karlp>
and they're probably right..
<zapb_>
Ah okay, I expected something more fancy :)
<karlp>
I can see mis coloured lines occasionally that say radio when it's not on the spikes, but it's probably good enough for a lot of work...
<zapb_>
But why do you need the extended version of AEM? Because it is more accurate?
<karlp>
well, the energy profiler inside ssv5 seems to have a resolution down to about 80usecs
<karlp>
the most I can get out of the 0x10001 channel appears to be around 1ms.
<zapb_>
Okay, I see
<karlp>
so yeah, more, more better, if you want to look at radio spikes and timer interrupts firing for processing and back to sleep
<karlp>
you can _read_ the 0x10001 channel faster, but you get the same data duplicated until the next ms..
renrelkha has quit [Quit: bye]
renrelkha has joined #openocd
c4017w__ has joined #openocd
c4017w_ has quit [Ping timeout: 256 seconds]
c4017w_ has joined #openocd
c4017w__ has quit [Ping timeout: 250 seconds]
crabbedhaloablut has quit [Remote host closed the connection]