[pull] master from torvalds:master#1943
Merged
pull[bot] merged 8 commits intor4b3rt:masterfrom Nov 29, 2024
Merged
Conversation
The scheduler added NEED_RESCHED_LAZY scheduling. Record this state as part of trace flags and expose it in the need_resched field. Record and expose NEED_RESCHED_LAZY. [bigeasy: Commit description, documentation bits.] Cc: Peter Zijlstra <peterz@infradead.org> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Link: https://lore.kernel.org/20241122202849.7DfYpJR0@linutronix.de Reviewed-by: Ankur Arora <ankur.a.arora@oracle.com> Reviewed-by: Steven Rostedt (Google) <rostedt@goodmis.org> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
When introducing __DO_TRACE_CALL(), the iteration over it_func moved from __DO_TRACE() to __tracepoint_iter_##_name(), but the comment relevant for this iterator was left in its original location. Move the comment to the relevant context. Fixes: d25e37d ("tracepoint: Optimize using static_call()") Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Michael Jeanson <mjeanson@efficios.com> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Alexei Starovoitov <ast@kernel.org> Cc: Yonghong Song <yhs@fb.com> Cc: Paul E. McKenney <paulmck@kernel.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Andrii Nakryiko <andrii.nakryiko@gmail.com> Cc: bpf@vger.kernel.org Cc: Joel Fernandes <joel@joelfernandes.org> Cc: Jordan Rife <jrife@google.com> Link: https://lore.kernel.org/20241123153031.2884933-2-mathieu.desnoyers@efficios.com Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Since the removal of SRCU-protected tracepoints, the __idx variable in __DO_TRACE is unused. Remove this variable. Fixes: 48bcda6 ("tracing: Remove definition of trace_*_rcuidle()") Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Michael Jeanson <mjeanson@efficios.com> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Alexei Starovoitov <ast@kernel.org> Cc: Yonghong Song <yhs@fb.com> Cc: Paul E. McKenney <paulmck@kernel.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Andrii Nakryiko <andrii.nakryiko@gmail.com> Cc: bpf@vger.kernel.org Cc: Joel Fernandes <joel@joelfernandes.org> Cc: Jordan Rife <jrife@google.com> Link: https://lore.kernel.org/20241123153031.2884933-3-mathieu.desnoyers@efficios.com Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Define a rcu_tasks_trace lock guard for use by the syscall enter/exit tracepoints. Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Michael Jeanson <mjeanson@efficios.com> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Alexei Starovoitov <ast@kernel.org> Cc: Yonghong Song <yhs@fb.com> Cc: Paul E. McKenney <paulmck@kernel.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Andrii Nakryiko <andrii.nakryiko@gmail.com> Cc: bpf@vger.kernel.org Cc: Joel Fernandes <joel@joelfernandes.org> Cc: Jordan Rife <jrife@google.com> Cc: linux-trace-kernel@vger.kernel.org Link: https://lore.kernel.org/20241123153031.2884933-4-mathieu.desnoyers@efficios.com Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Remove conditional locking by moving the __DO_TRACE() code into trace_##name(). When the faultable syscall tracepoints were implemented, __DO_TRACE() had a rcuidle argument which selected between SRCU and preempt disable. Therefore, the RCU tasks trace protection for faultable syscall tracepoints was introduced using the same pattern. At that point, it did not appear obvious that this feedback from Linus [1] applied here as well, because the __DO_TRACE() modification was extending a pre-existing pattern. Shortly before pulling the faultable syscall tracepoints modifications, Steven removed the rcuidle argument and SRCU protection scheme entirely from tracepoint.h: commit 48bcda6 ("tracing: Remove definition of trace_*_rcuidle()") This required a rebase of the faultable syscall tracepoints series, which missed a perfect opportunity to integrate the prior recommendation from Linus. In response to the pull request, Linus pointed out [2] that he was not pleased by the implementation, expecting this to be fixed in a follow up patch series. Move __DO_TRACE() code into trace_##name() within each of __DECLARE_TRACE() and __DECLARE_TRACE_SYSCALL(). Use a scoped guard to guard the preempt disable notrace and RCU tasks trace critical sections. Link: https://lore.kernel.org/all/CAHk-=wggDLDeTKbhb5hh--x=-DQd69v41137M72m6NOTmbD-cw@mail.gmail.com/ [1] Link: https://lore.kernel.org/lkml/CAHk-=witPrLcu22dZ93VCyRQonS7+-dFYhQbna=KBa-TAhayMw@mail.gmail.com/ [2] Fixes: a363d27 ("tracing: Allow system call tracepoints to handle page faults") Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Michael Jeanson <mjeanson@efficios.com> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Alexei Starovoitov <ast@kernel.org> Cc: Yonghong Song <yhs@fb.com> Cc: Paul E. McKenney <paulmck@kernel.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Andrii Nakryiko <andrii.nakryiko@gmail.com> Cc: bpf@vger.kernel.org Cc: Joel Fernandes <joel@joelfernandes.org> Cc: Jordan Rife <jrife@google.com> Cc: linux-trace-kernel@vger.kernel.org Link: https://lore.kernel.org/20241123153031.2884933-5-mathieu.desnoyers@efficios.com Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Syscall tracepoints do not require a "cond" argument, because they are meant to be used only for sys_enter and sys_exit instrumentation, which don't require condition evaluation. Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Michael Jeanson <mjeanson@efficios.com> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Alexei Starovoitov <ast@kernel.org> Cc: Yonghong Song <yhs@fb.com> Cc: Paul E. McKenney <paulmck@kernel.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Andrii Nakryiko <andrii.nakryiko@gmail.com> Cc: bpf@vger.kernel.org Cc: Joel Fernandes <joel@joelfernandes.org> Cc: Jordan Rife <jrife@google.com> Cc: linux-trace-kernel@vger.kernel.org Link: https://lore.kernel.org/20241123153031.2884933-6-mathieu.desnoyers@efficios.com Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Using scoped_guard() in the implementation of trace_##name() adds an unnecessary level of indentation. Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Michael Jeanson <mjeanson@efficios.com> Cc: Masami Hiramatsu <mhiramat@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Alexei Starovoitov <ast@kernel.org> Cc: Yonghong Song <yhs@fb.com> Cc: Paul E. McKenney <paulmck@kernel.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Andrii Nakryiko <andrii.nakryiko@gmail.com> Cc: bpf@vger.kernel.org Cc: Joel Fernandes <joel@joelfernandes.org> Cc: Jordan Rife <jrife@google.com> Link: https://lore.kernel.org/20241125142514.2897143-1-mathieu.desnoyers@efficios.com Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
…l/git/trace/linux-trace Pull more tracing updates from Steven Rostedt: - Add trace flag for NEED_RESCHED_LAZY Now that NEED_RESCHED_LAZY is upstream, add it to the status bits of the common_flags. This will now show when the NEED_RESCHED_LAZY flag is set that is used for debugging latency issues in the kernel via a trace. - Remove leftover "__idx" variable when SRCU was removed from the tracepoint code - Add rcu_tasks_trace guard To add a guard() around the tracepoint code, a rcu_tasks_trace guard needs to be created first. - Remove __DO_TRACE() macro and just call __DO_TRACE_CALL() directly The DO_TRACE() macro has conditional locking depending on what was passed into the macro parameters. As the guts of the macro has been moved to __DO_TRACE_CALL() to handle static call logic, there's no reason to keep the __DO_TRACE() macro around. It is better to just do the locking in place without the conditionals and call __DO_TRACE_CALL() from those locations. The "cond" passed in can also be moved out of that macro. This simplifies the code. - Remove the "cond" from the system call tracepoint macros The "cond" variable was added to allow some tracepoints to check a condition within the static_branch (jump/nop) logic. The system calls do not need this. Removing it simplifies the code. - Replace scoped_guard() with just guard() in the tracepoint logic guard() works just as well as scoped_guard() in the tracepoint logic and the scoped_guard() causes some issues. * tag 'trace-v6.13-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: tracing: Use guard() rather than scoped_guard() tracing: Remove cond argument from __DECLARE_TRACE_SYSCALL tracing: Remove conditional locking from __DO_TRACE() rcupdate_trace: Define rcu_tasks_trace lock guard tracing: Remove __idx variable from __DO_TRACE tracing: Move it_func[0] comment to the relevant context tracing: Record task flag NEED_RESCHED_LAZY.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
See Commits and Changes for more details.
Created by
pull[bot] (v2.0.0-alpha.1)
Can you help keep this open source service alive? 💖 Please sponsor : )