Initialize TLS for idle task

[jounathaen]

Fixes #2225

I'm not 100% sure if this is the best way to do it , and this is only yet implemented for x86_64, as I haven't yet looked into how TLS is implemented on arm & RISC-V

Questions:

• Shall we only do this in combination with the instrument feature?
• Shall we initialize a smaller TLS? I think we only need 8 bytes for rftrace.

[jounathaen]

Ok, for arm the TLS apparently needs to be stored in tpidr_el0; for RISC-V in the TP register.

[github-actions]

Benchmark Results

Details

Benchmark	Current: ec9652b	Previous: af77eb8	Performance Ratio
startup_benchmark Build Time	100.83 s	96.76 s	1.04 ❗
startup_benchmark File Size	0.86 MB	0.82 MB	1.06 ❗
Startup Time - 1 core	0.94 s (±0.03 s)	0.93 s (±0.03 s)	1.01
Startup Time - 2 cores	0.96 s (±0.03 s)	0.93 s (±0.03 s)	1.03
Startup Time - 4 cores	0.97 s (±0.03 s)	0.95 s (±0.04 s)	1.02
multithreaded_benchmark Build Time	97.43 s	94.96 s	1.03 ❗
multithreaded_benchmark File Size	0.96 MB	0.96 MB	1.00 ❗
Multithreaded Pi Efficiency - 2 Threads	91.53 % (±8.28 %)	88.34 % (±8.83 %)	1.04
Multithreaded Pi Efficiency - 4 Threads	44.59 % (±3.54 %)	43.96 % (±2.87 %)	1.01
Multithreaded Pi Efficiency - 8 Threads	26.20 % (±1.76 %)	25.67 % (±1.88 %)	1.02
micro_benchmarks Build Time	107.38 s	107.03 s	1.00 ❗
micro_benchmarks File Size	0.97 MB	0.97 MB	1.00 ❗
Scheduling time - 1 thread	68.01 ticks (±3.46 ticks)	68.40 ticks (±2.80 ticks)	0.99
Scheduling time - 2 threads	39.51 ticks (±5.17 ticks)	36.75 ticks (±3.77 ticks)	1.08
Micro - Time for syscall (getpid)	3.01 ticks (±0.25 ticks)	3.72 ticks (±0.30 ticks)	0.81 ❗
Memcpy speed - (built_in) block size 4096	67624.66 MByte/s (±48174.20 MByte/s)	65936.06 MByte/s (±46962.77 MByte/s)	1.03
Memcpy speed - (built_in) block size 1048576	29642.26 MByte/s (±24468.98 MByte/s)	29711.39 MByte/s (±24468.80 MByte/s)	1.00
Memcpy speed - (built_in) block size 16777216	28463.72 MByte/s (±23700.48 MByte/s)	28572.31 MByte/s (±23801.85 MByte/s)	1.00
Memset speed - (built_in) block size 4096	68500.32 MByte/s (±48761.11 MByte/s)	66224.66 MByte/s (±47148.82 MByte/s)	1.03
Memset speed - (built_in) block size 1048576	30405.49 MByte/s (±24905.59 MByte/s)	30455.87 MByte/s (±24906.53 MByte/s)	1.00
Memset speed - (built_in) block size 16777216	29099.21 MByte/s (±24018.80 MByte/s)	29305.64 MByte/s (±24204.32 MByte/s)	0.99
Memcpy speed - (rust) block size 4096	58999.27 MByte/s (±43346.09 MByte/s)	59811.41 MByte/s (±43763.94 MByte/s)	0.99
Memcpy speed - (rust) block size 1048576	29540.25 MByte/s (±24439.40 MByte/s)	29629.02 MByte/s (±24504.10 MByte/s)	1.00
Memcpy speed - (rust) block size 16777216	28315.82 MByte/s (±23599.19 MByte/s)	28401.51 MByte/s (±23669.15 MByte/s)	1.00
Memset speed - (rust) block size 4096	59830.04 MByte/s (±43865.04 MByte/s)	60851.37 MByte/s (±44540.62 MByte/s)	0.98
Memset speed - (rust) block size 1048576	30292.03 MByte/s (±24868.53 MByte/s)	30426.48 MByte/s (±24959.19 MByte/s)	1.00
Memset speed - (rust) block size 16777216	29089.30 MByte/s (±24040.80 MByte/s)	29175.93 MByte/s (±24111.52 MByte/s)	1.00
alloc_benchmarks Build Time	104.28 s	102.72 s	1.02 ❗
alloc_benchmarks File Size	0.94 MB	0.89 MB	1.05 ❗
Allocations - Allocation success	100.00 %	100.00 %	1
Allocations - Deallocation success	100.00 %	100.00 %	1
Allocations - Pre-fail Allocations	100.00 %	100.00 %	1
Allocations - Average Allocation time	9458.16 Ticks (±847.78 Ticks)	8547.70 Ticks (±113.80 Ticks)	1.11 ❗
Allocations - Average Allocation time (no fail)	9458.16 Ticks (±847.78 Ticks)	8547.70 Ticks (±113.80 Ticks)	1.11 ❗
Allocations - Average Deallocation time	1765.18 Ticks (±1155.59 Ticks)	1117.79 Ticks (±513.75 Ticks)	1.58
mutex_benchmark Build Time	110.73 s	111.82 s	0.99 ❗
mutex_benchmark File Size	0.97 MB	0.97 MB	1.00 ❗
Mutex Stress Test Average Time per Iteration - 1 Threads	12.80 ns (±0.72 ns)	13.38 ns (±0.85 ns)	0.96
Mutex Stress Test Average Time per Iteration - 2 Threads	15.68 ns (±0.90 ns)	15.68 ns (±0.79 ns)	1

This comment was automatically generated by workflow using github-action-benchmark.