1 files changed, 74 insertions, 0 deletions
diff --git a/results/classifier/zero-shot/118/assembly-x86/1377 b/results/classifier/zero-shot/118/assembly-x86/1377
new file mode 100644
index 00000000..5f1dd9fc
--- /dev/null
+++ b/results/classifier/zero-shot/118/assembly-x86/1377
@@ -0,0 +1,74 @@
+x86: 0.997
+i386: 0.900
+assembly: 0.847
+architecture: 0.838
+device: 0.773
+vnc: 0.768
+graphic: 0.742
+kernel: 0.698
+ppc: 0.615
+risc-v: 0.554
+debug: 0.506
+boot: 0.462
+socket: 0.412
+permissions: 0.405
+arm: 0.386
+TCG: 0.372
+PID: 0.328
+KVM: 0.326
+semantic: 0.321
+performance: 0.309
+network: 0.279
+VMM: 0.264
+hypervisor: 0.205
+virtual: 0.186
+peripherals: 0.177
+files: 0.113
+register: 0.113
+mistranslation: 0.073
+user-level: 0.021
+--------------------
+x86: 1.000
+i386: 0.977
+assembly: 0.946
+debug: 0.242
+kernel: 0.071
+TCG: 0.049
+semantic: 0.042
+risc-v: 0.034
+virtual: 0.033
+user-level: 0.033
+register: 0.029
+hypervisor: 0.027
+files: 0.025
+performance: 0.024
+device: 0.012
+PID: 0.012
+network: 0.012
+architecture: 0.010
+peripherals: 0.009
+boot: 0.005
+permissions: 0.003
+socket: 0.002
+VMM: 0.001
+graphic: 0.001
+vnc: 0.001
+KVM: 0.001
+mistranslation: 0.001
+ppc: 0.000
+arm: 0.000
+
+x86 CVT* series instructions fault
+Description of problem:
+For example, CVTSD2SS instruction converts SRC[63:0] double precision floating point to DEST[31:0] single precision floating point. Although the CVTSD2SS instruction uses only 8 bytes, if it overlaps page boundary, I think QEMU tries to access over the valid memory and crashes.
+Steps to reproduce:
+1. Compile this code
+```
+void main() {
+    mmap(0x555555559000, 0x1000, flag, ~~, 0);
+    asm("cvtsd2ss xmm1, qword ptr [0x555555559ff8]");
+}
+```
+2. Execute. QEMU crashes but CPU does not.
+Additional information:
+This bug is discovered by research conducted by KAIST SoftSec.