| *** BCMM <BCMM!~BCMM@unaffiliated/bcmm> has quit IRC | 00:08 | |
| *** armpit <armpit!~armpit@2601:202:4180:c33:29af:1aac:5242:8221> has joined #yocto | 00:10 | |
| *** everclear <everclear!~martin@dslb-178-002-165-167.178.002.pools.vodafone-ip.de> has joined #yocto | 00:43 | |
| *** ferlzc <ferlzc!~ferlzc@177.83.99.221> has quit IRC | 01:18 | |
| *** ferlzc <ferlzc!~ferlzc@177.83.99.221> has joined #yocto | 01:18 | |
| *** gtristan <gtristan!~tristanva@24.48.71.248> has joined #yocto | 03:31 | |
| *** chandana73 <chandana73!~ckalluri@149.199.62.129> has joined #yocto | 03:39 | |
| *** ferlzc <ferlzc!~ferlzc@177.83.99.221> has quit IRC | 04:58 | |
| *** georgem_home <georgem_home!uid210681@gateway/web/irccloud.com/x-zywvmfokojijueav> has quit IRC | 05:23 | |
| *** harisokanovic <harisokanovic!~harisokan@130.164.62.193> has quit IRC | 05:38 | |
| *** harisokanovic <harisokanovic!~harisokan@130.164.62.193> has joined #yocto | 05:51 | |
| *** chandana73 <chandana73!~ckalluri@149.199.62.129> has quit IRC | 06:11 | |
| *** camus <camus!~Instantbi@183.128.236.162> has joined #yocto | 06:48 | |
| *** kaspter <kaspter!~Instantbi@183.156.69.203> has quit IRC | 06:50 | |
| *** camus is now known as kaspter | 06:50 | |
| *** |King_InuYasha| <|King_InuYasha|!~King_InuY@ool-18e49371.dyn.optonline.net> has joined #yocto | 07:13 | |
| *** King_InuYasha <King_InuYasha!~King_InuY@fedora/ngompa> has quit IRC | 07:16 | |
| *** tprrt <tprrt!~tprrt@ram31-1-82-234-79-177.fbx.proxad.net> has joined #yocto | 08:49 | |
| *** kroon <kroon!~kroon@37-247-29-68.customers.ownit.se> has joined #yocto | 09:00 | |
| *** kroon <kroon!~kroon@37-247-29-68.customers.ownit.se> has quit IRC | 09:12 | |
| *** nighty- <nighty-!~nighty@b157153.ppp.asahi-net.or.jp> has joined #yocto | 10:02 | |
| *** kroon <kroon!~kroon@37-247-29-68.customers.ownit.se> has joined #yocto | 10:09 | |
| *** BCMM <BCMM!~BCMM@unaffiliated/bcmm> has joined #yocto | 10:28 | |
| *** bluelightning <bluelightning!~paul@pdpc/supporter/professional/bluelightning> has joined #yocto | 10:30 | |
| yocti | New news from stackoverflow: Bitbake command to locate path to installed toolchain <https://stackoverflow.com/questions/54387026/bitbake-command-to-locate-path-to-installed-toolchain> | 10:32 |
|---|---|---|
| *** tprrt <tprrt!~tprrt@ram31-1-82-234-79-177.fbx.proxad.net> has quit IRC | 11:10 | |
| *** slips <slips!~slips@2.51-174-212.customer.lyse.net> has quit IRC | 11:59 | |
| *** bluelightning <bluelightning!~paul@pdpc/supporter/professional/bluelightning> has quit IRC | 12:00 | |
| *** ferlzc <ferlzc!~ferlzc@177.83.99.221> has joined #yocto | 13:23 | |
| *** wooosaiii <wooosaiii!~woo@cpe-90-157-180-95.static.amis.net> has joined #yocto | 14:42 | |
| *** wooosaiiii <wooosaiiii!~woo@cpe-90-157-180-95.static.amis.net> has quit IRC | 14:43 | |
| *** wooosaii <wooosaii!~woo@cpe-90-157-180-95.static.amis.net> has joined #yocto | 14:47 | |
| *** wooosaiii <wooosaiii!~woo@cpe-90-157-180-95.static.amis.net> has quit IRC | 14:49 | |
| *** wooosaiii <wooosaiii!~woo@cpe-90-157-180-95.static.amis.net> has joined #yocto | 14:52 | |
| *** wooosaii <wooosaii!~woo@cpe-90-157-180-95.static.amis.net> has quit IRC | 14:54 | |
| *** BCMM <BCMM!~BCMM@unaffiliated/bcmm> has quit IRC | 15:08 | |
| *** BCMM <BCMM!~BCMM@unaffiliated/bcmm> has joined #yocto | 15:09 | |
| *** chandana73 <chandana73!~ckalluri@149.199.62.131> has joined #yocto | 15:25 | |
| *** sagner <sagner!~ags@2a02:169:34b6::b11> has joined #yocto | 15:39 | |
| *** chandana73 <chandana73!~ckalluri@149.199.62.131> has quit IRC | 15:41 | |
| *** tprrt <tprrt!~tprrt@ram31-1-82-234-79-177.fbx.proxad.net> has joined #yocto | 15:47 | |
| *** nighty- <nighty-!~nighty@b157153.ppp.asahi-net.or.jp> has quit IRC | 16:09 | |
| khem | RP: Can you give a try to top two patches from http://git.openembedded.org/openembedded-core-contrib/log/?h=yoe/mut | 16:39 |
| khem | RP: I want to check if this patch is technically causing no regressions | 16:40 |
| khem | RP: and next week glibc 2.29 is releasing so I will send the patch the day it comes out | 16:40 |
| khem | I have been testing master all along so should be a simple recipe upgrade | 16:40 |
| RP | khem: that patch isn't correct :( | 16:52 |
| RP | khem: I can tell just by looking at it | 16:53 |
| RP | khem: cool on 2.29 :) | 16:53 |
| khem | RP: I think the question is should be allow -march or not | 17:04 |
| khem | I think I agree to provide an option to override MCPU is not ideal | 17:05 |
| RP | khem: I think we have to have the option | 17:05 |
| khem | I was trying to address Andre's feedback | 17:05 |
| RP | khem: the trouble with that patch is it breaks the way the tune code is meant to dynamically adjust | 17:05 |
| khem | RP: so may be add some generic tunes then ? | 17:05 |
| RP | khem: has implications for multilib and perhaps other corner cases | 17:05 |
| khem | and not let people include the arm/arch-arm.inc exclusively | 17:05 |
| RP | khem: your patch is ok, we just need to fix the dynamic adjustment | 17:06 |
| khem | explain to me the dynamic adjustments and I can see what we can do | 17:07 |
| RP | khem: if I were to include some of your tune files but select a base tune, we'd get inconsistent results | 17:15 |
| RP | khem: e.g. tune-arm1136jf-s.inc clears TUNE_MARCH even if that tune isn't selected | 17:15 |
| khem | RP: but thats the point isnt it, arm1136jf-s is a CPU imeplementation based on armv6 architecture | 17:18 |
| khem | so if you are including a tune for this then we are using that as base ISA | 17:18 |
| khem | via -mcpu | 17:18 |
| RP | khem: Its perfectly legal to include 1136jf but then set the selected tune to armv6t | 17:20 |
| khem | yes | 17:20 |
| RP | khem: your patch breaks that | 17:20 |
| khem | that will be added as separate option via -mthumb or -marm option | 17:20 |
| RP | khem: since 1136jf isn't selected, the mcpu option won't be added but with your change, the march won't be either | 17:21 |
| khem | its not represented via -mcpu option | 17:21 |
| khem | or march for that matter | 17:21 |
| RP | khem: well, what option should be passed to the compiler if the "armv6t" tune is selected? | 17:22 |
| khem | -mthumb-interwork as per OE's semantics | 17:22 |
| khem | ISA is set based on another variable | 17:23 |
| khem | ARM_INSTRUCTION_SET | 17:23 |
| RP | khem: so you're saying there shouldn't be any march, mtune or mpcu set for "armv6t" ? | 17:23 |
| khem | yes that was the point I mentioned in email thread as well.. should we demand include of a tune-*.inc file | 17:24 |
| khem | and not tune-arm.inc directly in machine configs | 17:24 |
| RP | khem: no, we shouldn't, that doesn't make sense | 17:24 |
| khem | because that assumes that we need you to define a CPU more than an arch | 17:25 |
| khem | but in a way thats what we are doing today | 17:25 |
| RP | I agree what we;re doing today breaks | 17:25 |
| khem | I am just making it explicit and logically correct | 17:25 |
| RP | no, you break some of it | 17:25 |
| RP | you do fix some cases but break others | 17:26 |
| khem | we expect mcpu to override march but in real its the opposite in gcc | 17:26 |
| RP | khem: on that I agree and understand | 17:26 |
| RP | khem: but the patch as it stands breaks some of the other tunes | 17:26 |
| *** ravi <ravi!~ravi@2a02:908:698:68a0:584:157:c422:7169> has joined #yocto | 17:26 | |
| khem | thats what I was asking, and no one answered | 17:27 |
| RP | khem: I did reply? | 17:27 |
| khem | does someone has cases where they dont include tune* but arch file directly | 17:27 |
| RP | khem: the people who understand this fully can probably be counted on one hand at this point :( | 17:27 |
| RP | We'll see complaints with about a 6-12 month lag as people upgrade | 17:27 |
| khem | right :) but then we need to do the right thing | 17:28 |
| khem | so choice to make | 17:28 |
| khem | do we want to give up some cpu specific opts and use march/mtune and fix it once for all | 17:28 |
| khem | should be leave out the cases where mcpu is not defined in gcc | 17:29 |
| RP | khem: or design this so both work? | 17:29 |
| khem | and lets then define tunes where they are allowed to add -march | 17:29 |
| khem | both could work I have thought of it and that is to use -march= with extentions | 17:30 |
| khem | but it means the package feed arch will be roughly equal | 17:30 |
| khem | to site you an example | 17:30 |
| khem | use -march=armv7ve+fp+neon+vfpv4 | 17:32 |
| khem | instread of -mcpu=cortex-a53 | 17:32 |
| RP | khem: Looking at the code and specifically at 1136jf, could what you want not work if we change TUNE_FEATURES_tune-arm1136jfs = "${TUNE_FEATURES_tune-armv6} arm1136jfs" | 17:33 |
| RP | to be TUNE_FEATURES_tune-arm1136jfs = "arm arm1136jfs" ? | 17:33 |
| khem | we can add the fragments based on tune features | 17:33 |
| RP | that way the armv6 is dropped | 17:33 |
| RP | khem: in other words I think the bug is that armv6 is included there and shouldn't be | 17:34 |
| *** kroon <kroon!~kroon@37-247-29-68.customers.ownit.se> has quit IRC | 17:35 | |
| khem | for armv6 the lowest level of marches are ‘armv6’, ‘armv6j’, ‘armv6k’, ‘armv6kz’, ‘armv6t2’, ‘armv6z’, ‘armv6zk’ | 17:37 |
| khem | so we need to incrementally construction the march | 17:37 |
| khem | where TUNE_FEATURES_tune-armv6 seeds it with 'armv6' | 17:38 |
| khem | and then we keep appending to it based on tune features | 17:38 |
| RP | khem: I think we can do something simpler as above | 17:38 |
| RP | khem: we don't have to append it though | 17:38 |
| RP | that is currently how the files are structured but its the problem | 17:39 |
| RP | khem: nothing says "armv6" has to be in TUNE_FEATURES | 17:39 |
| RP | Its an internal hook on how we construct the values | 17:39 |
| khem | so you think if we break the include chain ? | 17:40 |
| RP | khem: I think it will then do what we want | 17:41 |
| RP | khem: let me see if I can write a patch to test | 17:41 |
| khem | so you mean replace TUNE_FEATURES_tune-arm1136jfs = "${TUNE_FEATURES_tune-armv6} arm1136jfs" with TUNE_FEATURES_tune-arm1136jfs = "armv6 arm1136jfs" | 17:42 |
| RP | khem: no, I don't | 17:43 |
| RP | khem: http://git.yoctoproject.org/cgit.cgi/poky-contrib/commit/?h=rpurdie/t222&id=3200ec0b60a90ee2a10c06f12e463297b20887f6 | 17:44 |
| RP | khem: I just did four examples | 17:44 |
| RP | comments need fixing :) | 17:44 |
| khem | ah I see | 17:44 |
| khem | yes so basically define concrete tunes in tune files | 17:45 |
| khem | define concrete tune features | 17:45 |
| RP | yes | 17:46 |
| RP | its fine to inherit if the values are right but they're not | 17:47 |
| khem | this will flatten the structure and I think I like it but it also means the whole thing will change | 17:47 |
| RP | khem: I think that is ok | 17:48 |
| khem | so idea is define atomic base tune features and then assemble them as needed in tune files | 17:48 |
| RP | khem: yes | 17:48 |
| RP | khem: we don't need to do this everywhere, only where we have the mcpu/march conflicts | 17:49 |
| khem | I think we should do it across to be consistent | 17:49 |
| RP | khem: I disagree, we just need to fix the places its broken. There is no value in duplicating this everywhere, just where it needs to be spelt out | 17:50 |
| khem | since when we use -mcpu and -match together gcc becomes smart and chooses the march but then compares it with internally generated march from mtune | 17:50 |
| RP | khem: lets just add a sanity test which throws a bb.fatal if we see it | 17:50 |
| khem | and complains if it finds it compatible it uses march silently | 17:50 |
| khem | so we will be chasing the ghost always | 17:51 |
| RP | i.e. ban the combination | 17:51 |
| RP | test is trivial | 17:51 |
| RP | khem: we could find out which tunes people are using this way :) | 17:52 |
| khem | so if mcpu and march are in CCARGS then throw a fatal ? | 17:52 |
| * RP is only half joking | 17:52 | |
| RP | khem: yes | 17:52 |
| khem | but that means we have to stop tune-* to recurse into arch-arm* | 17:53 |
| *** kroon <kroon!~kroon@37-247-29-68.customers.ownit.se> has joined #yocto | 17:53 | |
| khem | when it comes to TUNE_CCARGS | 17:53 |
| RP | khem: why? | 17:54 |
| RP | khem: ll we need to do is set TUNE_FEATURES correctly in the broken cases, add the sanity test and I think we're good. | 17:55 |
| *** chandana73 <chandana73!~ckalluri@149.199.62.131> has joined #yocto | 18:01 | |
| RP | khem: Need to head afk but will have a go at a patch later | 18:06 |
| *** gtristan <gtristan!~tristanva@24.48.71.248> has quit IRC | 18:08 | |
| khem | RP: looking at code I think if we avoid adding armv6 to TUNE_FEATURES then -march=armv6 wont be added to armv6 derivatives | 18:18 |
| *** tprrt <tprrt!~tprrt@ram31-1-82-234-79-177.fbx.proxad.net> has quit IRC | 18:21 | |
| *** kroon <kroon!~kroon@37-247-29-68.customers.ownit.se> has quit IRC | 18:26 | |
| RP | khem: I guess I should come up with a patch to show what I mean... | 18:33 |
| khem | RP: yes it will be a biggish patch but I think it will be right thing to do | 18:34 |
| khem | right now this recursive includes means DEFAULTTUNE constructs various settings as it includes other tunes | 18:36 |
| RP | khem: right, that will continue to work | 18:37 |
| *** kroon <kroon!~kroon@37-247-29-68.customers.ownit.se> has joined #yocto | 18:39 | |
| *** WillMiles <WillMiles!~Will@static-209-87-231-80.storm.ca> has joined #yocto | 18:41 | |
| khem | RP: I think we should also change -mfpu=name to use 'auto' | 18:50 |
| khem | so let compiler deduce the right FPU from -mcpu | 18:50 |
| khem | but thats for another day :) | 18:51 |
| *** ravi <ravi!~ravi@2a02:908:698:68a0:584:157:c422:7169> has quit IRC | 18:53 | |
| RP | khem: yes! :) | 18:54 |
| khem | since this will be a prerequisite for that to work | 18:54 |
| khem | RP: btw there is -mcpu=generic-arch e.g. -mcpu=generic-armv7-a | 18:56 |
| khem | which is equivalent to -march=arch -mtune=generic-arch | 18:57 |
| khem | so may be we should replace the -march with these values | 18:59 |
| khem | in arch specific includes | 18:59 |
| RP | khem: not sure which is more confusing! | 19:05 |
| khem | using -mcpu across board is consistent IMO | 19:05 |
| RP | khem: not really convinced... | 19:06 |
| RP | khem: one step at a time I guess, lets get tis first change right | 19:07 |
| khem | but we can again postpone that | 19:08 |
| khem | after getting this first thing going | 19:08 |
| khem | nod | 19:08 |
| khem | using these fairly esoteric options would also mean I have to check what clang is doing :) | 19:09 |
| *** remcycles <remcycles!~remcycles@mobile-166-176-185-247.mycingular.net> has joined #yocto | 19:13 | |
| khem | RP: http://git.openembedded.org/openembedded-core-contrib/commit/?h=yoe/mut&id=3e44bd14417702936421c0e4336ad8bc14ccd791 | 19:28 |
| khem | we need to emulate this behaviour | 19:29 |
| *** WillMiles <WillMiles!~Will@static-209-87-231-80.storm.ca> has quit IRC | 19:32 | |
| *** aleblanc <aleblanc!~aleblanc@2607:f2c0:95d9:8500:5de2:6581:5c40:650> has quit IRC | 19:39 | |
| *** aleblanc <aleblanc!~aleblanc@2607:f2c0:95d9:8500:158d:788b:8f6c:7350> has joined #yocto | 19:40 | |
| khem | and this is portion ( for cortexa5 ) that you were thinking http://git.openembedded.org/openembedded-core-contrib/log/?h=yoe/mut | 19:54 |
| khem | RP: I see http://git.yoctoproject.org/cgit.cgi/poky-contrib/commit/?h=rpurdie/t222 and we also need to append armv* to OVERRIDES there is lot of places where it is used | 20:04 |
| *** sagner <sagner!~ags@2a02:169:34b6::b11> has quit IRC | 20:10 | |
| *** |King_InuYasha| <|King_InuYasha|!~King_InuY@ool-18e49371.dyn.optonline.net> has quit IRC | 20:10 | |
| *** King_InuYasha <King_InuYasha!~King_InuY@fedora/ngompa> has joined #yocto | 20:11 | |
| *** sagner <sagner!~ags@2a02:169:34b6::d58> has joined #yocto | 20:15 | |
| *** RB2 <RB2!~RB2@c-73-178-160-56.hsd1.nj.comcast.net> has quit IRC | 20:23 | |
| *** RB2 <RB2!~RB2@c-73-178-160-56.hsd1.nj.comcast.net> has joined #yocto | 20:25 | |
| *** chandana73 <chandana73!~ckalluri@149.199.62.131> has quit IRC | 20:45 | |
| *** vmeson <vmeson!~rmacleod@24-52-238-240.cable.teksavvy.com> has quit IRC | 21:02 | |
| khem | sent a v3 | 21:04 |
| *** TurBoss <TurBoss!turbossmat@gateway/shell/matrix.org/x-jjkjcaybfguhedrl> has left #yocto | 21:08 | |
| *** TurBoss <TurBoss!turbossmat@gateway/shell/matrix.org/x-ewrigbehjgtytqyo> has joined #yocto | 21:26 | |
| *** andrey_utkin <andrey_utkin!~andrey_ut@gentoo/developer/andrey-utkin> has joined #yocto | 21:32 | |
| *** andrey_utkin <andrey_utkin!~andrey_ut@gentoo/developer/andrey-utkin> has left #yocto | 21:33 | |
| *** andrey_utkin <andrey_utkin!~andrey_ut@gentoo/developer/andrey-utkin> has joined #yocto | 21:33 | |
| *** remcycles <remcycles!~remcycles@mobile-166-176-185-247.mycingular.net> has quit IRC | 21:38 | |
| *** kroon <kroon!~kroon@37-247-29-68.customers.ownit.se> has quit IRC | 21:49 | |
| *** georgem_home <georgem_home!uid210681@gateway/web/irccloud.com/x-dowzvmkjsvmspzbn> has joined #yocto | 21:50 | |
| *** andrey_utkin <andrey_utkin!~andrey_ut@gentoo/developer/andrey-utkin> has quit IRC | 21:55 | |
| *** camus <camus!~Instantbi@183.128.236.162> has joined #yocto | 22:02 | |
| *** kaspter <kaspter!~Instantbi@183.128.236.162> has quit IRC | 22:06 | |
| *** camus is now known as kaspter | 22:06 | |
| *** chandana73 <chandana73!~ckalluri@149.199.62.130> has joined #yocto | 22:07 | |
| *** ravi <ravi!~ravi@2a02:908:698:68a0:584:157:c422:7169> has joined #yocto | 22:18 | |
| *** everclear <everclear!~martin@dslb-178-002-165-167.178.002.pools.vodafone-ip.de> has quit IRC | 22:20 | |
| * armpit hahah, now thats a first.. eclipse built but nothing else | 22:24 | |
| *** sagner <sagner!~ags@2a02:169:34b6::d58> has quit IRC | 22:27 | |
| *** ravi <ravi!~ravi@2a02:908:698:68a0:584:157:c422:7169> has quit IRC | 22:28 | |
| RP | khem: just saw the patch, beat me to it. Looks good | 22:29 |
| RP | khem: fired a test run (inc the valgrind fix) | 22:36 |
| armpit | RP, stable/thud-next clean on qa-full | 23:07 |
| * armpit working on sumo-next with testimage and qa updates | 23:07 | |
| armpit | RP, will send pull request later | 23:08 |
| *** vmeson <vmeson!~rmacleod@24-52-238-240.cable.teksavvy.com> has joined #yocto | 23:09 | |
| *** andrey_utkin <andrey_utkin!~andrey_ut@gentoo/developer/andrey-utkin> has joined #yocto | 23:36 | |
Generated by irclog2html.py 2.11.0 by Marius Gedminas - find it at mg.pov.lt!