diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2018-08-13 22:55:49 +0300 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2018-08-13 22:55:49 +0300 |
commit | 8603596a327c978534f5c45db135e6c36b4b1425 (patch) | |
tree | 7270af18e4d1c42986672eb5673d55c7dfa678cf /Documentation | |
parent | de5d1b39ea0b38a9f4dfb08966042b7b91e2df30 (diff) | |
parent | ec2cb7a526d49b65576301e183448fb51ee543a6 (diff) | |
download | linux-8603596a327c978534f5c45db135e6c36b4b1425.tar.xz |
Merge branch 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull perf update from Thomas Gleixner:
"The perf crowd presents:
Kernel updates:
- Removal of jprobes
- Cleanup and consolidatation the handling of kprobes
- Cleanup and consolidation of hardware breakpoints
- The usual pile of fixes and updates to PMUs and event descriptors
Tooling updates:
- Updates and improvements all over the place. Nothing outstanding,
just the (good) boring incremental grump work"
* 'perf-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (103 commits)
perf trace: Do not require --no-syscalls to suppress strace like output
perf bpf: Include uapi/linux/bpf.h from the 'perf trace' script's bpf.h
perf tools: Allow overriding MAX_NR_CPUS at compile time
perf bpf: Show better message when failing to load an object
perf list: Unify metric group description format with PMU event description
perf vendor events arm64: Update ThunderX2 implementation defined pmu core events
perf cs-etm: Generate branch sample for CS_ETM_TRACE_ON packet
perf cs-etm: Generate branch sample when receiving a CS_ETM_TRACE_ON packet
perf cs-etm: Support dummy address value for CS_ETM_TRACE_ON packet
perf cs-etm: Fix start tracing packet handling
perf build: Fix installation directory for eBPF
perf c2c report: Fix crash for empty browser
perf tests: Fix indexing when invoking subtests
perf trace: Beautify the AF_INET & AF_INET6 'socket' syscall 'protocol' args
perf trace beauty: Add beautifiers for 'socket''s 'protocol' arg
perf trace beauty: Do not print NULL strarray entries
perf beauty: Add a generator for IPPROTO_ socket's protocol constants
tools include uapi: Grab a copy of linux/in.h
perf tests: Fix complex event name parsing
perf evlist: Fix error out while applying initial delay and LBR
...
Diffstat (limited to 'Documentation')
-rw-r--r-- | Documentation/kprobes.txt | 35 |
1 files changed, 27 insertions, 8 deletions
diff --git a/Documentation/kprobes.txt b/Documentation/kprobes.txt index cb3b0de83fc6..10f4499e677c 100644 --- a/Documentation/kprobes.txt +++ b/Documentation/kprobes.txt @@ -80,6 +80,26 @@ After the instruction is single-stepped, Kprobes executes the "post_handler," if any, that is associated with the kprobe. Execution then continues with the instruction following the probepoint. +Changing Execution Path +----------------------- + +Since kprobes can probe into a running kernel code, it can change the +register set, including instruction pointer. This operation requires +maximum care, such as keeping the stack frame, recovering the execution +path etc. Since it operates on a running kernel and needs deep knowledge +of computer architecture and concurrent computing, you can easily shoot +your foot. + +If you change the instruction pointer (and set up other related +registers) in pre_handler, you must return !0 so that kprobes stops +single stepping and just returns to the given address. +This also means post_handler should not be called anymore. + +Note that this operation may be harder on some architectures which use +TOC (Table of Contents) for function call, since you have to setup a new +TOC for your function in your module, and recover the old one after +returning from it. + Return Probes ------------- @@ -262,7 +282,7 @@ is optimized, that modification is ignored. Thus, if you want to tweak the kernel's execution path, you need to suppress optimization, using one of the following techniques: -- Specify an empty function for the kprobe's post_handler or break_handler. +- Specify an empty function for the kprobe's post_handler. or @@ -474,7 +494,7 @@ error occurs during registration, all probes in the array, up to the bad probe, are safely unregistered before the register_*probes function returns. -- kps/rps/jps: an array of pointers to ``*probe`` data structures +- kps/rps: an array of pointers to ``*probe`` data structures - num: the number of the array entries. .. note:: @@ -566,12 +586,11 @@ the same handler) may run concurrently on different CPUs. Kprobes does not use mutexes or allocate memory except during registration and unregistration. -Probe handlers are run with preemption disabled. Depending on the -architecture and optimization state, handlers may also run with -interrupts disabled (e.g., kretprobe handlers and optimized kprobe -handlers run without interrupt disabled on x86/x86-64). In any case, -your handler should not yield the CPU (e.g., by attempting to acquire -a semaphore). +Probe handlers are run with preemption disabled or interrupt disabled, +which depends on the architecture and optimization state. (e.g., +kretprobe handlers and optimized kprobe handlers run without interrupt +disabled on x86/x86-64). In any case, your handler should not yield +the CPU (e.g., by attempting to acquire a semaphore, or waiting I/O). Since a return probe is implemented by replacing the return address with the trampoline's address, stack backtraces and calls |