- 19 Oct, 2018 2 commits
-
-
Waiman Long authored
Make the frequently used lockdep global variable debug_locks read-mostly. As debug_locks_silent is sometime used together with debug_locks, it is also made read-mostly so that they can be close together. With false cacheline sharing, cacheline contention problem can happen depending on what get put into the same cacheline as debug_locks. Signed-off-by:
Waiman Long <longman@redhat.com> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Will Deacon <will.deacon@arm.com> Link: http://lkml.kernel.org/r/1539913518-15598-2-git-send-email-longman@redhat.comSigned-off-by:
Ingo Molnar <mingo@kernel.org>
-
Waiman Long authored
It was found that when debug_locks was turned off because of a problem found by the lockdep code, the system performance could drop quite significantly when the lock_stat code was also configured into the kernel. For instance, parallel kernel build time on a 4-socket x86-64 server nearly doubled. Further analysis into the cause of the slowdown traced back to the frequent call to debug_locks_off() from the __lock_acquired() function probably due to some inconsistent lockdep states with debug_locks off. The debug_locks_off() function did an unconditional atomic xchg to write a 0 value into debug_locks which had already been set to 0. This led to severe cacheline contention in the cacheline that held debug_locks. As debug_locks is being referenced in quite a few different places in the kernel, this greatly slow down the system performance. To prevent that trashing of debug_locks cacheline, lock_acquired() and lock_contended() now checks the state of debug_locks before proceeding. The debug_locks_off() function is also modified to check debug_locks before calling __debug_locks_off(). Signed-off-by:
-
- 17 Oct, 2018 2 commits
-
-
Waiman Long authored
The qspinlock code supports up to 4 levels of slowpath nesting using four per-CPU mcs_spinlock structures. For 64-bit architectures, they fit nicely in one 64-byte cacheline. For para-virtualized (PV) qspinlocks it needs to store more information in the per-CPU node structure than there is space for. It uses a trick to use a second cacheline to hold the extra information that it needs. So PV qspinlock needs to access two extra cachelines for its information whereas the native qspinlock code only needs one extra cacheline. Freshly added counter profiling of the qspinlock code, however, revealed that it was very rare to use more than two levels of slowpath nesting. So it doesn't make sense to penalize PV qspinlock code in order to have four mcs_spinlock structures in the same cacheline to optimize for a case in the native qspinlock code that rarely happens. Extend the per-CPU node structure to have two more long words when PV qspinlock locks are configured to hold the extra data that it needs. As a result, the PV qspinlock code will enjoy the same benefit of using just one extra cacheline like the native counterpart, for most cases. [ mingo: Minor changelog edits. ] Signed-off-by:
-
Waiman Long authored
Queued spinlock supports up to 4 levels of lock slowpath nesting - user context, soft IRQ, hard IRQ and NMI. However, we are not sure how often the nesting happens. So add 3 more per-CPU stat counters to track the number of instances where nesting index goes to 1, 2 and 3 respectively. On a dual-socket 64-core 128-thread Zen server, the following were the new stat counter values under different circumstances: State slowpath index1 index2 index3 ----- -------- ------ ------ ------- After bootup 1,012,150 82 0 0 After parallel build + perf-top 125,195,009 82 0 0 So the chance of having more than 2 levels of nesting is extremely low. [ mingo: Minor changelog edits. ] Signed-off-by:
-
- 16 Oct, 2018 6 commits
-
-
Peter Zijlstra authored
On x86 we cannot do fetch_or() with a single instruction and thus end up using a cmpxchg loop, this reduces determinism. Replace the fetch_or() with a composite operation: tas-pending + load. Using two instructions of course opens a window we previously did not have. Consider the scenario: CPU0 CPU1 CPU2 1) lock trylock -> (0,0,1) 2) lock trylock /* fail */ 3) unlock -> (0,0,0) 4) lock trylock -> (0,0,1) 5) tas-pending -> (0,1,1) load-val <- (0,1,0) from 3 6) clear-pending-set-locked -> (0,0,1) FAIL: _2_ owners where 5) is our new composite operation. When we consider each part of the qspinlock state as a separate variable (as we can when _Q_PENDING_BITS == 8) then the above is entirely possible, because tas-pending will only RmW the pending byte, so the later load is able to observe prior tail and lock state (but not earlier than its own trylock, which operates on the whole word, due to coherence). To avoid this we need 2 things: - the load must come after the tas-pending (obviously, otherwise it can trivially observe prior state). - the tas-pending must be a full word RmW instruction, it cannot be an XCHGB for example, such that we cannot observe other state prior to setting pending. On x86 we can realize this by using "LOCK BTS m32, r32" for tas-pending followed by a regular load. Note that observing later state is not a problem: - if we fail to observe a later unlock, we'll simply spin-wait for that store to become visible. - if we observe a later xchg_tail(), there is no difference from that xchg_tail() having taken place before the tas-pending. Suggested-by:
Will Deacon <will.deacon@arm.com> Reported-by:
Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by:
-
Peter Zijlstra authored
Currently the GEN_*_RMWcc() macros include a return statement, which pretty much mandates we directly wrap them in a (inline) function. Macros with return statements are tricky and, as per the above, limit use, so remove the return statement and make them statement-expressions. This allows them to be used more widely. Also, shuffle the arguments a bit. Place the @cc argument as 3rd, this makes it consistent between UNARY and BINARY, but more importantly, it makes the @arg0 argument last. Since the @arg0 argument is now last, we can do CPP trickery and make it an optional argument, simplifying the users; 17 out of 18 occurences do not need this argument. Finally, change to asm symbolic names, instead of the numeric ordering of operands, which allows us to get rid of __BINARY_RMWcc_ARG and get cleaner code overall. Signed-off-by:
-
Peter Zijlstra authored
While working my way through the code again; I felt the comments could use help. Signed-off-by:
-
Peter Zijlstra authored
Flip the branch condition after atomic_fetch_or_acquire(_Q_PENDING_VAL) such that we loose the indent. This also result in a more natural code flow IMO. Signed-off-by:
-
Ingo Molnar authored
Signed-off-by: -
Waiman Long authored
Remove the duplicated 'lock_class_ops' percpu array that is not used anywhere. Signed-off-by:
-
- 10 Oct, 2018 1 commit
-
-
Adam Borowski authored
A spanking new machine I just got has all but one USB ports wired as 3.0. Booting defconfig resulted in no keyboard or mouse, which was pretty uncool. Let's enable that -- USB3 is ubiquitous rather than an oddity. As 'y' not 'm' -- recovering from initrd problems needs a keyboard. Also add it to the 32-bit defconfig. Signed-off-by:
Adam Borowski <kilobyte@angband.pl> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-usb@vger.kernel.org Link: http://lkml.kernel.org/r/20181009062803.4332-1-kilobyte@angband.plSigned-off-by:
-
- 09 Oct, 2018 2 commits
-
-
Lance Roy authored
lockdep_assert_held() is better suited for checking locking requirements, since it won't get confused when the lock is held by some other task. This is also a step towards possibly removing spin_is_locked(). Signed-off-by:
Lance Roy <ldr709@gmail.com> Signed-off-by:
-
Waiman Long authored
A sizable portion of the CPU cycles spent on the __lock_acquire() is used up by the atomic increment of the class->ops stat counter. By taking it out from the lock_class structure and changing it to a per-cpu per-lock-class counter, we can reduce the amount of cacheline contention on the class structure when multiple CPUs are trying to acquire locks of the same class simultaneously. To limit the increase in memory consumption because of the percpu nature of that counter, it is now put back under the CONFIG_DEBUG_LOCKDEP config option. So the memory consumption increase will only occur if CONFIG_DEBUG_LOCKDEP is defined. The lock_class structure, however, is reduced in size by 16 bytes on 64-bit archs after ops removal and a minor restructuring of the fields. This patch also fixes a bug in the increment code as the counter is of the 'unsigned long' type, but atomic_inc() was used to increment it. Signed-off-by:
Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Will Deacon <will.deacon@arm.com> Link: http://lkml.kernel.org/r/d66681f3-8781-9793-1dcf-2436a284550b@redhat.comSigned-off-by:
-
- 06 Oct, 2018 4 commits
-
-
Nadav Amit authored
As described in: 77b0bf55: ("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 - which is also a minor cleanup for the jump-label code. As a result the code size is slightly increased, but inlining decisions are better: text data bss dec hex filename 18163528 10226300 2957312 31347140 1de51c4 ./vmlinux before 18163608 10227348 2957312 31348268 1de562c ./vmlinux after (+1128) And functions such as intel_pstate_adjust_policy_max(), kvm_cpu_accept_dm_intr(), kvm_register_readl() are inlined. Tested-by:
Kees Cook <keescook@chromium.org> Signed-off-by:
Nadav Amit <namit@vmware.com> Acked-by:
-
Nadav Amit authored
As described in: 77b0bf55: ("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 - which is pretty pointless indirection in the static_cpu_has() case, but is worth it to improve overall inlining quality. The patch slightly increases the kernel size: text data bss dec hex filename 18162879 10226256 2957312 31346447 1de4f0f ./vmlinux before 18163528 10226300 2957312 31347140 1de51c4 ./vmlinux after (+693) And enables the inlining of function such as free_ldt_pgtables(). Tested-by:
-
Nadav Amit authored
As described in: 77b0bf55: ("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 - which is also a minor cleanup for the exception table code. Text size goes up a bit: text data bss dec hex filename 18162555 10226288 2957312 31346155 1de4deb ./vmlinux before 18162879 10226256 2957312 31346447 1de4f0f ./vmlinux after (+292) But this allows the inlining of functions such as nested_vmx_exit_reflected(), set_segment_reg(), __copy_xstate_to_user() which is a net benefit. Tested-by:
-
Ingo Molnar authored
Signed-off-by:
-
- 05 Oct, 2018 1 commit
-
-
Ingo Molnar authored
Signed-off-by:
-
- 04 Oct, 2018 8 commits
-
-
Nadav Amit authored
As described in: 77b0bf55: ("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.) In this patch we wrap the paravirt call section tricks in a macro, to hide it from GCC. The effect of the patch is a more aggressive inlining, which also causes a size increase of kernel. text data bss dec hex filename 18147336 10226688 2957312 31331336 1de1408 ./vmlinux before 18162555 10226288 2957312 31346155 1de4deb ./vmlinux after (+14819) The number of static text symbols (non-inlined functions) goes down: Before: 40053 After: 39942 (-111) [ mingo: Rewrote the changelog. ] Tested-by:
Juergen Gross <jgross@suse.com> Acked-by:
-
Nadav Amit authored
As described in: 77b0bf55: ("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:
-
Nadav Amit authored
As described in: 77b0bf55: ("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 - i.e. to macrify the affected block. As a result GCC considers the inline assembly block as a single instruction. This patch handles the LOCK prefix, allowing more aggresive inlining: text data bss dec hex filename 18140140 10225284 2957312 31322736 1ddf270 ./vmlinux before 18146889 10225380 2957312 31329581 1de0d2d ./vmlinux after (+6845) This is the reduction in non-inlined functions: Before: 40286 After: 40218 (-68) Tested-by:
-
Nadav Amit authored
As described in: 77b0bf55: ("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 allows GCC to inline simple functions such as __get_seccomp_filter(). To no-one's surprise the result is that GCC performs more aggressive (read: correct) inlining decisions in these senarios, which reduces the kernel size and presumably also speeds it up: text data bss dec hex filename 18140970 10225412 2957312 31323694 1ddf62e ./vmlinux before 18140140 10225284 2957312 31322736 1ddf270 ./vmlinux after (-958) 16 fewer static text symbols: Before: 40302 After: 40286 (-16) these got inlined instead. Functions such as kref_get(), free_user(), fuse_file_get() now get inlined. Hurray! [ mingo: Rewrote the changelog. ] Tested-by:
-
Nadav Amit authored
As described in: 77b0bf55: ("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. In the case of objtool the resulting borkage can be significant, since all the annotations of objtool are discarded during linkage and never inlined, yet GCC bogusly considers most functions affected by objtool annotations as 'too large'. 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 increases the kernel size slightly: text data bss dec hex filename 18140829 10224724 2957312 31322865 1ddf2f1 ./vmlinux before 18140970 10225412 2957312 31323694 1ddf62e ./vmlinux after (+829) The number of static text symbols (i.e. non-inlined functions) is reduced: Before: 40321 After: 40302 (-19) [ mingo: Rewrote the changelog. ] Tested-by:
Josh Poimboeuf <jpoimboe@redhat.com> Acked-by:
-
Nadav Amit authored
kbuild/Makefile: Prepare for using macros in inline assembly code to work around asm() related GCC inlining bugs Using macros in inline assembly allows us to work around bugs in GCC's inlining decisions. Compile macros.S and use it to assemble all C files. Currently only x86 will use it. Background: The inlining pass of GCC doesn't include an assembler, so it's not aware of basic properties of the generated code, such as its size in bytes, or that there are such things as discontiuous blocks of code and data due to the newfangled linker feature called 'sections' ... Instead GCC uses a lazy and fragile heuristic: it does a linear count of certain syntactic and whitespace elements in inlined assembly block source code, such as a count of new-lines and semicolons (!), as a poor substitute for "code size and complexity". Unsurprisingly this heuristic falls over and breaks its neck whith certain common types of kernel code that use inline assembly, such as the frequent practice of putting useful information into alternative sections. As a result of this fresh, 20+ years old GCC bug, GCC's inlining decisions are effectively disabled for inlined functions that make use of such asm() blocks, because GCC thinks those sections of code are "large" - when in reality they are often result in just a very low number of machine instructions. This absolute lack of inlining provess when GCC comes across such asm() blocks both increases generated kernel code size and causes performance overhead, which is particularly noticeable on paravirt kernels, which make frequent use of these inlining facilities in attempt to stay out of the way when running on baremetal hardware. Instead of fixing the compiler we use a workaround: we set an assembly macro and call it from the inlined assembly block. As a result GCC considers the inline assembly block as a single instruction. (Which it often isn't but I digress.) This uglifies and bloats the source code - for example just the refcount related changes have this impact: Makefile | 9 +++++++-- arch/x86/Makefile | 7 +++++++ arch/x86/kernel/macros.S | 7 +++++++ scripts/Kbuild.include | 4 +++- scripts/mod/Makefile | 2 ++ 5 files changed, 26 insertions(+), 3 deletions(-) Yay readability and maintainability, it's not like assembly code is hard to read and maintain ... We also hope that GCC will eventually get fixed, but we are not holding our breath for that. Yet we are optimistic, it might still happen, any decade now. [ mingo: Wrote new changelog describing the background. ] Tested-by:
Masahiro Yamada <yamada.masahiro@socionext.com> Acked-by:
-
Nadav Amit authored
Define the LINKER_SCRIPT when building the linker script as being done in other architectures. This is required, because upcoming Makefile changes would otherwise break things. Signed-off-by:
Max Filippov <jcmvbkbc@gmail.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Chris Zankel <chris@zankel.net> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Masahiro Yamada <yamada.masahiro@socionext.com> Cc: Michal Marek <michal.lkml@markovi.net> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-xtensa@linux-xtensa.org Link: http://lkml.kernel.org/r/20181003213100.189959-2-namit@vmware.comSigned-off-by:
-
Ingo Molnar authored
Signed-off-by:
-
- 03 Oct, 2018 3 commits
-
-
Waiman Long authored
When __lock_release() is called, the most likely unlock scenario is on the innermost lock in the chain. In this case, we can skip some of the checks and provide a faster path to completion. Signed-off-by:
-
Waiman Long authored
The static __lock_acquire() function has only two callers: 1) lock_acquire() 2) reacquire_held_locks() In lock_acquire(), raw_local_irq_save() is called beforehand. So IRQs must have been disabled. So the check: DEBUG_LOCKS_WARN_ON(!irqs_disabled()) is kind of redundant in this case. So move the above check to reacquire_held_locks() to eliminate redundant code in the lock_acquire() path. Signed-off-by:
-
Waiman Long authored
The inline function add_chain_cache_classes() is defined, but has no caller. Just remove it. Signed-off-by:
-
- 02 Oct, 2018 11 commits
-
-
https://github.com/bzolnier/linuxGreg Kroah-Hartman authored
Bartlomiej writes: "fbdev fixes for v4.19-rc7: - fix OMAPFB_MEMORY_READ ioctl to not leak kernel memory in omapfb driver (Tomi Valkeinen) - add missing prepare/unprepare clock operations in pxa168fb driver (Lubomir Rintel) - add nobgrt option in efifb driver to disable ACPI BGRT logo restore (Hans de Goede) - fix spelling mistake in fall-through annotation in stifb driver (Gustavo A. R. Silva) - fix URL for uvesafb repository in the documentation (Adam Jackson)" * tag 'fbdev-v4.19-rc7' of https://github.com/bzolnier/linux: video/fbdev/stifb: Fix spelling mistake in fall-through annotation uvesafb: Fix URLs in the documentation efifb: BGRT: Add nobgrt option fbdev/omapfb: fix omapfb_memory_read infoleak pxa168fb: prepare the clock -
git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmcGreg Kroah-Hartman authored
Ulf writes: "MMC core: - Fixup conversion of debounce time to/from ms/us MMC host: - sdhi: Fixup whitelisting for Gen3 types" * tag 'mmc-v4.19-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/ulfh/mmc: mmc: slot-gpio: Fix debounce time to use miliseconds again mmc: core: Fix debounce time to use microseconds mmc: sdhi: sys_dmac: check for all Gen3 types when whitelisting -
Andrew Murray authored
Fix incorrect line number in example output Signed-off-by:
Andrew Murray <andrew.murray@arm.com> Cc: Jiri Kosina <trivial@kernel.org> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Will Deacon <will.deacon@arm.com> Cc: linux-doc@vger.kernel.org Link: http://lkml.kernel.org/r/1538391663-54524-1-git-send-email-andrew.murray@arm.comSigned-off-by:
-
Andrea Parri authored
Amend the changes in commit: 1f03e8d2 ("locking/barriers: Replace smp_cond_acquire() with smp_cond_load_acquire()") ... by updating the documentation accordingly. Also remove some obsolete information related to the implementation. Signed-off-by:
Andrea Parri <andrea.parri@amarulasolutions.com> Signed-off-by:
Paul E. McKenney <paulmck@linux.vnet.ibm.com> Acked-by:
Alan Stern <stern@rowland.harvard.edu> Cc: Akira Yokosawa <akiyks@gmail.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Boqun Feng <boqun.feng@gmail.com> Cc: Daniel Lustig <dlustig@nvidia.com> Cc: David Howells <dhowells@redhat.com> Cc: Jade Alglave <j.alglave@ucl.ac.uk> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Luc Maranget <luc.maranget@inria.fr> Cc: Nicholas Piggin <npiggin@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Vince Weaver <vincent.weaver@maine.edu> Cc: linux-arch@vger.kernel.org Cc: parri.andrea@gmail.com Link: http://lkml.kernel.org/r/20180926182920.27644-5-paulmck@linux.ibm.comSigned-off-by:
-
Paul E. McKenney authored
This commit adds more detail about compiler optimizations and not-yet-modeled Linux-kernel APIs. Signed-off-by:
-
SeongJae Park authored
This commit fixes a duplicate-"the" typo in README. Signed-off-by:
SeongJae Park <sj38.park@gmail.com> Signed-off-by:
-
Alan Stern authored
More than one kernel developer has expressed the opinion that the LKMM should enforce ordering of writes by locking. In other words, given the following code: WRITE_ONCE(x, 1); spin_unlock(&s): spin_lock(&s); WRITE_ONCE(y, 1); the stores to x and y should be propagated in order to all other CPUs, even though those other CPUs might not access the lock s. In terms of the memory model, this means expanding the cumul-fence relation. Locks should also provide read-read (and read-write) ordering in a similar way. Given: READ_ONCE(x); spin_unlock(&s); spin_lock(&s); READ_ONCE(y); // or WRITE_ONCE(y, 1); the load of x should be executed before the load of (or store to) y. The LKMM already provides this ordering, but it provides it even in the case where the two accesses are separated by a release/acquire pair of fences rather than unlock/lock. This would prevent architectures from using weakly ordered implementations of release and acquire, which seems like an unnecessary restriction. The patch therefore removes the ordering requirement from the LKMM for that case. There are several arguments both for and against this change. Let us refer to these enhanced ordering properties by saying that the LKMM would require locks to be RCtso (a bit of a misnomer, but analogous to RCpc and RCsc) and it would require ordinary acquire/release only to be RCpc. (Note: In the following, the phrase "all supported architectures" is meant not to include RISC-V. Although RISC-V is indeed supported by the kernel, the implementation is still somewhat in a state of flux and therefore statements about it would be premature.) Pros: The kernel already provides RCtso ordering for locks on all supported architectures, even though this is not stated explicitly anywhere. Therefore the LKMM should formalize it. In theory, guaranteeing RCtso ordering would reduce the need for additional barrier-like constructs meant to increase the ordering strength of locks. Will Deacon and Peter Zijlstra are strongly in favor of formalizing the RCtso requirement. Linus Torvalds and Will would like to go even further, requiring locks to have RCsc behavior (ordering preceding writes against later reads), but they recognize that this would incur a noticeable performance degradation on the POWER architecture. Linus also points out that people have made the mistake, in the past, of assuming that locking has stronger ordering properties than is currently guaranteed, and this change would reduce the likelihood of such mistakes. Not requiring ordinary acquire/release to be any stronger than RCpc may prove advantageous for future architectures, allowing them to implement smp_load_acquire() and smp_store_release() with more efficient machine instructions than would be possible if the operations had to be RCtso. Will and Linus approve this rationale, hypothetical though it is at the moment (it may end up affecting the RISC-V implementation). The same argument may or may not apply to RMW-acquire/release; see also the second Con entry below. Linus feels that locks should be easy for people to use without worrying about memory consistency issues, since they are so pervasive in the kernel, whereas acquire/release is much more of an "experts only" tool. Requiring locks to be RCtso is a step in this direction. Cons: Andrea Parri and Luc Maranget think that locks should have the same ordering properties as ordinary acquire/release (indeed, Luc points out that the names "acquire" and "release" derive from the usage of locks). Andrea points out that having different ordering properties for different forms of acquires and releases is not only unnecessary, it would also be confusing and unmaintainable. Locks are constructed from lower-level primitives, typically RMW-acquire (for locking) and ordinary release (for unlock). It is illogical to require stronger ordering properties from the high-level operations than from the low-level operations they comprise. Thus, this change would make while (cmpxchg_acquire(&s, 0, 1) != 0) cpu_relax(); an incorrect implementation of spin_lock(&s) as far as the LKMM is concerned. In theory this weakness can be ameliorated by changing the LKMM even further, requiring RMW-acquire/release also to be RCtso (which it already is on all supported architectures). As far as I know, nobody has singled out any examples of code in the kernel that actually relies on locks being RCtso. (People mumble about RCU and the scheduler, but nobody has pointed to any actual code. If there are any real cases, their number is likely quite small.) If RCtso ordering is not needed, why require it? A handful of locking constructs (qspinlocks, qrwlocks, and mcs_spinlocks) are built on top of smp_cond_load_acquire() instead of an RMW-acquire instruction. It currently provides only the ordinary acquire semantics, not the stronger ordering this patch would require of locks. In theory this could be ameliorated by requiring smp_cond_load_acquire() in combination with ordinary release also to be RCtso (which is currently true on all supported architectures). On future weakly ordered architectures, people may be able to implement locks in a non-RCtso fashion with significant performance improvement. Meeting the RCtso requirement would necessarily add run-time overhead. Overall, the technical aspects of these arguments seem relatively minor, and it appears mostly to boil down to a matter of opinion. Since the opinions of senior kernel maintainers such as Linus, Peter, and Will carry more weight than those of Luc and Andrea, this patch changes the model in accordance with the maintainers' wishes. Signed-off-by:
-
Paul E. McKenney authored
This commit documents the scheme used to generate the names for the litmus tests. [ paulmck: Apply feedback from Andrea Parri and Will Deacon. ] Signed-off-by:
-
Matthew Wilcox authored
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:
Matthew Wilcox <willy@infradead.org> Signed-off-by:
-
Ard Biesheuvel authored
Commit 19483677 ("jump_label: Annotate entries that operate on __init code earlier") refactored the code that manages runtime patching of jump labels in modules that are tied to static keys defined in other modules or in the core kernel. In the latter case, we may iterate over the static_key_mod linked list until we hit the entry for the core kernel, whose 'mod' field will be NULL, and attempt to dereference it to get at its 'state' member. So let's add a non-NULL check: this forces the 'init' argument of __jump_label_update() to false for static keys that are defined in the core kernel, which is appropriate given that __init annotated jump_label entries in the core kernel should no longer be active at this point (i.e., when loading modules). Fixes: 19483677 ("jump_label: Annotate entries that operate on ...") Reported-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
Ard Biesheuvel <ard.biesheuvel@linaro.org> Signed-off-by:
-
Ard Biesheuvel authored
Commit e872267b ("jump_table: move entries into ro_after_init region") moved the __jump_table input section into the __ro_after_init output section, but inadvertently put the macro in the wrong place in the s390 linker script. Let's fix that. Fixes: e872267b ("jump_table: move entries into ro_after_init region") Reported-by:
Guenter Roeck <linux@roeck-us.net> Signed-off-by:
Heiko Carstens <heiko.carstens@de.ibm.com> Tested-by:
-
