Kernel Address sanitizer (KASan) is a dynamic memory error detector. It
provides fast and comprehensive solution for finding use-after-free and
out-of-bounds bugs.
KASAN uses compile-time instrumentation for checking every memory access,
therefore GCC > v4.9.2 required. v4.9.2 almost works, but has issues with
putting symbol aliases into the wrong section, which breaks kasan
instrumentation of globals.
This patch only adds infrastructure for kernel address sanitizer. It's
not available for use yet. The idea and some code was borrowed from [1].
Basic idea:
The main idea of KASAN is to use shadow memory to record whether each byte
of memory is safe to access or not, and use compiler's instrumentation to
check the shadow memory on each memory access.
Address sanitizer uses 1/8 of the memory addressable in kernel for shadow
memory and uses direct mapping with a scale and offset to translate a
memory address to its corresponding shadow address.
Here is function to translate address to corresponding shadow address:
unsigned long kasan_mem_to_shadow(unsigned long addr)
{
return (addr >> KASAN_SHADOW_SCALE_SHIFT) + KASAN_SHADOW_OFFSET;
}
where KASAN_SHADOW_SCALE_SHIFT = 3.
So for every 8 bytes there is one corresponding byte of shadow memory.
The following encoding used for each shadow byte: 0 means that all 8 bytes
of the corresponding memory region are valid for access; k (1 <= k <= 7)
means that the first k bytes are valid for access, and other (8 - k) bytes
are not; Any negative value indicates that the entire 8-bytes are
inaccessible. Different negative values used to distinguish between
different kinds of inaccessible memory (redzones, freed memory) (see
mm/kasan/kasan.h).
To be able to detect accesses to bad memory we need a special compiler.
Such compiler inserts a specific function calls (__asan_load*(addr),
__asan_store*(addr)) before each memory access of size 1, 2, 4, 8 or 16.
These functions check whether memory region is valid to access or not by
checking corresponding shadow memory. If access is not valid an error
printed.
Historical background of the address sanitizer from Dmitry Vyukov:
"We've developed the set of tools, AddressSanitizer (Asan),
ThreadSanitizer and MemorySanitizer, for user space. We actively use
them for testing inside of Google (continuous testing, fuzzing,
running prod services). To date the tools have found more than 10'000
scary bugs in Chromium, Google internal codebase and various
open-source projects (Firefox, OpenSSL, gcc, clang, ffmpeg, MySQL and
lots of others): [2] [3] [4].
The tools are part of both gcc and clang compilers.
We have not yet done massive testing under the Kernel AddressSanitizer
(it's kind of chicken and egg problem, you need it to be upstream to
start applying it extensively). To date it has found about 50 bugs.
Bugs that we've found in upstream kernel are listed in [5].
We've also found ~20 bugs in out internal version of the kernel. Also
people from Samsung and Oracle have found some.
[...]
As others noted, the main feature of AddressSanitizer is its
performance due to inline compiler instrumentation and simple linear
shadow memory. User-space Asan has ~2x slowdown on computational
programs and ~2x memory consumption increase. Taking into account that
kernel usually consumes only small fraction of CPU and memory when
running real user-space programs, I would expect that kernel Asan will
have ~10-30% slowdown and similar memory consumption increase (when we
finish all tuning).
I agree that Asan can well replace kmemcheck. We have plans to start
working on Kernel MemorySanitizer that finds uses of unitialized
memory. Asan+Msan will provide feature-parity with kmemcheck. As
others noted, Asan will unlikely replace debug slab and pagealloc that
can be enabled at runtime. Asan uses compiler instrumentation, so even
if it is disabled, it still incurs visible overheads.
Asan technology is easily portable to other architectures. Compiler
instrumentation is fully portable. Runtime has some arch-dependent
parts like shadow mapping and atomic operation interception. They are
relatively easy to port."
Comparison with other debugging features:
========================================
KMEMCHECK:
- KASan can do almost everything that kmemcheck can. KASan uses
compile-time instrumentation, which makes it significantly faster than
kmemcheck. The only advantage of kmemcheck over KASan is detection of
uninitialized memory reads.
Some brief performance testing showed that kasan could be
x500-x600 times faster than kmemcheck:
$ netperf -l 30
MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to localhost (127.0.0.1) port 0 AF_INET
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
no debug: 87380 16384 16384 30.00 41624.72
kasan inline: 87380 16384 16384 30.00 12870.54
kasan outline: 87380 16384 16384 30.00 10586.39
kmemcheck: 87380 16384 16384 30.03 20.23
- Also kmemcheck couldn't work on several CPUs. It always sets
number of CPUs to 1. KASan doesn't have such limitation.
DEBUG_PAGEALLOC:
- KASan is slower than DEBUG_PAGEALLOC, but KASan works on sub-page
granularity level, so it able to find more bugs.
SLUB_DEBUG (poisoning, redzones):
- SLUB_DEBUG has lower overhead than KASan.
- SLUB_DEBUG in most cases are not able to detect bad reads,
KASan able to detect both reads and writes.
- In some cases (e.g. redzone overwritten) SLUB_DEBUG detect
bugs only on allocation/freeing of object. KASan catch
bugs right before it will happen, so we always know exact
place of first bad read/write.
[1] https://code.google.com/p/address-sanitizer/wiki/AddressSanitizerForKernel
[2] https://code.google.com/p/address-sanitizer/wiki/FoundBugs
[3] https://code.google.com/p/thread-sanitizer/wiki/FoundBugs
[4] https://code.google.com/p/memory-sanitizer/wiki/FoundBugs
[5] https://code.google.com/p/address-sanitizer/wiki/AddressSanitizerForKernel#Trophies
Based on work by Andrey Konovalov.
Signed-off-by: Andrey Ryabinin <a.ryabinin@samsung.com>
Acked-by: Michal Marek <mmarek@suse.cz>
Signed-off-by: Andrey Konovalov <adech.fo@gmail.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Konstantin Serebryany <kcc@google.com>
Cc: Dmitry Chernenkov <dmitryc@google.com>
Cc: Yuri Gribov <tetra2005@gmail.com>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Cc: Sasha Levin <sasha.levin@oracle.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull s390 updates from Martin Schwidefsky:
- The remaining patches for the z13 machine support: kernel build
option for z13, the cache synonym avoidance, SMT support,
compare-and-delay for spinloops and the CES5S crypto adapater.
- The ftrace support for function tracing with the gcc hotpatch option.
This touches common code Makefiles, Steven is ok with the changes.
- The hypfs file system gets an extension to access diagnose 0x0c data
in user space for performance analysis for Linux running under z/VM.
- The iucv hvc console gets wildcard spport for the user id filtering.
- The cacheinfo code is converted to use the generic infrastructure.
- Cleanup and bug fixes.
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux: (42 commits)
s390/process: free vx save area when releasing tasks
s390/hypfs: Eliminate hypfs interval
s390/hypfs: Add diagnose 0c support
s390/cacheinfo: don't use smp_processor_id() in preemptible context
s390/zcrypt: fixed domain scanning problem (again)
s390/smp: increase maximum value of NR_CPUS to 512
s390/jump label: use different nop instruction
s390/jump label: add sanity checks
s390/mm: correct missing space when reporting user process faults
s390/dasd: cleanup profiling
s390/dasd: add locking for global_profile access
s390/ftrace: hotpatch support for function tracing
ftrace: let notrace function attribute disable hotpatching if necessary
ftrace: allow architectures to specify ftrace compile options
s390: reintroduce diag 44 calls for cpu_relax()
s390/zcrypt: Add support for new crypto express (CEX5S) adapter.
s390/zcrypt: Number of supported ap domains is not retrievable.
s390/spinlock: add compare-and-delay to lock wait loops
s390/tape: remove redundant if statement
s390/hvc_iucv: add simple wildcard matches to the iucv allow filter
...
If the kernel is compiled with function tracer support the -pg compile option
is passed to gcc to generate extra code into the prologue of each function.
This patch replaces the "open-coded" -pg compile flag with a CC_FLAGS_FTRACE
makefile variable which architectures can override if a different option
should be used for code generation.
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
The introduction of the uapi directories in v3.7-rc1 moved some of the
generated headers from arch/*/include/generated to the uapi directory,
keeping the #include directives intact.
This creates a problem when bisecting, because the unversioned files are
not cleaned automatically by git and the compiler might include stale
headers as a result. Instead of cleaning them in the Makefiles, promote
arch/*/include/generated/uapi in the search path. Under normal
circumstances, there is no overlap between this uapi subdirectory and
its parent, so the include choices remain the same. We keep
arch/*/include/generated/uapi in the USERINCLUDE variable so that it is
usable standalone.
Note that we cannot completely swap the order of the uapi and
kernel-only directories, since the headers in include/uapi/asm-generic
are meant to be wrapped by their include/asm-generic counterparts when
building kernel code.
Reported-by: "Nicholas A. Bellinger" <nab@linux-iscsi.org>
Reported-by: David Drysdale <dmd@lurklurk.org>
Signed-off-by: Michal Marek <mmarek@suse.cz>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull kbuild updates from Michal Marek:
"Here are the kbuild changes for v3.19-rc1:
- Cleanups and deduplication in the main Makefile and
scripts/Makefile.*
- Sort the output of *config targets in make help
- Old <linux/version.h> is always removed to avoid a surprise during
bisecting
- Warning fix in kconfig"
* 'kbuild' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild:
kbuild: remove redundant -rR flag of hdr-inst
kbuild: Fix make help-<board series> on powerpc
kbuild: Automatically remove stale <linux/version.h> file
kconfig: Fix warning "‘jump’ may be used uninitialized"
Makefile: sort list of defconfig targets in make help output
kbuild: Remove duplicate $(cmd) definition in Makefile.clean
kbuild: collect shorthands into scripts/Kbuild.include
make ARCH=powerpc help-<board series> should not require a cofigured
source tree. Also, sort the boards in the output.
Signed-off-by: Michal Marek <mmarek@suse.cz>
In 3.7, the file moved from include/linux/ to
include/generated/uapi/linux/. The path in the #include directive
remained the same for compatibility reasons, but this created a problem
when bisecting. Commit 9c8cdb71 (kbuild: unconditionally clobber
include/linux/version.h on distclean) fixes this, provided the user does
make distclean between builds. Better not rely on the user and delete
the stale file each time make is invoked.
Cc: Paul Gortmaker <paul.gortmaker@windriver.com>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Michal Marek <mmarek@suse.cz>
Without sorting this list is completely unreadable for ARCH=arm.
Signed-off-by: Konstantin Khlebnikov <k.khlebnikov@samsung.com>
Signed-off-by: Michal Marek <mmarek@suse.cz>
The shorthand "clean" is defined in both the top Makefile and
scripts/Makefile.clean. Likewise, the "hdr-inst" is defined in
both the top Makefile and scripts/Makefile.headersinst.
To reduce code duplication, this commit collects them into
scripts/Kbuild.include like the "build" and "modbuiltin" shorthands.
It requires scripts/Makefile.clean to include scripts/Kbuild.include,
but its impact on the performance of "make clean" should be
negligible.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Signed-off-by: Michal Marek <mmarek@suse.cz>
Sasha Levin reports:
"gcc5 changes the default standard to c11, which makes kernel build
unhappy
Explicitly define the kernel standard to be gnu89 which should keep
everything working exactly like it was before gcc5"
There are multiple small issues with the new default, but the biggest
issue seems to be that the old - and very useful - GNU extension to
allow a cast in front of an initializer has gone away.
Patch updated by Kirill:
"I'm pretty sure all gcc versions you can build kernel with supports
-std=gnu89. cc-option is redunrant.
We also need to adjust HOSTCFLAGS otherwise allmodconfig fails for me"
Note by Andrew Pinski:
"Yes it was reported and both problems relating to this extension has
been added to gnu99 and gnu11. Though there are other issues with the
kernel dealing with extern inline have different semantics between
gnu89 and gnu99/11"
End result: we may be able to move up to a newer stdc model eventually,
but right now the newer models have some annoying deficiencies, so the
traditional "gnu89" model ends up being the preferred one.
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Singed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull kbuild changes from Michal Marek:
- fix for handling dependencies of *-objs targets by Masahiro Yamada
- lots of cleanups in the kbuild machinery, also by Masahiro
- fixes for the kconfig build to use an UTF-8 capable ncurses library
if possible and to build on not-so-standard installs
- some more minor fixes
* 'kbuild' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild:
kbuild: Do not reference *-n variables in the Makefile
kbuild: simplify build, clean, modbuiltin shorthands
kbuild: arm: Do not define "comma" twice
kbuild: remove obj-n and lib-n handling
kbuild: remove unnecessary variable initializaions
kbuild: remove unnecessary "obj- := dummy.o" trick
kbuild: handle C=... and M=... after entering into build directory
kbuild: use $(Q) for sub-make target
kbuild: fake the "Entering directory ..." message more simply
kconfig/lxdialog: get ncurses CFLAGS with pkg-config
kconfig: nconfig: fix multi-byte UTF handling
kconfig: lxdialog: fix spelling
kbuild: Make scripts executable
kbuild: remove redundant clean-files from scripts/kconfig/Makefile
kbuild: refactor script/kconfig/Makefile
kbuild: handle the dependency of multi-objs hostprogs appropriately
kbuild: handle multi-objs dependency appropriately
Pull documentation updates from Jiri Kosina:
"Updates to kernel documentation.
I took this over (hopefully temporarily) from Randy who was not
willing to maintain it any longer. This pile mostly is a relay of
queue that Randy already had in his tree"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/doc:
Documentation: fix broken v4l-utils URL
Documentation: update include path for mpssd
Documentation: correct parameter error for dma_mapping_error
MAINTAINERS: update location of linux-doc tree
Documentation: remove networking/.gitignore
tools: add more endian.h macros
Make Documenation depend on headers_install
Docs: this_cpu_ops: remove redundant add forms
Documentation: disable vdso_test to avoid breakage with old glibc
Documentation: update vDSO makefile to build portable examples
Documentation: update .gitignore files
Documentation: support glibc versions without htole macros
v4l2-pci-skeleton: Only build if PCI is available
Documentation: fix misc. warnings
Documentation: make functions static to avoid prototype warnings
Documentation: add makefiles for more targets
Documentation: use subdir-y to avoid unnecessary built-in.o files
Cheers,
Rusty.
PS. My virtio-next tree is empty: DaveM took the patches I had. There might
be a virtio-rng starvation fix, but so far it's a bit voodoo so I will
get to that in the next two days or it will wait.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1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=QzCn
-----END PGP SIGNATURE-----
Merge tag 'modules-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux
Pull module update from Rusty Russell:
"Nothing major: support for compressing modules, and auto-tainting
params.
PS. My virtio-next tree is empty: DaveM took the patches I had. There
might be a virtio-rng starvation fix, but so far it's a bit voodoo
so I will get to that in the next two days or it will wait"
* tag 'modules-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux:
moduleparam: Resolve missing-field-initializer warning
kbuild: handle module compression while running 'make modules_install'.
modinst: wrap long lines in order to enhance cmd_modules_install
modsign: lookup lines ending in .ko in .mod files
modpost: simplify file name generation of *.mod.c files
modpost: reduce visibility of symbols and constify r/o arrays
param: check for tainting before calling set op.
drm/i915: taint the kernel if unsafe module parameters are set
module: add module_param_unsafe and module_param_named_unsafe
module: make it possible to have unsafe, tainting module params
module: rename KERNEL_PARAM_FL_NOARG to avoid confusion
$(if $(KBUILD_SRC),$(srctree)/) was a useful strategy
to omit a long absolute path for in-source-tree build
prior to commit 890676c65d
(kbuild: Use relative path when building in the source tree).
Now $(srctree) is "." when building in the source tree.
It would not be annoying to add "$(srctree)/" all the time.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Signed-off-by: Michal Marek <mmarek@suse.cz>
This commit avoids processing C=... and M=... twice
when O=... is also given.
Besides, we can also remove KBUILD_EXTMOD="$(KBUILD_EXTMOD)"
in the sub-make target.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Acked-by: Peter Foley <pefoley2@pefoley.com>
Signed-off-by: Michal Marek <mmarek@suse.cz>
Since commit 066b7ed955
(kbuild: Do not print the build directory with make -s),
"Q" is defined above the sub-make target.
This commit takes advantage of that and replaces
"$(if $(KBUILD_VERBOSE:1=),@)" with "$(Q)".
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Acked-by: Peter Foley <pefoley2@pefoley.com>
Signed-off-by: Michal Marek <mmarek@suse.cz>
Commit c2e28dc975
(kbuild: Print the name of the build directory)
added a gimmick to show the "Entering directory ...".
Instead of echoing the hard-coded message (that is, we need to know
the exact message), moving --no-print-directory would be easier.
Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Acked-by: Peter Foley <pefoley2@pefoley.com>
Signed-off-by: Michal Marek <mmarek@suse.cz>
Since module-init-tools (gzip) and kmod (gzip and xz) support compressed
modules, it could be useful to include a support for compressing modules
right after having them installed. Doing this in kbuild instead of per
distro can permit to make this kind of usage more generic.
This patch add a Kconfig entry to "Enable loadable module support" menu
and let you choose to compress using gzip (default) or xz.
Both gzip and xz does not used any extra -[1-9] option since Andi Kleen
and Rusty Russell prove no gain is made using them. gzip is called with -n
argument to avoid storing original filename inside compressed file, that
way we can save some more bytes.
On a v3.16 kernel, 'make allmodconfig' generated 4680 modules for a
total of 378MB (no strip, no sign, no compress), the following table
shows observed disk space gain based on the allmodconfig .config :
| time |
+-------------+-----------------+
| manual .ko | make | size | percent
| compression | modules_install | | gain
+-------------+-----------------+------+--------
- | | 18.61s | 378M |
GZIP | 3m16s | 3m37s | 102M | 73.41%
XZ | 5m22s | 5m39s | 77M | 79.83%
The gain for restricted environnement seems to be interesting while
uncompress can be time consuming but happens only while loading a module,
that is generally done only once.
This is fully compatible with signed modules while the signed module is
compressed. module-init-tools or kmod handles decompression
and provide to other layer the uncompressed but signed payload.
Reviewed-by: Willy Tarreau <w@1wt.eu>
Signed-off-by: Bertrand Jacquin <beber@meleeweb.net>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Pull misc kbuild updates from Michal Marek:
"This is the non-critical part of kbuild for 3.17-rc1:
- make help hint to use make -s with make kernelrelease et al.
- moved a kbuild document to Documentation/kbuild where it belongs
- four new Coccinelle scripts, one dropped and one fixed
- new make kselftest target to run various tests on the kernel"
* 'misc' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild:
kbuild: kselftest - new make target to build and run kernel selftests
Coccinelle: Script to replace if and BUG with BUG_ON
Coccinelle: Script to detect incorrect argument to sizeof
Coccinelle: Script to use ARRAY_SIZE instead of division of two sizeofs
Coccinelle: Script to detect cast after memory allocation
coccinelle/null: solve parse error
Documentation: headers_install.txt is part of kbuild
kbuild: make -s should be used with kernelrelease/kernelversion/image_name
Pull kbuild updates from Michal Marek:
- make clean also considers $(extra-m) and $(extra-) to be consistent
- cleanup and fixes in scripts/Makefile.host
- allow to override the name of the Python 2 executable with make
PYTHON=... (only needed for ia64 in practice)
- option to split debugingo into *.dwo files to save disk space if the
compiler supports it (CONFIG_DEBUG_INFO_SPLIT)
- option to use dwarf4 debuginfo if the compiler supports it
(CONFIG_DEBUG_INFO_DWARF4)
- fix for disabling certain warnings with clang
- fix for unneeded rebuild with dash when a command contains
backslashes
* 'kbuild' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild:
kbuild: Fix handling of backslashes in *.cmd files
kbuild, LLVMLinux: Supress warnings unless W=1-3
Kbuild: Add a option to enable dwarf4 v2
kbuild: Support split debug info v4
kbuild: allow to override Python command name
kbuild: clean-up and bug fix of scripts/Makefile.host
kbuild: clean up scripts/Makefile.host
kbuild: drop shared library support from Makefile.host
kbuild: fix a bug of C++ host program handling
kbuild: fix a typo in scripts/Makefile.host
scripts/Makefile.clean: clean also $(extra-m) and $(extra-)
Add a new make target "kselftest" to enable kernel testing. This
new target builds and runs kernel selftests. Running as root is
recommended for a complete test run as some tests don't run when
run by non-root user. Build, install, and boot kernel before
running kselftest on it.
Signed-off-by: Shuah Khan <shuah.kh@samsung.com>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Michal Marek <mmarek@suse.cz>
We have been chasing a memory corruption bug, which turned out to be
caused by very old gcc (4.3.4), which happily turned conditional load
into a non-conditional one, and that broke correctness (the condition
was met only if lock was held) and corrupted memory.
This particular problem with that particular code did not happen when
never gccs were used. I've brought this up with our gcc folks, as I
wanted to make sure that this can't really happen again, and it turns
out it actually can.
Quoting Martin Jambor <mjambor@suse.cz>:
"More current GCCs are more careful when it comes to replacing a
conditional load with a non-conditional one, most notably they check
that a store happens in each iteration of _a_ loop but they assume
loops are executed. They also perform a simple check whether the
store cannot trap which currently passes only for non-const
variables. A simple testcase demonstrating it on an x86_64 is for
example the following:
$ cat cond_store.c
int g_1 = 1;
int g_2[1024] __attribute__((section ("safe_section"), aligned (4096)));
int c = 4;
int __attribute__ ((noinline))
foo (void)
{
int l;
for (l = 0; (l != 4); l++) {
if (g_1)
return l;
for (g_2[0] = 0; (g_2[0] >= 26); ++g_2[0])
;
}
return 2;
}
int main (int argc, char* argv[])
{
if (mprotect (g_2, sizeof(g_2), PROT_READ) == -1)
{
int e = errno;
error (e, e, "mprotect error %i", e);
}
foo ();
__builtin_printf("OK\n");
return 0;
}
/* EOF */
$ ~/gcc/trunk/inst/bin/gcc cond_store.c -O2 --param allow-store-data-races=0
$ ./a.out
OK
$ ~/gcc/trunk/inst/bin/gcc cond_store.c -O2 --param allow-store-data-races=1
$ ./a.out
Segmentation fault
The testcase fails the same at least with 4.9, 4.8 and 4.7. Therefore
I would suggest building kernels with this parameter set to zero. I
also agree with Jikos that the default should be changed for -O2. I
have run most of the SPEC 2k6 CPU benchmarks (gamess and dealII
failed, at -O2, not sure why) compiled with and without this option
and did not see any real difference between respective run-times"
Hopefully the default will be changed in newer gccs, but let's force it
for kernel builds so that we are on a safe side even when older gcc are
used.
The code in question was out-of-tree printk-in-NMI (yeah, surprise
suprise, once again) patch written by Petr Mladek, let me quote his
comment from our internal bugzilla:
"I have spent few days investigating inconsistent state of kernel ring buffer.
It went out that it was caused by speculative store generated by
gcc-4.3.4.
The problem is in assembly generated for make_free_space(). The functions is
called the following way:
+ vprintk_emit();
+ log = MAIN_LOG; // with logbuf_lock
or
log = NMI_LOG; // with nmi_logbuf_lock
cont_add(log, ...);
+ cont_flush(log, ...);
+ log_store(log, ...);
+ log_make_free_space(log, ...);
If called with log = NMI_LOG then only nmi_log_* global variables are safe to
modify but the generated code does store also into (main_)log_* global
variables:
<log_make_free_space>:
55 push %rbp
89 f6 mov %esi,%esi
48 8b 05 03 99 51 01 mov 0x1519903(%rip),%rax # ffffffff82620868 <nmi_log_next_id>
44 8b 1d ec 98 51 01 mov 0x15198ec(%rip),%r11d # ffffffff82620858 <log_next_idx>
8b 35 36 60 14 01 mov 0x1146036(%rip),%esi # ffffffff8224cfa8 <log_buf_len>
44 8b 35 33 60 14 01 mov 0x1146033(%rip),%r14d # ffffffff8224cfac <nmi_log_buf_len>
4c 8b 2d d0 98 51 01 mov 0x15198d0(%rip),%r13 # ffffffff82620850 <log_next_seq>
4c 8b 25 11 61 14 01 mov 0x1146111(%rip),%r12 # ffffffff8224d098 <log_buf>
49 89 c2 mov %rax,%r10
48 21 c2 and %rax,%rdx
48 8b 1d 0c 99 55 01 mov 0x155990c(%rip),%rbx # ffffffff826608a0 <nmi_log_buf>
49 c1 ea 20 shr $0x20,%r10
48 89 55 d0 mov %rdx,-0x30(%rbp)
44 29 de sub %r11d,%esi
45 29 d6 sub %r10d,%r14d
4c 8b 0d 97 98 51 01 mov 0x1519897(%rip),%r9 # ffffffff82620840 <log_first_seq>
eb 7e jmp ffffffff81107029 <log_make_free_space+0xe9>
[...]
85 ff test %edi,%edi # edi = 1 for NMI_LOG
4c 89 e8 mov %r13,%rax
4c 89 ca mov %r9,%rdx
74 0a je ffffffff8110703d <log_make_free_space+0xfd>
8b 15 27 98 51 01 mov 0x1519827(%rip),%edx # ffffffff82620860 <nmi_log_first_id>
48 8b 45 d0 mov -0x30(%rbp),%rax
48 39 c2 cmp %rax,%rdx # end of loop
0f 84 da 00 00 00 je ffffffff81107120 <log_make_free_space+0x1e0>
[...]
85 ff test %edi,%edi # edi = 1 for NMI_LOG
4c 89 0d 17 97 51 01 mov %r9,0x1519717(%rip) # ffffffff82620840 <log_first_seq>
^^^^^^^^^^^^^^^^^^^^^^^^^^
KABOOOM
74 35 je ffffffff81107160 <log_make_free_space+0x220>
It stores log_first_seq when edi == NMI_LOG. This instructions are used also
when edi == MAIN_LOG but the store is done speculatively before the condition
is decided. It is unsafe because we do not have "logbuf_lock" in NMI context
and some other process migh modify "log_first_seq" in parallel"
I believe that the best course of action is both
- building kernel (and anything multi-threaded, I guess) with that
optimization turned off
- persuade gcc folks to change the default for future releases
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Cc: Martin Jambor <mjambor@suse.cz>
Cc: Petr Mladek <pmladek@suse.cz>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Marek Polacek <polacek@redhat.com>
Cc: Jakub Jelinek <jakub@redhat.com>
Cc: Steven Noonan <steven@uplinklabs.net>
Cc: Richard Biener <richard.guenther@gmail.com>
Cc: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This adds a hopefully helpful comment above the (seemingly weird) compiler
flag selection logic.
Signed-off-by: Kees Cook <keescook@chromium.org>
Suggested-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Michal Marek <mmarek@suse.cz>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
clang has more warnings enabled by default. Turn them off unless W is
set. This patch fixes a logic bug where warnings in clang were disabled
when W was set.
Signed-off-by: Behan Webster <behanw@converseincode.com>
Signed-off-by: Jan-Simon Möller <dl9pf@gmx.de>
Signed-off-by: Mark Charlebois <charlebm@gmail.com>
Cc: bp@alien8.de
Signed-off-by: Michal Marek <mmarek@suse.cz>
