next-0402: build fails at mm/debug-pagealloc.c
Subsystems (8)
mm/debug-pagealloc.c: In function 'set_page_poison':mm/debug-pagealloc.c:8: error: 'struct page' has no member named 'debug_flags' mm/debug-pagealloc.c: In function 'clear_page_poison':mm/debug-pagealloc.c:13: error: 'struct page' has no member named 'debug_flags' mm/debug-pagealloc.c: In function 'page_poison':mm/debug-pagealloc.c:18: error: 'struct page' has no member named 'debug_flags' mm/debug-pagealloc.c: At top level:mm/debug-pagealloc.c:120: error: redefinition of 'kernel_map_pages' include/linux/mm.h:1278: error: previous definition of 'kernel_map_pages' was heremm/debug-pagealloc.c: In function 'kernel_map_pages':mm/debug-pagealloc.c:122: error: 'debug_pagealloc_enabled' undeclared (first use in this function)
Should debug_flags really be in vm_area_struct instead of struct page ?
PATCH 1/1 build fix for generic debug pagealloc
Quoted 5 lines
mm/debug-pagealloc.c: In function 'set_page_poison':mm/debug-pagealloc.c:8: error: 'struct page' has no member named 'debug_flags'mm/debug-pagealloc.c: In function 'clear_page_poison':mm/debug-pagealloc.c:13: error: 'struct page' has no member named 'debug_flags'mm/debug-pagealloc.c: In function 'page_poison':
Quoted 8 lines
mm/debug-pagealloc.c:18: error: 'struct page' has no member named 'debug_flags'mm/debug-pagealloc.c: At top level:mm/debug-pagealloc.c:120: error: redefinition of 'kernel_map_pages'include/linux/mm.h:1278: error: previous definition of 'kernel_map_pages' was heremm/debug-pagealloc.c: In function 'kernel_map_pages':mm/debug-pagealloc.c:122: error: 'debug_pagealloc_enabled' undeclared (first use in this function)
This patch fixes build failure with generic debug pagealloc:
- debug_flags should be in struct page
- define DEBUG_PAGEALLOC config option for all architectures
--- arch/powerpc/Kconfig.debug | 10 ---------- arch/s390/Kconfig.debug | 9 --------- arch/sparc/Kconfig.debug | 9 --------- arch/x86/Kconfig.debug | 9 --------- include/linux/mm_types.h | 6 +++--- mm/Kconfig.debug | 9 +++++++++ 6 files changed, 12 insertions(+), 40 deletions(-)
arch/powerpc/Kconfig.debug +0 -10
@@ -27,16 +27,6 @@ config DEBUG_STACK_USAGE This option will slow down process creation somewhat. -config DEBUG_PAGEALLOC - bool "Debug page memory allocations" - depends on DEBUG_KERNEL && !HIBERNATION - depends on ARCH_SUPPORTS_DEBUG_PAGEALLOC - help - Unmap pages from the kernel linear mapping after free_pages(). - This results in a large slowdown, but helps to find certain types - of memory corruptions. - - config HCALL_STATS bool "Hypervisor call instrumentation" depends on PPC_PSERIES && DEBUG_FS
arch/s390/Kconfig.debug +0 -9
@@ -6,13 +6,4 @@ config TRACE_IRQFLAGS_SUPPORT source "lib/Kconfig.debug" -config DEBUG_PAGEALLOC - bool "Debug page memory allocations" - depends on DEBUG_KERNEL - depends on ARCH_SUPPORTS_DEBUG_PAGEALLOC - help - Unmap pages from the kernel linear mapping after free_pages(). - This results in a slowdown, but helps to find certain types of - memory corruptions. - endmenu
arch/sparc/Kconfig.debug +0 -9
@@ -22,15 +22,6 @@ config DEBUG_DCFLUSH config STACK_DEBUG bool "Stack Overflow Detection Support" -config DEBUG_PAGEALLOC - bool "Debug page memory allocations" - depends on DEBUG_KERNEL && !HIBERNATION - depends on ARCH_SUPPORTS_DEBUG_PAGEALLOC - help - Unmap pages from the kernel linear mapping after free_pages(). - This results in a large slowdown, but helps to find certain types - of memory corruptions. - config MCOUNT bool depends on SPARC64
arch/x86/Kconfig.debug +0 -9
@@ -72,15 +72,6 @@ config DEBUG_STACK_USAGE This option will slow down process creation somewhat. -config DEBUG_PAGEALLOC - bool "Debug page memory allocations" - depends on DEBUG_KERNEL - depends on ARCH_SUPPORTS_DEBUG_PAGEALLOC - ---help--- - Unmap pages from the kernel linear mapping after free_pages(). - This results in a large slowdown, but helps to find certain types - of memory corruptions. - config DEBUG_PER_CPU_MAPS bool "Debug access to per_cpu maps" depends on DEBUG_KERNEL
include/linux/mm_types.h +3 -3
@@ -95,6 +95,9 @@ struct page {void *virtual;/* Kernel virtual address (NULL if not kmapped, ie. highmem) */ #endif /* WANT_PAGE_VIRTUAL */ +#ifdef CONFIG_WANT_PAGE_DEBUG_FLAGS +unsigned long debug_flags;/* Use atomic bitops on this */ +#endif };/*
@@ -175,9 +178,6 @@ struct vm_area_struct {#ifdef CONFIG_NUMA struct mempolicy *vm_policy;/* NUMA policy for the VMA */ #endif -#ifdef CONFIG_WANT_PAGE_DEBUG_FLAGS -unsigned long debug_flags;/* Use atomic bitops on this */ -#endif };struct core_thread {
mm/Kconfig.debug +9 -0
@@ -1,3 +1,12 @@ +config DEBUG_PAGEALLOC + bool "Debug page memory allocations" + depends on DEBUG_KERNEL && ARCH_SUPPORTS_DEBUG_PAGEALLOC + depends on !HIBERNATION || !PPC && !SPARC + ---help--- + Unmap pages from the kernel linear mapping after free_pages(). + This results in a large slowdown, but helps to find certain types + of memory corruptions. + config WANT_PAGE_DEBUG_FLAGS bool
Andrew Morton
maintainerRe: [PATCH] build fix for generic debug pagealloc
Quoted 38 lines · On Fri, 3 Apr 2009 01:31:52 +0900 Akinobu Mita <[email protected]> wrote:
Quoted 5 lines
mm/debug-pagealloc.c: In function 'set_page_poison':mm/debug-pagealloc.c:8: error: 'struct page' has no member named 'debug_flags'mm/debug-pagealloc.c: In function 'clear_page_poison':mm/debug-pagealloc.c:13: error: 'struct page' has no member named 'debug_flags'mm/debug-pagealloc.c: In function 'page_poison':Quoted 8 lines
mm/debug-pagealloc.c:18: error: 'struct page' has no member named 'debug_flags'mm/debug-pagealloc.c: At top level:mm/debug-pagealloc.c:120: error: redefinition of 'kernel_map_pages'include/linux/mm.h:1278: error: previous definition of 'kernel_map_pages' was heremm/debug-pagealloc.c: In function 'kernel_map_pages':mm/debug-pagealloc.c:122: error: 'debug_pagealloc_enabled' undeclared (first use in this function)This patch fixes build failure with generic debug pagealloc:
-
debug_flagsshould be in struct page
- defineDEBUG_PAGEALLOCconfig option for all architectures...
arch/x86/Kconfig.debug +0
-0bool "Debug access to per_cpu maps" depends on DEBUG_KERNELinclude/linux/mm_types.h +3
-0@@ -95,6 +95,9 @@ struct page {void *virtual;/* Kernel virtual address (NULL if not kmapped, ie. highmem) */ #endif /* WANT_PAGE_VIRTUAL */ +#ifdef CONFIG_WANT_PAGE_DEBUG_FLAGS +unsigned long debug_flags;/* Use atomic bitops on this */ +#endif };/*
include/linux/mm_types.h struct vm_area_struct { · line 181
Quoted 7 lines
include/linux/mm_types.h +0
-3@@ -175,9 +178,6 @@ struct vm_area_struct {#ifdef CONFIG_NUMA struct mempolicy *vm_policy;/* NUMA policy for the VMA */ #endif -#ifdef CONFIG_WANT_PAGE_DEBUG_FLAGS -unsigned long debug_flags;/* Use atomic bitops on this */ -#endif };
dammit.
Quoted 6 lines
include/linux/mm_types.h +0
-0struct core_thread {mm/Kconfig.debug +0
-0
mm/Kconfig.debug line 9
Quoted 9 lines
mm/Kconfig.debug +9
-0@@ -1,3 +1,12 @@ +config DEBUG_PAGEALLOC + bool "Debug page memory allocations" + depends on DEBUG_KERNEL && ARCH_SUPPORTS_DEBUG_PAGEALLOC + depends on !HIBERNATION || !PPC && !SPARC + ---help--- + Unmap pages from the kernel linear mapping after free_pages(). + This results in a large slowdown, but helps to find certain types + of memory corruptions. +
I guess this is why it wasn't discovered in my compile testing. It's quite regrettable that `make allmodconfig' didn't enable the feature. That's what allmodconfig is for.
Re: [PATCH] build fix for generic debug pagealloc
Quoted 5 lines · Andrew Morton
Quoted 5 lines
mm/Kconfig.debug +5
-0+config DEBUG_PAGEALLOC + bool "Debug page memory allocations" + depends on DEBUG_KERNEL && ARCH_SUPPORTS_DEBUG_PAGEALLOC + depends on !HIBERNATION || !PPC && !SPARC + ---help---
Quoted 8 lines
mm/Kconfig.debug line 9
Quoted 4 lines
mm/Kconfig.debug +4
-0+ Unmap pages from the kernel linear mapping after free_pages(). + This results in a large slowdown, but helps to find certain types + of memory corruptions. +I guess this is why it wasn't discovered in my compile testing. It's quite regrettable that `make allmodconfig' didn't enable the feature. That's what allmodconfig is for.
Hm, i _do_ remember some hibernation magic fix from Rafael, in the area of debugalloc. Many months ago.
Rafael, is it supposed to work? If yes, i think we should unblock it.
Ingo
Re: [PATCH] build fix for generic debug pagealloc
Quoted 17 lines · Ingo Molnar
Quoted 5 lines · Andrew Morton
Quoted 5 lines
mm/Kconfig.debug +5
-0+config DEBUG_PAGEALLOC + bool "Debug page memory allocations" + depends on DEBUG_KERNEL && ARCH_SUPPORTS_DEBUG_PAGEALLOC + depends on !HIBERNATION || !PPC && !SPARC + ---help---Quoted 8 lines
mm/Kconfig.debug line 9
Quoted 4 lines
mm/Kconfig.debug +4
-0+ Unmap pages from the kernel linear mapping after free_pages(). + This results in a large slowdown, but helps to find certain types + of memory corruptions. +I guess this is why it wasn't discovered in my compile testing. It's quite regrettable that `make allmodconfig' didn't enable the feature. That's what allmodconfig is for.
Quoted 5 lines
Hm, i _do_ remember some hibernation magic fix from Rafael, in the area of debugalloc. Many months ago.
Rafael, is it supposed to work? If yes, i think we should unblock it.
Yes, it is, at least on x86.
ISTR using DEBUG_PAGEALLOC with hibernation quite some time ago already.
Thanks,
Rafael
Re: [PATCH] build fix for generic debug pagealloc
Quoted 21 lines · Rafael J. Wysocki
Quoted 20 lines · Ingo Molnar
Quoted 5 lines · Andrew Morton
Quoted 5 lines
mm/Kconfig.debug +5
-0+config DEBUG_PAGEALLOC + bool "Debug page memory allocations" + depends on DEBUG_KERNEL && ARCH_SUPPORTS_DEBUG_PAGEALLOC + depends on !HIBERNATION || !PPC && !SPARC + ---help---Quoted 8 lines
mm/Kconfig.debug line 9
Quoted 4 lines
mm/Kconfig.debug +4
-0+ Unmap pages from the kernel linear mapping after free_pages(). + This results in a large slowdown, but helps to find certain types + of memory corruptions. +I guess this is why it wasn't discovered in my compile testing. It's quite regrettable that `make allmodconfig' didn't enable the feature. That's what allmodconfig is for.
Hm, i _do_ remember some hibernation magic fix from Rafael, in the area of debugalloc. Many months ago.
Quoted 7 lines
Quoted 2 lines
Rafael, is it supposed to work? If yes, i think we should unblock it.
Yes, it is, at least on x86.
ISTR using
DEBUG_PAGEALLOCwith hibernation quite some time ago already.
Could you please send a patch to Andrew then to remove the above dependency?
Ingo