#dri-devel on 2021-10-29 — irc logs at oftc.irclog.whitequark.org

2021-07-26 22:56 ChanServ changed the topic of #dri-devel to: <ajax> nothing involved with X should ever be unable to find a bar

00:15 Lucretia has quit []

00:48 camus has joined #dri-devel

00:53 camus1 has quit [Ping timeout: 480 seconds]

01:00 pnowack has quit [Quit: pnowack]

01:06 columbarius has joined #dri-devel

01:07 co1umbarius has quit [Ping timeout: 480 seconds]

01:09 ybogdano has quit [Ping timeout: 480 seconds]

01:16 shashank1202_ has quit [Quit: Connection closed for inactivity]

01:25 rasterman has quit [Quit: Gettin' stinky!]

01:36 nchery has quit [Quit: Leaving]

01:59 camus has quit [Remote host closed the connection]

01:59 camus has joined #dri-devel

02:04 orbea has quit [Read error: Connection reset by peer]

02:06 agners has quit [Remote host closed the connection]

02:07 agners has joined #dri-devel

02:07 orbea has joined #dri-devel

02:11 boistordu_old has joined #dri-devel

02:12 f11f13 has quit [Remote host closed the connection]

02:13 f11f13 has joined #dri-devel

02:17 boistordu_ex has quit [Ping timeout: 480 seconds]

02:24 mclasen has quit [Ping timeout: 480 seconds]

02:24 The_Company has quit []

02:28 lemonzest has joined #dri-devel

02:44 kenjigashu has joined #dri-devel

03:02 camus1 has joined #dri-devel

03:06 camus has quit [Ping timeout: 480 seconds]

03:25 vivijim has quit [Ping timeout: 480 seconds]

03:34 <jekstrand> dcbaker: Can I get your eyes on https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13156 tomorrow?

03:34 <jekstrand> dcbaker: It should be an easy review and maybe you can help me untangle the python a bit. We really need to land that before my sync stuff.

03:35 gpuman has joined #dri-devel

03:57 mattrope has quit [Remote host closed the connection]

03:59 aravind has joined #dri-devel

04:19 rpigott has quit [Read error: Connection reset by peer]

04:20 rpigott has joined #dri-devel

04:22 <dcbaker> jekstrand: sure, I'll have a look tomorrow

04:49 kenjigashu has quit [Ping timeout: 480 seconds]

04:50 Duke`` has joined #dri-devel

05:11 craftyguy has quit [Remote host closed the connection]

05:12 craftyguy has joined #dri-devel

05:13 craftyguy has quit []

05:16 craftyguy has joined #dri-devel

05:16 craftyguy has quit []

05:18 craftyguy has joined #dri-devel

05:33 itoral has joined #dri-devel

05:33 sdutt has quit [Read error: Connection reset by peer]

05:35 shashank1202_ has joined #dri-devel

05:38 reductum has quit [Remote host closed the connection]

05:39 reductum has joined #dri-devel

05:39 rcf has quit [Remote host closed the connection]

05:39 rcf has joined #dri-devel

05:39 sadlerap has quit [Remote host closed the connection]

05:39 kallisti5 has quit [Remote host closed the connection]

05:39 kallisti5 has joined #dri-devel

05:41 sadlerap has joined #dri-devel

05:55 i-garrison has joined #dri-devel

05:59 Hi-Angel has joined #dri-devel

06:00 nroberts has quit [Quit: Gxis!]

06:07 pnowack has joined #dri-devel

06:08 <mareko> I kinda have a tex indirection scheduler in NIR/SSA, it reorders everything and groups tex opcodes based on their indirection level

06:21 danvet has joined #dri-devel

06:24 camus has joined #dri-devel

06:25 camus1 has quit [Remote host closed the connection]

06:27 alanc has quit [Remote host closed the connection]

06:32 thellstrom has joined #dri-devel

06:33 alanc has joined #dri-devel

06:51 f11f13 has quit [Remote host closed the connection]

06:51 f11f13 has joined #dri-devel

07:10 camus1 has joined #dri-devel

07:10 camus has quit [Read error: Connection reset by peer]

07:11 <mripard> airlied: there's no new patch in drm-misc-fixes, you can enjoy your day off :)

07:16 bgs has quit [Ping timeout: 480 seconds]

07:21 bgs has joined #dri-devel

07:26 cef has joined #dri-devel

07:39 jkrzyszt has joined #dri-devel

07:39 camus has joined #dri-devel

07:40 camus1 has quit [Ping timeout: 480 seconds]

07:44 <ishitatsuyuki> trying out the GOOG present timing patch and the timestamps seems to freak out with VSync off (faster than refresh rendering)

07:45 <ishitatsuyuki> does XCB_PRESENT_EVENT_IDLE_NOTIFY always happen after XCB_PRESENT_EVENT_COMPLETE_NOTIFY? I suspect that a swapchain image is being reused before the timings are finalized

07:46 rgallaispou has joined #dri-devel

08:02 f11f13 has quit [Remote host closed the connection]

08:03 f11f13 has joined #dri-devel

08:03 <MrCooper> ishitatsuyuki: there's no defined ordering between those events

08:04 <ishitatsuyuki> I guess I need to clarify the question

08:04 bgs has quit [Read error: Connection reset by peer]

08:04 <MrCooper> when not synchronizing to vblank, the completion timestamp may correspond to the end of last vblank

08:05 <ishitatsuyuki> can the IDLE event arrive first without being immediately followed by a COMPLETE?

08:05 <MrCooper> yes

08:05 <ishitatsuyuki> so yeah that's the problem

08:06 camus1 has joined #dri-devel

08:06 <ishitatsuyuki> will comment on !12086 later

08:06 <MrCooper> IDLE signals when the source pixmap buffer can be reused, which can be significantly earlier than when the presentation actually completes

08:07 bgs has joined #dri-devel

08:10 camus has quit [Ping timeout: 480 seconds]

08:16 <ishitatsuyuki> actually, the code says `if (render_time > ust) render_time = ust;`. not sure why but having timestamps backwards in time seems to be the intended behavior...

08:19 aravind has quit [Ping timeout: 480 seconds]

08:20 camus has joined #dri-devel

08:20 linearcannon has quit [Read error: Connection reset by peer]

08:22 linearcannon has joined #dri-devel

08:22 camus1 has quit [Ping timeout: 480 seconds]

08:23 <MrCooper> intended behaviour of what? (No context for what render_time / ust mean either)

08:24 pochu has joined #dri-devel

08:27 linearcannon is now known as Guest4346

08:28 linearcannon has joined #dri-devel

08:32 Guest4346 has quit [Ping timeout: 480 seconds]

08:33 gawin has joined #dri-devel

08:35 pcercuei has joined #dri-devel

08:37 <ishitatsuyuki> https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12086#note_1122868

08:39 hikiko has quit [Quit: Leaving!]

08:48 join_subline has quit [Ping timeout: 480 seconds]

09:05 rasterman has joined #dri-devel

09:09 ella-0 has quit []

09:10 ella-0 has joined #dri-devel

09:17 gawin has quit [Ping timeout: 480 seconds]

09:19 camus1 has joined #dri-devel

09:23 camus has quit [Ping timeout: 480 seconds]

09:31 Lucretia has joined #dri-devel

09:52 aravind has joined #dri-devel

10:17 nikitalita4 has quit [Ping timeout: 480 seconds]

10:26 nikitalita4 has joined #dri-devel

10:36 kts has quit [Quit: Konversation terminated!]

10:37 gawin has joined #dri-devel

10:38 flacks has quit [Quit: Quitter]

10:38 hansg has joined #dri-devel

10:41 flacks has joined #dri-devel

10:49 frieder has joined #dri-devel

11:04 join_subline has joined #dri-devel

11:04 hansg has quit [Quit: Leaving]

11:08 tarceri has quit [Remote host closed the connection]

11:08 tarceri has joined #dri-devel

11:21 frieder has quit [Remote host closed the connection]

11:26 mclasen has joined #dri-devel

11:32 cef has quit [Ping timeout: 480 seconds]

11:37 camus has joined #dri-devel

11:38 camus1 has quit [Read error: Connection reset by peer]

11:44 shashank1202_ has quit [Quit: Connection closed for inactivity]

11:45 <arnd> I saw a build regression with clang-14 that was fixed in linux-next with https://patchwork.freedesktop.org/patch/459304/, but now the patch is gone from linux-next again?

11:45 <arnd> is this intentional, or did some stuff get dropped by accident?

12:01 probablymoony has quit [Remote host closed the connection]

12:02 moony has joined #dri-devel

12:04 aravind has quit [Ping timeout: 480 seconds]

12:05 vivijim has joined #dri-devel

12:07 dliviu has quit [Ping timeout: 480 seconds]

12:10 dliviu has joined #dri-devel

12:13 tarceri has quit [Remote host closed the connection]

12:14 tarceri has joined #dri-devel

12:26 cef has joined #dri-devel

12:32 itoral has quit []

12:35 camus1 has joined #dri-devel

12:39 camus has quit [Ping timeout: 480 seconds]

12:41 gpuman_ has joined #dri-devel

12:42 gpuman has quit [Read error: Connection reset by peer]

12:44 mclasen has quit [Remote host closed the connection]

12:45 mclasen has joined #dri-devel

13:04 <cwabbott> jekstrand: so apparently you broke image barriers on qualcomm with the nir_var_mem_image MR and CI didn't notice

13:05 <cwabbott> I guess the compiler doesn't reorder stuff so it just led to a tiny bit of flakiness

13:06 nsneck has joined #dri-devel

13:06 ella-0 has quit []

13:07 ella-0 has joined #dri-devel

13:09 <cwabbott> jekstrand: actually, I should say image barriers on turnip, since you forgot to fixup the scoped barrier handling

13:11 <cwabbott> I made !13593

13:32 shashank1202_ has joined #dri-devel

13:41 Company has joined #dri-devel

13:43 Lucretia has quit []

13:46 camus1 has quit []

13:47 Lucretia has joined #dri-devel

13:53 alyssa has joined #dri-devel

13:57 gawin has quit [Ping timeout: 480 seconds]

14:01 sdutt has joined #dri-devel

14:05 sdutt has quit []

14:05 sdutt has joined #dri-devel

14:15 <jekstrand> cwabbott: ack

14:16 <jekstrand> dcbaker: Thanks! I've got quite a few vulkan common runtime MRs now and they're starting to pile up. That one's not necessarily block development but it does block merging for some of them.

14:16 tarceri has quit [Remote host closed the connection]

14:17 tarceri has joined #dri-devel

14:38 kmn has quit [Ping timeout: 480 seconds]

14:40 vivijim has quit [Read error: Connection reset by peer]

14:46 vivijim has joined #dri-devel

14:56 <xlei> 3

15:00 mattrope has joined #dri-devel

15:01 fcarrijo has joined #dri-devel

15:04 fcarrijo has quit []

15:04 jewins has joined #dri-devel

15:21 <jekstrand> .14

15:37 hansg has joined #dri-devel

15:39 gpuman_ has quit [Remote host closed the connection]

15:40 gpuman has joined #dri-devel

15:40 shashank1202_ has quit [Quit: Connection closed for inactivity]

15:56 tarceri has quit [Remote host closed the connection]

15:56 rgallaispou has quit [Ping timeout: 480 seconds]

15:56 tarceri has joined #dri-devel

16:08 luckyxxl has joined #dri-devel

16:12 vivijim has quit [Quit: Lost terminal]

16:13 luckyxxl_ has joined #dri-devel

16:19 luckyxxl has quit [Ping timeout: 480 seconds]

16:24 f11f13 has quit [Remote host closed the connection]

16:24 f11f13 has joined #dri-devel

16:28 luckyxxl has joined #dri-devel

16:35 luckyxxl_ has quit [Ping timeout: 480 seconds]

16:50 <dcbaker> jekstrand: review left

16:50 <dcbaker> It's very strange to read untyped python code

16:51 <jekstrand> dcbaker: untyped?

16:51 <dcbaker> without type annotations

16:56 <jekstrand> dcbaker: Thanks!

16:56 luckyxxl has quit [Ping timeout: 480 seconds]

16:56 <jekstrand> dcbaker: Any suggestions for pulling the runtime-specfici generators into runtime/ while importing from util/?

16:57 <jekstrand> This feels like something dependencies should be able to sort out....

16:57 <jekstrand> Maybe meson needs a python_path field in dependency()

16:57 thellstrom has quit [Quit: thellstrom]

16:59 <jekstrand> dcbaker: What I did in that MR (which is kind-of rubbish) is to leave the generators in util/ even if they're invoked from runtime/

17:00 <dcbaker> I almost feel like we just just put all of the generators in a gen/ folder and just call them from wherever we want

17:00 <jekstrand> :-/

17:01 <dcbaker> custom_target has an `env` option

17:01 <dcbaker> you could build an `Environment` object with `PYTHON_PATH` set

17:02 vivijim has joined #dri-devel

17:08 gouchi has joined #dri-devel

17:10 <jekstrand> can one stick arbitrary stuff in a dependency object?

17:11 kenjigashu has joined #dri-devel

17:11 kenjigashu has quit [Remote host closed the connection]

17:11 <dcbaker> You can stick whatever is listed here: https://jobs.intel.com/ShowJob/Id/3005363/Graphics-Software-Engineer

17:12 <dcbaker> or, here: https://mesonbuild.com/Reference-manual_functions.html#declare_dependency

17:12 <dcbaker> sigh, why does copy and highlight paste not do the same thing?

17:12 <jekstrand> copy+paste is hard

17:12 <dcbaker> so you can attach random string:string mappings in the variables

17:12 <dcbaker> key, but otherwise not really

17:13 <jekstrand> Ok, that might work.

17:13 <jekstrand> Can I declare my own functions?

17:13 <dcbaker> in meson? no meson doesn't have functions by design

17:14 <jekstrand> Hrm... Maybe it's ok without

17:14 <dcbaker> we've been talking about custom_target templates which might also be a solution here

17:14 orbea has quit [Ping timeout: 480 seconds]

17:14 <jekstrand> Here's what I'm thinking:

17:14 gawin has joined #dri-devel

17:15 <jekstrand> Declare a dependency which lists the python files that would be imported to use that dependency as the sources

17:15 <dcbaker> ie, you could write something like:... (full message at https://matrix.org/_matrix/media/r0/download/matrix.org/YPVGfDieqbhQTzlqVlAAugjn)

17:15 <jekstrand> The dep would also have a python_path in the string:string map

17:17 <dcbaker> Hold on, there might be a way to do this

17:19 <gawin> what is easiest way to run deqp tests now? (that linked repo is in rough shape)

17:20 <gawin> also how to use piglit with profiles(?)? I mean these "ci" folders

17:20 <dcbaker> jekstrand: So, there's this basically undocumented feature that is actually what you want (I think):... (full message at https://matrix.org/_matrix/media/r0/download/matrix.org/aytQPcFZZvuDwvHUevcGssQn)

17:20 <dcbaker> you can pass a dictionary of keyword arguments as a dictionary

17:21 <jekstrand> dcbaker: Matrix is giving me links to your messages, FYI

17:21 <dcbaker> you can pass keyword options as a dictionary

17:21 <dcbaker> so you can put the Environment object with the PYTHON_PATH and depend_files in a dictionary, then pass that to each custom_target

17:22 <dcbaker> basically like a dependency, but you can actually just put whatever you want in it

17:24 <jekstrand> Right

17:24 <jekstrand> It just won't combine well if you have two of them

17:24 ybogdano has joined #dri-devel

17:26 <dcbaker> meson dicts can be added together, though I don't know if that merges or replaces tbh

17:27 <jekstrand> It won't merge PYTHON_PATH properly

17:27 <dcbaker> seems likely

17:28 <jekstrand> The template thing sounds like it could work

17:28 <jekstrand> but I'm not sure

17:28 <dcbaker> plus I'd have to write the code :D

17:28 <dcbaker> since that's still a "we should do this, it would be useful"

17:31 orbea has joined #dri-devel

17:47 ybogdano has quit [Ping timeout: 480 seconds]

17:48 orbea has quit [Ping timeout: 480 seconds]

17:49 anujp has joined #dri-devel

17:50 Peste_Bubonica has joined #dri-devel

17:50 pochu has quit [Quit: leaving]

17:52 orbea has joined #dri-devel

17:53 fxkamd has joined #dri-devel

17:56 ybogdano has joined #dri-devel

18:00 mbrost has joined #dri-devel

18:01 f11f13 has quit [Read error: Connection reset by peer]

18:01 f11f13 has joined #dri-devel

18:06 orbea has quit [Quit: You defeated orbea! 2383232 XP gained!]

18:08 orbea has joined #dri-devel

18:18 <jekstrand> dcbaker: I've updated with your suggestions. A couple of them are in new commits.

18:27 lemonzest has quit [Quit: WeeChat 3.2]

18:28 gpoo has quit [Ping timeout: 480 seconds]

18:28 gpoo has joined #dri-devel

18:37 garrison has joined #dri-devel

18:37 i-garrison has quit [Read error: Connection reset by peer]

18:37 tzimmermann has joined #dri-devel

18:45 tzimmermann has quit [Quit: Leaving]

18:48 gpoo has quit [Ping timeout: 480 seconds]

18:59 gpoo has joined #dri-devel

19:06 garrison has quit []

19:11 tarceri has quit [Remote host closed the connection]

19:11 tarceri has joined #dri-devel

19:59 hansg has quit [Quit: Leaving]

20:16 dllud has quit [Ping timeout: 480 seconds]

20:21 dllud has joined #dri-devel

20:22 X-Scale` has joined #dri-devel

20:28 Net147 has quit [Ping timeout: 480 seconds]

20:28 X-Scale has quit [Ping timeout: 480 seconds]

20:29 <jekstrand> dcbaker: ping ^^

20:36 <zmike> dcbaker: any chance you could check out the windows fail on https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/13588 ? rumor has it this is a meson dependency issue

20:36 doras has joined #dri-devel

20:39 Net147 has joined #dri-devel

20:43 <dcbaker> zmike: that sure looks like you need to add `sha1_h` to the sources libllvmpipe sources

20:50 Duke`` has quit [Ping timeout: 480 seconds]

20:52 jkrzyszt has quit [Ping timeout: 480 seconds]

20:55 i-garrison has joined #dri-devel

21:08 f11f11 has joined #dri-devel

21:14 f11f13 has quit [Ping timeout: 480 seconds]

21:22 Net147 has quit [Read error: Connection reset by peer]

21:23 Net147 has joined #dri-devel

21:29 danvet has quit [Ping timeout: 480 seconds]

21:36 ybogdano has quit [Ping timeout: 480 seconds]

21:37 Bennett has joined #dri-devel

21:38 co1umbarius has joined #dri-devel

21:40 columbarius has quit [Ping timeout: 480 seconds]

21:47 Peste_Bubonica has quit [Remote host closed the connection]

21:48 craftyguy has quit [Ping timeout: 480 seconds]

21:49 tarceri has quit [Remote host closed the connection]

21:49 tarceri has joined #dri-devel

22:01 f11f11 has quit [Remote host closed the connection]

22:02 f11f11 has joined #dri-devel

22:02 anarsoul has quit [Quit: ZNC 1.8.2 - https://znc.in]

22:02 alyssa has quit [Quit: Thanks for all the fish]

22:03 anarsoul has joined #dri-devel

22:18 Hi-Angel has quit [Ping timeout: 480 seconds]

22:34 agners has quit [Quit: WeeChat 3.3]

22:43 gouchi has quit [Remote host closed the connection]

22:51 shashank1202_ has joined #dri-devel

22:55 Net147 has quit [Ping timeout: 480 seconds]

23:00 craftyguy has joined #dri-devel

23:00 Net147 has joined #dri-devel

23:10 ybogdano has joined #dri-devel

23:11 gawin has quit [Quit: Konversation terminated!]

23:15 pcercuei has quit [Quit: dodo]

23:17 Net147_ has joined #dri-devel

23:21 Net147 has quit [Ping timeout: 480 seconds]

23:26 X-Scale has joined #dri-devel

23:30 rasterman has quit [Quit: Gettin' stinky!]

23:31 X-Scale` has quit [Ping timeout: 480 seconds]