summary refs log tree commit diff stats
path: root/results/classifier/gemma3:12b/kernel/2146
diff options
context:
space:
mode:
authorChristian Krinitsin <mail@krinitsin.com>2025-07-03 07:27:52 +0000
committerChristian Krinitsin <mail@krinitsin.com>2025-07-03 07:27:52 +0000
commitd0c85e36e4de67af628d54e9ab577cc3fad7796a (patch)
treef8f784b0f04343b90516a338d6df81df3a85dfa2 /results/classifier/gemma3:12b/kernel/2146
parent7f4364274750eb8cb39a3e7493132fca1c01232e (diff)
downloademulator-bug-study-d0c85e36e4de67af628d54e9ab577cc3fad7796a.tar.gz
emulator-bug-study-d0c85e36e4de67af628d54e9ab577cc3fad7796a.zip
add deepseek and gemma results
Diffstat (limited to 'results/classifier/gemma3:12b/kernel/2146')
-rw-r--r--results/classifier/gemma3:12b/kernel/2146115
1 files changed, 115 insertions, 0 deletions
diff --git a/results/classifier/gemma3:12b/kernel/2146 b/results/classifier/gemma3:12b/kernel/2146
new file mode 100644
index 00000000..9b1d15b1
--- /dev/null
+++ b/results/classifier/gemma3:12b/kernel/2146
@@ -0,0 +1,115 @@
+
+qemu-system-aarch64 Segfaults
+Description of problem:
+Never finishes the script below always segfaults after a few hours
+in seemingly random functions.
+Steps to reproduce:
+This is what i did with qemu version 8.2.1
+inside test directory:
+1. wget https://download.qemu.org/qemu-8.2.1.tar.xz
+2. tar xvJf qemu-8.2.1.tar.xz
+3. cd qemu-8.2.1
+4. ./configure --target-list="aarch64-linux-user, aarch64-softmmu" --enable-slirp (crashes with and without --enable-debug)
+5. make -j$(nproc)
+6. ln -sf "$PWD/build/qemu-system-aarch64" "../qemu-system-aarch64"
+7. cd ..
+
+Now the VM
+1. wget -O installer-linux https://deb.debian.org/debian/dists/bookworm/main/installer-arm64/current/images/netboot/debian-installer/arm64/linux
+2. wget -O installer-initrd.gz https://deb.debian.org/debian/dists/bookworm/main/installer-arm64/current/images/netboot/debian-installer/arm64/initrd.gz
+3. qemu-img create -f qcow2 hda.qcow2 15G
+4. ./qemu-system-aarch64 -M virt -m 6G -cpu cortex-a72 \
+      -kernel installer-linux \
+      -initrd installer-initrd.gz \
+      -drive if=none,file=hda.qcow2,format=qcow2,id=hd \
+      -device virtio-blk-pci,drive=hd \
+      -netdev user,id=mynet \
+      -device virtio-net-pci,netdev=mynet \
+      -nographic -no-reboot \
+      -accel tcg,thread=multi \
+      -smp 8
+5. Install minimal debian inside the VM
+6. sudo virt-copy-out -a hda.qcow2 /boot/vmlinuz-6.1.0-17-arm64 /boot/initrd.img-6.1.0-17-arm64 .
+7. ./qemu-system-aarch64 -M virt -m 6G -cpu cortex-a72 \
+      -kernel vmlinuz-6.1.0-17-arm64 \
+      -initrd initrd.img-6.1.0-17-arm64 \
+      -append 'root=/dev/vda2' \
+      -drive if=none,file=hda.qcow2,format=qcow2,id=hd \
+      -device virtio-blk-pci,drive=hd \
+      -netdev user,id=mynet,hostfwd=tcp::10022-:22 \
+      -device virtio-net-pci,netdev=mynet \
+      -nographic \
+      -accel tcg,thread=multi \
+      -smp 8
+8. Now run this script inside some directory inside the VM(you might need to install gcc first)
+
+#!/bin/bash
+
+wget --no-clobber https://sourceware.org/pub/binutils/releases/binutils-2.41.tar.xz   
+wget --no-clobber https://ftp.gnu.org/gnu/mpfr/mpfr-4.2.0.tar.xz   
+wget --no-clobber https://ftp.gnu.org/gnu/gmp/gmp-6.3.0.tar.xz    
+wget --no-clobber https://ftp.gnu.org/gnu/mpc/mpc-1.3.1.tar.gz    
+wget --no-clobber https://ftp.gnu.org/gnu/gcc/gcc-13.2.0/gcc-13.2.0.tar.xz   
+
+BUG_TARGET="$(uname -m)-bug-linux-gnu"
+
+tar -xf binutils-2.41.tar.xz   
+cd binutils-2.41   
+mkdir -vp build   
+cd build   
+../configure --prefix=$PWD        \
+             --with-sysroot=$PWD  \
+             --target=$BUG_TARGET \
+             --disable-nls        \
+             --enable-gprofng=no  \
+             --disable-werror     \
+             --disable-gdb
+make --jobs $(nproc)   
+cd ../..   
+rm -rf binutils   
+
+tar -xf gcc-13.2.0.tar.xz   
+cd gcc-13.2.0   
+tar -xf ../mpfr-4.2.0.tar.xz   
+tar -xf ../gmp-6.3.0.tar.xz   
+tar -xf ../mpc-1.3.1.tar.gz   
+mv mpfr-4.2.0 mpfr   
+mv gmp-6.3.0 gmp   
+mv mpc-1.3.1 mpc   
+mkdir -vp build   
+cd build   
+../configure --prefix=$PWD             \
+             --with-sysroot=$PWD       \
+             --target=$BUG_TARGET      \
+             --with-glibc-version=2.38 \
+             --with-newlib             \
+             --without-headers         \
+             --enable-default-pie      \
+             --enable-default-ssp      \
+             --disable-nls             \
+             --disable-shared          \
+             --disable-multilib        \
+             --disable-threads         \
+             --disable-libatomic       \
+             --disable-libgomp         \
+             --disable-libquadmath     \
+             --disable-libssp          \
+             --disable-libvtv          \
+             --disable-libstdcxx       \
+             --enable-languages=c,c++
+make --jobs $(nproc)   
+cd ../..   
+rm -rf gcc
+Additional information:
+I tried all the versions listed above, 6.2 usually segfaults in binutils while the other two run further.
+
+Example:
+```
+Program terminated with signal SIGSEGV, Segmentation fault.
+#0  0x000055555615dd37 in tlb_index (cpu=<Cannot access memory at address 0x7fffefffe1c8>,
+    mmu_idx=<Cannot access memory at address 0x7fffefffe1c0>,
+    addr=<Cannot access memory at address 0x7fffefffe1b8>)
+    at qemu-8.2.1/include/exec/cpu_ldst.h:367
+367	    uintptr_t size_mask = cpu->neg.tlb.f[mmu_idx].mask >> CPU_TLB_ENTRY_BITS;
+[Current thread is 1 (LWP 857562)]
+```