[PATCH net-next v2] page_pool: fix build on powerpc with GCC 14

[Matthew Wilcox]

On Fri, Sep 13, 2024 at 09:33:51PM +0000, Mina Almasry wrote:
> Building net-next with powerpc with GCC 14 compiler results in this
> build error:
> 
> /home/sfr/next/tmp/ccuSzwiR.s: Assembler messages:
> /home/sfr/next/tmp/ccuSzwiR.s:2579: Error: operand out of domain (39 is
> not a multiple of 4)
> make[5]: *** [/home/sfr/next/next/scripts/Makefile.build:229:
> net/core/page_pool.o] Error 1
> 
> Root caused in this thread:
> https://lore.kernel.org/netdev/913e2fbd-d318-4c9b-aed2-4d333a1d5cf0@cs-soprasteria.com/

It would be better to include a direct link to the GCC bugzilla.

> We try to access offset 40 in the pointer returned by this function:
> 
> static inline unsigned long _compound_head(const struct page *page)
> {
>         unsigned long head = READ_ONCE(page->compound_head);
> 
>         if (unlikely(head & 1))
>                 return head - 1;
>         return (unsigned long)page_fixed_fake_head(page);
> }
> 
> The GCC 14 (but not 11) compiler optimizes this by doing:
> 
> ld page + 39
> 
> Rather than:
> 
> ld (page - 1) + 40
> 
> And causing an unaligned load. Get around this by issuing a READ_ONCE as
> we convert the page to netmem.  That disables the compiler optimizing the
> load in this way.
> 
> Cc: Simon Horman <horms at kernel.org>
> Cc: Stephen Rothwell <sfr at canb.auug.org.au>
> Cc: Jakub Kicinski <kuba at kernel.org>
> Cc: David Miller <davem at davemloft.net>
> Cc: Paolo Abeni <pabeni at redhat.com>
> Cc: Networking <netdev at vger.kernel.org>
> Cc: Linux Kernel Mailing List <linux-kernel at vger.kernel.org>
> Cc: Linux Next Mailing List <linux-next at vger.kernel.org>
> Cc: Arnd Bergmann <arnd at arndb.de>
> Cc: "linuxppc-dev at lists.ozlabs.org" <linuxppc-dev at lists.ozlabs.org>
> Cc: Matthew Wilcox <willy at infradead.org>
> 
> Suggested-by: Jakub Kicinski <kuba at kernel.org>
> Signed-off-by: Mina Almasry <almasrymina at google.com>
> 
> ---
> 
> v2: https://lore.kernel.org/netdev/20240913192036.3289003-1-almasrymina@google.com/
> 
> - Work around this issue as we convert the page to netmem, instead of
>   a generic change that affects compound_head().
> ---
>  net/core/page_pool.c | 15 ++++++++++++++-
>  1 file changed, 14 insertions(+), 1 deletion(-)
> 
> diff --git a/net/core/page_pool.c b/net/core/page_pool.c
> index a813d30d2135..74ea491d0ab2 100644
> --- a/net/core/page_pool.c
> +++ b/net/core/page_pool.c
> @@ -859,12 +859,25 @@ void page_pool_put_page_bulk(struct page_pool *pool, void **data,
>  {
>  	int i, bulk_len = 0;
>  	bool allow_direct;
> +	netmem_ref netmem;
> +	struct page *page;
>  	bool in_softirq;
>  
>  	allow_direct = page_pool_napi_local(pool);
>  
>  	for (i = 0; i < count; i++) {
> -		netmem_ref netmem = page_to_netmem(virt_to_head_page(data[i]));
> +		page = virt_to_head_page(data[i]);
> +
> +		/* GCC 14 powerpc compiler will optimize reads into the
> +		 * resulting netmem_ref into unaligned reads as it sees address
> +		 * arithmetic in _compound_head() call that the page has come
> +		 * from.
> +		 *
> +		 * The READ_ONCE here gets around that by breaking the
> +		 * optimization chain between the address arithmetic and later
> +		 * indexing.
> +		 */
> +		netmem = page_to_netmem(READ_ONCE(page));
>  
>  		/* It is not the last user for the page frag case */
>  		if (!page_pool_is_last_ref(netmem))
> -- 
> 2.46.0.662.g92d0881bb0-goog
>