**** BEGIN LOGGING AT Thu Jun 25 02:59:57 2020 Jun 25 12:59:15 lorforlinux: do you know how to rebase with git? Jun 25 14:31:19 Hello all Jun 25 14:31:28 Hello! Jun 25 14:32:04 Hi Jun 25 14:33:50 *going through the codes Jun 25 14:34:05 one of the prime blockers that deepankar pointed out was about which shared RAM to use. He's shared some sample code on using the OCMC on the ARM SS. Jun 25 14:35:11 upon a small discussion with Abhishek earlier today the conclusion is that the shared RAM space on the PRUss is better to start with Jun 25 14:37:02 jkridner: if you're around, could you please comment on the same? Jun 25 14:37:21 Yeah today was fruitfull in that sense cleared out several things. Voip discussions are much better Jun 25 14:37:59 i have one more doubt how is a gpio chip driver a better choice than a char driver for our application Jun 25 14:38:32 https://github.com/beagleboard/linux/blob/4.19/drivers/rpmsg/rpmsg_pru.c https://github.com/beagleboard/linux/blob/4.19/drivers/tty/serial/pru_suart.c are some places to look for inspiration Jun 25 14:38:33 https://github.com/beagleboard/linux/blob/4.19/arch/arm/boot/dts/am5729-beagleboneai.dts Jun 25 14:39:15 I will spend tonight going through the code base and then probably will come up with some implementation tommorow Jun 25 15:01:48 It really just depends on how much memory is needed and which processor needs fast access. The shared data ram in the PRUSS will be much faster to access from the PRUs. Jun 25 15:01:57 ...but they are smaller. Jun 25 15:02:02 er, it is smaller. Jun 25 15:08:24 It is 12KB which i think should be enough Jun 25 15:08:30 21KB Jun 25 15:08:36 * 12KB Jun 25 15:09:31 yes, it is smaller than the 64 KB on ARM ss Jun 25 15:09:37 but access times and access mechanisms are a trade off Jun 25 15:10:17 one final vote and deepankar will should have no blockers to get going Jun 25 15:10:28 * one final vote and deepankar should have no blockers to get going Jun 25 15:10:40 * one final vote on this and deepankar should have no blockers to get going Jun 25 15:49:04 * nwan[m] sent a long message: < https://matrix.org/_matrix/media/r0/download/matrix.org/lQrXIKOuJGVzIKhwdnDtAUSZ > Jun 25 15:50:39 pdp7: Do you know if there is a possibility to receive software interrupts from incoming ptp timestamps and driving a timer by this? Jun 25 16:12:08 I guess using, $ git checkout then $ git rebase master. Right? Jun 25 16:41:36 nwan[m]: of is the meeting now now? Jun 25 16:42:36 oh it was earlier. I was in bootlin graphics training Jun 25 16:43:25 nwan[m]: I've never used ptp. What is creating ptp timestamps? Jun 25 16:44:34 there is some external hardware? Jun 25 16:45:08 @pdp7 the network card of BBB and BBAI is able to create timestamps via hardware when packets arrive. And I am wondering if there is any way to issue a software interrupt through linux kernel when this happens Jun 25 16:46:16 https://training.ti.com/intro-ptp-procsdk-linux-series Jun 25 16:46:20 that is an interesting question Jun 25 16:46:50 I will look a bit deeper into this. Maybe i find something Jun 25 16:47:01 Thought you might know Jun 25 16:49:11 I'm not familiar with ptp but it does seem like that should be possible. It is an interesting idea so let me know how it goes. Jun 25 16:49:27 jkridner: do you have any knowledge of ptp? Jun 25 16:52:02 @pdp7 allright I will inform you when I find something usefull. Jun 25 16:52:44 here's an idea, post about it on the omap list Jun 25 16:52:45 https://lore.kernel.org/linux-omap/ Jun 25 16:53:02 I've seen people ask timer questions, and tony lindgren is very helpful Jun 25 16:53:32 like this Jun 25 16:53:33 https://lore.kernel.org/linux-omap/2019546.2yLtp7J43K@linux-5fgm.suse/ Jun 25 16:53:33 Thank you :) Jun 25 16:54:40 oh, and there is #linux-omap on freenode Jun 25 16:54:59 though i never see an activity, there are people there. you could ask over there too Jun 25 16:55:13 okay thanks Jun 25 16:56:01 and NishanthM is there, he is a TI kernel engineer Jun 25 16:57:52 Cool. Will look into that tomorrow. Thank you very much for your help Jun 25 17:00:33 no problem Jun 25 17:01:13 it sounds like a useful thing to have on the AM3358 Jun 25 17:48:07 I compiled gcc-10 for PRU target, with arch=arm32 https://github.com/VedantParanjape/pru-gcc-arm32/tree/master Abhishek_ Trying this now Jun 25 17:50:26 * I compiled gcc-10 for PRU target, with arch=arm32 https://github.com/VedantParanjape/pru-gcc-arm32/tree/master Abhishek_ Trying this now. earlier contacted dinuxbg, helped me out with this **** ENDING LOGGING AT Fri Jun 26 03:01:39 2020