On Fri, Mar 29, 2024 at 12:43:08AM +0100, Mark Wielaard wrote:
> Some new tests and one configure file weren't in .gitignore. Also
> we made a copy of libelf.h in tests/ which should be an symlink.
>
> * config/.gitignore: Add profile.fish.
> * t
d by RASU JSC.
Sorry, but I am going to stop reviewing your patches till you have
responded to some of the previous reviews. All your suggested changes
are (subtly) wrong. In this case gelf_fsize cannot return zero for
example.
Please better explain your proposed changes and provide
testcases/examples of how the current code is wrong.
Thanks,
Mark
hrough dwarf_getsrcfiles, and the file names
>will be broken. */
>
> - if (__libdw_getsrclines (table->dbg, line_offset, table->comp_dir,
> -table->address_size, NULL, &table->files) < 0)
> + if (__libdw_getsrcfiles (table->dbg, line_offset, table->comp_dir,
> +table->address_size, &table->files) < 0)
> table->files = (void *) -1;
> }
OK.
> diff --git a/libdw/libdwP.h b/libdw/libdwP.h
> index c1c84ed3..e55ff50a 100644
> --- a/libdw/libdwP.h
> +++ b/libdw/libdwP.h
> @@ -1108,6 +1108,16 @@ int __libdw_getsrclines (Dwarf *dbg, Dwarf_Off
> debug_line_offset,
>internal_function
>__nonnull_attribute__ (1);
>
> +/* Load .debug_line unit at DEBUG_LINE_OFFSET. COMP_DIR is a value of
> + DW_AT_comp_dir or NULL if that attribute is not available. Caches
> + the loaded unit and set *FILESP with loaded information. Returns 0
> + for success or a negative value for failure. */
> +int __libdw_getsrcfiles (Dwarf *dbg, Dwarf_Off debug_line_offset,
> + const char *comp_dir, unsigned address_size,
> + Dwarf_Files **filesp)
> + internal_function
> + __nonnull_attribute__ (1);
> +
> /* Load and return value of DW_AT_comp_dir from CUDIE. */
> const char *__libdw_getcompdir (Dwarf_Die *cudie);
OK.
> diff --git a/tests/get-files.c b/tests/get-files.c
> index 04091733..fa65aa93 100644
> --- a/tests/get-files.c
> +++ b/tests/get-files.c
> @@ -24,6 +24,7 @@
> #include ELFUTILS_HEADER(dw)
> #include
> #include
> +#include "../libdw/libdwP.h"
>
>
> int
> @@ -76,6 +77,13 @@ main (int argc, char *argv[])
> break;
> }
>
> + if (die->cu->lines != NULL)
> + {
> + printf ("%s: dwarf_getsrcfiles should not get lines\n",
> argv[cnt]);
> + result = 1;
> + break;
> + }
> +
> const char *const *dirs;
> size_t ndirs;
> if (dwarf_getsrcdirs (files, &dirs, &ndirs) != 0)
> diff --git a/tests/get-lines.c b/tests/get-lines.c
> index 188d0162..77fb3c54 100644
> --- a/tests/get-lines.c
> +++ b/tests/get-lines.c
> @@ -26,6 +26,7 @@
> #include
> #include
> #include
> +#include "../libdw/libdwP.h"
>
>
> int
> @@ -69,6 +70,24 @@ main (int argc, char *argv[])
> }
> old_cuoff = cuoff;
>
> + Dwarf_Files *files;
> + size_t nfiles;
> +
> + /* Get files first to test that lines are read separately. */
> + if (dwarf_getsrcfiles (&die, &files, &nfiles) != 0)
> + {
> + printf ("%s: cannot get files\n", argv[cnt]);
> + result = 1;
> + break;
> + }
> +
> + if (die.cu->lines != NULL)
> + {
> + printf ("%s: dwarf_getsrcfiles should not get lines\n",
> argv[cnt]);
> + result = 1;
> + break;
> + }
> +
> Dwarf_Lines *lb;
> size_t nlb;
> if (dwarf_getsrclines (&die, &lb, &nlb) != 0)
> @@ -103,7 +122,6 @@ main (int argc, char *argv[])
>
> /* Getting the file path through the Dwarf_Files should
>result in the same path. */
> - Dwarf_Files *files;
> size_t idx;
> if (dwarf_line_file (l, &files, &idx) != 0)
> {
OK. Tests do use libdwP.h internals. Which isn't great. But OK in this
case because there is no other way to test this functionality.
tests/run-get-files.sh and tests/run-get-lines.sh also contain
testfiles using dwo and dwp files, which all still pass.
Thanks,
Mark
r `old`.
OK, but that doesn't really make sense. old isn't a dangling
pointer. It is a local pointer that is freed before returning from the
function. What do you try to accomplish by assigning it the value
NULL?
What real issue are you trying to fix?
Thanks,
Mark
> пт, 29 мар. 2024
On Tue, Apr 02, 2024 at 11:32:50PM +0300, Maks Mishin wrote:
> Dynamic memory, referenced by 'naddr', is allocated at segment.c:66
> by calling function 'realloc' and lost at segment.c:92.
It isn't lost, it is assigned to dwfl->lookup_addr at segment.c:77
>
> Found by RASU JSC.
>
> Signed-off-b
GElf_Word reserved_flags = ~(EF_HEXAGON_TINY | EF_HEXAGON_MACH);
>
> return (flags & reserved_flags) == 0;
> }
>
> ... implies a new EF_HEXAGON_TINY 0x8000 definition BTW.
You obviously know this architecture better than me. But is the TINY
flag (bit 15) appropriate for all machines? It looks like it can only
be set on V67 and V71. If so maybe just check those two explicitly?
Cheers,
Mark
agon.ko.bz2.
* tests/hello_hexagon.ko.bz2: New test file.
* tests/run-strip-reloc-ko.sh: Add hello_hexagon.ko.
This is mainly for my own review, so I know all changes were actually
intended.
Pushed,
Mark
be in those cases?
OK, if there could be other TINY variants then either the check
suggested by Matheus or you would be more correct. I do like the idea
to have a self documenting EF_HEXAGON_TINY constant for this.
Thanks,
Mark
ded ChangeLog to commit message.
> - Implemented better machine_flag_check operation, as suggested by
> bcain.
> - Extracted only patch 2/2 as 1/2 was already merged.
This looks good.
Pushed,
Mark
-core-.sh with
tests/tests/backtrace..{exec,core}.bz2 so it can be tested from
a non-MIPS setup (as opposed to the tests/run-backtrace-native.sh and
tests/run-backtrace-native-core.sh tests).
Thanks,
Mark
tly
this backend only handles the 64 bit little endian one?
Again it would be nice if you could come up with non-native variants
of the tests so the implementation can also be tested on non-MIPS
arches. Specifically for tests/run-allregs.sh, tests/run-funcretval.sh
and/or tests/run-readelf-mixed-corenote.sh.
Thanks,
Mark
ractive, but it would be a lot of work recreating
the existing binary test files. And nobody has tried yet, so we don't
know if this (or some other tool) is expressive enough.
Cheers,
Mark
y understand the
code.
In this case if debugelf == NULL then gelf_newehdr will also return
NULL to indicate the earlier error.
This is a common pattern in the libelf code.
Thanks,
Mark
> Found by RASU JSC.
>
> Signed-off-by: Maks Mishin
> ---
> src/strip.c | 4 +++-
>
Hi,
On Thu, Mar 28, 2024 at 12:25:32PM -0400, Aaron Merey wrote:
> This code of conduct includes a committee to handle any complaints or
> concerns that fall under the code. Mark and I have offered to be on this
> committee. If anyone else would like to be a member of the elfutils
S) $(sqlite3_LIBS)
> $(libarchive_LIBS) $(rpm_LIBS) -lpthread -ldl
>
> debuginfod_find_SOURCES = debuginfod-find.c
> debuginfod_find_LDADD = $(libdw) $(libelf) $(libeu) $(libdebuginfod)
> $(argp_LDADD) $(fts_LIBS)
> @@ -97,7 +97,7 @@ libdebuginfod_so_LIBS = libdebuginfod_pic.a
t;
> + file[7] = "include/linux/sched.h"
> + file[8] = "include/asm/thread_info.h"
> + file[9] = "include/linux/thread_info.h"
> + file[10] = "include/asm/atomic.h"
> + file[11] = "include/linux/list.h"
> + file[12] = "include/linux/cpumask.h"
> + file[13] = "include/linux/rbtree.h"
> + file[14] = "include/asm/page.h"
> + file[15] = "include/linux/rwsem.h"
> + file[16] = "include/asm/rwsem.h"
> + file[17] = "include/asm/spinlock.h"
> + file[18] = "include/linux/completion.h"
> + file[19] = "include/linux/wait.h"
> + file[20] = "include/linux/aio.h"
> + file[21] = "include/linux/workqueue.h"
> + file[22] = "include/linux/timer.h"
> + file[23] = "include/linux/types.h"
> + file[24] = "include/asm/posix_types.h"
> + file[25] = "include/linux/pid.h"
> + file[26] = "include/linux/time.h"
> + file[27] = "include/linux/capability.h"
> + file[28] = "include/linux/signal.h"
> + file[29] = "include/linux/resource.h"
> + file[30] = "include/linux/sem.h"
> + file[31] = "include/asm/fpu.h"
> + file[32] = "include/linux/fs_struct.h"
> + file[33] = "include/asm/signal.h"
> + file[34] = "include/asm/siginfo.h"
> + file[35] = "include/asm-generic/siginfo.h"
> + file[36] = "include/asm/nodedata.h"
> + file[37] = "include/linux/mmzone.h"
> + file[38] = "include/linux/jiffies.h"
> + file[39] = "include/asm/io.h"
> + file[40] = "include/asm/machvec.h"
> + file[41] = "include/asm/smp.h"
> + file[42] = "include/asm/numa.h"
> + file[43] = "include/linux/slab.h"
> +EOF
> +
> +# Files should be printed 3 times, followed by the files from
> DW_LNE_define_file
> +cat files > define-files.out
> +cat files >> define-files.out
> +cat files >> define-files.out
> +echo ' file[44] = "include/asm/abc.c"' >> define-files.out
> +echo ' file[45] = "include/linux/01.c"' >> define-files.out
> +
> +cat define-files.out | testrun_compare ${abs_builddir}/get-files-define-file
> testfile-define-file
> +
> exit 0
Nice.
So testfile-define-file is actually testfile36.debug but with a new
line program? How did you edit/insert that one?
Cheers,
Mark
am
> that was easy to replace with two DW_LNE_define_file opcodes without
> corrupting things.
Fun. Could you add a small description to tests/run-get-files.sh where
testfile-define-file is use so future hackers know how the file was
created?
All looks good BTW. Please do push (if possible with the above change).
Thanks,
Mark
ial replacement of files). Relying on https and CRC checking
seems simpler to understand for that. But you are right that there
isn't really anyhing for the main executables. For the source files
you could check the lenght and time stamps (or the MD5 sums if added
by the DWARF producer).
&g
.
Signed-off-by: Mark Wielaard
---
src/ar.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/src/ar.c b/src/ar.c
index e6d6d58f2b3b..fcb8bfb90a9f 100644
--- a/src/ar.c
+++ b/src/ar.c
@@ -41,6 +41,7 @@
#include
#include
+#include "libeu.h"
#include "arlib.h&
efore calling
print_form_data.
* tests/testfileclangmacro.bz2: New testfile.
* tests/Makefile.am (EXTRA_DIST): Add testfileclangmacro.bz2.
* tests/run-readelf-macro.sh: Add testfileclangmacro output.
Signed-off-by: Mark Wielaard
---
src/readelf.c| 13 +-
Hi Aaron,
On Fri, May 10, 2024 at 05:16:06PM -0400, Aaron Merey wrote:
> On Tue, Apr 30, 2024 at 10:39 AM Mark Wielaard wrote:
> >
> > This alloca use is inside a lexical block and is used to replace one
> > element of argv. Use a function local variable, xmalloc and fr
Hi,
On Sun, 2024-05-05 at 00:15 +0200, Mark Wielaard wrote:
> print_form_data didn't take the offset_len (4 or 8 bytes) into account
> causing the wrong entry to be read from .debug_str_offsets.
> print_debug_macro_section did sanity checking before calling
> print_form_data,
and formulate the security attack
we are trying to defend against with ima verification first.
Cheers,
Mark
ed as "not collegial and
exaggerated", then maybe next time also provide an example of how to
formulate the request that sounds better. Tone is a difficult thing,
especially on the mailinglist.
Cheers,
Mark
when the elf.h change is accepted in
glibc, then we'll sync and integrate this.
Thanks,
Mark
>
> Signed-off-by: Luca Boccassi
> ---
> libebl/eblobjnote.c | 9 +++--
> libebl/eblobjnotetypename.c | 3 +++
> libelf/elf.h| 4
> 3 files c
Hi Luca,
On Thu, 2024-05-30 at 11:45 +0100, Luca Boccassi wrote:
> On Tue, 14 May 2024 at 22:18, Mark Wielaard wrote:
> > Not a fan of json, feels very un-ELF. But it is what it is. The patch
> > looks OK. Could you let us know when the elf.h change is accepted in
> > glibc
* libelf/elf.h: Adds AT_HWCAP{3,4}, NT_FDO_DLOPEN_METADATA
and R_LARCH_TLS_DESC{32,64}.
Signed-off-by: Mark Wielaard
---
libelf/elf.h | 12 +++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git a/libelf/elf.h b/libelf/elf.h
index f2206e5c065d..081742a9c38c
From: Luca Boccassi
The node ID and the string format are now fixed, even if the content
of the string might change, it will still be a string.
* libebl/eblobjnote.c (ebl_object_note): Handle both type
being NT_FDO_PACKAGING_METADATA or NT_FDO_DLOPEN_METADATA when
name is "FDO".
Hi,
Adding elfutils-devel to CC to keep everyone up to date on the state of
the patches.
On Mon, 2024-06-10 at 23:36 -0700, Tony Ambardar wrote:
> On Mon, Jun 03, 2024 at 08:47:24PM -0700, Tony Ambardar wrote:
> > On Mon, Jun 03, 2024 at 09:18:33PM +0200, Mark Wielaard wrote:
> >
at the rest of the tree
> behaves).
Yes, I'll sent a patch.
> But I think the root cause is more the continued reliance
> on "make dist", made necessary by the exclusion of generated autoconf*
> materiel in the source tree.
I really don't think putting generated file into git is a good idea.
You get all those issues that gcc, binutils and gdb have where you need
all this extra CI just to make sure people use the "correct" tools to
regenerate the files.
Cheers,
Mark
dist targets shouldn't depend on configure conditionals.
* doc/Makefile.am: Remove DEBUGINFOD and LIBDEBUGINFOD from
dist targets. Add comments about which man pages go into which
section.
Signed-off-by: Mark Wielaard
---
doc/Makefile.am | 33 +++
> unsatisfactory version, any other developer (or a bot) can regen and
> fix it. It seems to me a non-problem.
Having each committer push generated files with their own variant of
the tools causes generates updates that are hard to sanity check by
others.
Cheers,
Mark
Hi,
On Tue, Jun 11, 2024 at 06:48:49PM +0200, Mark Wielaard wrote:
> dist targets shouldn't depend on configure conditionals.
>
> * doc/Makefile.am: Remove DEBUGINFOD and LIBDEBUGINFOD from
> dist targets. Add comments about which man pages go into which
> sect
/281
>
> Build state: worker cannot substantiate
> Revision: (unknown)
> Worker: bb1-1
> Build Reason: (unknown)
> Blamelist: Mark Wielaard
Turns out current fedora rawhide is uninstallable because of a partial
python transition. I have disabled the rawhide container builders for
now.
Cheers,
Mark
RIPTION
> +The \fBelf_version\fP function sets the library's ELF version to the
> specified value. This function must be called before any other libelf
> functions are used.
> +
> +.SH PARAMETERS
> +.TP
> +.I version
> +An \fIunsigned int\fP value specifying the desired ELF library version. This
> is should be set to \fBEV_CURRENT\fP to indicate the current version of the
> library. At this time the only supported \fIversion\fP values are
> \fBEV_CURRENT\fP and \fBEV_NONE\fP.
> +
> +.SH RETURN VALUE
> +The \fBelf_version\fP function returns \fBEV_CURRENT\fP when \fIversion\fP
> is supported. If the specified version is not supported, it returns
> \fBEV_NONE\fP.
> +
> +.SH REPORTING BUGS
> +Report bugs to or
> https://sourceware.org/bugzilla/.
>
elf_version is kind of funny. EV_CURRENT is 1 and I don't believe we
will ever see another ELF version.
Cheers,
Mark
always called with *_whole being NULL.
* libdwfl/gzip.c (unzip): Set mapped = state.input_buffer
when *whole is not NULL.
Signed-off-by: Mark Wielaard
---
libdwfl/gzip.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/libdwfl/gzip.c b/libdwfl/gzip.c
index 002afc4e916b
Found by GCC -fanalyzer.
When allocating the notcvt buffer fails we leak the shdr. goto
free_and_out on malloc failure.
* libelf/elf32_getshdr.c (load_shdr_wrlock): goto
free_and_out on second malloc failure.
Signed-off-by: Mark Wielaard
---
libelf/elf32_getshdr.c
): Check
whether id or version is already set before resetting them.
Signed-off-by: Mark Wielaard
---
debuginfod/debuginfod-client.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/debuginfod/debuginfod-client.c b/debuginfod/debuginfod-client.c
index 95f2a92b701c
to -1.
(do_oper_delete): Likewise.
(do_oper_insert): Likewise.
* src/ranlib.c (handle_file): Likewise.
Signed-off-by: Mark Wielaard
---
src/ar.c | 12 ++--
src/ranlib.c | 4 ++--
2 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/src/ar.c b/src/ar.c
Hi,
On Sun, Jun 23, 2024 at 01:50:10AM +0200, Mark Wielaard wrote:
> Found by GCC14 -Wanalyzer-null-argument.
Aaron took a quick look on irc and said all 4 patches looked
reasonable. So I pushed them all to main now.
Cheers,
Mark
NT is the only supported version.
When given EV_NONE it won't initialize the library, but return the
highest supported version (which is EV_CURRENT obviously).
> +.SH RETURN VALUE
> +The \fBelf_version\fP function returns \fBEV_CURRENT\fP when \fIversion\fP
> is supported. If the specified version is not supported, it returns
> \fBEV_NONE\fP.
^ If the version given is not EV_NONE.
If the version given is EV_NONE it will return EV_CURRENT.
Cheers,
Mark
rom zero to
lastix retval is set when mix == firstix. So when firstix < mix retval
is definitely set.
Cheers,
Mark
plain
why you believe the tool reports are correct.
https://inbox.sourceware.org/elfutils-devel/20240405190452.gh1...@gnu.wildebeest.org/
Cheers,
Mark
On Mon, Jul 01, 2024 at 11:53:13PM +0300, Maks Mishin wrote:
> Pointer `symstrdata` is dereferenced at unstrip.c:1977 without checking for
> NULL.
If there has to be a NULL check then it needs to be when symstrdata is
assigned a value from elf_getdata.
Cheers,
Mark
.c
> +++ b/src/readelf.c
> @@ -3597,6 +3597,7 @@ print_hash_info (Ebl *ebl, Elf_Scn *scn, GElf_Shdr
> *shdr, size_t shstrndx,
> {
>error (0, 0, _("invalid sh_link value in section %zu"),
>elf_ndxscn (scn));
> + free (counts);
> return;
> }
That looks correct. Pushed.
Thanks,
Mark
be big enough as the comment says:
/* NAME should be big enough to hold any possible comma-separated
list (no repetitions allowed) of attribute names from one of the
arrays above. */
static char name[32*17+32+1];
Maybe there is a way to add a static assert for this?
Cheers,
Mark
views/questions.
https://patchwork.sourceware.org/project/elfutils/list/?series=15310&state=*
Thanks,
Mark
Hi Aleksei,
On Thu, 2024-07-11 at 20:35 +, Aleksei Vetrov wrote:
> elf_memory open mode recently changed from ELF_C_READ to
> ELF_C_READ_MMAP.
That was this commit:
commit cc44ac6740797a23cd0af0cb22bd828d569224b8
Author: Mark Wielaard
Date: Thu Feb 1 14:56:18 2024 +0100
libelf:
{'b', 'p'},
+ {'s', 'i'},
+ {'d', 'i'},
+ {'i', 'p'},
};
case 4:
?
Since the use of this array is basically just:
name[0] = 'r';
name[1] = baseregs[regno][0];
name[2] = baseregs[regno][1];
I think I prefer the individual bytes init way. It makes more clear
what we really use these arrays for. imho.
Thanks,
Mark
to.
This looks good. I have pushed it to main.
Will this warning be enabled by -Wall or -Wextra (both of which are
enabled by default for elfutils builds)? Or would be need to enable it
explicitly?
Thanks,
Mark
On Wed, 2024-07-17 at 18:34 -0400, Aaron Merey wrote:
> From: Heather McIntyre
>
> Apply locking during __libelf_readall.
>
> Signed-off-by: Heather S. McIntyre
> Signed-off-by: Aaron Merey
> Signed-off-by: Mark Wielaard
>
> ---
> v2 changes:
>
>
Hi,
On Wed, 2024-07-17 at 18:34 -0400, Aaron Merey wrote:
> From: Heather McIntyre
>
> * libelf/elf_cntl.c (elf_cntl): Move rwlock_wrlock, rwlock_unlock,
> inside case switch statements.
>
> Signed-off-by: Heather S. McIntyre
> Signed-off-by: Aaron Merey
Hi,
On Wed, Jul 17, 2024 at 06:34:03PM -0400, Aaron Merey wrote:
> From: Heather McIntyre
>
> * libdw/dwarf_getalt.c (dwarf_getalt): Add locking.
>
> Signed-off-by: Heather S. McIntyre
> Signed-off-by: Aaron Merey
> Signed-off-by: Mark Wielaard
>
> v2 c
Hi,
On Wed, Jul 17, 2024 at 06:34:04PM -0400, Aaron Merey wrote:
> From: Heather McIntyre
>
> Signed-off-by: Heather S. McIntyre
> Signed-off-by: Aaron Merey
> Signed-off-by: Mark Wielaard
>
> ---
>
> v2 changes:
> This replaces patch "libdw: Add l
Hi,
On Wed, Jul 17, 2024 at 06:34:05PM -0400, Aaron Merey wrote:
> From: Heather McIntyre
>
> * (__libdw_find_split_unit): Add lock for cu->split.
>
> Signed-off-by: Heather S. McIntyre
> Signed-off-by: Aaron Merey
> Signed-off-by: Mark Wielaard
>
>
tyre
> Signed-off-by: Aaron Merey
> Signed-off-by: Mark Wielaard
>
> ---
> v2 changes:
> Use per-Dwarf_CU lock instead of a global lock.
Are you sure this description and the ChangeLog entry are correct?
This patch doesn't contain a change to use eu_tfind (there is already
one
Move the rwlock_* macros out of eu-config.h and into a new header file
> locks.h. This was done so that the rwlock_* macros can be included
> in libdwP.h without having to also include the rest of eu-config.h.
>
> Signed-off-by: Heather S. McIntyre
> Signed-off-by: Aaron Merey
Hi Aaron,
Some small comments below.
On Thu, Jul 18, 2024 at 10:03:02PM -0400, Aaron Merey wrote:
> Add man pages for elf32_offscn.3, elf64_offscn.3, elf_getscn.3 and
> elf_ndxscn.3.
>
> Signed-off-by: Aaron Merey
> ---
>
> Reposting this patch with the new man pages added to doc/Makefile.am.
uld be served by the current debuginfod protocol?
Also no. Not in memory. There are the dwp files though.
Cheers,
Mark
Hi Aaron,
On Fri, 2024-08-02 at 19:38 -0400, Aaron Merey wrote:
> From: Heather McIntyre
>
> Apply locking during __libelf_readall.
>
> Signed-off-by: Heather S. McIntyre
> Signed-off-by: Aaron Merey
> Signed-off-by: Mark Wielaard
>
> ---
> v3 changes:
>
&
McIntyre
> Signed-off-by: Aaron Merey
> Signed-off-by: Mark Wielaard
>
> ---
> v3 changes:
> Remove unnecessary early return that would require locking to
> check condition.
The code looks good and clean now.
Thanks,
Mark
* elf_getdata_rawchunck.c: Ditto.
> * libelfP.h (struct Elf): Replace void * rawchuncks memeber with
> search_tree rawchunk_tree.
Nice Changelog. (note the memeber typo)
> Signed-off-by: Heather S. McIntyre
> Signed-off-by: Aaron Merey
> Signed-off-by: Mark Wielaard
dwarf_getalt
for the dwarf_lock, dwarf_formref_die and __libdw_dieabbrev for
abbrev_lock).
It might also be a good idea to explicitly document which structure
fields are "protected" by the separate locks to make reasoning about
them a little easier.
Cheers,
Mark
On Fri, 2024-08-02 at 19:38 -0400, Aaron Merey wrote:
> From: Heather McIntyre
>
> * libdw/dwarf_getalt.c (dwarf_getalt): Add locking.
> * libdw/dwarf_setalt.c (dwarf_setalt): Ditto.
>
> Signed-off-by: Heather S. McIntyre
> Signed-off-by: Aaron Merey
> Sign
efined.
> (eu_search_die_LDFLAGS): Likewise.
> (eu_search_lines_LDFLAGS): Likewise.
> (eu_search_macros_LDFLAGS): Likewise.
>
> Signed-off-by: Heather S. McIntyre
> Signed-off-by: Aaron Merey
> Signed-off-by: Mark Wielaard
>
> ---
> No changes betwee
cIntyre
> Signed-off-by: Aaron Merey
> Signed-off-by: Mark Wielaard
>
> ---
> v3 changes:
> Move Dwarf split_lock code from patch 3/9 v2 to this patch.
> Assign return value before releasing lock.
Looks good this way.
Thanks,
Mark
> libdw/dwarf_end.c
Hi,
On Fri, Aug 02, 2024 at 07:38:07PM -0400, Aaron Merey wrote:
> From: Heather McIntyre
>
> * libdw/libdw_findcu.c (__libdw_findcu): Add locking.
>
> Signed-off-by: Heather S. McIntyre
> Signed-off-by: Aaron Merey
> Signed-off-by: Mark Wielaard
>
&
igned-off-by: Heather S. McIntyre
> Signed-off-by: Aaron Merey
> Signed-off-by: Mark Wielaard
>
> v3 changes:
> Move dwarf_setalt.c changes to patch 5/9 v3.
> Set __libdw_dieabbrev result before releasing lock.
> ---
> libdw/dwarf_formref_die
with the
glibc elf/elf.h. Since they recently added the same (and more)
NT_ARM_* constants I'll sync that file seperately before incorporating
your patch.
Did you test the output on a core file that has these new regsets and
core notes with eu-readelf to see if it looks correct?
Thanks,
Mark
* libelf/elf.h: Adds NT_ARM_{SSVE,ZA,ZT,FPRM}
Signed-off-by: Mark Wielaard
---
libelf/elf.h | 4
1 file changed, 4 insertions(+)
diff --git a/libelf/elf.h b/libelf/elf.h
index 081742a9c38c..33aea7f743b8 100644
--- a/libelf/elf.h
+++ b/libelf/elf.h
@@ -831,6 +831,10 @@ typedef
agging-extension.html
Implementation looks correct.
Pushed,
Mark
>
> Signed-off-by: Kuan-Ying Lee
> ---
> backends/aarch64_corenote.c | 11 ++-
> libebl/eblcorenotetypename.c | 1 +
> 2 files changed, 11 insertions(+), 1 deletion(-)
>
> diff --git a/backends
cannot find the
definition. This assumes it is just one 8 byte word. Is that correct?
Can it be more than one?
Thanks,
Mark
> Signed-off-by: Kuan-Ying Lee
> ---
> backends/aarch64_corenote.c | 24 +++-
> libebl/eblcorenotetypename.c | 2 ++
> 2 files changed
Hi,
On Wed, Aug 14, 2024 at 04:51:20PM +0800, Kuan-Ying Lee wrote:
> Recognize names of some new core note types in ebl_core_note_type_name.
Thanks. Pushed the eblcorenotetypename.c part.
I pushed a glibc elf.h sync separately (commit 0abdd2e79).
Cheers,
Mark
14085134.109500-3-kuan-ying@canonical.com/
Luckily the patches agree on the definition of the the pac_items
(modulo the name data_mask/insn_mask vs pauth_dmask/pauth_cmask).
This patch doesn't introduce a regset for ARM_PAC_ENABLED_KEYS that
the other one does. Is this not necessary?
Tha
ial_registers_tid);
>HOOK (eh, unwind);
That should still be 97. RA_SIGN_STATE is regno 34, which was reserved
before, followed by 29 Resered regs (35-43).
Cheers,
Mark
result)))
> + goto out;
> + fs->regs[DW_AARCH64_RA_SIGN_STATE].value ^= 0x1;
> }
> else
> {
Funny we already had some code here to handle
DW_CFA_AARCH64_negate_ra_state (and how unfortunate it overlaps with
DW_CFA_GNU_window_save, but there is already an em check).
Also looks good.
Thanks,
Mark
der). Since this is really just a private detail for CFI
could we define it in cfi.h (which is a private elfutils header)?
Cheers,
Mark
gly. But given we can hide most of it in
architecture specific/private code better than special case public
architecture functions.
Cheers,
Mark
When doing a --enable-gcov build we link all binaries static.
libdebuginfod.so now depends on crypto an jsonc. So also add those
when linking against libdebuginfod.a
debuginfod/Makefile.am (libdebuginfod): Add $(crypto_LIBS)
$(jsonc_LIBS) when BUILD_STATIC.
Signed-off-by: Mark
libeu is a static library with internal helper functions normally
included in all shared libraries. But when linking static (with
--enable-gcov) we need to explicitly link it into the test binaries.
* tests/Makefile.am (libelf): Add $(libeu) when BUILD_STATIC.
Signed-off-by: Mark Wielaard
Hi,
On Wed, Aug 21, 2024 at 03:32:59PM +0200, Mark Wielaard wrote:
> When doing a --enable-gcov build we link all binaries static.
> libdebuginfod.so now depends on crypto an jsonc. So also add those
> when linking against libdebuginfod.a
>
> debuginfod/Makefile.am (libd
Hi,
On Wed, Aug 21, 2024 at 03:52:20PM +0200, Mark Wielaard wrote:
> libeu is a static library with internal helper functions normally
> included in all shared libraries. But when linking static (with
> --enable-gcov) we need to explicitly link it into the test binaries.
>
>
BUTES
> +For an explanation of the terms used in this section, see
> +.BR attributes (7).
> +.TS
> +allbox;
> +lbx lb lb
> +l l l.
> +InterfaceAttribute Value
> +T{
> +.na
> +.nh
> +.BR elf32_checksum ()
> +T} Thread safety MT-Safe
> +.TE
> +
> +.SH REPORTING BUGS
> +Report bugs to or
> https://sourceware.org/bugzilla/.
Cheers,
Mark
e
> +.BR attributes (7).
> +.TS
> +allbox;
> +lbx lb lb
> +l l l.
> +InterfaceAttribute Value
> +T{
> +.na
> +.nh
> +.BR elf32_fsize ()
> +T} Thread safety MT-Safe
> +.TE
> +
> +.SH REPORTING BUGS
> +Report bugs to or
> https://sourceware.org/bugzilla/.
Cheers,
Mark
hdr
> +first appeared in elfutils 0.165. This elfutils libelf function may not be
> found in other libelf implementations.
Yes, but it has been standardized:
http://www.linker-aliens.org/blogs/ali/entry/elf_section_compression/
Cheers,
Mark
gt; +.BR libelf (3),
> +.BR elf (5)
Add elf_getphdrnum and elf32_newphdr.
> +.SH ATTRIBUTES
> +For an explanation of the terms used in this section, see
> +.BR attributes (7).
> +.TS
> +allbox;
> +lbx lb lb
> +l l l.
> +InterfaceAttribute Value
> +T{
> +.na
> +.nh
> +.BR elf32_getphdr ()
> +T} Thread safety MT-Safe
> +.TE
> +
> +.SH REPORTING BUGS
> +Report bugs to or
> https://sourceware.org/bugzilla/.
Cheers,
Mark
.BR elf (5)
Add elf_flagshdr.
> +.SH ATTRIBUTES
> +For an explanation of the terms used in this section, see
> +.BR attributes (7).
> +.TS
> +allbox;
> +lbx lb lb
> +l l l.
> +InterfaceAttribute Value
> +T{
> +.na
> +.nh
> +.BR elf32_getshdr ()
> +T} Thread safety MT-Safe
> +.TE
> +
> +.SH REPORTING BUGS
> +Report bugs to or
> https://sourceware.org/bugzilla/.
Cheers,
Mark
IBUTES
> +For an explanation of the terms used in this section, see
> +.BR attributes (7).
> +.TS
> +allbox;
> +lbx lb lb
> +l l l.
> +InterfaceAttribute Value
> +T{
> +.na
> +.nh
> +.BR elf32_newehdr ()
> +T} Thread safety MT-Safe
> +.TE
> +
> +.SH REPORTING BUGS
> +Report bugs to or
> https://sourceware.org/bugzilla/.
Cheers,
Mark
+.BR elf_errno (3),
> +.BR elf64_newphdr (3),
> +.BR libelf (3),
> +.BR elf (5)
> +
> +.SH ATTRIBUTES
> +For an explanation of the terms used in this section, see
> +.BR attributes (7).
> +.TS
> +allbox;
> +lbx lb lb
> +l l l.
> +InterfaceAttribute Value
> +T{
> +.na
> +.nh
> +.BR elf32_newphdr ()
> +T} Thread safety MT-Safe
> +.TE
> +
> +.SH REPORTING BUGS
> +Report bugs to or
> https://sourceware.org/bugzilla/.
Cheers,
Mark
f the terms used in this section, see
> +.BR attributes (7).
> +.TS
> +allbox;
> +lbx lb lb
> +l l l.
> +InterfaceAttribute Value
> +T{
> +.na
> +.nh
> +.BR elf32_xlatetof ()
> +T} Thread safety MT-Safe
> +.TE
> +
> +.SH REPORTING BUGS
> +Report bugs to or
> https://sourceware.org/bugzilla/.
Cheers,
Mark
number of records in case of ELF_T_NHDR[8].
Signed-off-by: Mark Wielaard
---
libelf/elf32_xlatetof.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/libelf/elf32_xlatetof.c b/libelf/elf32_xlatetof.c
index ab857409329a..ab3e609ee5d2 100644
--- a/libelf/elf32_xlatetof.c
++
Check that src and dest aren't NULL and that src->d_type < ELF_T_NUM.
* elf32_xlatetof.c (elfw2(LIBELFBITS, xlatetof)): Check src, dest
and src->d_type are valid.
* elf32_xlatetom.c (elfw2(LIBELFBITS, xlatetom)): Likewise.
Signed-off-by: Mark Wielaar
Hi,
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 uncompress some sections (which changes the section header).
With ELF_C_READ_MMAP lib
about is the usage of
struct user_pac_mask in aarch64_initreg.c. It might not be defined on
older aarch64 systems. Hopefully the buildbot (or some aarch64 user)
will catch that if it really is an issue.
Thanks,
Mark
> Cheers,
> --
> Steve
>
> German Gomez (3):
> aarch64: Crea
Hi Kuan-Ying,
On Mon, 2024-08-19 at 17:43 +0800, Kuan-Ying Lee wrote:
> On Sat, Aug 17, 2024 at 01:36:33AM +0200, Mark Wielaard wrote:
> > On Wed, Aug 14, 2024 at 04:51:19PM +0800, Kuan-Ying Lee wrote:
> > > Add the NT_ARM_PAC_MASK and NT_ARM_PAC_ENABLED_KEYS for aarch64.
&
table
offsets that were the same as the address offset.
Signed-off-by: Mark Wielaard
---
src/readelf.c | 2 +-
tests/run-readelf-gdb_index.sh | 8
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/src/readelf.c b/src/readelf.c
index 48035264be13
end.
Signed-off-by: Mark Wielaard
---
src/readelf.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/readelf.c b/src/readelf.c
index 8d96443daa83..21cacd7b6c06 100644
--- a/src/readelf.c
+++ b/src/readelf.c
@@ -11679,7 +11679,7 @@ print_gdb_index_section (Dwfl_Module *dwflmod
ENABLE_LIBDEBUGINFOD
after no_backup check.
* tests/Makefile.am (TESTS): Only add run-srcfiles-self.sh
if DEBUGINFOD and !DUMMY_LIBDEBUGINFOD.
* tests/run-srcfiles.self.sh: Use local DEBUGINFOD_CACHE_PATH.
https://sourceware.org/bugzilla/show_bug.cgi?id=32155
Signed-off-by: Mark Wielaard
ight, they are not fully compatible ways to detect zstd
library support. I wonder if it is reasonable now to just check for
some 1.4.0 zstd symbol to get both libdw decompression and libelf
compression support at the same time. Then you only need the eu_ZIPLIB
macro and can forget about the pkgconfig version check.
Cheers,
Mark
Hi,
On Sun, 2024-09-08 at 00:00 +0200, Mark Wielaard wrote:
> The .gdb_index const table can be empty, if no symbol table entry
> refers to a name or CU. Which technically means the symbol table is
> empty or has only zero members.
>
> * src/readelf.c (print_gdb_index_
501 - 600 of 3433 matches
Mail list logo
