Skip to contentSkip to conversation

kconfig: consolidate arch-specific seccomp options

PATCH In review Dave Hansen 5 participants 14 messages Jan 2, 2014 – Feb 5, 2014

Subsystems (12)
lore
Share & tools
lore
#0 lore raw

PATCH 1/3 kconfig: consolidate arch-specific seccomp options

From: Dave Hansen <[email protected]>

There are 7 architecures with "config SECCOMP". They all have virtually the same help text except for those referencing the /proc interface which was removed in 2007.

There is *NOTHING* architecture-specific about SECCOMP except that the syscalls have per-architecture definitions, like every other syscall. It is absurd to have the option in the arch-specific menus.

Move it to the security menu, consolidate the 7 down to one, and remove the embarassingly-ancient references to the /proc interface.

---

 linux.git-davehans/arch/arm/Kconfig        |   15 +--------------
 linux.git-davehans/arch/microblaze/Kconfig |   18 +-----------------
 linux.git-davehans/arch/mips/Kconfig       |   18 +-----------------
 linux.git-davehans/arch/powerpc/Kconfig    |   18 +-----------------
 linux.git-davehans/arch/s390/Kconfig       |   18 +-----------------
 linux.git-davehans/arch/sh/Kconfig         |   17 +----------------
 linux.git-davehans/arch/sparc/Kconfig      |   18 +-----------------
 linux.git-davehans/arch/x86/Kconfig        |   17 +----------------
 linux.git-davehans/security/Kconfig        |   21 ++++++++++++++++++++-
 9 files changed, 28 insertions(+), 132 deletions(-)
arch/arm/Kconfig renamed from arch/arm/Kconfig~consolidate-seccomp-options +1 -14
diff -puN arch/arm/Kconfig~consolidate-seccomp-options arch/arm/Kconfig
@@ -26,6 +26,7 @@ config ARM
 	select HAVE_ARCH_JUMP_LABEL if !XIP_KERNEL
 	select HAVE_ARCH_KGDB
 	select HAVE_ARCH_SECCOMP_FILTER if (AEABI && !OABI_COMPAT)
+	select HAVE_ARCH_SECCOMP
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_BPF_JIT
 	select HAVE_CONTEXT_TRACKING
@@ -1842,20 +1843,6 @@ config UACCESS_WITH_MEMCPY
 	  However, if the CPU data cache is using a write-allocate mode,
 	  this option is unlikely to provide any performance gain.
 
-config SECCOMP
-	bool
-	prompt "Enable seccomp to safely compute untrusted bytecode"
-	---help---
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via prctl(PR_SET_SECCOMP), it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
 config CC_STACKPROTECTOR
 	bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)"
 	help
arch/microblaze/Kconfig renamed from arch/microblaze/Kconfig~consolidate-seccomp-options +1 -17
diff -puN arch/microblaze/Kconfig~consolidate-seccomp-options arch/microblaze/Kconfig
@@ -11,6 +11,7 @@ config MICROBLAZE
 	select ARCH_WANT_OPTIONAL_GPIOLIB
 	select HAVE_OPROFILE
 	select HAVE_ARCH_KGDB
+	select HAVE_ARCH_SECCOMP
 	select HAVE_DMA_ATTRS
 	select HAVE_DMA_API_DEBUG
 	select TRACING_SUPPORT
@@ -106,23 +107,6 @@ config CMDLINE_FORCE
 	  Set this to have arguments from the default kernel command string
 	  override those passed by the boot loader.
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	default y
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 endmenu
 
 menu "Advanced setup"
arch/mips/Kconfig renamed from arch/mips/Kconfig~consolidate-seccomp-options +1 -17
diff -puN arch/mips/Kconfig~consolidate-seccomp-options arch/mips/Kconfig
@@ -10,6 +10,7 @@ config MIPS
 	select PERF_USE_VMALLOC
 	select HAVE_ARCH_KGDB
 	select HAVE_ARCH_TRACEHOOK
+	select HAVE_ARCH_SECCOMP
 	select ARCH_HAVE_CUSTOM_GPIO_H
 	select HAVE_FUNCTION_TRACER
 	select HAVE_FUNCTION_TRACE_MCOUNT_TEST
@@ -2305,23 +2306,6 @@ config PHYSICAL_START
 	  specified in the "crashkernel=YM@XM" command line boot parameter
 	  passed to the panic-ed kernel).
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	default y
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 config CC_STACKPROTECTOR
 	bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)"
 	help
arch/powerpc/Kconfig renamed from arch/powerpc/Kconfig~consolidate-seccomp-options +1 -17
diff -puN arch/powerpc/Kconfig~consolidate-seccomp-options arch/powerpc/Kconfig
@@ -101,6 +101,7 @@ config PPC
 	select HAVE_EFFICIENT_UNALIGNED_ACCESS if !CPU_LITTLE_ENDIAN
 	select HAVE_KPROBES
 	select HAVE_ARCH_KGDB
+	select HAVE_ARCH_SECCOMP
 	select HAVE_KRETPROBES
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_MEMBLOCK
@@ -626,23 +627,6 @@ config ARCH_WANTS_FREEZER_CONTROL
 
 source kernel/power/Kconfig
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	default y
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 endmenu
 
 config ISA_DMA_API
arch/s390/Kconfig renamed from arch/s390/Kconfig~consolidate-seccomp-options +1 -17
diff -puN arch/s390/Kconfig~consolidate-seccomp-options arch/s390/Kconfig
@@ -105,6 +105,7 @@ config S390
 	select HAVE_ALIGNED_STRUCT_PAGE if SLUB
 	select HAVE_ARCH_JUMP_LABEL if !MARCH_G5
 	select HAVE_ARCH_SECCOMP_FILTER
+	select HAVE_ARCH_SECCOMP
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_ARCH_TRANSPARENT_HUGEPAGE if 64BIT
 	select HAVE_BPF_JIT if 64BIT && PACK_STACK
@@ -608,23 +609,6 @@ menu "Executable file formats / Emulatio
 
 source "fs/Kconfig.binfmt"
 
-config SECCOMP
-	def_bool y
-	prompt "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y.
-
 endmenu
 
 menu "Power Management"
arch/sh/Kconfig renamed from arch/sh/Kconfig~consolidate-seccomp-options +1 -16
diff -puN arch/sh/Kconfig~consolidate-seccomp-options arch/sh/Kconfig
@@ -10,6 +10,7 @@ config SUPERH
 	select HAVE_OPROFILE
 	select HAVE_GENERIC_DMA_COHERENT
 	select HAVE_ARCH_TRACEHOOK
+	select HAVE_ARCH_SECCOMP
 	select HAVE_DMA_API_DEBUG
 	select HAVE_DMA_ATTRS
 	select HAVE_PERF_EVENTS
@@ -679,22 +680,6 @@ config PHYSICAL_START
 	  where the fail safe kernel needs to run at a different address
 	  than the panic-ed kernel.
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via prctl, it cannot be disabled and the task is only
-	  allowed to execute a few safe syscalls defined by each seccomp
-	  mode.
-
-	  If unsure, say N.
-
 config CC_STACKPROTECTOR
 	bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)"
 	depends on SUPERH32
arch/sparc/Kconfig renamed from arch/sparc/Kconfig~consolidate-seccomp-options +1 -17
diff -puN arch/sparc/Kconfig~consolidate-seccomp-options arch/sparc/Kconfig
@@ -66,6 +66,7 @@ config SPARC64
 	select HAVE_SYSCALL_TRACEPOINTS
 	select HAVE_CONTEXT_TRACKING
 	select HAVE_DEBUG_KMEMLEAK
+	select HAVE_ARCH_SECCOMP if PROC_FS
 	select RTC_DRV_CMOS
 	select RTC_DRV_BQ4802
 	select RTC_DRV_SUN4V
@@ -222,23 +223,6 @@ config EARLYFB
 	help
 	  Say Y here to enable a faster early framebuffer boot console.
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on SPARC64 && PROC_FS
-	default y
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 config HOTPLUG_CPU
 	bool "Support for hot-pluggable CPUs"
 	depends on SPARC64 && SMP
arch/x86/Kconfig renamed from arch/x86/Kconfig~consolidate-seccomp-options +1 -16
diff -puN arch/x86/Kconfig~consolidate-seccomp-options arch/x86/Kconfig
@@ -101,6 +101,7 @@ config X86
 	select GENERIC_SMP_IDLE_THREAD
 	select ARCH_WANT_IPC_PARSE_VERSION if X86_32
 	select HAVE_ARCH_SECCOMP_FILTER
+	select HAVE_ARCH_SECCOMP
 	select BUILDTIME_EXTABLE_SORT
 	select GENERIC_CMOS_UPDATE
 	select HAVE_ARCH_SOFT_DIRTY
@@ -1601,22 +1602,6 @@ config EFI_STUB
 
 	  See Documentation/efi-stub.txt for more information.
 
-config SECCOMP
-	def_bool y
-	prompt "Enable seccomp to safely compute untrusted bytecode"
-	---help---
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via prctl(PR_SET_SECCOMP), it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 config CC_STACKPROTECTOR
 	bool "Enable -fstack-protector buffer overflow detection"
 	---help---
security/Kconfig renamed from security/Kconfig~consolidate-seccomp-options +20 -1
diff -puN security/Kconfig~consolidate-seccomp-options security/Kconfig
@@ -167,5 +167,24 @@ config DEFAULT_SECURITY
 	default "yama" if DEFAULT_SECURITY_YAMA
 	default "" if DEFAULT_SECURITY_DAC
 
-endmenu
+config HAVE_ARCH_SECCOMP
+	bool
+
+config SECCOMP
+	bool
+	default y
+	prompt "Enable seccomp to safely compute untrusted bytecode"
+	---help---
+	  This kernel feature is useful for number crunching applications
+	  that may need to compute untrusted bytecode during their
+	  execution. By using pipes or other transports made available to
+	  the process as file descriptors supporting the read/write
+	  syscalls, it's possible to isolate those applications in
+	  their own address space using seccomp. Once seccomp is
+	  enabled via prctl(PR_SET_SECCOMP), it cannot be disabled
+	  and the task is only allowed to execute a few safe syscalls
+	  defined by each seccomp mode.
 
+	  If unsure, say Y. Only embedded should say N here.
+
+endmenu

_

#1 lore raw

PATCH 2/3 Kconfig: collapse initrd in to own menu

From: Dave Hansen <[email protected]>

Initrd support takes up 8 lines of space in the general menu, pushing down options like OProfile or the VM event counters. Stick it in its own menu to trim it down.

---

 linux.git-davehans/init/Kconfig |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
init/Kconfig renamed from init/Kconfig~kconfig-initrd-collapse-in-to-menu +1 -1
diff -puN init/Kconfig~kconfig-initrd-collapse-in-to-menu init/Kconfig
@@ -1223,7 +1223,7 @@ config RELAY
 
 	  If unsure, say N.
 
-config BLK_DEV_INITRD
+menuconfig BLK_DEV_INITRD
 	bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
 	depends on BROKEN || !FRV
 	help

_

#2 lore raw

PATCH 3/3 Kconfig: organize memory-related config options

In reply to Dave Hansen

From: Dave Hansen <[email protected]>

This continues in a series of patches to clean up the configuration menus. I believe they've become really hard to navigate and there are some simple things we can do to make things easier to find.

This creates a "Memory Options" menu and moves some things like swap and slab configuration under them. It also moves SLUB_DEBUG to the debugging menu.

After this patch, the menu has the following options:

[ ] Memory placement aware NUMA scheduler [*] Enable VM event counters for /proc/vmstat [ ] Disable heap randomization
[*] Support for paging of anonymous memory (swap)
Choose SLAB allocator (SLUB (Unqueued Allocator))
[*] SLUB per cpu partial cache
[*] SLUB: attempt to use double-cmpxchg operations

---

 linux.git-davehans/init/Kconfig     |  243 ++++++++++++++++++------------------
 linux.git-davehans/mm/Kconfig.debug |   11 +
 2 files changed, 135 insertions(+), 119 deletions(-)
init/Kconfig renamed from init/Kconfig~organize-memory-config-options +124 -119
diff -puN init/Kconfig~organize-memory-config-options init/Kconfig
@@ -208,16 +208,6 @@ config DEFAULT_HOSTNAME
 	  but you may wish to use a different default here to make a minimal
 	  system more usable with less configuration.
 
-config SWAP
-	bool "Support for paging of anonymous memory (swap)"
-	depends on MMU && BLOCK
-	default y
-	help
-	  This option allows you to choose whether you want to have support
-	  for so called swap devices or swap files in your kernel that are
-	  used to provide more virtual memory than the actual RAM present
-	  in your computer.  If unsure say Y.
-
 config SYSVIPC
 	bool "System V IPC"
 	---help---
@@ -760,6 +750,130 @@ endchoice
 
 endmenu # "RCU Subsystem"
 
+menu "Memory Options"
+
+config NUMA_BALANCING
+	bool "Memory placement aware NUMA scheduler"
+	depends on ARCH_SUPPORTS_NUMA_BALANCING
+	depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
+	depends on SMP && NUMA && MIGRATION
+	help
+	  This option adds support for automatic NUMA aware memory/task placement.
+	  The mechanism is quite primitive and is based on migrating memory when
+	  it has references to the node the task is running on.
+
+	  This system will be inactive on UMA systems.
+
+config VM_EVENT_COUNTERS
+	default y
+	bool "Enable VM event counters for /proc/vmstat" if EXPERT
+	help
+	  VM event counters are needed for event counts to be shown.
+	  This option allows the disabling of the VM event counters
+	  on EXPERT systems.  /proc/vmstat will only show page counts
+	  if VM event counters are disabled.
+
+config COMPAT_BRK
+	bool "Disable heap randomization"
+	default y
+	help
+	  Randomizing heap placement makes heap exploits harder, but it
+	  also breaks ancient binaries (including anything libc5 based).
+	  This option changes the bootup default to heap randomization
+	  disabled, and can be overridden at runtime by setting
+	  /proc/sys/kernel/randomize_va_space to 2.
+
+	  On non-ancient distros (post-2000 ones) N is usually a safe choice.
+
+config SWAP
+	bool "Support for paging of anonymous memory (swap)"
+	depends on MMU && BLOCK
+	default y
+	help
+	  This option allows you to choose whether you want to have support
+	  for so called swap devices or swap files in your kernel that are
+	  used to provide more virtual memory than the actual RAM present
+	  in your computer.  If unsure say Y.
+
+choice
+	prompt "Choose SLAB allocator"
+	default SLUB
+	help
+	   This option allows to select a slab allocator.
+
+config SLAB
+	bool "SLAB"
+	help
+	  The regular slab allocator that is established and known to work
+	  well in all environments. It organizes cache hot objects in
+	  per cpu and per node queues.
+
+config SLUB
+	bool "SLUB (Unqueued Allocator)"
+	help
+	   SLUB is a slab allocator that minimizes cache line usage
+	   instead of managing queues of cached objects (SLAB approach).
+	   Per cpu caching is realized using slabs of objects instead
+	   of queues of objects. SLUB can use memory efficiently
+	   and has enhanced diagnostics. SLUB is the default choice for
+	   a slab allocator.
+
+config SLOB
+	depends on EXPERT
+	bool "SLOB (Simple Allocator)"
+	help
+	   SLOB replaces the stock allocator with a drastically simpler
+	   allocator. SLOB is generally more space efficient but
+	   does not perform as well on large systems.
+
+endchoice
+
+config SLUB_CPU_PARTIAL
+	default y
+	depends on SLUB && SMP
+	bool "SLUB per cpu partial cache"
+	help
+	  Per cpu partial caches accellerate objects allocation and freeing
+	  that is local to a processor at the price of more indeterminism
+	  in the latency of the free. On overflow these caches will be cleared
+	  which requires the taking of locks that may cause latency spikes.
+	  Typically one would choose no for a realtime system.
+
+config SLUB_ATTEMPT_CMPXCHG_DOUBLE
+	default y
+	depends on SLUB && HAVE_CMPXCHG_DOUBLE
+	bool "SLUB: attempt to use double-cmpxchg operations"
+	help
+	  Some CPUs support instructions that let you do a large double-word
+	  atomic cmpxchg operation.  This keeps the SLUB fastpath from
+	  needing to disable interrupts.
+
+	  If you are unsure, say y.
+
+config MMAP_ALLOW_UNINITIALIZED
+	bool "Allow mmapped anonymous memory to be uninitialized"
+	depends on EXPERT && !MMU
+	default n
+	help
+	  Normally, and according to the Linux spec, anonymous memory obtained
+	  from mmap() has it's contents cleared before it is passed to
+	  userspace.  Enabling this config option allows you to request that
+	  mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
+	  providing a huge performance boost.  If this option is not enabled,
+	  then the flag will be ignored.
+
+	  This is taken advantage of by uClibc's malloc(), and also by
+	  ELF-FDPIC binfmt's brk and stack allocator.
+
+	  Because of the obvious security issues, this option should only be
+	  enabled on embedded devices where you control what is run in
+	  userspace.  Since that isn't generally a problem on no-MMU systems,
+	  it is normally safe to say Y here.
+
+	  See Documentation/nommu-mmap.txt for more information.
+
+endmenu # "Memory Optionse
+
 config IKCONFIG
 	tristate "Kernel .config support"
 	---help---
@@ -840,18 +954,6 @@ config NUMA_BALANCING_DEFAULT_ENABLED
 	  If set, automatic NUMA balancing will be enabled if running on a NUMA
 	  machine.
 
-config NUMA_BALANCING
-	bool "Memory placement aware NUMA scheduler"
-	depends on ARCH_SUPPORTS_NUMA_BALANCING
-	depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
-	depends on SMP && NUMA && MIGRATION
-	help
-	  This option adds support for automatic NUMA aware memory/task placement.
-	  The mechanism is quite primitive and is based on migrating memory when
-	  it has references to the node the task is running on.
-
-	  This system will be inactive on UMA systems.
-
 menuconfig CGROUPS
 	boolean "Control Group support"
 	depends on EVENTFD
@@ -1529,103 +1631,6 @@ config DEBUG_PERF_USE_VMALLOC
 
 endmenu
 
-config VM_EVENT_COUNTERS
-	default y
-	bool "Enable VM event counters for /proc/vmstat" if EXPERT
-	help
-	  VM event counters are needed for event counts to be shown.
-	  This option allows the disabling of the VM event counters
-	  on EXPERT systems.  /proc/vmstat will only show page counts
-	  if VM event counters are disabled.
-
-config SLUB_DEBUG
-	default y
-	bool "Enable SLUB debugging support" if EXPERT
-	depends on SLUB && SYSFS
-	help
-	  SLUB has extensive debug support features. Disabling these can
-	  result in significant savings in code size. This also disables
-	  SLUB sysfs support. /sys/slab will not exist and there will be
-	  no support for cache validation etc.
-
-config COMPAT_BRK
-	bool "Disable heap randomization"
-	default y
-	help
-	  Randomizing heap placement makes heap exploits harder, but it
-	  also breaks ancient binaries (including anything libc5 based).
-	  This option changes the bootup default to heap randomization
-	  disabled, and can be overridden at runtime by setting
-	  /proc/sys/kernel/randomize_va_space to 2.
-
-	  On non-ancient distros (post-2000 ones) N is usually a safe choice.
-
-choice
-	prompt "Choose SLAB allocator"
-	default SLUB
-	help
-	   This option allows to select a slab allocator.
-
-config SLAB
-	bool "SLAB"
-	help
-	  The regular slab allocator that is established and known to work
-	  well in all environments. It organizes cache hot objects in
-	  per cpu and per node queues.
-
-config SLUB
-	bool "SLUB (Unqueued Allocator)"
-	help
-	   SLUB is a slab allocator that minimizes cache line usage
-	   instead of managing queues of cached objects (SLAB approach).
-	   Per cpu caching is realized using slabs of objects instead
-	   of queues of objects. SLUB can use memory efficiently
-	   and has enhanced diagnostics. SLUB is the default choice for
-	   a slab allocator.
-
-config SLOB
-	depends on EXPERT
-	bool "SLOB (Simple Allocator)"
-	help
-	   SLOB replaces the stock allocator with a drastically simpler
-	   allocator. SLOB is generally more space efficient but
-	   does not perform as well on large systems.
-
-endchoice
-
-config SLUB_CPU_PARTIAL
-	default y
-	depends on SLUB && SMP
-	bool "SLUB per cpu partial cache"
-	help
-	  Per cpu partial caches accellerate objects allocation and freeing
-	  that is local to a processor at the price of more indeterminism
-	  in the latency of the free. On overflow these caches will be cleared
-	  which requires the taking of locks that may cause latency spikes.
-	  Typically one would choose no for a realtime system.
-
-config MMAP_ALLOW_UNINITIALIZED
-	bool "Allow mmapped anonymous memory to be uninitialized"
-	depends on EXPERT && !MMU
-	default n
-	help
-	  Normally, and according to the Linux spec, anonymous memory obtained
-	  from mmap() has it's contents cleared before it is passed to
-	  userspace.  Enabling this config option allows you to request that
-	  mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
-	  providing a huge performance boost.  If this option is not enabled,
-	  then the flag will be ignored.
-
-	  This is taken advantage of by uClibc's malloc(), and also by
-	  ELF-FDPIC binfmt's brk and stack allocator.
-
-	  Because of the obvious security issues, this option should only be
-	  enabled on embedded devices where you control what is run in
-	  userspace.  Since that isn't generally a problem on no-MMU systems,
-	  it is normally safe to say Y here.
-
-	  See Documentation/nommu-mmap.txt for more information.
-
 config PROFILING
 	bool "Profiling support"
 	help
mm/Kconfig.debug renamed from mm/Kconfig.debug~organize-memory-config-options +11 -0
diff -puN mm/Kconfig.debug~organize-memory-config-options mm/Kconfig.debug
@@ -27,3 +27,14 @@ config PAGE_POISONING
 config PAGE_GUARD
 	bool
 	select WANT_PAGE_DEBUG_FLAGS
+
+config SLUB_DEBUG
+	default y
+	bool "Enable SLUB debugging support" if EXPERT
+	depends on SLUB && SYSFS
+	help
+	  SLUB has extensive debug support features. Disabling these can
+	  result in significant savings in code size. This also disables
+	  SLUB sysfs support. /sys/slab will not exist and there will be
+	  no support for cache validation etc.
+

_

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

#5 lore raw
In reply to Dave Hansen
Quoted 310 lines · Dave Hansen

From: Dave Hansen <[email protected]>

There are 7 architecures with "config SECCOMP". They all have virtually the same help text except for those referencing the /proc interface which was removed in 2007.

There is *NOTHING* architecture-specific about SECCOMP except that the syscalls have per-architecture definitions, like every other syscall. It is absurd to have the option in the arch-specific menus.

Move it to the security menu, consolidate the 7 down to one, and remove the embarassingly-ancient references to the /proc interface.

---

 linux.git-davehans/arch/arm/Kconfig        |   15 +--------------
 linux.git-davehans/arch/microblaze/Kconfig |   18 +-----------------
 linux.git-davehans/arch/mips/Kconfig       |   18 +-----------------
 linux.git-davehans/arch/powerpc/Kconfig    |   18 +-----------------
 linux.git-davehans/arch/s390/Kconfig       |   18 +-----------------
 linux.git-davehans/arch/sh/Kconfig         |   17 +----------------
 linux.git-davehans/arch/sparc/Kconfig      |   18 +-----------------
 linux.git-davehans/arch/x86/Kconfig        |   17 +----------------
 linux.git-davehans/security/Kconfig        |   21 ++++++++++++++++++++-
 9 files changed, 28 insertions(+), 132 deletions(-)

diff -puN arch/arm/Kconfig~consolidate-seccomp-options arch/arm/Kconfig

arch/arm/Kconfig renamed from arch/arm/Kconfig~consolidate-seccomp-options +1 -14
@@ -26,6 +26,7 @@ config ARM
 	select HAVE_ARCH_JUMP_LABEL if !XIP_KERNEL
 	select HAVE_ARCH_KGDB
 	select HAVE_ARCH_SECCOMP_FILTER if (AEABI && !OABI_COMPAT)
+	select HAVE_ARCH_SECCOMP
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_BPF_JIT
 	select HAVE_CONTEXT_TRACKING
@@ -1842,20 +1843,6 @@ config UACCESS_WITH_MEMCPY
 	  However, if the CPU data cache is using a write-allocate mode,
 	  this option is unlikely to provide any performance gain.
 
-config SECCOMP
-	bool
-	prompt "Enable seccomp to safely compute untrusted bytecode"
-	---help---
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via prctl(PR_SET_SECCOMP), it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
 config CC_STACKPROTECTOR
 	bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)"
 	help

diff -puN arch/microblaze/Kconfig~consolidate-seccomp-options arch/microblaze/Kconfig

arch/microblaze/Kconfig renamed from arch/microblaze/Kconfig~consolidate-seccomp-options +1 -17
@@ -11,6 +11,7 @@ config MICROBLAZE
 	select ARCH_WANT_OPTIONAL_GPIOLIB
 	select HAVE_OPROFILE
 	select HAVE_ARCH_KGDB
+	select HAVE_ARCH_SECCOMP
 	select HAVE_DMA_ATTRS
 	select HAVE_DMA_API_DEBUG
 	select TRACING_SUPPORT
@@ -106,23 +107,6 @@ config CMDLINE_FORCE
 	  Set this to have arguments from the default kernel command string
 	  override those passed by the boot loader.
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	default y
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 endmenu
 
 menu "Advanced setup"

diff -puN arch/mips/Kconfig~consolidate-seccomp-options arch/mips/Kconfig

arch/mips/Kconfig renamed from arch/mips/Kconfig~consolidate-seccomp-options +1 -17
@@ -10,6 +10,7 @@ config MIPS
 	select PERF_USE_VMALLOC
 	select HAVE_ARCH_KGDB
 	select HAVE_ARCH_TRACEHOOK
+	select HAVE_ARCH_SECCOMP
 	select ARCH_HAVE_CUSTOM_GPIO_H
 	select HAVE_FUNCTION_TRACER
 	select HAVE_FUNCTION_TRACE_MCOUNT_TEST
@@ -2305,23 +2306,6 @@ config PHYSICAL_START
 	  specified in the "crashkernel=YM@XM" command line boot parameter
 	  passed to the panic-ed kernel).
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	default y
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 config CC_STACKPROTECTOR
 	bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)"
 	help

diff -puN arch/powerpc/Kconfig~consolidate-seccomp-options arch/powerpc/Kconfig

arch/powerpc/Kconfig renamed from arch/powerpc/Kconfig~consolidate-seccomp-options +1 -17
@@ -101,6 +101,7 @@ config PPC
 	select HAVE_EFFICIENT_UNALIGNED_ACCESS if !CPU_LITTLE_ENDIAN
 	select HAVE_KPROBES
 	select HAVE_ARCH_KGDB
+	select HAVE_ARCH_SECCOMP
 	select HAVE_KRETPROBES
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_MEMBLOCK
@@ -626,23 +627,6 @@ config ARCH_WANTS_FREEZER_CONTROL
 
 source kernel/power/Kconfig
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	default y
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 endmenu
 
 config ISA_DMA_API

diff -puN arch/s390/Kconfig~consolidate-seccomp-options arch/s390/Kconfig

arch/s390/Kconfig renamed from arch/s390/Kconfig~consolidate-seccomp-options +1 -17
@@ -105,6 +105,7 @@ config S390
 	select HAVE_ALIGNED_STRUCT_PAGE if SLUB
 	select HAVE_ARCH_JUMP_LABEL if !MARCH_G5
 	select HAVE_ARCH_SECCOMP_FILTER
+	select HAVE_ARCH_SECCOMP
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_ARCH_TRANSPARENT_HUGEPAGE if 64BIT
 	select HAVE_BPF_JIT if 64BIT && PACK_STACK
@@ -608,23 +609,6 @@ menu "Executable file formats / Emulatio
 
 source "fs/Kconfig.binfmt"
 
-config SECCOMP
-	def_bool y
-	prompt "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y.
-
 endmenu
 
 menu "Power Management"

diff -puN arch/sh/Kconfig~consolidate-seccomp-options arch/sh/Kconfig

arch/sh/Kconfig renamed from arch/sh/Kconfig~consolidate-seccomp-options +1 -16
@@ -10,6 +10,7 @@ config SUPERH
 	select HAVE_OPROFILE
 	select HAVE_GENERIC_DMA_COHERENT
 	select HAVE_ARCH_TRACEHOOK
+	select HAVE_ARCH_SECCOMP
 	select HAVE_DMA_API_DEBUG
 	select HAVE_DMA_ATTRS
 	select HAVE_PERF_EVENTS
@@ -679,22 +680,6 @@ config PHYSICAL_START
 	  where the fail safe kernel needs to run at a different address
 	  than the panic-ed kernel.
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via prctl, it cannot be disabled and the task is only
-	  allowed to execute a few safe syscalls defined by each seccomp
-	  mode.
-
-	  If unsure, say N.
-
 config CC_STACKPROTECTOR
 	bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)"
 	depends on SUPERH32

diff -puN arch/sparc/Kconfig~consolidate-seccomp-options arch/sparc/Kconfig

arch/sparc/Kconfig renamed from arch/sparc/Kconfig~consolidate-seccomp-options +1 -17
@@ -66,6 +66,7 @@ config SPARC64
 	select HAVE_SYSCALL_TRACEPOINTS
 	select HAVE_CONTEXT_TRACKING
 	select HAVE_DEBUG_KMEMLEAK
+	select HAVE_ARCH_SECCOMP if PROC_FS
 	select RTC_DRV_CMOS
 	select RTC_DRV_BQ4802
 	select RTC_DRV_SUN4V
@@ -222,23 +223,6 @@ config EARLYFB
 	help
 	  Say Y here to enable a faster early framebuffer boot console.
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on SPARC64 && PROC_FS
-	default y
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 config HOTPLUG_CPU
 	bool "Support for hot-pluggable CPUs"
 	depends on SPARC64 && SMP

diff -puN arch/x86/Kconfig~consolidate-seccomp-options arch/x86/Kconfig

arch/x86/Kconfig renamed from arch/x86/Kconfig~consolidate-seccomp-options +1 -16
@@ -101,6 +101,7 @@ config X86
 	select GENERIC_SMP_IDLE_THREAD
 	select ARCH_WANT_IPC_PARSE_VERSION if X86_32
 	select HAVE_ARCH_SECCOMP_FILTER
+	select HAVE_ARCH_SECCOMP
 	select BUILDTIME_EXTABLE_SORT
 	select GENERIC_CMOS_UPDATE
 	select HAVE_ARCH_SOFT_DIRTY
@@ -1601,22 +1602,6 @@ config EFI_STUB
 
 	  See Documentation/efi-stub.txt for more information.
 
-config SECCOMP
-	def_bool y
-	prompt "Enable seccomp to safely compute untrusted bytecode"
-	---help---
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via prctl(PR_SET_SECCOMP), it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 config CC_STACKPROTECTOR
 	bool "Enable -fstack-protector buffer overflow detection"
 	---help---

diff -puN security/Kconfig~consolidate-seccomp-options security/Kconfig

security/Kconfig config DEFAULT_SECURITY · line 174

Quoted 9 lines
security/Kconfig renamed from security/Kconfig~consolidate-seccomp-options +5 -1
@@ -167,5 +167,24 @@ config DEFAULT_SECURITY
 	default "yama" if DEFAULT_SECURITY_YAMA
 	default "" if DEFAULT_SECURITY_DAC
 
-endmenu
+config HAVE_ARCH_SECCOMP
+	bool
+
+config SECCOMP
+	bool

Hi Dave,

I haven't looked at the other 'CONFIG_HAVE' options, but shouldn't 'HAVE_ARCH_SECCOMP' be dependent on 'SECCOMP'?

Mimi

Quoted 22 lines
security/Kconfig renamed from security/Kconfig~consolidate-seccomp-options +15 -0
+	default y
+	prompt "Enable seccomp to safely compute untrusted bytecode"
+	---help---
+	  This kernel feature is useful for number crunching applications
+	  that may need to compute untrusted bytecode during their
+	  execution. By using pipes or other transports made available to
+	  the process as file descriptors supporting the read/write
+	  syscalls, it's possible to isolate those applications in
+	  their own address space using seccomp. Once seccomp is
+	  enabled via prctl(PR_SET_SECCOMP), it cannot be disabled
+	  and the task is only allowed to execute a few safe syscalls
+	  defined by each seccomp mode.
 
+	  If unsure, say Y. Only embedded should say N here.
+
+endmenu

_
--
To unsubscribe from this list: send the line "unsubscribe linux-security-module" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

#6 lore raw
Quoted 8 lines · Mimi Zohar
Quoted 5 lines

Review

Quoted 5 lines
(quoted diff) +5 -0
+config HAVE_ARCH_SECCOMP
+	bool
+
+config SECCOMP
+	bool

I haven't looked at the other 'CONFIG_HAVE' options, but shouldn't 'HAVE_ARCH_SECCOMP' be dependent on 'SECCOMP'?

Ahh, you're backwards, but right. :)

I forgot to make sure that SECCOMP depends on the architecture option being set. Fixed patch is attached.

From: Dave Hansen <[email protected]>

There are 7 architecures with "config SECCOMP". They all have virtually the same help text except for those referencing the /proc interface which was removed in 2007.

There is *NOTHING* architecture-specific about SECCOMP except that the syscalls have per-architecture definitions, like every other syscall. It is absurd to have the option in the arch-specific menus.

Move it to the security menu, consolidate the 7 down to one, and remove the embarassingly-ancient references to the /proc interface.

---

 linux.git-davehans/arch/arm/Kconfig        |   15 +--------------
 linux.git-davehans/arch/microblaze/Kconfig |   18 +-----------------
 linux.git-davehans/arch/mips/Kconfig       |   18 +-----------------
 linux.git-davehans/arch/powerpc/Kconfig    |   18 +-----------------
 linux.git-davehans/arch/s390/Kconfig       |   18 +-----------------
 linux.git-davehans/arch/sh/Kconfig         |   17 +----------------
 linux.git-davehans/arch/sparc/Kconfig      |   18 +-----------------
 linux.git-davehans/arch/x86/Kconfig        |   17 +----------------
 linux.git-davehans/security/Kconfig        |   21 ++++++++++++++++++++-
 9 files changed, 28 insertions(+), 132 deletions(-)
arch/arm/Kconfig renamed from arch/arm/Kconfig~consolidate-seccomp-options +1 -14
diff -puN arch/arm/Kconfig~consolidate-seccomp-options arch/arm/Kconfig
@@ -26,6 +26,7 @@ config ARM
 	select HAVE_ARCH_JUMP_LABEL if !XIP_KERNEL
 	select HAVE_ARCH_KGDB
 	select HAVE_ARCH_SECCOMP_FILTER if (AEABI && !OABI_COMPAT)
+	select HAVE_ARCH_SECCOMP
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_BPF_JIT
 	select HAVE_CONTEXT_TRACKING
@@ -1842,20 +1843,6 @@ config UACCESS_WITH_MEMCPY
 	  However, if the CPU data cache is using a write-allocate mode,
 	  this option is unlikely to provide any performance gain.
 
-config SECCOMP
-	bool
-	prompt "Enable seccomp to safely compute untrusted bytecode"
-	---help---
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via prctl(PR_SET_SECCOMP), it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
 config CC_STACKPROTECTOR
 	bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)"
 	help
arch/microblaze/Kconfig renamed from arch/microblaze/Kconfig~consolidate-seccomp-options +1 -17
diff -puN arch/microblaze/Kconfig~consolidate-seccomp-options arch/microblaze/Kconfig
@@ -11,6 +11,7 @@ config MICROBLAZE
 	select ARCH_WANT_OPTIONAL_GPIOLIB
 	select HAVE_OPROFILE
 	select HAVE_ARCH_KGDB
+	select HAVE_ARCH_SECCOMP
 	select HAVE_DMA_ATTRS
 	select HAVE_DMA_API_DEBUG
 	select TRACING_SUPPORT
@@ -106,23 +107,6 @@ config CMDLINE_FORCE
 	  Set this to have arguments from the default kernel command string
 	  override those passed by the boot loader.
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	default y
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 endmenu
 
 menu "Advanced setup"
arch/mips/Kconfig renamed from arch/mips/Kconfig~consolidate-seccomp-options +1 -17
diff -puN arch/mips/Kconfig~consolidate-seccomp-options arch/mips/Kconfig
@@ -10,6 +10,7 @@ config MIPS
 	select PERF_USE_VMALLOC
 	select HAVE_ARCH_KGDB
 	select HAVE_ARCH_TRACEHOOK
+	select HAVE_ARCH_SECCOMP
 	select ARCH_HAVE_CUSTOM_GPIO_H
 	select HAVE_FUNCTION_TRACER
 	select HAVE_FUNCTION_TRACE_MCOUNT_TEST
@@ -2305,23 +2306,6 @@ config PHYSICAL_START
 	  specified in the "crashkernel=YM@XM" command line boot parameter
 	  passed to the panic-ed kernel).
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	default y
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 config CC_STACKPROTECTOR
 	bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)"
 	help
arch/powerpc/Kconfig renamed from arch/powerpc/Kconfig~consolidate-seccomp-options +1 -17
diff -puN arch/powerpc/Kconfig~consolidate-seccomp-options arch/powerpc/Kconfig
@@ -101,6 +101,7 @@ config PPC
 	select HAVE_EFFICIENT_UNALIGNED_ACCESS if !CPU_LITTLE_ENDIAN
 	select HAVE_KPROBES
 	select HAVE_ARCH_KGDB
+	select HAVE_ARCH_SECCOMP
 	select HAVE_KRETPROBES
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_MEMBLOCK
@@ -626,23 +627,6 @@ config ARCH_WANTS_FREEZER_CONTROL
 
 source kernel/power/Kconfig
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	default y
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 endmenu
 
 config ISA_DMA_API
arch/s390/Kconfig renamed from arch/s390/Kconfig~consolidate-seccomp-options +1 -17
diff -puN arch/s390/Kconfig~consolidate-seccomp-options arch/s390/Kconfig
@@ -105,6 +105,7 @@ config S390
 	select HAVE_ALIGNED_STRUCT_PAGE if SLUB
 	select HAVE_ARCH_JUMP_LABEL if !MARCH_G5
 	select HAVE_ARCH_SECCOMP_FILTER
+	select HAVE_ARCH_SECCOMP
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_ARCH_TRANSPARENT_HUGEPAGE if 64BIT
 	select HAVE_BPF_JIT if 64BIT && PACK_STACK
@@ -608,23 +609,6 @@ menu "Executable file formats / Emulatio
 
 source "fs/Kconfig.binfmt"
 
-config SECCOMP
-	def_bool y
-	prompt "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y.
-
 endmenu
 
 menu "Power Management"
arch/sh/Kconfig renamed from arch/sh/Kconfig~consolidate-seccomp-options +1 -16
diff -puN arch/sh/Kconfig~consolidate-seccomp-options arch/sh/Kconfig
@@ -10,6 +10,7 @@ config SUPERH
 	select HAVE_OPROFILE
 	select HAVE_GENERIC_DMA_COHERENT
 	select HAVE_ARCH_TRACEHOOK
+	select HAVE_ARCH_SECCOMP
 	select HAVE_DMA_API_DEBUG
 	select HAVE_DMA_ATTRS
 	select HAVE_PERF_EVENTS
@@ -679,22 +680,6 @@ config PHYSICAL_START
 	  where the fail safe kernel needs to run at a different address
 	  than the panic-ed kernel.
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on PROC_FS
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via prctl, it cannot be disabled and the task is only
-	  allowed to execute a few safe syscalls defined by each seccomp
-	  mode.
-
-	  If unsure, say N.
-
 config CC_STACKPROTECTOR
 	bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)"
 	depends on SUPERH32
arch/sparc/Kconfig renamed from arch/sparc/Kconfig~consolidate-seccomp-options +1 -17
diff -puN arch/sparc/Kconfig~consolidate-seccomp-options arch/sparc/Kconfig
@@ -66,6 +66,7 @@ config SPARC64
 	select HAVE_SYSCALL_TRACEPOINTS
 	select HAVE_CONTEXT_TRACKING
 	select HAVE_DEBUG_KMEMLEAK
+	select HAVE_ARCH_SECCOMP if PROC_FS
 	select RTC_DRV_CMOS
 	select RTC_DRV_BQ4802
 	select RTC_DRV_SUN4V
@@ -222,23 +223,6 @@ config EARLYFB
 	help
 	  Say Y here to enable a faster early framebuffer boot console.
 
-config SECCOMP
-	bool "Enable seccomp to safely compute untrusted bytecode"
-	depends on SPARC64 && PROC_FS
-	default y
-	help
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via /proc/<pid>/seccomp, it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 config HOTPLUG_CPU
 	bool "Support for hot-pluggable CPUs"
 	depends on SPARC64 && SMP
arch/x86/Kconfig renamed from arch/x86/Kconfig~consolidate-seccomp-options +1 -16
diff -puN arch/x86/Kconfig~consolidate-seccomp-options arch/x86/Kconfig
@@ -101,6 +101,7 @@ config X86
 	select GENERIC_SMP_IDLE_THREAD
 	select ARCH_WANT_IPC_PARSE_VERSION if X86_32
 	select HAVE_ARCH_SECCOMP_FILTER
+	select HAVE_ARCH_SECCOMP
 	select BUILDTIME_EXTABLE_SORT
 	select GENERIC_CMOS_UPDATE
 	select HAVE_ARCH_SOFT_DIRTY
@@ -1601,22 +1602,6 @@ config EFI_STUB
 
 	  See Documentation/efi-stub.txt for more information.
 
-config SECCOMP
-	def_bool y
-	prompt "Enable seccomp to safely compute untrusted bytecode"
-	---help---
-	  This kernel feature is useful for number crunching applications
-	  that may need to compute untrusted bytecode during their
-	  execution. By using pipes or other transports made available to
-	  the process as file descriptors supporting the read/write
-	  syscalls, it's possible to isolate those applications in
-	  their own address space using seccomp. Once seccomp is
-	  enabled via prctl(PR_SET_SECCOMP), it cannot be disabled
-	  and the task is only allowed to execute a few safe syscalls
-	  defined by each seccomp mode.
-
-	  If unsure, say Y. Only embedded should say N here.
-
 config CC_STACKPROTECTOR
 	bool "Enable -fstack-protector buffer overflow detection"
 	---help---
security/Kconfig renamed from security/Kconfig~consolidate-seccomp-options +20 -1
diff -puN security/Kconfig~consolidate-seccomp-options security/Kconfig
@@ -167,5 +167,24 @@ config DEFAULT_SECURITY
 	default "yama" if DEFAULT_SECURITY_YAMA
 	default "" if DEFAULT_SECURITY_DAC
 
-endmenu
+config HAVE_ARCH_SECCOMP
+	bool
+
+config SECCOMP
+	def_bool y
+	depends on HAVE_ARCH_SECCOMP
+	prompt "Enable seccomp to safely compute untrusted bytecode"
+	---help---
+	  This kernel feature is useful for number crunching applications
+	  that may need to compute untrusted bytecode during their
+	  execution. By using pipes or other transports made available to
+	  the process as file descriptors supporting the read/write
+	  syscalls, it's possible to isolate those applications in
+	  their own address space using seccomp. Once seccomp is
+	  enabled via prctl(PR_SET_SECCOMP), it cannot be disabled
+	  and the task is only allowed to execute a few safe syscalls
+	  defined by each seccomp mode.
 
+	  If unsure, say Y. Only embedded should say N here.
+
+endmenu

_

#7 lore raw
In reply to Dave Hansen

Hi Dave,

Quoted 11 lines · On Thu, 02 Jan 2014 12:20:14 -0800 Dave Hansen <[email protected]> wrote:

From: Dave Hansen <[email protected]>

There are 7 architecures with "config SECCOMP". They all have virtually the same help text except for those referencing the /proc interface which was removed in 2007.

There is *NOTHING* architecture-specific about SECCOMP except that the syscalls have per-architecture definitions, like every other syscall. It is absurd to have the option in the arch-specific menus.

You seem to have (mostly) lost the dependency some of the architecture versions of config SECCOMP had on PROC_FS ...

Signature · 2 lines
Cheers,
Stephen Rothwell <[email protected]>
#8 lore raw
Quoted 9 lines · Stephen Rothwell
Quoted 5 lines · On Thu, 02 Jan 2014 12:20:14 -0800 Dave Hansen <[email protected]> wrote:

From: Dave Hansen <[email protected]>
There is *NOTHING* architecture-specific about SECCOMP except that the syscalls have per-architecture definitions, like every other syscall. It is absurd to have the option in the arch-specific menus.

You seem to have (mostly) lost the dependency some of the architecture versions of config SECCOMP had on PROC_FS ...

I _believe_ the /proc interface has gone away. I can't find any reference to /proc/<pid>/seccomp in any of the code. Is there some /proc dependency that I'm missing outside of the removed /proc/<pid>/seccomp interface?

#9 lore raw

Hi Dave,

Quoted 9 lines · On Sat, 04 Jan 2014 11:04:20 -0800 Dave Hansen <[email protected]> wrote:
Quoted 7 lines · Stephen Rothwell
Quoted 5 lines · On Thu, 02 Jan 2014 12:20:14 -0800 Dave Hansen <[email protected]> wrote:

From: Dave Hansen <[email protected]>
There is *NOTHING* architecture-specific about SECCOMP except that the syscalls have per-architecture definitions, like every other syscall. It is absurd to have the option in the arch-specific menus.

Quoted 7 lines
Quoted 2 lines

You seem to have (mostly) lost the dependency some of the architecture versions of config SECCOMP had on PROC_FS ...

I _believe_ the /proc interface has gone away. I can't find any reference to /proc/<pid>/seccomp in any of the code. Is there some /proc dependency that I'm missing outside of the removed /proc/<pid>/seccomp interface?

I don't know, but if it has gone then it is worth mentioning in the commit message ... and you did preserve the dependency in the sparc64 case.

Signature · 2 lines
Cheers,
Stephen Rothwell                    [email protected]
#10 lore raw
Quoted 7 lines · Stephen Rothwell
Quoted 4 lines

I _believe_ the /proc interface has gone away. I can't find any reference to /proc/<pid>/seccomp in any of the code. Is there some /proc dependency that I'm missing outside of the removed /proc/<pid>/seccomp interface?

I don't know, but if it has gone then it is worth mentioning in the commit message

Maybe like this? ;)

Quoted 3 lines

There are 7 architecures with "config SECCOMP". They all have virtually the same help text except for those referencing the /proc interface which was removed in 2007.

I guess I could call it out a _bit_ more clearly that I removed the dependency because the code is gone.

Quoted 1 line

... and you did preserve the dependency in the sparc64 case.

Ahh, I did. I'll remove it. Thanks!

#11 lore raw

Hi Dave,

Quoted 6 lines · On Sat, 04 Jan 2014 16:41:31 -0800 Dave Hansen <[email protected]> wrote:
Quoted 3 lines · Stephen Rothwell

I don't know, but if it has gone then it is worth mentioning in the commit message

Quoted 8 lines

Maybe like this? ;)

Quoted 3 lines

There are 7 architecures with "config SECCOMP". They all have virtually the same help text except for those referencing the /proc interface which was removed in 2007.

I guess I could call it out a _bit_ more clearly that I removed the dependency because the code is gone.

Ah, yes, I must have been half asleep :-)

Quoted 3 lines
Quoted 1 line

... and you did preserve the dependency in the sparc64 case.

Ahh, I did. I'll remove it. Thanks!

And that probably confused me as well.

Signature · 2 lines
Cheers,
Stephen Rothwell                    [email protected]
#12 lore raw
In reply to Dave Hansen
Quoted 19 lines · Dave Hansen

From: Dave Hansen <[email protected]>

---

 linux.git-davehans/arch/arm/Kconfig        |   15 +--------------
 linux.git-davehans/arch/microblaze/Kconfig |   18 +-----------------
 linux.git-davehans/arch/mips/Kconfig       |   18 +-----------------
 linux.git-davehans/arch/powerpc/Kconfig    |   18 +-----------------
 linux.git-davehans/arch/s390/Kconfig       |   18 +-----------------
 linux.git-davehans/arch/sh/Kconfig         |   17 +----------------
 linux.git-davehans/arch/sparc/Kconfig      |   18 +-----------------
 linux.git-davehans/arch/x86/Kconfig        |   17 +----------------
 linux.git-davehans/security/Kconfig        |   21 ++++++++++++++++++++-
 9 files changed, 28 insertions(+), 132 deletions(-)

diff -puN security/Kconfig~consolidate-seccomp-options security/Kconfig

security/Kconfig config DEFAULT_SECURITY · line 175

Quoted 10 lines
security/Kconfig renamed from security/Kconfig~consolidate-seccomp-options +6 -1
@@ -167,5 +167,24 @@ config DEFAULT_SECURITY
 	default "yama" if DEFAULT_SECURITY_YAMA
 	default "" if DEFAULT_SECURITY_DAC
 
-endmenu
+config HAVE_ARCH_SECCOMP
+	bool
+
+config SECCOMP
+	bool
+	default y

Prefer
def_bool y

Quoted 15 lines
security/Kconfig renamed from security/Kconfig~consolidate-seccomp-options +14 -0
+	prompt "Enable seccomp to safely compute untrusted bytecode"
+	---help---
+	  This kernel feature is useful for number crunching applications
+	  that may need to compute untrusted bytecode during their
+	  execution. By using pipes or other transports made available to
+	  the process as file descriptors supporting the read/write
+	  syscalls, it's possible to isolate those applications in
+	  their own address space using seccomp. Once seccomp is
+	  enabled via prctl(PR_SET_SECCOMP), it cannot be disabled
+	  and the task is only allowed to execute a few safe syscalls
+	  defined by each seccomp mode.
 
+	  If unsure, say Y. Only embedded should say N here.
+
+endmenu
Signature · 1 line
~Randy
#13 lore raw
Quoted 5 lines · Randy Dunlap
Quoted 3 lines

security/Kconfig config DEFAULT_SECURITY · line 175

Quoted 3 lines
security/Kconfig renamed from security/Kconfig~consolidate-seccomp-options +3 -0
+config SECCOMP
+	bool
+	default y

Prefer
def_bool y

I've actually got that already in my updated set that I'll send out when the merge window opens.

#3 lore raw

Re: [PATCH 3/3] Kconfig: organize memory-related config options

In reply to Dave Hansen
Quoted 14 lines · Dave Hansen

From: Dave Hansen <[email protected]>

This continues in a series of patches to clean up the configuration menus. I believe they've become really hard to navigate and there are some simple things we can do to make things easier to find.

This creates a "Memory Options" menu and moves some things like swap and slab configuration under them. It also moves SLUB_DEBUG to the debugging menu.

After this patch, the menu has the following options:

Quoted 7 lines

[ ] Memory placement aware NUMA scheduler [*] Enable VM event counters for /proc/vmstat [ ] Disable heap randomization
[*] Support for paging of anonymous memory (swap)
Choose SLAB allocator (SLUB (Unqueued Allocator))
[*] SLUB per cpu partial cache
[*] SLUB: attempt to use double-cmpxchg operations

Is there any reason to keep them in init/Kconfig rather than mm/Kconfig? It would sound like a logical place to have them all, no?

Quoted 307 lines

---

 linux.git-davehans/init/Kconfig     |  243 ++++++++++++++++++------------------
 linux.git-davehans/mm/Kconfig.debug |   11 +
 2 files changed, 135 insertions(+), 119 deletions(-)

diff -puN init/Kconfig~organize-memory-config-options init/Kconfig

init/Kconfig renamed from init/Kconfig~organize-memory-config-options +124 -119
@@ -208,16 +208,6 @@ config DEFAULT_HOSTNAME
 	  but you may wish to use a different default here to make a minimal
 	  system more usable with less configuration.
 
-config SWAP
-	bool "Support for paging of anonymous memory (swap)"
-	depends on MMU && BLOCK
-	default y
-	help
-	  This option allows you to choose whether you want to have support
-	  for so called swap devices or swap files in your kernel that are
-	  used to provide more virtual memory than the actual RAM present
-	  in your computer.  If unsure say Y.
-
 config SYSVIPC
 	bool "System V IPC"
 	---help---
@@ -760,6 +750,130 @@ endchoice
 
 endmenu # "RCU Subsystem"
 
+menu "Memory Options"
+
+config NUMA_BALANCING
+	bool "Memory placement aware NUMA scheduler"
+	depends on ARCH_SUPPORTS_NUMA_BALANCING
+	depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
+	depends on SMP && NUMA && MIGRATION
+	help
+	  This option adds support for automatic NUMA aware memory/task placement.
+	  The mechanism is quite primitive and is based on migrating memory when
+	  it has references to the node the task is running on.
+
+	  This system will be inactive on UMA systems.
+
+config VM_EVENT_COUNTERS
+	default y
+	bool "Enable VM event counters for /proc/vmstat" if EXPERT
+	help
+	  VM event counters are needed for event counts to be shown.
+	  This option allows the disabling of the VM event counters
+	  on EXPERT systems.  /proc/vmstat will only show page counts
+	  if VM event counters are disabled.
+
+config COMPAT_BRK
+	bool "Disable heap randomization"
+	default y
+	help
+	  Randomizing heap placement makes heap exploits harder, but it
+	  also breaks ancient binaries (including anything libc5 based).
+	  This option changes the bootup default to heap randomization
+	  disabled, and can be overridden at runtime by setting
+	  /proc/sys/kernel/randomize_va_space to 2.
+
+	  On non-ancient distros (post-2000 ones) N is usually a safe choice.
+
+config SWAP
+	bool "Support for paging of anonymous memory (swap)"
+	depends on MMU && BLOCK
+	default y
+	help
+	  This option allows you to choose whether you want to have support
+	  for so called swap devices or swap files in your kernel that are
+	  used to provide more virtual memory than the actual RAM present
+	  in your computer.  If unsure say Y.
+
+choice
+	prompt "Choose SLAB allocator"
+	default SLUB
+	help
+	   This option allows to select a slab allocator.
+
+config SLAB
+	bool "SLAB"
+	help
+	  The regular slab allocator that is established and known to work
+	  well in all environments. It organizes cache hot objects in
+	  per cpu and per node queues.
+
+config SLUB
+	bool "SLUB (Unqueued Allocator)"
+	help
+	   SLUB is a slab allocator that minimizes cache line usage
+	   instead of managing queues of cached objects (SLAB approach).
+	   Per cpu caching is realized using slabs of objects instead
+	   of queues of objects. SLUB can use memory efficiently
+	   and has enhanced diagnostics. SLUB is the default choice for
+	   a slab allocator.
+
+config SLOB
+	depends on EXPERT
+	bool "SLOB (Simple Allocator)"
+	help
+	   SLOB replaces the stock allocator with a drastically simpler
+	   allocator. SLOB is generally more space efficient but
+	   does not perform as well on large systems.
+
+endchoice
+
+config SLUB_CPU_PARTIAL
+	default y
+	depends on SLUB && SMP
+	bool "SLUB per cpu partial cache"
+	help
+	  Per cpu partial caches accellerate objects allocation and freeing
+	  that is local to a processor at the price of more indeterminism
+	  in the latency of the free. On overflow these caches will be cleared
+	  which requires the taking of locks that may cause latency spikes.
+	  Typically one would choose no for a realtime system.
+
+config SLUB_ATTEMPT_CMPXCHG_DOUBLE
+	default y
+	depends on SLUB && HAVE_CMPXCHG_DOUBLE
+	bool "SLUB: attempt to use double-cmpxchg operations"
+	help
+	  Some CPUs support instructions that let you do a large double-word
+	  atomic cmpxchg operation.  This keeps the SLUB fastpath from
+	  needing to disable interrupts.
+
+	  If you are unsure, say y.
+
+config MMAP_ALLOW_UNINITIALIZED
+	bool "Allow mmapped anonymous memory to be uninitialized"
+	depends on EXPERT && !MMU
+	default n
+	help
+	  Normally, and according to the Linux spec, anonymous memory obtained
+	  from mmap() has it's contents cleared before it is passed to
+	  userspace.  Enabling this config option allows you to request that
+	  mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
+	  providing a huge performance boost.  If this option is not enabled,
+	  then the flag will be ignored.
+
+	  This is taken advantage of by uClibc's malloc(), and also by
+	  ELF-FDPIC binfmt's brk and stack allocator.
+
+	  Because of the obvious security issues, this option should only be
+	  enabled on embedded devices where you control what is run in
+	  userspace.  Since that isn't generally a problem on no-MMU systems,
+	  it is normally safe to say Y here.
+
+	  See Documentation/nommu-mmap.txt for more information.
+
+endmenu # "Memory Optionse
+
 config IKCONFIG
 	tristate "Kernel .config support"
 	---help---
@@ -840,18 +954,6 @@ config NUMA_BALANCING_DEFAULT_ENABLED
 	  If set, automatic NUMA balancing will be enabled if running on a NUMA
 	  machine.
 
-config NUMA_BALANCING
-	bool "Memory placement aware NUMA scheduler"
-	depends on ARCH_SUPPORTS_NUMA_BALANCING
-	depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
-	depends on SMP && NUMA && MIGRATION
-	help
-	  This option adds support for automatic NUMA aware memory/task placement.
-	  The mechanism is quite primitive and is based on migrating memory when
-	  it has references to the node the task is running on.
-
-	  This system will be inactive on UMA systems.
-
 menuconfig CGROUPS
 	boolean "Control Group support"
 	depends on EVENTFD
@@ -1529,103 +1631,6 @@ config DEBUG_PERF_USE_VMALLOC
 
 endmenu
 
-config VM_EVENT_COUNTERS
-	default y
-	bool "Enable VM event counters for /proc/vmstat" if EXPERT
-	help
-	  VM event counters are needed for event counts to be shown.
-	  This option allows the disabling of the VM event counters
-	  on EXPERT systems.  /proc/vmstat will only show page counts
-	  if VM event counters are disabled.
-
-config SLUB_DEBUG
-	default y
-	bool "Enable SLUB debugging support" if EXPERT
-	depends on SLUB && SYSFS
-	help
-	  SLUB has extensive debug support features. Disabling these can
-	  result in significant savings in code size. This also disables
-	  SLUB sysfs support. /sys/slab will not exist and there will be
-	  no support for cache validation etc.
-
-config COMPAT_BRK
-	bool "Disable heap randomization"
-	default y
-	help
-	  Randomizing heap placement makes heap exploits harder, but it
-	  also breaks ancient binaries (including anything libc5 based).
-	  This option changes the bootup default to heap randomization
-	  disabled, and can be overridden at runtime by setting
-	  /proc/sys/kernel/randomize_va_space to 2.
-
-	  On non-ancient distros (post-2000 ones) N is usually a safe choice.
-
-choice
-	prompt "Choose SLAB allocator"
-	default SLUB
-	help
-	   This option allows to select a slab allocator.
-
-config SLAB
-	bool "SLAB"
-	help
-	  The regular slab allocator that is established and known to work
-	  well in all environments. It organizes cache hot objects in
-	  per cpu and per node queues.
-
-config SLUB
-	bool "SLUB (Unqueued Allocator)"
-	help
-	   SLUB is a slab allocator that minimizes cache line usage
-	   instead of managing queues of cached objects (SLAB approach).
-	   Per cpu caching is realized using slabs of objects instead
-	   of queues of objects. SLUB can use memory efficiently
-	   and has enhanced diagnostics. SLUB is the default choice for
-	   a slab allocator.
-
-config SLOB
-	depends on EXPERT
-	bool "SLOB (Simple Allocator)"
-	help
-	   SLOB replaces the stock allocator with a drastically simpler
-	   allocator. SLOB is generally more space efficient but
-	   does not perform as well on large systems.
-
-endchoice
-
-config SLUB_CPU_PARTIAL
-	default y
-	depends on SLUB && SMP
-	bool "SLUB per cpu partial cache"
-	help
-	  Per cpu partial caches accellerate objects allocation and freeing
-	  that is local to a processor at the price of more indeterminism
-	  in the latency of the free. On overflow these caches will be cleared
-	  which requires the taking of locks that may cause latency spikes.
-	  Typically one would choose no for a realtime system.
-
-config MMAP_ALLOW_UNINITIALIZED
-	bool "Allow mmapped anonymous memory to be uninitialized"
-	depends on EXPERT && !MMU
-	default n
-	help
-	  Normally, and according to the Linux spec, anonymous memory obtained
-	  from mmap() has it's contents cleared before it is passed to
-	  userspace.  Enabling this config option allows you to request that
-	  mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
-	  providing a huge performance boost.  If this option is not enabled,
-	  then the flag will be ignored.
-
-	  This is taken advantage of by uClibc's malloc(), and also by
-	  ELF-FDPIC binfmt's brk and stack allocator.
-
-	  Because of the obvious security issues, this option should only be
-	  enabled on embedded devices where you control what is run in
-	  userspace.  Since that isn't generally a problem on no-MMU systems,
-	  it is normally safe to say Y here.
-
-	  See Documentation/nommu-mmap.txt for more information.
-
 config PROFILING
 	bool "Profiling support"
 	help

diff -puN mm/Kconfig.debug~organize-memory-config-options mm/Kconfig.debug

mm/Kconfig.debug renamed from mm/Kconfig.debug~organize-memory-config-options +11 -0
@@ -27,3 +27,14 @@ config PAGE_POISONING
 config PAGE_GUARD
 	bool
 	select WANT_PAGE_DEBUG_FLAGS
+
+config SLUB_DEBUG
+	default y
+	bool "Enable SLUB debugging support" if EXPERT
+	depends on SLUB && SYSFS
+	help
+	  SLUB has extensive debug support features. Disabling these can
+	  result in significant savings in code size. This also disables
+	  SLUB sysfs support. /sys/slab will not exist and there will be
+	  no support for cache validation etc.
+

_

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

Signature · 8 lines
Michal Hocko
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
#4 lore raw

Re: [PATCH 3/3] Kconfig: organize memory-related config options

Quoted 20 lines · Michal Hocko
Quoted 11 lines · Dave Hansen

This continues in a series of patches to clean up the configuration menus. I believe they've become really hard to navigate and there are some simple things we can do to make things easier to find.

This creates a "Memory Options" menu and moves some things like swap and slab configuration under them. It also moves SLUB_DEBUG to the debugging menu.

After this patch, the menu has the following options:

Quoted 7 lines

[ ] Memory placement aware NUMA scheduler [*] Enable VM event counters for /proc/vmstat [ ] Disable heap randomization
[*] Support for paging of anonymous memory (swap)
Choose SLAB allocator (SLUB (Unqueued Allocator))
[*] SLUB per cpu partial cache
[*] SLUB: attempt to use double-cmpxchg operations

Quoted 2 lines

Is there any reason to keep them in init/Kconfig rather than mm/Kconfig? It would sound like a logical place to have them all, no?

These options are the memory-related ones that fall under the "General setup" menu and the mm/Kconfig ones fall in to "Processor type and features". I've been hesitant to move these over to mm/Kconfig just because I don't want to put more stuff in the arch-specific menus.

You raise a good point, though, that there isn't a great logical separation about what should go where. Things like zram and KSM end up in "Processor type and features" when they're really pretty architecture-neutral.