From: Ingo Molnar <mingo@elte.hu>
To: Marti Raudsepp <marti@juffo.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>,
Paul Mackerras <paulus@samba.org>,
Frederic Weisbecker <fweisbec@gmail.com>,
Arnaldo Carvalho de Melo <acme@redhat.com>,
Arjan van de Ven <arjan@linux.intel.com>,
Mike Galbraith <efault@gmx.de>,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] perf tools: add compatibility with libelf 0.8 and autodetect
Date: Fri, 23 Oct 2009 23:52:39 +0200 [thread overview]
Message-ID: <20091023215239.GB29403@elte.hu> (raw)
In-Reply-To: <1256334501-15755-1-git-send-email-marti@juffo.org>
* Marti Raudsepp <marti@juffo.org> wrote:
> On Sat, Oct 24, 2009 at 12:02 AM, Ingo Molnar <mingo@elte.hu> wrote:
> > Mind doing a small change:
> >
> > I think we want a small cleanup here: a perf_elf_begin() wrapper in a
> > header file to hide this #ifdef. (That's how Git wraps environmental
> > libraries as well.)
>
> Ok, does this patch do what you had in mind? I wasn't quite sure how to
> implement; changing the cmd behind someone's back might be considered
> surprising.
>
> Another take would be to #define our own ELF_C_READ_MMAP, but that would be
> lying, too...
>
> > Also, this would be an urgent fix for v2.6.32 too, to make perf build on
> > Arch Linux, agreed?
>
> Hopefully :)
>
> Regards,
> Marti
>
>
> [PATCH] perf tools: add compatibility with libelf 0.8 and autodetect
>
> The Makefile now automatically defines LIBELF_NO_MMAP when libelf 0.8 is
> detected. libelf 0.8 is still maintained and some distributions such as
> Arch Linux use it instead of elfutils.
>
> Signed-off-by: Marti Raudsepp <marti@juffo.org>
> ---
> tools/perf/Makefile | 6 +++++-
> tools/perf/util/symbol.c | 9 +++++----
> tools/perf/util/symbol.h | 11 +++++++++++
> 3 files changed, 21 insertions(+), 5 deletions(-)
>
> diff --git a/tools/perf/Makefile b/tools/perf/Makefile
> index 742a32e..46e877b 100644
> --- a/tools/perf/Makefile
> +++ b/tools/perf/Makefile
> @@ -422,7 +422,11 @@ ifeq ($(uname_S),Darwin)
> PTHREAD_LIBS =
> endif
>
> -ifneq ($(shell sh -c "(echo '\#include <libelf.h>'; echo 'int main(void) { Elf * elf = elf_begin(0, ELF_C_READ_MMAP, 0); return (long)elf; }') | $(CC) -x c - $(ALL_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -o /dev/null $(ALL_LDFLAGS) > /dev/null 2>&1 && echo y"), y)
> +ifeq ($(shell sh -c "(echo '\#include <libelf.h>'; echo 'int main(void) { Elf * elf = elf_begin(0, ELF_C_READ, 0); return (long)elf; }') | $(CC) -x c - $(ALL_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -o /dev/null $(ALL_LDFLAGS) > /dev/null 2>&1 && echo y"), y)
> + ifneq ($(shell sh -c "(echo '\#include <libelf.h>'; echo 'int main(void) { Elf * elf = elf_begin(0, ELF_C_READ_MMAP, 0); return (long)elf; }') | $(CC) -x c - $(ALL_CFLAGS) -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -o /dev/null $(ALL_LDFLAGS) > /dev/null 2>&1 && echo y"), y)
> + BASIC_CFLAGS += -DLIBELF_NO_MMAP
> + endif
> +else
> msg := $(error No libelf.h/libelf found, please install libelf-dev/elfutils-libelf-devel and glibc-dev[el]);
> endif
>
> diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
> index 47ea060..ec69c42 100644
> --- a/tools/perf/util/symbol.c
> +++ b/tools/perf/util/symbol.c
> @@ -5,7 +5,6 @@
>
> #include "debug.h"
>
> -#include <libelf.h>
> #include <gelf.h>
> #include <elf.h>
>
> @@ -413,7 +412,8 @@ static int dso__synthesize_plt_symbols(struct dso *self, int v)
> if (fd < 0)
> goto out;
>
> - elf = elf_begin(fd, ELF_C_READ_MMAP, NULL);
> + elf = perf_elf_begin(fd, ELF_C_READ, NULL);
> +
> if (elf == NULL)
> goto out_close;
>
> @@ -533,7 +533,8 @@ static int dso__load_sym(struct dso *self, int fd, const char *name,
> Elf *elf;
> int nr = 0, kernel = !strcmp("[kernel]", self->name);
>
> - elf = elf_begin(fd, ELF_C_READ_MMAP, NULL);
> + elf = perf_elf_begin(fd, ELF_C_READ, NULL);
> +
> if (elf == NULL) {
> if (v)
> fprintf(stderr, "%s: cannot read %s ELF file.\n",
> @@ -675,7 +676,7 @@ static char *dso__read_build_id(struct dso *self, int v)
> if (fd < 0)
> goto out;
>
> - elf = elf_begin(fd, ELF_C_READ_MMAP, NULL);
> + elf = perf_elf_begin(fd, ELF_C_READ, NULL);
> if (elf == NULL) {
> if (v)
> fprintf(stderr, "%s: cannot read %s ELF file.\n",
> diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h
> index 6e84907..50750e9 100644
> --- a/tools/perf/util/symbol.h
> +++ b/tools/perf/util/symbol.h
> @@ -8,6 +8,8 @@
> #include "module.h"
> #include "event.h"
>
> +#include <libelf.h>
> +
> #ifdef HAVE_CPLUS_DEMANGLE
> extern char *cplus_demangle(const char *, int);
>
> @@ -27,6 +29,15 @@ static inline char *bfd_demangle(void __used *v, const char __used *c,
> #endif
> #endif
>
> +static inline Elf *perf_elf_begin(int fd, Elf_Cmd cmd, Elf *ref)
> +{
> +#ifndef LIBELF_NO_MMAP
> + if (cmd == ELF_C_READ)
> + cmd = ELF_C_READ_MMAP;
> +#endif
> + return elf_begin(fd, cmd, ref);
> +}
> +
> #ifndef DMGL_PARAMS
> #define DMGL_PARAMS (1 << 0) /* Include function args */
> #define DMGL_ANSI (1 << 1) /* Include const, volatile, etc */
Looks nice!
Mind adding a comment as well to document the version dependency?
Also, there's a few probably unintended whitespaces in the patch not
matching surrounding style (which is the kernel coding style) - please
run the patch through scripts/checkpatch.pl.
Ingo
next prev parent reply other threads:[~2009-10-23 21:52 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-10-23 20:37 [PATCH] perf tools: add compatibility with libelf 0.8 and autodetect Marti Raudsepp
2009-10-23 21:02 ` Ingo Molnar
2009-10-23 21:48 ` Marti Raudsepp
2009-10-23 21:52 ` Ingo Molnar [this message]
2009-10-23 22:10 ` Marti Raudsepp
2009-10-24 1:07 ` Lucas De Marchi
2009-10-24 1:18 ` Marti Raudsepp
2009-10-24 8:30 ` Ingo Molnar
2009-10-24 13:26 ` Lucas De Marchi
2009-10-24 15:13 ` Marti Raudsepp
2009-10-24 15:19 ` Marti Raudsepp
2009-10-24 15:41 ` Ingo Molnar
2009-10-24 16:10 ` Marti Raudsepp
2009-10-23 21:56 ` Lucas De Marchi
2009-10-23 21:59 ` Marti Raudsepp
2009-10-23 22:16 ` Lucas De Marchi
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=20091023215239.GB29403@elte.hu \
--to=mingo@elte.hu \
--cc=a.p.zijlstra@chello.nl \
--cc=acme@redhat.com \
--cc=arjan@linux.intel.com \
--cc=efault@gmx.de \
--cc=fweisbec@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=marti@juffo.org \
--cc=paulus@samba.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.