For radix we split the mapping into smaller page sizes (at the cost of
additional TLB overhead), but for hash its best to print a warning. In
the case of hash and no-relocation, the kernel should be well aligned
to provide the least overhead with the current linear mapping size (16M)
Signed-off-by: Balbir Singh <bsinghar...@gmail.com>
---
arch/powerpc/mm/pgtable-hash64.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/mm/pgtable-hash64.c b/arch/powerpc/mm/pgtable-hash64.c
index 0809102b..7c2479d 100644
--- a/arch/powerpc/mm/pgtable-hash64.c
+++ b/arch/powerpc/mm/pgtable-hash64.c
@@ -438,6 +438,11 @@ void hash__mark_rodata_ro(void)
return;
}
+ if (PHYSICAL_START > MEMORY_START)
+ pr_warn("Detected relocation and CONFIG_STRICT_KERNEL_RWX "
+ "permissions are best effort, some non-text area "
+ "might still be left as executable");
+
shift = mmu_psize_defs[mmu_linear_psize].shift;
step = 1 << shift;
@@ -448,7 +453,7 @@ void hash__mark_rodata_ro(void)
start, end, step);
if (start == end) {
- pr_warn("could not set rodata ro, relocate the start"
+ pr_warn("Could not set rodata ro, relocate the start"
" of the kernel to a 0x%x boundary\n", step);
return;
}
--
2.9.4
