compiler-gcc.h: add more comments to RELOC_HIDE

Requested by C. Lameter

Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Mike Travis <travis@sgi.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Richard Henderson <rth@twiddle.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

authored by

Andi Kleen and committed by
Linus Torvalds
85c210ed 2d96d105

+12 -2
+12 -2
include/linux/compiler-gcc.h
··· 11 11 /* The "volatile" is due to gcc bugs */ 12 12 #define barrier() __asm__ __volatile__("": : :"memory") 13 13 14 - /* This macro obfuscates arithmetic on a variable address so that gcc 15 - shouldn't recognize the original var, and make assumptions about it */ 16 14 /* 15 + * This macro obfuscates arithmetic on a variable address so that gcc 16 + * shouldn't recognize the original var, and make assumptions about it. 17 + * 18 + * This is needed because the C standard makes it undefined to do 19 + * pointer arithmetic on "objects" outside their boundaries and the 20 + * gcc optimizers assume this is the case. In particular they 21 + * assume such arithmetic does not wrap. 22 + * 23 + * A miscompilation has been observed because of this on PPC. 24 + * To work around it we hide the relationship of the pointer and the object 25 + * using this macro. 26 + * 17 27 * Versions of the ppc64 compiler before 4.1 had a bug where use of 18 28 * RELOC_HIDE could trash r30. The bug can be worked around by changing 19 29 * the inline assembly constraint from =g to =r, in this particular