February 2017 - CRUX-ARM - lists.crux-arm.nu

Re: [crux-arm] crux-arm 3.3?
by Victor Martinez 26 Feb '17

26 Feb '17

* Guido Caliandro (guido.caliandro(a)gmail.com) wrote: > Hi Victor, > Hey Guido! > first of all thank you very much for the great job that you and all staff > are doing. > No problem :) > if it could be of any interest, I can test it on raspberry pi 1b, b+, PI2 > and PI3, and also on some olinuxino arm with A10 and A20 soc. > I hope in the next weeks I can put hands on optimizations for some devices, A10 and A20. My Rpi 1b is far and I'm not sure if I'll be able to start with it in at least 3 weeks. If you want to deploy the generic release and look for errors, improvements or anything, it could be greati, also to debug and probably to start optimizations in some of your devices like Rpi2/Rpi3. If you need any point to start doing these optimizations like we do, just tell me and I'll try to guide you in the proccess. There is some information here: https://crux-arm.nu/Documentation/HowToContribute > ciao > Guido > Regards, --- Victor Martinez Learning bit by bit | http://vjml.es

1 0

crux-arm 3.3?
by Guido Caliandro 25 Feb '17

25 Feb '17

Hi all, Is there any news about that? ;-) thank you bye Guido

2 2

Re: [crux-arm] crux-arm setup firefox error - segmentation fault
by Víctor Martínez 07 Feb '17

07 Feb '17

El 03/02/17 a las 07:47, Milan Buška escribió: > Hi Victor Hey Milan, > nspr => enable-64bit on disable-64bit > firefox => delete line ac_add_options --with-pthreads I've make an overlay for nspr to avoid the enable-64bit option. Thank you for this catch. About firefox, I was deeping in the problem this last weekend with the segfault and removing pthread support isn't a good way to solve the problem, since it wasn't really related to firefox, and doing this is like avoiding the problem. The problem was the nss update to 3.28.1 which I made it wrong. I don't know how I checked upstream commit because I didn't see all the changes made by Fredrick... and this broke firefox. I've updated the port and only a rebuild of nss is needed. I recommend you to check it and let firefox with pthreads support because it will work right now. > I deleted a line ac_add_options --with-pthreads > maybe it should just be pthread - I do not know > But after adjustments and subsequent compilation Firefox starts without errors. Thank you very much for your report and help provided. > Have a nice day > Milan > Have a nice web surfing with firefox! Regards, -- Victor Martinez Learning bit by bit | http://vjml.es

3 2

Re: [crux-arm] crux-arm setup firefox error - segmentation fault
by matti christensen 07 Feb '17

07 Feb '17

Hello i've been following your discussion because i used to have serious issues w ff and arm boards while ago. after version 48.n.n building has gone smoothly on rpi which i use for writing this. as Victor may remember i'm really aschetic w my systems = i only install bare minimum binary using the only system suitable for that ( CRUX ARM ), then i remove stuff i don't need ( there are many + binary package manager ) and start building stuff like X etc.. attached you find my .mozconfig for 48.0.2 ( please note that most lines are just because of my minimal needs ); /mc Quoting Victor Martinez <pitillo(a)crux-arm.nu>: > * Milan Bu??ka (milan.buska(a)gmail.com) wrote: >> Hi Victor >> >> 2017-02-02 15:17 GMT+01:00 Victor Martinez <pitillo(a)crux-arm.nu>: >>> * Milan Bu??ka (milan.buska(a)gmail.com) wrote: >>>> Hi Victor >>>> >> > Hey! > >>> Hello again Milan, >>> >>>> Compile with options O0 gives the same error. >>> I can confirm the same. I've built firefox without optimizations >>> and it's still segfaulting. >>> >>>> My suggestion is to move the package from nspr opt to opt-arm >>>> and adjust Pkgfile for detailed specification of the ARM 32-bit >>>> architecture. >>> Good catch I'll review nspr and give a try with a rebuild. >>> >>>> I suspect that this error depends on compilations since nspr. >>>> I'll try to compile firefox with modified nspr and nss packages. >>>> >>>> Have a nice day >>> Same there. I'll update with results. >>> >>>> Milan >>>> >> >> nspr => enable-64bit on disable-64bit >> firefox => delete line ac_add_options --with-pthreads >> >> I deleted a line ac_add_options --with-pthreads >> maybe it should just be pthread - I do not know >> But after adjustments and subsequent compilation Firefox starts >> without errors. >> > Very good catch Milan. I've modified nspr, rebuilt nss and firefox > and the problem persists. I'll update firefox port with your > suggestions and I'll check tomorrow, if it works right, I'll push > changes for nspr and firefox at saturday night. > > I was near to revert nss and firefox ports because I wasn't abke to > deep in to the problem more than removing optimizations. > > Thank you very much for your research and fixes provided. You have > Flyspray and IRC too to contact us, but ML is fine (just as a > reminder). > >> Have a nice day >> Milan >> > Best Regards, > > --- > Victor Martinez > Learning bit by bit | http://vjml.es > _______________________________________________ > crux-arm mailing list > crux-arm(a)crux-arm.nu > https://crux-arm.nu/mailman/listinfo/crux-arm

2 1

Re: [crux-arm] crux-arm setup firefox error - segmentation fault
by Victor Martinez 03 Feb '17

03 Feb '17

* Milan Bu??ka (milan.buska(a)gmail.com) wrote: > Hi Victor > > 2017-02-02 15:17 GMT+01:00 Victor Martinez <pitillo(a)crux-arm.nu>: > > * Milan Bu??ka (milan.buska(a)gmail.com) wrote: > >> Hi Victor > >> > Hey! > > Hello again Milan, > > > >> Compile with options O0 gives the same error. > > I can confirm the same. I've built firefox without optimizations and it's still segfaulting. > > > >> My suggestion is to move the package from nspr opt to opt-arm > >> and adjust Pkgfile for detailed specification of the ARM 32-bit architecture. > > Good catch I'll review nspr and give a try with a rebuild. > > > >> I suspect that this error depends on compilations since nspr. > >> I'll try to compile firefox with modified nspr and nss packages. > >> > >> Have a nice day > > Same there. I'll update with results. > > > >> Milan > >> > > nspr => enable-64bit on disable-64bit > firefox => delete line ac_add_options --with-pthreads > > I deleted a line ac_add_options --with-pthreads > maybe it should just be pthread - I do not know > But after adjustments and subsequent compilation Firefox starts without errors. > Very good catch Milan. I've modified nspr, rebuilt nss and firefox and the problem persists. I'll update firefox port with your suggestions and I'll check tomorrow, if it works right, I'll push changes for nspr and firefox at saturday night. I was near to revert nss and firefox ports because I wasn't abke to deep in to the problem more than removing optimizations. Thank you very much for your research and fixes provided. You have Flyspray and IRC too to contact us, but ML is fine (just as a reminder). > Have a nice day > Milan > Best Regards, --- Victor Martinez Learning bit by bit | http://vjml.es

1 0

crux-arm setup firefox error - segmentation fault
by Milan Buška 02 Feb '17

02 Feb '17

Hello. Can you advise me where to start looking for a mistake why not start firefox. When run in a terminal am getting 1485791758698 addons.manager ERROR Exception loading default provider "resource://gre/modules/addons/XPIProvider.jsm": [Exception... "Component returned failure code: 0x80570016 (NS_ERROR_XPC_GS_RETURNED_FAILURE) [nsIJSCID.getService]" nsresult: "0x80570016 (NS_ERROR_XPC_GS_RETURNED_FAILURE)" location: "JS frame :: resource://gre/modules/addons/XPIProvider.jsm :: <TOP_LEVEL> :: line 1769" data: no] Stack trace: resource://gre/modules/addons/XPIProvider.jsm:1769 < AddonManagerInternal.startup()@resource://gre/modules/AddonManager.jsm:925 < this.AddonManagerPrivate.startup()@resource://gre/modules/AddonManager.jsm:3016 < amManager.prototype.observe()@resource://gre/components/addonManager.js:71 Assertion failure: NS_IsMainThread(), at /usr/ports/opt-arm/firefox/work/src/firefox-51.0/security/manager/ssl/nsNSSComponent.cpp:259 Segmentation fault nss version 3.28.1 machine tegra jetson tk-1 Regards, Milan -- ---------------------------------------------------------------------------- Remember, no question is too stupid and no problem too small--We've all been beginners

3 4