ruby-changes:40766
From: nobu <ko1@a...>
Date: Wed, 2 Dec 2015 09:55:55 +0900 (JST)
Subject: [ruby-changes:40766] nobu:r52845 (trunk): explicit_bzero.c: fix comments
nobu 2015-12-02 09:55:50 +0900 (Wed, 02 Dec 2015) New Revision: 52845 http://svn.ruby-lang.org/cgi-bin/viewvc.cgi?view=revision&revision=52845 Log: explicit_bzero.c: fix comments * missing/explicit_bzero.c: fix comments. Microsoft Visual Studio does not provide memset_s() in 12.0 at least. Modified files: trunk/missing/explicit_bzero.c Index: missing/explicit_bzero.c =================================================================== --- missing/explicit_bzero.c (revision 52844) +++ missing/explicit_bzero.c (revision 52845) @@ -5,17 +5,18 @@ https://github.com/ruby/ruby/blob/trunk/missing/explicit_bzero.c#L5 #include <windows.h> #endif -/* Similar to bzero(), but have a guarantee not to be eliminated from compiler +/* Similar to bzero(), but has a guarantee not to be eliminated from compiler optimization. */ /* OS support note: - * BSD have explicit_bzero(). - * Windows, OS-X have memset_s(). + * BSDs have explicit_bzero(). + * OS-X has memset_s(). + * Windows has SecureZeroMemory() since XP. * Linux has none. *Sigh* */ /* - * Following URL explain why memset_s is added to the standard. + * Following URL explains why memset_s is added to the standard. * http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1381.pdf */ @@ -40,7 +41,7 @@ explicit_bzero(void *b, size_t len) https://github.com/ruby/ruby/blob/trunk/missing/explicit_bzero.c#L41 #elif defined HAVE_FUNC_WEAK -/* A weak function never be optimization away. Even if nobody use it. */ +/* A weak function never be optimized away. Even if nobody uses it. */ WEAK(void ruby_explicit_bzero_hook_unused(void *buf, size_t len)); void ruby_explicit_bzero_hook_unused(void *buf, size_t len) @@ -64,8 +65,8 @@ explicit_bzero(void *b, size_t len) https://github.com/ruby/ruby/blob/trunk/missing/explicit_bzero.c#L65 { /* * volatile is not enough if compiler have a LTO (link time - * optimization). At least, the standard provide no guarantee. - * However, gcc and major other compiler never optimization a volatile + * optimization). At least, the standard provides no guarantee. + * However, gcc and major other compiler never optimize a volatile * variable away. So, using volatile is practically ok. */ volatile char* p = (volatile char*)b; -- ML: ruby-changes@q... Info: http://www.atdot.net/~ko1/quickml/