On Sun, May 11, 2003 at 10:08:41AM +0200, Andi Kleen wrote:
> put_page_dirty must use the page protection of the stack VMA, not hardcoded
> PAGE_COPY. They can be different e.g. when the stack is set non executable
> via VM_STACK_FLAGS.
> I added an fallback path for now because I'm not sure if the stack VMA
> is always predowngrowed here, if the printk better it may be also needed
> to add an stack extension here.
We know which vma is involved at the callsite and what we just set its
vma->vm_page_prot to; I suggest this patch instead.
-- wli
diff -prauN linux-2.5.69-1/fs/exec.c exec-2.5.69-1/fs/exec.c
--- linux-2.5.69-1/fs/exec.c Wed Apr 16 15:34:51 2003
+++ exec-2.5.69-1/exec.c Sun May 11 01:27:47 2003
@@ -314,7 +314,7 @@
}
lru_cache_add_active(page);
flush_dcache_page(page);
- set_pte(pte, pte_mkdirty(pte_mkwrite(mk_pte(page, PAGE_COPY))));
+ set_pte(pte, pte_mkdirty(pte_mkwrite(mk_pte(page, protection_map[VM_STACK_FLAGS & 0x7]))));
pte_chain = page_add_rmap(page, pte, pte_chain);
pte_unmap(pte);
tsk->mm->rss++;
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Thu May 15 2003 - 22:00:35 EST