Hi Geert !
On 11/5/24 15:03, Geert Uytterhoeven wrote:
Hi Jean-Michel,
On Wed, Oct 16, 2024 at 6:18 PM Jean-Michel Hautbois
<jeanmichel.hautbois@xxxxxxxxxx> wrote:
The jump_label_init() should be called from setup_arch() very
early for proper functioning of jump label support.
Signed-off-by: Jean-Michel Hautbois <jeanmichel.hautbois@xxxxxxxxxx>
Thanks for your patch!
--- a/arch/m68k/kernel/setup_mm.c
+++ b/arch/m68k/kernel/setup_mm.c
@@ -249,7 +249,11 @@ void __init setup_arch(char **cmdline_p)
process_uboot_commandline(&m68k_command_line[0], CL_SIZE);
*cmdline_p = m68k_command_line;
memcpy(boot_command_line, *cmdline_p, CL_SIZE);
-
+ /*
+ * Initialise the static keys early as they may be enabled by the
+ * cpufeature code and early parameters.
+ */
+ jump_label_init();
parse_early_param();
switch (m68k_machtype) {
This is indeed what some (but not all) other architectures are doing, so
Reviewed-by: Geert Uytterhoeven <geert@xxxxxxxxxxxxxx>
I assume you saw some "static key used before call to jump_label_init()"
warning[1]? Since I never saw such a message, can you please elaborate
and explain your use case, so I can add that to the patch description
when applying?
Indeed ! I saw this when I was passing the "threadirqs" parameter to the
kernel commandline and the "select IRQ_FORCED_THREADING" line to the
Kconfig.
I suspect this might be true for other keys.
BTW, threaded IRQs work fine ;-).
Thanks,
JM
Thanks!
[1] https://elixir.bootlin.com/linux/v6.11.6/source/include/linux/jump_label.h#L81
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
