From ba22f13563de5773701fc318ccaaa37b1fb6d294 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Sat, 25 Mar 2006 16:31:31 +0100 Subject: [PATCH] x86_64: Remove CONFIG_UNORDERED_IO It was a failed experiment - all benchmarks done with it on both AMD and Intel showed it was a loss. That was probably because the store buffers of the CPUs for write combining traffic weren't large enough. Signed-off-by: Andi Kleen Signed-off-by: Linus Torvalds --- arch/x86_64/Kconfig | 10 ---------- 1 file changed, 10 deletions(-) (limited to 'arch/x86_64') diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index 6fc63145719..31bab721cb7 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -520,16 +520,6 @@ config PCI_MMCONFIG bool "Support mmconfig PCI config space access" depends on PCI && ACPI -config UNORDERED_IO - bool "Unordered IO mapping access" - depends on EXPERIMENTAL - help - Use unordered stores to access IO memory mappings in device drivers. - Still very experimental. When a driver works on IA64/ppc64/pa-risc it should - work with this option, but it makes the drivers behave differently - from i386. Requires that the driver writer used memory barriers - properly. - source "drivers/pci/pcie/Kconfig" source "drivers/pci/Kconfig" -- cgit v1.2.3