4 files changed, 97 insertions, 18 deletions

diff --git a/util/coroutine-sigaltstack.c b/util/coroutine-sigaltstack.c
index a7c3366553..f6fc49a0e5 100644
--- a/util/coroutine-sigaltstack.c
+++ b/util/coroutine-sigaltstack.c
@@ -33,8 +33,9 @@
 typedef struct {
     Coroutine base;
     void *stack;
+    size_t stack_size;
     sigjmp_buf env;
-} CoroutineUContext;
+} CoroutineSigAltStack;
 
 /**
  * Per-thread coroutine bookkeeping
@@ -44,7 +45,7 @@ typedef struct {
     Coroutine *current;
 
     /** The default coroutine */
-    CoroutineUContext leader;
+    CoroutineSigAltStack leader;
 
     /** Information for the signal handler (trampoline) */
     sigjmp_buf tr_reenter;
@@ -89,7 +90,7 @@ static void __attribute__((constructor)) coroutine_init(void)
  * (from the signal handler when it is not signal handling, read ahead
  * for more information).
  */
-static void coroutine_bootstrap(CoroutineUContext *self, Coroutine *co)
+static void coroutine_bootstrap(CoroutineSigAltStack *self, Coroutine *co)
 {
     /* Initialize longjmp environment and switch back the caller */
     if (!sigsetjmp(self->env, 0)) {
@@ -109,7 +110,7 @@ static void coroutine_bootstrap(CoroutineUContext *self, Coroutine *co)
  */
 static void coroutine_trampoline(int signal)
 {
-    CoroutineUContext *self;
+    CoroutineSigAltStack *self;
     Coroutine *co;
     CoroutineThreadState *coTS;
 
@@ -143,8 +144,7 @@ static void coroutine_trampoline(int signal)
 
 Coroutine *qemu_coroutine_new(void)
 {
-    const size_t stack_size = 1 << 20;
-    CoroutineUContext *co;
+    CoroutineSigAltStack *co;
     CoroutineThreadState *coTS;
     struct sigaction sa;
     struct sigaction osa;
@@ -164,7 +164,8 @@ Coroutine *qemu_coroutine_new(void)
      */
 
     co = g_malloc0(sizeof(*co));
-    co->stack = g_malloc(stack_size);
+    co->stack_size = COROUTINE_STACK_SIZE;
+    co->stack = qemu_alloc_stack(&co->stack_size);
     co->base.entry_arg = &old_env; /* stash away our jmp_buf */
 
     coTS = coroutine_get_thread_state();
@@ -189,7 +190,7 @@ Coroutine *qemu_coroutine_new(void)
      * Set the new stack.
      */
     ss.ss_sp = co->stack;
-    ss.ss_size = stack_size;
+    ss.ss_size = co->stack_size;
     ss.ss_flags = 0;
     if (sigaltstack(&ss, &oss) < 0) {
         abort();
@@ -251,17 +252,17 @@ Coroutine *qemu_coroutine_new(void)
 
 void qemu_coroutine_delete(Coroutine *co_)
 {
-    CoroutineUContext *co = DO_UPCAST(CoroutineUContext, base, co_);
+    CoroutineSigAltStack *co = DO_UPCAST(CoroutineSigAltStack, base, co_);
 
-    g_free(co->stack);
+    qemu_free_stack(co->stack, co->stack_size);
     g_free(co);
 }
 
 CoroutineAction qemu_coroutine_switch(Coroutine *from_, Coroutine *to_,
                                       CoroutineAction action)
 {
-    CoroutineUContext *from = DO_UPCAST(CoroutineUContext, base, from_);
-    CoroutineUContext *to = DO_UPCAST(CoroutineUContext, base, to_);
+    CoroutineSigAltStack *from = DO_UPCAST(CoroutineSigAltStack, base, from_);
+    CoroutineSigAltStack *to = DO_UPCAST(CoroutineSigAltStack, base, to_);
     CoroutineThreadState *s = coroutine_get_thread_state();
     int ret;
 
diff --git a/util/coroutine-ucontext.c b/util/coroutine-ucontext.c
index 2bb7e10d4b..6621f3f692 100644
--- a/util/coroutine-ucontext.c
+++ b/util/coroutine-ucontext.c
@@ -34,6 +34,7 @@
 typedef struct {
     Coroutine base;
     void *stack;
+    size_t stack_size;
     sigjmp_buf env;
 
 #ifdef CONFIG_VALGRIND_H
@@ -82,7 +83,6 @@ static void coroutine_trampoline(int i0, int i1)
 
 Coroutine *qemu_coroutine_new(void)
 {
-    const size_t stack_size = 1 << 20;
     CoroutineUContext *co;
     ucontext_t old_uc, uc;
     sigjmp_buf old_env;
@@ -101,17 +101,18 @@ Coroutine *qemu_coroutine_new(void)
     }
 
     co = g_malloc0(sizeof(*co));
-    co->stack = g_malloc(stack_size);
+    co->stack_size = COROUTINE_STACK_SIZE;
+    co->stack = qemu_alloc_stack(&co->stack_size);
     co->base.entry_arg = &old_env; /* stash away our jmp_buf */
 
     uc.uc_link = &old_uc;
     uc.uc_stack.ss_sp = co->stack;
-    uc.uc_stack.ss_size = stack_size;
+    uc.uc_stack.ss_size = co->stack_size;
     uc.uc_stack.ss_flags = 0;
 
 #ifdef CONFIG_VALGRIND_H
     co->valgrind_stack_id =
-        VALGRIND_STACK_REGISTER(co->stack, co->stack + stack_size);
+        VALGRIND_STACK_REGISTER(co->stack, co->stack + co->stack_size);
 #endif
 
     arg.p = co;
@@ -149,7 +150,7 @@ void qemu_coroutine_delete(Coroutine *co_)
     valgrind_stack_deregister(co);
 #endif
 
-    g_free(co->stack);
+    qemu_free_stack(co->stack, co->stack_size);
     g_free(co);
 }
 
diff --git a/util/coroutine-win32.c b/util/coroutine-win32.c
index 02e28e825f..de6bd4fd3e 100644
--- a/util/coroutine-win32.c
+++ b/util/coroutine-win32.c
@@ -71,7 +71,7 @@ static void CALLBACK coroutine_trampoline(void *co_)
 
 Coroutine *qemu_coroutine_new(void)
 {
-    const size_t stack_size = 1 << 20;
+    const size_t stack_size = COROUTINE_STACK_SIZE;
     CoroutineWin32 *co;
 
     co = g_malloc0(sizeof(*co));
diff --git a/util/oslib-posix.c b/util/oslib-posix.c
index f2d4e9e592..aaec1891f5 100644
--- a/util/oslib-posix.c
+++ b/util/oslib-posix.c
@@ -50,6 +50,10 @@
 
 #include "qemu/mmap-alloc.h"
 
+#ifdef CONFIG_DEBUG_STACK_USAGE
+#include "qemu/error-report.h"
+#endif
+
 int qemu_get_thread_id(void)
 {
 #if defined(__linux__)
@@ -499,3 +503,76 @@ pid_t qemu_fork(Error **errp)
     }
     return pid;
 }
+
+void *qemu_alloc_stack(size_t *sz)
+{
+    void *ptr, *guardpage;
+#ifdef CONFIG_DEBUG_STACK_USAGE
+    void *ptr2;
+#endif
+    size_t pagesz = getpagesize();
+#ifdef _SC_THREAD_STACK_MIN
+    /* avoid stacks smaller than _SC_THREAD_STACK_MIN */
+    long min_stack_sz = sysconf(_SC_THREAD_STACK_MIN);
+    *sz = MAX(MAX(min_stack_sz, 0), *sz);
+#endif
+    /* adjust stack size to a multiple of the page size */
+    *sz = ROUND_UP(*sz, pagesz);
+    /* allocate one extra page for the guard page */
+    *sz += pagesz;
+
+    ptr = mmap(NULL, *sz, PROT_READ | PROT_WRITE,
+               MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
+    if (ptr == MAP_FAILED) {
+        abort();
+    }
+
+#if defined(HOST_IA64)
+    /* separate register stack */
+    guardpage = ptr + (((*sz - pagesz) / 2) & ~pagesz);
+#elif defined(HOST_HPPA)
+    /* stack grows up */
+    guardpage = ptr + *sz - pagesz;
+#else
+    /* stack grows down */
+    guardpage = ptr;
+#endif
+    if (mprotect(guardpage, pagesz, PROT_NONE) != 0) {
+        abort();
+    }
+
+#ifdef CONFIG_DEBUG_STACK_USAGE
+    for (ptr2 = ptr + pagesz; ptr2 < ptr + *sz; ptr2 += sizeof(uint32_t)) {
+        *(uint32_t *)ptr2 = 0xdeadbeaf;
+    }
+#endif
+
+    return ptr;
+}
+
+#ifdef CONFIG_DEBUG_STACK_USAGE
+static __thread unsigned int max_stack_usage;
+#endif
+
+void qemu_free_stack(void *stack, size_t sz)
+{
+#ifdef CONFIG_DEBUG_STACK_USAGE
+    unsigned int usage;
+    void *ptr;
+
+    for (ptr = stack + getpagesize(); ptr < stack + sz;
+         ptr += sizeof(uint32_t)) {
+        if (*(uint32_t *)ptr != 0xdeadbeaf) {
+            break;
+        }
+    }
+    usage = sz - (uintptr_t) (ptr - stack);
+    if (usage > max_stack_usage) {
+        error_report("thread %d max stack usage increased from %u to %u",
+                     qemu_get_thread_id(), max_stack_usage, usage);
+        max_stack_usage = usage;
+    }
+#endif
+
+    munmap(stack, sz);
+}