Merge branch 'sverker/halt-unicode-slogan/OTP-14553' into maint

* sverker/halt-unicode-slogan/OTP-14553:
  erts: Allow any unicode string as crash dump slogan
  erts: Generate crash_dump slogan string as UTF8
  erts: Refactor erts_unicode_list_to_buf

5 files changed, 65 insertions, 52 deletions

diff --git a/erts/doc/src/erlang.xml b/erts/doc/src/erlang.xml
index 105734d5b2..11f2d36a01 100644
--- a/erts/doc/src/erlang.xml
+++ b/erts/doc/src/erlang.xml
@@ -1943,23 +1943,26 @@ os_prompt%</pre>
           <item>The runtime system exits with integer value
             <c><anno>Status</anno></c>
             as status code to the calling environment (OS).
+	    <note>
+	      <p>On many platforms, the OS supports only status
+	      codes 0-255. A too large status code is truncated by clearing
+	      the high bits.</p>
+	    </note>
           </item>
           <tag>string()</tag>
           <item>An Erlang crash dump is produced with <c><anno>Status</anno></c>
             as slogan. Then the runtime system exits with status code <c>1</c>.
-            Note that only code points in the range 0-255 may be used
-	    and the string will be truncated if longer than 200 characters.
+	    The string will be truncated if longer than 200 characters.
+	    <note>
+	      <p>Before ERTS 9.1 (OTP-20.1) only code points in the range 0-255
+	         was accepted in the string. Now any unicode string is valid.</p>
+	    </note>
           </item>
           <tag><c>abort</c></tag>
           <item>The runtime system aborts producing a core dump, if that is
             enabled in the OS.
           </item>
         </taglist>
-        <note>
-          <p>On many platforms, the OS supports only status
-            codes 0-255. A too large status code is truncated by clearing
-            the high bits.</p>
-        </note>
         <p>For integer <c><anno>Status</anno></c>, the Erlang runtime system
           closes all ports and allows async threads to finish their
           operations before exiting. To exit without such flushing, use
diff --git a/erts/emulator/beam/bif.c b/erts/emulator/beam/bif.c
index 890277a3ba..52842a26eb 100644
--- a/erts/emulator/beam/bif.c
+++ b/erts/emulator/beam/bif.c
@@ -3018,17 +3018,17 @@ BIF_RETTYPE list_to_atom_1(BIF_ALIST_1)
 {
     Eterm res;
     byte *buf = (byte *) erts_alloc(ERTS_ALC_T_TMP, MAX_ATOM_SZ_LIMIT);
-    Sint i = erts_unicode_list_to_buf(BIF_ARG_1, buf, MAX_ATOM_CHARACTERS);
-
+    Sint written;
+    int i = erts_unicode_list_to_buf(BIF_ARG_1, buf, MAX_ATOM_CHARACTERS,
+                                     &written);
     if (i < 0) {
 	erts_free(ERTS_ALC_T_TMP, (void *) buf);
-	i = erts_list_length(BIF_ARG_1);
-	if (i > MAX_ATOM_CHARACTERS) {
+	if (i == -2) {
 	    BIF_ERROR(BIF_P, SYSTEM_LIMIT);
 	}
 	BIF_ERROR(BIF_P, BADARG);
     }
-    res = erts_atom_put(buf, i, ERTS_ATOM_ENC_UTF8, 1);
+    res = erts_atom_put(buf, written, ERTS_ATOM_ENC_UTF8, 1);
     ASSERT(is_atom(res));
     erts_free(ERTS_ALC_T_TMP, (void *) buf);
     BIF_RET(res);
@@ -3039,8 +3039,9 @@ BIF_RETTYPE list_to_atom_1(BIF_ALIST_1)
 BIF_RETTYPE list_to_existing_atom_1(BIF_ALIST_1)
 {
     byte *buf = (byte *) erts_alloc(ERTS_ALC_T_TMP, MAX_ATOM_SZ_LIMIT);
-    Sint i = erts_unicode_list_to_buf(BIF_ARG_1, buf, MAX_ATOM_CHARACTERS);
-
+    Sint written;
+    int i = erts_unicode_list_to_buf(BIF_ARG_1, buf, MAX_ATOM_CHARACTERS,
+                                     &written);
     if (i < 0) {
     error:
 	erts_free(ERTS_ALC_T_TMP, (void *) buf);
@@ -3048,7 +3049,7 @@ BIF_RETTYPE list_to_existing_atom_1(BIF_ALIST_1)
     } else {
 	Eterm a;
 	
-	if (erts_atom_get((char *) buf, i, &a, ERTS_ATOM_ENC_UTF8)) {
+	if (erts_atom_get((char *) buf, written, &a, ERTS_ATOM_ENC_UTF8)) {
 	    erts_free(ERTS_ALC_T_TMP, (void *) buf);
 	    BIF_RET(a);
 	} else {
@@ -3967,9 +3968,6 @@ BIF_RETTYPE display_nl_0(BIF_ALIST_0)
 /**********************************************************************/
 
 
-#define HALT_MSG_SIZE	200
-static char halt_msg[HALT_MSG_SIZE+1];
-
 /* stop the system with exit code and flags */
 BIF_RETTYPE halt_2(BIF_ALIST_2)
 {
@@ -4019,16 +4017,17 @@ BIF_RETTYPE halt_2(BIF_ALIST_2)
 	erts_smp_proc_unlock(BIF_P, ERTS_PROC_LOCK_MAIN);
 	erts_exit(ERTS_ABORT_EXIT, "");
     }
-    else if (is_string(BIF_ARG_1) || BIF_ARG_1 == NIL) {
-	Sint i;
+    else if (is_list(BIF_ARG_1) || BIF_ARG_1 == NIL) {
+#       define HALT_MSG_SIZE 200
+        static byte halt_msg[4*HALT_MSG_SIZE+1];
+        Sint written;
 
-        if ((i = intlist_to_buf(BIF_ARG_1, halt_msg, HALT_MSG_SIZE)) == -1) {
+        if (erts_unicode_list_to_buf(BIF_ARG_1, halt_msg, HALT_MSG_SIZE,
+                                     &written) == -1 ) {
             goto error;
         }
-        if (i == -2) /* truncated string */
-            i = HALT_MSG_SIZE;
-        ASSERT(i >= 0 && i <= HALT_MSG_SIZE);
-	halt_msg[i] = '\0';
+        ASSERT(written >= 0 && written < sizeof(halt_msg));
+	halt_msg[written] = '\0';
 	VERBOSE(DEBUG_SYSTEM,
 		("System halted by BIF halt(%T, %T)\n", BIF_ARG_1, BIF_ARG_2));
 	erts_smp_proc_unlock(BIF_P, ERTS_PROC_LOCK_MAIN);
diff --git a/erts/emulator/beam/global.h b/erts/emulator/beam/global.h
index 2105ee7a6c..717ecf5e6a 100644
--- a/erts/emulator/beam/global.h
+++ b/erts/emulator/beam/global.h
@@ -1288,7 +1288,7 @@ int erts_utf8_to_latin1(byte* dest, const byte* source, int slen);
 
 void bin_write(fmtfn_t, void*, byte*, size_t);
 Sint intlist_to_buf(Eterm, char*, Sint); /* most callers pass plain char*'s */
-Sint erts_unicode_list_to_buf(Eterm list, byte *buf, Sint len);
+int erts_unicode_list_to_buf(Eterm list, byte *buf, Sint len, Sint* written);
 
 struct Sint_buf {
 #if defined(ARCH_64)
diff --git a/erts/emulator/beam/utils.c b/erts/emulator/beam/utils.c
index 0fb25c2082..cb8d93549c 100644
--- a/erts/emulator/beam/utils.c
+++ b/erts/emulator/beam/utils.c
@@ -3635,30 +3635,40 @@ intlist_to_buf(Eterm list, char *buf, Sint len)
     return -2;			/* not enough space */
 }
 
-/* Fill buf with the contents of the unicode list.
- * Return the number of bytes in the buffer,
- * or -1 for type error,
- * or -2 for not enough buffer space (buffer contains truncated result).
+/** @brief Fill buf with the UTF8 contents of the unicode list
+ * @param len Max number of characters to write.
+ * @param written NULL or bytes written.
+ * @return 0 ok,
+ *        -1 type error,
+ *        -2 list too long, only \c len characters written
  */
-Sint
-erts_unicode_list_to_buf(Eterm list, byte *buf, Sint len)
+int
+erts_unicode_list_to_buf(Eterm list, byte *buf, Sint len, Sint* written)
 {
     Eterm* listptr;
     Sint sz = 0;
+    Sint val;
+    int res;
 
-    if (is_nil(list)) {
-	return 0;
-    }
-    if (is_not_list(list)) {
-	return -1;
-    }
-    listptr = list_val(list);
+    while (1) {
+        if (is_nil(list)) {
+            res = 0;
+            break;
+        }
+        if (is_not_list(list)) {
+            res = -1;
+            break;
+        }
+        listptr = list_val(list);
 
-    while (len-- > 0) {
-	Sint val;
+        if (len-- <= 0) {
+            res = -2;
+            break;
+        }
 
 	if (is_not_small(CAR(listptr))) {
-	    return -1;
+	    res = -1;
+            break;
 	}
 	val = signed_val(CAR(listptr));
 	if (0 <= val && val < 0x80) {
@@ -3670,7 +3680,8 @@ erts_unicode_list_to_buf(Eterm list, byte *buf, Sint len)
 	    sz += 2;
 	} else if (val < 0x10000UL) {
 	    if (0xD800 <= val && val <= 0xDFFF) {
-		return -1;
+		res = -1;
+                break;
 	    }
 	    buf[sz+0] = 0xE0 | (val >> 12);
 	    buf[sz+1] = 0x80 | ((val >> 6) & 0x3F);
@@ -3683,18 +3694,15 @@ erts_unicode_list_to_buf(Eterm list, byte *buf, Sint len)
 	    buf[sz+3] = 0x80 | (val & 0x3F);
 	    sz += 4;
 	} else {
-	    return -1;
+            res = -1;
+            break;
 	}
 	list = CDR(listptr);
-	if (is_nil(list)) {
-	    return sz;
-	}
-	if (is_not_list(list)) {
-	    return -1;
-	}
-	listptr = list_val(list);
     }
-    return -2;			/* not enough space */
+
+    if (written)
+        *written = sz;
+    return res;
 }
 
 /*
diff --git a/erts/emulator/test/bif_SUITE.erl b/erts/emulator/test/bif_SUITE.erl
index 339c827602..4908ac410c 100644
--- a/erts/emulator/test/bif_SUITE.erl
+++ b/erts/emulator/test/bif_SUITE.erl
@@ -691,6 +691,9 @@ erlang_halt(Config) when is_list(Config) ->
     {badrpc,nodedown} = rpc:call(N3, erlang, halt, [0,[]]),
     {ok,N4} = slave:start(H, halt_node4),
     {badrpc,nodedown} = rpc:call(N4, erlang, halt, [lists:duplicate(300,$x)]),
+    %% Test unicode slogan
+    {ok,N4} = slave:start(H, halt_node4),
+    {badrpc,nodedown} = rpc:call(N4, erlang, halt, [[339,338,254,230,198,295,167,223,32,12507,12531,12480]]),
 
     % This test triggers a segfault when dumping a crash dump
     % to make sure that we can handle it properly.