Page MenuHomePhabricator

raster (Carsten Haitzler (Rasterman))Administrator
Keeper of the rabid stoat

Projects (9)

Today

  • Clear sailing ahead.

Tomorrow

  • Clear sailing ahead.

Tuesday

  • Clear sailing ahead.

User Details

User Since
Jan 27 2013, 4:24 PM (320 w, 6 d)
Roles
Administrator
Availability
Available

Behold... the rabid stoat! Prostrate yourselves before it before it is too late!

Recent Activity

Tue, Mar 19

raster accepted D8106: canvas map: introduce a new texture mapping for better quality..

aaah ok. that's a good selection there. thanks! :) so impact can vary from almost nothing to "pretty big". at least it's an option - aa and smooth on at the same time i think is good enough. if it was not that much more i was going to suggest we adopt it as the default "smooth". :)

Tue, Mar 19, 4:06 PM · efl
raster raised the priority of T7685: Video wallpapers for e broken - doesn't seem to propagate update rects from High to Showstopper Issues.
Tue, Mar 19, 4:03 PM · efl

Fri, Mar 15

raster added a comment to T7360: Evas/Edje animations not in sync .

I'm not going to have time for a few weeks...

Fri, Mar 15, 2:54 AM · efl: layout engine, efl: rendering, efl (efl-1.22), regression

Tue, Mar 12

raster triaged T7737: efl-git 1.22.0beta1.61484 + enlightenment-git: icon menu properties not diplayed (wayland) as High priority.
Tue, Mar 12, 10:39 PM · enlightenment-git, efl (efl-1.22)

Sun, Mar 10

raster added a comment to T7734: windows : could not open src\lib\ecore\efl_loop_timer.eo.legacy.h.eo.legacy.h.

i can't reproduce... now that autotools is working enough again...

Sun, Mar 10, 6:16 AM · Restricted Project
raster added a comment to T7734: windows : could not open src\lib\ecore\efl_loop_timer.eo.legacy.h.eo.legacy.h.

autotools was broken for me anyway. so i couldn't build with it. iut literally deleted files committed to git and failed to build part way thru

Sun, Mar 10, 5:05 AM · Restricted Project

Sat, Mar 9

raster abandoned D8099: evas: make convert_rgb_32 keep consistency.
Sat, Mar 9, 7:22 AM · efl
raster commandeered D8099: evas: make convert_rgb_32 keep consistency.

going to close this as 4758f06e637239f981eedbaaf8c0d613b78e4417 solves it better by moving to tiled rendering

Sat, Mar 9, 7:22 AM · efl
raster added a comment to D8099: evas: make convert_rgb_32 keep consistency.

on arm32 the tiled rotate is 2.4x faster vs the neon. the neon interestingly is also about 1.8x faster than the plain C code.

Sat, Mar 9, 6:37 AM · efl
raster requested changes to D8099: evas: make convert_rgb_32 keep consistency.

ok i checked on arm32 and yeah - the neon is broken. but... it's half the speed of the tiled rotator... so why fix something that is slow when we can just switch to the faster one. :) i think its best to reject this and instead just remove any rotate code/paths that are not the tiled rotates... :)

Sat, Mar 9, 5:49 AM · efl

Fri, Mar 8

raster added a comment to D8095: Add json-c dependency.

forecasts doesn't work for me - only says it supports us zip codes (and i don't live there) and can't enter anything in the entry box there in settings anyway, so can't test it.

Fri, Mar 8, 7:27 AM
raster added a comment to D8099: evas: make convert_rgb_32 keep consistency.

justr one thing i noted... why aren't we just doing tiled rotation by default anyway? in terms of pixels/sec its almost 2x the speed vs ye olde C. i'm not sure the neon would even be 2x the speed of C... so... anyway. i need to check more...

Fri, Mar 8, 4:59 AM · efl
raster added a comment to D8099: evas: make convert_rgb_32 keep consistency.

argh... just realized this isn't used on aarch64 - only 32bit arm... i need more time to check on that as now my more accessible arms are all aarch64 :)

Fri, Mar 8, 4:57 AM · efl

Thu, Mar 7

raster added a comment to D8106: canvas map: introduce a new texture mapping for better quality..

what's the performance difference on a range of devices/architectures?

Thu, Mar 7, 12:35 AM · efl

Wed, Mar 6

raster added a comment to D8103: efl.file: use ENOENT as default error if file opening fails and errno isn't set.

do we need this if we have D8104 ?

Wed, Mar 6, 2:17 AM · efl

Tue, Mar 5

raster added a comment to D8099: evas: make convert_rgb_32 keep consistency.

give me a bit... just saw this and need to look at it more, BUT the if ((w & 1) || (h & 1)) is a simple "if width or height is odd (not a multiple of 2) in size). this is kind of lazy as the neon can still be used EXCEPT for the "edge pixels" where it can fall back to some C. I'll need to read this a bit carefully to figure out what's wrong (and maybe try some things out).

Tue, Mar 5, 4:38 PM · efl

Mon, Mar 4

raster closed T7097: something broken with gl when lots of windows open with x11 engine as Wontfix.

probably not even an error we can fix. this is probably something to bring up with the appropriate driver developers, but even then may never be fixed. also note you may hit a limit at 126/127/128 or so or maybe 254/255 as xorg has a limit on client connections and last time i hit it, it was 128 clients (e will have 1 connection at least itself to start). so you'll hit a wall/limit like that that will probably be far before you hit a ram limit these days. as i said the 20/30 limit smells like a driver/gpu limit in maximum number of contexts or something similar. we impose no limit in efl and even if we did it's be per process, not globally.

Mon, Mar 4, 11:23 AM · efl: display system, efl: rendering

Fri, Mar 1

raster added a comment to T7710: Impossible to build EFL on FreeBSD.

i couldn't say anymore - fbsd has since my build shot itself in the face. perhaps i did have some inotify lib installed... but i managed to build efl, e, rage and terminology...

Fri, Mar 1, 4:34 PM · efl (efl-1.22)
raster added a comment to T7710: Impossible to build EFL on FreeBSD.

err sorry

Fri, Mar 1, 10:51 AM · efl (efl-1.22)
raster added a comment to T7710: Impossible to build EFL on FreeBSD.

are you building with meson there? looks like it.

Fri, Mar 1, 8:21 AM · efl (efl-1.22)
raster added a comment to T7710: Impossible to build EFL on FreeBSD.

bbdb2e5c23139d96b6fa57633f9a73e85ea6e7d6 i hope should fix this. i also found a few other loose ends regarding environ related things and fixed those too. my fbsd vm tho is having other issues (meson cant find libintl.h and it isn't respecting CFLAGS when looking for libintl.h ...:() so my fix above is a guess, but a decent one based on the fact that efl_core_proc_env.c was indeed missing the extern char **environ declaration and uour comment above that it works.

Fri, Mar 1, 2:36 AM · efl (efl-1.22)

Wed, Feb 27

raster added a comment to T7710: Impossible to build EFL on FreeBSD.

likely related to @bu5hm4n 's changes

Wed, Feb 27, 12:30 AM · efl (efl-1.22)
raster assigned T7710: Impossible to build EFL on FreeBSD to bu5hm4n.
Wed, Feb 27, 12:29 AM · efl (efl-1.22)

Tue, Feb 26

raster added a comment to D7926: Eina: implement strtod in C locale and remove linkl against msvcr100..

see inline questions :)

Tue, Feb 26, 2:37 PM · efl

Sun, Feb 24

raster added a comment to T7707: termpty: compute exact line length in termpty_cellrow_get() breaks jed display.

hehehe yeah - tho .. xterm vs terminology should be enough to compare :)

Sun, Feb 24, 12:47 PM · Terminology
raster triaged T7707: termpty: compute exact line length in termpty_cellrow_get() breaks jed display as High priority.
Sun, Feb 24, 3:20 AM · Terminology

Feb 15 2019

raster added a comment to D7926: Eina: implement strtod in C locale and remove linkl against msvcr100..

keep the tests simple then - don't go too far in trying to test it as it just can't pass on all platforms. like you say - with large numbers it might not be that useful etc.

Feb 15 2019, 2:42 AM · efl
raster added a comment to D7934: Efl.Gfx.Stack: Rename raise and lower.

actually raise and lower mean to top and bottom commonly. xrwaisewindow and xlowerwindow certainly do. in Qt its raise() and lower() methods.

Feb 15 2019, 2:35 AM · efl

Feb 13 2019

raster updated the task description for T6726: E/EFL Flat Theme.
Feb 13 2019, 7:37 AM · efl: widgets, Restricted Project, enlightenment-git, Restricted Project
raster updated the task description for T6726: E/EFL Flat Theme.
Feb 13 2019, 7:32 AM · efl: widgets, Restricted Project, enlightenment-git, Restricted Project
raster updated the task description for T6726: E/EFL Flat Theme.
Feb 13 2019, 7:31 AM · efl: widgets, Restricted Project, enlightenment-git, Restricted Project
raster updated the task description for T6726: E/EFL Flat Theme.
Feb 13 2019, 7:25 AM · efl: widgets, Restricted Project, enlightenment-git, Restricted Project
raster updated the task description for T6726: E/EFL Flat Theme.
Feb 13 2019, 7:23 AM · efl: widgets, Restricted Project, enlightenment-git, Restricted Project
raster updated the task description for T6726: E/EFL Flat Theme.
Feb 13 2019, 6:32 AM · efl: widgets, Restricted Project, enlightenment-git, Restricted Project

Feb 12 2019

raster added a comment to D7516: ecore: here comes a command line object.

this makes the c definitely worse exactly as i described and it is not niche usage. the ownership is even mixed up between strings and the array.

Feb 12 2019, 12:15 PM · efl

Feb 10 2019

raster added a commit to T7687: alt-tab switching order is broken: rEc1c3af7412e1: fix alt tab winlist focus ordering.
Feb 10 2019, 3:32 AM · enlightenment-git
raster closed T7687: alt-tab switching order is broken as Resolved.

fixed by c1c3af7412e1f451bbd7f4df57226f4847d1b446

Feb 10 2019, 3:32 AM · enlightenment-git
raster added a comment to T7687: alt-tab switching order is broken.

oh this is very simple it seems... there's a double "next client" when showing the winlist... :) it's not efl - it's an e thing.

Feb 10 2019, 3:30 AM · enlightenment-git

Feb 8 2019

raster updated the task description for T7690: drm (both gl and sw) are broken with rotation.
Feb 8 2019, 9:04 AM · efl
raster triaged T7690: drm (both gl and sw) are broken with rotation as High priority.
Feb 8 2019, 9:03 AM · efl
raster created T7687: alt-tab switching order is broken.
Feb 8 2019, 1:45 AM · enlightenment-git

Feb 7 2019

raster triaged T7685: Video wallpapers for e broken - doesn't seem to propagate update rects as High priority.
Feb 7 2019, 6:18 AM · efl

Feb 2 2019

raster closed T7674: setting wallpaper for lock screen is broken using non edj files as Resolved by committing rE11b0a6e6708a: theme wallapaper other than edj fix.
Feb 2 2019, 2:43 AM · enlightenment-git
raster added a comment to T7674: setting wallpaper for lock screen is broken using non edj files.

pointing me to this leads me to "argh there is broken stuff here". just the way that the bg is placed in the canvas, how it decides to do multiple screen wallpapers etc. - i think i need to redo a lot of this code... but i've fixed this immediate issue

Feb 2 2019, 2:39 AM · enlightenment-git

Feb 1 2019

raster closed D7834: replace hton and ntoh family functions with ones defined in eina.
Feb 1 2019, 6:30 AM · efl
raster accepted D7834: replace hton and ntoh family functions with ones defined in eina.

i had to make fixes as you removed some include blocks that are needed still for other inet funcs and missed adding linking/headers to other code.

Feb 1 2019, 6:26 AM · efl
raster closed D7858: efl-cxx: Fix compilation error when using a ptr to const any_value.
Feb 1 2019, 2:51 AM · efl
raster accepted D7858: efl-cxx: Fix compilation error when using a ptr to const any_value.

fixed. tnx.

Feb 1 2019, 2:51 AM · efl

Jan 30 2019

raster added a comment to T7097: something broken with gl when lots of windows open with x11 engine.

i don't even know that we get an error or can detect it. we do things like check the return of glxcreatewindow and error out and complain what failed. it seemingly is not failing. if we don't see things failing from the glx/egl api then we can't do anything about it.

Jan 30 2019, 3:24 PM · efl: display system, efl: rendering
raster added a comment to T7662: Flat theme - notification popup cant be close.

;)

Jan 30 2019, 2:35 AM · Restricted Project
raster closed T7662: Flat theme - notification popup cant be close as Resolved by committing rEFLdb0d0d706cd5: TH - fix popup close.
Jan 30 2019, 2:10 AM · Restricted Project

Jan 29 2019

raster added a comment to T7662: Flat theme - notification popup cant be close.

what popup?

Jan 29 2019, 3:10 AM · Restricted Project

Jan 25 2019

raster added a comment to D7736: Batman Udev: Don't include batteries who have unknown status, have never had a full charge, and have no charge rate..

I have some data for hid devices with batteries. these are real devices with rechargeable batteries and ... I actually see proper reports. Note that my mouse (MX Master) showed "Unknown" until I unplugged it to run wirelessly for a while - then it started showing Full as the charge level. The Craft Keyboard shows Full. Here they were plugged in:

Jan 25 2019, 2:12 AM · enlightenment-git

Jan 24 2019

raster added a comment to D7514: efl_task: remove env from this object.

Saying "that's what OO is about - you inherit and modify the behavior of things appropriately for that object type. ", without any form of constraint is just wrong and dangerous.

Jan 24 2019, 11:52 AM · efl
raster requested changes to D7516: ecore: here comes a command line object.

as cedric said. it should be properties.

Jan 24 2019, 11:19 AM · efl
raster updated the task description for T6726: E/EFL Flat Theme.
Jan 24 2019, 2:07 AM · efl: widgets, Restricted Project, enlightenment-git, Restricted Project

Jan 23 2019

raster accepted D7736: Batman Udev: Don't include batteries who have unknown status, have never had a full charge, and have no charge rate..
Jan 23 2019, 9:06 AM · enlightenment-git

Jan 21 2019

raster accepted D7706: elm.hoversel: ensure sd is not NULL.

though one question - when does this happen? like during shutdown or something like that?

Jan 21 2019, 5:30 AM · efl

Jan 16 2019

raster closed D7569: mixer: Add cards settings into pulseaudio backend.
Jan 16 2019, 9:30 AM · enlightenment-git
raster closed T3127: Add Configure option to emixer. as Resolved by committing rE1527599b789f: mixer: Add cards settings into pulseaudio backend.
Jan 16 2019, 9:30 AM · enlightenment-git
raster accepted D7569: mixer: Add cards settings into pulseaudio backend.

thumbs up

Jan 16 2019, 9:29 AM · enlightenment-git
raster updated the summary of D7646: elm_config: Free data returned from eet_read.
Jan 16 2019, 8:09 AM · efl
raster accepted D7646: elm_config: Free data returned from eet_read.

ACK - go land it.

Jan 16 2019, 8:05 AM · efl
raster updated the summary of D7646: elm_config: Free data returned from eet_read.
Jan 16 2019, 8:02 AM · efl
raster closed D7638: emotion: gstreamer: fix deprecation warning for g_type_class_add_private.
Jan 16 2019, 3:26 AM · efl
raster accepted D7638: emotion: gstreamer: fix deprecation warning for g_type_class_add_private.

WFM. :)

Jan 16 2019, 3:25 AM · efl
raster triaged T7633: efl.ui.box test broken - homogeneous not implemented as High priority.
Jan 16 2019, 12:52 AM · efl
raster added a comment to T6462: git f*ck up? can't find revision (has 2 hashes) and file is unchanged on server.
git log src/lib/ecore/efl_io_closer_fd.c
Jan 16 2019, 12:48 AM
raster added a comment to T6462: git f*ck up? can't find revision (has 2 hashes) and file is unchanged on server.

to be more specific...

Jan 16 2019, 12:47 AM
raster added a comment to T6462: git f*ck up? can't find revision (has 2 hashes) and file is unchanged on server.

what @bu5hm4n said - i looked at the history of the file - the commit went in - changes there then at a later point it was changed again by other commits. so in GIT it seemed all fine to me.

Jan 16 2019, 12:33 AM
raster added a comment to T6462: git f*ck up? can't find revision (has 2 hashes) and file is unchanged on server.

it's just the cgit gui web - git has all the right state/content. that's what matters at the end of the day.

Jan 16 2019, 12:16 AM

Jan 15 2019

raster added a comment to D7351: elm_config: remove profile name reading logic from data dir.

this never should have even passed review. it's obviously wrong - the code path is used. a simple look at the for loop being removed would have shown that

Jan 15 2019, 12:47 PM · efl
raster added a comment to T7097: something broken with gl when lots of windows open with x11 engine.

this smells here like a video driver issue - like there are a limited number of "contexts" (video ports) for output rendering and once you hit the limit things fail. that's what it smells like at any rate... so either a hardware or driver design limit. the numbers above hint that the limit might be 64 which smells like the kind of number that might be such a limit well some ports are used implicitly e.g. the screen/compositor uses 1 then it gets 63 mroe of them (port 0, 1, 2, ... 62) then the 65'th fails.

Jan 15 2019, 4:19 AM · efl: display system, efl: rendering
raster added a comment to T6247: Please install engine headers again.

i think we aren't - the implication is code should use the higher level api wrappers. we had to expose these structs because we used to have efl broken up and had to expose an api from one lib to another... :)

Jan 15 2019, 3:59 AM · efl
raster added a comment to T7409: Terminology does not start on xwayland.

Just tried here and forced terminology to use X11 (unset WAYLAND_DISPLAY and ELM_DISPLAY terminology comes up fine in x11 mode as the only display info it has int he env is for x11). i checked with xwininfo -tree -root to see the x11 window as there in xwayland...

Jan 15 2019, 3:42 AM · regression, efl, Terminology

Jan 13 2019

raster added a comment to T7224: eina safety check failed, obj == NULL in evas_object_event_callback_del_full in lib/evas/canvas/evas_callbacks.c:539.

oh = i didnt even know about this ticket. i was fixing error logs i found from running enlightenment so i could have some more peace and quiet... :) i'll flag this.

Jan 13 2019, 4:10 PM · Restricted Project, efl: canvas

Jan 12 2019

raster closed T7153: DND broken now on non elm objects. as Wontfix.

well then elm needs some invisible event object then because how else can you just listen to dnd events on areas of a ui irrespective of the content that happens to be there? well ok- you can use buttons and set their color to 0 0 0 0 but that's a hack.

Jan 12 2019, 3:51 AM · Restricted Project, efl: widgets
raster added a comment to T6462: git f*ck up? can't find revision (has 2 hashes) and file is unchanged on server.

ummm errrr? i don't know. cgit might have caches like @cedric says and these somehow got stale or unwritable or... ?

Jan 12 2019, 3:44 AM

Jan 8 2019

raster added a comment to T6028: Port per window volume control on wayland.

This has to be done by the app PROCESS on wayland, but it doesn't need to be done by the APP. This could be done in EFL.

Jan 8 2019, 7:22 AM · enlightenment-git, Restricted Project

Jan 4 2019

raster added a comment to D7517: ecore: get rid of commands in efl_task..

I am wondering right now why this thing went into the place where noone finds it ?

Jan 4 2019, 3:58 AM · efl
raster added a comment to D7510: ecore: here comes a env object.

The biggest part of this code is copied from efl_app & efl_task, so i am not too sure where it is really more complexity

Jan 4 2019, 3:48 AM · efl

Dec 31 2018

raster added a comment to D7441: eo: add the ability to get the size of object of a certain class..

@kimcinoo - good question because getting size of the class doesn't get the size of every object as i mentioned above - you could store sub-structs, lists (and the list nodes and what they point to) strings etc. so every obj could be a different size. this just gets base absolute minimum size. i'm not so sure it's really a great and useful thing as any old malloc debugger can probably tell you this kind of info (with track trace to the allocation etc.). a proper "get the mem used now of THIS object to the best of its knowledge" is more useful IMHO. it would mean every class has to implement this (track its own allocations and add them up then ask parent class etc.) it'd be a lot of extra work though.

Dec 31 2018, 9:01 PM · efl

Dec 29 2018

raster added a comment to T7530: Fix eina promise data store.

so basically a convention of namespacing the string so people only mess with "their data" right? for example the efl loop promise might use "_efl/loop.promise" or something right? this does fix the need for eina_promise_data_set_cb_set()

Dec 29 2018, 4:48 AM · efl (efl-1.22), efl: data types, Restricted Project

Dec 28 2018

raster added a comment to D7441: eo: add the ability to get the size of object of a certain class..

I don't have much idea on how to write a useful test at all here.

Dec 28 2018, 2:27 AM · efl

Dec 27 2018

raster requested changes to D7510: ecore: here comes a env object.

eina hash just can't work well. please see my comments on D7514 - you end up with the same "it works differently in different places", a less efficient implementation as you have to scan the hash and the environment every time you set it (after modifications) to sync it etc. etc. but eina_hash is worse because now you don't have any ownership of the strings. do you have to stringshare_add them when adding to hash? or just cosnt char *'s? if the latter then we have all the nasty problems putenv() does (see manual pages - but specifically that the string ptr is used AS-IS int he env so you cannot release the string you pass in - it's not duplicated and so tracking this string's lifetime and ensuring it's duplicated and kept alive as long as in use etc. etc. also nukes the idea of an eina hash).

Dec 27 2018, 10:26 AM · efl
raster requested changes to D7517: ecore: get rid of commands in efl_task..

Note that the usage in efl_thread.c should and could be removed. the problem with its usage is that when the ARGUMENTS event is fired, noone ever had the chance to subscribe to the loop of the thread yet. So all in all this is unneccessary, since noone could ever touch that.

Dec 27 2018, 10:20 AM · efl
raster added a comment to T7520: "Digital Clock Gadget" date row is not shown with Flat theme.

aaah ok. i don't have one of those set up... i'll see. shelf seems fine though

Dec 27 2018, 10:11 AM · Restricted Project
raster added a comment to D7514: efl_task: remove env from this object.

The old API is not really definite, the API specifies that the API sometimes mirrors to environment variables, sometimes not. Considering the situation of this within a thread is quite a problem, since i don't know if i am altering the environment variables of the process or not, same argument for the efl_exe class.

Dec 27 2018, 5:00 AM · efl
raster added a comment to T7520: "Digital Clock Gadget" date row is not shown with Flat theme.

shelf or bryce/gadget bar?

Dec 27 2018, 4:32 AM · Restricted Project

Dec 26 2018

raster requested changes to D7514: efl_task: remove env from this object.
Dec 26 2018, 11:16 AM · efl
raster added a comment to D7514: efl_task: remove env from this object.

how is this better? it makes creation of a process with a modified env more work. multiple objects for no apparent reason (surely not given here in this patch). having env be part of a task is realistically what an environment is. it's part of a process (executable) and threads (they share the same env and it's implemented this way to modify the same shared env as that is the definition of an env within a process) and on an exe it's setting up an env in preparation for the actual execution (after which it can't be modified just like an environment actually works). why an object vs part of an existing class/interface? why is this better?

Dec 26 2018, 11:15 AM · efl

Dec 20 2018

raster added a comment to D7461: optimize glyph images data copy into 4 byte aligned images.

ok. that's something more concrete. 512 is a huge size... like insanely huge. but ok - some info as to the kind of sizes you are looking at. thanks!

Dec 20 2018, 8:57 AM · efl
raster added a comment to D7471: eina blist data structure proposal.

I updated my benchmarks for the intl Atom - the system wasn't completely idle. fixed that now. was idle. still tells the same story relatively speaking of blist vs list.

Dec 20 2018, 3:08 AM · efl
raster added a comment to D7471: eina blist data structure proposal.

@ProhtMeyhet you seem to have slightly older hardware than me, but that indeed shows the same trend as my benchmarks.The trade-off of cache friendliness vs some extra work in shuffling small arrays around seems to pay off. Well in all cases except @ManMower :) Thanks for the extra numbers.

Dec 20 2018, 2:52 AM · efl
raster added a comment to D7471: eina blist data structure proposal.

@ManMower - it seems your machine is an outlier so far, but i guess the principle still counts. I should modify the brenchmark to pollute cache at times, but then the benchmark needs to do its own timings as it has to ignore the pollution bits, so it's going to become a fair bit more complex than it is.

Dec 20 2018, 2:49 AM · efl

Dec 19 2018

raster added a comment to D7471: eina blist data structure proposal.

I did some benchmarking across many machines, generations and architectures. I scripted it as part of a build script. All binaries were compiled with -O3 -march=native (except on aarch64 where -march=native is not valid). The results are consistent: That BList always wins, and the wins get much bigger the longer the list and the less the list fits into cache. These are run times so the shorter the run time, the better. On the 32bit ARM systems and on the Baytrail Atom system they did 1/10th the number of loops so they complete in a reasonable time.

Dec 19 2018, 7:10 AM · efl
raster added a comment to D7471: eina blist data structure proposal.

Admittedly my numbers were on aarch64. But don't misunderstand list length being the real factor. It's cache hits that is. BList is intended to reduce the downside of cache misses very specifically. It's reliant on that. So lists that don't stay in cache very much are going to be far worse on Eina List vs BList. The length in this test is more a measure for "how much of the list will be in l1/l2/l3 cache". The bigger the number of items, the less will be in caches (or less in L1 vs L2, less in L2 vs L3, less in L3 vs RAM). I maybe could have putt in some code to pollute caches regularly between ops as well as well as a "how often to pollute caches and by how much" as part of the benchmark. I could adapt the test to do this explicitly. I was relying on a very simple factor of list size determining cache hits vs. misses.

Dec 19 2018, 4:09 AM · efl

Dec 17 2018

raster requested review of D7471: eina blist data structure proposal.
Dec 17 2018, 9:17 AM · efl
raster added a comment to D7461: optimize glyph images data copy into 4 byte aligned images.

yeah - but what are these benchmarks? where does this ACTUALLY happen enough?

Dec 17 2018, 4:39 AM · efl
raster added a comment to D7461: optimize glyph images data copy into 4 byte aligned images.

I'm curious... in what kind of benchmark/situation did this actually make a measurable difference? :) I never saw this function in any profiles (well not near the top of anything) so it never was looked into.

Dec 17 2018, 1:56 AM · efl