Skip to contentSkip to conversation

tools: Remove conflicting BITS_PER_LONG define

PATCH In review Alexander Sverdlin 3 participants 8 messages Sep 12, 2018 – Sep 24, 2018

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 tools: Remove conflicting BITS_PER_LONG define

CC .../tools/objtool/builtin-check.o
...
In file included from .../tools/arch/x86/include/uapi/asm/bitsperlong.h:11:0,

                 from .../tools/include/asm-generic/bitops/__ffs.h:6,
                 from .../tools/include/asm-generic/bitops.h:16,
                 from .../tools/include/linux/bitops.h:35,
                 from .../tools/include/linux/hashtable.h:13,
                 from elf.h:24,
                 from check.h:22,
                 from builtin-check.c:30:

.../tools/include/asm-generic/bitsperlong.h:8:0: error: "BITS_PER_LONG" redefined [-Werror]
#define BITS_PER_LONG (__CHAR_BIT__ * __SIZEOF_LONG__)

Include <asm/bitsperlong.h> instead as other headers do.

---

Seen during x86_64 build. Seems that most of the compilers do not define __SIZEOF_LONG__, but my does.

 tools/include/linux/bitops.h | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)
tools/include/linux/bitops.h +1 -4
index acc704b..4ae5232 100644
@@ -3,16 +3,13 @@
 #define _TOOLS_LINUX_BITOPS_H_
 
 #include <asm/types.h>
+#include <asm/bitsperlong.h>
 #include <linux/compiler.h>
 
 #ifndef __WORDSIZE
 #define __WORDSIZE (__SIZEOF_LONG__ * 8)
 #endif
 
-#ifndef BITS_PER_LONG
-# define BITS_PER_LONG __WORDSIZE
-#endif
-
 #define BIT_MASK(nr)		(1UL << ((nr) % BITS_PER_LONG))
 #define BIT_WORD(nr)		((nr) / BITS_PER_LONG)
 #define BITS_PER_BYTE		8
Signature · 1 line
2.4.6
#1 lore raw

Em Wed, Sep 12, 2018 at 07:02:32PM +0200, Alexander Sverdlin escreveu:

Quoted 10 lines

CC .../tools/objtool/builtin-check.o
...
In file included from .../tools/arch/x86/include/uapi/asm/bitsperlong.h:11:0,

                 from .../tools/include/asm-generic/bitops/__ffs.h:6,
                 from .../tools/include/asm-generic/bitops.h:16,
                 from .../tools/include/linux/bitops.h:35,
                 from .../tools/include/linux/hashtable.h:13,
                 from elf.h:24,
                 from check.h:22,
                 from builtin-check.c:30:
Quoted 4 lines

.../tools/include/asm-generic/bitsperlong.h:8:0: error: "BITS_PER_LONG" redefined [-Werror]
#define BITS_PER_LONG (__CHAR_BIT__ * __SIZEOF_LONG__)

Include <asm/bitsperlong.h> instead as other headers do.

Please try test building all tools in tools/

This broke make -C tools/perf/

- Arnaldo

#2 lore raw

Em Wed, Sep 12, 2018 at 04:01:07PM -0300, Arnaldo Carvalho de Melo escreveu:

Quoted 14 lines

Em Wed, Sep 12, 2018 at 07:02:32PM +0200, Alexander Sverdlin escreveu:

Quoted 13 lines

CC .../tools/objtool/builtin-check.o
...
In file included from .../tools/arch/x86/include/uapi/asm/bitsperlong.h:11:0,

                 from .../tools/include/asm-generic/bitops/__ffs.h:6,
                 from .../tools/include/asm-generic/bitops.h:16,
                 from .../tools/include/linux/bitops.h:35,
                 from .../tools/include/linux/hashtable.h:13,
                 from elf.h:24,
                 from check.h:22,
                 from builtin-check.c:30:

.../tools/include/asm-generic/bitsperlong.h:8:0: error: "BITS_PER_LONG" redefined [-Werror]
#define BITS_PER_LONG (__CHAR_BIT__ * __SIZEOF_LONG__)

Quoted 5 lines
Quoted 1 line

Include <asm/bitsperlong.h> instead as other headers do.

Please try test building all tools in tools/

This broke make -C tools/perf/

Also, where is the build of objtool failing?

[acme@jouet linux]$ git log --oneline -3
96eddb810b14 (HEAD -> master, torvalds/master) Merge tag 'riscv-for-linus-4.19-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/riscv-linux
aeb5427218a7 Merge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
5e335542de83 Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid
[acme@jouet linux]$ make -C tools/objtool/
make: Entering directory '/home/acme/git/linux/tools/objtool'
make -C /home/acme/git/linux/tools/build CFLAGS= LDFLAGS= fixdep
make[1]: Entering directory '/home/acme/git/linux/tools/build'

  HOSTCC   fixdep.o
  HOSTLD   fixdep-in.o
  LINK     fixdep

make[1]: Leaving directory '/home/acme/git/linux/tools/build'
make -C /home/acme/git/linux/tools/lib/subcmd/ OUTPUT=/home/acme/git/linux/tools/objtool/ make[1]: Entering directory '/home/acme/git/linux/tools/lib/subcmd'
make -C /home/acme/git/linux/tools/build CFLAGS= LDFLAGS= /home/acme/git/linux/tools/objtool/fixdep

  HOSTCC   /home/acme/git/linux/tools/objtool/fixdep.o
  HOSTLD   /home/acme/git/linux/tools/objtool/fixdep-in.o
  LINK     /home/acme/git/linux/tools/objtool/fixdep
  CC       /home/acme/git/linux/tools/objtool/exec-cmd.o
  CC       /home/acme/git/linux/tools/objtool/help.o
  CC       /home/acme/git/linux/tools/objtool/pager.o
  CC       /home/acme/git/linux/tools/objtool/parse-options.o
  CC       /home/acme/git/linux/tools/objtool/run-command.o
  CC       /home/acme/git/linux/tools/objtool/sigchain.o
  CC       /home/acme/git/linux/tools/objtool/subcmd-config.o
  LD       /home/acme/git/linux/tools/objtool/libsubcmd-in.o
  AR       /home/acme/git/linux/tools/objtool/libsubcmd.a
make[1]: Leaving directory '/home/acme/git/linux/tools/lib/subcmd'
make[1]: Entering directory '/home/acme/git/linux/tools/objtool'
make[2]: Entering directory '/home/acme/git/linux/tools/objtool'
  GEN      arch/x86/lib/inat-tables.c
  CC       arch/x86/decode.o
  LD       arch/x86/objtool-in.o
make[2]: Leaving directory '/home/acme/git/linux/tools/objtool'
  CC       builtin-check.o
  CC       builtin-orc.o
  CC       check.o
  CC       orc_gen.o
  CC       orc_dump.o
  CC       elf.o
  CC       special.o
  CC       objtool.o
  CC       libstring.o
  CC       str_error_r.o
  LD       objtool-in.o

make[1]: Leaving directory '/home/acme/git/linux/tools/objtool'
LINK objtool
make: Leaving directory '/home/acme/git/linux/tools/objtool' [acme@jouet linux]$

Now building it without -C:

$ make -C tools/clean
[acme@jouet objtool]$ make
make -C /home/acme/git/linux/tools/build CFLAGS= LDFLAGS= fixdep
make[1]: Entering directory '/home/acme/git/linux/tools/build'
  HOSTCC   fixdep.o
  HOSTLD   fixdep-in.o
  LINK     fixdep
make[1]: Leaving directory '/home/acme/git/linux/tools/build'
make -C /home/acme/git/linux/tools/lib/subcmd/ OUTPUT=/home/acme/git/linux/tools/objtool/
make[1]: Entering directory '/home/acme/git/linux/tools/lib/subcmd'
make -C /home/acme/git/linux/tools/build CFLAGS= LDFLAGS= /home/acme/git/linux/tools/objtool/fixdep
  HOSTCC   /home/acme/git/linux/tools/objtool/fixdep.o
  HOSTLD   /home/acme/git/linux/tools/objtool/fixdep-in.o
  LINK     /home/acme/git/linux/tools/objtool/fixdep
  CC       /home/acme/git/linux/tools/objtool/exec-cmd.o
  CC       /home/acme/git/linux/tools/objtool/help.o
  CC       /home/acme/git/linux/tools/objtool/pager.o
  CC       /home/acme/git/linux/tools/objtool/parse-options.o
  CC       /home/acme/git/linux/tools/objtool/run-command.o
  CC       /home/acme/git/linux/tools/objtool/sigchain.o
  CC       /home/acme/git/linux/tools/objtool/subcmd-config.o
  LD       /home/acme/git/linux/tools/objtool/libsubcmd-in.o
  AR       /home/acme/git/linux/tools/objtool/libsubcmd.a
make[1]: Leaving directory '/home/acme/git/linux/tools/lib/subcmd'
make[1]: Entering directory '/home/acme/git/linux/tools/objtool'
make[2]: Entering directory '/home/acme/git/linux/tools/objtool'
  GEN      arch/x86/lib/inat-tables.c
  CC       arch/x86/decode.o
  LD       arch/x86/objtool-in.o
make[2]: Leaving directory '/home/acme/git/linux/tools/objtool'
  CC       builtin-check.o
  CC       builtin-orc.o
  CC       check.o
  CC       orc_gen.o
  CC       orc_dump.o
  CC       elf.o
  CC       special.o
  CC       objtool.o
  CC       libstring.o
  CC       str_error_r.o
  LD       objtool-in.o
make[1]: Leaving directory '/home/acme/git/linux/tools/objtool'
  LINK     objtool
[acme@jouet objtool]$

So where is the failure?

- Arnaldo

#3 lore raw
Quoted 15 lines · Arnaldo Carvalho de Melo

Em Wed, Sep 12, 2018 at 04:01:07PM -0300, Arnaldo Carvalho de Melo escreveu:

Quoted 14 lines

Em Wed, Sep 12, 2018 at 07:02:32PM +0200, Alexander Sverdlin escreveu:

Quoted 13 lines

CC .../tools/objtool/builtin-check.o
...
In file included from .../tools/arch/x86/include/uapi/asm/bitsperlong.h:11:0,

                 from .../tools/include/asm-generic/bitops/__ffs.h:6,
                 from .../tools/include/asm-generic/bitops.h:16,
                 from .../tools/include/linux/bitops.h:35,
                 from .../tools/include/linux/hashtable.h:13,
                 from elf.h:24,
                 from check.h:22,
                 from builtin-check.c:30:

.../tools/include/asm-generic/bitsperlong.h:8:0: error: "BITS_PER_LONG" redefined [-Werror]
#define BITS_PER_LONG (__CHAR_BIT__ * __SIZEOF_LONG__)

Quoted 7 lines
Quoted 5 lines
Quoted 1 line

Include <asm/bitsperlong.h> instead as other headers do.

Please try test building all tools in tools/

This broke make -C tools/perf/

Also, where is the build of objtool failing?

I've prepared some examples for you with "-fdirectives-only -save-temps":

This is what happens ultimately:

In file included from .../tools/arch/x86/include/uapi/asm/bitsperlong.h:11:0,

                 from .../tools/include/asm-generic/bitops/__ffs.h:6,
                 from .../tools/include/asm-generic/bitops.h:16,
                 from .../tools/include/linux/bitops.h:35,
                 from .../tools/include/linux/hashtable.h:13,
                 from elf.h:24,
                 from check.h:22,
                 from builtin-check.c:30:

.../tools/include/asm-generic/bitsperlong.h:8:0: error: "BITS_PER_LONG" redefined [-Werror]
#define BITS_PER_LONG (__CHAR_BIT__ * __SIZEOF_LONG__)

In file included from .../tools/include/linux/hashtable.h:13:0,

                 from elf.h:24,
                 from check.h:22,
                 from builtin-check.c:30:
.../tools/include/linux/bitops.h:13:0: note: this is the location of the previous definition
 # define BITS_PER_LONG __WORDSIZE

And this are relevant parts from builtin-check.i file:

# 1 "builtin-check.c"
# 1 ".../tools/objtool//"
# 1 "<built-in>"

...

#define __SIZEOF_LONG__ 4

...

# 1 ".../sys-root/usr/include/bits/wordsize.h" 1 3 4

...

#define __WORDSIZE 32

...

# 1 ".../sys-root/usr/include/bits/wordsize.h" 1 3 4

...

#define __WORDSIZE 32

...

# 1 ".../sys-root/usr/include/bits/wordsize.h" 1 3 4

...

#define __WORDSIZE 32

...

# 1 ".../sys-root/usr/include/bits/wordsize.h" 1 3 4

...

#define __WORDSIZE 32

...

# 1 ".../tools/include/linux/bitops.h" 1

...

#define BITS_PER_LONG __WORDSIZE

...

# 6 ".../tools/include/asm-generic/bitsperlong.h" 2

...

#define BITS_PER_LONG (__CHAR_BIT__ * __SIZEOF_LONG__)

It could be that your compiler doesn't define __SIZEOF_LONG__, while my does. It could be that it's caused by the fact that I'm using a "cross compiler" (i686 compiler to build x86_64 kernel), maybe I'm just using a newer compiler than you.

Signature · 2 lines
Best regards,
Alexander Sverdlin.
#4 lore raw

Em Wed, Sep 19, 2018 at 02:28:29PM +0200, Alexander Sverdlin escreveu:

Quoted 99 lines · Arnaldo Carvalho de Melo
Quoted 15 lines

Em Wed, Sep 12, 2018 at 04:01:07PM -0300, Arnaldo Carvalho de Melo escreveu:

Quoted 14 lines

Em Wed, Sep 12, 2018 at 07:02:32PM +0200, Alexander Sverdlin escreveu:

Quoted 13 lines

CC .../tools/objtool/builtin-check.o
...
In file included from .../tools/arch/x86/include/uapi/asm/bitsperlong.h:11:0,

                 from .../tools/include/asm-generic/bitops/__ffs.h:6,
                 from .../tools/include/asm-generic/bitops.h:16,
                 from .../tools/include/linux/bitops.h:35,
                 from .../tools/include/linux/hashtable.h:13,
                 from elf.h:24,
                 from check.h:22,
                 from builtin-check.c:30:

.../tools/include/asm-generic/bitsperlong.h:8:0: error: "BITS_PER_LONG" redefined [-Werror]
#define BITS_PER_LONG (__CHAR_BIT__ * __SIZEOF_LONG__)

Quoted 7 lines
Quoted 5 lines
Quoted 1 line

Include <asm/bitsperlong.h> instead as other headers do.

Please try test building all tools in tools/

This broke make -C tools/perf/

Also, where is the build of objtool failing?

I've prepared some examples for you with "-fdirectives-only -save-temps":

This is what happens ultimately:

In file included from .../tools/arch/x86/include/uapi/asm/bitsperlong.h:11:0,

                 from .../tools/include/asm-generic/bitops/__ffs.h:6,
                 from .../tools/include/asm-generic/bitops.h:16,
                 from .../tools/include/linux/bitops.h:35,
                 from .../tools/include/linux/hashtable.h:13,
                 from elf.h:24,
                 from check.h:22,
                 from builtin-check.c:30:

.../tools/include/asm-generic/bitsperlong.h:8:0: error: "BITS_PER_LONG" redefined [-Werror]
#define BITS_PER_LONG (__CHAR_BIT__ * __SIZEOF_LONG__)

In file included from .../tools/include/linux/hashtable.h:13:0,

                 from elf.h:24,
                 from check.h:22,
                 from builtin-check.c:30:
.../tools/include/linux/bitops.h:13:0: note: this is the location of the previous definition
 # define BITS_PER_LONG __WORDSIZE

And this are relevant parts from builtin-check.i file:

# 1 "builtin-check.c"
# 1 ".../tools/objtool//"
# 1 "<built-in>"

...

#define __SIZEOF_LONG__ 4

...

# 1 ".../sys-root/usr/include/bits/wordsize.h" 1 3 4

...

#define __WORDSIZE 32

...

# 1 ".../sys-root/usr/include/bits/wordsize.h" 1 3 4

...

#define __WORDSIZE 32

...

# 1 ".../sys-root/usr/include/bits/wordsize.h" 1 3 4

...

#define __WORDSIZE 32

...

# 1 ".../sys-root/usr/include/bits/wordsize.h" 1 3 4

...

#define __WORDSIZE 32

...

# 1 ".../tools/include/linux/bitops.h" 1

...

#define BITS_PER_LONG __WORDSIZE

...

# 6 ".../tools/include/asm-generic/bitsperlong.h" 2

Quoted 8 lines

...

#define BITS_PER_LONG (__CHAR_BIT__ * __SIZEOF_LONG__)

It could be that your compiler doesn't define __SIZEOF_LONG__, while my does. It could be that it's caused by the fact that I'm using a "cross compiler" (i686 compiler to build x86_64 kernel), maybe I'm just using a newer compiler than you.

That is indeed a cross build environment I'm not regularly testing, I'm trying these cross builds:

   9 android-ndk:r12b-arm          : Ok   arm-linux-androideabi-gcc (GCC) 4.9.x 20150123 (prerelease)
  10 android-ndk:r15c-arm          : Ok   arm-linux-androideabi-gcc (GCC) 4.9.x 20150123 (prerelease)
  19 debian:experimental-x-arm64   : Ok   aarch64-linux-gnu-gcc (Debian 8.2.0-4) 8.2.0
  20 debian:experimental-x-mips    : Ok   mips-linux-gnu-gcc (Debian 8.1.0-12) 8.1.0
  21 debian:experimental-x-mips64  : Ok   mips64-linux-gnuabi64-gcc (Debian 8.1.0-12) 8.1.0
  22 debian:experimental-x-mipsel  : Ok   mipsel-linux-gnu-gcc (Debian 8.1.0-12) 8.1.0
  28 fedora:24-x-ARC-uClibc        : Ok   arc-linux-gcc (ARCompact ISA Linux uClibc toolchain 2017.09-rc2) 7.1.1 20170710
  46 ubuntu:14.04.4-x-linaro-arm64 : Ok   aarch64-linux-gnu-gcc (Linaro GCC 5.5-2017.10) 5.5.0
  48 ubuntu:16.04-x-arm            : Ok   arm-linux-gnueabihf-gcc (Ubuntu/Linaro 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
  49 ubuntu:16.04-x-arm64          : Ok   aarch64-linux-gnu-gcc (Ubuntu/Linaro 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
  50 ubuntu:16.04-x-powerpc        : Ok   powerpc-linux-gnu-gcc (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
  51 ubuntu:16.04-x-powerpc64      : Ok   powerpc64-linux-gnu-gcc (Ubuntu/IBM 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
  52 ubuntu:16.04-x-powerpc64el    : Ok   powerpc64le-linux-gnu-gcc (Ubuntu/IBM 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
  53 ubuntu:16.04-x-s390           : Ok   s390x-linux-gnu-gcc (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
  57 ubuntu:18.04-x-arm            : Ok   arm-linux-gnueabihf-gcc (Ubuntu/Linaro 7.3.0-16ubuntu3) 7.3.0
  58 ubuntu:18.04-x-arm64          : Ok   aarch64-linux-gnu-gcc (Ubuntu/Linaro 7.3.0-16ubuntu3) 7.3.0
  59 ubuntu:18.04-x-m68k           : Ok   m68k-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
  60 ubuntu:18.04-x-powerpc        : Ok   powerpc-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
  61 ubuntu:18.04-x-powerpc64      : Ok   powerpc64-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
  62 ubuntu:18.04-x-powerpc64el    : Ok   powerpc64le-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
  63 ubuntu:18.04-x-riscv64        : Ok   riscv64-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
  64 ubuntu:18.04-x-s390           : Ok   s390x-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
  65 ubuntu:18.04-x-sh4            : Ok   sh4-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
  66 ubuntu:18.04-x-sparc64        : Ok   sparc64-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0

I'll try and get one for building a x86_64 tools/perf, tools/lib/{api,bpf,traceevent} to see if I manage to reproduce the problem you're reporting.

- Arnaldo

#5 lore raw

Hi!

Quoted 2 lines · Arnaldo Carvalho de Melo

That is indeed a cross build environment I'm not regularly testing, I'm trying these cross builds:

Probably my assumption was wrong about cross compiler.
HOST tool fails (objtool) and it's being build by separate i686->i686 compiler.

Quoted 16 lines
   9 android-ndk:r12b-arm          : Ok   arm-linux-androideabi-gcc (GCC) 4.9.x 20150123 (prerelease)
  10 android-ndk:r15c-arm          : Ok   arm-linux-androideabi-gcc (GCC) 4.9.x 20150123 (prerelease)
  19 debian:experimental-x-arm64   : Ok   aarch64-linux-gnu-gcc (Debian 8.2.0-4) 8.2.0
  20 debian:experimental-x-mips    : Ok   mips-linux-gnu-gcc (Debian 8.1.0-12) 8.1.0
  21 debian:experimental-x-mips64  : Ok   mips64-linux-gnuabi64-gcc (Debian 8.1.0-12) 8.1.0
  22 debian:experimental-x-mipsel  : Ok   mipsel-linux-gnu-gcc (Debian 8.1.0-12) 8.1.0
  28 fedora:24-x-ARC-uClibc        : Ok   arc-linux-gcc (ARCompact ISA Linux uClibc toolchain 2017.09-rc2) 7.1.1 20170710
  46 ubuntu:14.04.4-x-linaro-arm64 : Ok   aarch64-linux-gnu-gcc (Linaro GCC 5.5-2017.10) 5.5.0
  48 ubuntu:16.04-x-arm            : Ok   arm-linux-gnueabihf-gcc (Ubuntu/Linaro 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
  49 ubuntu:16.04-x-arm64          : Ok   aarch64-linux-gnu-gcc (Ubuntu/Linaro 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
  50 ubuntu:16.04-x-powerpc        : Ok   powerpc-linux-gnu-gcc (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
  51 ubuntu:16.04-x-powerpc64      : Ok   powerpc64-linux-gnu-gcc (Ubuntu/IBM 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
  52 ubuntu:16.04-x-powerpc64el    : Ok   powerpc64le-linux-gnu-gcc (Ubuntu/IBM 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
  53 ubuntu:16.04-x-s390           : Ok   s390x-linux-gnu-gcc (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
  57 ubuntu:18.04-x-arm            : Ok   arm-linux-gnueabihf-gcc (Ubuntu/Linaro 7.3.0-16ubuntu3) 7.3.0
  58 ubuntu:18.04-x-arm64          : Ok   aarch64-linux-gnu-gcc (Ubuntu/Linaro 7.3.0-16ubuntu3) 7.3.0
Quoted 8 lines
  59 ubuntu:18.04-x-m68k           : Ok   m68k-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
  60 ubuntu:18.04-x-powerpc        : Ok   powerpc-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
  61 ubuntu:18.04-x-powerpc64      : Ok   powerpc64-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
  62 ubuntu:18.04-x-powerpc64el    : Ok   powerpc64le-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
  63 ubuntu:18.04-x-riscv64        : Ok   riscv64-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
  64 ubuntu:18.04-x-s390           : Ok   s390x-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
  65 ubuntu:18.04-x-sh4            : Ok   sh4-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0
  66 ubuntu:18.04-x-sparc64        : Ok   sparc64-linux-gnu-gcc (Ubuntu 7.3.0-16ubuntu3) 7.3.0

I think the big list is mostly irrelevant. I was wondering why only x86 fails from my list of different targets (ARM, ARM64, MIPS64, x86, x86_64, PPC), but it turns out that objtool is only built for HAVE_STACK_VALIDATION targets (means, x86 only?).

I've tried to build it for MIPS64 with make tools/objtool and it fails in the same way.

One special thing about my compiler is really old glibc it's being built against (hello RHEL :) maybe this causes reverse order of includes.

Nevertheless two conflicting defines of BITS_PER_LONG will only work with one particular order of includes but not opposite.

Quoted 3 lines

I'll try and get one for building a x86_64 tools/perf, tools/lib/{api,bpf,traceevent} to see if I manage to reproduce the problem you're reporting.

Signature · 2 lines
Best regards,
Alexander Sverdlin.
#6 lore raw
In reply to Arnaldo Carvalho de Melo

Hello Arnaldo,

Quoted 13 lines
Quoted 13 lines
Quoted 13 lines
Quoted 13 lines

Em Wed, Sep 12, 2018 at 07:02:32PM +0200, Alexander Sverdlin escreveu:

Quoted 12 lines

CC .../tools/objtool/builtin-check.o
...
In file included from .../tools/arch/x86/include/uapi/asm/bitsperlong.h:11:0,

                 from .../tools/include/asm-generic/bitops/__ffs.h:6,
                 from .../tools/include/asm-generic/bitops.h:16,
                 from .../tools/include/linux/bitops.h:35,
                 from .../tools/include/linux/hashtable.h:13,
                 from elf.h:24,
                 from check.h:22,
                 from builtin-check.c:30:

.../tools/include/asm-generic/bitsperlong.h:8:0: error: "BITS_PER_LONG" redefined [-Werror]
#define BITS_PER_LONG (__CHAR_BIT__ * __SIZEOF_LONG__)

[...]

finally I more or less know what happens here. In the actual Linux we have two files defining the same define:

Quoted 1 line
Quoted 1 line

# 1 ".../tools/include/linux/bitops.h" 1

#ifndef BITS_PER_LONG

# define BITS_PER_LONG __WORDSIZE
#endif
Quoted 1 line
Quoted 1 line

# 6 ".../tools/include/asm-generic/bitsperlong.h" 2

#ifdef __SIZEOF_LONG__
#define BITS_PER_LONG (__CHAR_BIT__ * __SIZEOF_LONG__) #else
#define BITS_PER_LONG __WORDSIZE
#endif

So the two defines only work together if bitsperlong.h is included first. In objtool both files are included and for most people bitsperlong.h is indeed included first.

Quoted 3 lines

I'll try and get one for building a x86_64 tools/perf, tools/lib/{api,bpf,traceevent} to see if I manage to reproduce the problem you're reporting.

One way to reproduce the reverted include order is to take a HOST compiler built against old Linux headers, namely 2.6.30 and older.
This is because of the changes in asm/types.h.

It would be possible to put a guard into bitsperlong.h (#ifndef BITS_PER_LONG), but it just doesn't look correct to me that we have two files defining the same thing once quite simple, in the second case even more tricky.

Signature · 2 lines
Best regards,
Alexander Sverdlin.
#7 lore raw
In reply to Arnaldo Carvalho de Melo
Quoted 13 lines · Arnaldo Carvalho de Melo
Quoted 13 lines

CC .../tools/objtool/builtin-check.o
...
In file included from .../tools/arch/x86/include/uapi/asm/bitsperlong.h:11:0,

                 from .../tools/include/asm-generic/bitops/__ffs.h:6,
                 from .../tools/include/asm-generic/bitops.h:16,
                 from .../tools/include/linux/bitops.h:35,
                 from .../tools/include/linux/hashtable.h:13,
                 from elf.h:24,
                 from check.h:22,
                 from builtin-check.c:30:

.../tools/include/asm-generic/bitsperlong.h:8:0: error: "BITS_PER_LONG" redefined [-Werror]
#define BITS_PER_LONG (__CHAR_BIT__ * __SIZEOF_LONG__)

Quoted 4 lines
Quoted 1 line

Include <asm/bitsperlong.h> instead as other headers do.

Please try test building all tools in tools/

This broke make -C tools/perf/

I agree, this brakes perf, so we need to work further on that...

Signature · 2 lines
Best regards,
Alexander Sverdlin.