Skip to content

Instantly share code, notes, and snippets.

View stblr's full-sized avatar

Pablo Stebler stblr

  • Neuchâtel, Switzerland
  • 13:05 (UTC +02:00)
View GitHub Profile
[ 0.000000] Booting Linux on physical CPU 0x0
[ 0.000000] Linux version 4.14.83-g2457335fb8d4-dirty (pablo@e7470) (gcc version 4.9.x 20150123 (prerelease) (GCC)) #1 SMP PREEMPT Thu Dec 19 10:48:29 CET 2019
[ 0.000000] Boot CPU: AArch64 Processor [51af8014]
[ 0.000000] Machine: SoMC Lilac-ROW(MSM8998 v2.1)
[ 0.000000] Reserved memory: created CMA memory pool at 0x000000017e000000, size 4 MiB
[ 0.000000] OF: reserved mem: initialized node qseecom_ta_region, compatible id shared-dma-pool
[ 0.000000] Reserved memory: created CMA memory pool at 0x00000000fe800000, size 20 MiB
[ 0.000000] OF: reserved mem: initialized node qseecom_region, compatible id shared-dma-pool
[ 0.000000] Reserved memory: created CMA memory pool at 0x00000000fe000000, size 8 MiB
[ 0.000000] OF: reserved mem: initialized node adsp_region, compatible id shared-dma-pool
This file has been truncated, but you can view the full file.
--------- beginning of main
03-12 10:47:03.071 429 429 I SELinux : SELinux: Loaded service_contexts from:
03-12 10:47:03.075 429 429 I SELinux : /system/etc/selinux/plat_service_contexts
03-12 10:47:03.083 431 431 I SELinux : SELinux: Loaded service_contexts from:
03-12 10:47:03.083 431 431 I SELinux : /vendor/etc/selinux/vndservice_contexts
03-12 10:47:03.089 432 432 D QSEECOMD: : qseecom listener services process entry PPID = 1
03-12 10:47:03.090 432 432 E QSEECOMD: : Listener: index = 0, hierarchy = 0
03-12 10:47:03.094 432 432 D QSEECOMD: : Init dlopen(librpmb.so, RTLD_NOW) succeeds
03-12 10:47:03.094 432 432 D QSEECOMD: : Init::Init dlsym(g_FSHandle rpmb_init_service) succeeds
03-12 10:47:03.095 432 432 I rpmb_ufs: RPMB Mult (512-byte sector) = 32768, Rel_sec_cnt = 32

GSOC 2019 with VideoLAN - final report

Initial goals

This summer I worked on dav1d, the AV1 decoder in VLC, Firefox, Chromium and many others. My task was to try to add GPU support to it, either because it may be faster, it may use lower power consumption, or simply because the CPU can do other stuff during this time. I wanted to write a shader for one of the decoding stage, and invoke it using the different graphics APIs. Then I planned to make some testing and after that do the same for another decoding stage.

Some preparation

To be accepted in GSoC VideoLAN asked me to write a simple patch to dav1d. I worked to add a few benchmarking options to the dav1d CLI, notably a realtime mode that can be used to benchmark power consumption in realistic playback conditions.

--- TLP 1.1.907 --------------------------------------------
+++ Configured Settings: /etc/default/tlp
TLP_ENABLE=1
TLP_DEFAULT_MODE=AC
TLP_PERSISTENT_DEFAULT=0
DISK_IDLE_SECS_ON_AC=0
DISK_IDLE_SECS_ON_BAT=2
MAX_LOST_WORK_SECS_ON_AC=15
MAX_LOST_WORK_SECS_ON_BAT=60
--- TLP 1.1.907 --------------------------------------------
+++ Configured Settings: /etc/default/tlp
TLP_ENABLE=1
TLP_DEFAULT_MODE=AC
TLP_PERSISTENT_DEFAULT=0
DISK_IDLE_SECS_ON_AC=0
DISK_IDLE_SECS_ON_BAT=2
MAX_LOST_WORK_SECS_ON_AC=15
MAX_LOST_WORK_SECS_ON_BAT=60

Keybase proof

I hereby claim:

  • I am pablostebler on github.
  • I am pablostebler (https://keybase.io/pablostebler) on keybase.
  • I have a public key ASDt6H8v_mGgetHpVlDHw02HHJ7nAMXgs6RFhePUm94Nowo

To claim this, I am signing this object: