lwn.git/include/linux/kernel.h, branch docs-4.15

linux/kernel.h: move DIV_ROUND_DOWN_ULL() macro

2017-09-09T01:26:47+00:00

This macro is useful to avoid link error on 32-bit systems. We have the same definition in two drivers, so move it to include/linux/kernel.h While we are here, refactor DIV_ROUND_UP_ULL() by using DIV_ROUND_DOWN_ULL(). Link: http://lkml.kernel.org/r/1500945156-12907-1-git-send-email-yamada.masahiro@socionext.com Signed-off-by: Masahiro Yamada Acked-by: Mark Brown Cc: Cyrille Pitchen Cc: Jaroslav Kysela Cc: Takashi Iwai Cc: Liam Girdwood Cc: Boris Brezillon Cc: Marek Vasut Cc: Brian Norris Cc: Richard Weinberger Cc: David Woodhouse Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

locking/refcounts, x86/asm: Implement fast refcount overflow protection

2017-08-17T08:40:26+00:00

This implements refcount_t overflow protection on x86 without a noticeable performance impact, though without the fuller checking of REFCOUNT_FULL. This is done by duplicating the existing atomic_t refcount implementation but with normally a single instruction added to detect if the refcount has gone negative (e.g. wrapped past INT_MAX or below zero). When detected, the handler saturates the refcount_t to INT_MIN / 2. With this overflow protection, the erroneous reference release that would follow a wrap back to zero is blocked from happening, avoiding the class of refcount-overflow use-after-free vulnerabilities entirely. Only the overflow case of refcounting can be perfectly protected, since it can be detected and stopped before the reference is freed and left to be abused by an attacker. There isn't a way to block early decrements, and while REFCOUNT_FULL stops increment-from-zero cases (which would be the state _after_ an early decrement and stops potential double-free conditions), this fast implementation does not, since it would require the more expensive cmpxchg loops. Since the overflow case is much more common (e.g. missing a "put" during an error path), this protection provides real-world protection. For example, the two public refcount overflow use-after-free exploits published in 2016 would have been rendered unexploitable: http://perception-point.io/2016/01/14/analysis-and-exploitation-of-a-linux-kernel-vulnerability-cve-2016-0728/ http://cyseclabs.com/page?n=02012016 This implementation does, however, notice an unchecked decrement to zero (i.e. caller used refcount_dec() instead of refcount_dec_and_test() and it resulted in a zero). Decrements under zero are noticed (since they will have resulted in a negative value), though this only indicates that a use-after-free may have already happened. Such notifications are likely avoidable by an attacker that has already exploited a use-after-free vulnerability, but it's better to have them reported than allow such conditions to remain universally silent. On first overflow detection, the refcount value is reset to INT_MIN / 2 (which serves as a saturation value) and a report and stack trace are produced. When operations detect only negative value results (such as changing an already saturated value), saturation still happens but no notification is performed (since the value was already saturated). On the matter of races, since the entire range beyond INT_MAX but before 0 is negative, every operation at INT_MIN / 2 will trap, leaving no overflow-only race condition. As for performance, this implementation adds a single "js" instruction to the regular execution flow of a copy of the standard atomic_t refcount operations. (The non-"and_test" refcount_dec() function, which is uncommon in regular refcount design patterns, has an additional "jz" instruction to detect reaching exactly zero.) Since this is a forward jump, it is by default the non-predicted path, which will be reinforced by dynamic branch prediction. The result is this protection having virtually no measurable change in performance over standard atomic_t operations. The error path, located in .text.unlikely, saves the refcount location and then uses UD0 to fire a refcount exception handler, which resets the refcount, handles reporting, and returns to regular execution. This keeps the changes to .text size minimal, avoiding return jumps and open-coded calls to the error reporting routine. Example assembly comparison: refcount_inc() before: .text: ffffffff81546149: f0 ff 45 f4 lock incl -0xc(%rbp) refcount_inc() after: .text: ffffffff81546149: f0 ff 45 f4 lock incl -0xc(%rbp) ffffffff8154614d: 0f 88 80 d5 17 00 js ffffffff816c36d3 ... .text.unlikely: ffffffff816c36d3: 48 8d 4d f4 lea -0xc(%rbp),%rcx ffffffff816c36d7: 0f ff (bad) These are the cycle counts comparing a loop of refcount_inc() from 1 to INT_MAX and back down to 0 (via refcount_dec_and_test()), between unprotected refcount_t (atomic_t), fully protected REFCOUNT_FULL (refcount_t-full), and this overflow-protected refcount (refcount_t-fast): 2147483646 refcount_inc()s and 2147483647 refcount_dec_and_test()s: cycles protections atomic_t 82249267387 none refcount_t-fast 82211446892 overflow, untested dec-to-zero refcount_t-full 144814735193 overflow, untested dec-to-zero, inc-from-zero This code is a modified version of the x86 PAX_REFCOUNT atomic_t overflow defense from the last public patch of PaX/grsecurity, based on my understanding of the code. Changes or omissions from the original code are mine and don't reflect the original grsecurity/PaX code. Thanks to PaX Team for various suggestions for improvement for repurposing this code to be a refcount-only protection. Signed-off-by: Kees Cook Reviewed-by: Josh Poimboeuf Cc: Alexey Dobriyan Cc: Andrew Morton Cc: Arnd Bergmann Cc: Christoph Hellwig Cc: David S. Miller Cc: Davidlohr Bueso Cc: Elena Reshetova Cc: Eric Biggers Cc: Eric W. Biederman Cc: Greg KH Cc: Hans Liljestrand Cc: James Bottomley Cc: Jann Horn Cc: Linus Torvalds Cc: Manfred Spraul Cc: Peter Zijlstra Cc: Rik van Riel Cc: Serge E. Hallyn Cc: Thomas Gleixner Cc: arozansk@redhat.com Cc: axboe@kernel.dk Cc: kernel-hardening@lists.openwall.com Cc: linux-arch Link: http://lkml.kernel.org/r/20170815161924.GA133115@beast Signed-off-by: Ingo Molnar

kernel.h: handle pointers to arrays better in container_of()

2017-07-12T23:25:59+00:00

If the first parameter of container_of() is a pointer to a non-const-qualified array type (and the third parameter names a non-const-qualified array member), the local variable __mptr will be defined with a const-qualified array type. In ISO C, these types are incompatible. They work as expected in GNU C, but some versions will issue warnings. For example, GCC 4.9 produces the warning "initialization from incompatible pointer type". Here is an example of where the problem occurs: ------------------------------------------------------- #include #include MODULE_LICENSE("GPL"); struct st { int a; char b[16]; }; static int __init example_init(void) { struct st t = { .a = 101, .b = "hello" }; char (*p)[16] = &t.b; struct st *x = container_of(p, struct st, b); printk(KERN_DEBUG "%p %p\n", (void *)&t, (void *)x); return 0; } static void __exit example_exit(void) { } module_init(example_init); module_exit(example_exit); ------------------------------------------------------- Building the module with gcc-4.9 results in these warnings (where '{m}' is the module source and '{k}' is the kernel source): ------------------------------------------------------- In file included from {m}/example.c:1:0: {m}/example.c: In function `example_init': {k}/include/linux/kernel.h:854:48: warning: initialization from incompatible pointer type const typeof( ((type *)0)->member ) *__mptr = (ptr); \ ^ {m}/example.c:14:17: note: in expansion of macro `container_of' struct st *x = container_of(p, struct st, b); ^ {k}/include/linux/kernel.h:854:48: warning: (near initialization for `x') const typeof( ((type *)0)->member ) *__mptr = (ptr); \ ^ {m}/example.c:14:17: note: in expansion of macro `container_of' struct st *x = container_of(p, struct st, b); ^ ------------------------------------------------------- Replace the type checking performed by the macro to avoid these warnings. Make sure `*(ptr)` either has type compatible with the member, or has type compatible with `void`, ignoring qualifiers. Raise compiler errors if this is not true. This is stronger than the previous behaviour, which only resulted in compiler warnings for a type mismatch. [arnd@arndb.de: fix new warnings for container_of()] Link: http://lkml.kernel.org/r/20170620200940.90557-1-arnd@arndb.de Link: http://lkml.kernel.org/r/20170525120316.24473-7-abbotti@mev.co.uk Signed-off-by: Ian Abbott Signed-off-by: Arnd Bergmann Acked-by: Michal Nazarewicz Acked-by: Kees Cook Cc: Hidehiro Kawai Cc: Borislav Petkov Cc: Rasmus Villemoes Cc: Johannes Berg Cc: Peter Zijlstra Cc: Alexander Potapenko Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

init: Introduce SYSTEM_SCHEDULING state

2017-05-23T08:01:38+00:00

might_sleep() debugging and smp_processor_id() debugging should be active right after the scheduler starts working. The init task can invoke smp_processor_id() from preemptible context as it is pinned on the boot cpu until sched_smp_init() removes the pinning and lets it schedule on all non isolated cpus. Add a new state which allows to enable those checks earlier and add it to the xen do_poweroff() function. No functional change. Tested-by: Mark Rutland Signed-off-by: Thomas Gleixner Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Boris Ostrovsky Acked-by: Mark Rutland Cc: Greg Kroah-Hartman Cc: Juergen Gross Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Steven Rostedt Link: http://lkml.kernel.org/r/20170516184736.196214622@linutronix.de Signed-off-by: Ingo Molnar

Merge branch 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6

2017-05-02T22:53:46+00:00

Pull crypto updates from Herbert Xu: "Here is the crypto update for 4.12: API: - Add batch registration for acomp/scomp - Change acomp testing to non-unique compressed result - Extend algorithm name limit to 128 bytes - Require setkey before accept(2) in algif_aead Algorithms: - Add support for deflate rfc1950 (zlib) Drivers: - Add accelerated crct10dif for powerpc - Add crc32 in stm32 - Add sha384/sha512 in ccp - Add 3des/gcm(aes) for v5 devices in ccp - Add Queue Interface (QI) backend support in caam - Add new Exynos RNG driver - Add ThunderX ZIP driver - Add driver for hardware random generator on MT7623 SoC" * 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (101 commits) crypto: stm32 - Fix OF module alias information crypto: algif_aead - Require setkey before accept(2) crypto: scomp - add support for deflate rfc1950 (zlib) crypto: scomp - allow registration of multiple scomps crypto: ccp - Change ISR handler method for a v5 CCP crypto: ccp - Change ISR handler method for a v3 CCP crypto: crypto4xx - rename ce_ring_contol to ce_ring_control crypto: testmgr - Allow ecb(cipher_null) in FIPS mode Revert "crypto: arm64/sha - Add constant operand modifier to ASM_EXPORT" crypto: ccp - Disable interrupts early on unload crypto: ccp - Use only the relevant interrupt bits hwrng: mtk - Add driver for hardware random generator on MT7623 SoC dt-bindings: hwrng: Add Mediatek hardware random generator bindings crypto: crct10dif-vpmsum - Fix missing preempt_disable() crypto: testmgr - replace compression known answer test crypto: acomp - allow registration of multiple acomps hwrng: n2 - Use devm_kcalloc() in n2rng_probe() crypto: chcr - Fix error handling related to 'chcr_alloc_shash' padata: get_next is never NULL crypto: exynos - Add new Exynos RNG driver ...

linux/kernel.h: Add ALIGN_DOWN macro

2017-04-21T12:30:44+00:00

Few parts of kernel define their own macro for aligning down so provide a common define for this, with the same usage and assumptions as existing ALIGN. Convert also three existing implementations to this one. Signed-off-by: Krzysztof Kozlowski Signed-off-by: Herbert Xu

boot/param: Move next_arg() function to lib/cmdline.c for later reuse

2017-04-18T08:37:13+00:00

next_arg() will be used to parse boot parameters in the x86/boot/compressed code, so move it to lib/cmdline.c for better code reuse. No change in functionality. Signed-off-by: Baoquan He Cc: Andrew Morton Cc: Gustavo Padovan Cc: Jens Axboe Cc: Jessica Yu Cc: Johannes Berg Cc: Josh Triplett Cc: Larry Finger Cc: Linus Torvalds Cc: Niklas Söderlund Cc: Peter Zijlstra Cc: Petr Mladek Cc: Rasmus Villemoes Cc: Thomas Gleixner Cc: dan.j.williams@intel.com Cc: dave.jiang@intel.com Cc: dyoung@redhat.com Cc: keescook@chromium.org Cc: zijun_hu Link: http://lkml.kernel.org/r/1492436099-4017-2-git-send-email-bhe@redhat.com Signed-off-by: Ingo Molnar

linux/kernel.h: fix DIV_ROUND_CLOSEST to support negative divisors

2017-02-25T01:46:57+00:00

While working on a thermal driver I encounter a scenario where the divisor could be negative, instead of adding local code to handle this I though I first try to add support for this in DIV_ROUND_CLOSEST. Add support to DIV_ROUND_CLOSEST for negative divisors if both dividend and divisor variable types are signed. This should not alter current behavior for users of the macro as previously negative divisors where not supported. Before: DIV_ROUND_CLOSEST( 59, 4) = 15 DIV_ROUND_CLOSEST( 59, -4) = -14 DIV_ROUND_CLOSEST( -59, 4) = -15 DIV_ROUND_CLOSEST( -59, -4) = 14 After: DIV_ROUND_CLOSEST( 59, 4) = 15 DIV_ROUND_CLOSEST( 59, -4) = -15 DIV_ROUND_CLOSEST( -59, 4) = -15 DIV_ROUND_CLOSEST( -59, -4) = 15 [akpm@linux-foundation.org: fix comment, per Guenter] Link: http://lkml.kernel.org/r/20161222102217.29011-1-niklas.soderlund+renesas@ragnatech.se Signed-off-by: Niklas Söderlund Reviewed-by: Guenter Roeck Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

taint/module: Fix problems when out-of-kernel driver defines true or false

2017-01-17T18:56:45+00:00

Commit 7fd8329ba502 ("taint/module: Clean up global and module taint flags handling") used the key words true and false as character members of a new struct. These names cause problems when out-of-kernel modules such as VirtualBox include their own definitions of true and false. Fixes: 7fd8329ba502 ("taint/module: Clean up global and module taint flags handling") Signed-off-by: Larry Finger Cc: Petr Mladek Cc: Jessica Yu Cc: Rusty Russell Reported-by: Valdis Kletnieks Reviewed-by: Petr Mladek Acked-by: Rusty Russell Signed-off-by: Jessica Yu

Merge tag 'modules-for-v4.10' of git://git.kernel.org/pub/scm/linux/kernel/git/jeyu/linux

2016-12-15T04:12:43+00:00

Pull modules updates from Jessica Yu: "Summary of modules changes for the 4.10 merge window: - The rodata= cmdline parameter has been extended to additionally apply to module mappings - Fix a hard to hit race between module loader error/clean up handling and ftrace registration - Some code cleanups, notably panic.c and modules code use a unified taint_flags table now. This is much cleaner than duplicating the taint flag code in modules.c" * tag 'modules-for-v4.10' of git://git.kernel.org/pub/scm/linux/kernel/git/jeyu/linux: module: fix DEBUG_SET_MODULE_RONX typo module: extend 'rodata=off' boot cmdline parameter to module mappings module: Fix a comment above strong_try_module_get() module: When modifying a module's text ignore modules which are going away too module: Ensure a module's state is set accordingly during module coming cleanup code module: remove trailing whitespace taint/module: Clean up global and module taint flags handling modpost: free allocated memory