linux-stable.git/tools/perf/util/annotate.c, branch linux-4.18.y

perf annotate: Fix parsing aarch64 branch instructions after objdump update

2018-10-10T06:55:59+00:00

[ Upstream commit 4e67b2a5df5d3f341776d12ee575e00ca3ef92de ]

Starting with binutils 2.28, aarch64 objdump adds comments to the
disassembly output to show the alternative names of a condition code
[1].

It is assumed that commas in objdump comments could occur in other
arches now or in the future, so this fix is arch-independent.

The fix could have been done with arm64 specific jump__parse and
jump__scnprintf functions, but the jump__scnprintf instruction would
have to have its comment character be a literal, since the scnprintf
functions cannot receive a struct arch easily.

This inconvenience also applies to the generic jump__scnprintf, which is
why we add a raw_comment pointer to struct ins_operands, so the __parse
function assigns it to be re-used by its corresponding __scnprintf
function.

Example differences in 'perf annotate --stdio2' output on an aarch64
perf.data file:

BEFORE: → b.cs   ffff200008133d1c   // b.hs, dffff7ecc47b
AFTER : ↓ b.cs   18c

BEFORE: → b.cc   ffff200008d8d9cc   // b.lo, b.ul, dffff727295b
AFTER : ↓ b.cc   31c

The branch target labels 18c and 31c also now appear in the output:

BEFORE:        add    x26, x29, #0x80
AFTER : 18c:   add    x26, x29, #0x80

BEFORE:        add    x21, x21, #0x8
AFTER : 31c:   add    x21, x21, #0x8

The Fixes: tag below is added so stable branches will get the update; it
doesn't necessarily mean that commit was broken at the time, rather it
didn't withstand the aarch64 objdump update.

Tested no difference in output for sample x86_64, power arch perf.data files.

[1] https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;a=commit;h=bb7eff5206e4795ac79c177a80fe9f4630aaf730

Signed-off-by: Kim Phillips 
Tested-by: Arnaldo Carvalho de Melo 
Cc: Alexander Shishkin 
Cc: Anton Blanchard 
Cc: Christian Borntraeger 
Cc: Jiri Olsa 
Cc: Mark Rutland 
Cc: Namhyung Kim 
Cc: Peter Zijlstra 
Cc: Ravi Bangoria 
Cc: Robin Murphy 
Cc: Taeung Song 
Cc: linux-arm-kernel@lists.infradead.org
Fixes: b13bbeee5ee6 ("perf annotate: Fix branch instruction with multiple operands")
Link: http://lkml.kernel.org/r/20180827125340.a2f7e291901d17cea05daba4@arm.com
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman

perf annotate: Properly interpret indirect call

2018-10-10T06:55:59+00:00

[ Upstream commit 1dc27f63303db58ce1b1a6932d1825305f86d574 ]

The patch changes the parsing of:

	callq  *0x8(%rbx)

from:

  0.26 │     → callq  *8

to:

  0.26 │     → callq  *0x8(%rbx)

in this case an address is followed by a register, thus one can't parse
only the address.

Committer testing:

1) run 'perf record sleep 10'
2) before applying the patch, run:

     perf annotate --stdio2 > /tmp/before

3) after applying the patch, run:

     perf annotate --stdio2 > /tmp/after

4) diff /tmp/before /tmp/after:
#  --- /tmp/before 2018-08-28 11:16:03.238384143 -0300
#  +++ /tmp/after  2018-08-28 11:15:39.335341042 -0300
#  @@ -13274,7 +13274,7 @@
#                ↓ jle    128
#                  hash_value = hash_table->hash_func (key);
#                  mov    0x8(%rsp),%rdi
#  -  0.91       → callq  *30
#  +  0.91       → callq  *0x30(%r12)
#                  mov    $0x2,%r8d
#                  cmp    $0x2,%eax
#                  node_hash = hash_table->hashes[node_index];
#  @@ -13848,7 +13848,7 @@
#                   mov    %r14,%rdi
#                   sub    %rbx,%r13
#                   mov    %r13,%rdx
#  -              → callq  *38
#  +              → callq  *0x38(%r15)
#                   cmp    %rax,%r13
#     1.91        ↓ je     240
#            1b4:   mov    $0xffffffff,%r13d
#  @@ -14026,7 +14026,7 @@
#                   mov    %rcx,-0x500(%rbp)
#                   mov    %r15,%rsi
#                   mov    %r14,%rdi
#  -              → callq  *38
#  +              → callq  *0x38(%rax)
#                   mov    -0x500(%rbp),%rcx
#                   cmp    %rax,%rcx
#                 ↓ jne    9b0


Signed-off-by: Martin Liška 
Tested-by: Arnaldo Carvalho de Melo 
Tested-by: Kim Phillips 
Cc: Jiri Olsa 
Link: http://lkml.kernel.org/r/bd1f3932-be2b-85f9-7582-111ee0a43b07@suse.cz
Signed-off-by: Arnaldo Carvalho de Melo 
Signed-off-by: Sasha Levin 
Signed-off-by: Greg Kroah-Hartman

perf annnotate: Make symbolinc_addr_samples handle src->histograms == NULL

2018-06-06T15:52:08+00:00

Making it a bit more robust, this took place here when a sample appeared
right after:

  ffffffff8a925000 D __nosave_end

And before the next considered symbol, which, using kallsyms make us
over guess the size of __nosave_end, and then the sequence:

  hist_entry__inc_addr_samples ->
    symbol__inc_addr_samples ->
      symbol__hists ->
        annotated_source__alloc_histograms

Ends up not liking to allocate gigabytes of ram for annotation...

This will be alleviated by considering BSS symbols, which we should but
don't so far, and then we should investigate those samples further.

The testcase was to have:

   perf top -e cycles/call-graph=fp/,cache-misses/call-graph=dwarf/,instructions

Running for a while till it segfaulted trying to access NULL notes->src->histograms.

Cc: Adrian Hunter 
Cc: David Ahern 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Wang Nan 
Link: https://lkml.kernel.org/n/tip-ndfjtpiop3tdcnyjgp320ra8@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf annotate: Move objdump_path to struct annotation_options

2018-06-04T13:28:53+00:00

One more step in grouping annotation options.

Cc: Adrian Hunter 
Cc: David Ahern 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Wang Nan 
Link: https://lkml.kernel.org/n/tip-sogzdhugoavm6fyw60jnb0vs@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf annotate: Move disassembler_style global to annotation_options

2018-06-04T13:28:53+00:00

Continuing to group annotation specific stuff into a struct.

Cc: Adrian Hunter 
Cc: David Ahern 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Wang Nan 
Link: https://lkml.kernel.org/n/tip-p3cdhltj58jt0byjzg3g7obx@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf annotate: Adopt anotation options from symbol_conf

2018-06-04T13:28:53+00:00

Continuing to group annotation options in an annotation specific struct.

Cc: Adrian Hunter 
Cc: David Ahern 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Wang Nan 
Link: https://lkml.kernel.org/n/tip-astei92tzxp4yccag5pxb2h7@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf annotate: Pass annotation_options to symbol__annotate()

2018-06-04T13:28:53+00:00

Now all callers to symbol__disassemble() can hand it the per-tool
annotation_options, which will allow us to remove lots of stuff
from symbol_options, the kitchen sink of perf configs, reducing its
size and getting annotation specific stuff grouped together.

Cc: Adrian Hunter 
Cc: David Ahern 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Wang Nan 
Link: https://lkml.kernel.org/n/tip-vpr7ys7ggvs2fzpg8wbjcw7e@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf annotate stdio: Use annotation_options consistently

2018-06-04T13:28:52+00:00

Accross all the routines, this way we can have eventually have a
consistent set of defaults for all UIs.

Cc: Adrian Hunter 
Cc: David Ahern 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Wang Nan 
Link: https://lkml.kernel.org/n/tip-6qgtixurjgdk5u0n3rw78ges@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf annotate: Replace symbol__alloc_hists() with symbol__hists()

2018-06-04T13:28:52+00:00

Its a bit shorter, so ditch the old symbol__alloc_hists() function.

Cc: Adrian Hunter 
Cc: David Ahern 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Wang Nan 
Link: https://lkml.kernel.org/n/tip-m7tienxk7dijh5ln62yln1m9@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

perf annotate: Stop using symbol_conf.nr_events global in symbol__hists()

2018-06-04T13:28:52+00:00

Since now we have evsel->evlist->nr_entries in the single place calling
this function, use it.

Cc: Adrian Hunter 
Cc: David Ahern 
Cc: Jiri Olsa 
Cc: Namhyung Kim 
Cc: Wang Nan 
Link: https://lkml.kernel.org/n/tip-9mgosbqa977h39j4i9ys8t75@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo

linux-stable.git/tools/perf/util/annotate.c, branch linux-4.18.y

perf annotate: Fix parsing aarch64 branch instructions after objdump update

perf annotate: Properly interpret indirect call

perf annnotate: Make __symbol__inc_addr_samples handle src->histograms == NULL

perf annotate: Move objdump_path to struct annotation_options

perf annotate: Move disassembler_style global to annotation_options

perf annotate: Adopt anotation options from symbol_conf

perf annotate: Pass annotation_options to symbol__annotate()

perf annotate stdio: Use annotation_options consistently

perf annotate: Replace symbol__alloc_hists() with symbol__hists()

perf annotate: Stop using symbol_conf.nr_events global in symbol__hists()

perf annnotate: Make symbolinc_addr_samples handle src->histograms == NULL