Hi Michael,
On Wed, 2024-10-16 at 19:53 +, Michael Pratt wrote:
> These "unlocked" Linux Standard Base variants of standard functions
> are not available on some systems that are still capable
> of building Linux and ELFs.
>
> The difference is negligible for simple printing to stdout.
>
> P
Track the method used to unwind each Dwfl_Frame using an enum field
unwound_source; provide access to the field. Then use that in
eu-stacktrace to display the unwind methods used for each process.
This is an important diagnostic to verify how many processes are
adequately covered by the eh_frame u
Since we obtain diagnostics about unwind method, another logical place
to show them is in eu-stack.
* src/stack.c (show_unwound_source): New global variable.
(struct frame): Add unwound_source field.
(frame_callback): Copy over unwound_source from Dwfl_Frame.
(print_frame): Take unwound_sour
Signed-off-by: Serhei Makarov
---
NEWS | 7 +++
1 file changed, 7 insertions(+)
diff --git a/NEWS b/NEWS
index 410bccc1..01909c0c 100644
--- a/NEWS
+++ b/NEWS
@@ -8,6 +8,13 @@ debuginfod: Add per-file signature verification for integrity
debuginfod: New API for metadata queries: file name
eu-stacktrace is a utility to process a stream of raw stack
samples (such as those obtained from the Linux kernel's
PERF_SAMPLE_STACK facility) into a stream of stack traces (such as
those obtained from PERF_SAMPLE_CALLCHAIN), freeing other profiling
utilities from having to implement their own bac
Due to the x86-specific code in the initial version the configury has
significant restrictions. If --enable-stacktrace is not explicitly
provided, then eu-stacktrace will be disabled by default.
The way we test for x86 is a bit unusual. What we actually care about
is that the register file provide
Hi Serhei,
On Thu, Oct 17, 2024 at 02:54:34PM -0400, Serhei Makarov wrote:
> Since we obtain diagnostics about unwind method, another logical place
> to show them is in eu-stack.
>
> * src/stack.c (show_unwound_source): New global variable.
> (struct frame): Add unwound_source field.
> (frame
On Thu, Oct 17, 2024 at 2:51 PM Serhei Makarov wrote:
>
> Due to the x86-specific code in the initial version the configury has
> significant restrictions. If --enable-stacktrace is not explicitly
> provided, then eu-stacktrace will be disabled by default.
>
> The way we test for x86 is a bit unus
On Thu, Oct 17, 2024 at 2:53 PM Serhei Makarov wrote:
>
> Track the method used to unwind each Dwfl_Frame using an enum field
> unwound_source; provide access to the field. Then use that in
> eu-stacktrace to display the unwind methods used for each process.
>
> This is an important diagnostic to
On Thu, Oct 17, 2024 at 2:56 PM Serhei Makarov wrote:
>
> Signed-off-by: Serhei Makarov
> ---
> NEWS | 7 +++
> 1 file changed, 7 insertions(+)
>
> diff --git a/NEWS b/NEWS
> index 410bccc1..01909c0c 100644
> --- a/NEWS
> +++ b/NEWS
> @@ -8,6 +8,13 @@ debuginfod: Add per-file signature verif
On Thu, Oct 17, 2024 at 3:41 PM Mark Wielaard wrote:
>
> Hi Serhei,
>
> On Thu, Oct 17, 2024 at 02:54:34PM -0400, Serhei Makarov wrote:
> > Since we obtain diagnostics about unwind method, another logical place
> > to show them is in eu-stack.
> >
> > * src/stack.c (show_unwound_source): New globa
On Thu, Oct 17, 2024 at 2:49 PM Serhei Makarov wrote:
>
> eu-stacktrace is a utility to process a stream of raw stack
> samples (such as those obtained from the Linux kernel's
> PERF_SAMPLE_STACK facility) into a stream of stack traces (such as
> those obtained from PERF_SAMPLE_CALLCHAIN), freeing
From: Di Chen
Provide a public function for removing debug section relocations.
eu-strip previously contained the code to remove debug section
relocations. This patch moves that code into
dwelf_elf_remove_debug_relocs.
https://sourceware.org/bugzilla/show_bug.cgi?id=31447
Signed-off-by: Di Ch
On Thu, Oct 17, 2024 at 11:59 AM Mark Wielaard wrote:
>
> elf_compress would compress all (new) data chunks, but didn't reset
> the section data_list. This would cause extra data to be returned
> after decompression or create bad compressed data. Add a new testcase
> for this and explicitly zap th
Signed-off-by: Aaron Merey
---
config/elfutils.spec.in | 1 +
doc/Makefile.am | 3 +-
doc/libelf.3| 363
3 files changed, 366 insertions(+), 1 deletion(-)
create mode 100644 doc/libelf.3
diff --git a/config/elfutils.spec.in b/conf
On Thu, Oct 17, 2024 at 11:28 AM Serhei Makarov wrote:
>
> On Wed, Oct 16, 2024, at 11:38 PM, Aaron Merey wrote:
> > Using less wide indentation throughout would help make the configure.ac
> > changes more readable. Otherwise this patch LGTM.
> Ok, my code imitated the offset for the debuginfod c
On Wed, Oct 16, 2024, at 11:37 PM, Aaron Merey wrote:
> Please add an eu-stacktrace entry to NEWS.
Ok, will do.
> I tried to run this but I kept getting:
> Failed to complete recording: GDBus.Error[...]: Action
> org.gnome.sysprof3.profile is not registered
Right. To have sysprof work requ
https://sourceware.org/bugzilla/show_bug.cgi?id=32102
Mark Wielaard changed:
What|Removed |Added
Assignee|unassigned at sourceware dot org |mark at klomp dot org
On Thu, Oct 17, 2024 at 11:50 AM Serhei Makarov wrote:
>
> > I'm comfortable with this being merged for the release as long as any
> > build errors are sorted out and we emphasize in the usage text and
> > NEWS that eu-stacktrace is an experimental feature with a possibly
> > unstable ABI.
> I thi
Hi Serhei,
On Tue, 2024-10-15 at 11:26 -0400, Serhei Makarov wrote:
> Since we obtain diagnostics about unwind method, another logical place
> to show them is in eu-stack.
>
> This requires an additional pseudo-unwind type DWFL_UNWOUND_INLINE to
> clarify what happens with entries for inlined fun
On Wed, Oct 16, 2024, at 11:39 PM, Aaron Merey wrote:
> Hi Serhei,
>
> On Tue, Oct 15, 2024 at 11:28 AM Serhei Makarov wrote:
>>
>> Track the method used to unwind each Dwfl_Frame using an enum field
>> unwound_source; provide access to the field. Then use that in
>> eu-stacktrace to display t
On Wed, Oct 16, 2024, at 11:38 PM, Aaron Merey wrote:
> Using less wide indentation throughout would help make the configure.ac
> changes more readable. Otherwise this patch LGTM.
Ok, my code imitated the offset for the debuginfod configury immediately above,
but I'll update the patch to use 3
elf_compress would compress all (new) data chunks, but didn't reset
the section data_list. This would cause extra data to be returned
after decompression or create bad compressed data. Add a new testcase
for this and explicitly zap the scn->data_list before resetting the
elf section raw data after
On Thu, Oct 17, 2024, at 12:18 PM, Aaron Merey wrote:
> I'm seeing eu-stacktrace built by default on x86 with no
> --enable-stacktrace given to configure.
Now I'm waffling back and forth on what to do.
Mark wants to enable eu-stacktrace build on one of the buildbots.
The current behaviour (whic
On Thu, Oct 17, 2024, at 1:00 PM, Mark Wielaard wrote:
> This part looks a little iffy IMHO.
> It feels like a layering violation.
> The meaning of DWFL_UNWOUND_INLINE is also different from the others.
> There is no worst/best ordering here.
> They are not connected to frames.
>
> Is there a wa
Hi Serhei,
On Thu, 2024-10-17 at 13:13 -0400, Serhei Makarov wrote:
> On Thu, Oct 17, 2024, at 1:00 PM, Mark Wielaard wrote:
> > This part looks a little iffy IMHO.
> > It feels like a layering violation.
> > The meaning of DWFL_UNWOUND_INLINE is also different from the others.
> > There is no wor
On Thu, Oct 17, 2024 at 1:09 PM Serhei Makarov wrote:
> On Thu, Oct 17, 2024, at 12:18 PM, Aaron Merey wrote:
> > I'm seeing eu-stacktrace built by default on x86 with no
> > --enable-stacktrace given to configure.
> Now I'm waffling back and forth on what to do.
>
> Mark wants to enable eu-stackt
On Thu, Oct 17, 2024, at 1:22 PM, Mark Wielaard wrote:
> I think (but haven't tried) only print_frame has to have a char *
> argument, print_inline_frames and print_frames can still pass around an
> Dwfl_Unwound_Source. Both print_frames and print_inline_frames can then
> call print_frame with t
28 matches
Mail list logo
