[PATCH v5 0/3] Optimize code generation during context switching

Xie Yuanbin qq570070308 at gmail.com
Mon Dec 15 05:10:01 PST 2025


There is some additional testing information that needs to be
supplemented:
1.
On Mon, 15 Dec 2025 03:09:04 +0800, Xie Yuanbin Wrote:
> 2. Testing config:
> x86-64: `make x86_64_defconfig` first, then menuconfig setting:
> CONFIG_HZ=100
> CONFIG_DEBUG_ENTRY=n
> CONFIG_X86_DEBUG_FPU=n
> CONFIG_EXPERT=y
> CONFIG_MODIFY_LDT_SYSCALL=n
> CONFIG_STACKPROTECTOR=n
> CONFIG_BLK_DEV_NVME=y (just for boot)
>
> arm64: `make defconfig` first, then menuconfig setting:
> CONFIG_KVM=n
> CONFIG_HZ=100
> CONFIG_SHADOW_CALL_STACK=y
>
> arm32: `make multi_v7_defconfig` first, then menuconfig setting:
> CONFIG_ARCH_OMAP2PLUS_TYPICAL=n
> CONFIG_HIGHMEM=n

The above is the config for performance test, when testing for size, the
config is: `make x86_64_defconfig` first, then menuconfig setting:
CONFIG_SCHED_CORE=y
CONFIG_NO_HZ_FULL=y
CONFIG_CC_OPTIMIZE_FOR_SIZE=y (optional)

Enable CONFIG_SCHED_CORE and CONFIG_NO_HZ_FULL to test the modified code.

2. When testing with Raspberry Pi 3b, in order to make the test result
stable, the CPU frequency should be fixed. In detail, setting in
config.txt:
```config.txt
arm_boost=0
core_freq_fixed=1
arm_freq=1200
gpu_freq=250
sdram_freq=400
arm_freq_min=1200
gpu_freq_min=250
sdram_freq_min=400
```

Thanks!



More information about the linux-riscv mailing list