Commit Graph

22 Commits

Author SHA1 Message Date
Nikolay Borisov
1d35b6054a scripts/bloat-o-meter: Support comparing library archives
Library archives (.a) usually contain multiple object files so their
output of nm --size-sort contains lines like:

<omitted for brevity>
00000000000003a8 t run_test

extent-map-tests.o:
<omitted for brevity>

bloat-o-meter currently doesn't handle them which results in errors when
calling .split() on them.  Fix this by simply ignoring them.  This enables
diffing subsystems which generate built-in.a files.

Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Link: http://lkml.kernel.org/r/20200603103513.3712-1-nborisov@suse.com
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-08-07 11:33:21 -07:00
Rasmus Villemoes
e0b2475a3f bloat-o-meter: ignore __addressable_ symbols
Since __LINE__ is part of the symbol created by __ADDRESSABLE, almost
any change causes those symbols to disappear and get reincarnated, e.g.

add/remove: 4/4 grow/shrink: 0/3 up/down: 32/-171 (-139)
Function                                     old     new   delta
__addressable_tracing_set_default_clock8649       -       8      +8
__addressable_tracer_init_tracefs8631          -       8      +8
__addressable_ftrace_dump8383                  -       8      +8
__addressable_clear_boot_tracer8632            -       8      +8
__addressable_tracing_set_default_clock8650       8       -      -8
__addressable_tracer_init_tracefs8632          8       -      -8
__addressable_ftrace_dump8384                  8       -      -8
__addressable_clear_boot_tracer8633            8       -      -8
trace_default_header                         663     642     -21
tracing_mark_raw_write                       406     355     -51
tracing_mark_write                           624     557     -67
Total: Before=63889, After=63750, chg -0.22%

They're small and in .discard, so ignore them, leading to more useful

add/remove: 0/0 grow/shrink: 0/3 up/down: 0/-139 (-139)
Function                                     old     new   delta
trace_default_header                         663     642     -21
tracing_mark_raw_write                       406     355     -51
tracing_mark_write                           624     557     -67
Total: Before=63721, After=63582, chg -0.22%

Link: http://lkml.kernel.org/r/20181102210030.8383-1-linux@rasmusvillemoes.dk
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Dominik Brodowski <linux@dominikbrodowski.net>
Cc: Maninder Singh <maninder1.s@samsung.com>
Cc: Matteo Croce <mcroce@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2018-12-28 12:11:44 -08:00
Dominik Brodowski
5ac9efa3c5 syscalls/core, syscalls/x86: Clean up compat syscall stub naming convention
Tidy the naming convention for compat syscall subs. Hints which describe
the purpose of the stub go in front and receive a double underscore to
denote that they are generated on-the-fly by the COMPAT_SYSCALL_DEFINEx()
macro.

For the generic case, this means:

t            kernel_waitid	# common C function (see kernel/exit.c)

    __do_compat_sys_waitid	# inlined helper doing the actual work
				# (takes original parameters as declared)

T   __se_compat_sys_waitid	# sign-extending C function calling inlined
				# helper (takes parameters of type long,
				# casts them to unsigned long and then to
				# the declared type)

T        compat_sys_waitid      # alias to __se_compat_sys_waitid()
				# (taking parameters as declared), to
				# be included in syscall table

For x86, the naming is as follows:

t            kernel_waitid	# common C function (see kernel/exit.c)

    __do_compat_sys_waitid	# inlined helper doing the actual work
				# (takes original parameters as declared)

t   __se_compat_sys_waitid      # sign-extending C function calling inlined
				# helper (takes parameters of type long,
				# casts them to unsigned long and then to
				# the declared type)

T __ia32_compat_sys_waitid	# IA32_EMULATION 32-bit-ptregs -> C stub,
				# calls __se_compat_sys_waitid(); to be
				# included in syscall table

T  __x32_compat_sys_waitid	# x32 64-bit-ptregs -> C stub, calls
				# __se_compat_sys_waitid(); to be included
				# in syscall table

If only one of IA32_EMULATION and x32 is enabled, __se_compat_sys_waitid()
may be inlined into the stub __{ia32,x32}_compat_sys_waitid().

Suggested-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20180409105145.5364-3-linux@dominikbrodowski.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2018-04-09 16:47:28 +02:00
Dominik Brodowski
e145242ea0 syscalls/core, syscalls/x86: Clean up syscall stub naming convention
Tidy the naming convention for compat syscall subs. Hints which describe
the purpose of the stub go in front and receive a double underscore to
denote that they are generated on-the-fly by the SYSCALL_DEFINEx() macro.

For the generic case, this means (0xffffffff prefix removed):

 810f08d0 t     kernel_waitid	# common C function (see kernel/exit.c)

 <inline>     __do_sys_waitid	# inlined helper doing the actual work
				# (takes original parameters as declared)

 810f1aa0 T   __se_sys_waitid	# sign-extending C function calling inlined
				# helper (takes parameters of type long;
				# casts them to the declared type)

 810f1aa0 T        sys_waitid	# alias to __se_sys_waitid() (taking
				# parameters as declared), to be included
				# in syscall table

For x86, the naming is as follows:

 810efc70 t     kernel_waitid	# common C function (see kernel/exit.c)

 <inline>     __do_sys_waitid	# inlined helper doing the actual work
				# (takes original parameters as declared)

 810efd60 t   __se_sys_waitid	# sign-extending C function calling inlined
				# helper (takes parameters of type long;
				# casts them to the declared type)

 810f1140 T __ia32_sys_waitid	# IA32_EMULATION 32-bit-ptregs -> C stub,
				# calls __se_sys_waitid(); to be included
				# in syscall table

 810f1110 T        sys_waitid	# x86 64-bit-ptregs -> C stub, calls
				# __se_sys_waitid(); to be included in
				# syscall table

For x86, sys_waitid() will be re-named to __x64_sys_waitid in a follow-up
patch.

Suggested-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20180409105145.5364-2-linux@dominikbrodowski.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2018-04-09 16:47:27 +02:00
Matteo Croce
61fc470814 scripts/bloat-o-meter: fix typos in help
The bloat-o-meter script has two typos in the help, fix both.

Fixes: 192efb7a1f ("bloat-o-meter: provide 3 different arguments for data, function and All")
Signed-off-by: Matteo Croce <mcroce@redhat.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
2018-03-09 01:12:31 +09:00
Andy Shevchenko
edbddb83a1 scripts/bloat-o-meter: don't fail with division by 0
Under some circumstances it's possible to get a divider 0 which crashes
the script.

  Traceback (most recent call last):
    File "linux/scripts/bloat-o-meter", line 98, in <module>
      print_result("Function", "tTdDbBrR", 2)
    File "linux/scripts/bloat-o-meter", line 87, in print_result
      (otot, ntot, (ntot - otot)*100.0/otot))
  ZeroDivisionError: float division by zero

Hide this by checking the divider first.

Link: http://lkml.kernel.org/r/20171123171219.31453-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Vaneet Narang <v.narang@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-29 18:40:43 -08:00
Maninder Singh
192efb7a1f bloat-o-meter: provide 3 different arguments for data, function and All
This patch provides 3 new arguments for bloat-o-meter
 1) -c -> for all (showing function and data differently)
 2) -d -> data
 3) -t -> function

output:

  ./scripts/bloat-o-meter  -c "file1" "file2"
  add/remove: 0/0 grow/shrink: 0/1 up/down: 0/-152 (-152)
  Function                                     old     new   delta
  main                                         412     260    -152
  Total: Before=548, After=396, chg -27.74%
  ##########################################################
  add/remove: 1/0 grow/shrink: 1/0 up/down: 84/0 (84)
  Data                                         old     new   delta
  arr                                            -      64     +64
  backtrace                                     60      80     +20
  Total: Before=109, After=193, chg +77.06%
  ##########################################################
  add/remove: 0/1 grow/shrink: 0/0 up/down: 0/-64 (-64)
  RO Data                                      old     new   delta
  arr                                           64       -     -64
  Total: Before=68, After=4, chg -94.12%

[maninder1.s@samsung.com: v1 -> v2]
  Link: http://lkml.kernel.org/r/1506569402-24787-1-git-send-email-maninder1.s@samsung.com
Link: http://lkml.kernel.org/r/1506336313-27187-1-git-send-email-maninder1.s@samsung.com
Signed-off-by: Vaneet Narang <v.narang@samsung.com>
Signed-off-by: Maninder Singh <maninder1.s@samsung.com>
Cc: Amit Sahrawat <a.sahrawat@samsung.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Michal Marek <mmarek@suse.cz>
Cc: <pankaj.m@samsung.com>
Cc: <a.sahrawat@samsung.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2017-11-15 18:21:00 -08:00
Alexey Dobriyan
0d7bbb4364 scripts/bloat-o-meter: compile .NUMBER regex
Every often used regex is better be compiled in Python.

Speedup is about ~9.8% (whee!)

    $ perf stat -r 16 taskset -c 15 ./scripts/bloat-o-meter ../vmlinux-000 ../obj/vmlinux >/dev/null
    7.091202853 seconds time elapsed                         ( +-  0.15% )

    +re.compile
    6.397564973 seconds time elapsed                         ( +-  0.34% )

Link: http://lkml.kernel.org/r/20161119004417.GB1200@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-12 18:55:06 -08:00
Alexey Dobriyan
3af06fd96a scripts/bloat-o-meter: don't use readlines()
readlines() conses whole list before doing anything which is slower for
big object files.  Use per line iterator.

Speed up is ~2% on "allyesconfig" type of kernel.

    $ perf stat -r 16 taskset -c 15 ./scripts/bloat-o-meter ../vmlinux-000 ../obj/vmlinux >/dev/null
	...

  Before:  7.247708646 seconds time elapsed                ( +-  0.28% )
  After:   7.091202853 seconds time elapsed                ( +-  0.15% )

Link: http://lkml.kernel.org/r/20161119004143.GA1200@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-12-12 18:55:06 -08:00
Alexey Dobriyan
eef06b82f1 scripts/bloat-o-meter: fix SIGPIPE
Fix piping output to a program which quickly exits (read: head -n1)

	$ ./scripts/bloat-o-meter ../vmlinux-000 ../obj/vmlinux | head -n1
	add/remove: 0/0 grow/shrink: 9/60 up/down: 124/-305 (-181)
	close failed in file object destructor:
	sys.excepthook is missing
	lost sys.stderr

Link: http://lkml.kernel.org/r/20161028204618.GA29923@avx2
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-11-11 08:12:37 -08:00
Riku Voipio
8cde0daf6c scripts/bloat-o-meter: fix percent on <1% changes
Python divisions are integer divisions unless at least one parameter is
a float.  The current bloat-o-meter fails to print sub-percentage
changes:

  Total: Before=10515408, After=10604060, chg 0.000000%

Force float division by using one float and pretty the print to two
significant decimals:

  Total: Before=10515408, After=10604060, chg +0.84%

Link: http://lkml.kernel.org/r/1465980311-23814-1-git-send-email-riku.voipio@linaro.org
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Michal Marek <mmarek@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-07-26 16:19:19 -07:00
Vineet Gupta
b21e91c305 scripts/bloat-o-meter: print percent change
This adds an additional line of output (to reduce the chances of
breaking any existing output parsers) which prints the total size before
and after and the relative difference.

  add/remove: 39/0 grow/shrink: 12408/55 up/down: 362227/-1430 (360797)
  function                                     old     new   delta
  ext4_fill_super                            10556   12590   +2034
  _fpadd_parts                                   -    1186   +1186
  ntfs_fill_super                             5340    6164    +824
  ...
  ...
  __divdf3                                     752     386    -366
  unlzma                                      3682    3274    -408
  Total: Before=5023101, After=5383898, chg 7.000000%
  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Link: http://lkml.kernel.org/r/1463124110-30314-1-git-send-email-vgupta@synopsys.com
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
Cc: Josh Triplett <josh@joshtriplett.org>
Cc: Michal Marek <mmarek@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-05-19 19:12:14 -07:00
Sergey Senozhatsky
72214a24a7 scripts/bloat-o-meter: fix python3 syntax error
In Python3+ print is a function so the old syntax is not correct
anymore:

  $ ./scripts/bloat-o-meter vmlinux.o vmlinux.o.old
    File "./scripts/bloat-o-meter", line 61
      print "add/remove: %s/%s grow/shrink: %s/%s up/down: %s/%s (%s)" % \
                                                                     ^
  SyntaxError: invalid syntax

Fix by calling print as a function.

Tested on python 2.7.11, 3.5.1

Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-01-14 16:00:49 -08:00
Josh Triplett
b25c2ff547 bloat-o-meter: Ignore syscall aliases SyS_ and compat_SyS_
This avoids double-counting size changes in syscall implementations.

Signed-off-by: Josh Triplett <josh@joshtriplett.org>
2014-08-23 13:08:48 -07:00
Josh Triplett
c2e182fab0 scripts/bloat-o-meter: use .startswith rather than fragile slicing
str.startswith has existed since at least Python 2.0, in 2000; use it
rather than a fragile comparison against an initial slice of a string,
which requires hard-coding the length of the string to compare against.

Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Michal Marek <mmarek@suse.cz>
2013-11-07 10:47:13 +01:00
Josh Triplett
5a7b2d2796 scripts/bloat-o-meter: ignore changes in the size of linux_banner
linux_banner can change size due to changes in the compiler, build number,
or the user@host the system was compiled on; ignore size changes in
linux_banner entirely.

Signed-off-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Michal Marek <mmarek@suse.cz>
2013-11-07 10:46:54 +01:00
Andi Kleen
21cf6e584c kbuild, bloat-o-meter: fix static detection
Disable static detection: the static currently drops a lot of useful
information including clones generated by gcc. Drop this. The statics
will appear now without static. prefix.

But remove the LTO .NUMBER postfixes that look ugly

Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Michal Marek <mmarek@suse.cz>
2013-11-06 22:25:59 +01:00
Jean Delvare
c50e3f512a bloat-o-meter: include read-only data section in report
I'm not sure why the read-only data section is excluded from the report,
it seems as relevant as the other data sections (b and d).

I've stripped the symbols starting with __mod_ as they can have their
names dynamically generated and thus comparison between binaries is not
possible.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Andi Kleen <andi@firstfloor.org>
Acked-by: Nathan Lynch <ntl@pobox.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2011-03-22 17:44:17 -07:00
Nathan Lynch
8998979cc1 fix bloat-o-meter for ppc64
bloat-o-meter assumes that a '.' anywhere in a symbol's name means that it
is static and prepends 'static.' to the first part of the symbol name,
discarding the portion of the name that follows the '.'.  However, the
names of function entry points begin with '.' in the ppc64 ABI.  This
causes all function text size changes to be accounted to a single 'static.'
entry in the output when comparing ppc64 kernels.

Change getsizes() to ignore the first character of the symbol name when
searching for '.'.

Signed-off-by: Nathan Lynch <ntl@pobox.com>
Cc: Matt Mackall <mpm@selenic.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2007-12-17 19:28:17 -08:00
Rob Landley
51849738cf [PATCH] bloat-o-meter: gcc-4 fix
Upgrade scripts/bloat-o-meter to handle the names gcc 4 gives static
symbols.

Signed-off-by: Rob Landley <rob@landley.net>
Signed-off-by: Matt Mackall <mpm@selenic.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-06-25 10:01:00 -07:00
Paolo 'Blaisorblade' Giarrusso
71378be91f kbuild: fix mode of checkstack.pl and other files.
Make it executable like it should be. Do the same for other files intended to be
executed by the user - the ones called by the build process needn't be
executable as they already work (as argument to their interpreter).

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Sam Ravnborg <sam@ravnborg.org>
2006-04-11 13:37:07 +02:00
Matt Mackall
d960600df3 [PATCH] tiny: Add bloat-o-meter to scripts
This is a rewrite of Andi Kleen's bloat-o-meter with sorting and reporting of
gainers/decliners.  Sample output:

add/remove: 0/8 grow/shrink: 2/0 up/down: 88/-4424 (-4336)
function                                     old     new   delta
__copy_to_user_ll                             59     103     +44
__copy_from_user_ll                           59     103     +44
fill_note                                     32       -     -32
maydump                                       58       -     -58
dump_seek                                     67       -     -67
writenote                                    180       -    -180
elf_dump_thread_status                       274       -    -274
fill_psinfo                                  308       -    -308
fill_prstatus                                466       -    -466
elf_core_dump                               3039       -   -3039

The summary line says:
 no functions added, 8 removed
 two functions grew, none shrunk
 we gained 88 bytes and lost 4424 (or -4336 net)

This work was sponsored in part by CE Linux Forum

Signed-off-by: Matt Mackall <mpm@selenic.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-01-08 20:14:10 -08:00