Supplying entropy from etext has created a regression on arm where we

get an alignment fault while copying the data.  Turns out that since we
have .rodata in the text segment, it's very easily possible that etext
remains unaligned.  Work around this by word-aligning etext.  The next
step is to split .rodata out of the text segment.

ok deraadt@

diff --git a/sys/arch/arm/conf/kern.ldscript b/sys/arch/arm/conf/kern.ldscript
index 77fc994..d844dd6 100644 (file)
--- a/sys/arch/arm/conf/kern.ldscript
+++ b/sys/arch/arm/conf/kern.ldscript
@@ -1,4 +1,4 @@
-/*     $OpenBSD: kern.ldscript,v 1.1 2018/02/10 06:52:47 patrick Exp $ */
+/*     $OpenBSD: kern.ldscript,v 1.2 2018/02/10 07:02:21 patrick Exp $ */
 /*     $NetBSD: ldscript.evbarm,v 1.2 2003/03/05 23:54:22 thorpej Exp $        */
 
 OUTPUT_ARCH(arm)
@@ -24,6 +24,7 @@ SECTIONS
     *(.glue_7t) *(.glue_7)
     *(.rodata) *(.rodata.*)
   } =0
+  . = ALIGN(4);
   PROVIDE (__etext = .);
   PROVIDE (_etext = .);
   PROVIDE (etext = .);