[compiler-rt] r279768 - [lsan] one more test fix to please the Debian Sid bot (this time, confirmed on the proper machine). Apparently, newer glibc uses slightly more stack

[Kostya Serebryany via llvm-commits]

Author: kcc
Date: Thu Aug 25 14:08:10 2016
New Revision: 279768

URL: http://llvm.org/viewvc/llvm-project?rev=279768&view=rev
Log:
[lsan] one more test fix to please the Debian Sid bot (this time, confirmed on the proper machine). Apparently, newer glibc uses slightly more stack

Modified:
    compiler-rt/trunk/test/lsan/TestCases/guard-page.c

Modified: compiler-rt/trunk/test/lsan/TestCases/guard-page.c
URL: http://llvm.org/viewvc/llvm-project/compiler-rt/trunk/test/lsan/TestCases/guard-page.c?rev=279768&r1=279767&r2=279768&view=diff
==============================================================================
--- compiler-rt/trunk/test/lsan/TestCases/guard-page.c (original)
+++ compiler-rt/trunk/test/lsan/TestCases/guard-page.c Thu Aug 25 14:08:10 2016
@@ -22,7 +22,7 @@ static void die(const char* msg, int err
 static void ctxfunc() {
   pthread_mutex_lock(&mutex);
   ctxfunc_started = 1;
-  printf("ctxfunc\n");
+  // printf("ctxfunc\n");
   pthread_cond_signal(&cond);
   pthread_mutex_unlock(&mutex);
   // Leave this context alive when the program exits.
@@ -36,11 +36,11 @@ static void* thread(void* arg) {
 
   if (getcontext(&ctx) < 0)
     die("getcontext", 0);
-  stack = malloc(1 << 12);
+  stack = malloc(1 << 11);
   if (stack == NULL)
     die("malloc", 0);
   ctx.uc_stack.ss_sp = stack;
-  ctx.uc_stack.ss_size = 1 << 12;
+  ctx.uc_stack.ss_size = 1 << 11;
   makecontext(&ctx, ctxfunc, 0);
   setcontext(&ctx);
   die("setcontext", 0);