Is iounmap(NULL) safe or not?

[Alexey Khoroshilov]

Hello,

It seems thare are many places where code assumes iounmap(NULL) is safe.
Also there are several patches that state it explicitly:
  ff6defa6a8fa ("ALSA: Deletion of checks before the function call "iounmap")
  e24bb0ed8179 ("staging: dgnc: remove NULL test")

At the same time it seems PPC implementation generates a warning in this case:
  3bfafd6b136b ("netxen: avoid invalid iounmap")

  arch/powerpc/mm/pgtable_64.c:
	if ((unsigned long)addr < ioremap_bot) {
		printk(KERN_WARNING "Attempt to iounmap early bolted mapping"
		       " at 0x%p\n", addr);
		return;
	}

Could you please clarify if iounmap(NULL) safe or not.
I guess it would be less errorprone if the answer is architecture independent.

--
Thank you,
Alexey Khoroshilov
Linux Verification Center, ISPRAS
web: http://linuxtesting.org