IRC log of #maemo for Monday, 2013-12-30

ShadowJKI discovered a dns issue in my setup today00:01
ShadowJKPrimary dns was typoed to 127.0.1.100:01
ShadowJKSo everything not recently accessed too 5 seconds for 127.0.1.1 to timeout before asking secondary dns server. :)00:02
*** beford has joined #maemo00:04
*** OkropNick has quit IRC00:05
*** jmlich has quit IRC00:06
*** Timmy has quit IRC00:08
*** zammy has quit IRC00:09
*** dhbiker has quit IRC00:11
*** Timmy has joined #maemo00:19
*** robotanarchy_ has joined #maemo00:23
*** robotanarchy has quit IRC00:23
*** natsukao1 has quit IRC00:23
*** robotanarchy_ is now known as robotanarchy00:23
*** Timmy has quit IRC00:23
*** larrfed2 has quit IRC00:31
*** MohammadAG has quit IRC00:32
*** arcean_ has quit IRC00:35
*** MohammadAG has joined #maemo00:36
*** larrfed has joined #maemo00:38
*** kolp has quit IRC00:46
*** louisdk has joined #maemo00:47
*** Venemo has quit IRC00:50
*** zamn900 is now known as zamn900AWAY01:00
*** peetah has joined #maemo01:01
*** Agge has joined #maemo01:04
*** Agge has quit IRC01:05
*** Aggese has joined #maemo01:05
*** disco_stu has quit IRC01:10
*** louisdk has quit IRC01:15
*** Aggese is now known as Agge01:20
*** Timmy has joined #maemo01:24
DocScrutinizer05ShadowJK: you know they hack eMMC controller firmware now?01:27
ShadowJKnop01:27
DocScrutinizer05http://forum.xda-developers.com/showthread.php?t=209604501:28
ShadowJKSome samsung microsds that bunniestudios decapped contained an ARM core inside :)01:28
DocScrutinizer05M301:28
DocScrutinizer05M4?01:29
*** rcg has quit IRC01:30
ShadowJKSpeedEvil: ^ should be interesting for you too01:30
DocScrutinizer05I really start giggling when thinking about RMS and his FSF-certified laptop that for sure also has at least 2 or 3 controllers that are considered "chips" but actually can change their firmware01:30
*** dafox has joined #maemo01:32
*** eijk has quit IRC01:32
*** Timmy has quit IRC01:33
*** skule has quit IRC01:34
*** Kabouik has joined #maemo01:34
DocScrutinizer05ShadowJK: (127.0.1.1) sounds pretty much similar to my problem, judging by symptoms. OTOH I think 127.x.x.x is all supposed to be LOCALHOST01:35
*** Kabouik_ has quit IRC01:36
DocScrutinizer05actually localhost (servernames and uppercase don't go together well)01:37
*** rcg has joined #maemo01:38
DocScrutinizer05I learned the hard way when trying to name a server L2.domain.tld01:38
Hurrianwhoa, that's pretty cool - they put ARM cores in eMMCs now?01:39
DocScrutinizer05which broke munin01:39
kerioHurrian: it's cheaper than shipping perfect flash chips01:39
DocScrutinizer05sure01:39
Hurrianlooks like SSD-style wear leveling has hit the low-end :D01:39
DocScrutinizer05always did01:39
ShadowJKHurrian; well01:40
ShadowJKCF, SD, uSD have had arm cores and other uCs for 5-10 years01:40
ShadowJKKingston actually has a controller desgiend from scratch, but they rarely use it, and it sucks01:41
*** shentey has joined #maemo01:41
HurrianCounting down to when people use them as complete system-on-package...01:41
ShadowJKOh, that exists01:42
ShadowJKwifi+linux system in a SD card form factor, iirc01:43
Hurrianthose WiFi SD cards run Linux?01:43
ShadowJK"electric imp"01:45
Hurrianwow, there are ARM cores in everything now...01:47
*** Venusaur has quit IRC01:49
ShadowJKworth keeping in mind that arm scales very small01:50
*** Venusaur has joined #maemo01:50
*** amonk has joined #maemo01:55
*** valeriusL has quit IRC02:00
*** valerius has quit IRC02:01
*** amizraa has quit IRC02:05
*** amizraa has joined #maemo02:06
*** rcg has quit IRC02:09
*** shanttu has quit IRC02:11
*** Kabouik has quit IRC02:11
*** louisdk has joined #maemo02:11
*** valeriusL has joined #maemo02:13
*** florian has quit IRC02:14
*** valerius has joined #maemo02:16
*** amonk has quit IRC02:16
*** fortytwo has quit IRC02:17
*** fortytwo has joined #maemo02:18
*** amonk has joined #maemo02:20
*** M4rtinK2 has quit IRC02:22
b1101there isn't a buy/sell n900 thing in this channel right ?02:30
ShadowJKit's probably a bit low volume for that02:31
b1101How much should I pay for a mint n900 with all accessories ? Looking at one on ebay. I've already contributed to the neo900 project, but can't stand using android anymore. I want my rights back :).02:33
ShadowJKI would personally pay 150 without accessories02:33
ShadowJKmaybe 20002:34
b1101I like that this one is legitimate. Good pics show it's made in finland and all. I purchased a fake months ago, and was really upset.02:36
*** louisdk has quit IRC02:38
*** amonk has quit IRC02:38
*** triggerhappy has joined #maemo02:40
bef0rdthey could use legit pics and fake device though02:40
bef0rdI guess ebay/paypal protects you from that02:40
b1101ya, ebay is usually on the customers side02:41
*** amonk has joined #maemo02:42
ShadowJKThe main batch were actually made in Korea, iirc02:45
*** shentey has quit IRC03:14
*** NIN101 has quit IRC03:14
SpeedEvilShadowJK: indeed.03:14
* SpeedEvil sighs and wishes he had the energy to do stuff that required coherent thought for more than a few minutes a day.03:16
SpeedEvil(Hence lack of doing interesting stuff like paying more attention to neo900 or proper involvment with jolla, or...)03:18
*** xes has quit IRC03:18
ShadowJKyes03:22
* ShadowJK has spent his past 2 or so weeks of time off work mostly in bed silently cursing himself for not getting anything useful done03:23
b1101^in the same situation03:27
*** goldkatze has quit IRC03:43
*** Agge has quit IRC03:52
*** Aggese has joined #maemo03:52
* SpeedEvil has had his AC for ~2 weeks or so, and been unable to install it.04:09
SpeedEvilOn a more sort-of-on-topic point - it raises the other issue (which is not strictly required but is made easier) - attacks based on fuzzing responses from the card - not data, but return codes, and packet formats.04:10
*** mschlens_ has quit IRC04:15
*** mschlens has joined #maemo04:16
*** dafox has quit IRC04:18
DocScrutinizer05SpeedEvil: AC?04:25
*** qwazix has quit IRC04:25
*** dos1 has quit IRC04:27
SpeedEvilAir conditioning - for heating. Heat pump - COP 5.7 - to make it so I can actually heat this place to a reasonable level.04:28
SpeedEvil(10C in here)04:28
FIQiirc only prototype N900s were made in finland?04:29
*** qwazix has joined #maemo04:30
DocScrutinizer05sounds about right04:31
*** Humpelstilzchen has quit IRC04:32
DocScrutinizer05SpeedEvil: I think a heating should improve your activity04:33
SpeedEvilDocScrutinizer05: It doesn't.04:33
DocScrutinizer05at 10°C I also wouldn't be able to concentrate on anything04:34
SpeedEvilI'm on the sofa in a nice warm electric blanket.04:34
DocScrutinizer05yeah, but that also automatically immobilzes you04:34
SpeedEvilActually too hot at the moment. The problem is that I wobble when I get up to do anything, and my head is feeling packed by cotton wool.04:35
DocScrutinizer05dang04:35
*** Defiant has joined #maemo04:35
*** muelli has joined #maemo04:55
*** muellisoft has quit IRC04:57
*** muelli has quit IRC05:03
*** triggerhappy has quit IRC05:42
*** triggerhappy has joined #maemo05:46
*** jeverling_ has quit IRC05:48
*** jeverling has joined #maemo05:52
*** Aggese is now known as Agge05:53
*** jeverling has quit IRC05:56
*** psycho_oreos has quit IRC06:01
*** psycho_oreos has joined #maemo06:05
*** uen| has joined #maemo06:06
*** uen has quit IRC06:10
*** uen| is now known as uen06:10
*** jeverling has joined #maemo06:13
*** at1as has quit IRC06:13
*** jeverling has quit IRC06:18
*** jeverling has joined #maemo06:27
*** maybeHere has joined #maemo06:30
*** jeverling has quit IRC06:31
*** LauRoman has quit IRC06:33
*** LauRoman|Alt has quit IRC06:33
*** maybeArgh has quit IRC06:33
*** Agge has quit IRC06:44
*** Aggese has joined #maemo06:44
*** jeverling has joined #maemo06:45
*** goldkatze has joined #maemo06:45
*** beford has quit IRC06:52
*** otep has joined #maemo06:52
*** goldkatze has quit IRC06:57
*** Herbstbert_ has joined #maemo06:59
*** Herbstbert has quit IRC07:02
*** triggerhappy_ has joined #maemo07:05
*** triggerhappy has quit IRC07:05
*** triggerhappy_ has quit IRC07:07
*** triggerhappy has joined #maemo07:45
*** Neutron3 has joined #maemo07:46
*** hubutm20 has joined #maemo07:52
*** rm_work|away is now known as rm_work08:37
*** RedW has joined #maemo08:39
*** RedM has quit IRC08:40
*** ZogG_laptop has quit IRC08:48
*** b1101 has quit IRC08:54
*** b1101 has joined #maemo08:55
*** b1101 has quit IRC09:05
*** b1101 has joined #maemo09:06
*** Timmy has joined #maemo09:15
*** Agge has joined #maemo09:37
*** Aggese has quit IRC09:39
*** jmlich has joined #maemo09:44
*** darkschneider has quit IRC10:03
*** Timmy has quit IRC10:06
*** Smily has joined #maemo10:08
*** sunny_s has joined #maemo10:09
*** zamn900AWAY is now known as zamn90010:19
*** ssvb has quit IRC10:35
*** sunny_s has quit IRC10:36
*** sunny_s has joined #maemo10:38
*** valerius has quit IRC10:39
*** valeriusL has quit IRC10:39
*** OkropNick has joined #maemo10:51
*** valeriusL has joined #maemo10:52
*** valerius has joined #maemo10:53
*** rcg has joined #maemo10:53
*** sunny_s has quit IRC11:01
*** jmlich has quit IRC11:01
*** jmlich has joined #maemo11:03
*** dhbiker has joined #maemo11:11
*** zammy has joined #maemo11:13
*** zamn900 has quit IRC11:20
*** valeriusL has quit IRC11:28
*** trx has quit IRC11:28
*** valerius has quit IRC11:29
*** LauRoman has joined #maemo11:30
*** rm_work is now known as rm_work|away11:30
*** LauRoman|Alt has joined #maemo11:32
*** LauRoman has quit IRC11:38
*** rcg has quit IRC11:40
*** arcean has joined #maemo11:41
*** LauRoman has joined #maemo11:42
*** valeriusL has joined #maemo11:43
*** valerius has joined #maemo11:44
*** valerius has quit IRC11:54
*** valeriusL has quit IRC11:54
*** NIN101 has joined #maemo12:00
*** Timmy has joined #maemo12:02
*** sequantz has joined #maemo12:02
*** valeriusL has joined #maemo12:07
*** eijk has joined #maemo12:07
*** hubutm20 has quit IRC12:12
*** hubutm20 has joined #maemo12:13
*** valerius has joined #maemo12:14
*** florian has joined #maemo12:15
*** xes has joined #maemo12:24
DocScrutinizer05moinmaemoians12:34
*** zemmy has joined #maemo12:35
*** trx has joined #maemo12:36
*** zummy has joined #maemo12:36
*** zammy has quit IRC12:38
*** arcean_ has joined #maemo12:42
*** zummy is now known as zammy12:43
*** amonk has quit IRC12:44
*** arcean has quit IRC12:46
*** shentey has joined #maemo12:49
*** muelli has joined #maemo12:49
*** louisdk has joined #maemo12:59
*** Timmy has quit IRC13:00
*** amonk has joined #maemo13:03
*** eMHa__ has quit IRC13:16
*** valeriusL has quit IRC13:17
*** valerius has quit IRC13:17
*** at1as has joined #maemo13:20
*** jmlich has quit IRC13:23
*** Herbstbert_ is now known as Herbstbert13:26
*** dafox has joined #maemo13:27
*** valeriusL has joined #maemo13:31
*** valerius has joined #maemo13:42
*** chem|st_ is now known as chem|st13:54
*** eMHa__ has joined #maemo13:56
*** torindel has joined #maemo13:58
*** nslu2-log_ has joined #maemo14:01
*** Malinux_ has joined #maemo14:03
*** louisdk has quit IRC14:04
*** inz_ has joined #maemo14:05
*** mattaust1n has joined #maemo14:05
*** torindel_ has joined #maemo14:05
*** joga_ has joined #maemo14:05
*** oniongar1ic has joined #maemo14:05
*** erstazi_ has joined #maemo14:05
*** qwazix_nc has joined #maemo14:05
*** qwazix_nc has quit IRC14:05
*** qwazix_nc has joined #maemo14:05
*** erstazi_ has quit IRC14:05
*** erstazi_ has joined #maemo14:05
*** netkat_ has joined #maemo14:05
*** muellisoft has joined #maemo14:05
*** cyborg-o1e has joined #maemo14:05
*** dhbiker_ has joined #maemo14:05
*** ot0to has joined #maemo14:06
*** Milhouse has quit IRC14:06
*** valerius has quit IRC14:06
*** muelli has quit IRC14:06
*** Defiant has quit IRC14:06
*** qwazix has quit IRC14:06
*** oniongarlic has quit IRC14:06
*** erstazi has quit IRC14:06
*** ototo has quit IRC14:06
*** inz has quit IRC14:06
*** netkat has quit IRC14:06
*** mattaustin has quit IRC14:06
*** AndrewX192 has quit IRC14:06
*** torindel has quit IRC14:06
*** NIN101 has quit IRC14:07
*** dhbiker has quit IRC14:07
*** robotanarchy has quit IRC14:07
*** shamus has quit IRC14:07
*** nslu2-log has quit IRC14:07
*** Malinux has quit IRC14:07
*** cyborg-one has quit IRC14:07
*** joga has quit IRC14:07
*** AndrewX192 has joined #maemo14:07
*** qwazix_nc is now known as qwazix14:07
*** inz_ is now known as inz14:07
*** nslu2-log_ is now known as nslu2-log14:07
*** RaBBit- is now known as RaBBit14:07
*** AndrewX192 is now known as Guest5361114:07
*** shamus has joined #maemo14:07
*** Humpelstilzchen has joined #maemo14:07
*** NIN102 has joined #maemo14:07
*** torindel_ is now known as torindel14:08
*** netkat_ is now known as netkat14:08
*** ot0to is now known as ototo14:08
*** ototo has quit IRC14:08
*** ototo has joined #maemo14:08
*** florian has quit IRC14:11
*** florian has joined #maemo14:13
*** sequantz has quit IRC14:14
*** Milhouse has joined #maemo14:17
*** valerius has joined #maemo14:22
*** robotanarchy has joined #maemo14:25
*** sequantz has joined #maemo14:43
*** phlixi has quit IRC14:52
*** hubutm20 has quit IRC14:59
*** GeneralAntilles has quit IRC15:02
*** GeneralAntilles has joined #maemo15:05
*** dafox has quit IRC15:05
*** phlixi has joined #maemo15:05
*** valeriusL has quit IRC15:10
*** valerius has quit IRC15:10
*** zemmy has quit IRC15:19
*** valeriusL has joined #maemo15:22
*** dos1 has joined #maemo15:24
*** valerius has joined #maemo15:25
*** Humpelstilzchen has quit IRC16:03
*** Defiant has joined #maemo16:07
*** SmilyOrg has joined #maemo16:17
*** iDont has joined #maemo16:19
*** Smily has quit IRC16:21
*** shentey has quit IRC16:23
*** Venemo has joined #maemo16:39
*** iDont has quit IRC16:40
*** NIN102 is now known as NIN10116:40
*** netkat has quit IRC16:44
*** Ninja has quit IRC16:44
*** Gh0sty has quit IRC16:49
*** netkat has joined #maemo16:50
*** Ninja has joined #maemo16:50
*** Gh0sty has joined #maemo16:51
*** muellisoft has quit IRC16:59
*** dafox has joined #maemo17:02
*** dafox is now known as Guest8160417:02
*** Agge has quit IRC17:11
*** Aggese has joined #maemo17:11
*** Agge has joined #maemo17:35
*** FlameReaper has joined #maemo17:36
*** Aggese has quit IRC17:37
*** goldkatze has joined #maemo17:38
*** fuz_ has quit IRC17:38
*** shanttu has joined #maemo17:41
*** FReaper has joined #maemo17:42
*** Guest53611 has quit IRC17:42
*** AndrewX192 has joined #maemo17:43
*** FlameReaper has quit IRC17:44
*** FReaper has quit IRC17:47
*** sLumPia has joined #maemo17:50
*** zemmy has joined #maemo17:52
*** zemmy has quit IRC17:53
sLumPiasixwheeledbeast: Does ereswap support swapfile?17:56
sLumPiaI only have one swap partition, but I would like to use ereswap17:56
*** Gatta_Negra has quit IRC17:59
*** jmlich has joined #maemo18:00
*** fuz_ has joined #maemo18:05
DocScrutinizer05swapfile isn't working on stock maemo kernel afaik18:12
sLumPiaI am not using stock kernel18:12
DocScrutinizer05dunno about KP18:12
freemangordoneven if it works, I don;t think it makes much of a sense18:13
*** at1as has quit IRC18:13
sLumPiaYeah I know18:14
sLumPiaI thought at least ereswap can turn off my swap partition for a minute, use swapfile, then turn on swap partition again18:15
*** Agge has quit IRC18:15
*** Aggese has joined #maemo18:15
DocScrutinizer05dividing the existing 768MB swap partition into two halves should work great, it seems Nokia just used 768MB for marketing and fragmentation reasons. A system that actually is in a status where it needs 768MB swap isn't usable by any metrics anyway18:16
*** solofight has joined #maemo18:16
DocScrutinizer05fragmentation would become irrelevant with reswap18:16
sLumPiaWell, I guess there is also that option18:17
sLumPiaSo 389 MB enough for Maemo's swap partition?18:18
sLumPiaWe don't really need big swap partition?18:18
DocScrutinizer05well, I personally never needed 768MB18:19
sLumPiaI thought Maemo need bigger swap, because its lack of ram18:19
DocScrutinizer05actually it's hard to get the printout that shows swap usage >400MB, since the system already came to a grinding halt18:19
sLumPiaI'm sorry, how do you know how much swap already used by Maemo?18:21
DocScrutinizer05err, e.g. look at htop or top18:21
freemangordonfree?18:21
DocScrutinizer05or that18:21
freemangordoncat /proc/swaps?18:21
DocScrutinizer05or that18:21
freemangordon:)18:21
sLumPialol, I don't know free show swap usage.18:22
sLumPiaSilly me18:22
sLumPiaI only look at memory usage when use free18:22
sLumPiaOk guys18:22
sLumPiaThanks18:22
sLumPiaI will just devide my swap partition18:23
*** Kabouik has joined #maemo18:28
*** sLumPia has quit IRC18:28
*** Kabouik has quit IRC18:28
solofightI opened the application manager and there is a list of apps for which i need to click, open the terms agreement and the click on details, and description to read about the app.... can i see th elist of apps on a web page somewhere to quickly check them ?18:31
solofighti have installed recaller which records all calls into a folder.. the problem is those recorded voices are getting played inbetween songs with the music player18:42
solofightany ways to avoid that ?18:42
solofight3. My friends view the gallery often, how can i avoid showing private pictures as a part of the gallery ?18:43
*** sunny_s has joined #maemo18:49
Arkenoidamn. my *THIRD* n950 suffers from screen rot :-(18:49
*** fuz_ has quit IRC18:50
SpeedEvil:(18:56
*** triggerhappy has quit IRC18:57
ShadowJKscreen rot$18:57
*** solofight has quit IRC18:58
*** Agge has joined #maemo18:59
ShadowJK?19:00
*** Aggese has quit IRC19:02
DocScrutinizer05Arkenoi: how the heck did you manage to get THREE n950?19:02
DocScrutinizer05Arkenoi: I guess you found out the hard way about why N950 never went to official sales19:04
*** winocm has quit IRC19:04
ShadowJKYou know how some people create a 2partition on ssd, 10-20% size, and leave it unused, in order to improve ssd perform19:05
*** fuz_ has joined #maemo19:05
DocScrutinizer05the hinge pushing hard against LCD's back, ruining it when you're not gentle and careful when opening up the kbd19:05
ShadowJKwhy I have 2g swap on uSD19:05
*** goldkatze has quit IRC19:06
*** goldkatze has joined #maemo19:07
DocScrutinizer05ShadowJK: we discussed that recently, no? the controller on SD needs to erase pages that don't have data, so they can take new data written to them. This erasing can happen async on a background task as long as there are sufficient pages already erased so a write can copy&modify the data from one page to an empty new page19:08
DocScrutinizer05obviously that background erasing of empty pages depends on empty pages _existing_19:09
DocScrutinizer05otherwise worst case (when _all_ pages hold relevant data) the controller needs to read-modify-erase-write the same erasepage on flash19:10
DocScrutinizer05which slows down each write to ~50ms or whatever it takes to erase a page19:10
DocScrutinizer05might also be 100 or even 200ms, I can't recall19:11
ShadowJKI would say nothing happens in background on SD/emmc19:11
DocScrutinizer05while in the long run the performance on SD can't get inproved by keeping empty pages, they nevertheless make a convenient stash of instantly writable pages that can get used up by write action, before that throttling for erase kicks in19:12
DocScrutinizer05ShadowJK: I'd say when they don't do that background erasing, they are pretty dull19:13
DocScrutinizer05why wouldn't they implement flash management in eactly this way?19:14
DocScrutinizer05it seems obvious and natural19:14
ShadowJKhost devices drop voltage and clock when sd is idle, so it pretty much doesn't even run at all if the host isn't actively writing/reading, no opportunity for background tasks19:15
DocScrutinizer05I'm sure you know that it's not the writing of "1" bits that takes time, but the erasing of whole page to all-zeroes19:16
*** louisdk has joined #maemo19:16
ShadowJKSure19:16
DocScrutinizer05well, then they can't continue their background task when voltage dropped. That's pretty trivial. It however doesn't defeat the original idea19:17
ShadowJKBut the read-write-modify cycles are what hurts performance the worst19:17
DocScrutinizer05hm?19:18
DocScrutinizer05the OS may write to several sectors on same erasepage. During that the SD can write to that erasepage without erasing it, and in background can erase "unused" erasepages. The very moment the OS wants to write a 0 to a bit that previously had a 1, the controller on SD will need to do a read-modify-write action and use an erased erasepage as destination for that. Same moment the controller schedules a deferred page-erase on the original,19:21
DocScrutinizer05now unused, erasepage19:21
ShadowJKFor example, host writes a tiny file to ext3, 4k written for tiny file, 4k to journal, 4k to ext3 metadata. 3 places in different locations. The first 4k gets written to a page in a newly erased block, when the next request comes in, the SD controller has to "close" the current block, so it reads in 4M-4k from the old block, writes it to its new block, and proceeds on with the second of the 3 4k requests19:22
ShadowJKthis is what happens in practice on most SDs19:22
ShadowJKThis describes most Kingston cards actually19:23
DocScrutinizer05whatever happens in practice, the controller can always follow the "erase pages in background" policy19:23
ShadowJKwhile high-end cards are able to keep 8-12 open simultaneously19:23
ShadowJKsure, but it's the 4M read-modify-write that hurts more than erases..19:24
DocScrutinizer05I seriously doubt that19:24
DocScrutinizer05it's known that erase takes AGES19:24
DocScrutinizer05while write is relatively fast compared to that19:24
ShadowJKWell it's a factor of 2 or 319:25
*** zemmy has joined #maemo19:25
ShadowJKWhile read-modify-write write amplification is a factor 1000 speed difference19:25
DocScrutinizer05and actually it doesn't matter how much the erase hurts, since you can not compensate the one with the other. When you can avoid sync erase, you will do it19:26
DocScrutinizer05and what you tell about your read-modify-write is actually exactly the problem that write takes an erased page19:27
DocScrutinizer05otherwise you can't write modified content19:27
ShadowJKThe amount of "half erased/written" blocks a sd can keep track of is a major influence on speed19:28
DocScrutinizer05and since your unused partition reserves some erase pages for unused, it speeds up exactly those read-modify-([no]erase)-write cycles19:28
ShadowJKIt helps the kernel more really :)19:29
DocScrutinizer05there's no such thing like a half-erased erasepage19:29
*** zemmy has joined #maemo19:29
DocScrutinizer05erasing a page is a parallel process19:29
ShadowJKpartially written to19:29
*** Wizzup has quit IRC19:29
ShadowJKSome cards can deal with two sequential write streams, some can't19:30
DocScrutinizer05maybe19:30
DocScrutinizer05I don't see how that contrdict what I said19:30
ShadowJKthe net effect, however, is that non-sequential writes are horribly slow, and sequential writes are fast19:31
DocScrutinizer05and basically a controller doesn't need at all keep track of write fill of a page. It just needs to check if modify changed a bit from 1 to 0, which qualifies for a copy from old to new empty erasepage19:32
DocScrutinizer05unless you use a layer above that to tak invalid sectors and use a next sector in same hash chain that may or may not be on same erasepage. But even then the basic principle on lowest level still stays unchanged and applies19:33
DocScrutinizer05s/tak /tag /19:34
infobotDocScrutinizer05 meant: unless you use a layer above that to tag invalid sectors and use a next sector in same hash chain that may or may not be on same erasepage. But even then the basic principle on lowest level still stays unchanged and applies19:34
ShadowJKMaemo kernel's swapout tries to write sequentially, it basically checks swap area for free space, sets a pointer to start of biggest continous chunk of free space, and writes sequentially into that chunk until it reaches the end, and then looks for another free chunk. With time the data gets spread out almost uniformly across the swap area, so the free area becomes smaller and smaller. Each time crossing an erase block boundary costs anout the same in terms of19:34
ShadowJK performance penalty, and it will be doing it much more often when it only had fragmented little spaces to write into19:34
ShadowJKWith a huge swap partition but same 120-180M or so swapped out, the space in between occupied sectors grows larger, and it doesn't cross erase page boundaries as often19:35
DocScrutinizer05yes, sure19:35
DocScrutinizer05but how is that related to flash controller now?19:36
DocScrutinizer05and particularly to keeping an unused partition to speed up stuff19:37
ShadowJKIf it was as capable as a full SSD, it wouldn't matter in which order host wrote stuff19:37
DocScrutinizer05errr, mhm19:37
ShadowJKProper SSDs like Intel's 3700 reorders everything, constantly. :-)19:38
DocScrutinizer05sorry, can't parse that19:39
DocScrutinizer05aah, yep, can19:39
*** Agge has quit IRC19:40
*** Aggese has joined #maemo19:40
ShadowJKInternally they work almost like maemo swapout, except they gc when stuff gets fragmented. And like maemo kernel swapout, the less actual data, the easier time it has to gc and reorder19:40
DocScrutinizer05maybe they reorder everything, constantly. How would I even know? I however would think it would be silly to do additional writes and page erases just to reorder something19:40
DocScrutinizer05I fail to remember what we're discussing about19:41
DocScrutinizer05I suggested that a user shall split his swap into two equal halves as maemo never uses 768MB anyway19:42
DocScrutinizer05and I think that's been a correct advice19:42
*** MrPingu has joined #maemo19:42
ShadowJKI'm just saying with huge swap partition, swap fragmentation effects become minimal enough there's no noticeable benefit to swapoff/swapon19:43
DocScrutinizer05while the initial statement of this discussion seems [2013-12-30 18:05:11] <ShadowJK> You know how some people create a 2partition on ssd, 10-20% size, and leave it unused, in order to improve ssd perform19:43
DocScrutinizer05which, on a second thought, I seem to have problems to parse correctly19:44
DocScrutinizer05I don't know how they do that19:44
DocScrutinizer05but I guess it's done to give the flash controller some headroom to do page copy and background erase19:45
ShadowJKI found it interesting, that people independent of myself use the same method as I do with swap, to help the SSD controller maintain performance19:45
ShadowJKBut then, the ssd controller and the maemo kernel swapout are operating with similar constraints, so it shouldn't be too surprising that similar tricks help both19:46
ShadowJKMost USB3 flash keys illustrate the constraints beautifully. 100 megabytes/sec write speed for sequential writes, 0.01 megabyte/sec for nonsequential writing :-)19:48
DocScrutinizer05watching a defragmentizer with graphical progress display doing its job on a disl fille at 75% and a disk filled at 98% gives you an instant idea why it helps to keep some space unused on storage19:49
ShadowJK(Except "Sandisk Extreme", they put an SSD inside it, and it eats 500mA of current)19:50
DocScrutinizer05s/disl fille/disk filled/19:50
infobotDocScrutinizer05 meant: watching a defragmentizer with graphical progress display doing its job on a disk filled at 75% and a disk filled at 98% gives you an instant idea why it helps to keep some space unused on storage19:50
DocScrutinizer05(([2013-12-30 18:43:18] <ShadowJK> I'm just saying with huge swap partition, swap fragmentation effects become minimal enough there's no noticeable benefit to swapoff/swapon)) Wasn't it you who shown impressive diagrams about fragmentation kicking in after all available swap space been written once sequentially?20:00
DocScrutinizer05and those diagrams didn't look like they want to suggest that increasing swap size could fundamentally change that situation20:00
*** ssvb has joined #maemo20:01
DocScrutinizer05it just takes longer until all available swap got used one time and this swap "filled up" with random data fragments cluttered all over the place20:01
ShadowJKImagine 128M swap in use, worst case fragmentation of that 128M divided into 4k pieces and spread uniformly across the swapspace. If the swapspace is 256M big, that would mean worst case fragmentation is 4k free, 4k used, interleaved across entire device. With 1 gig swapspace, the situation improves to 4k pieces interleaved with 16k free20:03
DocScrutinizer05and when the swap daemon want to reuse a freed slot that previously been used for storing some data, this is a modify on the erasepage and thus needs a page erase and page copy20:03
ShadowJKyes20:04
ShadowJKhm, would be interesting to make the kernel provide statistics on this :-)20:04
DocScrutinizer05btw the "problem" is that SD is a persistent storage. There's basically never a situation of a "virgin" storage section, unless you use those special new commands that tell the controller to simply erase and leave *empty* the rest of the erasepage20:12
DocScrutinizer05otherwise each erased page will inherit all the obsolete garbage of its ancestor, on read-modify-copywrite20:12
DocScrutinizer05since the flash jas no idea which of the data blocks are considered "free" and which contain valid swap data20:13
DocScrutinizer05considered by the swapd for example, that is20:14
*** arcean_ is now known as arcean20:14
*** SAiF has joined #maemo20:14
DocScrutinizer05thus the whole "fragmentation" actually only happens in swapd's RAM based allocation table20:15
DocScrutinizer05while flash/SD always is "fragmented"20:15
DocScrutinizer05afer a certain timespan of usage, with sufficient IO20:16
DocScrutinizer05nd I guess not even a mkfs or other format program usually will fix that flash fragmentation, unless it's a specialized flash formatter20:16
*** zummy has joined #maemo20:17
ShadowJKFor simplicity, I always consider SD fragmented and having only one block it can erase20:18
*** sviridova has joined #maemo20:18
*** zemmy has quit IRC20:19
DocScrutinizer05I honestly wonder if writing all zeroes to flash would help20:19
*** sviridova has quit IRC20:19
DocScrutinizer05when controller is smart (and doesn't work erase-to-ones) it might actually help a lot with write speed20:20
ShadowJKI have cards where it helps20:22
ShadowJKBut the speed boost is pretty short lived20:22
ShadowJKI also have cards where writing anything at all sequentially gives boost to future random write :-)20:23
*** Aggese is now known as Agge20:23
ShadowJKThose cards typically have a small log-structured SLC cache or "journal" internally, small writes go into its cache, big writes blows away the cache and leaves it free20:25
*** bef0rd has quit IRC20:25
ShadowJKIt's obviously empty when new, and its size seems to be carefully sized to match the workload presented by running crystaldiskmark, a popular disk benchmark, with the default settings20:26
ShadowJKsecond run of crystaldiskmark gives entirely different and worse results :D20:26
*** murova has joined #maemo20:27
*** murova has quit IRC20:27
*** bef0rd has joined #maemo20:32
*** zummy has quit IRC20:34
*** LauRoman|Alt has quit IRC20:40
*** LauRoman has quit IRC20:49
MrPinguHi guys, long time no see!20:49
ShadowJKhai20:49
SpeedEvilhai20:49
MrPinguBeen a bit busy and were not able to play and use my N900, hope I haven't missed some fun20:52
MrPinguSaw that MCE is REed, anymore big news?20:54
MrPinguJust the headlines, I can find the rest myself ^^20:55
*** eMHa__ has quit IRC21:07
*** eMHa__ has joined #maemo21:08
*** eMHa__ has quit IRC21:11
* ShadowJK reads tmo threads from 201221:14
ShadowJKI guess 2010 or 2011 was last time I read tmo21:14
* sixwheeledbeast wonders why sLumPia was asking me about ereswap.21:16
DocScrutinizer05MrPingu: hi!21:18
DocScrutinizer05http://talk.maemo.org/showthread.php?t=9114221:19
DocScrutinizer05MrPingu: ^^^21:19
MrPinguThat's some interesting project.21:21
*** ssvb has quit IRC21:21
sixwheeledbeastalso #neo900 :)21:24
MrPinguSeems I need to plant another money-tree...21:24
MrPingu:P21:24
*** at1as has joined #maemo21:26
*** SAiF has quit IRC21:27
*** Timmy has joined #maemo21:27
*** ssvb has joined #maemo21:27
*** nox- has joined #maemo21:33
*** jmlich has quit IRC21:33
*** Agge has quit IRC21:36
*** Aggese has joined #maemo21:36
anYcdoes anyone else has two battery icons after the latest cssu update?21:38
*** Agge has joined #maemo21:39
*** Aggese has quit IRC21:41
MrPinguanYc, do you have advanced power installed?21:43
*** arcean has quit IRC21:45
anYchm, I think yes21:47
*** rcg has joined #maemo21:51
MrPinguUninstall that ;)21:52
MrPinguThe new batteryicon give the same information AFAIK21:52
anYcokay, I'll try that, thanks :)21:58
*** freemangordon has quit IRC21:58
*** freemangordon has joined #maemo21:59
*** messerting has joined #maemo22:00
MrPinguanYc: You're welcome22:00
*** eMHa has joined #maemo22:04
*** Timmy has quit IRC22:06
*** NIN101 has quit IRC22:09
*** valerius has quit IRC22:14
*** valeriusL has quit IRC22:14
*** arcean has joined #maemo22:16
*** int_ua has joined #maemo22:24
*** at1as has quit IRC22:24
*** NIN101 has joined #maemo22:25
*** valeriusL has joined #maemo22:27
*** shentey has joined #maemo22:29
*** valerius has joined #maemo22:29
*** int_ua has quit IRC22:33
*** Guest81604 has quit IRC22:45
*** eijk has quit IRC22:50
*** eijk has joined #maemo22:50
*** amonk has quit IRC22:51
*** Guest81604 has joined #maemo23:01
*** sequantz has quit IRC23:07
*** sunny_s has quit IRC23:08
*** amonk has joined #maemo23:09
*** Agge has quit IRC23:13
*** Aggese has joined #maemo23:13
*** mvp_ has joined #maemo23:17
*** ALoGeNo has joined #maemo23:30
*** ALoGeNo has joined #maemo23:30
*** eijk_ has joined #maemo23:39
*** eijk has quit IRC23:41
*** sequantz has joined #maemo23:42
*** OkropNick has quit IRC23:46
*** Venemo has quit IRC23:49
*** croppa has joined #maemo23:50

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