2019-06-14 06:04:32

by Wei Yang

Subject: [PATCH] kernel/memremap.c: use ALIGN/ALIGN_DOWN to calculate align_start/end

The purpose of align_start/end is to expand to SECTION boundary. Use
ALIGN/ALIGN_DOWN directly is more self-explain and clean.

Signed-off-by: Wei Yang <[email protected]>
---
kernel/memremap.c | 7 +++----
1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/kernel/memremap.c b/kernel/memremap.c
index 1490e63f69a9..53cf751f0721 100644
--- a/kernel/memremap.c
+++ b/kernel/memremap.c
@@ -159,10 +159,9 @@ void *devm_memremap_pages(struct device *dev, struct dev_pagemap *pgmap)
if (!pgmap->ref || !pgmap->kill)
return ERR_PTR(-EINVAL);

- align_start = res->start & ~(SECTION_SIZE - 1);
- align_size = ALIGN(res->start + resource_size(res), SECTION_SIZE)
- - align_start;
- align_end = align_start + align_size - 1;
+ align_start = ALIGN_DOWN(res->start, SECTION_SIZE);
+ align_end = ALIGN(res->end, SECTION_SIZE) - 1;
+ align_size = align_end - align_start + 1;

conflict_pgmap = get_dev_pagemap(PHYS_PFN(align_start), NULL);
if (conflict_pgmap) {
--
2.19.1

2019-06-14 06:07:36

by Dan Williams

Subject: Re: [PATCH] kernel/memremap.c: use ALIGN/ALIGN_DOWN to calculate align_start/end

On Thu, Jun 13, 2019 at 11:02 PM Wei Yang <[email protected]> wrote:
>
> The purpose of align_start/end is to expand to SECTION boundary. Use
> ALIGN/ALIGN_DOWN directly is more self-explain and clean.

I'm actively trying to kill this code [1] so I don't see the need for
this patch.

[1]: https://lore.kernel.org/lkml/155977193326.2443951.14201009973429527491.stgit@dwillia2-desk3.amr.corp.intel.com/

2019-06-14 08:11:33

by Wei Yang

Subject: Re: [PATCH] kernel/memremap.c: use ALIGN/ALIGN_DOWN to calculate align_start/end

On Thu, Jun 13, 2019 at 11:06:13PM -0700, Dan Williams wrote:
>On Thu, Jun 13, 2019 at 11:02 PM Wei Yang <[email protected]> wrote:
>>
>> The purpose of align_start/end is to expand to SECTION boundary. Use
>> ALIGN/ALIGN_DOWN directly is more self-explain and clean.
>
>I'm actively trying to kill this code [1] so I don't see the need for
>this patch.
>
>[1]: https://lore.kernel.org/lkml/155977193326.2443951.14201009973429527491.stgit@dwillia2-desk3.amr.corp.intel.com/

Ah, thanks :-)

--
Wei Yang
Help you, Help me