linux.git/tools/objtool/elf.c, branch v6.18.21

objtool: Fix weak symbol detection

2025-12-18T13:02:28+00:00

[ Upstream commit 72567c630d32bc31f671977f78228c80937ed80e ]

find_symbol_hole_containing() fails to find a symbol hole (aka stripped
weak symbol) if its section has no symbols before the hole.  This breaks
weak symbol detection if -ffunction-sections is enabled.

Fix that by allowing the interval tree to contain section symbols, which
are always at offset zero for a given section.

Fixes a bunch of (-ffunction-sections) warnings like:

  vmlinux.o: warning: objtool: .text.__x64_sys_io_setup+0x10: unreachable instruction

Fixes: 4adb23686795 ("objtool: Ignore extra-symbol code")
Acked-by: Petr Mladek 
Tested-by: Joe Lawrence 
Signed-off-by: Josh Poimboeuf 
Signed-off-by: Sasha Levin

objtool: Speed up SHT_GROUP reindexing

2025-05-14T11:09:02+00:00

After elf_update_group_sh_info() was introduced, a prototype version of
"objtool klp diff" went from taking ~1s to several minutes, due to
looping almost endlessly in elf_update_group_sh_info() while creating
thousands of local symbols in a file with thousands of sections.

Dramatically improve the performance by marking all symbols' correlated
SHT_GROUP sections while reading the object.  That way there's no need
to search for it every time a symbol gets reindexed.

Fixes: 2cb291596e2c ("objtool: Fix up st_info in COMDAT group section")
Signed-off-by: Josh Poimboeuf 
Signed-off-by: Peter Zijlstra (Intel) 
Tested-by: Rong Xu 
Link: https://lkml.kernel.org/r/2a33e583c87e3283706f346f9d59aac20653b7fd.1746662991.git.jpoimboe@kernel.org

objtool: Fix up st_info in COMDAT group section

2025-04-30T11:58:34+00:00

When __elf_create_symbol creates a local symbol, it relocates the first
global symbol upwards to make space. Subsequently, elf_update_symbol()
is called to refresh the symbol table section. However, this isn't
sufficient, as other sections might have the reference to the old
symbol index, for instance, the sh_info field of an SHT_GROUP section.

This patch updates the `sh_info` field when necessary. This field
serves as the key for the COMDAT group. An incorrect key would prevent
the linker's from deduplicating COMDAT symbols, leading to duplicate
definitions in the final link.

Signed-off-by: Rong Xu 
Signed-off-by: Peter Zijlstra (Intel) 
Link: https://lkml.kernel.org/r/20250425200541.113015-1-xur@google.com

objtool: Change "warning:" to "error: " for fatal errors

2025-04-01T07:07:13+00:00

This is similar to GCC's behavior and makes it more obvious why the
build failed.

Signed-off-by: Josh Poimboeuf 
Signed-off-by: Ingo Molnar 
Cc: Linus Torvalds 
Link: https://lore.kernel.org/r/0ea76f4b0e7a370711ed9f75fd0792bb5979c2bf.1743481539.git.jpoimboe@kernel.org

objtool: Improve error handling

2025-03-25T08:20:27+00:00

Fix some error handling issues, improve error messages, properly
distinguish betwee errors and warnings, and generally try to make all
the error handling more consistent.

Signed-off-by: Josh Poimboeuf 
Signed-off-by: Ingo Molnar 
Cc: Linus Torvalds 
Link: https://lore.kernel.org/r/3094bb4463dad29b6bd1bea03848d1571ace771c.1742852846.git.jpoimboe@kernel.org

objtool: Fix detection of consecutive jump tables on Clang 20

2025-03-25T08:20:25+00:00

The jump table detection code assumes jump tables are in the same order
as their corresponding indirect branches.  That's apparently not always
true with Clang 20.

Fix that by changing how multiple jump tables are detected.  In the
first detection pass, mark the beginning of each jump table so the
second pass can tell where one ends and the next one begins.

Fixes the following warnings:

  vmlinux.o: warning: objtool: SiS_GetCRT2Ptr+0x1ad: stack state mismatch: cfa1=4+8 cfa2=5+16
  sound/core/seq/snd-seq.o: warning: objtool: cc_ev_to_ump_midi2+0x589: return with modified stack frame

Fixes: be2f0b1e1264 ("objtool: Get rid of reloc->jump_table_start")
Reported-by: kernel test robot 
Signed-off-by: Josh Poimboeuf 
Signed-off-by: Ingo Molnar 
Cc: Nathan Chancellor 
Cc: Nick Desaulniers 
Cc: Linus Torvalds 
Link: https://lore.kernel.org/r/141752fff614eab962dba6bdfaa54aa67ff03bba.1742852846.git.jpoimboe@kernel.org
Closes: https://lore.kernel.org/oe-kbuild-all/202503171547.LlCTJLQL-lkp@intel.com/
Closes: https://lore.kernel.org/oe-kbuild-all/202503200535.J3hAvcjw-lkp@intel.com/

objtool: Add --output option

2025-03-17T10:36:01+00:00

Add option to allow writing the changed binary to a separate file rather
than changing it in place.

Libelf makes this suprisingly hard, so take the easy way out and just
copy the file before editing it.

Also steal the -o short option from --orc.  Nobody will notice ;-)

Signed-off-by: Josh Poimboeuf 
Signed-off-by: Peter Zijlstra (Intel) 
Link: https://lore.kernel.org/r/0da308d42d82b3bbed16a31a72d6bde52afcd6bd.1741975349.git.jpoimboe@kernel.org

objtool: Fix unreachable instruction warnings for weak functions

2024-11-06T13:41:10+00:00

In the presence of both weak and strong function definitions, the
linker drops the weak symbol in favor of a strong symbol, but
leaves the code in place. Code in ignore_unreachable_insn() has
some heuristics to suppress the warning, but it does not work when
-ffunction-sections is enabled.

Suppose function foo has both strong and weak definitions.
Case 1: The strong definition has an annotated section name,
like .init.text. Only the weak definition will be placed into
.text.foo. But since the section has no symbols, there will be no
"hole" in the section.

Case 2: Both sections are without an annotated section name.
Both will be placed into .text.foo section, but there will be only one
symbol (the strong one). If the weak code is before the strong code,
there is no "hole" as it fails to find the right-most symbol before
the offset.

The fix is to use the first node to compute the hole if hole.sym
is empty. If there is no symbol in the section, the first node
will be NULL, in which case, -1 is returned to skip the whole
section.

Co-developed-by: Han Shen 
Signed-off-by: Han Shen 
Signed-off-by: Rong Xu 
Suggested-by: Sriraman Tallam 
Suggested-by: Krzysztof Pszeniczny 
Tested-by: Yonghong Song 
Tested-by: Yabin Cui 
Tested-by: Nathan Chancellor 
Reviewed-by: Kees Cook 
Acked-by: Josh Poimboeuf 
Signed-off-by: Masahiro Yamada

objtool: Remove max symbol name length limitation

2023-10-06T00:01:28+00:00

If one of the symbols processed by read_symbols() happens to have a
.cold variant with a name longer than objtool's MAX_NAME_LEN limit, the
build fails.

Avoid this problem by just using strndup() to copy the parent function's
name, rather than strncpy()ing it onto the stack.

Signed-off-by: Aaron Plattner 
Link: https://lore.kernel.org/r/41e94cfea1d9131b758dd637fecdeacd459d4584.1696355111.git.aplattner@nvidia.com
Signed-off-by: Josh Poimboeuf

objtool: initialize all of struct elf

2023-07-10T07:52:28+00:00

Function elf_open_read() only zero initializes the initial part of
allocated struct elf; num_relocs member was recently added outside the
zeroed part so that it was left uninitialized, resulting in build failures
on some systems.

The partial initialization is a relic of times when struct elf had large
hash tables embedded. This is no longer the case so remove the trap and
initialize the whole structure instead.

Fixes: eb0481bbc4ce ("objtool: Fix reloc_hash size")
Signed-off-by: Michal Kubecek 
Signed-off-by: Peter Zijlstra (Intel) 
Acked-by: Josh Poimboeuf 
Link: https://lore.kernel.org/r/20230629102051.42E8360467@lion.mk-sys.cz