diff options
-rw-r--r-- | arch/x86/Kconfig.cpu | 11 |
1 files changed, 10 insertions, 1 deletions
diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu index 018d68e3184e..e09a6b73a1aa 100644 --- a/arch/x86/Kconfig.cpu +++ b/arch/x86/Kconfig.cpu | |||
@@ -321,8 +321,17 @@ config X86_XADD | |||
321 | depends on X86_32 && !M386 | 321 | depends on X86_32 && !M386 |
322 | 322 | ||
323 | config X86_PPRO_FENCE | 323 | config X86_PPRO_FENCE |
324 | def_bool y | 324 | bool "PentiumPro memory ordering errata workaround" |
325 | depends on M686 || M586MMX || M586TSC || M586 || M486 || M386 || MGEODEGX1 | 325 | depends on M686 || M586MMX || M586TSC || M586 || M486 || M386 || MGEODEGX1 |
326 | help | ||
327 | Old PentiumPro multiprocessor systems had errata that could cause memory | ||
328 | operations to violate the x86 ordering standard in rare cases. Enabling this | ||
329 | option will attempt to work around some (but not all) occurances of | ||
330 | this problem, at the cost of much heavier spinlock and memory barrier | ||
331 | operations. | ||
332 | |||
333 | If unsure, say n here. Even distro kernels should think twice before enabling | ||
334 | this: there are few systems, and an unlikely bug. | ||
326 | 335 | ||
327 | config X86_F00F_BUG | 336 | config X86_F00F_BUG |
328 | def_bool y | 337 | def_bool y |