From e872267b8bcbb179e21ccc7118f258873d6e7a59 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Tue, 18 Sep 2018 23:51:43 -0700 Subject: jump_table: Move entries into ro_after_init region The __jump_table sections emitted into the core kernel and into each module consist of statically initialized references into other parts of the code, and with the exception of entries that point into init code, which are defused at post-init time, these data structures are never modified. So let's move them into the ro_after_init section, to prevent them from being corrupted inadvertently by buggy code, or deliberately by an attacker. Signed-off-by: Ard Biesheuvel Signed-off-by: Thomas Gleixner Reviewed-by: Kees Cook Acked-by: Jessica Yu Acked-by: Peter Zijlstra (Intel) Cc: linux-arm-kernel@lists.infradead.org Cc: linux-s390@vger.kernel.org Cc: Arnd Bergmann Cc: Heiko Carstens Cc: Will Deacon Cc: Catalin Marinas Cc: Steven Rostedt Cc: Martin Schwidefsky Link: https://lkml.kernel.org/r/20180919065144.25010-9-ard.biesheuvel@linaro.org --- include/asm-generic/vmlinux.lds.h | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'include/asm-generic') diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 7b75ff6e2fce..f09ee3c544bc 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -253,10 +253,6 @@ STRUCT_ALIGN(); \ *(__tracepoints) \ /* implement dynamic printk debug */ \ - . = ALIGN(8); \ - __start___jump_table = .; \ - KEEP(*(__jump_table)) \ - __stop___jump_table = .; \ . = ALIGN(8); \ __start___verbose = .; \ KEEP(*(__verbose)) \ @@ -300,6 +296,12 @@ . = __start_init_task + THREAD_SIZE; \ __end_init_task = .; +#define JUMP_TABLE_DATA \ + . = ALIGN(8); \ + __start___jump_table = .; \ + KEEP(*(__jump_table)) \ + __stop___jump_table = .; + /* * Allow architectures to handle ro_after_init data on their * own by defining an empty RO_AFTER_INIT_DATA. @@ -308,6 +310,7 @@ #define RO_AFTER_INIT_DATA \ __start_ro_after_init = .; \ *(.data..ro_after_init) \ + JUMP_TABLE_DATA \ __end_ro_after_init = .; #endif -- cgit v1.2.1 From 27df89689e257cccb604fdf56c91a75a25aa554a Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Mon, 20 Aug 2018 10:19:14 -0400 Subject: locking/spinlocks: Remove an instruction from spin and write locks Both spin locks and write locks currently do: f0 0f b1 17 lock cmpxchg %edx,(%rdi) 85 c0 test %eax,%eax 75 05 jne [slowpath] This 'test' insn is superfluous; the cmpxchg insn sets the Z flag appropriately. Peter pointed out that using atomic_try_cmpxchg_acquire() will let the compiler know this is true. Comparing before/after disassemblies show the only effect is to remove this insn. Take this opportunity to make the spin & write lock code resemble each other more closely and have similar likely() hints. Suggested-by: Peter Zijlstra Signed-off-by: Matthew Wilcox Signed-off-by: Peter Zijlstra (Intel) Acked-by: Will Deacon Cc: Arnd Bergmann Cc: Linus Torvalds Cc: Thomas Gleixner Cc: Waiman Long Link: http://lkml.kernel.org/r/20180820162639.GC25153@bombadil.infradead.org Signed-off-by: Ingo Molnar --- include/asm-generic/qrwlock.h | 7 ++++--- include/asm-generic/qspinlock.h | 16 +++++++++------- 2 files changed, 13 insertions(+), 10 deletions(-) (limited to 'include/asm-generic') diff --git a/include/asm-generic/qrwlock.h b/include/asm-generic/qrwlock.h index 0f7062bd55e5..36254d2da8e0 100644 --- a/include/asm-generic/qrwlock.h +++ b/include/asm-generic/qrwlock.h @@ -71,8 +71,8 @@ static inline int queued_write_trylock(struct qrwlock *lock) if (unlikely(cnts)) return 0; - return likely(atomic_cmpxchg_acquire(&lock->cnts, - cnts, cnts | _QW_LOCKED) == cnts); + return likely(atomic_try_cmpxchg_acquire(&lock->cnts, &cnts, + _QW_LOCKED)); } /** * queued_read_lock - acquire read lock of a queue rwlock @@ -96,8 +96,9 @@ static inline void queued_read_lock(struct qrwlock *lock) */ static inline void queued_write_lock(struct qrwlock *lock) { + u32 cnts = 0; /* Optimize for the unfair lock case where the fair flag is 0. */ - if (atomic_cmpxchg_acquire(&lock->cnts, 0, _QW_LOCKED) == 0) + if (likely(atomic_try_cmpxchg_acquire(&lock->cnts, &cnts, _QW_LOCKED))) return; queued_write_lock_slowpath(lock); diff --git a/include/asm-generic/qspinlock.h b/include/asm-generic/qspinlock.h index 9cc457597ddf..7541fa707f5b 100644 --- a/include/asm-generic/qspinlock.h +++ b/include/asm-generic/qspinlock.h @@ -66,10 +66,12 @@ static __always_inline int queued_spin_is_contended(struct qspinlock *lock) */ static __always_inline int queued_spin_trylock(struct qspinlock *lock) { - if (!atomic_read(&lock->val) && - (atomic_cmpxchg_acquire(&lock->val, 0, _Q_LOCKED_VAL) == 0)) - return 1; - return 0; + u32 val = atomic_read(&lock->val); + + if (unlikely(val)) + return 0; + + return likely(atomic_try_cmpxchg_acquire(&lock->val, &val, _Q_LOCKED_VAL)); } extern void queued_spin_lock_slowpath(struct qspinlock *lock, u32 val); @@ -80,11 +82,11 @@ extern void queued_spin_lock_slowpath(struct qspinlock *lock, u32 val); */ static __always_inline void queued_spin_lock(struct qspinlock *lock) { - u32 val; + u32 val = 0; - val = atomic_cmpxchg_acquire(&lock->val, 0, _Q_LOCKED_VAL); - if (likely(val == 0)) + if (likely(atomic_try_cmpxchg_acquire(&lock->val, &val, _Q_LOCKED_VAL))) return; + queued_spin_lock_slowpath(lock, val); } -- cgit v1.2.1 From f81f8ad56fd1c7b99b2ed1c314527f7d9ac447c6 Mon Sep 17 00:00:00 2001 From: Nadav Amit Date: Wed, 3 Oct 2018 14:30:56 -0700 Subject: x86/bug: Macrofy the BUG table section handling, to work around GCC inlining bugs As described in: 77b0bf55bc67: ("kbuild/Makefile: Prepare for using macros in inline assembly code to work around asm() related GCC inlining bugs") GCC's inlining heuristics are broken with common asm() patterns used in kernel code, resulting in the effective disabling of inlining. The workaround is to set an assembly macro and call it from the inline assembly block. As a result GCC considers the inline assembly block as a single instruction. (Which it isn't, but that's the best we can get.) This patch increases the kernel size: text data bss dec hex filename 18146889 10225380 2957312 31329581 1de0d2d ./vmlinux before 18147336 10226688 2957312 31331336 1de1408 ./vmlinux after (+1755) But enables more aggressive inlining (and probably better branch decisions). The number of static text symbols in vmlinux is much lower: Before: 40218 After: 40053 (-165) The assembly code gets harder to read due to the extra macro layer. [ mingo: Rewrote the changelog. ] Tested-by: Kees Cook Signed-off-by: Nadav Amit Acked-by: Peter Zijlstra (Intel) Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Josh Poimboeuf Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20181003213100.189959-7-namit@vmware.com Signed-off-by: Ingo Molnar --- include/asm-generic/bug.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include/asm-generic') diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h index 20561a60db9c..cdafa5edea49 100644 --- a/include/asm-generic/bug.h +++ b/include/asm-generic/bug.h @@ -17,10 +17,8 @@ #ifndef __ASSEMBLY__ #include -#ifdef CONFIG_BUG - -#ifdef CONFIG_GENERIC_BUG struct bug_entry { +#ifdef CONFIG_GENERIC_BUG #ifndef CONFIG_GENERIC_BUG_RELATIVE_POINTERS unsigned long bug_addr; #else @@ -35,8 +33,10 @@ struct bug_entry { unsigned short line; #endif unsigned short flags; -}; #endif /* CONFIG_GENERIC_BUG */ +}; + +#ifdef CONFIG_BUG /* * Don't use BUG() or BUG_ON() unless there's really no way out; one -- cgit v1.2.1