For example,

  $ scripts/objdiff record init drivers/usb

disassembles all the objects under init and drivers/usb directories.

This feature would be useful when we change various files under the
specific directory.
Signed-off-by: NMasahiro Yamada <yamada.m@jp.panasonic.com>
Acked-by: NJason Cooper <jason@lakedaemon.net>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

Signed-off-by: NMasahiro Yamada <yamada.m@jp.panasonic.com>
Acked-by: NJason Cooper <jason@lakedaemon.net>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

Prior to this commit, the command "scripts/objdiff record path/to/*.o"
disassembled the given object into ".tmp_objdiff/path/to/*.o" file.

The problem here is that recorded disassemblies are lost if we run
"make clean" because it removes all the *.o files.
Disassembled code should be dumped into *.dis instead of *.o files.
Signed-off-by: NMasahiro Yamada <yamada.m@jp.panasonic.com>
Acked-by: NJason Cooper <jason@lakedaemon.net>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

Prior to this commit, scripts/objdiff expected to be run at the top
directory and only the relative path of objects.

This commit provides more flexibility in terms of object path:

[1] scripts/objdiff can be run in any directory

For example,

  $ scripts/objdiff record init/main.o

and

  $ cd init; ../scripts/objdiff record main.o

produce the same result.

[2] Support absolute path for objects

  $ scripts/objdiff record /home/foo/bar/linux/init/main.o

work as well.
Signed-off-by: NMasahiro Yamada <yamada.m@jp.panasonic.com>
Acked-by: NJason Cooper <jason@lakedaemon.net>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

The directories for objdump is created by the code
a few lines below:

  [ ! -d "$OBJDIFFD/$dn" ] && mkdir -p "$OBJDIFFD/$dn"
Signed-off-by: NMasahiro Yamada <yamada.m@jp.panasonic.com>
Acked-by: NJason Cooper <jason@lakedaemon.net>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

Signed-off-by: NMasahiro Yamada <yamada.m@jp.panasonic.com>
Acked-by: NJason Cooper <jason@lakedaemon.net>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

This commit is a minor refactoring.

Temporary files for objdiff are stored in .tmp_objdiff directory
which is located at the top directory.

To get the path to this directory,

  SRCTREE=`git rev-parse --show-toplevel`
  TMPD=$SRCTREE/.tmp_objdiff

seems easier to understand than

  GIT_DIR=`git rev-parse --git-dir`
  TMPD=${GIT_DIR%git}tmp_objdiff

Besides, it is not always necessary to create .tmp_objdiff dicrectory.
It should be created only for "record" command.
Signed-off-by: NMasahiro Yamada <yamada.m@jp.panasonic.com>
Acked-by: NJason Cooper <jason@lakedaemon.net>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

The Debian s390x architecture has 64-bit userland whereas s390 has
32-bit userland.  A 64-bit kernel can be used with either.  Now that
Debian supports multiarch and officially supports s390x, it makes more
sense to assign a 64-bit kernel package to s390x.
Reported-by: NStephen Powell <zlinuxman@wowway.com>
References: https://bugs.debian.org/750925Signed-off-by: NBen Hutchings <ben@decadent.org.uk>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

This semantic patch looks for variables that are initialized with a
constant, are never updated, and are only used as parameter of return.
Return the constant instead of using a variable.
Signed-off-by: NPeter Senna Tschudin <peter.senna@gmail.com>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

Signed-off-by: NMasahiro Yamada <yamada.m@jp.panasonic.com>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

Signed-off-by: NMasahiro Yamada <yamada.m@jp.panasonic.com>

Failure to terminate an of_device_id table can lead to confusing
failures depending on where the compiler places the array. Add a
check to make sure these tables are terminated. Thanks to Mitchel
Humpherys for coming up with the pattern initially.

Cc: Mitchel Humpherys <mitchelh@codeaurora.org>
Cc: Gilles Muller <Gilles.Muller@lip6.fr>
Cc: Nicolas Palix <nicolas.palix@imag.fr>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: devicetree@vger.kernel.org
Signed-off-by: NStephen Boyd <sboyd@codeaurora.org>
Acked-by: NGrant Likely <grant.likely@linaro.org>
Signed-off-by: NJulia Lawall <julia.lawall@lip6.fr>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

Since commit 22d651dc ('selftests/powerpc:
Import Anton's memcpy / copy_tofrom_user tests'), some source files in the
tree appear as symlink.
Until commit 8c38a532 ('scripts/tags.sh:
ignore code of user space tools'), those symlinks made cscope report some
warnings:

    $ make ALLSOURCE_ARCHS=all O=./obj-cscope/ cscope
      GEN     cscope
    cscope: cannot find
file .../tools/testing/selftests/powerpc/copyloops/copyuser_power7.S
    cscope: cannot find
file .../tools/testing/selftests/powerpc/copyloops/memcpy_64.S
    cscope: cannot find
file .../tools/testing/selftests/powerpc/copyloops/memcpy_power7.S
    cscope: cannot find
file .../tools/testing/selftests/powerpc/copyloops/copyuser_64.S

In order to prevent the same kind of warnings to be triggered by future
addition of symlinks, the best option is to ignore all symlinks when
building the file list to be processed by cscope (and other tools
supported by scripts/tags.sh).

Ignoring symlinks won't hide source files from cscope (and others) as the
target of these symlinks already appear somewhere else in the tree, and,
as such, should be processed by cscope (or others).

Note that, cscope, when used with -R option to make it find the files to
process by itself, already skip symlinks: it's not expected that cscope
access source files through symlink.

On top of commit 8c38a532 ('scripts/tags.sh:
ignore code of user space tools'), scripts/tags.sh output from
"make cscope tags TAGS" is the same with and without this patch: it doesn't
seems to introduce any regression (on Fedora 20).

Link: http://lkml.kernel.org/r/1396530975.4361.28.camel@localhost.localdomain
Link: http://mid.gmane.org/534312F8.5090609@t-online.de
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Anton Blanchard <anton@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Hans-Bernhard Bröker <broeker@users.sourceforge.net>,
Cc: Hans-Bernhard Broeker <broeker@physik.rwth-aachen.de>,
Cc: Neil Horman <nhorman@users.sourceforge.net>
Cc: Neil Horman <nhorman@tuxdriver.com>
Cc: Gerhard Sittig <gsi@denx.de>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Signed-off-by: NYann Droneaud <ydroneaud@opteya.com>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

Currently, while using ctags to read code, we would get stumbled on
PageCgroup* symbols: no definition found. And it is quite dull to
manually dig it out.

This patch adds regular expression replacement pattern for such symbols,
like what have done for the PageXXX flag. It will teach ctags to find
out the definition for us.
Signed-off-by: NJianyu Zhan <nasa4836@gmail.com>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

Signed-off-by: NFathi Boudra <fathi.boudra@linaro.org>
Reviewed-by: NBen Hutchings <ben@decadent.org.uk>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

In cross-build environment, we expect to use the cross-compiler objcopy
instead of the host objcopy.

It fixes following build failures:
objcopy --only-keep-debug lib/modules/3.14/kernel/net/ipv6/xfrm6_mode_tunnel.ko /srv/build/linux/debian/dbgtmp/usr/lib/debug/lib/modules/3.14/kernel/net/ipv6/xfrm6_mode_tunnel.ko
objcopy: Unable to recognise the format of the input file `lib/modules/3.14/kernel/net/ipv6/xfrm6_mode_tunnel.ko'
Signed-off-by: NFathi Boudra <fathi.boudra@linaro.org>
Cc: stable <stable@vger.kernel.org> # 3.12+
Fixes: 810e8437 ('deb-pkg: split debug symbols in their own package')
Reviewed-by: NBen Hutchings <ben@decadent.org.uk>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

User space code in tools/ often reuses names of kernel constructions,
this confuses navigation in the normal kernel code. Let's fix this mess.
Signed-off-by: NKonstantin Khlebnikov <koct9i@gmail.com>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

Signed-off-by: NJianyu Zhan <nasa4836@gmail.com>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

When building the LINUX_COMPILER definition, instead of merely taking the last
line from "$(CC) -v", grep for ' version ' in the output. This supports both
gcc and clang.
Signed-off-by: NJan-Simon Möller <dl9pf@gmx.de>
Signed-off-by: NBehan Webster <behanw@converseincode.com>
Signed-off-by: NMark Charlebois <charlebm@gmail.com>
Cc: PaX Team <pageexec@freemail.hu>

When compiling kernel with clang, disable warnings which are too noisy, and
add the clang flag catch-undefined-behavior.
Signed-off-by: NJan-Simon Möller <dl9pf@gmx.de>
Signed-off-by: NBehan Webster <behanw@converseincode.com>
Signed-off-by: NMark Charlebois <mcharleb@gmail.com>
Cc: PaX Team <pageexec@freemail.hu>

PTR_RET is deprecated. Do not recommend its usage anymore.
Use PTR_ERR_OR_ZERO instead.
Signed-off-by: NSachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

Adding -header + help function like other .pl in /scripts.
Signed-off-by: NFabian Frederick <fabf@skynet.be>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

objdiff is useful when doing large code cleanups.  For example, when
removing checkpatch warnings and errors from new drivers in the staging
tree.

objdiff can be used in conjunction with a git rebase to confirm that
each commit made no changes to the resulting object code.  It has the
same return values as diff(1).

This was written specifically to support adding the skein and threefish
cryto drivers to the staging tree.  I needed a programmatic way to
confirm that commits changing >90% of the lines didn't inadvertently
change the code.

Temporary files (objdump output) are stored in

  /path/to/linux/.tmp_objdiff

'make mrproper' will remove this directory.
Signed-off-by: NJason Cooper <jason@lakedaemon.net>
Signed-off-by: NMichal Marek <mmarek@suse.cz>

"make allnoconfig" exists to ease testing of minimal configurations.
Documentation/SubmitChecklist includes a note to test with allnoconfig.
This helps catch missing dependencies on common-but-not-required
functionality, which might otherwise go unnoticed.

However, allnoconfig still leaves many symbols enabled, because they're
hidden behind CONFIG_EMBEDDED or CONFIG_EXPERT.  For instance, allnoconfig
still has CONFIG_PRINTK and CONFIG_BLOCK enabled, so drivers don't
typically get build-tested with those disabled.

To address this, introduce a new Kconfig option "allnoconfig_y", used on
symbols which only exist to hide other symbols.  Set it on CONFIG_EMBEDDED
(which then selects CONFIG_EXPERT).  allnoconfig will then disable all the
symbols hidden behind those.
Signed-off-by: NJosh Triplett <josh@joshtriplett.org>
Tested-by: NPaul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Michal Marek <mmarek@suse.cz>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This test prevents code from being aligned around the : for easy visual
counting of bitfield lengths.

ie:
	int foo		: 1,
	int bar		: 2,
	int foobar	:29;

should be acceptable so remove the test.
Signed-off-by: NJoe Perches <joe@perches.com>
Suggested-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Currently the parenthesis alignment test works only on misalignments of
if statements like

	if (foo(bar,
			baz)

Expand the test to find misalignments like:

static inline int foo(int bar,
			int baz)

and

	foo(bar,
			baz);

and

	foo = bar(baz,
			qux);

Expand the $Inline keyword for __inline and __inline__ too.
Add $Inline to $Declare so it also matches "static inline <foo>".

These checks are only performed with --strict.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

A commit hook for the Gerrit code review server [1] inserts change
identifiers so Gerrit can track patches through multiple revisions.
These identifiers are noise in the context of the upstream kernel.
(Many Gerrit servers are private.  Even given a public instance, given
only a Change-Id, one must guess which server a change was tracked on.
Patches submitted to the Linux kernel mailing lists should be able to
stand on their own.  If it's truly useful to reference code review on a
Gerrit server, a URL is a much clearer way to do so.) Thus, issue an
error when a Change-Id line is encountered before the Signed-off-by.

1. https://gerrit.googlesource.com/gerrit/+/master/gerrit-server/src/main/resources/com/google/gerrit/server/tools/root/hooks/commit-msgSigned-off-by: NChristopher Covington <cov@codeaurora.org>
Cc: Joe Perches <joe@perches.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Revert commit 7e4915e7 ("checkpatch: add warning of future
__GFP_NOFAIL use").

There are no plans to remove __GFP_NOFAIL.

__GFP_NOFAIL exists to

a) centralise the retry-allocation-for-ever operation into the core
   allocator, which is the appropriate implementation site and

b) permit us to identify code sites which aren't handling memory
   exhaustion appropriately.

Cc: David Rientjes <rientjes@google.com>
Cc: Joe Perches <joe@perches.com>
Cc: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Networking prefers this style, so warn when it's not used.

Networking uses:

    void foo(int bar)
    {
        int baz;

        code...
   }

not

    void foo(int bar)
    {
        int baz;
        code...
    }

There are a limited number of false positives when using macros to
declare variables like:

  WARNING: networking uses a blank line after declarations
  #330: FILE: net/ipv4/inet_hashtables.c:330:
  +	int dif = sk->sk_bound_dev_if;
  +	INET_ADDR_COOKIE(acookie, saddr, daddr)
Signed-off-by: NJoe Perches <joe@perches.com>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Improve the vendor name match in vendor-prefix.txt by only matching the
exact vendor name at the beginning of lines.
Signed-off-by: NFlorian Vaussard <florian.vaussard@epfl.ch>
Cc: Joe Perches <joe@perches.com>
Acked-by: NRob Herring <robh@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Look for ".compatible = "foo" strings not only in .dts files, but
in .c and .h too.
Signed-off-by: NFlorian Vaussard <florian.vaussard@epfl.ch>
Cc: Joe Perches <joe@perches.com>
Acked-by: NRob Herring <robh@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

With a compatible string like

  compatible = "foo";

checkpatch will currently try to find "foo" in vendor-prefixes.txt,
which is wrong since the vendor prefix is empty in this specific case.

Skip the vendor test if the compatible is not like

  compatible = "vendor,something";
Signed-off-by: NFlorian Vaussard <florian.vaussard@epfl.ch>
Cc: Joe Perches <joe@perches.com>
Acked-by: NRob Herring <robh@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The current vendor compatible check will not match vendors with dashes,
like:

  compatible="asahi-kasei"
Signed-off-by: NFlorian Vaussard <florian.vaussard@epfl.ch>
Reported-by: NJoe Perches <joe@perches.com>
Acked-by: NRob Herring <robh@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The current octal permissions test is very slow.

When patch ("checkpatch: add checks for constant non-octal permissions")
was added, processing time approximately tripled.

Regain almost all of the performance by not looping through all the
possible functions unless the line contains one of the functions.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Modify warning message when printk is used in a patch.  It mentions to
use subsystem_dbg instead of netdev_dbg as the first preferred format of
logging debug messages.
Signed-off-by: NYogesh Chaudhari <mr.yogesh@gmail.com>
Cc: Joe Perches <joe@perches.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This test is a bit noisy and opinions seem to agree that it should not
warn in a lot more situations.

It seems people agree that:

	return (foo || bar);
and
	return foo || bar;

are both acceptable style and checkpatch should be silent about them.

For now, it warns on parentheses around a simple constant or a single
function or a ternary.

	return (foo);
	return (foo(bar));
	return (foo ? bar : baz);

The last ternary test may be quieted in the future.

Modify the deparenthesize function to only strip balanced leading and
trailing parentheses.
Signed-off-by: NJoe Perches <joe@perches.com>
Cc: Julia Lawall <julia.lawall@lip6.fr>
Reviewed-by: NJosh Triplett <josh@joshtriplett.org>
Cc: Monam Agarwal <monamagarwal123@gmail.com>
Cc: Greg KH <gregkh@linuxfoundation.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

It's very common to have normal block comments for the initial comments
of a file description preface.

So for files in drivers/net and net/ don't emit a warning when the first
comment block in the file uses the normal block comment style and not
the networking block comment style.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Instead of array indexing $_, use temporary variables like all the other
subroutines in the script use.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

static const char* arrays create smaller text as each function call does
not have to populate the array.

Emit a warning when char *arrays aren't static const and the array is
not apparently global by being declared in the first column.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

checkpatch could not distinguish between a variable in a struct named
jiffies and the normal jiffies.

	foo->jiffies

would emit a "Comparing jiffies" arning.

Update the $Compare variable to do a negative look-behind for "-" when
finding a ">" so that a pointer dereference like -> isn't a comparison.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>