Re: irq-disabled vs vmap vs text_poke

[Masami Hiramatsu <mhiramat@redhat.com>]

Peter Zijlstra wrote:
> On Tue, 2009-02-17 at 04:03 +0100, Nick Piggin wrote:
> 
>> It is not a bad idea, but I don't think it quite goes far enough.
>> IMO we should reserve 2 pages of virtual memory for each CPU, and
>> then do the mapping/unmapping without locking, and with another
>> variant of unmap_kernel_range that does not do the global TLB
>> flush.
> 
> A bit like kmap_atomic() except for 2 pages and !highmem. Should work.

Hmm, I checked up the kmap_atomic(), but all kernel text
is not on the highmem.

So, what about changing kmap_atomic_prot() implementation as below?

- If the page is !highmem *and the page's prot is same as caller specified prot*,
  just returns page_address(page).
- Add 2 continuous km_types for text_poke().

Then, text_poke() can use it.
---
vaddr = kmap_atomic_prot(page[0], KM_TEXT_POKE0, PAGE_KERNEL);
if(nr_pages > 1)
	vaddr1 = kmap_atomic_prot(page[1], KM_TEXT_POKE1, PAGE_KERNEL);
< change text >
if(nr_pages > 1)
	kunmap_atomic(vaddr1, KM_TEXT_POKE1);
kunmap_atomic(vaddr, KM_TEXT_POKE0);
---

Thank you,

-- 
Masami Hiramatsu

Software Engineer
Hitachi Computer Products (America) Inc.
Software Solutions Division

e-mail: mhiramat@redhat.com