diff options
| author |   Andy Lutomirski <luto@kernel.org> | 2017-09-17 09:03:51 -0700 |
|---|---|---|
| committer |   Ingo Molnar <mingo@kernel.org> | 2017-09-17 18:59:09 +0200 |
| commit | 4ba55e65f471d011d3ba2ac2022180ea0877d68e (patch) | |
| tree | 1f2760ddbd6d5e8b97f824206b506ee9a892e2c4 /tools/objtool/arch | |
| parent | x86/mm/32: Move setup_clear_cpu_cap(X86_FEATURE_PCID) earlier (diff) | |
| download | linux-dev-4ba55e65f471d011d3ba2ac2022180ea0877d68e.tar.xz linux-dev-4ba55e65f471d011d3ba2ac2022180ea0877d68e.zip | |
x86/mm/32: Load a sane CR3 before cpu_init() on secondary CPUs
For unknown historical reasons (i.e. Borislav doesn't recall),
32-bit kernels invoke cpu_init() on secondary CPUs with
initial_page_table loaded into CR3. Then they set
current->active_mm to &init_mm and call enter_lazy_tlb() before
fixing CR3. This means that the x86 TLB code gets invoked while CR3
is inconsistent, and, with the improved PCID sanity checks I added,
we warn.
Fix it by loading swapper_pg_dir (i.e. init_mm.pgd) earlier.
Reported-by: Paul Menzel <pmenzel@molgen.mpg.de>
Reported-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Borislav Petkov <bpetkov@suse.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 72c0098d92ce ("x86/mm: Reinitialize TLB state on hotplug and resume")
Link: http://lkml.kernel.org/r/30cdfea504682ba3b9012e77717800a91c22097f.1505663533.git.luto@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Diffstat (limited to 'tools/objtool/arch')
0 files changed, 0 insertions, 0 deletions