kernel/linux.git/include/linux/module.h, branch v6.19.11

module: add helper function for reading module_buildid()

2026-02-26T23:01:07+00:00

[ Upstream commit acfdbb4ab2910ff6f03becb569c23ac7b2223913 ] Add a helper function for reading the optional "build_id" member of struct module. It is going to be used also in ftrace_mod_address_lookup(). Use "#ifdef" instead of "#if IS_ENABLED()" to match the declaration of the optional field in struct module. Link: https://lkml.kernel.org/r/20251128135920.217303-4-pmladek@suse.com Signed-off-by: Petr Mladek Reviewed-by: Daniel Gomez Reviewed-by: Petr Pavlu Cc: Aaron Tomlin Cc: Alexei Starovoitov Cc: Daniel Borkman Cc: John Fastabend Cc: Kees Cook Cc: Luis Chamberalin Cc: Marc Rutland Cc: "Masami Hiramatsu (Google)" Cc: Sami Tolvanen Cc: Steven Rostedt (Google) Signed-off-by: Andrew Morton Stable-dep-of: e8a1e7eaa19d ("kallsyms/ftrace: set module buildid in ftrace_mod_address_lookup()") Signed-off-by: Sasha Levin

module: Fix device table module aliases

2025-10-22T13:21:55+00:00

Commit 6717e8f91db7 ("kbuild: Remove 'kmod_' prefix from __KBUILD_MODNAME") inadvertently broke module alias generation for modules which rely on MODULE_DEVICE_TABLE(). It removed the "kmod_" prefix from __KBUILD_MODNAME, which caused MODULE_DEVICE_TABLE() to generate a symbol name which no longer matched the format expected by handle_moddevtable() in scripts/mod/file2alias.c. As a result, modpost failed to find the device tables, leading to missing module aliases. Fix this by explicitly adding the "kmod_" string within the MODULE_DEVICE_TABLE() macro itself, restoring the symbol name to the format expected by file2alias.c. Fixes: 6717e8f91db7 ("kbuild: Remove 'kmod_' prefix from __KBUILD_MODNAME") Reported-by: Alexander Stein Reported-by: Marek Szyprowski Reported-by: Mark Brown Reported-by: Cosmin Tanislav Signed-off-by: Josh Poimboeuf Signed-off-by: Peter Zijlstra (Intel) Tested-by: Cosmin Tanislav Tested-by: Marek Szyprowski Tested-by: Mark Brown Tested-by: Alexander Stein Tested-by: Chen-Yu Tsai Tested-by: Anders Roxell Link: https://patch.msgid.link/e52ee3edf32874da645a9e037a7d77c69893a22a.1760982784.git.jpoimboe@kernel.org

modpost: Create modalias for builtin modules

2025-09-24T16:10:45+00:00

For some modules, modalias is generated using the modpost utility and the section is added to the module file. When a module is added inside vmlinux, modpost does not generate modalias for such modules and the information is lost. As a result kmod (which uses modules.builtin.modinfo in userspace) cannot determine that modalias is handled by a builtin kernel module. $ cat /sys/devices/pci0000:00/0000:00:14.0/modalias pci:v00008086d0000A36Dsv00001043sd00008694bc0Csc03i30 $ modinfo xhci_pci name: xhci_pci filename: (builtin) license: GPL file: drivers/usb/host/xhci-pci description: xHCI PCI Host Controller Driver Missing modalias "pci:v*d*sv*sd*bc0Csc03i30*" which will be generated by modpost if the module is built separately. To fix this it is necessary to generate the same modalias for vmlinux as for the individual modules. Fortunately '.vmlinux.export.o' is already generated from which '.modinfo' can be extracted in the same way as for vmlinux.o. Signed-off-by: Masahiro Yamada Signed-off-by: Alexey Gladkov Tested-by: Stephen Rothwell Reviewed-by: Nicolas Schier Link: https://patch.msgid.link/28d4da3b0e3fc8474142746bcf469e03752c3208.1758182101.git.legion@kernel.org Signed-off-by: Nathan Chancellor

modpost: Add modname to mod_device_table alias

2025-09-24T16:10:45+00:00

At this point, if a symbol is compiled as part of the kernel, information about which module the symbol belongs to is lost. To save this it is possible to add the module name to the alias name. It's not very pretty, but it's possible for now. Cc: Miguel Ojeda Cc: Andreas Hindborg Cc: Danilo Krummrich Cc: Alex Gaynor Cc: rust-for-linux@vger.kernel.org Signed-off-by: Alexey Gladkov Acked-by: Danilo Krummrich Acked-by: Nicolas Schier Link: https://patch.msgid.link/1a0d0bd87a4981d465b9ed21e14f4e78eaa03ded.1758182101.git.legion@kernel.org Signed-off-by: Nathan Chancellor

Merge tag 'kbuild-v6.17-2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild

2025-08-06T04:32:52+00:00

Pull Kbuild updates from Masahiro Yamada: "This is the last pull request from me. I'm grateful to have been able to continue as a maintainer for eight years. From the next cycle, Nathan and Nicolas will maintain Kbuild. - Fix a shortcut key issue in menuconfig - Fix missing rebuild of kheaders - Sort the symbol dump generated by gendwarfsyms - Support zboot extraction in scripts/extract-vmlinux - Migrate gconfig to GTK 3 - Add TAR variable to allow overriding the default tar command - Hand over Kbuild maintainership" * tag 'kbuild-v6.17-2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild: (92 commits) MAINTAINERS: hand over Kbuild maintenance kheaders: make it possible to override TAR kbuild: userprogs: use correct linker when mixing clang and GNU ld kconfig: lxdialog: replace strcpy() with strncpy() in inputbox.c kconfig: lxdialog: replace strcpy with snprintf in print_autowrap kconfig: gconf: refactor text_insert_help() kconfig: gconf: remove unneeded variable in text_insert_msg kconfig: gconf: use hyphens in signals kconfig: gconf: replace GtkImageMenuItem with GtkMenuItem kconfig: gconf: Fix Back button behavior kconfig: gconf: fix single view to display dependent symbols correctly scripts: add zboot support to extract-vmlinux gendwarfksyms: order -T symtypes output by name gendwarfksyms: use preferred form of sizeof for allocation kconfig: qconf: confine {begin,end}Group to constructor and destructor kconfig: qconf: fix ConfigList::updateListAllforAll() kconfig: add a function to dump all menu entries in a tree-like format kconfig: gconf: show GTK version in About dialog kconfig: gconf: replace GtkHPaned and GtkVPaned with GtkPaned kconfig: gconf: replace GdkColor with GdkRGBA ...

module: Rename MAX_PARAM_PREFIX_LEN to __MODULE_NAME_LEN

2025-07-31T11:57:47+00:00

The maximum module name length (MODULE_NAME_LEN) is somewhat confusingly defined in terms of the maximum parameter prefix length (MAX_PARAM_PREFIX_LEN), when in fact the dependency is in the opposite direction. This split originates from commit 730b69d22525 ("module: check kernel param length at compile time, not runtime"). The code needed to use MODULE_NAME_LEN in moduleparam.h, but because module.h requires moduleparam.h, this created a circular dependency. It was resolved by introducing MAX_PARAM_PREFIX_LEN in moduleparam.h and defining MODULE_NAME_LEN in module.h in terms of MAX_PARAM_PREFIX_LEN. Rename MAX_PARAM_PREFIX_LEN to __MODULE_NAME_LEN for clarity. This matches the similar approach of defining MODULE_INFO in module.h and __MODULE_INFO in moduleparam.h. Signed-off-by: Petr Pavlu Reviewed-by: Daniel Gomez Link: https://lore.kernel.org/r/20250630143535.267745-6-petr.pavlu@suse.com Signed-off-by: Daniel Gomez

module: make structure definitions always visible

2025-07-31T11:51:25+00:00

To write code that works with both CONFIG_MODULES=y and CONFIG_MODULES=n it is convenient to use "if (IS_ENABLED(CONFIG_MODULES))" over raw #ifdef. The code will still fully typechecked but the unreachable parts are discarded by the compiler. This prevents accidental breakage when a certain kconfig combination was not specifically tested by the developer. This pattern is already supported to some extend by module.h defining empty stub functions if CONFIG_MODULES=n. However some users of module.h work on the structured defined by module.h. Therefore these structure definitions need to be visible, too. Many structure members are still gated by specific configuration settings. The assumption for those is that the code using them will be gated behind the same configuration setting anyways. Signed-off-by: Thomas Weißschuh Reviewed-by: Daniel Gomez Link: https://lore.kernel.org/r/20250711-kunit-ifdef-modules-v2-2-39443decb1f8@linutronix.de Signed-off-by: Daniel Gomez

module: move 'struct module_use' to internal.h

2025-07-31T11:40:46+00:00

The struct was moved to the public header file in commit c8e21ced08b3 ("module: fix kdb's illicit use of struct module_use."). Back then the structure was used outside of the module core. Nowadays this is not true anymore, so the structure can be made internal. Signed-off-by: Thomas Weißschuh Reviewed-by: Daniel Gomez Reviewed-by: Petr Pavlu Link: https://lore.kernel.org/r/20250711-kunit-ifdef-modules-v2-1-39443decb1f8@linutronix.de Signed-off-by: Daniel Gomez

Merge tag 'ftrace-v6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace

2025-07-30T23:04:10+00:00

Pull ftrace updates from Steven Rostedt: - Keep track of when fgraph_ops are registered or not Keep accounting of when fgraph_ops are registered as if a fgraph_ops is registered twice it can mess up the accounting and it will not work as expected later. Trigger a warning if something registers it twice as to catch bugs before they are found by things just not working as expected. - Make DYNAMIC_FTRACE always enabled for architectures that support it As static ftrace (where all functions are always traced) is very expensive and only exists to help architectures support ftrace, do not make it an option. As soon as an architecture supports DYNAMIC_FTRACE make it use it. This simplifies the code. - Remove redundant config HAVE_FTRACE_MCOUNT_RECORD The CONFIG_HAVE_FTRACE_MCOUNT was added to help simplify the DYNAMIC_FTRACE work, but now every architecture that implements DYNAMIC_FTRACE also has HAVE_FTRACE_MCOUNT set too, making it redundant with the HAVE_DYNAMIC_FTRACE. - Make pid_ptr string size match the comment In print_graph_proc() the pid_ptr string is of size 11, but the comment says /* sign + log10(MAX_INT) + '\0' */ which is actually 12. * tag 'ftrace-v6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: tracing: Remove redundant config HAVE_FTRACE_MCOUNT_RECORD ftrace: Make DYNAMIC_FTRACE always enabled for architectures that support it fgraph: Keep track of when fgraph_ops are registered or not fgraph: Make pid_str size match the comment

Merge tag 'probes-v6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace

2025-07-30T22:38:01+00:00

Pull probes updates from Masami Hiramatsu: "Stack usage reduction for probe events: - Allocate string buffers from the heap for uprobe, eprobe, kprobe, and fprobe events to avoid stack overflow - Allocate traceprobe_parse_context from the heap to prevent potential stack overflow - Fix a typo in the above commit New features for eprobe and tprobe events: - Add support for arrays in eprobes - Support multiple tprobes on the same tracepoint Improve efficiency: - Register fprobe-events only when it is enabled to reduce overhead - Register tracepoints for tprobe events only when enabled to resolve a lock dependency Code Cleanup: - Add kerneldoc for traceprobe_parse_event_name() and __get_insn_slot() - Sort #include alphabetically in the probes code - Remove the unused 'mod' field from the tprobe-event - Clean up the entry-arg storing code in probe-events Selftest update - Enable fprobe events before checking enable_functions in selftests" * tag 'probes-v6.17' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: tracing: trace_fprobe: Fix typo of the semicolon tracing: Have eprobes handle arrays tracing: probes: Add a kerneldoc for traceprobe_parse_event_name() tracing: uprobe-event: Allocate string buffers from heap tracing: eprobe-event: Allocate string buffers from heap tracing: kprobe-event: Allocate string buffers from heap tracing: fprobe-event: Allocate string buffers from heap tracing: probe: Allocate traceprobe_parse_context from heap tracing: probes: Sort #include alphabetically kprobes: Add missing kerneldoc for __get_insn_slot tracing: tprobe-events: Register tracepoint when enable tprobe event selftests: tracing: Enable fprobe events before checking enable_functions tracing: fprobe-events: Register fprobe-events only when it is enabled tracing: tprobe-events: Support multiple tprobes on the same tracepoint tracing: tprobe-events: Remove mod field from tprobe-event tracing: probe-events: Cleanup entry-arg storing code