With STRICT_KERNEL_RWX on in a relocatable kernel under the hash MMU,
if the position the kernel is loaded at is not 16M aligned things go
horribly wrong. Specifically hash__mark_initmem_nx() will call
hash__change_memory_range() which then aligns down the start address,
and due to the text not being 16M aligned causes some of the kernel
text to be marked non-executable.
We can avoid this when selecting the linear mapping size, so do so and
print a warning. I tested this for various alignments and as long as
the position is 64K aligned it's fine (the base requirement for
powerpc).
Signed-off-by: Russell Currey <ruscur@russell.cc>
[mpe: Add details of the failure mode]
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://lore.kernel.org/r/20191224064126.183670-1-ruscur@russell.cc
static void __init htab_init_page_sizes(void)
{
+ bool aligned = true;
init_hpte_page_sizes();
if (!debug_pagealloc_enabled()) {
* Pick a size for the linear mapping. Currently, we only
* support 16M, 1M and 4K which is the default
*/
- if (mmu_psize_defs[MMU_PAGE_16M].shift)
+ if (IS_ENABLED(STRICT_KERNEL_RWX) &&
+ (unsigned long)_stext % 0x1000000) {
+ if (mmu_psize_defs[MMU_PAGE_16M].shift)
+ pr_warn("Kernel not 16M aligned, "
+ "disabling 16M linear map alignment");
+ aligned = false;
+ }
+
+ if (mmu_psize_defs[MMU_PAGE_16M].shift && aligned)
mmu_linear_psize = MMU_PAGE_16M;
else if (mmu_psize_defs[MMU_PAGE_1M].shift)
mmu_linear_psize = MMU_PAGE_1M;