replica
/
hanchenye-llvm-project
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Attempt to stablize compiler-rt/test/asan/TestCases/strncpy-overflow.cpp 

This attempts to adjust the test to still exercise the expected codepath after D115904.  This test is fundementally rather fragile.

Unfortunately, I have not been able to confirm this workaround either does, or does not, work.  Attempting check-all with compiler-rt blows through an additional 30GB of disk space so my build config which exceeds my local disk space.
This commit is contained in:
Philip Reames 2021-12-20 17:51:11 -08:00
parent a268a63b1a
commit 9b955f77a1
1 changed files with 15 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
compiler-rt/test/asan/TestCases/strncpy-overflow.cpp
View File

@ -8,9 +8,23 @@
#include <string.h>
#include <stdlib.h>
// We need a way to prevent the optimize from eliminating the
// strncpy below (which otherwises writes to dead storage). We
// need the read to be out-of-line to prevent memory forwarding
// from making the memory dead again.
int sink_memory(int N, char *p) __attribute__((noinline));
int sink_memory(int N, char *p) {
int sum = 0;
for (int i = 0; i < N; i++)
sum += p[i];
return sum;
}
int main(int argc, char **argv) {
char *hello = (char*)malloc(6);
strcpy(hello, "hello");
int rval = sink_memory(6, hello);
char *short_buffer = (char*)malloc(9);
strncpy(short_buffer, hello, 10); // BOOM
// CHECK: {{WRITE of size 10 at 0x.* thread T0}}
@ -20,5 +34,5 @@ int main(int argc, char **argv) {
// CHECK: {{allocated by thread T0 here:}}
// CHECK: {{ #0 0x.* in .*malloc}}
// CHECK: {{ #1 0x.* in main .*strncpy-overflow.cpp:}}[[@LINE-8]]
return short_buffer[8];
return rval + sink_memory(9, short_buffer);
}