[browser] System.Memory.Tests is taking too long on FF

[pavelsavara]

investigate if we are swapping to disk and why

Easy start is to print memory usage during the test run.

• what the javaScript thinks
• what the wasm linear memory size is
• what the mono GC thinks about itself
• what the OS thinks about browser process

[dotnet-policy-service]

Tagging subscribers to 'arch-wasm': @lewing
See info in area-owners.md if you want to be subscribed.

[pavelsavara]

This took an hour

Log

[21:18:49] info: Starting:    System.Memory.Tests.dll
[22:22:17] info: Finished:    System.Memory.Tests.dll

[dotnet-policy-service]

Tagging subscribers to this area: @BrzVlad
See info in area-owners.md if you want to be subscribed.

[pavelsavara]

It still takes an hour Log

The OS stats before

               total        used        free      shared  buff/cache   available
Mem:            7929        2315         222           4        5392        5295
Swap:              0           0           0
===========================================================================================================
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
 1  0      0 227408 294284 5227436    0    0   468 10660 1034 5051 35  8 36 21  0
===========================================================================================================
TERM environment variable not set.
===========================================================================================================
USER         PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME COMMAND
helixbot       1  0.2  0.0   2892  1664 ?        Ss   12:30   0:00 /bin/sh /root/commands/helix_docker_work.sh
helixbot      13  0.0  0.0   2892  1664 ?        S    12:30   0:00 /bin/sh /root/helix/work/correlation/scripts/448797ac550b4576aa04cf2e36970e74/execute.sh
helixbot      65  0.0  0.0   7504  3584 ?        S    12:30   0:00 bash ./RunTests.sh
helixbot      71  0.0  0.0  10380  3584 ?        R    12:30   0:00 ps -aux
===========================================================================================================

And Os stats after

               total        used        free      shared  buff/cache   available
Mem:            7929        2310        2110           4        3508        5299
Swap:              0           0           0
===========================================================================================================
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa st
 2  0      0 2160752 301668 3291008    0    0   182  4179  626 2238 74  4 14  8  0
===========================================================================================================
TERM environment variable not set.
===========================================================================================================
USER         PID %CPU %MEM    VSZ   RSS TTY      STAT START   TIME COMMAND
helixbot       1  0.0  0.0   2892  1664 ?        Ss   12:30   0:00 /bin/sh /root/commands/helix_docker_work.sh
helixbot      13  0.0  0.0   2892  1664 ?        S    12:30   0:00 /bin/sh /root/helix/work/correlation/scripts/448797ac550b4576aa04cf2e36970e74/execute.sh
helixbot      65  0.0  0.0   7504  3584 ?        S    12:30   0:00 bash ./RunTests.sh
helixbot     501  0.0  0.0  10380  3456 ?        R    13:35   0:00 ps -aux
===========================================================================================================

I see no problem on OS/docker level from this

[pavelsavara]

C# GC.GetTotalMemory(false)

before :   4 199 696
after  : 244 450 240

JS Module.HEAPU8.length

before :  58 064 896
after  : 890 372 096

[pavelsavara]

It seems that debugLevel: -1 is preventing interp and jiterp optimizations.