1 files changed, 50 insertions, 0 deletions
diff --git a/results/classifier/accel-gemma3:12b/kvm/862 b/results/classifier/accel-gemma3:12b/kvm/862
new file mode 100644
index 000000000..5b9ca2367
--- /dev/null
+++ b/results/classifier/accel-gemma3:12b/kvm/862
@@ -0,0 +1,50 @@
+
+Using qemu+kvm is slower than using qemu in rv6(xv6 rust porting)
+Description of problem:
+Using qemu+kvm is slower than using qemu in rv6(xv6 rust porting)
+Steps to reproduce:
+```
+git clone https://github.com/kaist-cp/rv6
+cd rv6
+make clean
+TARGET=arm GIC_VERSION=3 KVM=yes make qemu
+```
+Additional information:
+We are currently working on the [rv6 project](https://github.com/kaist-cp/rv6) which is porting MIT's educational operating system [xv6](https://github.com/mit-pdos/xv6-public) to Rust.<br> Our code is located [here](https://github.com/kaist-cp/rv6/tree/main/kernel-rs).
+We use qemu and [qemu's virt platform](https://qemu.readthedocs.io/en/latest/system/arm/virt.html) to execute rv6, and it works well with using qemu.
+Executing command on arm machine is this:
+```
+RUST_MODE=release TARGET=arm KVM=yes GIC_VERSION=3
+qemu-system-aarch64 -machine virt -kernel kernel/kernel -m 128M -smp 80 -nographic -drive file=fs.img,if=none,format=raw,id=x0,copy-on-read=off -device virtio-blk-device,drive=x0,bus=virtio-mmio-bus.0 -cpu cortex-a53  -machine gic-version=3  -net none
+```
+To make some speed boost experiment with KVM, we made rv6 support the arm  architecture on arm machine. The arm architecture's driver code locates in [here](https://github.com/kaist-cp/rv6/tree/main/kernel-rs/src/arch/arm).
+The problem is, when we use qemu with kvm, the performance is significantly reduced.
+Executing command on arm machine with KVM is this:
+```
+qemu-system-aarch64 -machine virt -kernel kernel/kernel -m 128M -smp 80 -nographic -drive file=fs.img,if=none,format=raw,id=x0,copy-on-read=off -device virtio-blk-device,drive=x0,bus=virtio-mmio-bus.0 -cpu host -enable-kvm  -machine gic-version=3  -net none
+``` 
+We repeated 
+1. Write 500 bytes syscall 10,000 times and the result was: kvm disable: 4,500,000 us, kvm enable: 29,000,000 us. (> 5 times)
+2. Open/Close syscall 10,000 times result: kvm disable: 12,000,000 us, kvm enable: 29,000,000 us. (> 5 times)
+3. Getppid syscall 10,000 times result: kvm disable: 735,000 us, kvm enable: 825,000 us. (almost same)
+4. Simple calculation(a = a * 1664525 + 1013904223) 100 million times result: kvm disable: 2,800,000 us, kvm enable: 65,000,000 us. (> 20 times)
+
+And the elapsed time was estimated by [uptime_as_micro](https://github.com/kaist-cp/rv6/blob/90b84b60931327ae8635875b788b10280e47b99c/kernel-rs/src/arch/arm/timer.rs#L17) syscall in rv6.
+These results were so hard to understand. <br>So first we tried to find the bottleneck on rv6's booting process, because finding bottleneck during processing user program was so difficult.
+We found that the first noticeable bottleneck on rv6 booting process was [here](https://github.com/kaist-cp/rv6/blob/main/kernel-rs/src/kalloc.rs#L107-L108):
+```
+run.as_mut().init();
+self.runs().push_front(run.as_ref());
+```
+As far as we know, this part is just kind of "list initialization and push element" part. So we thought that by some reason, the KVM is not actually working and it makes worse result. And also this part is even before turn on some interrupts, so we thought [arm's GIC](https://developer.arm.com/documentation/dai0492/b/) or interrupt related thing is not related with problem.
+
+So, how can I get better performance when using kvm with qemu?
+
+To solve this problem, we tried these already:
+1. change qemu(4.2, 6.2), virt version, change [some command for qemu-kvm](https://linux.die.net/man/1/qemu-kvm) like cpu, drive cache, copy-on-read something, kernel_irqchip.., cpu core.. etc
+2. find some kvm hypercall to use - but not exists on arm64
+3. Run [lmbench](http://lmbench.sourceforge.net/) by ubuntu on qemu with kvm to check KVM itself is okay. - We found KVM with ubuntu is super faster than only using qemu.
+4. Check [16550a UART print code](https://github.com/kaist-cp/rv6/blob/main/kernel-rs/src/arch/arm/uart.rs) is really slow on enabling KVM which makes incorrect result on benchmark - Without bottleneck code, we found the progress time of rv6 booting were almost same with KVM enabled or not.
+5. Check other people who suffer same situation like us - but [this superuser page](https://superuser.com/questions/1317948/qemu-enable-kvm-slower-than-pure-emulation-for-x86-64) not works. Our clocksource is arch_sys_counter.
+
+Thank you for your help.