From: Namhyung Kim on
Until all supported versions of gcc recognize -fno-strict-overflow, we should
keep the RELOC_HIDE magic on __pa_symbol(). Comment it.

Signed-off-by: Namhyung Kim <namhyung(a)gmail.com>
Suggested-by: Ingo Molnar <mingo(a)elte.hu>

---

I hope I'm doing right. :-)

arch/x86/include/asm/page.h | 7 +++++++
1 files changed, 7 insertions(+), 0 deletions(-)

diff --git a/arch/x86/include/asm/page.h b/arch/x86/include/asm/page.h
index 625c3f0..06cb6f3 100644
--- a/arch/x86/include/asm/page.h
+++ b/arch/x86/include/asm/page.h
@@ -37,6 +37,13 @@ static inline void copy_user_page(void *to, void *from, unsigned long vaddr,
#define __pa_nodebug(x) __phys_addr_nodebug((unsigned long)(x))
/* __pa_symbol should be used for C visible symbols.
This seems to be the official gcc blessed way to do such arithmetic. */
+/*
+ * we need __phys_reloc_hide() here because gcc may assume that there is no
+ * overflow during __pa() calculation and can optimize it unexpectedly.
+ * Newer versions of gcc provide -fno-strict-overflow switch to handle this
+ * case properly. Once all supported versions of gcc understand it, we can
+ * remove this Voodoo magic stuff.
+ */
#define __pa_symbol(x) __pa(__phys_reloc_hide((unsigned long)(x)))

#define __va(x) ((void *)((unsigned long)(x)+PAGE_OFFSET))
--
1.7.0.4

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/