Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2012-03-01

00:07:47 Quit Strife1989 (Quit: Reboot back to Debian)
00:08:56dfktit's very nice that album art can now be a "backdrop" - but seems scrolling lines are still "eating" into the image - http://i.imgur.com/30yMf.png - is there a way around that?
00:10:14 Join enthdegree [0] (~BitchX@unaffiliated/enthdegree)
00:14:25 Join Strife89 [0] (~Strife89@207.144.201.128)
00:16:33 Quit adnap (Ping timeout: 244 seconds)
00:18:06 Join japc [0] (~japc@bl22-232-215.dsl.telepac.pt)
00:24:32 Join adnap [0] (~adnap@rrcs-71-42-140-57.sw.biz.rr.com)
00:25:04 Join evilnick [0] (~AndChat@rockbox/staff/evilnick)
00:26:06 Join rarog [0] (~chatzilla@p4FF5D3B8.dip.t-dialin.net)
00:26:32rarogis rockbox site down currently or is it just me?
00:28:41Tornelooks broken to me
00:30:23 Quit adnap (Ping timeout: 244 seconds)
00:30:45rarogdamn.... just wanted to update information on a bug...
00:30:45rarog:D
00:30:51rarogwill have to wait till morning then.
00:30:53evilnickYeah. It's down: http://www.downforeveryoneorjustme.com/rockbox.org
00:31:12 Join adnap [0] (~adnap@rrcs-71-42-140-57.sw.biz.rr.com)
00:33:58 Join pamaury [0] (~quassel@vit94-1-82-67-248-70.fbx.proxad.net)
00:34:05 Quit pamaury (Changing host)
00:34:05 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
00:35:51rarogpamaury: I think I know, why Windows sees only one drive with Fuze+
00:36:13 Quit lebellium (Quit: ChatZilla 0.9.88 [Firefox 11.0/20120222074758])
00:36:52rarogI checked lsusb -v on linux and saw that OF sends two configurations (bNumConfigurations =2 ...) while RB sends only one.
00:37:03rarogI think that this is the point.
00:37:12gevaertsno
00:37:30Torneonly one configuratio nis used at a time
00:37:31gevaertsbNumConfigurations has nothing to do with the number of drives you see
00:37:36Tornethe host selects the one it wants to use and uses it
00:38:46rarogHm... At what point then can I see with lsusb that two devices are reported to the system?
00:39:48Tornethere are not two devices
00:39:56rarog*two drives
00:40:03gevaertsYou can't. lsusb only reports standard descriptors. the number of LUNs isn't reported that way
00:41:13 Join jlbiasini [0] (~metaphys@d86-32-96-55.cust.tele2.at)
00:41:39rarogHm... ok "cat /proc/scsi/scsi" shows it at least...
00:43:11rarogThe only difference I see then, is that rockbox tells different Models for internal and sd card
00:43:12 Part jlbiasini
00:43:28rarogwhile OF tells 2x the same model
00:49:41rarogI'll change it to identical ids and will test it tomorrow...
00:50:15rarogon a windows computer
00:50:15 Quit rarog (Quit: ChatZilla 0.9.88 [SeaMonkey 2.4.1/20111011102430])
00:50:15 Quit bertrik (Ping timeout: 252 seconds)
00:50:33gevaertsThat's not the issue either. Different names are quite common and work with other rockbox players
01:00
01:00:50KiwiCamHi. Where do I find a breakdown of all the latest Skinnning engine features?
01:02:28dfktKiwiCam, i read the recent activity and customwps pages
01:03:57dfktalso, take a look at gevaerts' latest themes - they're a nice showcase of the new features
01:04:56gevaertsWell, not entirely. They may confuse people by trying to abuse lots of viewports for other stuff at the same time
01:05:16KiwiCamThat's actually what brought me here. I'm looking at a printout of the WPS. What's the difference between %pb and %px?
01:06:44KiwiCamI've somehow missed %px previously. It's been there for a long time.
01:07:22gevaerts%pb draws a progress bar. %px is the percentage played
01:08:10JdGordongevaerts: for your wavy theme... can you add the peak bars as two more vertical wavy fll in things?
01:08:14KiwiCamThat solves a problem for me. That explains why I'm getting no value (number result) for %pb. I should be using %ps.
01:08:55gevaertsJdGordon: that should work (I think... not sure about horizontal overlap). I'll have a go tomorrow
01:09:09JdGordonshould work the same way as your progress bar
01:09:53JdGordonI don't think i've seen any themes which use the split peak meter tags
01:10:48KiwiCamJdGordon: Doesn't this one? http://themes.rockbox.org/index.php?themeid=1486
01:11:00gevaertsThe thing is that due to the curve the width of the bitmap is a lot more than the width of the actual line
01:11:45JdGordonKiwiCam: does it?
01:13:34 Join CaptainKewl [0] (captainkew@207-237-110-248.c3-0.nyr-ubr2.nyr.ny.cable.rcn.com)
01:16:30KiwiCamJdGordon: I gather from you reaction it doesn't.
01:19:43dfktsplit peak meters means %pL and %pR?
01:20:35 Quit evilnick (Quit: Bye)
01:21:21JdGordonyes
01:21:27dfkti used that in a few themes
01:23:53KiwiCamI thought he had. My mistake.
01:25:18dfkt%VB only works with %V(...), not with %Vl(a,...), right?
01:25:33JdGordonno
01:31:23 Join Strife1989 [0] (~Strife89@207-144-19-39.cstel.net)
01:32:05KiwiCamActually. Quite a few of the Clip themes are using %pL and %pR, including the one I linked to above. I've got it one one I haven't uploaded yet. It's the Equaliser values, that I'd like to be able to use most. I've now printed out the latest CustomWPS, so I'll have a good read of that. Later folks.
01:33:18***Saving seen data "./dancer.seen"
01:34:41 Quit Strife89|AndChat (Ping timeout: 276 seconds)
01:38:01 Join Strife89|AndChat [0] (~Strife89@207.144.201.128)
01:38:06 Quit Strife89|AndChat (Client Quit)
01:40:26 Quit Strife1989 (Ping timeout: 252 seconds)
01:42:46 Quit dionoea (Ping timeout: 272 seconds)
01:44:20dfktthis is weird - when i draw the album art in the foreground, it doesn't do transparency in the peak/progess bars right - http://i.imgur.com/BoUtX.png - when i draw it in the backdrop layer, the transparency is right, but the rest isn't visible - http://i.imgur.com/XmxRx.png
01:45:59 Quit Utchy (Ping timeout: 240 seconds)
01:46:02 Quit perrikwp_ (Read error: Connection reset by peer)
01:47:15 Join perrikwp [0] (~quassel@cpe-024-163-024-033.triad.res.rr.com)
01:51:13 Join dionoea [0] (~dionoea@yop.chewa.net)
01:52:51dfkthah, peak/progress viewports have to span the whole screen width for album art in the backdrop to be visible
01:56:06 Quit zu (Ping timeout: 240 seconds)
01:57:00 Join zu [0] (~zu@ks387228.kimsufi.com)
01:58:11 Join Utchy [0] (~Utchy@rps6752.ovh.net)
02:00
02:09:46 Nick guymann_ is now known as guymann (~charles@69.182.30.34)
02:13:54 Quit Thra11 (Read error: Operation timed out)
02:15:36JdGordondfkt: gevaerts: using the 2 layers in the skins will take some getting used to.. you need to make sure both layers have a viewport covering the whole screen or it likely wont work
02:15:47JdGordoni.e both need a %V(0,0,-,-,-) viewport
02:16:05CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
02:16:05*gevaerts nods
02:16:06JdGordonand draw order probably matters if you want it to be perfect
02:17:23gevaertsJdGordon: the wavy theme has some bits being cleared in the list views on positions where an image would have been drawn if a conditional would have said something else than it does. The thing is that this "wave" effect uses lots and lots of viewports and conditionals to only draw in some, which arguably is asking for trouble
02:17:59gevaertsAnd as far as I can figure out you can't really conditionally load specific list item viewports
02:18:21gevaerts(which would be the clean way to do this)
02:19:04dfktJdGordon, thanks - i try that instead of making each used viewport full-sized (which won't work in all situations anyways)
02:19:06gevaertsOh, also, I can't get scrolling to work properly on skinned lists
02:19:42JdGordonwhy cant you conditionally chose the list viewport?
02:19:57JdGordonand scrolling was working fine eons ago when i was playing
02:20:17gevaertsMaybe the scrolling is another effect of me doing strange things
02:21:47gevaertsThe problem (as I see it) is that only viewports with the label specified by %Lb() get moved around to the right spot on the screen, so if you %Vl a viewport from a conditional in there, it will be in the wrong place
02:21:54gevaertsI admit that I haven't tried it
02:22:35*JdGordon cant really remember how it works
02:24:04gevaertsRight now I can't think of a good solution for this that doesn't involve hierarchical viewports (i.e. where viewport coordinates are relative to their parent), but I don't think we should go there
02:28:20JdGordoni thought thats how i did list viewports?
02:28:33*JdGordon really ddoesnt rememebr
02:29:27gevaertsFor each item it loops through all viewports, and if it sees one with the right label it (temporarily) changes its x and y coordinates to make it appear in the proper place and then renders it
02:30:42gevaertsAt least that's how I read the code
02:31:09gevaertsI'm not convinced that this is something that theme designers will actually have problems with though. The "wavy" design is basically driven by my curiosity to find out what I can get away with, which really isn't representative of what the skin engine should support
02:31:25JdGordon:)
02:33:05gevaertsI'd just like to figure out why exactly it has glitches. This may well involve lots of stepping through it in gdb :)
02:34:33JdGordonwith 50 viewports, thats gonna be fun!
02:34:33gevaertsOh, one more thing: I installed wavy and medieval on real hardware today, and it appears that skinned lists aren't initialised the very first time the menu is rendered after USB disconnect, so it shows the standard list until you touch a button
02:34:55JdGordonah, good bug, file it :)
02:35:10gevaertsTomorrow. It's getting late here :)
02:39:24gevaertsJdGordon: is there a debug option or something like that that makes the skin engine call lcd_update() more or less after every tag?
02:45:14JdGordonno
02:45:57JdGordonyou can add that at the end of skin_render_line() but it is insanely slow
02:46:33 Quit factor (Read error: Connection reset by peer)
03:00
03:03:33dfkti'd love to have album art size limit that considers the larger side of the screen - ie. on a 160x128 screen i could load 160x160 album art and 'crop' to a visible 160x128 portion (instead of being limited to 128x128)
03:03:41 Join factor [0] (~factor@r74-195-184-248.msk1cmtc01.mskgok.ok.dh.suddenlink.net)
03:09:40funmani wonder if i should rather ask the fund to buy me a nano2g from amazon
03:13:01 Nick scorche` is now known as scorche (~scorche@rockbox/administrator/scorche)
03:25:06 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
03:31:48 Quit pamaury (Ping timeout: 260 seconds)
03:33:00enthdegreeheggel geggel
03:33:08enthdegreefunman, don't. there are better players for the money
03:33:22***Saving seen data "./dancer.seen"
03:33:25enthdegreelike the Sansa Clip Zip
03:33:26funmanenthdegree: i have better players already, i need a nano2g to work on
03:33:36enthdegreeoh, for that reason.
03:34:20enthdegreeoh heavens you're a developer, I thought you were just an enduser.
03:35:32funmani also have had all the sansa targets in hand and most in my possession :)
03:35:44enthdegreewhoa!
03:36:18enthdegreewould anyone use a Ti-83/+/84+ emulator?
03:40:16enthdegreeThe only issue I can see with one is the amount of memory available in these devices
03:40:36 Quit japc (Quit: Ex-Chat)
03:40:36funmandoes it have audio output ?
03:40:58enthdegreewhat the calculator?
03:41:12funmanyeah
03:41:30enthdegreeI bet someone has hacked it out but IDK if you would want to
03:41:51funmanthen it's probably unrelated talk for #rockbox , move that to #rockbox-community please
03:42:51enthdegreewhat? I am speaking of a calculator emulator plugin for rockbox, not a port to a calculator
03:45:03enthdegreeis that still for #rockbox-community
03:45:42funmanooh sorry i didn't get it, yes it's on topic
03:46:11funmanof course it's interesting
03:47:03enthdegreeit would be amazing to be able to use that as an alternative to the crummy builtin calculator
03:48:46funmansome rockbox targets have 64MB of ram
03:50:28enthdegreeoh that shouldn't be an issue then; the calculators only have 128kb RAM and a 15MHz Z80
03:54:05enthdegreethe only real issue is text entry
03:54:56enthdegree96×64px screen is a joke
03:55:48 Quit Strife89 (Quit: Night, all!)
03:59:25 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
04:00
04:00:47JdGordonhey [Saint]
04:01:40JdGordonhttp://www.rockbox.org/irc/log-20120229#14:40:23
04:02:12 Join Riviera [0] (~Riviera@92.51.147.16)
04:02:27[Saint]Yeah, I vaguely recall reading some highlights before I fell asleep last night.
04:02:46[Saint]I'll be nosey after ingesting coffee :)
04:04:04JdGordonhehe
04:04:24 Quit remlap (Read error: Connection reset by peer)
04:06:11funman[Saint]: still no nano2g :(
04:12:27 Join remlap [0] (~Patrick@190.28.169.217.in-addr.arpa)
04:16:30 Join Rower85 [0] (husvagn@v-413-alfarv-90.bitnet.nu)
04:18:33 Quit amiconn (Disconnected by services)
04:18:34 Quit pixelma (Disconnected by services)
04:18:34 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:18:36 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:18:38 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:18:56 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:27:09 Quit DerPapst (Quit: Leaving.)
04:28:21 Quit dys (Ping timeout: 265 seconds)
04:29:04 Join dys [0] (~andreas@krlh-5f71c509.pool.mediaWays.net)
04:32:52 Join LambdaCalculus37 [0] (~rmenes@rockbox/staff/LambdaCalculus37)
04:36:19 Quit user890104 (Ping timeout: 272 seconds)
04:36:22 Quit [Saint] (Ping timeout: 260 seconds)
04:39:42 Join user890104 [0] (venci@Addicted.to.Minecraft.ipv6.6bez10.info)
04:40:36 Quit LambdaCalculus37 (Quit: FWUMP)
04:53:18 Quit anewuser_ (Read error: Connection reset by peer)
04:53:28 Join [Saint] [0] (~Saint]@101.98.158.103)
04:53:29 Quit [Saint] (Changing host)
04:53:29 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
04:55:26 Join passstab [0] (~v@c-68-80-37-73.hsd1.pa.comcast.net)
04:59:48 Quit TheSeven (Disconnected by services)
05:00
05:00:11 Join [7] [0] (~TheSeven@rockbox/developer/TheSeven)
05:15:56 Join Rob2223 [0] (~Miranda@p4FFD1906.dip.t-dialin.net)
05:17:53 Quit Rob2222 (Ping timeout: 252 seconds)
05:19:29 Quit passstab (Remote host closed the connection)
05:21:30 Join Minthe [0] (~Minthe@133.11.50.113)
05:33:24***Saving seen data "./dancer.seen"
05:50:48 Quit zenlunatic (Ping timeout: 276 seconds)
06:00
06:02:02 Quit Scromple (Read error: Connection reset by peer)
06:03:45 Join zenlunatic [0] (~justin@c-68-48-40-231.hsd1.md.comcast.net)
06:12:07 Quit Zoiah (Ping timeout: 240 seconds)
06:12:17 Join Zoiah [0] (zoiah@matryoshka.zoiah.net)
06:26:06 Quit curtism (Quit: Live Long and Prosper)
06:52:31pixelma[03:03:36] <dfkt> i'd love to have album art size limit that considers the larger side of the screen - ie. on a 160x128 screen i could load 160x160 album art and 'crop' to a visible 160x128 portion (instead of being limited to 128x128) <- I remember asking about that a loong time ago
06:54:46[Saint]Shouldn't that be what the "c" flags do?
06:55:47pixelmano, you are limited to LCD_WIDTH or HEIGHT - whichever is smaller
06:57:41pixelma[Saint]: and what are the "c" flags again?
06:59:24 Join hiptobecubic [0] (~john@unaffiliated/hiptobecubic)
07:00
07:00:01[Saint]they're supposed to center the art in the AA viewport. Exabple: "%Cl(0,0,100,100,c,c)"
07:01:00pixelmathat's not what we are talking about
07:01:25JdGordon[Saint]: any thought of the draw_rect tag?
07:01:50[Saint]pixelma: I thought it could be used in that way.
07:02:07[Saint]I thought it would crop if it needed to.
07:02:25pixelmathat would just mean that you can have a 128x128 album art centered (as currently the maximum limit), not a 160x160 one in the example
07:02:48pixelmaI think trying it out explains it better
07:04:38pixelmaI *believe* the tag doesn't crop, just aligns. At least in my c200 sbs I use the viewport for cropping (which works). That's a bit of a different story though
07:11:42 Quit nosa-j (Read error: Connection reset by peer)
07:12:13 Quit hiptobecubic (Ping timeout: 245 seconds)
07:12:20 Join nosa-j [0] (~m00k@adsl-74-235-42-18.clt.bellsouth.net)
07:33:27***Saving seen data "./dancer.seen"
07:34:05pixelmamaybe there was a way to get a fullscreen art in this case but only with stretching
07:37:24 Quit zenlunatic (Ping timeout: 276 seconds)
08:00
08:10:07 Quit jhMikeS ()
08:19:18 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
08:20:13 Join jhMikeS [0] (~jethead71@c-68-61-166-99.hsd1.mi.comcast.net)
08:20:13 Quit jhMikeS (Changing host)
08:20:13 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
08:25:14 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:37:49 Join ender` [0] (~ender@foo.eternallybored.org)
08:38:58[Saint]kugel: JdGordon: https://github.com/saint-lascivious/RaaA-CabbieV2/commits/master
08:39:08JdGordonThank You!
08:39:42 Quit enthdegree (Read error: Connection reset by peer)
08:40:12JdGordonthough i'm more interested in the draw rect skin tag right now :)
08:41:24 Join mortalis [0] (~mortalis@77.108.98.176)
08:42:11[Saint]I'll add a note to the Flyspray task reminding those using it its good and dead now, and that I'll still be doing periodic releases via DropBox and Google+ from now on then I'll poke someone to close the task.
08:42:47JdGordonthis is still quite pointless unless you're trying to get it actually commited
08:43:07JdGordonwhich I know is the ultimate goal, but im meaning "soon"
08:45:51[Saint]Well "soon" really depends on the amount of on-device testing I can muster, and the willingness of others to allow it to be committed, combined with the amount of things I'll have to piss around changing to facilitate said willingness.
08:49:02[Saint]I'm also told I need to do a "get you grievences out in the open" mail. (highly paraphrased).
08:50:16JdGordonarent there bigger issues like the convttf being broken? and wpsinstall.pl being broekn?
08:52:15[Saint]From my point of view, those are the bigger issues...but I get the feeling those aren't the issues people want to talk about.
08:56:31[Saint]wpsinstall isn't exactly "broken"...it just needs to know about application target vs normal target iiuc, and convttf being broken is still an issue but notsomuch if we did fonts in a saner way for building themes.
08:57:17[Saint]"sane" being "grab pre-converted fonts from a central directory as opposed to converting them when needed"
08:58:35GodEater_what's the flyspray # for the broken convttf issue?
08:59:41[Saint]It needs to be re-opened, it was a known issue when it was committed. Mentioned in the comments in there a few times.
09:00
09:01:43[Saint]I can search it out in a bit when I've finished messing around home. Though it probably is best to open a new task about it instead of opening the old one again as the old one is massive.
09:02:51 Quit bertrik (Ping timeout: 246 seconds)
09:03:07 Join pamaury [0] (~quassel@vit94-1-82-67-248-70.fbx.proxad.net)
09:03:08 Quit pamaury (Changing host)
09:03:08 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:03:34 Quit factor (Read error: Connection reset by peer)
09:05:44[Saint]I need to sit down again and force myself to use convttf again to remember what exactly all the issues with it are, but there was definitely more than one. Main issue was to with with a flag to set a maximum height and it not respecting that when ascent or descent was added or removed.
09:06:32[Saint]And the fact that ascent and descent were usually screwy in the first palce and needed manual trimming to begin with is an issue I suppose.
09:09:00[Saint]Creating a consistent looking font set with it was a nightmare, pretty much every font needed me to overview it in the terminal during conversion and adjust size/ascent/descent/ manually.
09:19:30 Quit pamaury (Ping timeout: 252 seconds)
09:19:51 Join factor [0] (~factor@r74-195-184-248.msk1cmtc01.mskgok.ok.dh.suddenlink.net)
09:21:21 Join randumbum [0] (~IRCclient@173-8-222-138-Oregon.hfc.comcastbusiness.net)
09:26:54 Join icarusfactor [0] (~factor@r74-195-184-248.msk1cmtc01.mskgok.ok.dh.suddenlink.net)
09:27:13 Quit factor (Disconnected by services)
09:27:16 Nick icarusfactor is now known as factor (~factor@r74-195-184-248.msk1cmtc01.mskgok.ok.dh.suddenlink.net)
09:32:36 Quit kadoban (Read error: Operation timed out)
09:33:31***Saving seen data "./dancer.seen"
09:44:06 Quit nosa-j (Read error: Connection reset by peer)
09:44:51 Join nosa-j [0] (~m00k@adsl-74-235-42-18.clt.bellsouth.net)
09:52:35 Quit Zarggg (Quit: Rebooting client...)
09:58:51 Quit CaptainKewl (Ping timeout: 244 seconds)
10:00
10:01:00 Quit Minthe (Remote host closed the connection)
10:17:42 Join pamaury [0] (~quassel@sphinx.lix.polytechnique.fr)
10:17:42 Quit pamaury (Changing host)
10:17:42 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:17:57 Quit mystica555_ (Ping timeout: 276 seconds)
10:18:05 Join wodz [0] (~wodz@89-76-160-35.dynamic.chello.pl)
10:21:17wodzamiconn, others: http://www.pastie.org/3494990 . This is generalized (to some degree) 4bit lut version of crc routine.
10:22:05wodzwith polynomial=0x04C11DB7, reflected=false it is equivalent to crc_32() from firmware/common/crc32.c
10:23:25wodzwith polynomial=0x04C11DB7, reflected=true it is equivalent to chksum_crc32() from firmware/common/crc32-mi4.c (with the exception that it uses 64bytes lut while crc32-mi4 uses 1k)
10:24:21wodzwith polynomial=0x04C10DB7, reflected=false it can calc checksum for RKW file format
10:24:22 Quit mortalis (Read error: Connection reset by peer)
10:26:44wodzAs crc_32() variant is the most frequently used we could optimize this a bit by declaring LUT static with prefilled values and add check whether we need to regenerate.
10:46:45 Join mortalis [0] (~mortalis@77.108.98.176)
11:00
11:07:46amiconnWhy not just include the needed lut(s) per target? 4 bit lut is 64 bytes, certainly less than the generator code (and faster)
11:08:35amiconnOn large mem targets we could use 8 bit lut(s) throughout.
11:16:00Zagorwhen did we last to a GPL version survey? do we have any v2-only code left?
11:16:16Zagormy grep fu is failing me
11:20:18amiconnIn the calculation part, the if(reflected) should be pulled out of the loop for performance. This makes it easier to ifdef it away as well, for targets where it's not needed (the parameter can be dropped using macro magic so that the crc calls don't need ifdefs)
11:26:19 Join hiptobecubic [0] (~john@unaffiliated/hiptobecubic)
11:26:51 Join wodz_ [0] (~wodz@iwl138.internetdsl.tpnet.pl)
11:28:13wodz_amiconn: Could you be a bit more specific? I fail to see how to do this with macros. And as you can see the difference is not only in LUTs
11:32:10wodz_and we can't decide at compile time which version to use as we need two in some cases
11:32:32amiconnDefine "some cases"
11:32:46amiconnIf "some cases" == "some targets" then sure we can
11:33:35***Saving seen data "./dancer.seen"
11:33:56amiconnDon't include code for a target were we know at compile time it won't be used.
11:35:04wodz_ok, so your opinion is that when I'll add RKW format support to rolo I should add another crc implementation conditionally compiled, right?
11:37:49amiconnYou don't need a completely separate implementation, it can be done by ifdefing a generic one
11:38:31wodz_that would be nightmare
11:38:42amiconnI doubt that
11:38:45wodz_prototypes will differ
11:39:05amiconnYes, so what? Just encapsulate the function call in a macro
11:39:17wodz_that what I call nightmare
11:39:32amiconnIt's just a bit of ifdefing in the header.
11:40:57amiconnOf course there is a limit in what would be acceptable to exclude unused code per target, e.g. if it requires ifdefing each call of a given function
11:41:19wodz_thats the point
11:41:41wodz_we call crc_32() in ~15 places
11:41:48 Join DerPapst [0] (~Adium@91-64-115-27-dynip.superkabel.de)
11:41:59wodz_all this calls would need to be mangled by preprocessor
11:42:25amiconnBut in this case, for example: #ifdef NEED_CRC32_R / #define CRC32(s, l, p) crc32_generic(s, l, p, false) / #define CRC32_R(s, l, p) crc32_generic(s, l, p, true) / #else / #define CRC32(s, l, p) crc32_generic(s, l, p) / #endif
11:42:49amiconnSimilar stuff can be done for the other parameters. In the caller you'd then just call the macro
11:43:00amiconn(s)
11:46:03wodz_in such approach it would be better to drop gen_tab part and include precalculated lut(s) for known variants
11:46:38amiconnIn this case it's actually pretty easy because one caller alway needs one variant, so you can ifdef out constants that way
11:48:29amiconnAnother method (if you don't want different macro names per variant) would be having one single macro and then just drop the parameters you don't need. But this is error prone, in that you rely on the macro user to know what he's doing (i.e. that some parameters might be dropped depending on target)
11:49:26wodz_different macro names are better IMO as not hiding true nature of what being called
11:49:54amiconnI'm all for dropping the gen_tab part as I already said. Sure it would save some binsize if a target needs many variants, or if we used 8 bit luts on large mem targets, but then a point of a large lut is to speed things up, and on large mem targets we can afford that, imo
11:50:58wodz_is it really that important? The difference in speed between 4bit LUT and 8bit LUT is ~2x in terms of speed
11:52:16amiconnDepends on how often it is called, or rather, for what amount of data
11:52:55amiconnIf going 8 bit means e.g. (re)booting 1 second faster on a large mem target it's worth it imo
11:54:43wodz_if by rebooting you mean roloing I dissagree.
11:55:00wodz_It is relatively rarely used feature
11:55:04amiconnAlso the speedup could easily be more than 2x, depending on the processor architecture. Somewhat educated guess is that this applies to both SH1 (there's no large mem target using that of course) and coldfire (and those are all large mem)
11:55:23amiconnIirc crc32 is also used in the database code
11:55:34wodz_in tagcache yes
11:56:21wodz_coldfire suffers from lack of dcache in 4bit lut version
11:57:31wodz_I guess it was somewhat mitigated in crc32.c by copying processed byte to local variable (which will be in iram) at the cost of additional mem access
11:59:50wodz_Anyway, I don't think this is THAT important for crc implementation to be as fast as possible in our case.
12:00
12:00:28amiconnI wasn't even thinking of the dcache issue. Shift-by-8 and or'ing can be done in one instruction on cf, for 4 bit you need to shift, mask and or()
12:01:51wodz_this is strictly academic discussion as I am not so sure gcc is smart enough to catch such nuance.
12:02:16amiconnBy (re)booting I mean all cases. Explicit rolo might be rarely used, but you need to boot the device every time you start unsing it. Also some operations require a reboot. The reasons for this became less, but afaik it's still necessary for database commit when not using dircache
12:03:47wodz_amiconn: Do we use crc check in booting routine? common.c from bootloader/ supports only simple checksum
12:05:36amiconnYou mentioned mi4 using crc
12:06:01wodz_but I think we relay on stock bootloader in this case.
12:07:20wodz_we calculate crc in rolo when running mi4 firmware
12:09:04wodz_hmm, no we use this in bootloader also
12:09:25wodz_http://git.rockbox.org/?p=rockbox.git;a=blob;f=bootloader/main-pp.c;h=28aee9cf90b6138415e484ccdf401b9b4166dc2c;hb=HEAD in load_mi4()
12:11:32 Quit wodz_ (Quit: Leaving)
12:15:17 Quit Keripo (Quit: Leaving.)
12:29:57 Quit DerPapst (Quit: Leaving.)
12:49:38wodzAfter thinking more about it I think I'll leave this all as is.
12:50:26 Join Topy [0] (~Topy44@f049045242.adsl.alicedsl.de)
12:54:36 Quit T44 (Ping timeout: 276 seconds)
12:59:36 Join anewuser [0] (~anewuser@190.207.129.184)
12:59:37 Quit anewuser (Changing host)
12:59:37 Join anewuser [0] (~anewuser@unaffiliated/anewuser)
13:00
13:05:15 Join dfkt [0] (dfkt@unaffiliated/dfkt)
13:11:39 Quit [Saint] (Read error: Connection reset by peer)
13:14:25 Quit hiptobecubic (Ping timeout: 252 seconds)
13:24:51 Join jlbiasini [0] (~metaphys@d86-32-96-55.cust.tele2.at)
13:32:14jlbiasinipamaury: I think that we should update the bootloader on the server before everyone start to brick its player installing the old version with rbutil!! Not being able anymore to boot into OF!!
13:32:35pamauryis the current version broken ?
13:33:09jlbiasiniyes the problem I add was by using rbutil installation
13:33:25jlbiasiniI couldn't boot anymore into OF
13:33:32pamaurythat looks strange, it has worked for some time now
13:33:38***Saving seen data "./dancer.seen"
13:33:57jlbiasinihum, let me try again to be sure
13:34:05pamauryare you sure you pressed power long enough with volume down ?
13:35:21jlbiasiniI suppose but I will try again the whole process. If that really breaks the bootloader it is worthy to check it before every one start to be mad at us! :D
13:38:32 Part Zagor
13:42:52 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
13:57:50jlbiasinipamaury: there is something new! with the actual Bootloader I can boot into OF but it doesn't go into update even if there is a firmware.sb on the root (I tested both with original firmware.sb and one from rockbox utility
13:58:08jlbiasiniactual bootloader=compiled from git
13:59:25pamauryhuh ? nothing changed, make sure to properly reboot and shutdown the player with rockbox
14:00
14:00:04jlbiasinithat's what i did, the firmware.sb file remains on the root
14:00:56jlbiasinithis just doesn't make sense
14:01:41pamauryremove it and retry, that probably an OF bug or the firmware.sb is incorrect
14:01:41 Join hiptobecubic [0] (~john@unaffiliated/hiptobecubic)
14:02:04jlbiasinii'm on it I'll let you know
14:04:17 Join Thra11 [0] (~thrall@87.114.84.113)
14:07:46jlbiasinipamaury: is it normal if I get a panic on unpluging after the dd command?
14:07:55pamauryyes
14:08:20pamaurywell, might be
14:08:46 Quit anewuser (Ping timeout: 245 seconds)
14:10:25pamaurydamn, why do we still compile tools with -pedantic
14:16:25jlbiasinipamaury: very strange I did the whole manual flashing again and had the same issue
14:16:56pamaurywhich issue ?
14:16:58jlbiasinithen I transfered the file from OF usb mode and this time it worked!!! ???/
14:17:55 Join anewuser [0] (~anewuser@190.207.129.184)
14:17:55 Quit anewuser (Changing host)
14:17:55 Join anewuser [0] (~anewuser@unaffiliated/anewuser)
14:18:24jlbiasiniAnd as pluged the player in OF usb mode the former firmware.sb file was still there!!!
14:18:51 Join petur [0] (~petur@rockbox/developer/petur)
14:19:23 Quit funman (Read error: Operation timed out)
14:19:39 Quit jlbiasini (Remote host closed the connection)
14:22:12 Join funman [0] (~fun@rockbox/developer/funman)
14:22:37 Quit anewuser (Ping timeout: 260 seconds)
14:23:41pamauryWho knows configure well ? I am wrong in saying that $toolset is actually never used
14:25:24 Join anewuser [0] (~anewuser@190.207.129.184)
14:25:24wodzAm I right that rolo is not going to work on targets without iram?
14:25:24 Quit anewuser (Changing host)
14:25:24 Join anewuser [0] (~anewuser@unaffiliated/anewuser)
14:27:12pamaurywodz: technically, I think is wrong. You can make it work as long as you are able to have the loading image disjoint from the running one
14:28:06pamaurythis might require that the loaded image move itself after been loaded
14:28:09wodzyeah, but in case of rb this doesn't hold.
14:28:21 Quit funman (Ping timeout: 246 seconds)
14:28:29 Join funman [0] (~fun@rockbox/developer/funman)
14:28:47pamaurywodz: depends. If the fuze+ for example, rolo loads it to the audio buffer
14:28:52pamaury*On
14:29:00pamauryI think
14:29:16wodzpamaury: on every target yes, but look what rolo_restart is doing
14:29:20 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
14:29:59wodzif rolo_restart() isn't in distinct space it will self overwrite
14:30:20pamauryyou are free to do differently
14:32:00wodzimo the most flexible solution would be reloc rolo_restart() so it wan't be overwritten for sure
14:32:03pamaurythe recent move to core_alloc makes it tricky though. But if you arrange so that loadaddress is far away from the code, that's ok
14:32:22wodzno it is not
14:32:34pamaury?
14:32:38wodzyou are not jumping into the code in filebuf
14:33:05wodzrolo_restart() first copy this code to the loadaddress
14:33:11pamauryyes, that's what I said
14:33:26pamauryif you set loadaddress so that it's not overwriting the code, it's ok
14:33:46pamauryof course, the code will then have to move itself back to the right location
14:34:09 Join jlbiasini [0] (~metaphys@d86-32-96-55.cust.tele2.at)
14:34:14wodzbut you usually want to load binary with the same load/run address as current one
14:34:22 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
14:34:29pamaurywodz: no
14:34:40wodzpamaury:?
14:34:51pamauryat least on the fuze+, the binary can be loaded at any address, it will just move itself to the right one at runtime
14:35:01pamaurythat solves the whole problem
14:35:07wodzby means of MMU?
14:35:13pamauryno, by mean of copying
14:35:49wodzyou check in crt0.S where are you or what?
14:36:33pamauryyes
14:36:57wodzis it in git so I could look at it?
14:37:15pamauryyes
14:37:20pamaurytarget/arm/imx233/crt0.S
14:37:23 Quit anewuser (Read error: Connection reset by peer)
14:37:47wodzhmm, but then you will need to hack rolo_restart() to simply jump and not copy
14:38:23pamauryno, just let it copy to loadaddress and set loadaddress to some place in the plugin buffer for example
14:39:05pamauryin the case of the fuze+, it's overwriting itself because it has iram but you just need to find a good location
14:39:30pamauryof course that get's tricky if you don't have enough DRAM to hold to copies
14:39:35wodzisn't it better to just skip the copying?
14:40:12pamauryI guess you can make it conditional if you want
14:40:29wodzI mean in rolo_restart()
14:40:46wodzthats really neat idea
14:40:54pamauryyes, just add #ifdef to skip the copy on your target
14:41:38pamauryor just use the plugin buffer if it's large enough, that will avoid conditional code
14:42:04wodzI don't get it
14:42:59pamauryyou want to skip the copy because you don't want the copy to overwrite the running code. I'm telling you that the simplest solution is to copy it to some safe location. The plugin buffer is a safe location
14:43:15wodzah ok
14:43:38pamaurythen the binary moves itself and your done. That's not the most efficient because you do two copies, admitted.
14:43:43wodzwhats wrong with alloc from bufflib?
14:45:02pamaurynothing, it's just that before, the code used to load the firmware to the audio buffer, so by setting loadadress=audio_buffer, you could bypass the copy if you wanted, iirc
14:46:29wodzok
14:48:02wodzpamaury: why _dramcopystart and _dramcopyend are defined in its own sections?
14:48:52jlbiasinipamaury: ok with a fresh install and bootloader from server i can boot into OF... I really wonder where all those problem where coming from.1) Anyway the actual bootloader compiled from up to date git get updated only if installed from OF usb.
14:49:22pamaury_dramcopystart could be put in .text, it's just to make it more obvious that it's the start of DRAM. Same thing for the end which could be put in .stack
14:49:24pamaurywodz: ^
14:50:03wodzok so purely aesthetic reasons :-)
14:50:07pamauryyes :)
14:50:10pamauryjlbiasini: I don't understand, why is 1) a problem ?
14:51:08wodzpamaury: thanks for inspiration
14:52:02jlbiasiniit isn't but its just something to know! If people want to update their device from an already rboxed device they could this issue. And this is really really weird!
14:54:05jlbiasininot further problem with the actual bootloader then
14:54:39pamauryjlbiasini: I don't understand you mean that putting firmware.sb with rockbox and then booting in the OF will not work ?
14:54:49jlbiasiniexactly
14:55:00jlbiasinithe of boot normaly
14:55:10jlbiasinibut the file is there!!!
14:55:15jlbiasiniI tested
14:55:17pamauryah, well, another OF stupidity
14:55:30 Quit hiptobecubic (Ping timeout: 276 seconds)
14:55:49jlbiasinibut I don't remember this was like that before (well not 100% sure)
14:59:35pamauryI don't remember too, I think I always tested with the OF
15:00
15:02:40 Join hiptobecubic [0] (~john@unaffiliated/hiptobecubic)
15:13:17jlbiasiniIt might have always been like that, we didn't changed the bootloader for a while and there used to be no rockbox usb
15:15:58pamauryusb has been there from the beginning
15:16:38pamauryfunman, wodz: any idea why configure sets up LD as arm-elf-eabi-*ar*
15:17:30wodzeee what?
15:18:04pamauryhttp://git.rockbox.org/?p=rockbox.git;a=blob;f=tools/configure;h=ee76ed73e261ade1bd43934b6ea6e5fcbff36332;hb=HEAD#l3648
15:19:34 Quit funman (Ping timeout: 252 seconds)
15:19:57wodzI guess this is pure error
15:20:30 Quit wodz (Quit: Leaving)
15:21:15 Join funman [0] (~fun@rockbox/developer/funman)
15:22:27pamauryit's strange that it went unnoticed
15:22:59 Quit randumbum (Quit: ChatZilla 0.9.88 [Firefox 10.0.2/20120216101208])
15:23:18pamauryshould we get more opinions before changing that ?
15:28:47pamauryI also would like to compile tools in c99 mode, there is no reason to keep them with -pedantic while the remaining of the code allows c99
15:33:40***Saving seen data "./dancer.seen"
15:44:04funmanpamaury: do we use .so ?
15:44:25pamauryfunman: no ? is that related ?
15:56:39 Join TheLemonMan [0] (~LemonBoy@adsl-ull-28-252.45-151.net24.it)
15:58:05funmanld is dynamic linker afaik
15:59:29gevaertsIt's not
15:59:37gevaertsld.so is
16:00
16:01:00funmanld can create static libraries ?
16:01:21Torneld is a linker, it can create anything
16:01:47Tornebut i suspect by "static library" you mena an ar archive
16:01:57Torneld doesn't directly do that, but it produces the object files inside the archive
16:02:03Tornear makes the actual archive
16:02:16Tornea static library is not really a specific thing on unix
16:02:25Torneit's just "some object files tha happen to be in an archive" :)
16:02:38funmanpamaury: what do we use "LD" for ?
16:03:18Tornethis sounds like the build system is being mangled in an unfortunatte and confusing way, tbh :)
16:03:21funmanrk27utils and imxtools which are not meant to be cross compiled afaict
16:03:37funmanso just remove it
16:03:38Torneif you use ar for $LD and some ar flags for $LDFLAGS then i can see that doing the "right" thing
16:03:42Tornebut not in a very natural way :)
16:04:17Tornewell, not right necessarily, since ar behaves differently to ld with respect to existing target files (it adds to the existing archive) and you normally need a different form of rule that rm's the target first
16:11:49funmanI also would like to compile tools in c99 mode, there is no reason to keep them with -pedantic while the remaining of the code allows c99
16:11:55funmanpamaury: ^ makes sense
16:12:38funmanpamaury: git grep '(LD)' returns only imxtools and rk27utils which don't use rockbox Makefile so deleting LD should be safe
16:20:24 Join kadoban [0] (~kadoban@ip98-165-177-158.ph.ph.cox.net)
16:30:03 Part jlbiasini
16:44:31 Join y4n [0] (y4n@unaffiliated/y4ndexx)
16:48:30pamauryfunman: on the contrary, fixing it would be more usefull if we ever need to use the makefile system for them
16:49:19pamauryfor example, ld is needed for mkamsboot and mkimxboot dualboot code
16:49:36pamauryand if we are to move them out of mk*boot, then $(LD) becomes useful
16:52:12CIA-44Commit dc572fb in rockbox by Amaury Pouly: configure: fix $(LD) which was incorrectly set as $(AR)
16:53:06 Quit maraz (Ping timeout: 260 seconds)
16:55:09CIA-44dc572fb build result: All green
17:00
17:03:58 Join DerPapst [0] (~Adium@91-64-115-27-dynip.superkabel.de)
17:04:34 Quit TheLemonMan (Quit: WeeChat 0.3.7)
17:07:45 Quit pamaury (Read error: Operation timed out)
17:12:56 Quit hiptobecubic (Ping timeout: 260 seconds)
17:14:32 Part Zagor
17:16:56 Join perrikwp_ [0] (~quassel@cpe-024-163-024-033.triad.res.rr.com)
17:17:32funmanpamaury: LD is not needed for mkamsboot, arm-elf-eabi-ld is
17:19:07 Quit perrikwp (Ping timeout: 240 seconds)
17:19:17 Join zenlunatic [0] (~justin@c-68-48-40-231.hsd1.md.comcast.net)
17:21:45 Join hiptobecubic [0] (~john@unaffiliated/hiptobecubic)
17:22:13 Join mshathlonxp [0] (msh@81.198.255.46)
17:22:25 Join maraz [0] (maraz@kapsi.fi)
17:23:39 Join passstab [0] (~v@c-68-80-37-73.hsd1.pa.comcast.net)
17:24:34 Part hiptobecubic
17:24:52 Quit kadoban (Ping timeout: 248 seconds)
17:30:07 Join jlbiasini [0] (~metaphys@d86-32-96-55.cust.tele2.at)
17:31:21 Join anewuser [0] (~anewuser@190.207.129.184)
17:31:21 Quit anewuser (Changing host)
17:31:21 Join anewuser [0] (~anewuser@unaffiliated/anewuser)
17:33:41***Saving seen data "./dancer.seen"
17:49:31 Join pamaury [0] (~quassel@vit94-1-82-67-248-70.fbx.proxad.net)
17:49:32 Quit pamaury (Changing host)
17:49:32 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
17:53:20 Join verbtim [0] (~verbtim@212.84.240.164)
17:58:18 Join evilnick [0] (~evilnick@host217-43-172-239.range217-43.btcentralplus.com)
17:59:14 Join enthdegree [0] (~BitchX@unaffiliated/enthdegree)
18:00
18:00:50verbtimI have the dc572fb build and The Rockbox Utility 1.2.12 is trying to download rockbox-fonts-dc572fb-120301.zip from download.rockbox.org/release/dc572fb-120301
18:00:56verbtimwhich isn't there
18:07:00gevaertsI think that might be fixed in the development tree. I'll test later
18:10:47jlbiasinigevaerts: verbtim: I just tested it is fixed in the development version
18:11:57 Quit passstab (Remote host closed the connection)
18:12:03jlbiasiniverbtim: you can find the development version there http://www.alice-dsl.net/dominik.riebeling/rockbox/rbutil/
18:18:41verbtimthanks. I downloaded the 3.10 fonts. I am on Xubuntu, so will try later to compile The Rockbox Utility.
18:19:07gevaertsI don't think the fonts have changed since 3.10
18:21:22 Join mystica555_ [0] (~Mike@mystica.mobi)
18:29:21 Join WalkGood [0] (~4@unaffiliated/walkgood)
18:33:54 Quit mystica555_ (Ping timeout: 276 seconds)
18:35:33 Join dhrasmus [0] (~dhrasmus@c-98-246-192-57.hsd1.or.comcast.net)
18:38:19 Quit nosa-j (Read error: Connection reset by peer)
18:39:28 Join nosa-j [0] (~m00k@adsl-74-235-42-18.clt.bellsouth.net)
18:41:24CIA-44Commit 4fbbbf1 in rockbox by Frank Gevaerts: Make %LC and %LR always have predictable and usable values.
18:44:12CIA-444fbbbf1 build result: All green
18:45:32pamauryIf someone can review http://gerrit.rockbox.org/r/#/c/152/ that would be nice. It adds support for the hidden attribute and it's not a complicated patch. In particular the apps/ part but another opinion on firmware/ would be useful too.
18:56:14funmanwhat's the button to see the whole diff on one page?
18:56:37funmanbah i'll just use the git fetch shortcut
18:57:09Tornethe gitweb link has the diff
18:57:10pamauryfunman: you can't :(
18:57:17Tornebut you can only see it
19:00
19:01:41funmanthat's what i want to do, just see it
19:02:51funmanpamaury: i don't know dircache at all, sorry
19:03:21pamauryfunman: I know dircache enough, that part is ok afaict
19:03:40pamaurybut I know nothing about apps/ on the other hand
19:07:03funmannot sure if we need 'hiding' and 'unhiding' strings
19:11:44 Join lebellium [0] (~chatzilla@g231084051.adsl.alicedsl.de)
19:14:34 Quit verbtim (Quit: Leaving)
19:17:01 Quit dhrasmus (Quit: Leaving)
19:24:15 Join rarog [0] (~chatzilla@p4FF5C6B2.dip.t-dialin.net)
19:27:15rarogpamaury: I'm still looking into usb configuration of OF vs RB. What is the 3rd endpoint in OF (EP 3 IN) for? MTP?
19:32:19 Join megal0maniac [0] (~megal0man@196.213.53.209)
19:33:14rarogjlbiasini: btw, how about bootloader v8 for delayed on button?
19:33:27 Quit megal0maniac (Client Quit)
19:33:45***Saving seen data "./dancer.seen"
19:34:12rarogoh
19:34:14rarogsorry.
19:34:18rarogit's already there. :D
19:39:09 Quit WalkGood (Quit: ♪ ♫ ♪ ♫ ♪ ♫ ♪)
19:39:32jlbiasiniragog ;)
19:39:46jlbiasini*rarog:
19:43:03 Quit mortalis (Quit: KVIrc 4.1.1 Equilibrium http://www.kvirc.net/)
20:00
20:07:55 Join bertrik [0] (~bertrik@ip117-49-211-87.adsl2.static.versatel.nl)
20:07:55 Quit bertrik (Changing host)
20:07:55 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
20:10:06jlbiasinibetrik: sorry for leaving you yesterday just after asking about this chinese stuff. I had to go and I was talking alone on the IRC for 1 hour so I thought that no one would answer anyway
20:10:37bluebrother^gevaerts: that font issue is fixed (see the ChangeLog on the RockboxUtility wiki page :)
20:10:58bluebrother^pamaury: IMO it would make sense to move the toggle functionality to the properties plugin
20:11:53gevaertsbluebrother^: I did look there, but not well enough :)
20:11:55jlbiasinibetrik: My point was that Chinese could get better supported and as we have a patch we should commit it :) should I do a git commit for you to commit it?
20:12:26 Join n1s [0] (~n1s@nl118-175-223.student.uu.se)
20:12:34 Quit n1s (Changing host)
20:12:34 Join n1s [0] (~n1s@rockbox/developer/n1s)
20:14:08bertrikjlbiasini, I think we should strip out the parts where the chinese phrase is exactly the same as the english (i.e. untranslated)
20:14:30jlbiasiniok I will give it a try
20:14:53 Join TheLemonMan [0] (~LemonBoy@adsl-ull-28-252.45-151.net24.it)
20:17:32 Quit [Saint] (Remote host closed the connection)
20:19:01gevaertsJdGordon: I can't reproduce the skinned lists and USB unplug issue anymore with a current build and the latest versions of my themes, so I won't report a bug :)
20:19:59 Quit anewuser (Read error: Connection reset by peer)
20:20:04gevaertsI also understand the glitch I was seeing on Wavy. I worked around it, but while investigating I did notice a possible issue
20:21:32gevaertsWhen a conditional changes, images that were previously drawn are cleared. If this is done in skinned lists, the way I understand the code the area that's cleared has moved down a bit to the next item, so this might clear the wrong area. I'll try to make a simple sbs to test this
20:22:09 Join anewuser [0] (~anewuser@190.207.129.184)
20:22:09 Quit anewuser (Changing host)
20:22:09 Join anewuser [0] (~anewuser@unaffiliated/anewuser)
20:25:48 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
20:31:22 Quit n1s (Read error: Connection timed out)
20:32:08 Join n1s [0] (~n1s@rockbox/developer/n1s)
20:33:03 Join Strife89 [0] (~Strife89@207.144.201.128)
20:35:46gevaertsJdGordon: it seems I can't make things go wrong the way I thought
20:38:27 Quit Thra11 (Ping timeout: 252 seconds)
20:39:19jlbiasinihere it goes for test_ plugins PLA g158
20:39:21fs-bluebotGerrit review #158 at http://gerrit.rockbox.org/r/#change,158 : test_X plugins PLA integration by Jean-Louis Biasini (changes/58/158/1)
20:40:19jlbiasinireview or push will be appreciated :)
20:47:30 Quit GodEater_ (Ping timeout: 245 seconds)
20:49:02gevaertsJdGordon: you can also ignore anything I said about scrolling. Scrolling works fine as long as you don't do weird things of the sort I do in Wavy
20:50:41pixelmahmm, I thought that if I used %Vi in the sbs that the viewports in a wps would be relative to that, not sure if that should be expected or not though
20:51:14pixelma(haven't thought it through, this is just my first reaction)
20:51:39 Join TomColler [0] (~thomas@net-93-144-191-188.cust.dsl.teletu.it)
20:52:23gevaertspixelma: %Vi basically defines the list area, so any non-list viewports have to be outside it, which can't be done if coordinates were relative to it
20:54:54 Join curtism [0] (~curtis@unaffiliated/programble)
21:00
21:03:23 Quit factor (Read error: Connection reset by peer)
21:07:24 Part TomColler
21:08:33 Quit y4n (Quit: PÆNTS ØLF!)
21:14:11jlbiasinibertrik: done g159 can someone push it?
21:14:13fs-bluebotGerrit review #159 at http://gerrit.rockbox.org/r/#change,159 : Updated some missing items for Simplified-Chinese translation (FS #12584) by Jean-Louis Biasini (changes/59/159/1)
21:16:46 Join passstab [0] (~v@c-68-80-37-73.hsd1.pa.comcast.net)
21:19:16 Quit antil33t| (Ping timeout: 252 seconds)
21:20:28 Join antil33t [0] (~Ahurhurr@101.98.150.103)
21:21:03 Quit lebellium (Ping timeout: 260 seconds)
21:21:34 Join factor [0] (~factor@r74-195-184-248.msk1cmtc01.mskgok.ok.dh.suddenlink.net)
21:23:21 Join lebellium [0] (~chatzilla@g231084051.adsl.alicedsl.de)
21:23:31pixelmagevaerts: ah, I see. Thanks for the explanation
21:27:20 Part Strife89 ("Leaving")
21:33:48***Saving seen data "./dancer.seen"
21:34:09 Join perrikwp [0] (~quassel@cpe-024-163-024-033.triad.res.rr.com)
21:37:12 Quit perrikwp_ (Ping timeout: 276 seconds)
21:38:09 Quit benedikt93 (Quit: Bye ;))
21:43:56 Quit yosafbridge (Quit: ERC Version 5.3 (IRC client for Emacs))
21:45:07 Quit [Saint] (Ping timeout: 240 seconds)
21:46:02 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
21:48:36 Quit anewuser (Ping timeout: 240 seconds)
21:50:00 Join yosafbridge [0] (~yosafbrid@li125-242.members.linode.com)
21:52:56 Quit TheLemonMan (Quit: WeeChat 0.3.7)
21:56:04 Join verbtim [0] (~verbtim@212.84.240.164)
22:00
22:02:09 Quit passstab (Remote host closed the connection)
22:04:05 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
22:06:26 Quit [Saint] (Ping timeout: 276 seconds)
22:10:34 Join [Saint] [0] (~Saint]@101.98.158.103)
22:10:34 Quit [Saint] (Changing host)
22:10:34 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
22:12:46verbtimI am doing the right thing and I am reading the manual for the Sansa Clip Zip. On "6.1 Volume" it says that the volume goes from -74 dB to +6 dB, but it currently goes to -81 dB.
22:13:30 Join bluebrother [0] (~dom@f053153192.adsl.alicedsl.de)
22:13:30 Quit bluebrother (Changing host)
22:13:30 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
22:13:49verbtimIn the simulator it goes to -73 dB, both are using 4fbbbf1
22:14:19bertrikI think we discussed that before (for other AMS/AMSv2 targets, but apparently it was never fixed :)
22:14:43 Quit fs-bluebot (Ping timeout: 260 seconds)
22:15:58 Join fs-bluebot [0] (~fs-bluebo@f053153192.adsl.alicedsl.de)
22:15:59verbtimI went to manual/configure_rockbox/sound_settings.tex but there are just the generic sansa/sansaAMS targets
22:16:15verbtimhow can I fix this? Is the target for the Zip sansazip?
22:16:41 Quit bluebrother^ (Ping timeout: 245 seconds)
22:16:53bertriksansaclipzip I think
22:21:57 Quit lebellium (Read error: Connection reset by peer)
22:22:26 Join lebellium [0] (~chatzilla@g231084051.adsl.alicedsl.de)
22:33:21dfkti have cleaned the image of the x5 simulator, it was kind of unappealingly dirty ;) ... would someone be so kind as to look at it and/or commit that, or should i put it on the tracker?
22:34:50dfkthttp://temp.elektrokrishna.com/UI256-old.bmp vs. http://temp.elektrokrishna.com/UI256.bmp
22:37:53 Join randumbum [0] (~IRCclient@c-71-193-176-198.hsd1.wa.comcast.net)
22:38:34 Quit petur (Quit: here today, gone tomorrow)
22:40:26jlbiasinipamaury: It seems that alot of files where not patched as you applied my patch for plugins keymaps ( I mean the big one with all game plugin inside
22:40:53pamauryI applied the patch as is
22:40:55jlbiasinior that it was reverted in some way recently
22:41:11jlbiasinisolitaire used to works and now no more
22:41:26jlbiasiniyes that is also what I remember
22:41:40jlbiasinibut solitaire used to work and now no more
22:41:41pamauryfeel free to resubmit on gerrit if you think there is a change, the patch is still on FS
22:42:04jlbiasiniso I took the fs patch and it succed nearly on all plugin
22:42:23jlbiasinistrange! :o ok I will do it
22:48:54pamauryI might have a made a mistake when applying it, I don't remember
22:48:57pamauryI'm sorry
22:49:26jlbiasiniI don't think so because it used to work for a while
22:50:23pamauryuse git blame or git log to see who touched it
22:51:00pamaurybluebrother: if you think hide/unhide should go to the property plugin, please add a comment to the task, I'm not the author !
22:51:00jlbiasiniI'm just noticing every change that occurs... because I just recieve my new fuze+ and can follow new change again
22:52:07jlbiasinipamaury: that might come from the pregit age (this seems like such a long time ago! :) )
22:52:16pamaurybluebrother: I've began some work to move dualboot images out of mk*boot, I should have finished tomorrow
22:52:58pamaurybut funman did not seem to be too enthusiastic about it
22:57:38 Quit pamaury (Remote host closed the connection)
23:00
23:01:26 Quit randumbum (Ping timeout: 246 seconds)
23:01:40 Join anewuser [0] (~anewuser@190.207.129.184)
23:01:41 Quit anewuser (Changing host)
23:01:41 Join anewuser [0] (~anewuser@unaffiliated/anewuser)
23:03:05funmani'm not opposed though since several people think it's a good idea
23:10:41 Join randumbum [0] (~IRCclient@c-71-193-176-198.hsd1.wa.comcast.net)
23:13:57 Quit [Saint_] (Quit: Quit)
23:13:59 Quit DerPapst (Quit: Leaving.)
23:14:57 Join DerPapst [0] (~Adium@91-64-115-27-dynip.superkabel.de)
23:19:31 Quit randumbum (Ping timeout: 252 seconds)
23:27:27 Join randumbum [0] (~IRCclient@c-71-193-176-198.hsd1.wa.comcast.net)
23:27:35jlbiasinipamaury: http://gerrit.rockbox.org/r/#/c/160/
23:27:38 Quit rarog (Quit: ChatZilla 0.9.88 [SeaMonkey 2.4.1/20111011102430])
23:27:50jlbiasinig160
23:27:51fs-bluebotGerrit review #160 at http://gerrit.rockbox.org/r/#change,160 : Fuze+: All games plugins keymaps by Jean-Louis Biasini (changes/60/160/1)
23:27:53gevaertsJdGordon: I also added the peakmeters you asked for
23:28:13 Part jlbiasini
23:29:50 Quit randumbum (Remote host closed the connection)
23:33:34 Join Scromple [0] (~Simon@119.225.209.134)
23:33:51***Saving seen data "./dancer.seen"
23:41:55 Quit n1s (Quit: Ex-Chat)
23:46:57 Quit factor (Read error: Connection reset by peer)
23:47:07verbtimI translated the remaining string for the Bulgarian language, here is a patch http://www.rockbox.org/tracker/task/12604
23:51:10 Quit lebellium (Quit: ChatZilla 0.9.88 [Firefox 11.0/20120222074758])
23:56:02 Quit adnap (Ping timeout: 246 seconds)
23:58:06 Join adnap [0] (~adnap@rrcs-71-42-140-57.sw.biz.rr.com)
23:58:13JdGordonverbtim: please use gerrit
23:58:38gevaertsJdGordon: for translation patches?
23:58:48JdGordonwhy not?

Previous day | Next day