From fa3d319ac6f648dc51cb672a31f482c80a8c2457 Mon Sep 17 00:00:00 2001
From: Pavel Machek <pavel@suse.cz>
Date: Thu, 26 Jun 2008 00:25:43 +0200
Subject: [PATCH] pci-gart_64.c: could we get better explanation?

Add better explanation to pci-gart.

Signed-off-by: Pavel Machek <pavel@suse.cz>
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Trivial patch monkey <trivial@kernel.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/pci-gart_64.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index f505c38903..f20c20a785 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -824,10 +824,10 @@ void __init gart_iommu_init(void)
 	wbinvd();
 
 	/*
-	 * Try to workaround a bug (thanks to BenH)
+	 * Try to workaround a bug (thanks to BenH):
 	 * Set unmapped entries to a scratch page instead of 0.
 	 * Any prefetches that hit unmapped entries won't get an bus abort
-	 * then.
+	 * then. (P2P bridge may be prefetching on DMA reads).
 	 */
 	scratch = get_zeroed_page(GFP_KERNEL);
 	if (!scratch)
-- 
2.39.5