[PATCH v3 1/3] x86: Move DEBUG_TLBFLUSH option.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



CONFIG_DEBUG_TLBFLUSH was added in

'commit 3df3212f9722 ("x86/tlb: add tlb_flushall_shift knob into debugfs")'
to support tlb_flushall_shift knob. The knob was removed in

'commit e9f4e0a9fe27 ("x86/mm: Rip out complicated, out-of-date, buggy
TLB flushing")'.
However, the debug option was never removed from Kconfig. It was reused
in commit

'9824cf9753ec ("mm: vmstats: tlb flush counters")'
but the commit text was never updated accordingly.

Update the Kconfig option description as per its current usage.

Take this opportunity to make this kconfig option a common option as it
touches the common vmstat code. Introduce another arch specific config
HAVE_ARCH_DEBUG_TLBFLUSH that can be selected to enable this config.

Signed-off-by: Atish Patra <atish.patra@xxxxxxx>
---
 arch/x86/Kconfig       |  1 +
 arch/x86/Kconfig.debug | 19 -------------------
 mm/Kconfig.debug       | 13 +++++++++++++
 3 files changed, 14 insertions(+), 19 deletions(-)

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 62fc3fda1a05..4c59f59e9491 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -132,6 +132,7 @@ config X86
 	select HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD if X86_64
 	select HAVE_ARCH_VMAP_STACK		if X86_64
 	select HAVE_ARCH_WITHIN_STACK_FRAMES
+	select HAVE_ARCH_DEBUG_TLBFLUSH		if DEBUG_KERNEL
 	select HAVE_CMPXCHG_DOUBLE
 	select HAVE_CMPXCHG_LOCAL
 	select HAVE_CONTEXT_TRACKING		if X86_64
diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index 15d0fbe27872..0c8f9931e901 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -124,25 +124,6 @@ config DOUBLEFAULT
 	  option saves about 4k and might cause you much additional grey
 	  hair.
 
-config DEBUG_TLBFLUSH
-	bool "Set upper limit of TLB entries to flush one-by-one"
-	depends on DEBUG_KERNEL
-	---help---
-
-	X86-only for now.
-
-	This option allows the user to tune the amount of TLB entries the
-	kernel flushes one-by-one instead of doing a full TLB flush. In
-	certain situations, the former is cheaper. This is controlled by the
-	tlb_flushall_shift knob under /sys/kernel/debug/x86. If you set it
-	to -1, the code flushes the whole TLB unconditionally. Otherwise,
-	for positive values of it, the kernel will use single TLB entry
-	invalidating instructions according to the following formula:
-
-	flush_entries <= active_tlb_entries / 2^tlb_flushall_shift
-
-	If in doubt, say "N".
-
 config IOMMU_DEBUG
 	bool "Enable IOMMU debugging"
 	depends on GART_IOMMU && DEBUG_KERNEL
diff --git a/mm/Kconfig.debug b/mm/Kconfig.debug
index e3df921208c0..e8622b26f0c2 100644
--- a/mm/Kconfig.debug
+++ b/mm/Kconfig.debug
@@ -111,3 +111,16 @@ config DEBUG_RODATA_TEST
     depends on STRICT_KERNEL_RWX
     ---help---
       This option enables a testcase for the setting rodata read-only.
+
+config HAVE_ARCH_DEBUG_TLBFLUSH
+	bool
+	depends on DEBUG_KERNEL
+
+config DEBUG_TLBFLUSH
+	bool "Save tlb flush statistics to vmstat"
+	depends on HAVE_ARCH_DEBUG_TLBFLUSH
+	help
+
+	Add tlbflush statistics to vmstat. It is really helpful understand tlbflush
+	performance and behavior. It should be enabled only for debugging purpose
+	by individual architectures explicitly by selecting HAVE_ARCH_DEBUG_TLBFLUSH.
-- 
2.21.0




[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux