IRC log of #maemo for Saturday, 2017-07-15

*** Vajb has joined #maemo00:11
*** louisdk has quit IRC00:15
*** Vajb has quit IRC00:16
*** Kilroo has joined #maemo00:27
*** dafox has quit IRC00:35
*** Vajb has joined #maemo00:40
*** Vajb has quit IRC00:46
Maxdamantus07:19:09 < DocScrutinizer05> kernel LOC 10,118,757(2.6.28)->12,532,667(2.6.32)00:47
Maxdamantus 21926 files changed, 3675432 insertions(+), 1262044 deletions(-)00:47
Maxdamantus(git diff --stat v2.6.28 v2.6.32)00:47
*** Vajb has joined #maemo00:48
enycMaxdamantus: what about 2.6.28 'stock' to maemo-power 2.6.28 ?00:49
Maxdamantus1321 files changed, 339414 insertions(+), 10614 deletions(-)00:49
Maxdamantusoh wait, that's not -power00:50
MaxdamantusThat's difference between mainline 2.6.28 and stock maemo 2.6.2800:50
enycMaxdamantus: more to the point, WHICH 2.6.28 version EXACTLY was maemo's kernel based on00:50
enycMaxdamantus: is there sort of thing, or does it not work like that ??00:50
Maxdamantusmainline 2.6.28 to -power53 is: 2010 files changed, 374761 insertions(+), 15117 deletions(-)00:50
MaxdamantusI guess I can keep comparing the patch releases to find which one is most similar.00:51
enycMaxdamantus: yes please =) has 2.6.28 upto
enycMaxdamantus: though does the maemo source just have a release/makefile with the extraversion in it00:52
MaxdamantusThe initial v2.6.28 release seems to be closest.00:54
Maxdamantus(the diff is from mainline to maemo 2.6.2800:55
Maxdamantusenyc: no. It has other changes too.00:55
MaxdamantusHm, there's a debian/changelog in my maemo tree with 14k lines in it though.00:56
MaxdamantusSo just subtract around 15k from all of those "insertions" counts.00:57
MaxdamantusHm. There are also additions under Documentation/arm/OMAP/, so maybe it's actually a fork of some omap branch somewhere.00:58
enycMaxdamantus: post patch/list ??00:58
enycMaxdamantus: i wonder how much is actually very much patches, and how much is just inserting huge piles of new drivers etc etc...00:58
MaxdamantusRandom modifications to a bunch of the fs drivers further suggests it's not a direct fork from mainline 2.6.2800:59
enycMaxdamantus: hrrm yes looking myself toooo puzzling01:01
enycMaxdamantus: i wonder whats going on01:01
enycMaxdamantus: that said, key is just to pickout whats' needed atop 2.6.32-latest 'stock'  and see if that patches relatively cleanly.....01:02
enycthat said, looks like kernel changes to 2.6. 29 30 31 32   added lots of OMAP support/changes anyway ....... good thing or bad thing01:09
MaxdamantusI can't find references to various blobs for the `fs` changes in any commits in my repository other than ones relating to N900.01:12
enychumm err arr err01:12
MaxdamantusWonder if has a way to search for blobs in .. all repositories .. probably not.01:12
enycdont quit e understand implication01:12
MaxdamantusWell, eg: `git rev-parse n900-2.6.28:fs/ext3/super.c` gives back "f4be66e3d62133a36f5b81714aa5f2b16f26e75b"01:13
enyci'm not familaiar wit that rev-parse01:13
enycwhat is this telling us?01:13
enycthat the 2.6.28 changes can't really be found anywhere?01:14
Maxdamantusso the 2.6.28 stock maemo kernel has an fs/ext3/super.c file with that checksum (basically; prepend "blob 84547\0" to teh beginning of the file)01:14
enycso whats the significance of those weird blobs on the beginning?01:14
Maxdamantusand I can't see that exact file in any reachable commit in my repository other than the ones I've made from those maemo sources and the branches I've fetched from Pali here.01:15
MaxdamantusThe significance is that I think it indicates that nokia's changes aren't a direct branch from mainline 2.6.2801:15
Maxdamantusand that there's probably another development branch somewhere that they used.01:16
Maxdamantusunless they themselves actually did those modifications to `fs` .. I guess that could be plausible, depending on what the changes aro.01:16
enycyes, dothey look significant?01:18
enyci'm thinking .... git // work-in-progress befre 2.6.28 itself released ?01:19
MaxdamantusHm. Some of the changes seem to be from after
*** dafox has joined #maemo01:20
Maxdamantuseg, the changes include those from commit 85c7859190c4197a7c34066db14c25903c401187, which is in the mainline master branch, based on v2.6.30-rc801:22
enycseems to suggest couldn't JUST use a latest 4.12 kernel compile right with old maemo because dsp//powervr drivers01:23
MaxdamantusRight, but I think you can use that 4.xx branch to run Maemo.01:24
MaxdamantusIt's just not compatible with mainline.01:24
enycwhat you mean?01:24
enycto run stock maemo ??01:24
enycor slow because driver missing?01:25
MaxdamantusI think you're meant to be able to use to run stock Maemo. I might be mistaken.01:25
* Maxdamantus has only tried a 3.18 or something kernel from there a year or two ago, and didn't try running stock Maemo.01:26
Maxdamantusactually, 3.1401:26
enycwill pali just tell us what the expected state is?!?!?01:26
enycfwiw,  <- interesting01:27
Paliwhat is the problem?01:28
*** florian has quit IRC01:28
*** navtis has joined #maemo01:30
enycPali: we'd like to know (a) what the limitainso of stock-maemo with a 4.12 etc n900 kernel latest are01:30
PaliHW support/kernel drivers is in table on
Paliyou can find in which version was what introduced in mainline kernel01:31
enycPali: yes, i see 2 red blobs, what practiceal problems that causes?01:31
enycPali: tidspbridge and PowerVR  won't be workable ?01:32
enycPali: is that it? does that make stock maemo fail because driver unavailable?01:32
Paliif you want to run Maemo, look at steps:
Pali-n900 branches in my git tree contains some patches01:33
Palithere is ported PowerVR driver01:33
Palifixes for wifi01:33
Palimaemo compatibility layer (where still needed)01:33
Palino idea if bluetooth is working...01:34
Palipatches for camera are not (finished) in my git repo01:35
Paliand tidspbride does not work at all01:35
enyci see didn't know all this01:35
enycwhats' the tidspbridge used for?01:36
Palipavelm now playing with camera, so we can expect working camera in new+1 or +2 mainline kernel version01:36
enyci see i see01:37
enycso a workable kernel is well on way...?01:37
Palitidspbridge is bridge driver for TI DSP processor01:37
enycyes, but whats that used for?01:37
PaliDSP processor used e.g. for video encoding/decoding01:38
Palimaybe there are also applications which uses it for audio01:39
Palior other multimedia processing01:39
Palido not know who else uses DSP in Maemo... if there are more applications in Maemo-Extras01:39
*** navtis has quit IRC01:42
enyctalking of wifi, is WPA host / 'hostapd' ever a possibility with that adapter?01:43
Palitheoretically with software implementation01:45
sunshavipali: Sebastian Reichel was working on bluetooth. But probably You know that. btw camera is very slow on n900, perhaps pavel version is quicker01:45
Paliwifi firmware supports packet injection and monitor mode which should be enough for implementing AP mode01:45
Paliwl1251 driver has both support01:46
Palibut I do not know any software-based wifi AP implementation01:46
PaliIIRC there are some problems with bluetooth on n900, do not know if those problems were fixed or not01:47
Palilast time I talked with pavel, he told me that camera for him is working... and I see patches periodically on mailing list01:47
Paliso we can expect full-working camera also in mainline01:48
sunshavipali: nice to know01:50
*** mike727 has joined #maemo01:53
*** navtis has joined #maemo01:58
*** th652 has quit IRC01:59
*** xy2_ has quit IRC02:09
*** navtis has left #maemo02:10
*** mike727 has quit IRC02:13
*** th652 has joined #maemo02:20
*** mavhc has quit IRC02:34
*** dafox has quit IRC02:34
*** xorly has quit IRC03:04
*** mavhc has joined #maemo03:13
*** Pali has quit IRC03:15
*** infobot has quit IRC03:19
*** infobot has joined #maemo03:20
*** guerby_ has joined #maemo04:36
*** guerby has quit IRC04:37
*** r00t|home has quit IRC04:54
*** r00t|home has joined #maemo04:54
*** pagurus has quit IRC06:22
*** pagurus has joined #maemo06:22
sicelosoftware based wifi ap - hostapd?07:38
DocScrutinizer05BUG!!  entropy runs dry in maemo5 (internet over WLAN, ssh session established). Though poolsize is 4096, it lingers at 177 and drops from 235 to 177 within minutes07:47
DocScrutinizer05damn! 13307:47
DocScrutinizer05though this isn't a problem per se it may have security implications. But particularly weird is what actually drains it07:49
DocScrutinizer05processes actually doing blocking read from /dev/random may freeze/lag07:50
DocScrutinizer05suggested measures: find (and fix) the process draining entropy pool. There should be no need for random numbers in normal idle system, IP stack should use its own less valuable source of entropy for sequence numbers etc.07:52
DocScrutinizer05run a process that monitors and - when needed - fills entropy pool from "unusual" sources like the used internet connection itself. E.G from a `iwlist scan`07:58
DocScrutinizer05other possible sources of entropy:  bq27200 battery voltage, current; audio sampling from microphone; audio sampling from FM-RX(!) white noise when tuned to no station; accelerometer (poor); ALS (poor, low bitcount). All of the aforementioned only lower significant few bits08:03
DocScrutinizer05the entropy-generating `ls /*` is probably a very poor quality entropy source, given the lack of any mechanical components or quantum effects involved in the whole process08:05
DocScrutinizer05actually this might be one source of hangs in ssh sessions and generally IP traffic08:08
MaxdamantusSuggested measures: ln -sf /dev/{u,}random08:15
MaxdamantusBecause random is practically never useful.08:16
MaxdamantusI'm pretty sure ssh won't use /dev/random btw08:23
Maxdamantusssh-keygen might.08:23
DocScrutinizer05freemangordon: ^^^ what do you think?08:25
MaxdamantusIf only Linux had some sort of CSPRNG that people could read the output of but not the state.08:27
DocScrutinizer05seems I fail to find a way to add to available entropy in /dev/(u)random08:32
DocScrutinizer05this is expected according to but it's pretty annoying nevertheless. Could we patch the device drover so it offer some writable node in /proc or /sys to actually *add* to the available entropy pool?
DocScrutinizer05also note that `iwlist scan` seems to block the WLAN for a short (0.5s?) period by itself, during which no data traffic is possible08:37
*** louisdk has joined #maemo08:57
DocScrutinizer05IroN900:~# lsof |grep random09:19
DocScrutinizer05csd         804       root    9r   CHR        1,9                1757 /dev/urandom09:19
DocScrutinizer05telepathy  2010       user    9r   CHR        1,9                1757 /dev/urandom09:19
DocScrutinizer05browserd   2712       user   12r   CHR        1,9                1757 /dev/urandom09:19
DocScrutinizer05browserd   4145       user   12r   CHR        1,9                1757 /dev/urandom09:19
*** th652 has quit IRC09:23
DocScrutinizer05>>The first non-Android device! It comes with a hardware keyboard, so one can type in the full disk encryption password directly. It uses pali's 4.6 kernel, which is relatively close to mainline already.<<09:26
KotCzarnyit might be a bit too much of tinfoil hat, but maybe nokia got crushed by international evil  forces for trying to make linux on phones happen09:28
DocScrutinizer05  to cure your tinfoil hat feelings10:02
DocScrutinizer05>>That can be used to turn your device silently into a surveillance device, because these components have direct access to your device's RAM (via DMA)<< not on N90010:03
DocScrutinizer05at least as far as anybody knows. And for sure not on Neo900 ;-)10:04
hurrianOne benefit of having a USB^H^H^H HSI modem ;)10:13
*** Konsieur has joined #maemo10:13
DocScrutinizer05I couldn't resist editing the nokia-rx51-(Nokia-N900) wiki page ;-D10:17
DocScrutinizer05FOSS radio stack, the most common pipe dream10:18
DocScrutinizer05will *never* happen10:19
DocScrutinizer05and even if id would, I wouldn't dare using it10:19
hurrianan open s/w interface to the baseband and AP-managed access is enough though :D10:19
DocScrutinizer05I regret not having bookmarked any of those stories about US citizens sentenced to unspeakable 6 digit fines for using non-certified RF transmitter equipment10:22
DocScrutinizer05like, USD300,000 plus some time in jail iirc10:23
hurrianguess that was why having a SDR module embedded in the Neo900 wasn't a design goal? :p10:23
DocScrutinizer05SDR is greedy and expensive10:24
hurrianyeah, the sheer power and RF board routing requirements would be crazy10:24
DocScrutinizer05and we actually have sort of SDR, for NFC10:24
hurrianbut then again, so would a phone with a software defined baseband10:24
DocScrutinizer05well, it's not _really_ SDR, it just has a MCU for protocols10:26
hurrianthe Neo900 has NFC?10:26
hurriancool, didn't know that. guessing some modification to the back cover would be required?10:27
DocScrutinizer05yes, probably. That's why we don't deliver an antenna with it per default10:28
DocScrutinizer05or, we still need to evaluate which antenna we might use10:28
DocScrutinizer05if we're extremely lucky, a flex foil PCB antenna will do10:29
DocScrutinizer05that wouldn't need any massive modificarions to battery lid10:29
hurrianmy guess is reuse of one of those wireless charging receiver+NFC coils for Android (usually Samsung) phones10:30
DocScrutinizer05that sort of stuff10:30
DocScrutinizer05it's RF. so voodoo, so only possible to test in real device10:31
hurrianand maybe bring two pogo pins through the back case of the N900, hoping they get close enough to the pads of said wireless charger pad10:31
DocScrutinizer05NFC_GND, NFC_ANT10:32
hurrianyep, something like that10:33
hurrianimagine how much fun could be had though given a few hundred thousand USD more in tooling (e.g. for plastic molds) could go- 5/6 inch N900?10:34
DocScrutinizer05  >> 4.3  Mechanical coupling with battery cover<<10:34
*** florian has joined #maemo11:00
*** florian has quit IRC11:05
*** xorly has joined #maemo11:07
*** aloril_ has quit IRC11:17
*** Pali has joined #maemo11:49
*** xy2_ has joined #maemo12:43
*** NeKit has joined #maemo13:29
*** TheKit has quit IRC13:31
*** xy2_ has quit IRC13:43
*** florian has joined #maemo14:02
*** xy2_ has joined #maemo14:05
*** florian has quit IRC14:11
*** florian has joined #maemo14:20
*** aloril has joined #maemo14:59
sicelooh gosh ... our mobile operator does not isolate clients, ie i can connect to other phones if they have any running services15:33
sicelosurely that's wrong?15:33
KotCzarnymoney stealer15:44
KotCzarnyunless you dont get charged for incoming packets15:45
KotCzarnywhich is unlikely ;)15:45
KotCzarnyudp flood someone ;)15:46
siceloyou don't get charged15:57
*** chfoo[m] has quit IRC15:59
*** chfoo[m] has joined #maemo16:04
*** Natch has quit IRC16:08
Maxdamantus"connect" as in just a normal TCP connection?16:11
MaxdamantusPresumably it should be the responsibility of the OS to not allow the ISP or anyone who happens to be on the internet to do bad things to it.16:11
MaxdamantusI imagine it would be unreasonable to make a phone OS that assumes it's always behind some NAT/firewall provided by the ISP.16:12
MaxdamantusI would also imagine the same set of "services" would generally be exposed over WiFi.16:14
KotCzarnyi would be afraid about possible costs of not putting phone behind the nat16:16
MaxdamantusHow would these "services" work behind a NAT anyway?16:22
*** florian has quit IRC16:25
siceloconnect normal tcp  connection .. yes. i can ssh my n900 from ny laptop using usb dongle with same cellular operator16:39
*** Vajb has quit IRC17:10
siceloat least it doesn't charge anything17:14
*** Vajb has joined #maemo17:32
*** Vajb has quit IRC17:40
*** Vajb has joined #maemo17:58
sicelothebpower consumption though!18:00
KotCzarnyrun tcpdump on it? :)18:00
sicelojust tested ... running an 'htop' this way makes standby current draw to rise to 200mA with screen off18:01
*** Vajb has quit IRC18:03
*** mavhc has quit IRC18:33
*** mavhc has joined #maemo18:35
*** Vajb has joined #maemo18:35
*** mavhc has quit IRC18:36
*** mavhc has joined #maemo18:37
*** Vajb has quit IRC18:40
DocScrutinizer05inbound pings and other shite can't get blocked, the modem _has_to_ send ack18:52
*** Vajb has joined #maemo18:52
DocScrutinizer05the only way to stop is to disable the APN18:53
*** LjL has quit IRC19:05
siceloinbound pings are actually blackholed by default on n900on gprs0, but allowed on the other connections19:06
sicelo /etc/network/if-up.d/00_disable_icmp_echo_reply19:08
*** Vajb has quit IRC19:10
*** xorly has quit IRC19:14
*** Vajb has joined #maemo19:18
DocScrutinizer05inbound data package needs an ACK always, on GPRS OTA protocol, independent from TCP OR IP or whatever19:23
*** LjL has joined #maemo19:24
DocScrutinizer05think of it as an ethernet through GPRS/UMTS tunnel19:24
*** Vajb has quit IRC19:34
*** Konsieur has quit IRC19:40
*** xy2_ has quit IRC19:44
*** xy2_ has joined #maemo19:44
*** Vajb has joined #maemo19:46
*** florian has joined #maemo19:48
*** xy2_ has quit IRC19:52
*** Vajb has quit IRC19:57
*** Vajb has joined #maemo19:57
*** florian has quit IRC20:02
*** Vajb has quit IRC20:06
*** xy2_ has joined #maemo20:19
*** sfa has joined #maemo20:28
*** sfa has quit IRC20:29
*** Vajb has joined #maemo20:33
*** Vajb has quit IRC20:44
*** Vajb has joined #maemo20:46
*** Vajb has quit IRC20:51
*** Vajb has joined #maemo21:14
*** Vajb has quit IRC21:24
*** florian has joined #maemo21:32
*** frals has quit IRC21:35
*** frals has joined #maemo21:40
*** frals has joined #maemo21:40
*** Vajb has joined #maemo21:49
*** Vajb has quit IRC21:53
*** xy2_ has quit IRC22:00
*** msava has joined #maemo22:05
*** xy2_ has joined #maemo22:05
*** xes_ has joined #maemo22:11
*** xes has quit IRC22:13
*** msava has quit IRC22:17
*** msava has joined #maemo22:18
*** Vajb has joined #maemo22:18
*** Vajb has quit IRC22:25
*** Vajb has joined #maemo22:25
*** Vajb has quit IRC22:29
*** Vajb has joined #maemo22:30
*** Vajb has quit IRC22:39
*** Vajb has joined #maemo22:41
*** Vajb has quit IRC22:46
*** frals has quit IRC22:46
*** Vajb has joined #maemo22:46
*** Vajb has quit IRC22:51
*** frals has joined #maemo22:57
*** frals has joined #maemo22:57
*** Vajb has joined #maemo23:09
*** merlin1991 has quit IRC23:17
*** merlin1991 has joined #maemo23:17
*** Vajb has quit IRC23:18
*** Vajb has joined #maemo23:48
*** dafox has joined #maemo23:48
*** Vajb has quit IRC23:54
*** Vajb has joined #maemo23:55

Generated by 2.15.1 by Marius Gedminas - find it at!