| *** pkill9 has quit IRC | 00:15 | |
| *** Pali has quit IRC | 01:00 | |
| *** atk has quit IRC | 02:00 | |
| *** atk has joined #maemo | 02:00 | |
| *** xorly has quit IRC | 02:28 | |
| *** Natch has quit IRC | 02:44 | |
| *** florian_kc has quit IRC | 02:46 | |
| *** Natch has joined #maemo | 02:47 | |
| *** pkill9 has joined #maemo | 03:04 | |
| *** pkill9 has quit IRC | 03:54 | |
| *** dafox has quit IRC | 04:10 | |
| *** Kabouik_ has quit IRC | 04:32 | |
| *** Maxdaman1us is now known as Maxdamantus | 07:17 | |
| *** FalconSpy has quit IRC | 07:43 | |
| *** FalconSpy has joined #maemo | 08:10 | |
| *** FalconSpy has joined #maemo | 08:10 | |
| *** spiiroin has quit IRC | 08:36 | |
| *** Oksanaa has quit IRC | 09:06 | |
| *** spiiroin has joined #maemo | 09:09 | |
| *** geaaru has quit IRC | 09:09 | |
| *** Marshall_Banana has joined #maemo | 09:13 | |
| *** herekun has quit IRC | 09:17 | |
| *** herekun has joined #maemo | 09:20 | |
| drathir | DocScrutinizer05: if i may ask n900 what type of simlock have? | 09:31 |
|---|---|---|
| *** xes__ is now known as xes | 09:40 | |
| *** Kabouik has joined #maemo | 09:44 | |
| bencoh | it usually comes sim-unlocked | 09:55 |
| KotCzarny | and since it has bb5 modem, all rules apply | 09:56 |
| *** Marshall_Banana has quit IRC | 10:02 | |
| *** florian has joined #maemo | 10:04 | |
| *** heroux has quit IRC | 10:22 | |
| *** heroux has joined #maemo | 10:38 | |
| *** Smily has joined #maemo | 10:41 | |
| *** Kabouik has quit IRC | 10:47 | |
| *** ketas has quit IRC | 11:00 | |
| *** ketas has joined #maemo | 11:04 | |
| *** jskarvad has joined #maemo | 11:08 | |
| *** jskarvad has quit IRC | 11:08 | |
| *** jskarvad has joined #maemo | 11:08 | |
| *** qwazix has quit IRC | 11:20 | |
| *** qwazix has joined #maemo | 11:21 | |
| *** Kilroo has quit IRC | 11:25 | |
| *** eijk_ has joined #maemo | 11:39 | |
| *** Kilroo has joined #maemo | 11:40 | |
| *** cyteen has quit IRC | 11:44 | |
| *** cyteen has joined #maemo | 11:48 | |
| *** cyteen has quit IRC | 12:01 | |
| *** cyteen has joined #maemo | 12:14 | |
| *** cyteen has quit IRC | 12:19 | |
| *** eMHa has joined #maemo | 12:28 | |
| *** pigeon has quit IRC | 12:34 | |
| *** pigeon has joined #maemo | 12:36 | |
| *** cyteen has joined #maemo | 12:44 | |
| *** opt1mus has quit IRC | 12:45 | |
| *** xorly has joined #maemo | 12:53 | |
| *** eMHa has quit IRC | 13:00 | |
| *** Wikiwide has joined #maemo | 13:21 | |
| Wikiwide | Why does my N900 have camera's videos mixed with all other films in Media Player, and not placed into category of "Recorded by device camera"? | 13:22 |
| *** xorly has quit IRC | 13:37 | |
| *** eMHa has joined #maemo | 13:39 | |
| *** freemangordon has quit IRC | 13:57 | |
| *** Wikiwide has left #maemo | 14:00 | |
| *** pigeon has quit IRC | 14:49 | |
| *** pigeon has joined #maemo | 14:52 | |
| *** eijk has joined #maemo | 14:52 | |
| *** eijk_ has quit IRC | 14:53 | |
| *** pkill9 has joined #maemo | 15:03 | |
| *** dafox has joined #maemo | 15:19 | |
| DocScrutinizer05 | I think because it doesn't filter subgroups out form "all" group | 15:31 |
| DocScrutinizer05 | beauty of fracker err tracker | 15:31 |
| DocScrutinizer05 | the tracker concept is broken by design, in my book | 15:32 |
| *** KotCzarny has quit IRC | 15:38 | |
| *** geaaru has joined #maemo | 15:46 | |
| *** mkz has joined #maemo | 15:53 | |
| mkz | is it ok if I mirror http://repository.maemo.org/ ? | 15:54 |
| mkz | that is, recursive wget or rsync | 15:54 |
| mkz | don't want to be that guy which hogs bandwidth without permission | 15:55 |
| bencoh | if anything, you should probably use apt-mirror | 15:56 |
| bencoh | (or something close to it) | 15:57 |
| bencoh | first sync will take ages, but then on it should be fine | 15:58 |
| DocScrutinizer05 | generally it's ok to mirror repos and whatever you like | 16:01 |
| DocScrutinizer05 | but listen to advice regarding best way to do so | 16:01 |
| *** spiiroin has quit IRC | 16:02 | |
| *** ketas- has joined #maemo | 16:13 | |
| *** spiiroin has joined #maemo | 16:16 | |
| Wizzup | new cssu in cssu testing it seems, cool. can't upgrade to it due some conflict with curl though (on cssu thumb) | 16:18 |
| enyc | Wizzup: hrrm somebody else whad some crazy dependency mess too | 16:54 |
| enyc | 15:23 * DocScrutinizer05 feels massive headache http://paste.opensuse.org/86359988 - prolly getting too old for such shit | 16:54 |
| enyc | ok maye that wasn't the rpoblem | 16:55 |
| *** pkill9 has quit IRC | 17:01 | |
| Wizzup | enyc: I think there needs to be a new curl build for thumb | 17:04 |
| Wizzup | so it might just take a bit more time | 17:04 |
| *** pkill9 has joined #maemo | 17:14 | |
| *** xorly has joined #maemo | 17:22 | |
| *** vectis3 has quit IRC | 17:34 | |
| *** vectis3 has joined #maemo | 17:34 | |
| *** auenf has quit IRC | 17:42 | |
| *** auenf has joined #maemo | 17:45 | |
| *** Kabouik has joined #maemo | 17:48 | |
| *** KotCzarny has joined #maemo | 17:53 | |
| *** Vajb has quit IRC | 18:33 | |
| *** Vajb has joined #maemo | 18:35 | |
| *** Pali has joined #maemo | 18:36 | |
| *** herekun has quit IRC | 18:59 | |
| *** herekun has joined #maemo | 19:04 | |
| *** ketar has joined #maemo | 19:08 | |
| *** ketar has quit IRC | 19:10 | |
| *** ketar has joined #maemo | 19:10 | |
| *** ketar has quit IRC | 19:11 | |
| *** herekun has quit IRC | 19:12 | |
| *** pagurus has quit IRC | 19:14 | |
| *** pagurus has joined #maemo | 19:15 | |
| *** herekun has joined #maemo | 19:32 | |
| *** pkill9 has quit IRC | 19:37 | |
| *** pkill9 has joined #maemo | 19:54 | |
| *** eMHa has quit IRC | 20:25 | |
| *** florian has quit IRC | 20:42 | |
| *** Natch has quit IRC | 21:01 | |
| *** pigeon has quit IRC | 21:11 | |
| *** pigeon has joined #maemo | 21:13 | |
| *** pkill9 has quit IRC | 21:20 | |
| *** freemangordon has joined #maemo | 21:22 | |
| *** eMHa has joined #maemo | 21:28 | |
| *** Natch has joined #maemo | 21:43 | |
| *** jonwil has joined #maemo | 21:48 | |
| *** jskarvad has quit IRC | 21:53 | |
| *** florian has joined #maemo | 22:05 | |
| *** Kabouik_ has joined #maemo | 22:17 | |
| *** Kabouik has quit IRC | 22:20 | |
| *** jonwil has quit IRC | 22:28 | |
| *** eijk has quit IRC | 22:43 | |
| *** Oksana has joined #maemo | 22:49 | |
| CatButts | hey DocScrutinizer05 | 23:02 |
| CatButts | you mentioned softfloat is not 'that' much slower than hardfloat | 23:03 |
| CatButts | how would it hold against something like DSP | 23:03 |
| CatButts | say, a multichannel softsynth | 23:04 |
| DocScrutinizer05 | CatButts: softFP uses whatever available, DSP or arithmetic coprocessor or even software implementation of float math | 23:04 |
| DocScrutinizer05 | softfp is hardfp wrapped into a library call ABI | 23:05 |
| DocScrutinizer05 | if the platform has a hw accel for FP | 23:05 |
| DocScrutinizer05 | so your question doesn't make much sense to me | 23:06 |
| CatButts | so it still ends up in hardware, but there is a capability check at runtime? | 23:06 |
| DocScrutinizer05 | not at runtime, at compile time | 23:06 |
| DocScrutinizer05 | at least for maemo, that's not done at runtime afaik | 23:07 |
| CatButts | so on the n900, I still benefit from hardware floats? | 23:08 |
| DocScrutinizer05 | softfp is a library, the implementation of that library may differ with target platform, the API stays the same | 23:08 |
| CatButts | ah, I see meow | 23:08 |
| DocScrutinizer05 | yes | 23:08 |
| DocScrutinizer05 | hardfp directly calls the float hw from your code, without any library in between | 23:09 |
| DocScrutinizer05 | (simplified picture) | 23:09 |
| CatButts | I initially thought that softfp meant software emulated float calculations | 23:09 |
| CatButts | strictly | 23:10 |
| DocScrutinizer05 | the overhead of softfp over hardfp is basically the library wrapper code, for maemo | 23:10 |
| CatButts | not a lib call | 23:10 |
| DocScrutinizer05 | afaik | 23:10 |
| CatButts | now it's clear in my head | 23:10 |
| DocScrutinizer05 | or let me put it that way: your maemo osftfp library is built using hardfp code | 23:11 |
| DocScrutinizer05 | softfp* | 23:11 |
| freemangordon | guys, it is not like that | 23:12 |
| freemangordon | it is about the calling convention | 23:13 |
| freemangordon | softfp means float args are passed through int registers | 23:13 |
| freemangordon | hardfp means float args are passed through FP registers | 23:13 |
| DocScrutinizer05 | yes, you call a copoc directly with parameters in coproc registers, for hardfp. You call a library with parameters on stack or whatever, for softfp | 23:13 |
| freemangordon | no, you call a library (or a local function) both ways | 23:14 |
| DocScrutinizer05 | https://wiki.debian.org/ArmHardFloatPort/VfpComparison | 23:15 |
| freemangordon | just that function parameters are put in CPU for softfp or FPU for hardfp registers | 23:15 |
| freemangordon | with softfp you can emulate FP in software | 23:15 |
| freemangordon | but you can use HW FPU if it is available as well | 23:16 |
| DocScrutinizer05 | "soft" vs "softfp" vs "hard" | 23:16 |
| freemangordon | DocScrutinizer05: remember how was 8087 emulated? | 23:16 |
| freemangordon | back then | 23:16 |
| DocScrutinizer05 | wasn't that softirq | 23:17 |
| DocScrutinizer05 | software interrupt | 23:17 |
| freemangordon | no, it was generatin "illegal instruction" interrupt, whatever it was called :) | 23:18 |
| freemangordon | that irq was handled by FPU emulation lib | 23:18 |
| DocScrutinizer05 | hmm, pretty much a sowtware interrupt | 23:18 |
| DocScrutinizer05 | an instruction causing an interupt | 23:19 |
| freemangordon | however, soft(fp) vs hardfp in terms of cortex a8 is just about where FP parameters are passed | 23:20 |
| DocScrutinizer05 | :-) | 23:20 |
| DocScrutinizer05 | so not as huge a difference as the uneduicated user might expect | 23:20 |
| freemangordon | well, it was a prefix at a start of every FPU instruction, 0x0F or somesuch | 23:20 |
| freemangordon | there is a huge diff for FPU intense function calls | 23:21 |
| freemangordon | some 20% iirc | 23:21 |
| freemangordon | but no difference for most of the system | 23:21 |
| DocScrutinizer05 | yeah, probably you need 30 instead 8 instructions for a complete FP op | 23:21 |
| DocScrutinizer05 | no 1000 instructions though | 23:22 |
| DocScrutinizer05 | like you'd need for software emulation | 23:22 |
| freemangordon | it is not the instructions themselves, but the memory access | 23:22 |
| DocScrutinizer05 | :nod: | 23:22 |
| freemangordon | I am not sure you can copy from integer reg to FPU reg directly, without going through memory, however, even if it is possible, it takes time | 23:24 |
| freemangordon | and leads to pipeline stalls | 23:24 |
| DocScrutinizer05 | I'd give ballpark estimations of "soft":1000; "softfp":1000 or 30 (depending if hw accel available);"hard" (only when available): 8 | 23:25 |
| freemangordon | CatButts: the last ~20 lines is a kind of an answer to your question | 23:25 |
| DocScrutinizer05 | whatever time unit | 23:25 |
| DocScrutinizer05 | according to you it's rather "soft":1000; "softfp":1000 or 12 (depending if hw accel available);"hard" (only when available): 10 | 23:26 |
| freemangordon | DocScrutinizer05: well, keep in mind that for bug FP functions penalty is relatively small comapred to function execution time | 23:26 |
| freemangordon | *big | 23:26 |
| CatButts | so it's a tad bit slowe, but no tragedy | 23:26 |
| CatButts | r | 23:26 |
| freemangordon | yes | 23:26 |
| CatButts | even with intensive float calculations? | 23:27 |
| freemangordon | float calculations themselves are not affected | 23:27 |
| DocScrutinizer05 | that's *only* looking at the float instructions | 23:27 |
| freemangordon | just the function calls | 23:27 |
| DocScrutinizer05 | the rest of the system doesn't care what your float does | 23:28 |
| DocScrutinizer05 | a program that doesn't use floats will not run any fatser on a hardfp system, | 23:28 |
| freemangordon | correct | 23:28 |
| DocScrutinizer05 | CatButts: so yes, *only* "with intensive float calculations" (100% intensity) | 23:29 |
| * DocScrutinizer05 idly wonders if you could wrap the softfp functions with an IRQ-lock and save all FP registers and restore them before leavinbg the softfp function, so hardfp and softfp code could coexist | 23:32 | |
| DocScrutinizer05 | sucks big time, but would make coexistence possible | 23:33 |
| * CatButts plays with cubic interpolation and manages to turn it into ear-raping noise | 23:33 | |
| DocScrutinizer05 | wait, wouldn't an IRQ either not touch FP registers at all, or (in case it's the schedukler) save the FP registers as well and restore them for the newly scheduled process? | 23:34 |
| DocScrutinizer05 | so except for shared libs, is there *anything* that comflicts between softfp-processes and hard(fp)-processes? | 23:35 |
| DocScrutinizer05 | and even for shared libs, would the ABI change for soft/hard? | 23:37 |
| *** thuttu77 has quit IRC | 23:37 | |
| DocScrutinizer05 | freemangordon: ^^^ | 23:37 |
| *** eijk has joined #maemo | 23:38 | |
| freemangordon | there is nothing that conflicts | 23:38 |
| DocScrutinizer05 | HAHAH >>FPU instructions could be emulated by the kernel so that FPU-less systems could run with this ABI but as far as we know this does not exist<< | 23:38 |
| freemangordon | not exactly true iirc, there are "some" instructions emulated in the kernel | 23:39 |
| DocScrutinizer05 | ooh, this direction of "not true" :-D | 23:39 |
| DocScrutinizer05 | oooh, "pipeline stalls" | 23:40 |
| DocScrutinizer05 | hmmm | 23:40 |
| freemangordon | iirc, some vfpv3 instructions are emulated if hw is vfpv2 | 23:40 |
| freemangordon | but don;t quote me on that one, it was a while I looked at kernel FP support code | 23:41 |
| *** pkill9 has joined #maemo | 23:41 | |
| DocScrutinizer05 | CatButts: (at runtime) I stand corrected: >>The GCC -mfloat-abi=softfp flag allows use of VFP while remaining compatible with soft-float code. This allows selection of appropriate routines at runtime based on the availability of VFP hardware.<< | 23:50 |
| DocScrutinizer05 | >>The runtime linker, ld.so, supports a mechanism for selecting runtime libraries based on features reported by the kernel. For instance, it's possible to provide two versions of libm, one in /lib and another one in /lib/vfp, and ld.so will select the /lib/vfp one on systems with VFP.<< | 23:51 |
| DocScrutinizer05 | BUT... >>...the ARM EABI and ELF specification has no way to tell which of soft, softfp, hard is used to build it<< | 23:53 |
| DocScrutinizer05 | HMMPH >>This needs to be looked into.. is it only a part of the version 5 EABI and not implemented? Could there possibly be scope for a GNU EABI extension for this, or a fix for the linker?<< | 23:56 |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!