]> www.infradead.org Git - users/dwmw2/linux.git/commit
x86/alternatives: Remove the confusing, inaccurate & unnecessary 'temp_mm_state_t...
authorIngo Molnar <mingo@kernel.org>
Fri, 11 Apr 2025 05:40:25 +0000 (07:40 +0200)
committerIngo Molnar <mingo@kernel.org>
Fri, 11 Apr 2025 09:01:33 +0000 (11:01 +0200)
commitf5afa2e8efda592ecc69cea7528ff660ac1d8096
treede7744938dcb10c9f645d691bb5ffcf98f0eff9c
parent5224f09a7b57fcf2024245d89dcb26b0756fb1c8
x86/alternatives: Remove the confusing, inaccurate & unnecessary 'temp_mm_state_t' abstraction

So the temp_mm_state_t abstraction used by use_temporary_mm() and
unuse_temporary_mm() is super confusing:

 - The whole machinery is about temporarily switching to the
   text_poke_mm utility MM that got allocated during bootup
   for text-patching purposes alone:

temp_mm_state_t prev;

        /*
         * Loading the temporary mm behaves as a compiler barrier, which
         * guarantees that the PTE will be set at the time memcpy() is done.
         */
        prev = use_temporary_mm(text_poke_mm);

 - Yet the value that gets saved in the temp_mm_state_t variable
   is not the temporary MM ... but the previous MM...

 - Ie. we temporarily put the non-temporary MM into a variable
   that has the temp_mm_state_t type. This makes no sense whatsoever.

 - The confusion continues in unuse_temporary_mm():

static inline void unuse_temporary_mm(temp_mm_state_t prev_state)

   Here we unuse an MM that is ... not the temporary MM, but the
   previous MM. :-/

Fix up all this confusion by removing the unnecessary layer of
abstraction and using a bog-standard 'struct mm_struct *prev_mm'
variable to save the MM to.

Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Juergen Gross <jgross@suse.com>
Cc: "H . Peter Anvin" <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250411054105.2341982-14-mingo@kernel.org
arch/x86/kernel/alternative.c