debuginfod: use single ipv4+ipv6 microhttpd daemon configuration
AUTHORS: Use generator script & git mailmap
Luca Boccassi (1):
libebl: recognize FDO Packaging Metadata ELF note
Mark Wielaard (71):
tests: Don't set DEBUGINFOD_TIMEOUT
tests: Add -rdynamic to dwfl_proc_attach_L
on, which
would be slightly ugly imho. I had hoped it could be a variant of the
func_addr_mask. But maybe this is too different to make more generic.
Cheers,
Mark
From: Mark Wielaard
make distcheck would always run with --enable-valgrind and
--enable-sanitize-undefined. Assuming the maintainer would run this just
before doing a release. But this makes distcheck really, really, really
slow. And we have various buildbots now that run various combinations
of
gt;
> Build state: build successful
> Revision: 6d0c3b57429add5b6494259c1c06499e52cff4d5
> Worker: fedora-x86_64
> Build Reason: (unknown)
> Blamelist: Mark Wielaard
This is a little cheating. I removed the --enable-valgrind from make
distcheck and this failed before because of s
csh/tcsh would warn about "Ambiguous output redirect" if not done inside
the sh -c command.
Fix-by: наб
https://bugzilla.redhat.com/show_bug.cgi?id=2080957
Signed-off-by: Mark Wielaard
---
config/ChangeLog | 4
config/profile.csh.in | 2 +-
2 files changed, 5 insert
y the asan or ubsan "interceptor". But if
so, why haven't we seen this before?
Cheers,
Mark
igned-off-by: Mark Wielaard
---
debuginfod/ChangeLog | 7
debuginfod/debuginfod.cxx | 74 +++
2 files changed, 58 insertions(+), 23 deletions(-)
diff --git a/debuginfod/ChangeLog b/debuginfod/ChangeLog
index 619ebd8c..e108f847 100644
--- a/debug
Also disable MHD_USE_THREAD_PER_CONNECTION when using MHD_USE_EPOLL.
https://sourceware.org/bugzilla/show_bug.cgi?id=29123
Signed-off-by: Mark Wielaard
---
debuginfod/ChangeLog | 6 ++
debuginfod/debuginfod.cxx | 15 ++-
2 files changed, 16 insertions(+), 5 deletions
This was the only place in debuginfod-client.c where we didn't check
the result of curl_easy_getinfo. Just check it to make things consistent.
Signed-off-by: Mark Wielaard
---
debuginfod/ChangeLog | 5 +
1 file changed, 5 insertions(+)
diff --git a/debuginfod/ChangeLog b/debug
This was the only place in debuginfod-client.c where we didn't check
the result of curl_easy_getinfo. Just check it to make things consistent.
Signed-off-by: Mark Wielaard
---
v2 This time with the actual code change...
debuginfod/ChangeLog | 5 +
debuginfod/debuginfod-cli
curl_easy_setup can fail for various reasons. Add a curl_easy_setopt_ck
macro to check all curl_easy_setopt calls and provides a human readable
error message in verbose mode.
Signed-off-by: Mark Wielaard
---
debuginfod/ChangeLog | 5 +++
debuginfod/debuginfod-client.c | 61
Don't just skip the block length, but check it is equal to the
op->number that we are going to use as length.
Signed-off-by: Mark Wielaard
---
libdw/ChangeLog | 4
libdw/dwarf_getlocation.c | 5 -
2 files changed, 8 insertions(+), 1 deletion(-)
diff --git a/libdw/Cha
When gelf_getshdr, gelf_getrela, gelf_getrel or gelf_getsymshndx
return NULL it is an internal error which we want to report instead of
crashing.
Signed-off-by: Mark Wielaard
---
src/ChangeLog | 5 +
src/strip.c | 12 ++--
2 files changed, 15 insertions(+), 2 deletions(-)
diff
always request both hostname and servname
with NI_NUMERICHOST | NI_NUMERICSERV.
Signed-off-by: Mark Wielaard
---
debuginfod/ChangeLog | 4
debuginfod/debuginfod.cxx | 15 ++-
2 files changed, 14 insertions(+), 5 deletions(-)
diff --git a/debuginfod/ChangeLog b/debuginfod
lso try to clean up the
cache dir. Because that is just duplicate work.
[PATCH 3/3] debuginfod: update mtime of interval_path as early as possible
Cheers,
Mark
d for either writing the default
value or reading the config file value.
Signed-off-by: Mark Wielaard
---
debuginfod/ChangeLog | 6 ++
debuginfod/debuginfod-client.c | 21 ++---
2 files changed, 20 insertions(+), 7 deletions(-)
diff --git a/debuginfod/Change
emove debuginfod_init_cache before calling
debuginfod_clean_cache.
Signed-off-by: Mark Wielaard
---
debuginfod/ChangeLog | 6
debuginfod/debuginfod-client.c | 61 +-
2 files changed, 15 insertions(+), 52 deletions(-)
diff --git a/debuginfod/Change
will try to delete the files in the same order.
Signed-off-by: Mark Wielaard
---
debuginfod/ChangeLog | 5 +
debuginfod/debuginfod-client.c | 7 +--
2 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/debuginfod/ChangeLog b/debuginfod/ChangeLog
index c9aa4fcf..209
The process_file code is a little tricky. Add sanity checks to make
sure shstrtab_name, shstrtab_newname and symstrents are only set once.
Signed-off-by: Mark Wielaard
---
src/ChangeLog | 5 +
src/elfcompress.c | 24
2 files changed, 29 insertions(+)
diff
Hi,
On Fri, May 06, 2022 at 12:37:35AM +0200, Mark Wielaard wrote:
> On a systems that have ipv6 disabled debuginfod doesn't start up
> anymore because libhttpd MHD_USE_DUAL_STACK only works if it can
> open an ipv6 socket. If MHD_start_daemon with MHD_USE_DUAL_STACK
> fails
On Fri, May 06, 2022 at 12:38:40AM +0200, Mark Wielaard wrote:
> Also disable MHD_USE_THREAD_PER_CONNECTION when using MHD_USE_EPOLL.
Pushed after OK from Frank on irc and some testing in Fedora.
Cheers,
Mark
Hi,
On Tue, May 10, 2022 at 12:35:01PM +0200, Mark Wielaard wrote:
> The process_file code is a little tricky. Add sanity checks to make
> sure shstrtab_name, shstrtab_newname and symstrents are only set once.
Pushed.
Cheers,
Mark
Hi,
On Mon, May 09, 2022 at 05:40:39PM +0200, Mark Wielaard wrote:
> This was the only place in debuginfod-client.c where we didn't check
> the result of curl_easy_getinfo. Just check it to make things consistent.
Pushed.
Cheers,
Mark
Hi,
On Mon, May 09, 2022 at 06:22:29PM +0200, Mark Wielaard wrote:
> curl_easy_setup can fail for various reasons. Add a curl_easy_setopt_ck
> macro to check all curl_easy_setopt calls and provides a human readable
> error message in verbose mode.
Pushed.
Cheers,
Mark
Hi,
On Mon, May 09, 2022 at 06:48:58PM +0200, Mark Wielaard wrote:
> Don't just skip the block length, but check it is equal to the
> op->number that we are going to use as length.
Pushed.
Cheers,
Mark
Hi,
On Mon, May 09, 2022 at 10:24:35PM +0200, Mark Wielaard wrote:
> When gelf_getshdr, gelf_getrela, gelf_getrel or gelf_getsymshndx
> return NULL it is an internal error which we want to report instead of
> crashing.
Pushed.
Cheers,
Mark
s/27
>
> Build state: failed test (failure)
> Revision: 0f2c8af3ba82ac54acc63ab7cb068578a6c0d96e
> Worker: debian-armhf
> Build Reason: (unknown)
> Blamelist: Mark Wielaard
>
> Steps:
> [...]
> - 7: make check ( failure )
> Logs:
> - stdio:
> http
Hi,
On Mon, May 09, 2022 at 11:10:44PM +0200, Mark Wielaard wrote:
> When getting the connection info getnameinfo is called getting the
> hostname and servname except when the sockaddr is a pure ipv6
> address. In that last case only hostname is requested. Since servname
> is stack a
Hi,
On Tue, May 10, 2022 at 12:57:20AM +0200, Mark Wielaard wrote:
> debuginfod-client would try to create the cache config files twice,
> once through debuginfod_init_cache, which was always called before the
> debuginfod_clean_cache check. Which called debuginfod_config_cache
> whic
Make clear that both the offset and sym arguments cannot be NULL.
https://bugzilla.redhat.com/show_bug.cgi?id=1986555
Signed-off-by: Mark Wielaard
---
libdwfl/ChangeLog | 5 +
libdwfl/libdwfl.h | 32
2 files changed, 21 insertions(+), 16 deletions(-)
diff
Hi,
On Thu, May 05, 2022 at 09:01:24PM +0800, Di Chen via Elfutils-devel wrote:
> From 3ac23c2584d76114deab0c0af6f4af99068dc7f4 Mon Sep 17 00:00:00 2001
> From: Di Chen
> Date: Thu, 28 Apr 2022 19:55:33 +0800
> Subject: [PATCH] readelf: Support --dynamic with --use-dynamic
>
> Currently, eu-read
Hi,
On Sun, 2022-05-15 at 22:00 +0200, Mark Wielaard wrote:
> Make clear that both the offset and sym arguments cannot be NULL.
>
> https://bugzilla.redhat.com/show_bug.cgi?id=1986555
I didn't get feedback on the actual wording, but I think this is better
than we had. So pushed.
Cheers,
Mark
chive_test c36708a78618d597dee15d0dc989f093ca5f9120
> /usr/src/debug/hello2-1.0-2.x86_64/hello.c $SHA
> archive_test 41a236eb667c362a1c4196018cc4581e09722b1b
> /usr/src/debug/hello2-1.0-2.x86_64/hello.c $SHA
> archive_test bc1febfd03ca05e030f0d205f7659db29f8a4b30
> /usr/src/debug/hello-1.0/hello.c $SHA
> archive_test f0aa15b8aba4f3c28cac3c2a73801fefa644a9f2
> /usr/src/debug/hello-1.0/hello.c $SHA
> archive_test bbbf92ebee5228310e398609c23c2d7d53f6e2f9
> /usr/src/debug/hello-1.0/hello.c $SHA
> -archive_test d44d42cbd7d915bc938c81333a21e355a6022fb7
> /usr/src/debug/hello-1.0/hello.c $SHA
OK, test that the hello.c source file is found in the scanned archives,
that the fdcache is used and that the sha1sum checks out.
> metrics=$(curl http://127.0.0.1:$PORT1/metrics)
> -regex="fdcache_bytes ([0-9]+).*fdcache_prefetch_bytes ([0-9]+)"
> +regex="fdcache_bytes ([0-9]+)"
> mb=$((1024*1024))
> if [[ $metrics =~ $regex ]]; then
>fdbytes=${BASH_REMATCH[1]}
> - pfbytes=${BASH_REMATCH[2]}
> - if [ $fdbytes -gt $mb ] || [ $pfbytes -gt $mb ]; then
> + if [ $fdbytes -gt $mb ] ; then
> err
>fi
OK, check that the prefetch cache isn't bigger than 10MB. Why that
number?
Thanks,
Mark
gt;
> Build state: failed test (failure)
> Revision: b8713b3fd0617415c76df8c9da70f8e2f26d3134
> Worker: centos-x86_64
> Build Reason: (unknown)
> Blamelist: Mark Wielaard
>
> Steps:
> [...]
> - 7: make check ( failure )
> Logs:
> - stdio:
> https://
6.6G
Swap: 511M 0B511M
So maybe / was slightly full. The buildbot itself runs under /srv/
I freed 2.5G from / and uninstalled abrt so it doesn't intercept
generating the core files.
Cheers,
Mark
s/33
>
> Build state: failed test (failure)
> Revision: b8713b3fd0617415c76df8c9da70f8e2f26d3134
> Worker: centos-x86_64
> Build Reason: (unknown)
> Blamelist: Mark Wielaard
>
> - 7: make check ( failure )
> Logs:
> - stdio:
> https://builder.sourceware.
On Sat, May 28, 2022 at 11:35:19AM +0200, Mark Wielaard wrote:
> Retrying with ulimit -c unlimited...
>
> O! It failed again, now with core
>
> (gdb) where
> #0 0x7f4d528be387 in __GI_raise (sig=sig@entry=6) at
> ../nptl/sysdeps/unix/sysv/linux/raise.c:55
> #
Hi Noah,
Hope you don't mind me adding elfutils-devel back to the CC.
On Fri, 2022-05-27 at 15:32 -0400, Noah Sanci wrote:
> On Fri, May 27, 2022 at 12:29 PM Mark Wielaard
> wrote:
>
> > On Wed, 2022-05-25 at 14:44 -0400, Noah Sanci via Elfutils-devel
> > wrot
possible. But it
would be good to at least try and discuss it. If only so that elfutils
eu-strip and binutils strip agree on how to handle/coordinate on such
special sections.
Cheers,
Mark
> Signed-off-by: Guillermo E. Martinez
> ---
> ChangeLog |
back cp is moved past the NUL byte. It should
move to the NUL byte. Otherwise one cannot anything but the first
added flag description. Finally some cosmetic changes (space after
each comma in the output).
Signed-off-by: Mark Wielaard
---
Code is by Ulrich, I just added a quick testcase.
On Sat, 2022-05-28 at 11:43 +0200, Mark Wielaard wrote:
> On Sat, May 28, 2022 at 11:35:19AM +0200, Mark Wielaard wrote:
> > Retrying with ulimit -c unlimited...
> >
> > O! It failed again, now with core
> >
> > (gdb) where
> > #0 0x7f4d528b
Hi Noah,
On Mon, 2022-05-30 at 17:36 +0200, Mark Wielaard wrote:
> > Please find the new patch attached.
>
> Thanks for adding all the comments. Looks good.
The only issue is that although it seems to work well on Fedora I am
having problems getting to test to work on an older Cent
Hi Noah,
Adding the mailinglist back to CC so other can jump in if there is
anything obvious from the logs.
On Fri, 2022-06-03 at 10:51 -0400, Noah Sanci wrote:
> Nothing immediately jumps out, would you mind sending the entire log?
Attached.
Cheers,
Mark
curl is /usr/bin/curl
rpm2cpio
Hi,
On Fri, 2022-06-03 at 17:02 +0200, Mark Wielaard wrote:
> On Fri, 2022-06-03 at 10:51 -0400, Noah Sanci wrote:
> > Nothing immediately jumps out, would you mind sending the entire
> > log?
>
> Attached.
And Noah figured it out. The machine where it failed has /tmp 92%
On Wed, 2022-06-01 at 16:31 +0200, Mark Wielaard wrote:
> From: Ulrich Drepper
>
> Arm needs to decode flags and I modeled it after the binutils code.
> The same messages are printed. Given the requirement of the
> interface
> and the ABIs the current version of the callb
In both profile.csh.in and profile.sh.in we set and then unset
prefix, but never use it.
Signed-off-by: Mark Wielaard
---
config/ChangeLog | 5 +
config/profile.csh.in | 2 --
config/profile.sh.in | 2 --
3 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/config/ChangeLog
quot;, which requires the shell variable $prefix
> to be set.
O, I see. Indeed the default is '${prefix}/etc'. But I never saw that
because in a package install configure is called with everything
expanded [*]. Should have tried by hand.
Patch retracted.
Thanks,
Mark
[*] --pref
e_only, 2, 3)’
50 | extern ssize_t __REDIRECT (__pread_alias,
|^~
cc1: all warnings being treated as errors
That needs investigation. The ar file format is a little odd, those
fields are in (non-zero terminated) ASCII decimal notation. Which
probably confuses t
OR (INVALID_REGISTER, N_("Invalid register")) \
> + DWFL_ERROR (INVALID_REGNO, N_("Invalid register number")) \
So maybe call this UNKNOWN_REGISTER_VALUE?
>DWFL_ERROR (PROCESS_MEMORY_READ, N_("Error reading process memory"))
> \
>DWFL_ERROR (PROCESS_NO_ARCH, N_("Couldn't find architecture of any
> ELF")) \
>DWFL_ERROR (PARSE_PROC, N_("Error parsing /proc filesystem")) \
> @@ -786,6 +787,7 @@ INTDECL (dwfl_getthread_frames)
> INTDECL (dwfl_getthreads)
> INTDECL (dwfl_thread_getframes)
> INTDECL (dwfl_frame_pc)
> +INTDECL (dwfl_frame_reg)
> INTDECL (dwfl_get_debuginfod_client)
>
> /* Leading arguments standard to callbacks passed a Dwfl_Module. */
Cheers,
Mark
7227: st_value out of bounds
> ```
>
> Does anyone know how this can happen? Is this a bug in the toolchain?
Try with eu-elflint --gnu which suppresses some known issues.
Also could you show those symbol values (1272, 3684, 25720, 27227) they
might have a special type, so their st_value isn't really an address?
Cheers,
Mark
error disappears.
Signed-off-by: Mark Wielaard
---
libdwfl/ChangeLog | 5 +
libdwfl/core-file.c | 26 --
2 files changed, 21 insertions(+), 10 deletions(-)
diff --git a/libdwfl/ChangeLog b/libdwfl/ChangeLog
index 75c53948..acdaa013 100644
--- a/libdwfl/ChangeLog
+++ b/li
On Thu, 2022-07-28 at 15:48 +0200, Mark Wielaard wrote:
> With GCC 12.1.1, glibc 2.3a, -fsanitize=undefined and
> -D_FORTIFY_SOURCE=3 we get the following error message:
Sorry for the typo, it is glibc 2.35. Basically an up to date Fedora 36
system (replicated on x86_64, ppc64le and s390x).
/#/builders?tags=elfutils-try
My workflow to use this looks like:
- git checkout -b ar_size
- hack, hack, hack... OK, looks good to submit
- git commit -a -m "Now who got an ar_size?"
- git push origin ar_size:users/mark/try-ar_size
- ... wait for the emails to come in or watch bui
by
setting a proxy environment variable so the requests goes through a
local proxy that is setup to do some kind of authentication
transparently? Or by simply defining the base DEBUGINFOD_URL with
https://user:p...@debuginfod.example.com/ ?
Thanks,
Mark
e symbol names?
Currently through the section tables, which point to the string table
section used. But Di Chen's patch would change that by going through
the dynamic segment and phdrs to find the strtab for the dynsym segment
(but will of course still need to go through the sections for the
.symtab symbols since those aren't accessible through the phdrs).
Cheers,
Mark
Hi,
On Thu, 2022-07-28 at 15:48 +0200, Mark Wielaard wrote:
> With GCC 12.1.1, glibc 2.3a, -fsanitize=undefined and
> -D_FORTIFY_SOURCE=3 we get the following error message:
>
> In file included from /usr/include/ar.h:22,
> from ../libelf/libelfP.h:33,
>
implified, since it assumes you just have one DEBUGINFOD_URL
that needs the Authorization, but that seems fine, it seems an obscure
enough feature that it doesn't really matter if other servers get the
Authorization header too, they will just ignore it and don't even know
against what other server they could use the Bearer token.
Would any of the above work for your use case?
Cheers,
Mark
, 2022 at 9:27 PM Di Chen wrote:
>
> > Thanks for the review, Mark.
> >
> > Re-pushed a patch for it.
> >
> > 1. NEWS entry added.
> > 2. __libdwfl_frame_reg_get/dwfl_frame_reg updated with return int.
> > 3. New DWFL_ERROR added: REGISTER_VAL_UNKNOWN
b_data->d_buf) + dyn->d_un.d_ptr;
+ }
That does the check whether dyn->d_un.d_ptr is valid early, so it
doesn't need to be checked in each case statement. Also it adds an
extra memrchr check to make sure the string is zero terminated.
Pushed with those changes.
Thanks,
M
e, but don't immediately see which mistake I
made. Maybe I should use d_val instead of d_ptr (but those are both
uint64_t so that shouldn't really matter).
Cheers,
Mark
Hi,
On Mon, Aug 01, 2022 at 02:09:52AM +0200, Mark Wielaard wrote:
> > - 7: make check ( failure )
> > Logs:
> > - stdio:
> > https://builder.sourceware.org/buildbot/#builders/43/builds/47/steps/7/logs/stdio
> > - test-suite.log:
> > htt
gt;
> Build state: build successful
> Revision: d0ff4e224738adf34eba38dc33ffda67e5da6634
> Worker: fedora-s390x
> Build Reason: (unknown)
> Blamelist: Di Chen , Mark Wielaard
That is good! \o/
> A new failure has been detected on builder elfutils-opensusetw-x86_64 while
> b
gress function and afterwards if there was an error during download
(then we seem to clean them). Which probably good, but should be documented.
Also I probably won't mind providing all header, at least during the progress
function callback. But you might be right that they aren't as informative.
Cheers,
Mark
f programs just reuse the debuginfod_client
objects.
Or maybe it should just be one control file that can have entries for
all of the variables.
Cheers,
Mark
Document which lock is held by which thread and how moving the
htab data is coordinated.
Signed-off-by: Mark Wielaard
---
lib/ChangeLog| 8
lib/dynamicsizehash_concurrent.c | 26 --
2 files changed, 28 insertions(+), 6 deletions(-)
diff
The po/pl.po file already just said Project-Id-Version: elfutils\n
Do the same for the other po files which had somewhat odd names.
The generated pot file will still include the version number too.
Signed-off-by: Mark Wielaard
---
po/ChangeLog | 7 +++
po/de.po | 2 +-
po/es.po | 2
special case.
>
> https://sourceware.org/bugzilla/show_bug.cgi?id=29450
Thanks for finding this and suggesting a workaround. But lets first
try to fix binutils. This seems like a pretty bad bug, lets hope it
gets fixed soon. So we don't need these kind of workarounds.
I added a comment to the binutils bug:
https://sourceware.org/bugzilla/show_bug.cgi?id=29451#c2
Cheers,
Mark
Hi,
On Sat, Aug 06, 2022 at 09:55:06AM +0100, Sergei Trofimovich via Elfutils-devel
wrote:
> Noticed when debugged test failure:
>
> lowpc: 8049000, highpc: 8049000lx
> ../sysdeps/i386/crti.S: [2def] '_init' highpc <= lowpc
Thanks, looks obviously correct.
Pushed.
Cheers,
Mark
Hi Noah,
On Fri, Aug 05, 2022 at 03:21:48PM -0400, Noah Sanci wrote:
> On Thu, Aug 4, 2022 at 9:12 AM Mark Wielaard wrote:
> >
> > Hi Noah,
> >
> > On Thu, 2022-07-14 at 11:32 -0400, Noah Sanci via Elfutils-devel wrote:
> > > Please find the patch for pr28284
ic_tag_name was updated to handle the new RELR dynamic tags.
Thanks, looks good. Pushed.
Cheers,
Mark
ng a DT_SYMTAB_COUNT or
DT_SYMTABSZ? It is not intuitive that one can find the number of
entries through the hash table.
I don't think DT_GNU_HASH can simply be made optional and replace
DT_HASH without adding something like DT_SYMTAB_COUNT or DT_SYMTABSZ
because the gnu hashtable doesn't come with a simple symbol count. To
find the symbol count you have to go through the whole gnu hash table
to count the number of entries (which is a non-trivial amount of
work).
Cheers,
Mark
On Tue, Aug 09, 2022 at 09:54:28AM +0200, Andreas Schwab wrote:
> Handle PT_RISCV_ATTRIBUTES, SHT_RISCV_ATTRIBUTES, DT_RISCV_VARIANT_CC.
Thanks, looks good. Added a ChangeLog entry and pushed.
Cheers,
Mark
Hi Andreas,
On Tue, Aug 09, 2022 at 10:29:53AM +0200, Andreas Schwab wrote:
> The RISCV_ATTRIBUTES segment is not meant to be loaded.
That makes sense. Pushed.
Thanks,
Mark
the whole series first before asking questions
on some of the previous commits. So this is why you (already) tried to
replace putchar_unlocked with fputc.
If we are trying to be more consistent then I think I would prefer we
use putchar, puts and printf, instead of adding stdout to all these
calls
libcpu/i386_lex.l (%top): add flex %top block
> and move config.h header inclusion to it.
configure already checks we have flex, so this is fine.
Pushed,
Mark
Looks good and is more consistent. Pushed.
There isn't much code in this file. What kind of build issue are you
seeing without this? Does config.h contain some construct that affects
the build of this file?
Thanks,
Mark
utchar_unlocked(),
> so consistently use the safer fputc() instead.
I agree the use of the _unlocked variants is inconsistent and it would
be better to consistently use the standard versions.
But shouldn't we simply replace putchar_unlocked with putchar instead
of using fputc (c, stdout)?
Cheers,
Mark
in the rest of the project.
>
> * libdw/Makefile.am: move MAINTAINERCLEANFILES
> variable to the end of the file
> as a non-conditional definition.
You are right, there is no reason to only define MAINTAINERCLEANFILES
conditionally. Pushed.
Thanks,
Mark
Hi Aaron,
On Wed, 2024-10-02 at 22:26 -0400, Aaron Merey wrote:
> Signed-off-by: Aaron Merey
>
> ---
> v2 changes:
> Reword description.
>
> State EV_CURRENT is the only valid version.
>
> Mention possibility of integer overflow.
>
> On Tue, Aug 27, 2024
found in other libelf implementations".
>
> On Tue, Aug 27, 2024 at 1:23 PM Mark Wielaard wrote:
> >
> > It should also mention what the Elf32_Chdr structure looks
> > like and what the meaning the fields have. What the legal values of
> > ch_type are, that ch
Hi Aaron,
On Wed, 2024-10-02 at 22:26 -0400, Aaron Merey wrote:
> Signed-off-by: Aaron Merey
>
> ---
> v2 changes:
> elf64_checksum.3 now refers to elf32_checksum.3.
>
> Explained which sections are used to calculate the checksum.
>
> Mentioned DT_CHECKSUM.
>
> Mentioned that the checksum is
Hi Aaron,
On Wed, 2024-10-02 at 22:26 -0400, Aaron Merey wrote:
> Signed-off-by: Aaron Merey
> ---
>
> v2 changes:
> Mention elf_flagshdr.
>
> Mention NULL is returned if scn is NULL.
This version looks good to me.
Thanks,
Mark
Hi Aaron,
On Wed, 2024-10-02 at 22:26 -0400, Aaron Merey wrote:
> v2 changes:
> Mention new program header table is zero'ed out and that
> NULL is returned if elf is NULL.
Looks good.
Thanks,
Mark
version looks good to me.
Thanks,
Mark
Hi Aaron,
On Wed, 2024-10-02 at 22:26 -0400, Aaron Merey wrote:
> v2 changes:
> Mention that the existing header will be returned if one is
> already present.
>
> Mention that an ELF header must be present before callling
> elf_newscn or elf_newphdr
Looks good to me.
Thanks,
Mark
nts of the original
> sections
> tempfiles ${BUILDID}.debug_info
> objcopy F/prog.debug -O binary --only-section=.debug_info
> --set-section-flags .debug_info=alloc $BUILDID.debug_info
> -cmp ${BUILDID}.debug_info
> ${DEBUGINFOD_CACHE_PATH}/${BUILDID}/section-.debug_info
> +cmp ${BUILDID}.debug_info
> ${DEBUGINFOD_CACHE_PATH}/${BUILDID}/section-*.debug_info
>
> tempfiles ${BUILDID}.text
> objcopy F/prog -O binary --only-section=.text ${BUILDID}.text
> -cmp ${BUILDID}.text ${DEBUGINFOD_CACHE_PATH}/${BUILDID}/section-.text
> +cmp ${BUILDID}.text ${DEBUGINFOD_CACHE_PATH}/${BUILDID}/section-*.text
>
> # Download the original debuginfo/executable files.
> DEBUGFILE=`env LD_LIBRARY_PATH=$ldpath
> ${abs_top_builddir}/debuginfod/debuginfod-find debuginfo $RPM_BUILDID`
> @@ -110,11 +110,11 @@ testrun ${abs_top_builddir}/debuginfod/debuginfod-find
> -vvv executable $BUILDID
> if test "$(arch)" == "x86_64"; then
>tempfiles DEBUGFILE.debug_info
>objcopy $DEBUGFILE -O binary --only-section=.debug_info
> --set-section-flags .debug_info=alloc DEBUGFILE.debug_info
> - testrun diff -u DEBUGFILE.debug_info
> ${DEBUGINFOD_CACHE_PATH}/${RPM_BUILDID}/section-.debug_info
> + testrun diff -u DEBUGFILE.debug_info
> ${DEBUGINFOD_CACHE_PATH}/${RPM_BUILDID}/section-*.debug_info
>
>tempfiles EXECFILE.text
>objcopy $EXECFILE -O binary --only-section=.text EXECFILE.text
> - testrun diff -u EXECFILE.text
> ${DEBUGINFOD_CACHE_PATH}/${RPM_BUILDID}/section-.text
> + testrun diff -u EXECFILE.text
> ${DEBUGINFOD_CACHE_PATH}/${RPM_BUILDID}/section-*.text
> fi
>
> # Kill the server.
> @@ -123,10 +123,10 @@ wait $PID1
> PID1=0
>
> # Delete the section files from the cache.
> -rm -f ${DEBUGINFOD_CACHE_PATH}/${RPM_BUILDID}/section-.text
> -rm -f ${DEBUGINFOD_CACHE_PATH}/${RPM_BUILDID}/section-.debug_info
> -rm -f ${DEBUGINFOD_CACHE_PATH}/${BUILDID}/section-.text
> -rm -f ${DEBUGINFOD_CACHE_PATH}/${BUILDID}/section-.debug_info
> +rm -f ${DEBUGINFOD_CACHE_PATH}/${RPM_BUILDID}/section-*.text
> +rm -f ${DEBUGINFOD_CACHE_PATH}/${RPM_BUILDID}/section-*.debug_info
> +rm -f ${DEBUGINFOD_CACHE_PATH}/${BUILDID}/section-*.text
> +rm -f ${DEBUGINFOD_CACHE_PATH}/${BUILDID}/section-*.debug_info
>
> # Verify that the client can extract sections from the debuginfo or
> executable
> # if they're already in the cache.
Indeed, looks like too specific an expected cache name.
Cheers,
Mark
IT (0) themselves should call this explicitly.
exit_cleanup()
{
rm -f $remove_files; cd ..; rmdir $test_dir
}
trap exit_cleanup 0
On clean exit we remove all (tempfiles) and then try to remove the
$test_dir. Note that this fails (the test) if any non-tmp-file is left
behind.
I think this works, but the cd .. doesn't end up where it originally
did. Maybe save the orig_dir="${PWD}" before creating the test_dir and
cd $orig_dir here?
Cheers,
Mark
Hi,
On Thu, 2024-08-29 at 15:47 +0200, Mark Wielaard wrote:
> So we changed elf_memory so it pretends the in-memory Elf image is
> read with ELF_C_READ_MMAP. This helps when calling elf_memory on
> read-only memory which still wants to change some things about the Elf
> like unc
sparent change to clients, who are not to make any
> assumptions about cache file naming structure. However, one existing
> test did make such assumptions, so is fixed with some globness. A new
> test is also added, using a pre-baked tarball with a very long srcfile
> name.
Looks good with or without the suggested s/PATH_MAX/NAME_MAX/ tweak
suggested above.
Thanks,
Mark
Hi Aaron,
On Fri, Oct 18, 2024 at 12:08:11AM -0400, Aaron Merey wrote:
> LGTM. The new testcases pass on the trybots. And when I revert the
> elf_compress.c changes and run the new tests they fail as expected, so
> the elf_compress.c changes do in fact fix PR32102.
Thanks, pushed.
Cheers,
Mark
Hi Aaron,
I think if still possible you should include this in the release.
I made some comments below, but they can be addressed later. They are
just clarifications.
On Fri, 2024-10-18 at 00:02 -0400, Aaron Merey wrote:
> Signed-off-by: Aaron Merey
> ---
> config/elfutils.spec.in | 1 +
> do
s of these print functions.
>
> There is inconsistent use of fputc_unlocked() with putc_unlocked(),
> so consistently use the safer fputc() instead.
Looks good. Pushed.
Thanks,
Mark
print_frames): Take unwound_source field from struct frame and pass
> it on.
> (parse_opt): Add --cfi-type,-c option to set show_unwound_source.
> (main): Ditto.
> * tests/run-stack-i-test.sh: Add testcase for --cfi-type.
I like this version.
Thanks,
Mark
rather than whatever-build-filesystem
> speed. Individual test scripts are all unaffected. (One could
> consider /tmp instead, which is a RAM disk on modern systems, except
> that some of the elfutils tests produce GB-sized temporary files.
> That's probably too big for RAM.)
Looks good to me.
Thanks,
Mark
other pending patches like the just posted Openwrt
compat patches.
Cheers,
Mark
> [1] https://patchwork.sourceware.org/project/elfutils/list
>
ima verification check.
https://sourceware.org/bugzilla/show_bug.cgi?id=32294
Signed-off-by: Mark Wielaard
---
configure.ac | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/configure.ac b/configure.ac
index f191488f9402..0a62fece137e 100644
--- a/configure.ac
+++ b
, add libeu_objects.
https://sourceware.org/bugzilla/show_bug.cgi?id=32293
Signed-off-by: Mark Wielaard
---
libelf/Makefile.am | 3 +++
1 file changed, 3 insertions(+)
diff --git a/libelf/Makefile.am b/libelf/Makefile.am
index 3402863ef174..2d3dbdf22756 100644
--- a/libelf/Makefile.am
+++ b/l
The run-debuginfod-seekable.sh test would run (and fail) if lzma
support was enabled, but debuginfod wasn't build. Only run it when
both are available.
* tests/Makefile.am (TESTS): Add run-debuginfod-seekable.sh if
DEBUGINFOD and LZMA.
Signed-off-by: Mark Wielaard
---
Hi Aaron,
On Thu, 2024-10-24 at 10:23 -0400, Aaron Merey wrote:
> On Thu, Oct 24, 2024 at 5:34 AM Mark Wielaard wrote:
> >
> > For testing that the eu-stacktrace and debuginfod ima verification
> > code builds correctly explicitly add --enable-stacktrace and
> &
);
| ^
This code won't be reached because if find_procfile doesn't initialize
elf_fd, it will return an error. But help the compiler by initializing
elf_fd to -1.
* src/stacktrace.c (sysprof_init_dwfl): Init elf_fd to -1.
Signed-off-by: Mark Wielaard
---
src/stacktrace.c | 2
2401 - 2500 of 3442 matches
Mail list logo
