https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #30 from Barto (mister.free...@laposte.net) ---
(In reply to Christian König from comment #26)
> As a band aid I will try to fix our algorithm when pages are freed again,
> but in general the driver stack or application should be fixed
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #29 from Christian König (christian.koe...@amd.com) ---
For this you need to call the driver IOCTL to create a buffer object directly.
Best is probably you use the Mesa code as and work from that backward, see here
https://cgit.freede
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #28 from Barto (mister.free...@laposte.net) ---
I made another test, by using a player like VLC and a video file ( 1080p, 60
FPS ) :
- with a video file 1080p@60 fps, GPU acceleration disabled ( XV output ) and
kernel 4.15rc8 : I can
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #23 from Barto (mister.free...@laposte.net) ---
so here is the output of "cat /sys/kernel/debug/dri/0/ttm_dma_page_pool" when
no youtube video are played ( firefox is not running ), with kernel 4.15rc8 :
pool refills
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #21 from Christian König (christian.koe...@amd.com) ---
(In reply to Barto from comment #18)
> pool refills pages freedinuse available name
>wc 801 0 2834 370 radeon
> 0
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #19 from Barto (mister.free...@laposte.net) ---
(In reply to Christian König from comment #16)
> Huge pages are always disabled on your system. It would be interesting to
> see what happens when you try to enable them, not disable the
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #25 from Barto (mister.free...@laposte.net) ---
Someone has made a suggestion to me : go to "about:support" in firefox for
ckecking some options, I can read this :
HW_COMPOSITING blocked by default: Acceleration blocked by platform
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #17 from Christian König (christian.koe...@amd.com) ---
What is the content of /sys/kernel/debug/dri/0/ttm_dma_page_pool while you are
playing a youtube video?
And can you try to compile the kernel with CONFIG_SWIOTLB disabled?
--
Y
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #24 from Barto (mister.free...@laposte.net) ---
I forgot the third case, here is the output after a youtube video has been
played on firefox 57, I closed firefox and now here is the output :
# cat /sys/kernel/debug/dri/0/ttm_dma_page_
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #18 from Barto (mister.free...@laposte.net) ---
here is the output of "cat /sys/kernel/debug/dri/0/ttm_dma_page_pool" with
kernel 4.15rc8 when I play a youtube video hd@60 fps in full screen mode :
pool refills pages
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #20 from Christian König (christian.koe...@amd.com) ---
(In reply to Barto from comment #19)
> AnonHugePages:106496 kB
> ShmemHugePages:0 kB
...
> but what means the line "AnonHugePages" ?
Ah crap you are right, AnonHugePa
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #16 from Christian König (christian.koe...@amd.com) ---
(In reply to Barto from comment #15)
> Hello Christian,
>
> I tried to disable "transparent huge page" with the kernel parameter
> "transparent_hugepage=never", but the bug is st
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #26 from Christian König (christian.koe...@amd.com) ---
(In reply to Barto from comment #25)
> but this advice seems to be a "cheat mode/workaround", it doesn't explain
> why your commit triggers this problem with firefox 57 when
> "la
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #22 from Barto (mister.free...@laposte.net) ---
(In reply to Christian König from comment #21)
> Best approach would be to gather the "while you play the video" numbers from
> a separate system over ssh, because when you put the firef
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #27 from Barto (mister.free...@laposte.net) ---
(In reply to Christian König from comment #26)
> Now what I see when I look at your numbers is that user space allocated and
> freed (13608−4514)*2M = 18.1GB of memory while playing yout
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #15 from Barto (mister.free...@laposte.net) ---
Hello Christian,
I tried to disable "transparent huge page" with the kernel parameter
"transparent_hugepage=never", but the bug is still here
here is the output of "cat /proc/meminfo" w
https://bugzilla.kernel.org/show_bug.cgi?id=198511
Christian König (christian.koe...@amd.com) changed:
What|Removed |Added
CC||christian.koe
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #13 from Barto (mister.free...@laposte.net) ---
the specs of my PC :
cpu : intel core 2 quad Q9650 ( 3 Ghz )
ram : 8 Gb DDR2
gpu : amd radeon HD4650 pcie
motherboard : gigabyte GA-P35-DS3L ( socket 775, chipset intel P35 )
window mana
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #12 from Barto (mister.free...@laposte.net) ---
Created attachment 273757
--> https://bugzilla.kernel.org/attachment.cgi?id=273757&action=edit
workaround by reverting changes made by the bad commit
I created a patch, it's a workarou
https://bugzilla.kernel.org/show_bug.cgi?id=198511
Barto (mister.free...@laposte.net) changed:
What|Removed |Added
Attachment #273743|0 |1
is obsolete
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #10 from Barto (mister.free...@laposte.net) ---
I think I found the real "first bad commit",
it's :
commit 648bc3574716400acc06f99915815f80d9563783
Author: Christian König
Date: Thu Jul 6 09:59:43 2017 +0200
drm/ttm: add tran
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #9 from Barto (mister.free...@laposte.net) ---
(In reply to Barto from comment #8)
> yes you are right, something went wrong with the bisect,
>
> because I revet commit cf2623d9 then the bug is still here
I meant "if I revert commit
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #8 from Barto (mister.free...@laposte.net) ---
Yes you (In reply to Alex Deucher from comment #7)
> I think something went wrong with your bisect. That commit touched a
> completely different driver that's not even loaded on your syst
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #7 from Alex Deucher (alexdeuc...@gmail.com) ---
I think something went wrong with your bisect. That commit touched a
completely different driver that's not even loaded on your system.
--
You are receiving this mail because:
You are
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #6 from Barto (mister.free...@laposte.net) ---
Created attachment 273743
--> https://bugzilla.kernel.org/attachment.cgi?id=273743&action=edit
bisect log
here is the bisect log
--
You are receiving this mail because:
You are watchi
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #5 from Barto (mister.free...@laposte.net) ---
I did the bisect,
the first bad commit is :
commit cf2623d951c1c52923a776e01cf2e2afc9d042a0
Author: Rex Zhu
Date: Mon Sep 4 18:11:52 2017 +0800
drm/amd/powerplay: refine powerpla
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #4 from Barto (mister.free...@laposte.net) ---
I will try(In reply to Alex Deucher from comment #3)
> Can you bisect?
I will try to bisect
--
You are receiving this mail because:
You are watching the assignee of the bug.
___
https://bugzilla.kernel.org/show_bug.cgi?id=198511
Alex Deucher (alexdeuc...@gmail.com) changed:
What|Removed |Added
CC||alexdeuc...@gmail.c
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #1 from Barto (mister.free...@laposte.net) ---
Created attachment 273695
--> https://bugzilla.kernel.org/attachment.cgi?id=273695&action=edit
glxinfo
glxinfo
--
You are receiving this mail because:
You are watching the assignee of
https://bugzilla.kernel.org/show_bug.cgi?id=198511
--- Comment #2 from Barto (mister.free...@laposte.net) ---
Created attachment 273697
--> https://bugzilla.kernel.org/attachment.cgi?id=273697&action=edit
dmesg kernel 4.14.13 good
dmesg kernel 4.14.13
with this version there is no problem of l
30 matches
Mail list logo