2012-05-07T00:07:13 *** xiangfu has quit IRC (Ping timeout: 276 seconds) 2012-05-07T00:25:22 *** xiangfu has joined #rtems 2012-05-07T01:01:46 *** mumptai has joined #rtems 2012-05-07T01:40:38 *** arvind_khadri has joined #rtems 2012-05-07T01:47:15 *** mumptai has quit IRC (Read error: Operation timed out) 2012-05-07T01:54:40 *** arvind_khadri has quit IRC (Remote host closed the connection) 2012-05-07T01:55:13 *** arvind_khadri has joined #rtems 2012-05-07T01:55:58 *** sebhub has joined #rtems 2012-05-07T02:04:52 *** xiangfu has quit IRC (Ping timeout: 276 seconds) 2012-05-07T02:15:02 *** arvind_khadri has quit IRC (Remote host closed the connection) 2012-05-07T02:17:23 *** arvind_khadri has joined #rtems 2012-05-07T02:17:24 *** arvind_khadri has joined #rtems 2012-05-07T02:21:10 *** xiangfu has joined #rtems 2012-05-07T03:07:42 *** arvind_khadri has quit IRC (Remote host closed the connection) 2012-05-07T03:09:48 *** arvind_khadri has joined #rtems 2012-05-07T03:09:48 *** arvind_khadri has joined #rtems 2012-05-07T03:16:00 *** xiangfu has quit IRC (Ping timeout: 260 seconds) 2012-05-07T03:27:45 *** xiangfu has joined #rtems 2012-05-07T05:36:04 *** QingPei has quit IRC (Ping timeout: 252 seconds) 2012-05-07T05:38:06 *** QingPei has joined #rtems 2012-05-07T08:29:30 *** cdcs has joined #rtems 2012-05-07T09:11:22 *** weiY has joined #rtems 2012-05-07T09:44:51 *** xiangfu has quit IRC (Remote host closed the connection) 2012-05-07T10:15:17 *** gedare has joined #rtems 2012-05-07T10:31:48 *** sebhub has quit IRC (Read error: Operation timed out) 2012-05-07T10:40:43 *** weiY has quit IRC (Ping timeout: 265 seconds) 2012-05-07T10:41:58 *** antgreen has joined #rtems 2012-05-07T10:47:23 *** DrJoel has joined #rtems 2012-05-07T10:47:23 *** DrJoel has joined #rtems 2012-05-07T10:47:23 *** ChanServ sets mode: +o DrJoel 2012-05-07T10:54:43 *** Hesham has joined #rtems 2012-05-07T11:04:12 *** QingPei has quit IRC (Ping timeout: 256 seconds) 2012-05-07T11:12:41 *** ash_ has joined #rtems 2012-05-07T11:16:09 *** Hesham has quit IRC (Read error: Connection reset by peer) 2012-05-07T11:17:08 *** Hesham has joined #rtems 2012-05-07T11:20:26 *** cdcs has quit IRC (Ping timeout: 245 seconds) 2012-05-07T11:23:50 *** QingPei has joined #rtems 2012-05-07T11:32:31 *** QingPei1 has joined #rtems 2012-05-07T11:32:55 *** QingPei has quit IRC (Quit: Leaving.) 2012-05-07T11:55:42 *** arvind_khadri has quit IRC (Ping timeout: 272 seconds) 2012-05-07T11:56:36 *** ash_ has quit IRC (Remote host closed the connection) 2012-05-07T11:59:24 *** mumptai has joined #rtems 2012-05-07T12:28:34 *** QingPei1 is now known as QingPei 2012-05-07T12:40:23 *** QingPei has quit IRC (Quit: Leaving.) 2012-05-07T12:41:27 *** QingPei has joined #rtems 2012-05-07T12:42:54 *** QingPei1 has joined #rtems 2012-05-07T12:42:57 *** QingPei has quit IRC (Read error: Connection reset by peer) 2012-05-07T12:47:58 *** arvind_khadri has joined #rtems 2012-05-07T12:55:11 *** sevikkk has quit IRC (Quit: Leaving.) 2012-05-07T13:01:21 *** antgreen has quit IRC (Ping timeout: 265 seconds) 2012-05-07T13:13:32 *** Hesham has quit IRC (Quit: Leaving.) 2012-05-07T13:13:39 *** Hesham has joined #rtems 2012-05-07T13:15:31 *** sevikkk has joined #rtems 2012-05-07T13:20:06 *** Hesham1 has joined #rtems 2012-05-07T13:20:06 *** Hesham has quit IRC (Read error: Connection reset by peer) 2012-05-07T13:21:19 *** QingPei1 is now known as QingPei 2012-05-07T13:37:15 *** QingPei has quit IRC (Ping timeout: 260 seconds) 2012-05-07T13:39:28 *** Hesham1 has quit IRC (Ping timeout: 256 seconds) 2012-05-07T15:13:22 *** arvind_khadri has quit IRC (Read error: Operation timed out) 2012-05-07T16:07:25 *** gedare has quit IRC (Quit: Leaving) 2012-05-07T17:05:37 *** DrJoel has quit IRC (Quit: Leaving) 2012-05-07T18:24:01 *** mumptai has quit IRC (Read error: Operation timed out) 2012-05-07T18:44:13 *** A1Sheds has quit IRC (Read error: Operation timed out) 2012-05-07T18:59:37 *** A1Sheds has joined #rtems 2012-05-07T18:59:37 *** A1Sheds has joined #rtems 2012-05-07T19:14:02 *** antgreen has joined #rtems 2012-05-07T19:37:14 *** xiangfu has joined #rtems 2012-05-07T20:41:21 *** xiangfu has quit IRC (Ping timeout: 255 seconds) 2012-05-07T20:53:08 *** xiangfu has joined #rtems 2012-05-07T21:20:43 *** antgreen has quit IRC (Remote host closed the connection) 2012-05-07T21:56:43 *** QingPei has joined #rtems 2012-05-08T00:12:31 *** sevikkk1 has joined #rtems 2012-05-08T00:21:25 *** sevikkk has quit IRC (*.net *.split) 2012-05-08T00:21:25 *** kiwichris has quit IRC (*.net *.split) 2012-05-08T00:21:27 *** tuxmaniac has quit IRC (*.net *.split) 2012-05-08T00:21:27 *** verm__ has quit IRC (*.net *.split) 2012-05-08T00:21:30 *** lcpfnvcy has quit IRC (*.net *.split) 2012-05-08T00:21:31 *** peerst has quit IRC (*.net *.split) 2012-05-08T00:21:32 *** ChanServ has quit IRC (*.net *.split) 2012-05-08T00:22:06 *** kiwichris has joined #rtems 2012-05-08T00:22:06 *** tuxmaniac has joined #rtems 2012-05-08T00:22:06 *** peerst has joined #rtems 2012-05-08T00:22:06 *** verm__ has joined #rtems 2012-05-08T00:22:06 *** lcpfnvcy has joined #rtems 2012-05-08T00:22:06 *** ChanServ has joined #rtems 2012-05-08T00:22:06 *** barjavel.freenode.net sets mode: +o ChanServ 2012-05-08T00:22:50 *** QingPei has left #rtems 2012-05-08T01:15:04 *** arvind_khadri has joined #rtems 2012-05-08T01:53:40 *** mumptai has joined #rtems 2012-05-08T02:12:18 *** arvind_k has joined #rtems 2012-05-08T02:13:00 *** arvind_khadri has quit IRC (Read error: Connection reset by peer) 2012-05-08T02:16:52 *** arvind_k is now known as arvind_khadri 2012-05-08T02:17:05 *** arvind_khadri has joined #rtems 2012-05-08T02:41:15 *** mumptai has quit IRC (Ping timeout: 260 seconds) 2012-05-08T03:36:40 *** arvind_khadri has quit IRC (Quit: Leaving) 2012-05-08T03:36:52 *** arvind_khadri has joined #rtems 2012-05-08T03:40:25 *** cdcs has joined #rtems 2012-05-08T04:44:55 *** A1Sheds has quit IRC (Ping timeout: 260 seconds) 2012-05-08T07:57:31 *** cdcs has quit IRC (Ping timeout: 245 seconds) 2012-05-08T08:48:04 *** antgreen has joined #rtems 2012-05-08T08:51:31 *** arvind_khadri has quit IRC (Ping timeout: 260 seconds) 2012-05-08T08:54:40 *** cdcs has joined #rtems 2012-05-08T09:10:09 *** gedare has joined #rtems 2012-05-08T09:13:55 *** arvind_khadri has joined #rtems 2012-05-08T09:13:55 *** arvind_khadri has joined #rtems 2012-05-08T10:17:32 *** antgreen has quit IRC (Ping timeout: 245 seconds) 2012-05-08T10:35:30 *** xiangfu has quit IRC (Ping timeout: 260 seconds) 2012-05-08T10:55:39 *** gedare has quit IRC (Quit: Ex-Chat) 2012-05-08T10:58:15 *** Hesham has joined #rtems 2012-05-08T11:12:57 *** arvind_khadri has quit IRC (Read error: Operation timed out) 2012-05-08T11:14:00 *** arvind_khadri has joined #rtems 2012-05-08T11:26:15 *** arvind_khadri has quit IRC (Ping timeout: 260 seconds) 2012-05-08T11:40:31 *** arvind_khadri has joined #rtems 2012-05-08T11:45:23 *** arvind_khadri has quit IRC (Ping timeout: 250 seconds) 2012-05-08T11:56:32 *** Hesham has quit IRC (Quit: Leaving.) 2012-05-08T12:08:26 *** mumptai has joined #rtems 2012-05-08T12:21:01 *** Hesham has joined #rtems 2012-05-08T12:21:33 *** gedare has joined #rtems 2012-05-08T12:21:40 hey hesham. 2012-05-08T12:21:50 hi gedare 2012-05-08T12:22:01 Hesham, I had to leave early. right now i need about 15 min and then we can chat 2012-05-08T12:22:13 take your time :) 2012-05-08T12:22:16 ok 2012-05-08T12:39:18 *** Hesham has quit IRC (Quit: Leaving.) 2012-05-08T12:39:30 *** Hesham has joined #rtems 2012-05-08T12:40:07 Hesham: hi 2012-05-08T12:40:15 hi 2012-05-08T12:40:33 How are you gedare ? 2012-05-08T12:40:44 good. ready to meet? 2012-05-08T12:40:50 yeah 2012-05-08T12:41:15 what i did from the last meeting 2012-05-08T12:41:31 is that i got skyeye working with csb337 BSP 2012-05-08T12:41:43 and studied rbheap 2012-05-08T12:41:54 skyeye-csb337: great!@ 2012-05-08T12:41:54 but i did not grasp all of it yet 2012-05-08T12:42:38 sebhub has stated that csb337 is "known to have problems in RTEMS 4.10 and later" so be a little careful there. 2012-05-08T12:42:52 * gedare pokes tuxmaniac 2012-05-08T12:43:09 just to get his tuxmaniac's attention so he can scroll back or watch in :) 2012-05-08T12:43:09 i have tried edb7312 with skyeye 2012-05-08T12:43:38 I saw your question about rbheap on the ml... 2012-05-08T12:43:50 you should try to add more context to your questions they are usually lacking in details 2012-05-08T12:43:50 is he Aanjhan ? 2012-05-08T12:43:53 yeah 2012-05-08T12:44:02 aha 2012-05-08T12:44:13 Sebastian replied 2012-05-08T12:45:00 "Would you mind elaborating the "protect task stacks" a bit more?" 2012-05-08T12:45:01 go back to the previous emails about rbheap (you can look for them in the archives, which are a little hard to search but I think you can use some google-fu somehow...) 2012-05-08T12:45:38 he (sebhub) has stated before that the purpose of rbheap is to provide protected task stacks, mainly though this was to prevent a stack overrun from affecting the allocator metadata 2012-05-08T12:45:49 so not the same thing as applying memory attributes to memory regions 2012-05-08T12:45:57 yeah i have read this 2012-05-08T12:46:28 but does rbheap deals with pages ? 2012-05-08T12:46:31 however, if you page-align the malloc of the chunk then you can apply page-based protection 2012-05-08T12:46:41 not explicitly, but that is why it has an alignment field 2012-05-08T12:46:46 so that you can page-align chunks 2012-05-08T12:46:53 sebhub used to call the chunks as pages 2012-05-08T12:47:07 that made me confused 2012-05-08T12:47:11 but i complained because they weren't really pages unless you page-aligned them and made the allocation size a multiple of page-size 2012-05-08T12:47:28 which is hw specific and rtems has no notion of it 2012-05-08T12:47:41 so you have to do it from the application level I guess 2012-05-08T12:47:50 i have troubles with understanding rbheap alignment 2012-05-08T12:47:53 that makes a lot of the memory protection a little tricky. 2012-05-08T12:48:50 rtems do not deal with pages in memory allocation yet ? 2012-05-08T12:48:52 the alignment just makes sure that every allocated chunk has a minimum alignment. 2012-05-08T12:48:56 correct 2012-05-08T12:49:43 The how malloc is implemented ? 2012-05-08T12:50:29 ok so ... 2012-05-08T12:50:54 you know i wanted to reach to the part that rbheap deal with pages to think how memory protection would be implemented by using MMU / domains / AP 2012-05-08T12:51:15 it uses "pages", but they aren't necessarily the pages you know of for virtual memory 2012-05-08T12:51:28 software pages 2012-05-08T12:51:44 yeah..but no translation to hw pages/framees 2012-05-08T12:52:01 it just carves up the memory... using a flat memory model 2012-05-08T12:52:08 but where are the attributes of each page or domain ? 2012-05-08T12:53:19 There are none ... 2012-05-08T12:53:44 Until we define some 2012-05-08T12:53:49 We will make them 2012-05-08T12:54:17 we will define them depending on software pages ? or actual pages ? 2012-05-08T12:54:39 We need to make them somewhat generic, since not all hw supports "pages" 2012-05-08T12:55:01 that is why there are multiple layers between the user API and the hw 2012-05-08T12:55:49 in order to translate between blocks of memory and hw protection mechanism whether that is pages or base/bounds registers or nonexistent 2012-05-08T12:56:47 we would make stubs for hw that does not support pages in that project ? 2012-05-08T12:57:17 *** antgreen has joined #rtems 2012-05-08T12:57:36 *** Hesham has quit IRC (Quit: Leaving.) 2012-05-08T12:57:45 *** Hesham has joined #rtems 2012-05-08T12:57:52 sorry 2012-05-08T12:57:57 did i miss anything ? 2012-05-08T12:58:01 Hesham: Possibly, or at least stubs that can be used generically. No I waiteed. 2012-05-08T12:58:40 these memory protection attributes would be added to rbheap ? 2012-05-08T12:59:04 maybe -- or something similar to it perhaps 2012-05-08T12:59:13 that we create it ? 2012-05-08T12:59:30 or perhaps not even care about the allocation, just enforce attributes on specific memory "zones" 2012-05-08T12:59:40 and it is up to the application to tell what zone has what attribute 2012-05-08T12:59:56 aha 2012-05-08T13:00:02 the easiest way for an application to create/allocate those would be to use an rbheap I think. 2012-05-08T13:00:20 *** cdcs has quit IRC (Quit: Page closed) 2012-05-08T13:00:23 Mostly this exercise of looking at allocators is for you to get familiar with score code, and to think about these kinds of problems with memory management 2012-05-08T13:00:35 i view the memory management as orthogonal to the memory protection/attributes. 2012-05-08T13:01:15 and did i get my feet wet till now or what ? 2012-05-08T13:01:20 but I also think there is an opportunity for you to design and implement a shared Allocation manager in Score. Whether it is part of Arena or is a separate thing I'm not sure. 2012-05-08T13:02:01 So keep plugging away at this. It will be useful 2012-05-08T13:02:07 i want to know the low-level implementation fr malloc 2012-05-08T13:02:13 for* 2012-05-08T13:02:22 check out the Heap Manager in score 2012-05-08T13:02:34 that implements the internal memory allocator for RTEMS 2012-05-08T13:03:00 and malloc invoked heap manager ? 2012-05-08T13:03:16 otherwise I think we rely on newlib. I actually don't know :) 2012-05-08T13:03:53 i can ask about that 2012-05-08T13:04:11 ok. make sure to give enough details ;) 2012-05-08T13:04:36 And about my question ( rbheap ) should i ask again or leave it for now ? 2012-05-08T13:06:01 Probably you should respond... 2012-05-08T13:06:21 that's ok 2012-05-08T13:06:33 I would say... say what you know about rbheap, and ask about what you don't 2012-05-08T13:06:50 Ok 2012-05-08T13:07:03 we will use rbheap for the rest of the project then ? 2012-05-08T13:07:05 you can ask if the alignment of chunks makes it possible to apply protection attributes. 2012-05-08T13:07:17 have to determine if it is the right thing to use. 2012-05-08T13:07:47 what about the superscore ? 2012-05-08T13:08:01 supercore 2012-05-08T13:08:25 I'm thinking the first step is to design a framework for memory allocation along the lines of what you already have been investigating 2012-05-08T13:09:17 The next step is to take the prior work with libmmu/memory protection and get it working on git head, then see what is missing.. 2012-05-08T13:10:03 it does not work ? 2012-05-08T13:10:10 I don't know if it works or not. 2012-05-08T13:10:36 I managed to pass some simple tests, but it is fragile 2012-05-08T13:11:17 ok 2012-05-08T13:11:28 Ultimately I think we will want an Arena Manager that uses an allocator that can support memory protection 2012-05-08T13:12:04 and the Arena will provide additional services including setting protection/attribute bits...maybe sharing arenas b/w tasks, context switching arenas, so on 2012-05-08T13:12:24 the allocator may not exist in RTEMS yet ? 2012-05-08T13:12:28 correct 2012-05-08T13:13:18 my intuition is that rbheap can be adapted for this purpose, but we need to know more about it to be certain. 2012-05-08T13:13:26 Then Arena manager will abandon Region/Partition Manager 2012-05-08T13:13:42 ? 2012-05-08T13:14:04 I don't know yet. But try to keep the idea of Allocation separate from Arena 2012-05-08T13:14:15 i think it will be great if we focused in just one manager or we can create our own allocator 2012-05-08T13:14:41 Yes. I believe that we will focus on one allocator. But first we want a framework for allocators 2012-05-08T13:15:12 that would need a research outside RTEMS docs right ? 2012-05-08T13:15:27 or to use RTEME alloctors ? 2012-05-08T13:15:40 I think the RTEMS allocators can be adapted for the job 2012-05-08T13:15:53 We'll at least try ;) 2012-05-08T13:16:12 what i know till now is heap alloctor 2012-05-08T13:16:16 what else ? 2012-05-08T13:16:33 Region and Partition are allocators with a little extra tacked on. 2012-05-08T13:16:42 right? 2012-05-08T13:16:58 yeah but i think Region is a Manager more that alloctor 2012-05-08T13:17:12 It rely on Heap manager 2012-05-08T13:17:33 So we could make use of heap manager directely 2012-05-08T13:19:24 Region Manager does not just pass allocation requests directly through to malloc/free 2012-05-08T13:20:04 Or does it? :) 2012-05-08T13:20:27 It does not 2012-05-08T13:20:55 it deals with objects 2012-05-08T13:20:59 ok, but the allocate/free parts are pretty thin wrappers for heap allocate/free though? 2012-05-08T13:22:03 region manager does not invoke any malloc/free 2012-05-08T13:22:08 functions 2012-05-08T13:22:23 it uses heap managers functions only 2012-05-08T13:22:29 and objects API 2012-05-08T13:22:50 That's what i got from studying it 2012-05-08T13:22:58 That sounds right 2012-05-08T13:23:23 So Heap Manager is one allocator... 2012-05-08T13:23:28 yeah 2012-05-08T13:23:35 Capturing the buffer management of partitions would be another... 2012-05-08T13:23:40 and then rbheap seems to be a third 2012-05-08T13:23:56 i will start with Heap manager 2012-05-08T13:24:09 it will be well documented i think 2012-05-08T13:24:31 probably. Generally speaking the interface for an allocator is quite simple: allocate, free, extend 2012-05-08T13:24:52 i care about its implementation 2012-05-08T13:25:39 buffer management rely on chain manager 2012-05-08T13:25:45 ok. my general question would be: which allocators can be used for memory protection, and how... 2012-05-08T13:26:41 I have to Study heap manager and get down to the low-level implementation to determine that 2012-05-08T13:27:00 the thing i did not got when i studied rbheap 2012-05-08T13:27:43 okay 2012-05-08T13:27:53 try not to get lost down in the weeds. 2012-05-08T13:28:23 i'll try 2012-05-08T13:28:55 after we determine the allocator ( if we found a suitable one ) 2012-05-08T13:29:10 we would create the Arena Manager right ? 2012-05-08T13:29:20 yeah. 2012-05-08T13:29:31 And make use of Region/Partition Manager as an example 2012-05-08T13:30:16 we might not even need an allocator if we leave it to the application to determine which memory can have protection bits applied 2012-05-08T13:30:26 but it would be good to have an idea of how the arena can be used 2012-05-08T13:30:36 and for that you need to understand the allocators 2012-05-08T13:31:18 i think so 2012-05-08T13:31:49 all right. sounds good 2012-05-08T13:31:51 So , now no need for the SuperCore work ? 2012-05-08T13:32:08 not sure yet. I still think it might be good 2012-05-08T13:32:48 let's put in the shift for the time being and care about alloctors and the ability to give memory protection 2012-05-08T13:32:55 ok 2012-05-08T13:33:00 or what you think ? 2012-05-08T13:33:10 right now is learning 2012-05-08T13:33:24 there is not too much more to learn for the supercore allocator framework 2012-05-08T13:33:40 What to do first ? Research for memory alloctors outside RTEMS then come back to RTEMS Allocators or what ? 2012-05-08T13:33:45 it might be nice to write-up what you learned 2012-05-08T13:33:55 only deal with RTEMS allocators for now 2012-05-08T13:34:01 i am writting it down 2012-05-08T13:34:02 we don't want to implement any new allocator 2012-05-08T13:34:20 I'd really like to see a design for an internal (supercore) allocator interface 2012-05-08T13:34:38 behind which you can put Heap Manager, Buffer Manager (partitions), and RBHeap 2012-05-08T13:35:38 And all of them can be used simultaneously right ? 2012-05-08T13:35:49 not like supercore scheduler 2012-05-08T13:36:14 yeah the application just carves up memory into start/end and the allocator works within that 2012-05-08T13:36:54 the application may use many alloctors , or just one ? 2012-05-08T13:37:44 can use many. it would just be a way to unify the interface. I'm not sure how useful yet though.. have to think some more 2012-05-08T13:38:55 Ok that would be like a report after the research of allocators 2012-05-08T13:39:05 yeah. 2012-05-08T13:39:32 also have you read the code for the previous MMU work? 2012-05-08T13:39:47 that's the next step 2012-05-08T13:39:53 it's on my todo list this week 2012-05-08T13:39:58 ok do you have the link? 2012-05-08T13:40:26 yeah i have some links for previous GSoC projects 2012-05-08T13:40:32 and reports about them 2012-05-08T13:40:42 ok 2012-05-08T13:41:03 but it would be good if you send all the work related to mmu work and code paths as a package 2012-05-08T13:41:09 http://code.google.com/p/gsoc2011-rtems-mmu-support-project/ 2012-05-08T13:41:16 its all available there as much as I know 2012-05-08T13:41:23 that's the most recent / up-to-date code 2012-05-08T13:41:38 i have that link 2012-05-08T13:41:45 ok. 2012-05-08T13:41:57 and there's one belongs to 2009 project 2012-05-08T13:42:06 that one is outdated by now. 2012-05-08T13:42:26 i think the 2011 project built from the 2009 project. and then I did a little extra work with it too 2012-05-08T13:42:45 ok i will work on 2011 project only 2012-05-08T13:42:49 good 2012-05-08T13:43:07 that would be separate from Arena design right now 2012-05-08T13:43:35 yes! design first and then see how you might adopt the code into your design 2012-05-08T13:43:41 until we could combine allocators & mmu work & our new memory protection definitions in just one box 2012-05-08T13:43:44 after the design is approved. 2012-05-08T13:44:11 Design of Arena manager right ? 2012-05-08T13:44:23 yup. eventually an application developer should have "one-stop shopping" for protection memory regions 2012-05-08T13:44:35 right. 2012-05-08T13:44:48 ok I think that's good for today 2012-05-08T13:44:56 good meeting 2012-05-08T13:45:12 yep. remember to submit your paperwork to google 2012-05-08T13:45:34 i did more that a week ago , thanks for reminding me :) 2012-05-08T13:45:38 than * 2012-05-08T13:45:40 ok great 2012-05-08T13:45:42 *** A1Sheds has joined #rtems 2012-05-08T13:45:42 *** A1Sheds has joined #rtems 2012-05-08T13:45:56 see you later. 2012-05-08T13:46:03 see you :) 2012-05-08T13:47:08 *** gedare has quit IRC (Quit: Leaving) 2012-05-08T13:51:46 *** Hesham has quit IRC (Quit: Leaving.) 2012-05-08T14:43:43 *** arvind_khadri has joined #rtems 2012-05-08T15:14:19 *** arvind_khadri has quit IRC (Ping timeout: 245 seconds) 2012-05-08T15:38:30 *** gedare has joined #rtems 2012-05-08T15:43:55 *** mumptai has quit IRC (Quit: Verlassend) 2012-05-08T16:03:23 *** Hesham has joined #rtems 2012-05-08T16:32:03 *** antgreen has quit IRC (Remote host closed the connection) 2012-05-08T16:40:44 *** Hesham has quit IRC (Quit: Leaving.) 2012-05-08T16:52:47 *** Hesham has joined #rtems 2012-05-08T16:53:05 *** Hesham has left #rtems 2012-05-08T17:45:14 *** gedare has quit IRC (Quit: Leaving) 2012-05-08T17:48:41 *** DrJoel has joined #rtems 2012-05-08T17:48:41 *** DrJoel has joined #rtems 2012-05-08T17:48:41 *** ChanServ sets mode: +o DrJoel 2012-05-08T17:52:10 *** DrJoel has quit IRC (Client Quit) 2012-05-08T20:33:30 *** xiangfu has joined #rtems 2012-05-08T20:33:56 *** xian9fu has joined #rtems 2012-05-08T20:38:05 *** xiangfu has quit IRC (Ping timeout: 260 seconds) 2012-05-09T01:58:45 *** sebhub has joined #rtems 2012-05-09T02:06:57 *** wangkun has joined #rtems 2012-05-09T02:07:05 good monring 2012-05-09T02:07:42 good monring 2012-05-09T02:07:55 *** wangkun has left #rtems 2012-05-09T02:15:46 *** wangkun has joined #rtems 2012-05-09T02:16:47 *** wangkun has left #rtems 2012-05-09T02:19:07 *** wangkun has joined #rtems 2012-05-09T02:19:22 *** wangkun has left #rtems 2012-05-09T02:43:53 sebhub, hi 2012-05-09T02:44:02 hi 2012-05-09T02:44:13 char a[1] has pos == 0 2012-05-09T02:44:28 the size is one more than the position 2012-05-09T02:44:44 yes 2012-05-09T02:44:53 but why pos + 1 and not pos + count? 2012-05-09T02:45:16 I would have to check in detail. The write follows ? 2012-05-09T02:45:22 so add the count 2012-05-09T02:45:33 It is the case of the lseek past the end 2012-05-09T02:45:53 so the write operations will grow the file implicitly? 2012-05-09T02:46:10 From the lseek'ed position 2012-05-09T02:46:20 Yes 2012-05-09T02:46:23 ah, ok, then the pos + 1 make sense 2012-05-09T02:46:35 Great. I will post to the list. 2012-05-09T02:49:17 Do we have support for limits which govern the file size ? 2012-05-09T02:49:50 I did not know of any. 2012-05-09T02:49:51 the dosfs seems to perform some checks, but we have not test cases yet 2012-05-09T02:50:20 posix has EFBIG 2012-05-09T02:50:56 also IMFS has some checks, but returns the wrong error code and is not overall posix conform 2012-05-09T02:52:31 Where is the DOSFS ? 2012-05-09T02:53:14 fat_file_write() function 2012-05-09T02:54:06 The return EIO ? 2012-05-09T02:54:12 at line 393 ? 2012-05-09T02:54:14 i think the [PATCH 2/3] Filesystem: PR1871: Fix O_APPEND is ok as it is 2012-05-09T02:54:26 It will not wokr. 2012-05-09T02:54:31 It will break 2012-05-09T02:54:34 the RFS 2012-05-09T02:54:41 yes line 393 2012-05-09T02:55:03 if ((iop->flags & LIBIO_FLAGS_APPEND) != 0) 2012-05-09T02:55:04 iop->offset = rtems_rfs_file_size (file); 2012-05-09T02:55:05 I suspect this due to the limits of FAT12 2012-05-09T02:55:29 You need to seek 2012-05-09T02:55:29 this will trigger the " if (pos >= rtems_rfs_file_size (file))" 2012-05-09T02:55:42 Hmmm ok. 2012-05-09T02:56:25 i will add a comment to this 2012-05-09T02:56:26 I have no problem with the seek happening in the read and write if the pos has moved 2012-05-09T02:56:49 This is independent from the issue you are talking about 2012-05-09T02:58:30 In the RFS I think EFBIG needs to come from rtems_rfs_block_map_grow 2012-05-09T02:59:26 do you prefer to move the lseek into the read/write handler? this will turn the RFS file lseek into the default file lseek 2012-05-09T02:59:28 rtems-rfs-group.c:37 2012-05-09T03:00:02 If a test is made in the RFS file layer where the pos is checked against the current handle then yes 2012-05-09T03:00:19 I think this should work ok. 2012-05-09T03:00:45 what happens if the open a file multiple times and read/write to it with different offsets? 2012-05-09T03:00:51 to write and read call the file layer seek and check the current value 2012-05-09T03:01:04 i think we need this lseek move for this case as well? 2012-05-09T03:01:10 The pos is in the handle while the common parts are in the shared parts 2012-05-09T03:01:39 ok, and the buffer are shared? 2012-05-09T03:01:53 A file handle in the RFS has a local part and the inode is shared with the buffers 2012-05-09T03:02:17 See the top of rtems-rfs-file.h 2012-05-09T03:02:33 ok 2012-05-09T03:02:34 This way the stat on a file is fast 2012-05-09T03:03:26 The bpos or block position is local to each handle 2012-05-09T03:04:24 bbl 2012-05-09T03:05:32 *** xian9fu has quit IRC (Ping timeout: 252 seconds) 2012-05-09T03:20:22 *** arvind_khadri has joined #rtems 2012-05-09T03:32:27 i have one more question regarding the "rtems_rfs_file_set_size (file, pos + 1);" what is if count == 0? since it seems that the rtems_rfs_file_io_end() will update the file size a "rtems_rfs_file_set_size (file, pos)" should be all right 2012-05-09T03:41:08 *** wangkun has joined #rtems 2012-05-09T03:41:15 test 2012-05-09T03:41:23 *** wangkun has left #rtems 2012-05-09T04:20:09 *** arvind_khadri has quit IRC (Ping timeout: 245 seconds) 2012-05-09T05:06:13 *** QingPei has joined #rtems 2012-05-09T05:26:01 *** xiangfu has joined #rtems 2012-05-09T05:56:02 *** xiangfu has quit IRC (Ping timeout: 255 seconds) 2012-05-09T06:33:31 *** weiY has joined #rtems 2012-05-09T06:35:27 Hi, chris. i have seen your mail. set the time for irc meeting thursday or now? 2012-05-09T06:39:36 *** rokka_ has joined #rtems 2012-05-09T06:42:10 *** rokka has quit IRC (Ping timeout: 252 seconds) 2012-05-09T06:43:16 *** arvind_khadri has joined #rtems 2012-05-09T06:43:22 *** arvind_khadri has joined #rtems 2012-05-09T07:18:04 *** arvind_khadri has quit IRC (Ping timeout: 245 seconds) 2012-05-09T07:41:24 *** weiY has quit IRC (Ping timeout: 245 seconds) 2012-05-09T07:57:36 *** weiY has joined #rtems 2012-05-09T08:05:02 *** cdcs has joined #rtems 2012-05-09T08:33:16 *** lcpfnvcy has quit IRC (Read error: Operation timed out) 2012-05-09T08:41:29 *** cdcs has quit IRC (Quit: Page closed) 2012-05-09T08:45:34 *** DrJoel has joined #rtems 2012-05-09T08:45:34 *** DrJoel has joined #rtems 2012-05-09T08:45:34 *** ChanServ sets mode: +o DrJoel 2012-05-09T08:48:33 sebhub, good morning/evening 2012-05-09T09:15:02 *** rtemsLogger has joined #rtems 2012-05-09T09:15:13 *** DrJoel has joined #rtems 2012-05-09T09:15:13 *** DrJoel has joined #rtems 2012-05-09T09:15:13 *** ChanServ sets mode: +o DrJoel 2012-05-09T09:24:36 *** gedare has joined #rtems 2012-05-09T09:25:46 good morning 2012-05-09T09:32:20 hey gedare 2012-05-09T09:32:36 how's it going 2012-05-09T09:32:53 is there any suggestion for/against using bitfields? 2012-05-09T09:33:52 generally frowned upon as questionably portable based on size of int, endianness, etc from my recollection. 2012-05-09T09:34:09 I don't know if situation has improved in recent years but .... 2012-05-09T09:38:40 *** sevikkk1 has quit IRC (Ping timeout: 272 seconds) 2012-05-09T09:47:00 bitfields are regularly broken on gcc is my impression 2012-05-09T09:55:26 ok i will avoid 2012-05-09T09:55:48 http://publications.gbdirect.co.uk/c_book/chapter6/bitfields.html 2012-05-09T09:56:06 The main use of bitfields is either to allow tight packing of data or to be able to specify the fields within some externally produced data files. C gives no guarantee of the ordering of fields within machine words, so if you do use them for the latter reason, you program will not only be non-portable, it will be compiler-dependent too. The Standard says that fields are packed into ‘storage units’, which are typically machine words. The packing ord 2012-05-09T09:56:07 d whether or not a bitfield may cross a storage unit boundary, are implementation defined. To force alignment to a storage unit boundary, a zero width field is used before the one that you want to have aligned. 2012-05-09T09:56:30 i still use bitfields for compact configuration data 2012-05-09T09:56:53 *** QingPei has left #rtems 2012-05-09T09:56:55 in case data space matters 2012-05-09T09:57:18 *** xiangfu has joined #rtems 2012-05-09T09:58:40 but not to touch HW. 2012-05-09T09:58:56 what's the git command to just get the 1 line summary of changes on your current branch 2012-05-09T09:59:23 yes, for hw i wouldn't use it 2012-05-09T10:01:55 *** xiangfu has quit IRC (Ping timeout: 260 seconds) 2012-05-09T10:06:13 git shortlog master..HEAD 2012-05-09T10:07:44 *** weiY has quit IRC (Ping timeout: 250 seconds) 2012-05-09T10:15:22 *** xiangfu has joined #rtems 2012-05-09T10:24:31 *** sevikkk has joined #rtems 2012-05-09T10:34:19 *** sebhub has quit IRC (Ping timeout: 245 seconds) 2012-05-09T10:36:17 *** xiangfu has quit IRC (Remote host closed the connection) 2012-05-09T10:58:26 *** arvind_khadri has joined #rtems 2012-05-09T11:11:12 DrJoel: do you know of any Application aware MMU desgins? 2012-05-09T11:11:44 DrJoel: I mean, some MMU like HW/design that is aware of which application has access to which part of memory? 2012-05-09T11:11:51 not in swoftware 2012-05-09T11:12:03 but in hardware. Yes makes the system static. But any pointers? 2012-05-09T11:17:43 ARINC-653 has configuration tables about which "partitions" have access to which parts of memory. But that is software layered on top of hardware. 2012-05-09T11:17:54 I don't know of any custom built hardware that is that system aware. 2012-05-09T11:59:28 DrJoel: thanks. 2012-05-09T12:13:41 was that actually an answer? :) 2012-05-09T12:15:44 DrJoel: if you didnt know, I assume there isnt any :P 2012-05-09T12:16:22 plus the timings of gedare and myself seem to be out of sync. He pings when I am not there and I ping when he is not there :) 2012-05-09T12:16:32 Have to follow up on the MMU ARM project 2012-05-09T12:16:35 am lagging behind 2012-05-09T12:16:36 hi 2012-05-09T12:16:40 gedare: WAIT 2012-05-09T12:16:42 :D 2012-05-09T12:16:53 gedare: finally 2012-05-09T12:16:54 :) 2012-05-09T12:17:01 gedare: so how can I be of help :) 2012-05-09T12:17:02 it's not so bad right now. i have hesham learning about score code 2012-05-09T12:17:33 is he documenting it as he learns somewhere? 2012-05-09T12:17:46 yeah i'll tell him to share his docs with you. i thought he already did 2012-05-09T12:17:55 tuxmaniac, I am not nearly as reliable as a thorough literature search. :D 2012-05-09T12:17:59 and i'm going to ask him to write a little report 2012-05-09T12:18:05 i have some pointers for you about that mmu stuff tuxmaniac 2012-05-09T12:18:19 gedare: am all ears 2012-05-09T12:18:20 :) 2012-05-09T12:18:22 DrJoel: :D 2012-05-09T12:18:26 If there is one thing about computer science to know, it is that you don't know everything and what you do know has a shelf life :) 2012-05-09T12:18:56 DrJoel: looks like gedare 's shelf is more up to date 2012-05-09T12:19:12 my friend is working on his thesis in a related area. we have tracked a few "application aware" hw-protection solutions 2012-05-09T12:19:21 one of the more famous is mondrian.. 2012-05-09T12:19:35 hmm 2012-05-09T12:19:37 http://groups.csail.mit.edu/cag/scale/mondriaan/index.html 2012-05-09T12:19:52 btw it is all academic 2012-05-09T12:19:57 no one in industry will do this 2012-05-09T12:20:13 well, you can do it maybe with asic/fpga designs 2012-05-09T12:20:26 gedare: haha. this is also for some academic work ;-) 2012-05-09T12:20:32 am collecting references 2012-05-09T12:21:01 ok. you can read the papers under "Hardware Containers for Software Components" section of my research page 2012-05-09T12:21:02 http://home.gwu.edu/~gedare/research.html# 2012-05-09T12:21:17 we have related work sections that should point you to some other hw solutions of interest 2012-05-09T12:21:22 oh cool 2012-05-09T12:21:33 thansk a lot. 2012-05-09T12:21:57 gedare, Just impractical/expensive/too custom for commercial? 2012-05-09T12:22:04 too custom 2012-05-09T12:22:20 no one will produce this hw at economy-of-scale prices 2012-05-09T12:22:44 DrJoel: too static 2012-05-09T12:22:49 Seems like it would require too much design and risk for too little payback unless you were doing something super secure and special 2012-05-09T12:22:54 but it can be done for fpga designs for special purpose. 2012-05-09T12:23:00 yup. 2012-05-09T12:23:14 But even then, there are probably MILS solutions with near-COTS boards that would meet the requirements 2012-05-09T12:23:16 DrJoel: exactly. probably for IMDs, etc 2012-05-09T12:23:24 IMDs? 2012-05-09T12:23:29 implantable medical devices 2012-05-09T12:23:47 *** antgreen has joined #rtems 2012-05-09T12:23:52 IMD was the name of Thomas' company before he merged with Peter and formed embedded brains :) 2012-05-09T12:24:04 :) 2012-05-09T12:24:43 I don't think IMDs get enough care to this kind of thing. Risk of doing something at this level is inconceivable .. the certification alone of such a unique solution would be a huge undertaking 2012-05-09T12:25:15 they need simple certifiable solutions.. too custom and time/money spent convincing FDA device is safe 2012-05-09T12:25:28 and the more complex it is, the more likely to have a bug and kill someone. 2012-05-09T12:25:48 That's why they have so little security on the tuneable IMDs. 2012-05-09T12:28:21 thats very true 2012-05-09T12:28:43 plus, these guys even though how simple their architecture is protect it like its something the world has never seen 2012-05-09T12:29:54 our work focuses on the problem of allowing memory protection for non-page-sized (smaller) chunks of memory 2012-05-09T12:30:23 but the problems / solutions end up being somewhat application-aware at least to get good perfromance 2012-05-09T12:32:09 are you thinking about general-purpose applications, or an application that is aware of all its tasks and their mmu requirements? 2012-05-09T12:32:17 *** tuxmaniac has quit IRC (Disconnected by services) 2012-05-09T12:32:27 guess i should have pinged him ;0 2012-05-09T12:32:36 *** tuxmaniac has joined #rtems 2012-05-09T12:32:51 *** tuxmaniac has joined #rtems 2012-05-09T12:40:54 sigh.. it is taking 7-11 minutes per BSP build on my computer. Admittedly this is with tests enabled but still... :( 2012-05-09T12:41:15 my laptop has ... Intel(R) Core(TM)2 Duo CPU T7500 @ 2.20GHz 2012-05-09T12:59:43 gedare... interesting buglet for you :) 2012-05-09T12:59:51 #define CONFIGURE_UNLIMITED 2012-05-09T12:59:51 #define CONFIGURE_UNIFIED_WOIRKSPACE 2012-05-09T12:59:51 #define CONFIGURE_RTEMS_INIT_TASKS_TABLE 2012-05-09T12:59:51 #define CONFIGURE_STACK_CHECKER_ENABLED 2012-05-09T12:59:51 #define CONFIGURE_INIT 2012-05-09T12:59:51 #include 2012-05-09T12:59:56 home/joel/newbsd/bsp-install/powerpc-rtems4.11/psim/lib/include/rtems/confdefs.h:2540:6: error: #error "CONFIGURATION ERROR: No tasks or threads configured!!" 2012-05-09T12:59:56 init.c:76:24: fatal error: demo-shell.h: No such file or directory 2012-05-09T13:00:06 hmmm 2012-05-09T13:00:53 try #define CONFIGURE_UNLIMITED_OBJECTS 2012-05-09T13:01:18 doh and it is woIrkspace :) 2012-05-09T13:01:30 heh that too. 2012-05-09T13:01:48 PICNIC 2012-05-09T13:01:51 :-) 2012-05-09T13:03:57 New TCP/IP stack update: Jennifer is making a sweep over the "interesting" PCI drivers to see if they all can be linked and not leave undefined symbols 2012-05-09T13:04:17 I am trying to hack together a simple test to initialize it. 2012-05-09T14:22:27 ah. i'm beating on red-black trees. 2012-05-09T14:25:22 freebsd has a nice single header implmentation for rb and splay trees 2012-05-09T14:25:26 if you haven't seen it already 2012-05-09T14:26:33 http://svnweb.freebsd.org/base/head/sys/sys/tree.h?view=log 2012-05-09T15:13:52 i have not. I implemented red-black trees in rtems a couple years back 2012-05-09T15:16:19 i like mine better:) 2012-05-09T15:17:18 that is interesting though. 2012-05-09T15:28:27 gedare, any comments on my CVS Ids branch? 2012-05-09T15:28:59 havne't looked since glancing. did you collapse into two changes? 2012-05-09T15:29:21 No. Trying to get entire tree to build again BEFORE I commit it. :( 2012-05-09T15:30:01 ah. 2012-05-09T15:30:01 19 small commits -- mostly cleanup ... but PowerPC C++ broke thanks to stdint.h change 2012-05-09T15:31:53 I was going to squash all scripted edits into 1 patch 2012-05-09T15:32:12 I think there are at least two other commits after that. 2012-05-09T15:32:14 ok 2012-05-09T15:32:20 1) scripted edits 2012-05-09T15:32:36 2) fixes to undo overly agressive scripted edits and make it build again :) 2012-05-09T15:32:47 3) manual removal of CVS Ids (misses) 2012-05-09T15:33:13 4) maybe .. other stuff I spotted along the way. 2012-05-09T15:34:50 sounds good try not to get too distracted with little things though 2012-05-09T15:35:00 better to make a list and deal with them later 2012-05-09T15:35:09 annoying, but better. 2012-05-09T15:36:11 hmm. i waited too long to go outside. going to rain soon 2012-05-09T15:36:16 and storm 2012-05-09T15:36:29 guess i get to work longer 2012-05-09T15:47:03 *** antgreen has quit IRC (Read error: Connection reset by peer) 2012-05-09T15:47:04 *** arvind_khadri has quit IRC (Excess Flood) 2012-05-09T15:47:28 *** arvind_khadri has joined #rtems 2012-05-09T16:22:05 gedare: ah, i didn't know that, what's the license? 2012-05-09T16:22:21 the rtems license 2012-05-09T16:22:32 ah, that counts that out then unfortunatly 2012-05-09T16:22:53 I'm not sure how separable it would be anyways. 2012-05-09T16:23:28 all my clients have banned gpl at the door due to gpl3 2012-05-09T16:23:40 oh, it's ingrained? 2012-05-09T16:23:51 yeah it's for kernel 2012-05-09T16:24:03 so is sys/tree.h in freebsd 2012-05-09T16:24:07 yeah 2012-05-09T16:24:23 it's still an insular implmentation though i collect these they come in handy 2012-05-09T16:24:51 yeah that's good. mine is pretty much integrated into rtems though. 2012-05-09T16:25:40 the github updater is broken i'll take a look at it tonight i keep getting this really odd git error no idea what's causing it 2012-05-09T16:25:46 :/ 2012-05-09T16:25:54 i deployed it at a half a dozen different places 2012-05-09T16:26:06 and this is the only one that broke.. 2012-05-09T16:31:50 *** lcpfnvcy has joined #rtems 2012-05-09T16:32:06 that's what rtems is good for! ;) 2012-05-09T17:03:37 squashing :) 2012-05-09T17:15:13 Down to 3 Miscellaneous - Clean up file headers so patterns followed 2012-05-09T17:15:13 Remove All CVS Id Strings Possible Using a Script 2012-05-09T17:15:13 cpukit/aclocal - Fix one mistake by script 2012-05-09T17:15:13 Remove CVS Id Strings (manual edits after script) 2012-05-09T17:15:19 4 I mean 2012-05-09T17:17:18 great. I say push a branch to your personal repo with those 4 commits and make an announcement, wait a day or two and then push to rtems 2012-05-09T17:17:37 the longer to wait the worse the drift and this touches so many files 2012-05-09T17:17:38 Are those good enough messages? 2012-05-09T17:17:46 EVERY file .. a very rare patch :) 2012-05-09T17:17:51 yeah i think so 2012-05-09T17:18:01 they're a sight better than Ralf's! 2012-05-09T17:18:25 hmmmm.... email list size may be issue 2012-05-09T17:18:27 $ du -s -k *patch 2012-05-09T17:18:27 4 0001-Miscellaneous-Clean-up-file-headers-so-patterns-foll.patch 2012-05-09T17:18:27 2876 0002-Remove-All-CVS-Id-Strings-Possible-Using-a-Script.patch 2012-05-09T17:18:27 4 0003-cpukit-aclocal-Fix-one-mistake-by-script.patch 2012-05-09T17:18:27 56 0004-Remove-CVS-Id-Strings-manual-edits-after-script.patch 2012-05-09T17:18:36 2.8 MB patch for largest. 2012-05-09T17:19:21 don't worry email list will cut it off 2012-05-09T17:19:25 don't mail those! 2012-05-09T17:19:28 :) 2012-05-09T17:19:36 push a branch to your personal repo like you did before 2012-05-09T17:19:44 and email a link to the branch to us 2012-05-09T17:19:47 I already have done that :) 2012-05-09T17:19:56 But rtems-vc will get the email eventually 2012-05-09T17:20:05 when you go to push to master it will cut off the emails on rtems-vc which is fine 2012-05-09T17:20:21 anyone can go to the repo to see the full info 2012-05-09T17:21:15 so don't worry about it :) 2012-05-09T17:26:44 i will setup patches tonight for the mailing lists 2012-05-09T17:27:11 i forgot to get jeff to create patches.rtems.org i'll send an email about that 2012-05-09T17:28:23 verm__: what do you mean 2012-05-09T17:28:38 http://patches.libav.org/project/FFmpeg-devel/list/ 2012-05-09T17:28:46 it listens to mailing lists then strips out patches 2012-05-09T17:28:52 it comes with a commandline utility where you can list 2012-05-09T17:28:55 Ah. Nice. 2012-05-09T17:29:00 or grab a patch remotely and apply it using git am 2012-05-09T17:29:05 That would make my life easier. when I have to go find patches I submitted and lost :) 2012-05-09T17:29:15 (does it on its own) you can also mark patches as 'in progress', 'invalid' etc 2012-05-09T17:29:30 yeah i'll go back and process all the old mailing list logs 2012-05-09T17:30:49 DrJoel: your set vector stuff is going to trouble me if i go back to the bsp_set_vector refactor :( 2012-05-09T17:31:09 which i really should do 2012-05-09T17:31:28 I just eliminated the ones which were dead. 2012-05-09T17:31:40 ok. it shouldnt' be too hard then to 'skip' conflicts 2012-05-09T17:31:53 We (for better or worse) have Simple Vectored and PIC interrupt models with different APIs. 2012-05-09T17:32:36 maybe better. harder to accidently misuse one for the other 2012-05-09T17:32:38 Most of those were PowerPC and MIPS BSPs which have dead code or (for MIPS) were still calling SV when we just changed them to PIC 2012-05-09T17:32:50 ok.. i squished my bug and am successful, time to clock out 2012-05-09T17:32:58 so, i like naming machines using a scheme that i administrate -- this makes it easier to refer to them 2012-05-09T17:32:58 good night 2012-05-09T17:33:01 gedare, fyi I also disabled something you care about 2012-05-09T17:33:06 uh oh 2012-05-09T17:33:09 what now 2012-05-09T17:33:10 gedare: wait a sec 2012-05-09T17:33:17 * gedare is popular 2012-05-09T17:33:42 SV had code in the score ISR and Classic Interrupt Manager. Only some of it was disabled on PIC models. 2012-05-09T17:33:54 chris has suggested a programming them: mutex, task, atomic, driver for example.. anyone have any issues with this? 2012-05-09T17:33:55 I tried to disable all that should be disabled but it pulled a thread I didn't want to get into 2012-05-09T17:33:59 otherwise i'll be picking these types of names 2012-05-09T17:34:01 DrJoel: ok i'll have to look into that 2012-05-09T17:34:09 first one will be called mutex. 2012-05-09T17:34:22 Primarily vector number, rtems_isr, and rtems_isr_entry types. Those are SV but show up in PIC code 2012-05-09T17:34:33 verm__: machine names with programming theme seems fine to me 2012-05-09T17:34:44 ok, good 2012-05-09T17:34:58 can we have deadlock? 2012-05-09T17:35:03 yep 2012-05-09T17:35:05 added 2012-05-09T17:35:07 DrJoel: yes i saw those when i went through the bsp_set_vector code originally. we can look at that. 2012-05-09T17:35:24 :) 2012-05-09T17:35:34 i'll need about 15 of them :) (each jail gets a name too that runs a major service) 2012-05-09T17:36:03 ok. i'm out 2012-05-09T17:36:11 *** gedare has quit IRC (Quit: Leaving) 2012-05-09T17:44:47 *** DrJoel has quit IRC (Quit: Leaving) 2012-05-09T18:28:04 *** arvind_khadri has quit IRC (Ping timeout: 245 seconds) 2012-05-09T18:28:04 *** peerst has quit IRC (Ping timeout: 245 seconds) 2012-05-09T18:36:35 *** arvind_khadri has joined #rtems 2012-05-09T19:42:43 *** ash__ has joined #rtems 2012-05-09T20:38:55 *** ash__ has quit IRC (Remote host closed the connection) 2012-05-09T21:09:58 *** zw_yao has joined #rtems 2012-05-09T21:13:32 *** zw_yao has quit IRC (Remote host closed the connection) 2012-05-09T21:50:07 *** arvind_khadri has quit IRC (Ping timeout: 250 seconds) 2012-05-09T21:52:28 *** QingPei1 has joined #rtems 2012-05-10T00:05:42 *** kuzew has quit IRC (Ping timeout: 272 seconds) 2012-05-10T00:21:52 *** arvind_khadri has joined #rtems 2012-05-10T00:33:05 *** kuzew has joined #rtems 2012-05-10T06:25:41 *** weiY has joined #rtems 2012-05-10T06:33:43 Hi chris, are you free now? 2012-05-10T07:05:09 *** xiangfu has joined #rtems 2012-05-10T07:09:21 *** sebhub has joined #rtems 2012-05-10T07:18:24 *** DrJoel has joined #rtems 2012-05-10T07:18:24 *** DrJoel has joined #rtems 2012-05-10T07:18:24 *** ChanServ sets mode: +o DrJoel 2012-05-10T07:19:27 sebhub, if you are around the keyboard, I have a question 2012-05-10T07:20:43 hi 2012-05-10T07:28:52 good afternoon 2012-05-10T07:29:06 How did you determine the set of code USB devices to initialize? 2012-05-10T07:29:56 We are building up the TCP/IP equivalent of what is now in rtems-bsd-config.h and wondered if you have pointers 2012-05-10T07:32:12 please wait two hours, i want to commit some test cases for SWI and CALLOUT 2012-05-10T07:32:43 i am a bit in a hurry now 2012-05-10T07:32:56 ok.. you can answer via email. so take your time 2012-05-10T07:33:03 I just wanted to catch you online and ask 2012-05-10T07:36:47 the initialization was broken, but works now locally 2012-05-10T07:36:54 i was delayed with the commit 2012-05-10T08:15:26 *** DrJoel has quit IRC (Quit: Leaving) 2012-05-10T09:00:02 *** xiangfu has quit IRC (Quit: Leaving) 2012-05-10T09:50:52 *** antgreen has joined #rtems 2012-05-10T10:01:09 *** gedare has joined #rtems 2012-05-10T10:30:08 *** sebhub has quit IRC (Ping timeout: 240 seconds) 2012-05-10T11:07:51 *** weiY has quit IRC () 2012-05-10T11:12:36 *** antgreen has quit IRC (Read error: Connection reset by peer) 2012-05-10T11:50:04 *** arvind_khadri has quit IRC (Ping timeout: 244 seconds) 2012-05-10T13:14:51 *** antgreen has joined #rtems 2012-05-10T14:03:41 *** gedare has quit IRC (Quit: Leaving) 2012-05-10T14:05:54 *** DrJoel has joined #rtems 2012-05-10T14:05:54 *** DrJoel has joined #rtems 2012-05-10T14:05:54 *** ChanServ sets mode: +o DrJoel 2012-05-10T14:07:52 *** antgreen has quit IRC (Read error: Connection reset by peer) 2012-05-10T14:53:55 *** gedare has joined #rtems 2012-05-10T15:04:40 hey gedare.. 2012-05-10T15:04:45 hi drjoel 2012-05-10T15:05:11 I think I've found why m32r (all bsps) and i386/pc386dx are broken with gcc 4.7 2012-05-10T15:05:39 good start... 2012-05-10T15:05:44 mistakes in libgcc/config.host which were not caught in "testing" the RPMs before being released 2012-05-10T15:05:51 ;-) 2012-05-10T15:06:03 super 2012-05-10T15:06:10 where's my eye-roll emote 2012-05-10T15:06:13 Have I given enough notice on the cvs id removal patch? 2012-05-10T15:06:44 * DrJoel is listening to Kraftwerk Man Machine .. 2012-05-10T15:06:46 i think any complaints would already have come in. besides it is all text modifications. it will cause problems with conflicts but there is not much to be done about that. 2012-05-10T15:07:15 no.. I am creating my own as I make things build 2012-05-10T15:08:00 fwiw I think reviewing each architecture to ensure it doesn't have bsp/driver code which use methods or types from the wrong ISR model support would be a good clean up 2012-05-10T15:08:26 Plus I am not sure what is the rule on PIC for knowing the API to use. Probably need to do a taxonomy 2012-05-10T15:08:31 and then reduce it :) 2012-05-10T15:09:23 yeah 2012-05-10T15:09:48 lots of little bits and pieces where things get copied, modified, and turn in to dead code 2012-05-10T15:09:57 then revived and misused 2012-05-10T15:10:12 I expect that for CVS Ids. 2012-05-10T15:10:29 It took me years to kill all the wrong URLs for the license after we got rtems.org 2012-05-10T15:11:07 And another FWIW .. some files have a line between the second and third line of this: 2012-05-10T15:11:09 * The license and distribution terms for this file may be 2012-05-10T15:11:09 * found in the file LICENSE in this distribution or at 2012-05-10T15:11:09 * http://www.rtems.com/license/LICENSE. 2012-05-10T15:11:44 I found a lot of those but am sure I didn't get them all. Any suggestions on finding them? 2012-05-10T15:12:19 *** mw1 is now known as mwalle 2012-05-10T15:12:30 All I can think of is another pass with a multi-line sed patter which is something like "distribution or at\n.*\*.*\n.*http:....." 2012-05-10T15:12:31 only thing i can think of is to look for two *s with blanks and newline between them and an http after the second. 2012-05-10T15:12:39 yeah 2012-05-10T15:12:50 And hope there aren't any in weird comment blocks :) 2012-05-10T15:12:53 heh. 2012-05-10T15:13:03 you can review it and should see pretty easily 2012-05-10T15:13:12 all the eliminated ones should be just : * 2012-05-10T15:13:12 the list of files I manually needed to remove Ids on is ripe for review 2012-05-10T15:13:35 those files lean heavily to being either badly formatted or obsolete 2012-05-10T15:16:58 yeah. well if we define a good style and formatting guide we can put together a list of such things for gci or people who want to do small tasks 2012-05-10T15:17:13 also good for gsoc students in the current "bonding" phase 2012-05-10T15:17:19 but need to have these things ready for them :0 2012-05-10T15:17:31 rtems-janitors ;) 2012-05-10T15:20:12 unfortunately a lot of janitorial duties require insight .. reformatting a file shouldn't but judging it obsolete does :) 2012-05-10T15:24:01 true. 2012-05-10T16:10:53 *** DrJoel has quit IRC (Quit: Leaving) 2012-05-10T19:01:58 *** gedare has quit IRC (Quit: Leaving) 2012-05-10T20:03:33 *** xiangfu has joined #rtems 2012-05-10T20:57:03 *** gedare has joined #rtems 2012-05-10T21:20:43 *** arvind_khadri has joined #rtems 2012-05-10T21:20:43 *** arvind_khadri has joined #rtems 2012-05-10T21:34:33 *** gedare has quit IRC (Quit: Leaving) 2012-05-10T21:36:38 *** arvind_khadri has quit IRC (Read error: Operation timed out) 2012-05-10T21:36:47 *** arvind_khadri has joined #rtems 2012-05-10T21:55:21 *** dr__house has quit IRC (Quit: leaving) 2012-05-10T21:55:55 *** dr__house has joined #rtems 2012-05-10T21:55:55 *** dr__house has joined #rtems 2012-05-10T22:32:06 *** arvind_khadri has quit IRC (Ping timeout: 248 seconds) 2012-05-10T22:44:08 *** xiangfu has quit IRC (Remote host closed the connection) 2012-05-10T22:48:45 *** xiangfu has joined #rtems 2012-05-10T22:51:01 *** arvind_khadri has joined #rtems 2012-05-10T23:13:43 *** arvind_khadri has quit IRC (Ping timeout: 252 seconds) 2012-05-10T23:44:23 *** dr__house has quit IRC (Remote host closed the connection) 2012-05-10T23:44:44 *** dr__house has joined #rtems 2012-05-10T23:44:44 *** dr__house has joined #rtems 2012-05-10T23:52:50 *** xiangfu has quit IRC (Ping timeout: 260 seconds) 2012-05-10T23:57:05 *** QingPei1 has left #rtems 2012-05-11T00:04:35 *** xiangfu has joined #rtems 2012-05-11T00:10:20 *** xiangfu has quit IRC (Ping timeout: 260 seconds) 2012-05-11T00:23:36 *** xiangfu has joined #rtems 2012-05-11T01:19:53 *** arvind_khadri has joined #rtems 2012-05-11T01:36:35 *** sebhub has joined #rtems 2012-05-11T01:52:26 good monring 2012-05-11T03:43:59 *** xiangfu has quit IRC (Ping timeout: 250 seconds) 2012-05-11T03:54:10 *** xiangfu has joined #rtems 2012-05-11T05:03:53 *** QingPei has joined #rtems 2012-05-11T06:42:02 *** Hesham has joined #rtems 2012-05-11T06:45:14 *** Hesham has quit IRC (Client Quit) 2012-05-11T07:10:32 *** Hesham has joined #rtems 2012-05-11T07:10:54 *** Hesham has left #rtems 2012-05-11T07:34:40 *** kristianpaul has quit IRC (Quit: leaving) 2012-05-11T07:35:23 *** kristianpaul has joined #rtems 2012-05-11T07:35:23 *** kristianpaul has joined #rtems 2012-05-11T08:05:59 *** xiangfu has quit IRC (Remote host closed the connection) 2012-05-11T08:10:28 *** sebhub has quit IRC (Ping timeout: 245 seconds) 2012-05-11T08:25:36 *** DrJoel has joined #rtems 2012-05-11T08:25:36 *** DrJoel has joined #rtems 2012-05-11T08:25:36 *** ChanServ sets mode: +o DrJoel 2012-05-11T08:44:00 * DrJoel is rebasing the "remove CVS Ids" branch 2012-05-11T08:44:09 and about to merge it... 2012-05-11T08:50:25 *** weiY has joined #rtems 2012-05-11T08:51:00 patch is merged 2012-05-11T08:53:36 *** gedare has joined #rtems 2012-05-11T08:57:56 drjoel: don't bother with git-send-email for very large patches. i'm not sure why your send-email is not working though in general. you should test it with some small patch and specify the "To" field as yourself until you get a reliable email sent... 2012-05-11T09:00:29 nice, the rtems-vc converted your large commit to an attachment :) 2012-05-11T09:01:42 I reconfigured it to use gmail so it should work now. That avoided authenticating through OAR and having to figure it out. gmail was well documented :) 2012-05-11T09:01:52 yeah 2012-05-11T09:01:53 I disabled the size limit for rtems-vc. :) 2012-05-11T09:02:18 ahhh 2012-05-11T09:05:48 *** gedare_ has joined #rtems 2012-05-11T09:05:59 *** gedare_ has quit IRC (Client Quit) 2012-05-11T09:06:49 verm__: is the github rtems working? 2012-05-11T09:06:57 or updater still broken? 2012-05-11T09:17:33 turning on warning flags without discussion or willingness to fix them is a bad thing :( 2012-05-11T09:22:53 ralf again? 2012-05-11T09:23:24 for awhile he was fixing some warning-induced issues. I wasn't really happy though the fixes seemed to be hacky. I should go back and look at those... 2012-05-11T09:35:44 The missing prototypes warnings are easy to fix but if turned on, their should have been a commitment to fix them 2012-05-11T09:47:51 *** gedare has quit IRC (Quit: Ex-Chat) 2012-05-11T10:03:52 * DrJoel remembers a friend who once told me to be careful when you poke a bear in the ass with a short stick... 2012-05-11T10:31:21 *** Hesham has joined #rtems 2012-05-11T10:58:46 *** arvind_khadri has quit IRC (Ping timeout: 248 seconds) 2012-05-11T11:10:08 *** weiY has quit IRC (Ping timeout: 240 seconds) 2012-05-11T11:48:14 *** gedare has joined #rtems 2012-05-11T11:48:44 hi hesham 2012-05-11T11:48:50 hi gedare 2012-05-11T11:49:11 bring me up to speed? 2012-05-11T11:49:25 i have to leave in awhile so now is a good time to cat 2012-05-11T11:49:26 chat 2012-05-11T11:49:49 it's ok if you are busy we can chat later 2012-05-11T11:50:11 i won't be around later today or much this weekend 2012-05-11T11:50:16 how is progress going? 2012-05-11T11:50:30 i have studied heap manager 2012-05-11T11:50:36 and chain manager 2012-05-11T11:50:57 heap manager has protection definitions but i could not grasp all of it 2012-05-11T11:51:13 i think that is to detect heap overruns 2012-05-11T11:51:44 It's not related to memory access ? 2012-05-11T11:52:47 also i have reviewed 2011 project report , it seems good so as i could see how heap or rbheap could make use of it and compare between them 2012-05-11T11:52:59 but i can not view the source code 2012-05-11T11:53:08 i think that's all for now 2012-05-11T11:53:09 great. note that i refactored the 2011 source code since that project report was written 2012-05-11T11:53:21 the source code should be available at the google code site 2012-05-11T11:53:34 i can not view it as i am not a member 2012-05-11T11:53:49 everyone can view the code i think... 2012-05-11T11:53:56 and check it out anonymously 2012-05-11T11:54:14 try a little harder :) the interface is a little clunky 2012-05-11T11:54:19 on-members may check out a read-only working copy anonymously over HTTP. 2012-05-11T11:54:32 isn't that sufficient? 2012-05-11T11:54:41 it's ok 2012-05-11T11:55:10 you'll import that code or adopt it into your own solution anyways 2012-05-11T11:55:23 don't need to be updating the 2011 code base anymore i think 2012-05-11T11:55:24 i see it now 2012-05-11T11:55:27 ok great 2012-05-11T11:56:20 what is on your plan now? 2012-05-11T11:56:20 i will study both ( this project ) and heap and make a report and maybe a design 2012-05-11T11:57:15 ok, remember that there is a difference between allocation and protection 2012-05-11T11:57:38 allocation make use of heap for example 2012-05-11T11:57:48 protection is what we will define or add 2012-05-11T11:57:54 both will form Arena 2012-05-11T11:58:09 yeah. but applying memory attributes to heap is hard 2012-05-11T11:58:46 i am thinking of making a 1:1 relation by domains and blocks 2012-05-11T11:59:17 just got the idea when i was reading the 2011 project report 2012-05-11T11:59:32 ok. we'll see what you come up with 2012-05-11T11:59:48 will you be available at Fri ? 2012-05-11T11:59:55 Tue sorry 2012-05-11T12:00:03 i think so yes. 2012-05-11T12:00:05 i'll email if not. 2012-05-11T12:00:24 it's ok , i will share any news with you 2012-05-11T12:01:50 sounds good. make sure you keep in mind the limitations of hw protection when designing your solution 2012-05-11T12:02:38 and what should be done if a given hw does not support pages or MMU 2012-05-11T12:02:46 I will keep that in mind 2012-05-11T12:03:01 super 2012-05-11T12:03:39 but i think it will take some time to study and review both 2011 project and the allocator 2012-05-11T12:04:31 yep 2012-05-11T12:04:36 but then you'll be up-to-speed 2012-05-11T12:06:09 i will try hard , but these days i have to do some reports and exams and interviews for the faculty so i'am scattered i think , but trying to manage my time 2012-05-11T12:06:30 ok well focus on getting through your school work. 2012-05-11T12:07:06 I will focus on both 2012-05-11T12:08:51 hopefully we can figure out something simple to do that you can code and submit 2012-05-11T12:09:02 the allocation work seemed promising 2012-05-11T12:09:34 the allocation part of the Arena ? 2012-05-11T12:09:55 combining region/partition/rbheap implementations in the score 2012-05-11T12:10:10 now i'm not so sure about that but keep thinking about that a bit too 2012-05-11T12:10:54 i think we could abandon object management by using heap/rbheap/chain 2012-05-11T12:11:10 what do you mean? 2012-05-11T12:11:39 i mean the score does not have to care about partition/region 2012-05-11T12:11:50 it cares about heap/rbheap/chain managers 2012-05-11T12:11:57 for example 2012-05-11T12:12:01 allocate 2012-05-11T12:12:05 may point to 2012-05-11T12:12:13 _heap_Allocate() 2012-05-11T12:12:25 _rbheap_Allocate() 2012-05-11T12:12:46 or chain 2012-05-11T12:13:04 rbheap does not deal with objects 2012-05-11T12:13:52 but we agreed to use just one allocator 2012-05-11T12:14:05 and the score will be a report 2012-05-11T12:14:10 right ? 2012-05-11T12:16:09 correct 2012-05-11T12:16:23 not sure what you mean by "the score will be a report" 2012-05-11T12:16:51 the last meeting we agreed that i will study allocator on RTEMS 2012-05-11T12:17:18 and how they could be combined in one score that deals with all of them 2012-05-11T12:17:30 and i said that will be like a report after i study them 2012-05-11T12:17:37 ah, ok 2012-05-11T12:18:50 i think it sounds right. 2012-05-11T12:18:58 after the design we will show it to the community if they agreed i will implement it 2012-05-11T12:19:14 yep 2012-05-11T12:19:54 for now i am studying the previous project and heap allocator 2012-05-11T12:20:51 ook 2012-05-11T12:26:57 gedare: i have to leave now 2012-05-11T12:27:49 see you later 2012-05-11T12:28:04 ok hesham. bye 2012-05-11T12:32:20 *** Hesham has quit IRC (Ping timeout: 260 seconds) 2012-05-11T12:40:44 *** gedare has quit IRC (Quit: Ex-Chat) 2012-05-11T13:07:44 *** antgreen has joined #rtems 2012-05-11T13:39:56 *** arvind_khadri has joined #rtems 2012-05-11T13:41:38 *** antgreen has quit IRC (Read error: Connection reset by peer) 2012-05-11T14:17:18 *** DrJoel has quit IRC (Quit: Leaving) 2012-05-11T15:08:29 *** arvind_khadri has quit IRC (Ping timeout: 245 seconds) 2012-05-11T20:04:45 *** xiangfu has joined #rtems 2012-05-11T20:32:19 *** weiY has joined #rtems 2012-05-11T21:51:24 *** QingPei1 has joined #rtems 2012-05-11T21:53:12 *** QingPei has quit IRC (Ping timeout: 252 seconds) 2012-05-11T22:39:48 *** dr__house` has joined #rtems 2012-05-11T22:39:48 *** dr__house` has joined #rtems 2012-05-11T22:39:55 *** dr__house` has left #rtems 2012-05-11T23:49:18 *** QingPei1 has quit IRC (Quit: Leaving.) 2012-05-11T23:56:15 *** QingPei has joined #rtems 2012-05-12T00:37:39 *** kiwichris has quit IRC (Ping timeout: 245 seconds) 2012-05-12T00:42:55 *** kiwichris has joined #rtems 2012-05-12T00:45:21 *** QingPei has left #rtems 2012-05-12T00:59:11 *** xiangfu has quit IRC (Ping timeout: 255 seconds) 2012-05-12T01:22:45 *** Hesham has joined #rtems 2012-05-12T01:56:21 *** Hesham has quit IRC (Ping timeout: 248 seconds) 2012-05-12T02:00:50 *** Hesham has joined #rtems 2012-05-12T02:01:18 *** Hesham has left #rtems 2012-05-12T02:57:58 *** arvind_khadri has joined #rtems 2012-05-12T03:39:10 *** arvind_khadri has quit IRC (Remote host closed the connection) 2012-05-12T03:41:18 *** arvind_khadri has joined #rtems 2012-05-12T03:41:18 *** arvind_khadri has joined #rtems 2012-05-12T03:57:08 *** weiY has quit IRC (Ping timeout: 240 seconds) 2012-05-12T04:19:48 *** xiangfu has joined #rtems 2012-05-12T04:24:08 *** xiangfu has quit IRC (Ping timeout: 240 seconds) 2012-05-12T04:33:00 *** weiY has joined #rtems 2012-05-12T04:36:13 *** xiangfu has joined #rtems 2012-05-12T05:44:41 *** weiY has quit IRC (Ping timeout: 244 seconds) 2012-05-12T05:46:31 *** weiY has joined #rtems 2012-05-12T06:40:57 *** weiY has quit IRC (Ping timeout: 252 seconds) 2012-05-12T06:44:34 *** arvind_khadri has quit IRC (Read error: Operation timed out) 2012-05-12T06:46:04 *** weiY has joined #rtems 2012-05-12T07:04:06 *** weiY has quit IRC (Ping timeout: 272 seconds) 2012-05-12T07:09:24 *** weiY has joined #rtems 2012-05-12T09:08:47 *** weiY has quit IRC (Ping timeout: 256 seconds) 2012-05-12T09:14:03 *** weiY has joined #rtems 2012-05-12T09:51:52 *** xiangfu has quit IRC (Remote host closed the connection) 2012-05-12T10:35:56 *** gedare has joined #rtems 2012-05-12T11:08:31 KILL AUTOTOOLS 2012-05-12T11:08:34 ;_) 2012-05-12T11:13:33 *** weiY has quit IRC () 2012-05-12T14:26:56 *** Hesham has joined #rtems 2012-05-12T14:39:06 *** Hesham has quit IRC (Quit: Leaving.) 2012-05-12T14:41:08 *** Hesham has joined #rtems 2012-05-12T14:58:16 *** Hesham1 has joined #rtems 2012-05-12T14:58:16 *** Hesham has quit IRC (Read error: Connection reset by peer) 2012-05-12T15:12:01 *** Hesham1 has quit IRC (Quit: Leaving.) 2012-05-12T15:28:06 *** gedare has quit IRC (Quit: Leaving) 2012-05-12T22:04:10 *** xiangfu has joined #rtems 2012-05-12T22:28:19 *** panzon has joined #rtems 2012-05-13T01:46:36 *** arvind_khadri has joined #rtems 2012-05-13T02:16:46 *** arvind_khadri has quit IRC (Ping timeout: 272 seconds) 2012-05-13T02:29:57 *** weiY has joined #rtems 2012-05-13T02:43:22 *** xiangfu has quit IRC (Ping timeout: 272 seconds) 2012-05-13T02:51:53 *** arvind_khadri has joined #rtems 2012-05-13T03:10:09 *** weiY has quit IRC (Ping timeout: 244 seconds) 2012-05-13T03:10:43 *** weiY has joined #rtems 2012-05-13T03:31:20 *** arvind_khadri has quit IRC (Ping timeout: 244 seconds) 2012-05-13T03:44:25 *** arvind_khadri has joined #rtems 2012-05-13T03:44:26 *** arvind_khadri has joined #rtems 2012-05-13T03:52:27 *** QingPei has joined #rtems 2012-05-13T05:10:03 *** arvind_khadri has quit IRC (Ping timeout: 245 seconds) 2012-05-13T05:10:17 *** xiangfu has joined #rtems 2012-05-13T05:14:40 *** xiangfu has quit IRC (Ping timeout: 244 seconds) 2012-05-13T05:33:51 *** arvind_khadri has joined #rtems 2012-05-13T05:33:51 *** arvind_khadri has joined #rtems 2012-05-13T06:47:14 *** weiY has quit IRC (Ping timeout: 255 seconds) 2012-05-13T06:56:46 *** weiY has joined #rtems 2012-05-13T07:28:42 *** xiangfu has joined #rtems 2012-05-13T07:30:16 *** weiY has quit IRC (Ping timeout: 272 seconds) 2012-05-13T07:48:38 *** kristianpaul has quit IRC (Ping timeout: 272 seconds) 2012-05-13T07:55:19 *** kristianpaul has joined #rtems 2012-05-13T08:01:05 *** kristianpaul has quit IRC (Ping timeout: 248 seconds) 2012-05-13T09:02:54 *** kristianpaul has joined #rtems 2012-05-13T09:05:57 *** gedare has joined #rtems 2012-05-13T09:25:20 *** xiangfu has quit IRC (Read error: Connection reset by peer) 2012-05-13T09:48:58 *** kristianpaul has quit IRC (Ping timeout: 272 seconds) 2012-05-13T09:49:20 *** kristianpaul has joined #rtems 2012-05-13T10:35:50 *** kristianpaul has quit IRC (Ping timeout: 272 seconds) 2012-05-13T10:36:03 *** kristianpaul has joined #rtems 2012-05-13T10:43:26 *** kristianpaul has quit IRC (Ping timeout: 272 seconds) 2012-05-13T10:43:29 *** kristian1aul has joined #rtems 2012-05-13T11:21:18 *** arvind_khadri has quit IRC (Ping timeout: 245 seconds) 2012-05-13T11:33:38 *** arvind_khadri has joined #rtems 2012-05-13T12:03:18 *** QingPei has left #rtems 2012-05-13T13:25:26 *** arvind_khadri has quit IRC (Ping timeout: 248 seconds) 2012-05-13T13:27:17 *** arvind_khadri has joined #rtems 2012-05-13T13:27:17 *** arvind_khadri has joined #rtems 2012-05-13T13:37:33 *** arvind_khadri has quit IRC (Ping timeout: 245 seconds) 2012-05-13T13:49:38 *** arvind_khadri has joined #rtems 2012-05-13T13:50:14 *** arvind_khadri has joined #rtems 2012-05-13T14:09:17 *** arvind_khadri has quit IRC (Ping timeout: 252 seconds) 2012-05-13T14:45:37 *** gedare has quit IRC (Quit: Leaving) 2012-05-13T16:42:35 *** kristian1aul has quit IRC (Quit: Reconnecting) 2012-05-13T16:42:44 *** kristianpaul has joined #rtems 2012-05-13T16:42:44 *** kristianpaul has joined #rtems 2012-05-13T16:52:04 *** kristian1aul has joined #rtems 2012-05-13T16:55:12 *** kristianpaul has quit IRC (Ping timeout: 272 seconds) 2012-05-13T17:40:41 *** kristian1aul has quit IRC (Quit: Reconnecting) 2012-05-13T17:40:48 *** kristianpaul has joined #rtems 2012-05-13T18:19:28 *** weiY has joined #rtems 2012-05-13T19:23:22 *** xiangfu has joined #rtems 2012-05-13T19:39:38 *** weiY has quit IRC (Ping timeout: 240 seconds) 2012-05-13T21:56:06 *** rtemsLogger has joined #rtems 2012-05-13T22:13:07 *** rtemsLogger has joined #rtems 2012-05-13T22:14:30 *** arvind_khadri has joined #rtems 2012-05-13T22:18:42 *** arvind_khadri has quit IRC (Read error: Operation timed out) 2012-05-13T23:25:57 *** xiangfu has quit IRC (Ping timeout: 248 seconds) 2012-05-13T23:31:06 *** xiangfu has joined #rtems