| --- Log opened Sun Jan 11 00:00:50 2015 | ||
| ysangkok | poke53282: there were some permissions errors in the new basefs.json, i fixed them myself and committed | 01:41 |
|---|---|---|
| olofk | poke53282: Do I need to build a separate binutils for each libc? | 08:43 |
| poke53282 | olofk: I guess not. There might be configure options, which one libc needs and the other not. | 09:41 |
| poke53282 | MAy I ask you what you are trying to do? | 09:41 |
| poke53282 | Another virtual machine image for the website? | 09:41 |
| poke53282 | ysangkok: I can't remember, but I guess, that I took the permissions from my archlinux distribution. | 09:43 |
| poke53282 | Is there a tool, which checks for the correct filesystem permissions? | 09:43 |
| stekern | hmm, seems like the lto problem is that it exchange a l.lbs from a var that is 0xff with a l.addi with the constant -1 | 09:47 |
| stekern | which of course isn't the same | 09:47 |
| olofk | poke53282: I'm partly trying to understand how these things fit together and partly just trying to install stuff on a fresh machine to update the build instructions if needed | 09:48 |
| stekern | the lbs will be 0x000000ff and the l.addi 0xffffffff | 09:48 |
| poke53282 | Stekern: wow, I didn't know, that lto is kind of doing this. | 09:50 |
| stekern | yeah, me neither, I've been quite ignorant to what lto does until now | 09:52 |
| poke53282 | but wait. | 09:55 |
| poke53282 | r[((ins >> 19) & 0x7C)>>2] = ((ReadMemory8(paddr|0)|0) << 24) >> 24; | 09:55 |
| poke53282 | This is my code. | 09:55 |
| poke53282 | and loading a -1 would be 0xffffffff | 09:55 |
| stekern | sorry, typo | 09:55 |
| stekern | l.lbz | 09:56 |
| stekern | otherwise everything would be in order, as you say | 09:56 |
| poke53282 | Yes | 09:56 |
| poke53282 | So I would look for a wrong signed or unsigned char. | 09:57 |
| poke53282 | Or update to 4.9.2 ;) | 09:57 |
| stekern | l.lbz r6,0x0(r5) r6 = 000000ff | 09:57 |
| stekern | l.addi r6,r0,-1 r6 = ffffffff | 09:57 |
| stekern | from or1ksim dumps | 09:58 |
| poke53282 | yes, I understand | 09:58 |
| poke53282 | and l.lbz is the correct one. | 09:58 |
| stekern | yes | 09:58 |
| poke53282 | the result is still correct if the register is treated as 8 Bit in the compiler. Then it does not matter. | 09:59 |
| poke53282 | So, this optimization strategy might still be valid. | 10:00 |
| stekern | yes, and this might be the assumption that is done | 10:01 |
| stekern | but it's then used in blueCmd's atomic stuff that shifts and ands around stuff | 10:01 |
| stekern | and that's probably opaque to the lto logic | 10:02 |
| poke53282 | Ok, then just two assumptions interfere here The one of blueCmd and the one of gcc. | 10:02 |
| stekern | so, I would be inclined to try to find a way to tell that no assumptions can be made on the ingoing arguments to blueCmd's stuff, if that's possible | 10:03 |
| poke53282 | sounds good | 10:05 |
| stekern | time to dig myself out from the house, there was obviously some sort of snowstorm here tonight | 10:10 |
| poke53282 | And we had yesterday 22°C. The hottest day in winter since the start time for recording | 10:14 |
| poke53282 | stekern: How do you update your Linux kernel with the official rep? Which git command? | 11:40 |
| poke53282 | git pull https://github.com/torvalds/linux.git gives me conflicts | 11:41 |
| stekern | like that and then resolve the conflicts | 12:09 |
| stekern | they are usually straight forward | 12:09 |
| poke53282 | Ok, done | 12:25 |
| ysangkok | poke53282: you must be back in canada? it's not 22 degrees in the Germany i'm in :P | 12:43 |
| poke53282 | Nope, Germany. I heard it in the radio. | 12:50 |
| stekern | maybe the right way to handle this is actually to treat the incoming argument as if the other parts of it than the ones that are to be used can contain whatever garbage | 13:05 |
| stekern | at least the failing test is fixed by doing so... | 13:07 |
| poke53282 | Sounds reasonable. | 13:21 |
| ysangkok | poke53282: ah ok :P | 13:26 |
| ysangkok | poke53282: it seems buildfs is missing some packages that are shipped in the sysroot | 13:26 |
| ysangkok | poke53282: is this intensional? | 13:27 |
| poke53282 | no, it is not updated yet. | 13:29 |
| ysangkok | poke53282: you know wget can follow links and read the directory listing itself? | 13:29 |
| stekern | blueCmd: are you up to context switching into some gcc atomic stuff? | 13:30 |
| poke53282 | well, the listing has also a lot, which is not included. | 13:30 |
| ysangkok | ah ok | 13:30 |
| stekern | blueCmd: http://pastie.org/9825355 | 13:31 |
| stekern | that's the patch I have, it's in "cmpxchg_mask" | 13:32 |
| poke53282 | ysangkok: Wait one second. I update buildfs | 13:32 |
| stekern | and the idea is that you don't make assumptions about the %4 register, but rather mask out the interesting bits | 13:33 |
| stekern | I just re-ran all the atomic tests and they all pass now | 13:34 |
| stekern | s/interesting/uninteresting | 13:36 |
| poke53282 | ysangkok: updated. The filesystem which is built, should be identical with the one in jor1k-sysroot | 13:36 |
| poke53282 | at least I hope so. | 13:37 |
| ysangkok | great, thanks | 13:39 |
| poke53282 | forgot two packages :) | 13:47 |
| poke53282 | ysangkok: finally Identical except of the src folder. | 13:54 |
| poke53282 | I had to update jor1k-sysroot json file too. | 13:54 |
| blueCmd | stekern: wsup? | 17:03 |
| stekern | blueCmd: read the log? ;) | 17:05 |
| blueCmd | yeah, i was unable to scroll | 17:06 |
| blueCmd | got myself a new keyboard now so here goes | 17:06 |
| stekern | I have done a minor change to "cmpxchg_mask", and wanted get some input on that | 17:06 |
| stekern | I can still see all I said in my irssi window, how small screen do you have? | 17:08 |
| blueCmd | hm, still cannt scroll - oh well | 17:08 |
| blueCmd | 15 lines or so | 17:08 |
| stekern | http://pastie.org/9825355 | 17:08 |
| blueCmd | yes, i see the hilighta | 17:08 |
| blueCmd | but nothing more :p | 17:08 |
| blueCmd | sorry | 17:08 |
| blueCmd | but adding another mask seems harmless enough | 17:10 |
| blueCmd | so if it helps / fixes stuff - i have nothing against it. is it for input sanitation? | 17:10 |
| stekern | the point is that it seems to be so, that you can't assume that the data in %4 is loaded as a 8-bit value | 17:10 |
| stekern | and where the testfailure occurs, lto change a l.lbz (that will load 0x000000ff) to a l.addi -1 | 17:12 |
| blueCmd | makes sense, probably possible to constrain it | 17:12 |
| blueCmd | ah hah, thats a bit naughty of it. | 17:12 |
| stekern | that was my first thought, but on the other hand, that lto optimization indicates that you should only *pick out* the 8-bit value from the incoming data | 17:13 |
| blueCmd | this should be outside the critical section anyway, so as long as 4 is marked as clobbered (maybe even early clobbered) we should be fine | 17:13 |
| blueCmd | this might be a problem with the other atomics as well, you might want to check that | 17:15 |
| stekern | yeah, that's the only thing I'm worried about, and that should probably be added | 17:15 |
| stekern | you mean the fetch_and_<op_name>_mask? | 17:16 |
| blueCmd | thanks for debugging this. really awesome | 17:16 |
| blueCmd | yeah, i dont recall how many _mask functions there are | 17:17 |
| stekern | I think that's the only one except the one I already changed | 17:17 |
| stekern | but you're right, looks like that needs the same treatment | 17:17 |
| poke53282 | ysangkok: Did you try your pull request with the video files? | 17:27 |
| ysangkok | poke53282: i tested it with a 55mb squashfs that i also mounted | 17:28 |
| ysangkok | using overlayfs :D | 17:28 |
| ysangkok | poke53282: is there a problem? | 17:28 |
| ysangkok | it only needed 4 chunks to mount :D | 17:29 |
| ysangkok | i didn't experiment with the chunk size... but maybe that should be configurable too.. if we add support for giving random json fs descriptions in the query string, people can make more use of it since they know what chunk sizes work best with their data | 17:30 |
| poke53282 | ysangkok: For some reason, he don't activate the lazy file mode for one file. | 17:38 |
| poke53282 | I added "lazy":1 at the orion mpg file | 17:39 |
| poke53282 | It works | 17:41 |
| ysangkok | ah good | 17:41 |
| ysangkok | usually i have problems cause it caches the metadata also | 17:41 |
| poke53282 | well, I have a button, which clears the cache | 17:41 |
| ysangkok | so i open the developer console (it has "disable cache" checked) while loading the basefs | 17:41 |
| ysangkok | poke53282: you're saying it works, but also that it doesn't work for one file... should i test again or not? | 17:42 |
| poke53282 | Now it works. My first test failed. Don't know why. | 17:43 |
| poke53282 | Doesn't matter | 17:43 |
| ysangkok | hmm ok | 17:43 |
| poke53282 | You can also use the ata device for the squash filesystem. | 17:51 |
| ysangkok | how ? | 17:52 |
| poke53282 | I used js/worker/dev/ata.js | 17:53 |
| poke53282 | The function SetBuffer | 17:53 |
| poke53282 | The ata device can also work with the whole buffer. | 17:54 |
| poke53282 | But I had a version, whcih worked with small junks on the server. | 17:54 |
| poke53282 | I never used it, because it was slow. | 17:55 |
| poke53282 | This was before I started the virtio/9p filesystem. | 17:55 |
| ysangkok | hmm ok | 17:55 |
| poke53282 | In principle the Lazyuintarray can be used as well for this device. | 17:56 |
| mor1kx | [mor1kx] bandvig pushed 1 new commit to withfpu: https://github.com/openrisc/mor1kx/commit/8b646f0f0fd3c5ce4d48731f85a6d261c5cae9c3 | 18:03 |
| mor1kx | mor1kx/withfpu 8b646f0 Andrey Bacherov: Size optimization by replace 64-bit adder to 48-bit one in 2nd stage of multiplier | 18:03 |
| ysangkok | well there aren't many disadvantages to the lazyuint8array i guess... cause even if access behaviour is random, it will always calculate the same chunk boundaries. so the chunks can be cached | 18:04 |
| poke53282 | no disadvantages. I just wanted to tell you, that there is an alternative if you just want to mount one big file. | 18:10 |
| olofk | stekern: Just finished Book of Unwritten tales today. You got to check it out. Best adventure game I've played in ages | 20:07 |
| LoneTech | olofk: on to the critter chronicles then? | 20:17 |
| olofk | LoneTech: Yes! I got both as part of a humble indie bundle | 20:18 |
| olofk | Seems like Book of Unwritten tales 2 is coming along as well. There is an alpha version on steam | 20:18 |
| LoneTech | yay (: | 20:18 |
| stekern | olofk: you really don't want me to work on these gcc test fails, do you? | 21:31 |
| olofk | stekern: Book of unwritten tales is your reward for a clean regression test suite | 21:32 |
| stekern | you're buying then too? ;) | 21:33 |
| stekern | I've got the back to the future game to play too | 21:33 |
| stekern | I can play that on highest settings on my laptop, something I can't on my ws that has a crappy graphics card | 21:35 |
| olofk | stekern: You're happy with the zenbook? | 21:40 |
| stekern | yes, the only bad thing I can say about it is the lack of dedicated pgup/pgdn/home/end | 21:45 |
| stekern | I've got multi-touch gestures working with touchegg too, hint if you haven't tried. | 21:48 |
| stekern | on the touch-pad | 21:48 |
| stekern | 115 gcc errors now, was 400+ the first run | 21:49 |
| olofk | stekern: Interesting. I remember seeing touchegg but didn't really understand how it fit into the rest of the system | 21:54 |
| stekern | you just run it and define your gestures, I'm using the defaults | 22:06 |
| stekern | you have to turn off the builtin synaptic gestures (like two finger scroll) | 22:07 |
| olofk | Ahh... maybe it was a bit work to build for gentoo, so I gave up | 22:07 |
| stekern | I especially like the three-finger move window gesture | 22:08 |
| olofk | Does it work with all window managers? | 22:12 |
| stekern | it should, but I recall reading that gnome have some own that might clash with it | 22:13 |
| stekern | but the only thing that happens then is that you just can't overrule them | 22:13 |
| stekern | or was it unity...? | 22:14 |
| olofk | xfce doesn't have anything like that built-in so that shouldn't be a problem | 22:14 |
| olofk | for me | 22:14 |
| stekern | I'm running plasma 5 on this | 22:14 |
| --- Log closed Mon Jan 12 00:00:52 2015 | ||
Generated by irclog2html.py 2.15.2 by Marius Gedminas - find it at mg.pov.lt!