linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Arnaldo Carvalho de Melo <arnaldo.melo@gmail.com>
To: Thomas Backlund <tmb@mageia.org>
Cc: Song Liu <liu.song.a23@gmail.com>,
	Linux List Kernel Mailing <linux-kernel@vger.kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@kernel.org>
Subject: Re: perf build broken in 5.1-rc7
Date: Wed, 1 May 2019 09:07:51 -0400	[thread overview]
Message-ID: <20190501130751.GB21436@kernel.org> (raw)
In-Reply-To: <b773df70-58e6-69f8-d566-282b0f7ae579@mageia.org>

Em Tue, Apr 30, 2019 at 04:31:14PM +0300, Thomas Backlund escreveu:
> 
> Den 30-04-2019 kl. 16:06, skrev Song Liu:
> > On Tue, Apr 30, 2019 at 12:55 AM Thomas Backlund <tmb@mageia.org> wrote:
> > > Den 30-04-2019 kl. 10:26, skrev Thomas Backlund:
> > > > Building perf in 5.1-rc5/6/7 fails:

> > > > Build start:

> > > >    make -s -C tools/perf NO_PERF_READ_VDSO32=1 NO_PERF_READ_VDSOX32=1
> > > > WERROR=0 NO_LIBUNWIND=1 HAVE_CPLUS_DEMANGLE=1 NO_GTK2=1 NO_STRLCPY=1
> > > > NO_BIONIC=1 NO_JVMTI=1 prefix=/usr lib=lib64 all
> > > >     BUILD:   Doing 'make -j32' parallel build
> > > >     HOSTCC   fixdep.o
> > > >     HOSTLD   fixdep-in.o
> > > >     LINK     fixdep
> > > > Warning: Kernel ABI header at 'tools/arch/x86/include/uapi/asm/vmx.h'
> > > > differs from latest version at 'arch/x86/include/uapi/asm/vmx.h'
> > > > diff -u tools/arch/x86/include/uapi/asm/vmx.h
> > > > arch/x86/include/uapi/asm/vmx.h
> > > > 
> > > > Auto-detecting system features:
> > > > ...                         dwarf: [ on  ]
> > > > ...            dwarf_getlocations: [ on  ]
> > > > ...                         glibc: [ on  ]
> > > > ...                          gtk2: [ on  ]
> > > > ...                      libaudit: [ on  ]
> > > > ...                        libbfd: [ on  ]
> > > > ...                        libelf: [ on  ]
> > > > ...                       libnuma: [ on  ]
> > > > ...        numa_num_possible_cpus: [ on  ]
> > > > ...                       libperl: [ on  ]
> > > > ...                     libpython: [ on  ]
> > > > ...                      libslang: [ on  ]
> > > > ...                     libcrypto: [ on  ]
> > > > ...                     libunwind: [ on  ]
> > > > ...            libdw-dwarf-unwind: [ on  ]
> > > > ...                          zlib: [ on  ]
> > > > ...                          lzma: [ on  ]
> > > > ...                     get_cpuid: [ on  ]
> > > > ...                           bpf: [ on  ]
> > > > ...                        libaio: [ on  ]
> > > > ...        disassembler-four-args: [ OFF ]
> > > > 
> > > > Makefile.config:473: No sys/sdt.h found, no SDT events are defined,
> > > > please install systemtap-sdt-devel or systemtap-sdt-dev
> > > > Makefile.config:853: No libbabeltrace found, disables 'perf data' CTF
> > > > format support, please install libbabeltrace-dev[el]/libbabeltrace-ctf-dev
> > > > 
> > > > 
> > > > And breaks with:
> > > > 
> > > > 
> > > > CC       ui/setup.o
> > > > util/annotate.c: In function 'symbol__disassemble_bpf':
> > > > util/annotate.c:1767:29: error: incompatible type for argument 1 of
> > > > 'disassembler'
> > > >     disassemble = disassembler(bfdf);
> > > >                                ^~~~
> > > > In file included from util/annotate.c:1689:
> > > > /usr/include/dis-asm.h:325:63: note: expected 'enum bfd_architecture'
> > > > but argument is of type 'bfd *' {aka 'struct bfd *'}
> > > >    extern disassembler_ftype disassembler (enum bfd_architecture arc,
> > > >                                            ~~~~~~~~~~~~~~~~~~~~~~^~~
> > > > util/annotate.c:1767:16: error: too few arguments to function
> > > > 'disassembler'
> > > >     disassemble = disassembler(bfdf);
> > > >                   ^~~~~~~~~~~~
> > > > In file included from util/annotate.c:1689:
> > > > /usr/include/dis-asm.h:325:27: note: declared here
> > > >    extern disassembler_ftype disassembler (enum bfd_architecture arc,
> > > >                              ^~~~~~~~~~~~
> > > >     CC       arch/x86/util/header.o
> > > >     CC       arch/x86/util/tsc.o
> > > >     CC       arch/x86/util/pmu.o
> > > > mv: cannot stat 'util/.annotate.o.tmp': No such file or directory
> > > >     CC       bench/futex-requeue.o
> > > >     CC       arch/x86/util/kvm-stat.o
> > > > make[4]: ***
> > > > [/work/rpmbuild/BUILD/kernel-x86_64/linux-5.0/tools/build/Makefile.build:97:
> > > > util/annotate.o] Error 1
> > > > make[4]: *** Waiting for unfinished jobs....
> > > >     CC       util/build-id.o
> > > > 
> > > > 
> > > > 
> > > 
> > > And I forgot...
> > > 
> > > Reverting:
> > >   From 6987561c9e86eace45f2dbb0c564964a63f4150a Mon Sep 17 00:00:00 2001
> > > From: Song Liu <songliubraving@fb.com>
> > > Date: Mon, 11 Mar 2019 22:30:48 -0700
> > > Subject: perf annotate: Enable annotation of BPF programs
> > > 
> > > Makes it build again.
> > > 
> > > --
> > > Thomas
> > > 
> > Hi Thomas,
> > 
> > Which system are you running this test on? I would like to repro it in a VM.
> > 
> > Thanks,
> > Song
> 
> 
> Mageia Cauldron currently stabilizing to become Mageia 7 in ~1 month.
> 
> 
> Basesystem is:
> 
> binutils-2.32-5.mga7
> (includes all fixes from upstream binutils-2_32-branch)
> 
> gcc-8.3.1-0.20190419.2.mga7
> 
> glibc-2.29-7.mga7
> (includes all fixes from upstream glibc release/2.29/master branch up to
> 2019-04-15 for now)
> 
> 
> kernel-desktop-5.1.0-0.rc7.1.mga7
> kernel-userspace-headers-5.1.0-0.rc7.1.mga7

Ok, so the steps are:

1) the feature test, the small C program that we try to build is:

[acme@quaco perf]$ cat tools/build/feature/test-disassembler-four-args.c
// SPDX-License-Identifier: GPL-2.0
#include <bfd.h>
#include <dis-asm.h>

int main(void)
{
	bfd *abfd = bfd_openr(NULL, NULL);

	disassembler(bfd_get_arch(abfd),
		     bfd_big_endian(abfd),
		     bfd_get_mach(abfd),
		     abfd);

	return 0;
}
[acme@quaco perf]$

And here in my fedora29 system it ends up producing the following file,
when built with:

$ make O=/tmp/build/perf  -C tools/perf install-bin

[acme@quaco perf]$ cat /tmp/build/perf/feature/test-disassembler-four-args.make.output 
[acme@quaco perf]$ 
[acme@quaco perf]$ file /tmp/build/perf/feature/test-disassembler-four-args.bin
/tmp/build/perf/feature/test-disassembler-four-args.bin: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked, interpreter /lib64/ld-linux-x86-64.so.2, for GNU/Linux 3.2.0, BuildID[sha1]=c9bd83db766a620c5cb6d756b0cd6991527641ff, not stripped, too many notes (256)
[acme@quaco perf]$ ldd /tmp/build/perf/feature/test-disassembler-four-args.bin
	linux-vdso.so.1 (0x00007ffebc5f9000)
	libz.so.1 => /lib64/libz.so.1 (0x00007fed2da04000)
	libdl.so.2 => /lib64/libdl.so.2 (0x00007fed2d9fe000)
	libc.so.6 => /lib64/libc.so.6 (0x00007fed2d838000)
	/lib64/ld-linux-x86-64.so.2 (0x00007fed2da3e000)
[acme@quaco perf]$

Meaning it built properly, so in this sytem the disassembler() function
has indeed four args, so we end up with:

[acme@quaco perf]$ grep disassembler /tmp/build/perf/FEATURE-DUMP 
feature-disassembler-four-args=1
[acme@quaco perf]$ 

Can you check the output for
/tmp/build/perf/feature/test-disassembler-four-args.make.output in your
system? And also check what is the prototype for the disassembler()
routine on mageia7?

Here I have:

[acme@quaco perf]$ rpm -q binutils
binutils-2.31.1-25.fc29.x86_64
[acme@quaco perf]$

Perhaps binutils 2.32 changed that prototype again and instead of
falling back to using just one arg we need to use some other number of
args, or even a different type for the N args it now maybe have?

- Arnaldo

  parent reply	other threads:[~2019-05-01 13:07 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-04-30  7:26 perf build broken in 5.1-rc7 Thomas Backlund
2019-04-30  7:53 ` Thomas Backlund
2019-04-30 13:06   ` Song Liu
2019-04-30 13:31     ` Thomas Backlund
2019-05-01  3:37       ` Song Liu
2019-05-01  6:42         ` Thomas Backlund
2019-05-01 13:07       ` Arnaldo Carvalho de Melo [this message]
2019-05-01 14:09         ` Thomas Backlund
2019-05-01 15:55           ` Thomas Backlund
2019-05-01 17:31           ` Arnaldo Carvalho de Melo
2019-05-01 20:20             ` Thomas Backlund
2019-05-01 20:26               ` Arnaldo Carvalho de Melo
2019-05-03  5:57             ` [tip:perf/urgent] tools build: Add -ldl to the disassembler-four-args feature test tip-bot for Arnaldo Carvalho de Melo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190501130751.GB21436@kernel.org \
    --to=arnaldo.melo@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=liu.song.a23@gmail.com \
    --cc=mingo@kernel.org \
    --cc=peterz@infradead.org \
    --cc=tmb@mageia.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).