From d2e5bc50742b66f6a8816c4036c6026726b87f01 Mon Sep 17 00:00:00 2001 From: "Frank Ch. Eigler" Date: Mon, 2 Jun 2008 15:32:24 -0400 Subject: PR6534: add a comment blurb explaining sufficiency UTS_RELEASE compile-time check --- translate.cxx | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/translate.cxx b/translate.cxx index 219e2c33..f411bd8f 100644 --- a/translate.cxx +++ b/translate.cxx @@ -1090,6 +1090,12 @@ c_unparser::emit_module_init () o->newline() << "{"; o->newline(1) << "const char* release = UTS_RELEASE;"; + // NB: This UTS_RELEASE compile-time macro directly checks only that + // the compile-time kbuild tree matches the compile-time debuginfo/etc. + // It does not check the run time kernel value. However, this is + // probably OK since the kbuild modversions system aims to prevent + // mismatches between kbuild and runtime versions at module-loading time. + // o->newline() << "const char* machine = UTS_MACHINE;"; // NB: We could compare UTS_MACHINE too, but on x86 it lies // (UTS_MACHINE=i386, but uname -m is i686). Sheesh. -- cgit