IRC log of #maemo for Saturday, 2016-11-05

*** N-Mi has quit IRC00:13
*** Cor-Ai has quit IRC00:20
*** Cor-Ai has joined #maemo00:22
*** enyc has quit IRC00:30
*** jonwil has joined #maemo00:42
*** troulouliou_div2 has joined #maemo00:50
jonwilhi01:04
CatButtsuhllo01:05
DocScrutinizer05hi jonwil01:07
jonwilStill no closer to figuring out how I am going to read/parse the GTFS data in the transit app I plan to write. I found a tool that converts GTFS data into a SQLite database but the tool requires libzip2 and libcsv3 neither of which are in the repos and I dont really know how to go about porting them to Maemo01:27
jonwilAnd then I would also need to come up with packaging bits for gtfs2db as well to get that into the repos01:27
jonwilMaybe I need to forget all this and write my own GTFS parser01:27
jonwilI am sure there is something in the repos that can read zip files01:28
jonwiland comma delimited files are easy enough to parse...01:28
jonwilor maybe I should just go ahead and get libzip, libcsv and gtfs2db into the repo :)01:29
*** dafox has joined #maemo01:36
*** troulouliou_div2 has quit IRC01:43
*** florian has quit IRC02:03
*** dafox has quit IRC02:17
jonwilhttp://talk.maemo.org/showthread.php?p=1518055#post151805502:32
*** cyphase has quit IRC02:43
*** cyphase has joined #maemo02:48
*** disco_stu has quit IRC02:58
*** robbiethe1st has joined #maemo03:23
*** krnlyng has quit IRC03:24
*** disco_stu has joined #maemo03:37
*** krnlyng has joined #maemo03:37
*** jonwil has quit IRC03:42
*** bruce_lee has quit IRC03:53
*** utanapischti has quit IRC03:59
*** drawkula has joined #maemo04:00
*** Maxdamantus has quit IRC05:54
*** lxp1 has joined #maemo06:01
*** lxp has quit IRC06:04
*** Maxdamantus has joined #maemo06:19
*** DocScrutinizer05 has quit IRC07:24
*** DocScrutinizer05 has joined #maemo07:24
*** sunshavi has quit IRC07:34
*** chainsawbike has quit IRC07:57
*** chainsawbike has joined #maemo07:59
*** gregoa has quit IRC09:20
*** gregoa has joined #maemo09:20
*** jonwil has joined #maemo09:40
KotCzarnyjonwil: zziplib09:55
KotCzarnyit allows you to tread zip archives as simple dirs/files09:55
jonwilwhat about zziplib?09:55
jonwilI know what it is, why are you mentioning it?09:56
KotCzarny00:28 #maemo jonwil   > I am sure there is something in the repos that can read zip files09:56
jonwiloh yeah09:56
KotCzarnyim using it in oscp to handle zip files09:57
KotCzarnyso it works nicely09:57
jonwilTurns out the best way to go for what I want is probably going to be to get gtfs2db and its dependencies into the repos and use that for dealing with the GTFS data09:57
jonwilPull down the GTFS zip file from the transit agency, run gtfs2db on it then use the resulting sqlite DB in my app09:58
jonwilQuerying a sqlite db should be easy enough especially since I probably wont need any really fancy SQL :)09:58
jonwilI just dont have the skills to do Debian packaging unfortunately :(09:59
KotCzarnyhehe09:59
KotCzarnyyeah, that part should be easy, but isnt so obvious09:59
KotCzarnyyou can always include those two libs in your package and compile them in statically10:00
KotCzarnyand as i've never heard about any of them, no great loss either10:01
brolin_empeyjonwil: Is SQLite even capable of “really fancy SQL”? ;-)10:18
jonwildont know :)10:18
siceloit has some shortcomings, but for most use cases, it's as good as any regular SQL implementation10:21
brolin_empeyjonwil: That was kind of a rhetorical question.  SQL database servers such as PostgreSQL or MySQL/MariaDB have far more capabilities/features than SQLite.10:22
jonwilyeah :)10:22
KotCzarnysooo, run psql on n900?10:26
KotCzarny;)10:26
* brolin_empey retired his N900 in 2014.10:28
KotCzarnychicken10:29
* CatButts patiently waits for his celery10:33
brolin_empeyKotCzarny: I am a chicken because I dislike the bad user experience I had with an N900 in Canada?10:39
KotCzarnyphone wise or smartphone wise?10:39
brolin_empeyKotCzarny: Both.  Many persons who have cellular service in Canada cannot even use the N900 as a literal telephone because the cellular modem used by the N900 cannot use the cellular networks of at least two (Telus Mobility, Bell Mobility, and all of their MVNOs) of the largest cellular network operators in Canada.  I have already discussed this subject in this channel months or years ago.10:48
KotCzarnyuhum10:48
KotCzarnypity10:49
sicelowhat replaced it (N900)?10:52
*** florian has joined #maemo10:59
brolin_empeysicelo: First a Geeksphone Revolution running the Android OS 4.2 on x86 + PCI but that delivered a bad user experience too so I finally bought a mainstream smartphone running the Android OS on ARM, a Samsung Galaxy Note 3 (variant for Canada) but I had to replace the stock OS with CyanogenMod because Samsung actively tries to restrict the users of their non-Nexus models.11:00
*** chainsawbike has quit IRC11:03
sicelocool :)11:04
KotCzarnynot so cool because of corporate greed ;)11:04
sicelomy dad gave me his Galaxy S4 ... i must say it packs some really nice hardware .. but .. it tends to feel like a ... toy :-/11:05
siceloso my N900 is still in constant use11:05
brolin_empeyKotCzarny: The Galaxy Note 7 was reportedly extremely literally uncool. ;-)11:06
*** chainsawbike has joined #maemo11:07
KotCzarny:)11:07
*** Vajb has quit IRC11:12
*** florian has quit IRC11:16
brolin_empeyMaybe I should run the Android OS on x86 + PCI on my mobile computer and run Windows NT (Windows 10) on ARM on my desktop/stationary computer to have the opposite of the usual combination. ;-)11:19
brolin_empeyMy High-density 5.25-inch flexible disc (mini floppy disc) drive still works with Windows 10 for x86-64 but I still use an LGA775/Socket T(Core 2 era) motherboard that still has legacy peripheral support such as headers on the motherboard for a flexible disc drive and PATA drives.  Current desktop motherboards for a socketed Core i? CPU seem to no longer have headers for a flexible disc drive nor PATA even though the motherboard may still use a Super I/O11:31
brolin_empeycontroller IC that includes a flexible disc drive controller function.  I can buy a PCI or maybe PCI Express card with a PATA host controller (and a PCI sound card with a game port) but I have not seen a PCI card with a flexible disc drive controller.11:31
KotCzarnyi bet there are usb adapters11:32
KotCzarnywhich would make it work a lot longer11:32
KotCzarnyi have usb-fdd and it works fine11:33
KotCzarnythough its 3.5"11:33
KotCzarnyhttp://www.deviceside.com/fc5025.html11:34
KotCzarnyyup11:34
* sicelo still uses same era motherboard :)11:34
KotCzarnyuh, readonly11:34
KotCzarnyo.O11:34
KotCzarnyim using thinkpad x40 ;)11:36
KotCzarnyand its the best!11:36
siceloi am using that too :-)11:36
sicelolove it to bits as well11:36
sicelobut since 4 weeks now, hibernation just doesn't work ..11:37
KotCzarnyits light, fast enough, with ultrabase allows for nice storage11:37
sicelono ide what's up11:37
sicelo*idea11:37
KotCzarnylinux hibernation or windoze?11:37
KotCzarnyalso, i've replace slow hdd with pata-msata adapter11:38
KotCzarnyand now it really flies11:38
sicelolinux11:38
sicelomine has 60GB SSD11:38
KotCzarnyhmm, tux-on-ice or swsusp?11:38
KotCzarnymsata is more versatile11:38
siceloKotCzarny: no idea .. (swsusp/tux-on-ice)11:38
sicelohow would i find out what's in use?11:39
KotCzarnyin /sys/power11:39
brolin_empeyI suppose that, if I was a far more skilled hardware hacker than I currently am, I could try to connect a header to the pins of the Super I/O controller IC that are used for the flexible disc drive controller function but, even if I succeeded at that, the closed-source firmware for the motherboard may not support a flexible disc drive using the original parallel interface instead of USB or may have this feature disabled in the build shipped by the11:39
brolin_empeymotherboard maker.  It might be like trying to use a flexible disc drive with the original parallel interface or maybe even only a PATA(PI) drive without a bridge/converter on an x86 Mac motherboard.11:39
KotCzarnythere should be some tuxonice specific files11:39
MaxdamantusYou don't need anything in userspace to do suspend-to-disk in Linux.11:41
Maxdamantusjust write the major:minor of the swap device to /sys/power/device then write `disk` to /sys/power/state11:42
sicelohttps://www.kernel.org/doc/Documentation/power/basic-pm-debugging.txt11:42
sicelothat's what i was testing (exactly as Maxdamantus says)11:42
* Maxdamantus only uses swsusp on his laptop because it's encrypted.11:42
KotCzarnymaxd: toi is a bit more flexible/advanced11:42
KotCzarnyunless things changed in a few last years11:43
sicelothe suspending works actually, but when resuming, immediately after loading the hibernation image, the laptop reboots .. haven't figured out why11:43
siceloif there was serial console maybe :)11:43
KotCzarnymight be intel drivers11:43
KotCzarnywith toi it was always graphics drivers to be the cause11:43
KotCzarny:)11:43
siceloso i can still use/try toi?11:44
KotCzarnyyup11:44
sicelowill try it .. it'd be nice to have hibernation again (battery isn't that good here now)11:45
sicelofwiw, it was working fine before11:46
siceloKotCzarny: the bluetooth, lid switch, and USB ports don't work on mine :)11:46
brolin_empeyKotCzarny: “Slow HDD” is redundant. ;-)11:46
KotCzarnybrolin, not necessarily ;)11:47
KotCzarny1.8" 4200 rpm drives are SLOOOOOW11:47
sicelothe USB problem solved with cardbus. when i feel in good mood, might open up the lappy for the other two things .. just in case11:47
brolin_empeyKotCzarny: Even the fastest HDD available, such as maybe a >=15 000 RPM model, still has a seek penalty that makes it slower than an SSD.11:49
KotCzarnyyes, that's true11:49
KotCzarnybut that particular drive from x40 is mega slow even for hdd11:49
*** Vajb has joined #maemo12:04
*** florian has joined #maemo12:04
brolin_empeyKotCzarny: I already found that FC5025 page months or years ago.  AFAIK, the reason for the lack of a read-write >=5.25-inch flexible disc drive with USB (Mass Storage) host interface as a COTS product is the perceived lack of sufficient market demand for the product to justify the cost of developing such a product and then producing and marketing it, not because it is technically impossible.  I believe the situation is similar with COTS products such as a12:13
brolin_empeymotorcycle with a compression-ignition engine instead of a spark-ignition engine or an electric motor or a smartphone with a medium-wave AM radio receiver in addition to an FM radio transceiver.12:13
sicelobleh .. debian's vanilla kernels don't support TOI, and I can't reach the TOI website12:14
KotCzarnyhttps://github.com/NigelCunningham/tuxonice-kernel12:15
KotCzarnycheck this12:15
KotCzarny4.6.0 o.o12:17
KotCzarnykinda old12:17
KotCzarnythere is 4.8 branch though12:18
KotCzarnyas for information try here http://www.thinkwiki.org/wiki/TuxOnIce12:19
KotCzarnymight be old too but has some useful information12:19
MaxdamantusSo what does it actually give you over the standard swsusp interface?12:21
siceloi'm doubtful it will help tbh :-/12:23
MaxdamantusHm. Reading the Wikipedia article it seems to list things that I'm pretty sure are part of the swsusp mechanism.12:23
sicelolooks like here i just need to find out what is crashing the system after loading the hibernate image12:23
KotCzarnymaxd: i donbt know if it still lacks it, but toi had compressible image12:24
MaxdamantusKotCzarny: afaik, you're meant to be able to just read the snapshot from the /dev/snapshot device after making the appropriate ioctls.12:25
MaxdamantusKotCzarny: what you do after you've read that should be up to you.12:25
sicelosuspend to ram resumes perfectly fine by the way12:26
siceloonly suspend to disk that fails12:27
Maxdamantus> Some efforts have been made over time to merge TuxOnIce into the vanilla kernel, but these have been opposed by Pavel,[1] primarily (it seems) because Pavel believes much of the suspend-to-disk process can and should be run from userspace (see uswsusp), while Nigel Cunningham thinks this code belongs in the kernel.12:28
MaxdamantusMaybe that's the distinction.12:28
Maxdamantusall this "more flexibility" stuff refers to teh fact that it just implements more things in the kernel.12:29
sicelowhich Pavel is that? :)12:29
MaxdamantusI'm guessing Pavel Maciek, or whatever.12:29
sicelo(uswsusp does do compression .. just checked)12:29
MaxdamantusMachek*12:29
Maxdamantususwsusp doesn't do compression.12:30
MaxdamantusUserspace tools that use uswsusp do.12:30
KotCzarnypavel machek afair12:33
MaxdamantusYes, it is him.12:33
KotCzarnyalso, is there any howto on using uswsusp with compression?12:33
Maxdamantushttps://lkml.org/lkml/2007/4/24/40512:33
MaxdamantusThe howto would be in suspend-tools or whatever you're using swsusp with.12:34
Maxdamantusswsusp is just the interface provided by Linux in /dev/snapshot12:34
*** florian has quit IRC12:36
KotCzarnythat link is just an arguing about which one is what, how to setup compression?12:38
sicelo12:35 < Maxdamantus> The howto would be in suspend-tools or whatever you're using swsusp with.12:41
Maxdamantushttps://git.kernel.org/cgit/linux/kernel/git/rafael/suspend-utils.git/tree/HOWTO12:44
MaxdamantusI have a feeling it uses it by default if you enable it at compilation.12:44
Maxdamantusotherwise you can pass that option, `compress`.12:44
*** Guest10767 has quit IRC13:34
*** Ariadeno has joined #maemo13:35
*** Ariadeno is now known as Guest6504013:35
*** troulouliou_div2 has joined #maemo14:27
*** troulouliou_div2 has quit IRC14:50
*** fuz_ has quit IRC15:21
brolin_empey“<sicelo> suspend to ram resumes perfectly fine by the way15:36
brolin_empey<sicelo> only suspend to disk that fails”15:36
brolin_empeyThen replace the volatile memory used as the main memory of the computer with non-volatile memory. ;-)  The assumption that the main memory of a computer as used in the Real World is volatile memory may become false as electronic memory evolves.15:36
*** fuz_ has joined #maemo15:40
brolin_empeyI guess that the problem in practice of using non-volatile memory as the main memory of a computer is longevity due to the software modifying (writing to) main memory much more often than it modifies (writes to) the secondary storage, which is necessarily non-volatile memory.15:41
brolin_empeyThe academic/scholarly/theoretical ideal of using the same type of memory for both primary storage (main memory) and secondary storage (disk) has already been achieved/accomplished/realised because solid-state drives that effectively use volatile memory as non-volatile memory by using RAM backed by a battery or (super/ultra)capacitor are already commercially available.15:48
*** troulouliou_div2 has joined #maemo15:57
brolin_empeyIf solid-state non-volatile memory with the same characteristics such as performance and longevity/write endurance as the RAM currently used as main memory becomes as available and at as low of a cost as current RAM, consumer-level computers can have non-volatile main memory, at least from a hardware design perspective.  It might still be decades away but it will likely happen, like how AMOLED displays are already commonly used in high-volume consumer16:05
brolin_empeyproducts such as mobile computers (marketed as smartphones or tablet computers).16:05
brolin_empeyMaybe (SO)DIMMs using non-volatile memory are already commercially available?  I have not checked.16:09
*** jonwil has quit IRC16:13
brolin_empeyMaybe computer designs based on RISC V instead of the ARM will even reach production, at least as a niche product for Free/libre hardware enthusiasts.16:17
brolin_empeyMaybe even rms will carry a computer with a constant cellular connection with him?16:20
brolin_empeyKotCzarny: sicelo: Maxdamantus: Anyway, back to discussing computing in current practice: Why do you need compression for suspend to disk/hibernation when you are using a computer that is probably limited to a maximum of 4 GiB or even only 2 GiB of main memory?  My completely solid-state Dell Latitude X1 with the maximum of 2 GiB of main memory installed has a 128-GB internal SSD;  why bother compressing the suspend-to-disk/hibernation image when it uses16:29
brolin_empeyonly up to 2 GiB of a 128-GB drive?16:29
brolin_empeyBecause compressing the image before writing it to the SSD reduces the writes to the flash memory?  I guess that could be beneficial in practice.16:33
*** flo_lap has joined #maemo16:36
*** flo_lap is now known as florian16:36
sicelobrolin_empey: i am not particularly interested in compression. it came up while going over the benefits of tuxonice over other implementations, suchs as swsusp16:44
brolin_empeysicelo: OK.16:49
brolin_empeyDocScrutinizer05: How do you pronounce your “real” name (joerg reisenweber)?16:59
DocScrutinizer05well, err17:10
DocScrutinizer05yerk ry-sen-whe-bear17:12
*** disco_stu has quit IRC17:13
*** disco_stu has joined #maemo17:13
DocScrutinizer05s/sen/zen/17:17
infobotDocScrutinizer05 meant: yerk ry-zen-whe-bear17:17
DocScrutinizer05Tesla explodes, 2 fatalities? ouch!17:18
Vajbsamsung batteries17:21
DocScrutinizer05nah, crash17:27
DocScrutinizer05http://tech.firstpost.com/auto/tesla-model-s-crashes-and-explodes-in-giant-fireball-two-fatalities-reported-345377.html17:27
* DocScrutinizer05 just doesn't get it why those battery blocks are not employing fuses (possibly with a blow control line to e.g. trigger a micro explosive cutting the wire) to connect the cells17:29
DocScrutinizer05honestly a catastrophic electrical short circuit MUST NOT happen in battery blocks, no matter what17:30
DocScrutinizer05to make such a block explode like that, you at least should have to pierce every single cell with a nail or a bullet17:31
*** disco_stu has quit IRC17:33
DocScrutinizer05also the cells should be wrapped into sort of foil that turns into foam (and become 50 times as thick) on ~90°C, providing increasing mechanical separation of cells as well as thermal isolation17:34
DocScrutinizer05the blow control trigger line, operated much like an airbag, would also conventiently remove any dangerous high voltages from car's electrics, to facilitate rescue measures like cutting the car17:38
* DocScrutinizer05 heads out to his patent lawyer ;-)17:39
*** trumee has quit IRC17:53
*** disco_stu has joined #maemo17:54
*** trumee has joined #maemo18:01
*** sunshavi has joined #maemo18:37
brolin_empeyDocScrutinizer05: OK, thank you.  I will try to remember how to pronounce your “real” name.18:39
kerioDocScrutinizer05: what do you mean yerk18:52
keriothere's a g18:53
kerio:<18:53
kerioDocScrutinizer05: no need for all of that when autopilot doesn't crash :^)18:53
DocScrutinizer05just that autopilot wasn't enabled and that's WHY it crashed18:54
keriotesla autopilot is pretty shit tho18:55
DocScrutinizer05brolin_empey: why would you want to pronounce my name? :-)18:55
DocScrutinizer05tesla actually has no autopilit, it has an assistant18:56
DocScrutinizer05pilot*18:56
kerioyeah but they call it autopilot18:57
*** disco_stu has quit IRC18:59
*** disco_stu has joined #maemo18:59
*** L29Ah has joined #maemo19:09
*** Pali has joined #maemo19:18
*** troulouliou_div2 has quit IRC19:19
ds3doesn't it come down to what 'auto' means in this context?19:32
ds3auto as in autonmous vs auto as in automobile19:33
*** reinob has quit IRC20:24
*** Pali has quit IRC20:29
KotCzarnybrolin_empey: hibernate/resume time20:47
KotCzarnyit really makes a difference on x4020:47
KotCzarnybecause its using crappy pata--msata bridge (~70MB/s)20:47
KotCzarnyminimizing writes isnt bad either20:48
*** trumee has quit IRC20:50
*** trumee has joined #maemo20:50
stryngsAnyone in here ever flash with slackware?22:18
stryngshttp://pastebin.com/uZFgbe6p22:19
CatButtsflash? I know flash22:20
CatButtsoh wait22:20
CatButtswrong flash22:20
CatButtshahahahaha22:20
*** valeriusL has quit IRC22:33
*** valeriusL has joined #maemo22:33
*** janemba has joined #maemo22:41
janembahi22:41
janembawhen trying to flash my n900 I have the following error : http://pastebin.com/BMhy6feY22:42
KotCzarnywhat is that?22:42
KotCzarnycmdline?22:42
stryngsKotCzarny: yeah22:44
KotCzarnyi was asking what cmdline he used22:44
stryngsKotCzarny: the initial flash22:44
stryngssudo ./flasher-3.5 -F RX-51_2009SE_10.2010.13-2.VANILLA_PR_EMMC_MR0_ARM.bin -f22:44
janemba ./flasher-3.5 -F RX-51_2009SE_10.2010.13-2.VANILLA_PR_EMMC_MR0_ARM.bin -f22:45
janembalol22:45
janembawhat's error code 1 btw ?22:46
KotCzarnytbh i dont remember ever seeing 'protocol mk ii'22:47
KotCzarnybut i might've overlooked it22:47
*** cyphase has quit IRC22:48
siceloovi's offscr backgammon is likely cheating :p22:49
KotCzarnylol22:50
*** cyphase has joined #maemo22:53
sicelothe odds are very ... strange :)22:59
siceloi should play it against gnubg one of the days when i'm in the mood23:00
*** LauRoman has quit IRC23:00
KotCzarnyor grab second n900?23:01
KotCzarny(might even be enough to run in emulator?)23:01
*** janemba has left #maemo23:12

Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!