[lkml]   [2008]   [Feb]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH] [5/8] Fix logic error in 64bit memory hotadd

* Andi Kleen <> wrote:

> The memory hotadd code assumes that the pud always exists already, but
> that might be not true. Allocate it if it isn't there.

ok, this seems an like an ancient memory-hotplug bug. Does anyone even
use memory hotplug currently? Did you find this bug via review, or did
it trigger in practice?

Also, your fix, while it solves a real bug we want to fix, is not quite
right for upstream integration yet. I can see 3 immediate problems with

> + if (!pud_present(*pud)) {
> + pud = (pud_t *)get_zeroed_page(GFP_ATOMIC);

the GFP_ATOMIC here can fail.

The proper solution is to instead extend init_memory_mapping() with a
gfp_t parameter and pass in GFP_ATOMIC from the early init code (where
we must not schedule and where GFP_ATOMIC will succeed anyway), but do a
GFP_KERNEL from arch_add_memory().

> + if (!pud)
> + panic("Out of memory");

the panic() here is very rude to the user in the hotplug usecase.

The proper solution is to extend init_memory_mapping() with a return
value, and to check in the caller. arch_add_memory() obviously does not
want to panic(), it wants to return -ENOMEM to mm/memory_hotplug.c.

and a small style nit while changing this code:

> + } else
> pud = alloc_low_page(&pud_phys);

please add curly braces to the 'else' branch too. (we generally prefer
symmetrical curly braces) Thanks,


 \ /
  Last update: 2008-02-11 13:49    [W:0.130 / U:1.540 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site