Re: linux-next: manual merge of the akpm-current tree with the slab tree

[Stephen Rothwell]

Hi all,

On Fri, 3 Dec 2021 18:22:55 +1100 Stephen Rothwell <sfr@xxxxxxxxxxxxxxxx> wrote:
>
> Today's linux-next merge of the akpm-current tree got a conflict in:
> 
>   mm/zsmalloc.c
> 
> between commit:
> 
>   5718b5cdd6e5 ("zsmalloc: Stop using slab fields in struct page")
> 
> from the slab tree and commit:
> 
>   efcc1718fe08 ("zsmalloc: move huge compressed obj from page to zspage")
> 
> from the akpm-current tree.
> 
> I fixed it up (see below) and can carry the fix as necessary. This
> is now fixed as far as linux-next is concerned, but any non trivial
> conflicts should be mentioned to your upstream maintainer when your tree
> is submitted for merging.  You may also want to consider cooperating
> with the maintainer of the conflicting tree to minimise any particularly
> complex conflicts.
> 
> diff --cc mm/zsmalloc.c
> index 0d3b65939016,7e03cc9363bb..000000000000
> --- a/mm/zsmalloc.c
> +++ b/mm/zsmalloc.c
> @@@ -824,10 -820,12 +820,12 @@@ static struct zspage *get_zspage(struc
>   
>   static struct page *get_next_page(struct page *page)
>   {
> - 	if (unlikely(PageHugeObject(page)))
> + 	struct zspage *zspage = get_zspage(page);
> + 
> + 	if (unlikely(ZsHugePage(zspage)))
>   		return NULL;
>   
>  -	return page->freelist;
>  +	return (struct page *)page->index;
>   }
>   
>   /**
> @@@ -900,8 -893,7 +893,7 @@@ static void reset_page(struct page *pag
>   	ClearPagePrivate(page);
>   	set_page_private(page, 0);
>   	page_mapcount_reset(page);
> - 	ClearPageHugeObject(page);
>  -	page->freelist = NULL;
>  +	page->index = 0;
>   }
>   
>   static int trylock_zspage(struct zspage *zspage)
> @@@ -1042,9 -1039,9 +1039,9 @@@ static void create_page_chain(struct si
>   			SetPagePrivate(page);
>   			if (unlikely(class->objs_per_zspage == 1 &&
>   					class->pages_per_zspage == 1))
> - 				SetPageHugeObject(page);
> + 				SetZsHugePage(zspage);
>   		} else {
>  -			prev_page->freelist = page;
>  +			prev_page->index = (unsigned long)page;
>   		}
>   		prev_page = page;
>   	}

This is now a conflict between the akpm-current tree and Linus' tree -
that slab tree commit is now commit ffedd09fa9b0 ("zsmalloc: Stop using
slab fields in struct page") in Linus' tree.

-- 
Cheers,
Stephen Rothwell

Attachment: pgpiFRHwbZqjs.pgp
Description: OpenPGP digital signature