IRC log for #openezx on 20090315

02:23.57*** join/#openezx Fosforo (n=Fosforo@20151040132.user.veloxzone.com.br)
03:10.13*** join/#openezx johanbr (n=j@blk-7-131-230.eastlink.ca)
04:08.31*** join/#openezx ysakaed (n=edward@118-166-239-91.dynamic.hinet.net)
04:10.48*** join/#openezx arne_ (i=nobody@dslc-082-082-067-086.pools.arcor-ip.net)
04:53.33*** part/#openezx ysakaed (n=edward@118-166-239-91.dynamic.hinet.net)
08:55.58*** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be)
09:10.04*** join/#openezx gmzhuo (i=gmzhuo@118.112.224.12)
09:36.25*** join/#openezx WyrM (n=wyrm@189.13.132.249)
09:36.36WyrMgood morning.
09:36.42WyrMgmzhuo: hi.
09:37.12gmzhuoWyrM,hi
09:37.26gmzhuohow about suspend/resume now?
09:37.47WyrMthere are a bunch of bugs.
09:37.56WyrMbut unrelated to the pxa2xx-spi bugs.
09:38.11WyrMand the gen2 blob patcher is done.
09:38.21gmzhuoyou mean in pcap,bp?
09:38.24WyrMhttp://people.openezx.org/wyrm/patch_blob
09:38.31WyrMyes, on both :)
09:38.43gmzhuohas anybody test it on E6:)
09:38.45WyrMits because of pcap that you can't resume twice.
09:38.58WyrMbecause you are using pcap irq as wakeup source
09:39.08WyrMonly on a1200 and a910.
09:39.16WyrMbut its not dangerous as on gen1.
09:39.31WyrMand remove the protection on blob partition on phonetools
09:39.35WyrMits not dangerous.
09:39.38WyrM(tested).
09:39.44gmzhuoOk
09:39.53*** join/#openezx pH5 (n=ph5@p5485FA9D.dip.t-dialin.net)
09:40.17WyrMgmzhuo: it's a makefile.
09:40.19gmzhuothen the protect area should be?
09:40.36WyrMput it on your boot_usb directory and run make -f patch_blob-gen2
09:40.49gmzhuoI have had a lookup on it:)
09:40.52WyrM0x0~0x60000
09:41.17WyrMand it will do nothing in case it detects something may go wrong :)
09:41.40WyrMalso, all phones tested up to now, has the same 1st stage blob.
09:41.55WyrM3xa910, 1xa1200.
09:43.37CIA-15OpenEZX: 03gmzhuo * r2440 10/trunk/src/host/phonetools/phonetools.cpp: change gen2 protect area
09:53.33WyrMgmzhuo: ahh, and sabrod's a910i has MBM at 0x0.
09:53.51WyrMthis invalidates ou gen detection code :/
10:51.04*** join/#openezx florian (n=fuchs@g227184197.adsl.alicedsl.de)
11:02.42gmzhuochecking for USB... no
11:02.42gmzhuoconfigure: error: *** Required libusb not installed ***
11:03.11gmzhuo[root@gbsoft boot_usb]# but /usr/local/lib/pkgconfig/libusb
11:03.11gmzhuolibusb-1.0.pc  libusb.pc
11:11.01*** join/#openezx ao2 (n=u@2001:1418:117:0:0:0:0:1)
11:15.03gmzhuoHi Wyrm
11:15.20gmzhuomd5="41d4087908214bef3381d7e815980cbc"; \
11:15.20gmzhuo<PROTECTED>
11:15.20gmzhuo<PROTECTED>
11:15.20gmzhuo<PROTECTED>
11:22.56*** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be)
11:27.26WyrMgmzhuo: your blob is different than mine.
11:27.43WyrMalso, it can't be hacked a second time... In case you did it on the first :)
11:27.51WyrMjust FYI.
11:29.37WyrMplease send me (email) the blob-original file that was left on the directory.
11:54.33*** join/#openezx xottoG (n=Gottox@s01.de)
11:56.36gmzhuoWyrM,http://people.openezx.org/gmzhuo/blob-original-e6
12:15.52WyrMgmzhuo: the same patch is ok for your blob too.
12:16.24WyrMthere is a single difference, at 0x3c.
12:16.31WyrM<   3c: e3a02005        mov     r2, #5  ; 0x5
12:16.32WyrM---
12:16.32WyrM>   3c: e3a02002        mov     r2, #2  ; 0x2
12:16.41WyrM< is a1200, > is e6.
12:17.24WyrM<PROTECTED>
12:17.24WyrM<PROTECTED>
12:17.24WyrM<PROTECTED>
12:19.38WyrMa different "LCD Buffer Strength Control register"
12:20.26WyrM7.5.18 of the manual. a1200 uses a value of 5, e6 uses 2.
12:20.43WyrMother than this, no differences. And the patch is still valid.
12:26.17WyrMscript updated to allow patching of this version too.
12:26.43WyrMgmzhuo: download the script again. it will allow execution this time.
12:33.57*** join/#openezx l403 (n=l@85.132.159.239)
12:35.53gmzhuoWyrM,we still need to reserve the first page mem for modified blob
12:36.46gmzhuo#ifdef MBM      //MBM module
12:36.46gmzhuo<PROTECTED>
12:36.46gmzhuo<PROTECTED>
12:37.20WyrMwhy?
12:37.26WyrMMBM is writing to it?
12:38.20gmzhuoyes
12:39.26WyrMthis is not good :/
12:39.26WyrMon gen1, as there is no mbm, its ok.
12:39.32gmzhuonot mbm write it
12:39.35gmzhuoIt's blob
12:39.39WyrMno.
12:39.46WyrMthen we just hack blob :)
12:40.20WyrM<PROTECTED>
12:40.20WyrM<PROTECTED>
12:40.37WyrMis powerup_reason_addr 0xa000000c?
12:40.41gmzhuoyes
12:41.01WyrMmbm passes the power up reason on r0.
12:41.07gmzhuoright
12:41.19WyrMso, I will hack it to save r0 value at the start.
12:41.43WyrMand only pass this if its NOT going to resume from suspend for openezx.
12:41.49WyrM(PSPR != 0).
12:42.07WyrMthere is still space on the patch (see all the nops...) :)
12:43.39gmzhuoOr we can jump to 0x80000,there has 0x800 space
12:43.44gmzhuoon E6
12:44.03WyrMnot needed.
12:44.09WyrMthe space is still enough.
12:44.24WyrMwant to do it?
12:44.30WyrMor want me to do?
12:44.48gmzhuoYou do it,because my e6 lost cammera key :)
12:44.55WyrMok.
12:45.27gmzhuoYou can use start.S in blob for e6,
12:45.45WyrMwell, it may be different.. :)
12:45.56WyrMit is different on gen1.
12:46.09WyrMand as I see, there are differences on gen2 too (a1200xe6).
12:46.19WyrMso I will just work on the binary.
12:55.02gmzhuo<PROTECTED>
12:55.02gmzhuo#if defined(MACAU) || defined(PENGLAI)
12:55.02gmzhuo<PROTECTED>
12:55.02gmzhuo#else
12:55.02gmzhuo<PROTECTED>
12:55.03gmzhuo#endif
12:55.05gmzhuo<PROTECTED>
12:55.26WyrMits done already.
13:07.20*** join/#openezx flameman (n=toor@net-93-65-10-119.cust.dsl.vodafone.it)
13:26.48WyrMI don't have a backup of a1200 original blob.
13:26.55WyrMlol
13:27.34gmzhuoYou can use E6's
13:27.57gmzhuoand you know what's the diffrent
13:28.09WyrMok, I found a backup
13:28.16WyrMI just need the md5sum :)
13:36.06WyrMoops, found an error.
13:38.51gmzhuowhat's error?
13:39.08WyrMmath error ;)
13:39.38WyrMif I hadn't spotted it, then the code would look for SLEEP_FLAG on the wrong place.
13:39.42WyrMno resume on 2.4.
13:39.46WyrMbut its ok now.
13:39.52*** join/#openezx xottoG (n=Gottox@s01.de)
13:39.54WyrMdouble checking the patch makefile
13:41.09gmzhuoMaybe you should give out the complete blob first stage,not the patch :)
13:41.33WyrMnah. a makefile to patch a bootloader is asm is more fun.
13:41.34WyrMlol
13:43.27WyrMjust updating the comments on the file.
13:43.30WyrMits ok now.
13:43.37WyrMI will test it.
13:45.22WyrMflashed.
13:45.27WyrMlets see if it works :)
13:46.59WyrMyes, its working.
13:47.36WyrMmakefiles on my people.openezx dir updated.
13:50.01gmzhuoMabe I will meet incorrect MD5
13:50.07WyrMno
13:50.12WyrMI updated the md5sums
13:50.17WyrMfor both a1200 and e6.
13:50.23gmzhuo:)
13:50.32gmzhuobut original will meet
13:50.50WyrMI updated the original too :)
13:52.38gmzhuomd5="15eba8cf027032ffbe2b3d2c893344a1"; \
13:52.38gmzhuo<PROTECTED>
13:52.38gmzhuo<PROTECTED>
13:52.40gmzhuo£¿£¿
13:53.08WyrMhuh?
13:53.18gmzhuoI said I will meet
13:53.28WyrMthis is the first md5
13:53.29gmzhuo41d is for original
13:53.42WyrMyes, this is the pre-ceck.
13:54.07WyrMit checks for the original bootloader md5sum before start, and after patching it checks the hacked bootloader md5.
13:54.19WyrMdo a "make clean"
13:54.23WyrMand try again :)
13:54.34WyrMI think you left some files from the previous script.
13:55.04gmzhuostill
13:56.21WyrMnot possible.
13:56.25gmzhuothere should have 4 md5 sum for a1200/oringal&firstpatched E6/oringal&firstpatched
13:56.35WyrMyes.
13:56.57gmzhuobut here only 2
13:57.06WyrMhttp://pastebin.com/m31c7d7fa
13:57.13WyrMno, its 4
13:57.20WyrM2 before and 2 after.
13:57.44WyrMsee the pastebin, its using the blob-original-e6 you sent me.
13:58.20gmzhuono after???
13:58.43gmzhuowhere is the after?
13:59.36WyrMsee pastebin
13:59.40WyrMbefore at line 13
13:59.45WyrMafter at line 73
13:59.57WyrMthis pastebin is from a e6 run.
14:01.12gmzhuoI mean the blob on my e6 is not oringal
14:01.30WyrMthe blob you sent me is the original isn't it?
14:01.51WyrMhave you patched it with the previous version of the script?
14:02.01WyrMdo you still have the original one?
14:02.34WyrMrun make -f patch_blob-gen2 clean
14:02.58WyrMthen cp my-backup-of-the-original-blob blob-original
14:03.00gmzhuono
14:03.07WyrMthen make -f patch_blob-gen2
14:03.29WyrMif blob-original is present, then it will not try to download from the phone again.
14:04.50gmzhuoI changed patch_blob-gen2
14:04.57WyrMplease don't.
14:04.58gmzhuodone
14:05.25gmzhuohow to test moto's suspend/resume?
14:05.45WyrMjust let it wait for some minutes
14:05.55WyrMand try to wake up.
14:06.04WyrMI really think you messed up things.
14:06.12WyrMthe md5sums are correct.
14:06.17gmzhuowhat's thing?
14:06.25WyrMyou really shouldn't change the script.
14:07.00gmzhuoyou mean patch is in diffrent address?
14:07.08WyrMno.
14:07.17WyrMwhat have you changed on the makefile?
14:07.28gmzhuoadd one md5
14:07.40WyrMthe md5sum for the original check, or for the after-patched check?
14:08.21WyrMon blob-1-hacked or on blob-hacked?
14:18.53*** join/#openezx Fosforo (n=Fosforo@20151040132.user.veloxzone.com.br)
14:41.34gmzhuoworked
14:43.50gmzhuoWyrM, are you here?
14:45.30gmzhuostably resume from lock key
14:45.37WyrMyes.
14:45.48WyrMI know, I tested it before ;)
14:45.53*** join/#openezx johanbr (n=j@blk-7-131-230.eastlink.ca)
14:45.54WyrMresume is ok.
14:46.07WyrMthe problem you were having before is with the pcap irq handler.
14:46.28WyrMif you resume with a pcap irq as the wake up source, then the irq handler gets confused.
14:46.46WyrMprobably, scheduling the workqueue at pcap_resume() will solve the issue :)
14:47.17WyrMwe need to take a look at some code to know the wakeup reason.
14:47.25gmzhuoso I can't resume from touch event,right?
14:47.31WyrMand only call the workqueue if the wakeup reason is the pcap irq.
14:47.35WyrMyou can.
14:47.47WyrMbut you will get no touch events after the resume ;)
14:47.48gmzhuoNo,I can't
14:47.58WyrMthe first time you can.
14:48.02gmzhuoright
14:48.07WyrMbut then touch events will stop working.
14:48.17WyrMbecause the ISR for touch is set on pcap.
14:48.21WyrMand never cleared.
14:48.40WyrMthat's why I told you to schedule pcap_work at pcap_resume.
14:48.48WyrMits not the proper fix, but will do for now.
14:49.38gmzhuoI think not get touch event before resume is not problem
14:49.52WyrMI don't understand what you mean.
14:50.09WyrMyou certainly will not get a touch event BEFORE resume.
14:50.11WyrMlol.
14:50.31WyrMyou mean not get the touch event that is used for resume?
14:50.41WyrMfor touch you are right.
14:50.53WyrMbut you can't lose a USB insert/remove event.
14:51.00WyrMgen1 usb is pcap, not eoc.
14:51.24WyrMalso, you can't lose a headphone jack insert/removal (this is valid for gen2 too).
14:51.25gmzhuoYou mean even after resume,we can't got touch event now?
14:51.36WyrMyes.
14:51.53gmzhuothat's not good
14:52.01WyrMif you wake up with a pcap event, the pcap event that you used to wake up the system will no longer work.
14:52.07WyrMI know how to fix it.
14:52.12WyrMand I told you already ;)
14:53.14WyrMthis happens because the pcapirq will not run on wakeup. So, ISR will never be cleared for the event that wake up the system.
14:53.40WyrMand because its not cleared it will no longer generate gpio interrupts to pxa.
14:54.01gmzhuoso the original clear all isr in pcap when resume
14:54.40*** join/#openezx Gottox (n=Gottox@s01.de)
14:55.10gmzhuonot roiginal,in 2.6.23/24/25
14:59.56WyrMplease don't copynpaste buggy code :)
15:00.01WyrMfix it properly.
15:00.18WyrMlearn how to know the wake up source.
15:00.41WyrMand, on pcap_resume(), if the wakeup source is the pcap irq pin, schedule pcap_work
15:04.20gmzhuoI'll not change this,but I let compile happy when no rfkill select
15:05.37gmzhuoI have local/debug based on ezx/current,and I switch to ezx/current,then push,will I add local/debug branch to remote?
15:09.19WyrMyes.
15:09.52WyrMif you do tg remote --populate it will upload all your topgit patches to openezx.git
15:10.09WyrMfor local branches, you should create git branches. Not topgit patches :)
15:10.31WyrMalso, have you run the script I sent to the mailing list?
15:10.44WyrMit deletes a lot of other branches that I don't want on openezx.git
15:10.51WyrMbecause they are already mainline.
15:11.54WyrMexport your local changes with "tg patch [name] > [name].patch"
15:12.01WyrMand recreate them as git branches.
15:13.24gmzhuowhere is the script?
15:13.36WyrMsent to openezx-devel
15:13.40gmzhuofind it
15:13.54WyrMit will delete your debug patches.
15:14.01WyrMso, save a backup first.
15:16.41gmzhuocleaned
15:16.58WyrMplease wait some minutes before commiting.
15:17.10gmzhuobut I got remote end hung up when git-push
15:17.11WyrMim uploading linus changes.
15:17.17WyrMdon't push now.
15:17.25WyrMand pull before push.
15:17.30gmzhuoOk
15:20.27WyrMgmzhuo: are you able to resume with pxa2xx-keypad keys?
15:20.30WyrMim not :/
15:20.41WyrMand I don't know why, nor how to solve it.
15:21.13WyrMlinus changes pushed to openezx.git
15:23.43*** join/#openezx ao2 (n=u@2001:1418:117:0:0:0:0:1)
15:25.11gmzhuoI'm not too
15:30.07WyrMok.
15:30.12WyrMI will look at this later.
15:30.13WyrM:)
15:33.08gmzhuodevice_may_wakeup(),what's meaning,should we set someting on pxa2xx_keyboard?
15:33.59gmzhuostill got remote end hung up unexpectly
15:34.35gmzhuoI use git clone git://git.openezx.org/openezx today
15:51.49WyrMgit.openezx is working here.
15:52.19WyrMdevice_may_wakeup is for a userspace toggle on sysfs.
15:52.34WyrM/sys/platform/...../power/wakeup
15:52.41WyrMor something like that :)
16:04.44gmzhuomaybe we need change pgsr3
16:05.28WyrMsearch on the code where pgsr is set.
16:05.30WyrM:)
16:05.51WyrMI guarantee that "PGSR3 =" is not allowed by mainline people :)
16:06.09gmzhuodose I use git clone git://git.openezx.org/openezx is right?
16:06.20WyrMgit pull is.
16:06.26WyrMno.
16:06.50WyrMin your case, you use git@git.openezx.org/openezx
16:06.56WyrMotherwise you can't push
16:07.23WyrMon .git/config
16:07.30WyrMurl = git@git.openezx.org:openezx.git
16:08.03gmzhuosee cm-x2xx.c
16:08.34WyrMwe can't do like this.
16:09.22gmzhuothen how can we do?
16:09.35WyrMI don't know :)
16:09.37gmzhuojust have a test?
16:10.04WyrMwell, it will probably work if that's what you want to know.
16:10.15WyrMbut it doesn't mean that its right.
16:10.27WyrMthere is an api for this.
16:10.40WyrMI don't know why its not working, but it should be used.
16:10.48WyrMfixed and used :)
16:11.03WyrMthere is irq_set_wake(), gpio_set_wake().
16:11.18WyrMand afaik its being set by pxa2xx-keypad.
16:11.34WyrMand the wakeup source is enabled on sysfs.
16:11.43WyrMwe have to debug this and find what is wrong.
16:12.37WyrM#hackndev
16:12.39WyrMoops
16:13.21WyrMmarex from #hackndev had some issues with pxa2xx-keypad as wakeup source too.
16:13.33WyrMbut he is not online now :/
16:15.12WyrMgmzhuo: also, read Documentation/arm/pxa/mfp.txt
16:15.41WyrM"WAKEUP_ON_LEVEL_HIGH - specifically for enabling of keypad GPIOs,"
16:22.10gmzhuobedtime bye
16:22.59WyrMbye, good night
16:23.57sabrodhi and good night.
16:24.07sabrod:)
16:26.48*** part/#openezx gmzhuo (i=gmzhuo@118.112.224.12)
16:44.09ao2WyrM, I keep getting "Adding branch ezx/defconfig_mainline" even after runnig your script, we have to remove also local top-bases refs
16:44.28ao2.git/refs/remotes/$REMOTE/top-bases/$p
16:47.40WyrMao2, I forgot to add this to the script.
16:49.11ao2WyrM, no problem, I'll start moving branches candidate for submission below ezx/mach/pcap
16:49.24WyrMok
17:00.45*** join/#openezx xottoG (n=Gottox@s01.de)
17:07.02CIA-15OpenEZX: 03Daniel Ribeiro <drwyrm@gmail.com> 07ezx/current * r0bc41302ab 10openezx/.topmsg: update patch header
17:08.39WyrMasoc/pxa-ssp-fix-SSPSP_SCMODE: I will look for some reason on why we need this.
17:09.01WyrMI suspect that "it causes noise on only some versions of our hardware" is not ok :)
17:09.42WyrMso I will review I2S specifications, and compare the data sampling times with what we get with and without SCMODE(2).
17:11.04WyrMahh, btw, this patch is not correct I think..
17:11.21WyrMscmode(2) should be only for DAIFMT_I2S.
17:18.55WyrMhum.
17:19.22WyrMwe are in trouble with this... the i2s specs says data is sampled on the clock rising edge.
17:19.30WyrMand this would be scmode(0). :/
17:20.45WyrMNo!
17:20.52WyrMthe lines on pxa manual is drive.
17:20.57WyrMso we are correct :)
17:33.25ao2:)
17:33.31yangWyrM: do you own A910 and do you get the battery error sometimes?
17:33.37yanghey ao2
17:33.53WyrMyang: no, never had battery errors here.
17:34.05ao2hi yang, me neither
17:34.15yangao2: I get "invalid battery" error - in italian language...
17:35.11yangsabrod said its common error with a910 - it has some memory effect
17:35.23WyrM<PROTECTED>
17:35.23ao2yang, is this a message from the linux os, do you get it whan trying to change via USB?
17:35.29yangThe mobile is U.K. made and I had Italian SIM inside so this is where error comes from
17:35.37yangno its motorola OS
17:35.57yangNo, I charge with power adapter
17:35.58ao2yes, for linux os I meant moto os, as opposed to MBM
17:36.02WyrM<PROTECTED>
17:36.09yangdamn
17:36.15yangso its common to some models
17:36.21WyrM<PROTECTED>
17:36.22sabrodI have this quite frequently, especially with original, neuf firmware...
17:36.30sabrod(on un flashed phones)
17:36.45yangit won't start to charge the battery
17:37.28yangbut I haven't tried to charege it via USb, will do that now
17:39.18yanghm, with USB i get green circle
17:39.29yangand on power adapter blue circle and error
17:41.42yangHA ! Ricarica batteria in corso (Battery charging ! :-)
17:49.53sabrodCool, yes, it works too here, if I insist.
17:49.57sabrod?
17:50.06sabrodweird, but...
17:51.01sabrodopenezx kernel's fault?
17:52.41*** join/#openezx johanbr (n=j@blk-7-131-230.eastlink.ca)
17:58.11yangsabrod: no its no openezx kernel installed
17:58.36sabrodNever booted it?
17:58.40yangadaptor fault
17:58.43yangI would say
17:58.44sabrodAh... :)
17:59.15yangdoes your A780 charges over USB?
17:59.22yangmine didn't want to
18:00.30yangao2: i am listening to "I santo California" do you know it?
18:01.12yangthere were quite popular, back in the 60s maybe
18:03.03yang70s actually
18:15.23WyrMis stereo sound working on openezx kernel?
18:16.34ao2WyrM, IIRC it is, I tested it with headphones once
18:16.43WyrMlol
18:17.16WyrMI remember I buggered broonie for a hack on pxa-ssp.c, the hack is there and we are not using it.
18:17.30WyrMand it was for stereo audio.
18:18.30WyrMI think I probably forgot to comit the changes to openezx.git or the hack was not needed after all..
18:21.57ao2WyrM, I can test it again later, just to be sure, I can't remember if my test was with code from git or svn
18:23.54WyrMim testing it now.
18:24.00WyrMstereo audio is working.
18:24.13WyrMthen the hack was not needed, or im missing something.
18:41.04sabrodWyrm: I could add my A910i first bytes to phonetools gen2/gen1 detection code ?
18:41.46WyrMvery likely it's the same from gen1.
18:42.12sabrodwell, I don't think so.
18:42.19sabrod0006 is on A780.
18:42.26sabrod0000 is on my A910i
18:42.35sabrod...
18:42.54WyrMthen its probably ok.
18:43.09WyrMbut this a dumb way to check what phone it is :)
18:43.11sabrodI hope, yes!
18:43.40sabrodwell, kct is always there at 60800 on all gen2 I know...
18:44.04sabrodBut I don't think it's smarter. :)
18:44.08bartva: 09.00 - 18.00
18:44.15WyrMyes, I tought that mbm would always be at the same place too ;)
18:44.58sabrodWhy the USB Id isn't used?
18:46.42sabrodWell, I don't know much about that, but looking at http://svn.openezx.org/trunk/src/host/phonetools/UsbPhoneFactory.cpp there is a poosibility, isn't it ?
18:48.11sabrodOk, for genblob it can't help, I forgot
18:48.22WyrMao2: capture with the mono codec is working too.
18:48.38WyrMao2: and I remember that on my last test it wasn't. :)
18:49.35ao2WyrM, very good, magis is useful sometimes...
18:49.40ao2magic, even
18:50.01WyrMmaybe I tested on a different phone. this is crazy.
18:55.42WyrMI think it was the noise. It is working but too noisy.
19:18.04*** join/#openezx Kensan_ (n=ken@gw.ptr-80-238-227-222.customer.ch.netstream.com)
19:43.26*** join/#openezx florian (n=fuchs@g227184197.adsl.alicedsl.de)
19:46.25*** join/#openezx pleemans (n=toi@d54C2AAB7.access.telenet.be)
20:33.03*** join/#openezx tombhadAC (n=tombhad@pD95DD6D0.dip.t-dialin.net)
22:14.16ao2mmh, bringing down a branch is more difficult than I thought, If i make ezx/mach/camera depend on ezx/mach/pxa2xx-flash instead of ezx/mach/udc and then call "tg update" the latter does nothing and it left udc stuff in the topic branch...
22:16.49ao2WyrM, maybe this is not possible at all, to do a "partial review" of the topic branch, I mena. I will end up exporting the patches and refreshing them manually, but a topgit way would have been great.
22:17.06ao2s/review/rewind/
22:17.26ao2and s/mena/mean :)
22:18.16WyrMao2: changing .topdeps is enough.
22:18.39ao2WyrM, should be
22:18.46WyrMfirst make the patch orphan, then add it where you want it to be.
22:19.45WyrMbut instead of calling tg update on ezx/current so it updates recursivelly, do tg update on the first child.
22:20.01ao2let's try the orphaning way
22:21.58WyrMthere is a great chance of conflicts, because you are changing the ezx/mach/* patches...
22:22.14WyrMbut its not hard do solve them :)
22:26.06*** join/#openezx sabrod (n=sab@lns-bzn-28-82-250-145-161.adsl.proxad.net)
23:28.57ao2WyrM, still something wrong, doing this http://people.openezx.org/ao2/tmp/tg-depend-delete.png camera changes are still in ezx/mach/bp ezx.c
23:29.36ao2they shouldn't
23:30.26ao2will try tomorrow again
23:31.03WyrMao2: yes, but it will not make a difference in the end :)
23:31.30WyrMyou are not deleting the camera branch, you will just move it nearer master.
23:33.10ao2ok, let's see what happens after all the relinking is done

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.