lkml.org 
[lkml]   [2014]   [Jul]   [22]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [PATCH 6/8] mm/highmem: make kmap cache coloring aware
From
On Tue, Jul 22, 2014 at 11:35 PM, Leonid Yegoshin
<Leonid.Yegoshin@imgtec.com> wrote:
> On 07/22/2014 12:01 PM, Max Filippov wrote:
>>
>> From: Leonid Yegoshin <Leonid.Yegoshin@imgtec.com>
>>
>> Provide hooks that allow architectures with aliasing cache to align
>> mapping address of high pages according to their color. Such architectures
>> may enforce similar coloring of low- and high-memory page mappings and
>> reuse existing cache management functions to support highmem.
>>
>> Cc: linux-mm@kvack.org
>> Cc: linux-arch@vger.kernel.org
>> Cc: linux-mips@linux-mips.org
>> Cc: David Rientjes <rientjes@google.com>
>> Signed-off-by: Leonid Yegoshin <Leonid.Yegoshin@imgtec.com>
>> [ Max: extract architecture-independent part of the original patch, clean
>> up checkpatch and build warnings. ]
>> Signed-off-by: Max Filippov <jcmvbkbc@gmail.com>
>> ---
>> Changes since the initial version:
>> - define set_pkmap_color(pg, cl) as do { } while (0) instead of /* */;
>> - rename is_no_more_pkmaps to no_more_pkmaps;
>> - change 'if (count > 0)' to 'if (count)' to better match the original
>> code behavior;
>>
>> mm/highmem.c | 19 ++++++++++++++++---
>> 1 file changed, 16 insertions(+), 3 deletions(-)
>>
>> diff --git a/mm/highmem.c b/mm/highmem.c
>> index b32b70c..88fb62e 100644
>> --- a/mm/highmem.c
>> +++ b/mm/highmem.c
>> @@ -44,6 +44,14 @@ DEFINE_PER_CPU(int, __kmap_atomic_idx);
>> */
>> #ifdef CONFIG_HIGHMEM
>> +#ifndef ARCH_PKMAP_COLORING
>> +#define set_pkmap_color(pg, cl) do { } while (0)
>> +#define get_last_pkmap_nr(p, cl) (p)
>> +#define get_next_pkmap_nr(p, cl) (((p) + 1) & LAST_PKMAP_MASK)
>> +#define no_more_pkmaps(p, cl) (!(p))
>> +#define get_next_pkmap_counter(c, cl) ((c) - 1)
>> +#endif
>> +
>> unsigned long totalhigh_pages __read_mostly;
>> EXPORT_SYMBOL(totalhigh_pages);
>> @@ -161,19 +169,24 @@ static inline unsigned long map_new_virtual(struct
>> page *page)
>> {
>> unsigned long vaddr;
>> int count;
>> + int color __maybe_unused;
>> +
>> + set_pkmap_color(page, color);
>> + last_pkmap_nr = get_last_pkmap_nr(last_pkmap_nr, color);
>> start:
>> count = LAST_PKMAP;
>> /* Find an empty entry */
>> for (;;) {
>> - last_pkmap_nr = (last_pkmap_nr + 1) & LAST_PKMAP_MASK;
>> - if (!last_pkmap_nr) {
>> + last_pkmap_nr = get_next_pkmap_nr(last_pkmap_nr, color);
>> + if (no_more_pkmaps(last_pkmap_nr, color)) {
>> flush_all_zero_pkmaps();
>> count = LAST_PKMAP;
>> }
>> if (!pkmap_count[last_pkmap_nr])
>> break; /* Found a usable entry */
>> - if (--count)
>> + count = get_next_pkmap_counter(count, color);
>> + if (count)
>> continue;
>> /*
>
> I would like to return back to "if (count >0)".
>
> The reason is in easy way to jump through the same coloured pages - next
> element is calculated via decrementing by non "1" value of colours and it
> can easy become negative on last page available:
>
> #define get_next_pkmap_counter(c,cl) (c - FIX_N_COLOURS)
>
> where FIX_N_COLOURS is a max number of page colours.

Initial value of c (i.e. LAST_PKMAP) should be a multiple of FIX_N_COLOURS,
so that should not be a problem.

> Besides that it is a good practice in stopping cycle.

But I agree with that.

--
Thanks.
-- Max


\
 
 \ /
  Last update: 2014-07-22 22:41    [W:0.080 / U:0.128 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site