**** BEGIN LOGGING AT Sat Aug 18 03:00:00 2012 Aug 18 03:00:14 KeatonT: it is funny on so many levels that i scarcely know where to begin...... Aug 18 03:00:33 Thank you for the support... Aug 18 03:01:49 KeatonT: not a matter of support Aug 18 03:02:50 heh Aug 18 03:02:59 pcb fab is not cheap Aug 18 03:03:10 they want bulk Aug 18 03:03:44 Hence the eventually. Aug 18 03:04:41 I really don't understand why it is so funny. It's not like I plan on creating custom design without all the proper resources. Aug 18 03:07:03 KeatonT: on face value your intention to create a custom design is not funny, but combining it with your statement about the RPi creates a "story" that tells someone about your "proper resources" Aug 18 12:20:31 Hello! I got a HDMI to DVI-I dual link adapter -- 28 pins... will it work if I use with BeagleBoard XM for outputting the DVI-D signal to a DVI based monitor? I see the manual mentioning DVI-D dual link 24 pins... I would like to know if DVI-I dual link would be back compatible with DVI-D dual link for my task? Aug 18 12:22:53 the board only produces dvi-d. the connector happens to be 'hdmi'. Aug 18 12:23:52 DVI-I is DVI-D + VGA Aug 18 12:24:11 the beagle output is single-link DVI-D Aug 18 12:24:37 using an HDMI type A connector Aug 18 12:25:40 so if your cable physically fits the socket, it will work Aug 18 12:37:37 Yes, thanks. Aug 18 12:39:43 I read that DVI - I supports both analog and digital. So, DVI-I(analog+digital) or DVI-D(digital only) both should work for BB since both supports digital, and BB outputs digital DVI. And, I checked the DVI port in the monitor, I checked the pin gaps and I think it will fit. Aug 18 12:42:22 neat, gdb 7.5 includes reverse debugging on ARM Aug 18 12:43:29 * mru is not impressed Aug 18 12:43:52 mru: have you tried it? Aug 18 12:43:58 I haven't yet Aug 18 12:44:01 no Aug 18 12:44:08 but it's nothing magical Aug 18 12:44:19 it's just an execution trace Aug 18 12:44:20 i'm curious what it means exactly Aug 18 12:44:57 stepping backwards afaik Aug 18 12:45:03 in a recorded trace Aug 18 12:45:09 nothing spectacular Aug 18 12:45:19 just a fancy way of scrolling through a log Aug 18 20:22:06 hey folks, i read on a google groups email from someone at TI that "the am355x PRU has a 32-bit hardware multiplier to perform a multiply in a single cycle"... is there an assembly instruction for this? :) Aug 18 20:24:15 why you dont write the someone from ti an email directly? Aug 18 20:36:42 not a bad idea Aug 18 20:38:37 actually, there isn't an instruction for that Aug 18 20:40:26 * mranostay wanders in Aug 18 20:40:50 * mru hands mranostay a cold beer Aug 18 20:42:58 mranostay: you're old enough to drink, right? Aug 18 20:48:32 he mru Aug 18 20:48:34 mru does arm11 support ldrex on shared memory? Aug 18 20:48:59 which arm11? Aug 18 20:50:02 hm ah Aug 18 20:50:03 this matters? Aug 18 20:50:03 its an armv6 Aug 18 20:50:04 which armv6 Aug 18 20:50:04 ah okay so it matters Aug 18 20:50:04 so maybee the qualcomm guys ah right Aug 18 20:50:04 are Aug 18 20:50:05 and using swp for spinlock Aug 18 20:50:36 oh wait, 32-bit ldrex is supported on all armv6 Aug 18 20:50:36 mru: i'm hungover. so no thanks... Aug 18 20:50:44 the other sizes are only in v6k (and v7) Aug 18 20:51:20 mru they claim for msm7x27 that ldrex and strexeq is not supported for shared mem Aug 18 20:51:36 shared in what sense? Aug 18 20:52:36 hi prpplague Aug 18 20:52:44 mranostay: greetings Aug 18 20:52:51 mranostay: hmm, so you're not old enough to drink responsibly Aug 18 20:53:24 mru -> https://www.codeaurora.org/gitweb/quic/le/?p=kernel/msm.git;a=blob;f=arch/arm/mach-msm/include/mach/remote_spinlock.h;h=75b70f3b5913e7f9f5f4fe451ce039c287be6e02;hb=msm-3.4 Aug 18 20:53:49 CONFIG_MSM_REMOTE_SPINLOCK_SWP Aug 18 20:55:00 in Kconfig swp version is selected Aug 18 20:56:20 and I wonder if I can safely use the ldrex variant for my phone with armv6 arm11 Aug 18 20:56:48 is that qualcomm thing an arm11 or something they built themselves? Aug 18 20:57:41 its msm7x27 Aug 18 20:58:17 woglinde: yeah, but it's qualcomm so you're lucky to even have the part number Aug 18 20:59:16 hm I will leave it Aug 18 20:59:29 and if I get the kernel run, I could test the other Aug 18 20:59:33 the issue here is whether the exclusive monitor is cleared by other system masters Aug 18 20:59:35 thanks anyway Aug 18 21:00:38 will the memory in question be accessed by more than one master? Aug 18 21:02:36 master is what? Aug 18 21:02:37 cpu? Aug 18 21:02:48 anything capable of initiating a memory access Aug 18 21:03:10 hm call hierachy is not so easy to see Aug 18 21:03:16 so dont know Aug 18 21:03:29 well, it's a single-core system, I assume Aug 18 21:06:27 mru: why drink responsible? :) Aug 18 21:06:49 there are other things that are more fun to do irresponsibly Aug 18 21:11:14 mru yes single-core Aug 18 21:12:47 [22:51:36] shared in what sense? <------ on qualcomm, memory is shared between baseband processor and app processor Aug 18 21:13:40 just like omap shares memory between arm and dsp Aug 18 21:17:58 the question is, do other masters write to memory regions used with ldrex/strex? Aug 18 21:18:07 hi phh ;) Aug 18 21:32:22 boxysean: I got the multiplier you are referring to working, just a couple of days ago. The problem is that it's not documented nor guaranteed to work. Aug 19 00:35:40 Heya, all. Aug 19 00:36:38 Where is a pointer to hook a 24C256 up to the BeagleBone? My A3 Bone isn't seeing the EEPROM off of the I2C port for it. Aug 19 01:01:21 Hold on, I had the two lines switched... Aug 19 01:01:29 There it goes. Cancel that question. **** ENDING LOGGING AT Sun Aug 19 03:00:01 2012