ci: upgrade to Ubuntu 26.04

[mkroening]

No description provided.

[github-actions]

Benchmark Results

Details

Benchmark	Current: df39a32	Previous: 80ddb41	Performance Ratio
startup_benchmark Build Time	82.69 s	84.08 s	0.98 ❗
startup_benchmark File Size	0.77 MB	0.77 MB	1.00 ❗
Startup Time - 1 core	0.80 s (±0.03 s)	0.78 s (±0.02 s)	1.03
Startup Time - 2 cores	0.81 s (±0.03 s)	0.78 s (±0.02 s)	1.04
Startup Time - 4 cores	0.78 s (±0.02 s)	0.79 s (±0.02 s)	1.00
multithreaded_benchmark Build Time	80.32 s	84.74 s	0.95 ❗
multithreaded_benchmark File Size	0.86 MB	0.86 MB	1.00 ❗
Multithreaded Pi Efficiency - 2 Threads	89.83 % (±7.94 %)	89.91 % (±7.04 %)	1.00
Multithreaded Pi Efficiency - 4 Threads	44.98 % (±2.76 %)	44.59 % (±3.96 %)	1.01
Multithreaded Pi Efficiency - 8 Threads	26.01 % (±1.42 %)	25.81 % (±1.77 %)	1.01
micro_benchmarks Build Time	80.59 s	84.85 s	0.95 ❗
micro_benchmarks File Size	0.87 MB	0.87 MB	1.00 ❗
Scheduling time - 1 thread	70.58 ticks (±4.27 ticks)	69.95 ticks (±3.60 ticks)	1.01
Scheduling time - 2 threads	38.02 ticks (±3.69 ticks)	38.89 ticks (±4.81 ticks)	0.98
Micro - Time for syscall (getpid)	4.38 ticks (±0.65 ticks)	4.52 ticks (±0.65 ticks)	0.97
Memcpy speed - (built_in) block size 4096	79291.97 MByte/s (±54830.88 MByte/s)	81501.17 MByte/s (±56311.67 MByte/s)	0.97
Memcpy speed - (built_in) block size 1048576	29971.42 MByte/s (±24381.54 MByte/s)	30207.13 MByte/s (±24513.35 MByte/s)	0.99
Memcpy speed - (built_in) block size 16777216	23643.86 MByte/s (±19429.90 MByte/s)	24558.71 MByte/s (±20265.22 MByte/s)	0.96
Memset speed - (built_in) block size 4096	79864.88 MByte/s (±55211.26 MByte/s)	82291.10 MByte/s (±56782.79 MByte/s)	0.97
Memset speed - (built_in) block size 1048576	30723.08 MByte/s (±24807.38 MByte/s)	30965.54 MByte/s (±24941.09 MByte/s)	0.99
Memset speed - (built_in) block size 16777216	23911.99 MByte/s (±19533.05 MByte/s)	25323.26 MByte/s (±20779.97 MByte/s)	0.94
Memcpy speed - (rust) block size 4096	71857.46 MByte/s (±50263.56 MByte/s)	70018.99 MByte/s (±49036.86 MByte/s)	1.03
Memcpy speed - (rust) block size 1048576	29891.88 MByte/s (±24414.50 MByte/s)	29906.33 MByte/s (±24373.88 MByte/s)	1.00
Memcpy speed - (rust) block size 16777216	24426.28 MByte/s (±20157.59 MByte/s)	25589.75 MByte/s (±21106.80 MByte/s)	0.95
Memset speed - (rust) block size 4096	72224.29 MByte/s (±50496.60 MByte/s)	70202.59 MByte/s (±49165.39 MByte/s)	1.03
Memset speed - (rust) block size 1048576	30644.27 MByte/s (±24835.37 MByte/s)	30681.42 MByte/s (±24831.49 MByte/s)	1.00
Memset speed - (rust) block size 16777216	24614.80 MByte/s (±20216.89 MByte/s)	26307.30 MByte/s (±21557.26 MByte/s)	0.94
alloc_benchmarks Build Time	78.99 s	77.85 s	1.01 ❗
alloc_benchmarks File Size	0.86 MB	0.86 MB	1.00 ❗
Allocations - Allocation success	91.32 %	91.32 %	1
Allocations - Deallocation success	100.00 %	100.00 %	1
Allocations - Pre-fail Allocations	61.46 %	61.46 %	1
Allocations - Average Allocation time	2878.36 Ticks (±261.76 Ticks)	2289.27 Ticks (±449.69 Ticks)	1.26 ❗
Allocations - Average Allocation time (no fail)	3862.25 Ticks (±223.78 Ticks)	3349.32 Ticks (±377.82 Ticks)	1.15 ❗
Allocations - Average Deallocation time	793.72 Ticks (±83.39 Ticks)	637.74 Ticks (±60.32 Ticks)	1.24 ❗
mutex_benchmark Build Time	79.90 s	78.82 s	1.01 ❗
mutex_benchmark File Size	0.87 MB	0.87 MB	1.00 ❗
Mutex Stress Test Average Time per Iteration - 1 Threads	13.02 ns (±0.71 ns)	12.66 ns (±0.79 ns)	1.03
Mutex Stress Test Average Time per Iteration - 2 Threads	45.62 ns (±3.82 ns)	42.98 ns (±3.81 ns)	1.06

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

[cagatay-y]

We should be able to uncomment these lines once this PR is merged

kernel/.github/workflows/ci.yml

Lines 295 to 301 in 5d52e43

	# FIXME: this is broken on QEMU 8.2.2
	# Enable this once CI reaches QEMU 8.2.3 or QEMU 9.0.0
	# - run: cargo xtask ci rs --arch ${{ matrix.arch }} --profile ${{ matrix.profile }} ${{ matrix.rs_flags }} --package httpd --features ci,hermit/dhcpv4,hermit/virtio-net qemu ${{ matrix.qemu_flags }} --devices virtio-net-pci
	# if: matrix.arch != 'riscv64'
	# env:
	# # The buffer is sized to be smaller than the packets received during DHCPv4 address acquisition and thus exercise the buffer merging code path.
	# HERMIT_MRG_RXBUF_SIZE: 200

[mkroening]

Yeah, that's one of the reasons I wanted to try this out. ^^

I am unsure whether we should land this now, though, since it is still a public preview only. 🤔