linux-toradex.git/kernel/events/hw_breakpoint.c, branch v5.7-rc7

perf/hw_breakpoint: Fix arch_hw_breakpoint use-before-initialization

2019-09-06T06:24:01+00:00

If we disable the compiler's auto-initialization feature, if
-fplugin-arg-structleak_plugin-byref or -ftrivial-auto-var-init=pattern
are disabled, arch_hw_breakpoint may be used before initialization after:

  9a4903dde2c86 ("perf/hw_breakpoint: Split attribute parse and commit")

On our ARM platform, the struct step_ctrl in arch_hw_breakpoint, which
used to be zero-initialized by kzalloc(), may be used in
arch_install_hw_breakpoint() without initialization.

Signed-off-by: Mark-PK Tsai 
Cc: Alexander Shishkin 
Cc: Alix Wu 
Cc: Arnaldo Carvalho de Melo 
Cc: Jiri Olsa 
Cc: Linus Torvalds 
Cc: Mark Rutland 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Cc: YJ Chiang 
Link: https://lkml.kernel.org/r/20190906060115.9460-1-mark-pk.tsai@mediatek.com
[ Minor edits. ]
Signed-off-by: Ingo Molnar

perf/hw_breakpoints: Convert to SPDX license identifier

2019-02-04T07:46:13+00:00

Replace the license boiler plate with a SPDX license identifier.

Signed-off-by: Thomas Gleixner 
Signed-off-by: Peter Zijlstra (Intel) 
Acked-by: Paul McKenney 
Cc: Alan Stern 
Cc: Alexander Shishkin 
Cc: Arnaldo Carvalho de Melo 
Cc: Frederic Weisbecker 
Cc: Greg Kroah-Hartman 
Cc: Jiri Olsa 
Cc: Kate Stewart 
Cc: Linus Torvalds 
Cc: Peter Zijlstra 
Link: https://lkml.kernel.org/r/20190116111308.105855650@linutronix.de
Signed-off-by: Ingo Molnar

perf: Fix typos in comments

2018-12-03T10:22:32+00:00

Fix two typos in kernel/events/*.

No change in functionality intended.

Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Cc: Linus Torvalds 
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar

perf/hw_breakpoint: Enable breakpoint in modify_user_hw_breakpoint

2018-08-30T17:49:23+00:00

Currently we enable the breakpoint back only if the breakpoint
modification was successful. If it fails we can leave the breakpoint in
disabled state with attr->disabled == 0.

We can safely enable the breakpoint back for both the fail and success
paths by checking the bp->attr.disabled, which either holds the new
'requested' disabled state or the original breakpoint state.

Suggested-by: Oleg Nesterov 
Signed-off-by: Jiri Olsa 
Acked-by: Frederic Weisbecker 
Acked-by: Oleg Nesterov 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Alexander Shishkin 
Cc: David Ahern 
Cc: Milind Chabbi 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/20180827091228.2878-5-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf/hw_breakpoint: Remove superfluous bp->attr.disabled = 0

2018-08-30T17:49:23+00:00

Once the breakpoint was succesfully modified, the attr->disabled value
is in bp->attr.disabled. So there's no reason to set it again, removing
that.

Signed-off-by: Jiri Olsa 
Acked-by: Frederic Weisbecker 
Acked-by: Oleg Nesterov 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Alexander Shishkin 
Cc: David Ahern 
Cc: Milind Chabbi 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/20180827091228.2878-4-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf/hw_breakpoint: Modify breakpoint even if the new attr has disabled set

2018-08-30T17:49:23+00:00

We need to change the breakpoint even if the attr with new fields has
disabled set to true.

Current code prevents following user code to change the breakpoint
address:

  ptrace(PTRACE_POKEUSER, child, offsetof(struct user, u_debugreg[0]), addr_1)
  ptrace(PTRACE_POKEUSER, child, offsetof(struct user, u_debugreg[0]), addr_2)
  ptrace(PTRACE_POKEUSER, child, offsetof(struct user, u_debugreg[7]), dr7)

The first PTRACE_POKEUSER creates the breakpoint with attr.disabled set
to true:

  ptrace_set_breakpoint_addr(nr = 0)
    struct perf_event *bp = t->ptrace_bps[nr];

    ptrace_register_breakpoint(..., disabled = true)
      ptrace_fill_bp_fields(..., disabled)
      register_user_hw_breakpoint

So the second PTRACE_POKEUSER will be omitted:

  ptrace_set_breakpoint_addr(nr = 0)
    struct perf_event *bp = t->ptrace_bps[nr];
    struct perf_event_attr attr = bp->attr;

    modify_user_hw_breakpoint(bp, &attr)
      if (!attr->disabled)
        modify_user_hw_breakpoint_check

Reported-by: Milind Chabbi 
Signed-off-by: Jiri Olsa 
Acked-by: Frederic Weisbecker 
Acked-by: Oleg Nesterov 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Alexander Shishkin 
Cc: David Ahern 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Link: http://lkml.kernel.org/r/20180827091228.2878-3-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf/hw_breakpoint: Clean up and consolidate modify_user_hw_breakpoint_check()

2018-06-26T07:07:59+00:00

Remove the dance around old and new attributes. Just don't modify the
previous breakpoint at all until we have verified everything.

Original-patch-by: Andy Lutomirski 
Reported-by: Linus Torvalds 
Signed-off-by: Frederic Weisbecker 
Cc: Alexander Shishkin 
Cc: Andy Lutomirski 
Cc: Arnaldo Carvalho de Melo 
Cc: Arnaldo Carvalho de Melo 
Cc: Benjamin Herrenschmidt 
Cc: Catalin Marinas 
Cc: Chris Zankel 
Cc: Jiri Olsa 
Cc: Joel Fernandes 
Cc: Mark Rutland 
Cc: Max Filippov 
Cc: Michael Ellerman 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Rich Felker 
Cc: Thomas Gleixner 
Cc: Will Deacon 
Cc: Yoshinori Sato 
Link: http://lkml.kernel.org/r/1529981939-8231-13-git-send-email-frederic@kernel.org
Signed-off-by: Ingo Molnar

perf/hw_breakpoint: Pass new breakpoint type to modify_breakpoint_slot()

2018-06-26T07:07:59+00:00

We soon won't be able to rely on bp->attr anymore to get the new
type of the modifying breakpoint because the new attributes are going
to be copied only once we successfully modified the breakpoint slot.

This will fix the current misdesigned layout where the new attr are
copied to the modifying breakpoint before we actually know if the
modification will be validated.

In order to prepare for that, allow modify_breakpoint_slot() to take
the new breakpoint type.

Signed-off-by: Frederic Weisbecker 
Cc: Alexander Shishkin 
Cc: Andy Lutomirski 
Cc: Arnaldo Carvalho de Melo 
Cc: Arnaldo Carvalho de Melo 
Cc: Benjamin Herrenschmidt 
Cc: Catalin Marinas 
Cc: Chris Zankel 
Cc: Jiri Olsa 
Cc: Joel Fernandes 
Cc: Linus Torvalds 
Cc: Mark Rutland 
Cc: Max Filippov 
Cc: Michael Ellerman 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Rich Felker 
Cc: Thomas Gleixner 
Cc: Will Deacon 
Cc: Yoshinori Sato 
Link: http://lkml.kernel.org/r/1529981939-8231-12-git-send-email-frederic@kernel.org
Signed-off-by: Ingo Molnar

perf/hw_breakpoint: Remove default hw_breakpoint_arch_parse()

2018-06-26T07:07:58+00:00

All architectures have implemented it, we can now remove the poor weak
version.

Signed-off-by: Frederic Weisbecker 
Cc: Alexander Shishkin 
Cc: Andy Lutomirski 
Cc: Arnaldo Carvalho de Melo 
Cc: Arnaldo Carvalho de Melo 
Cc: Benjamin Herrenschmidt 
Cc: Catalin Marinas 
Cc: Chris Zankel 
Cc: Jiri Olsa 
Cc: Joel Fernandes 
Cc: Linus Torvalds 
Cc: Mark Rutland 
Cc: Max Filippov 
Cc: Michael Ellerman 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Rich Felker 
Cc: Thomas Gleixner 
Cc: Will Deacon 
Cc: Yoshinori Sato 
Link: http://lkml.kernel.org/r/1529981939-8231-11-git-send-email-frederic@kernel.org
Signed-off-by: Ingo Molnar

perf/hw_breakpoint: Pass arch breakpoint struct to arch_check_bp_in_kernelspace()

2018-06-26T07:07:54+00:00

We can't pass the breakpoint directly on arch_check_bp_in_kernelspace()
anymore because its architecture internal datas (struct arch_hw_breakpoint)
are not yet filled by the time we call the function, and most
implementation need this backend to be up to date. So arrange the
function to take the probing struct instead.

Signed-off-by: Frederic Weisbecker 
Cc: Alexander Shishkin 
Cc: Andy Lutomirski 
Cc: Arnaldo Carvalho de Melo 
Cc: Arnaldo Carvalho de Melo 
Cc: Benjamin Herrenschmidt 
Cc: Catalin Marinas 
Cc: Chris Zankel 
Cc: Jiri Olsa 
Cc: Joel Fernandes 
Cc: Linus Torvalds 
Cc: Mark Rutland 
Cc: Max Filippov 
Cc: Michael Ellerman 
Cc: Namhyung Kim 
Cc: Paul Mackerras 
Cc: Peter Zijlstra 
Cc: Rich Felker 
Cc: Thomas Gleixner 
Cc: Will Deacon 
Cc: Yoshinori Sato 
Link: http://lkml.kernel.org/r/1529981939-8231-3-git-send-email-frederic@kernel.org
Signed-off-by: Ingo Molnar