[PATCH RFC 0/3] kernel/events/uprobes: uprobe_write_opcode() rewrite

[David Hildenbrand]

Against mm/mm-unstable.

There is interest in supporting uprobes on hugetlb pages [1]. Having
uprobe_write_opcode() implement COW-breaking itself is really far from
optimal, and having to add hugetlb COW-breaking to implement hugetlb
support really does sound wrong.

Further, I think the current code doesn't really handle some things
properly (see patch #3) when replacing/zapping pages.

Let's rewrite it, to leave COW-breaking to the fault handler, and handle
registration/unregistration by temporarily unmapping the anonymous page,
modifying it, and mapping it again. We still have to implement
zapping of anonymous pages ourselves, unfortunately.

Note that we now won't have to allocate another anonymous folio when
unregistering (which will be beneficial for hugetlb as well), we can simply
modify the already-mapped one from the registration (if any). When
registering a uprobe, we'll first trigger a write fault to break COW to
then modify the already-mapped page.

Briefly sanity tested with perf:
  [root at localhost ~]# perf probe -x /usr/bin/bash -a main
  ...
  [root at localhost ~]# perf record -e probe_bash:main -aR sleep 10 &
  [1] 2196
  [root at localhost ~]# bash
  [root at localhost ~]# exit
  exit
  [root at localhost ~]# bash
  [root at localhost ~]# exit
  exit
  [root at localhost ~]# [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.287 MB perf.data (8 samples) ]
  ...
  [root at localhost ~]# perf report --stdio
  # To display the perf.data header info, please use --header/--header-only optio>
  #
  #
  # Total Lost Samples: 0
  #
  # Samples: 8  of event 'probe_bash:main'
  # Event count (approx.): 8
  #
  # Overhead  Command      Shared Object  Symbol  
  # ........  ...........  .............  ........
  #
      75.00%  grepconf.sh  bash           [.] main
      25.00%  bash         bash           [.] main
  ...

Are there any uprobe tests / benchmarks that are worth running?

[1] https://lkml.kernel.org/r/ZiK50qob9yl5e0Xz@bender.morinfr.org

Cc: Andrew Morton <akpm at linux-foundation.org>
Cc: Guillaume Morin <guillaume at morinfr.org>
Cc: Russell King <linux at armlinux.org.uk>
Cc: Peter Zijlstra <peterz at infradead.org>
Cc: Ingo Molnar <mingo at redhat.com>
Cc: Arnaldo Carvalho de Melo <acme at kernel.org>
Cc: Namhyung Kim <namhyung at kernel.org>
Cc: Mark Rutland <mark.rutland at arm.com>
Cc: Alexander Shishkin <alexander.shishkin at linux.intel.com>
Cc: Jiri Olsa <jolsa at kernel.org>
Cc: Ian Rogers <irogers at google.com>
Cc: Adrian Hunter <adrian.hunter at intel.com>
Cc: "Liang, Kan" <kan.liang at linux.intel.com>

David Hildenbrand (3):
  kernel/events/uprobes: pass VMA instead of MM to install_breakpoint()
  kernel/events/uprobes: pass VMA to set_swbp(), set_orig_insn() and
    uprobe_write_opcode()
  kernel/events/uprobes: uprobe_write_opcode() rewrite

 arch/arm/probes/uprobes/core.c |   4 +-
 include/linux/uprobes.h        |   6 +-
 kernel/events/uprobes.c        | 387 +++++++++++++++++++--------------
 3 files changed, 223 insertions(+), 174 deletions(-)


base-commit: 065d3634d60843b8e338d405b844cc7f2e5e1c66
-- 
2.45.1