From: Yunsheng Lin <linyunsheng@huawei.com>
To: Paolo Abeni <pabeni@redhat.com>, <davem@davemloft.net>,
<kuba@kernel.org>
Cc: <netdev@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
Alexander Duyck <alexanderduyck@fb.com>,
Alexander Duyck <alexander.duyck@gmail.com>,
"Michael S. Tsirkin" <mst@redhat.com>,
Jason Wang <jasowang@redhat.com>,
Andrew Morton <akpm@linux-foundation.org>,
Eric Dumazet <edumazet@google.com>, <kvm@vger.kernel.org>,
<virtualization@lists.linux.dev>, <linux-mm@kvack.org>
Subject: Re: [PATCH net-next v4 2/5] page_frag: unify gfp bits for order 3 page allocation
Date: Fri, 2 Feb 2024 10:10:02 +0800 [thread overview]
Message-ID: <2e8606b1-81c2-6f3f-622c-607db5e90253@huawei.com> (raw)
In-Reply-To: <81c37127dda0f2f69a019d67d4420f62c995ee7f.camel@redhat.com>
On 2024/2/1 21:16, Paolo Abeni wrote:
> On Tue, 2024-01-30 at 19:37 +0800, Yunsheng Lin wrote:
>> Currently there seems to be three page frag implementions
>> which all try to allocate order 3 page, if that fails, it
>> then fail back to allocate order 0 page, and each of them
>> all allow order 3 page allocation to fail under certain
>> condition by using specific gfp bits.
>>
>> The gfp bits for order 3 page allocation are different
>> between different implementation, __GFP_NOMEMALLOC is
>> or'd to forbid access to emergency reserves memory for
>> __page_frag_cache_refill(), but it is not or'd in other
>> implementions, __GFP_DIRECT_RECLAIM is masked off to avoid
>> direct reclaim in skb_page_frag_refill(), but it is not
>> masked off in __page_frag_cache_refill().
>>
>> This patch unifies the gfp bits used between different
>> implementions by or'ing __GFP_NOMEMALLOC and masking off
>> __GFP_DIRECT_RECLAIM for order 3 page allocation to avoid
>> possible pressure for mm.
>>
>> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
>> Reviewed-by: Alexander Duyck <alexanderduyck@fb.com>
>> CC: Alexander Duyck <alexander.duyck@gmail.com>
>> ---
>> drivers/vhost/net.c | 2 +-
>> mm/page_alloc.c | 4 ++--
>> net/core/sock.c | 2 +-
>> 3 files changed, 4 insertions(+), 4 deletions(-)
>>
>> diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
>> index f2ed7167c848..e574e21cc0ca 100644
>> --- a/drivers/vhost/net.c
>> +++ b/drivers/vhost/net.c
>> @@ -670,7 +670,7 @@ static bool vhost_net_page_frag_refill(struct vhost_net *net, unsigned int sz,
>> /* Avoid direct reclaim but allow kswapd to wake */
>> pfrag->page = alloc_pages((gfp & ~__GFP_DIRECT_RECLAIM) |
>> __GFP_COMP | __GFP_NOWARN |
>> - __GFP_NORETRY,
>> + __GFP_NORETRY | __GFP_NOMEMALLOC,
>> SKB_FRAG_PAGE_ORDER);
>
>> if (likely(pfrag->page)) {
>> pfrag->size = PAGE_SIZE << SKB_FRAG_PAGE_ORDER;
>> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
>> index c0f7e67c4250..636145c29f70 100644
>> --- a/mm/page_alloc.c
>> +++ b/mm/page_alloc.c
>> @@ -4685,8 +4685,8 @@ static struct page *__page_frag_cache_refill(struct page_frag_cache *nc,
>> gfp_t gfp = gfp_mask;
>>
>> #if (PAGE_SIZE < PAGE_FRAG_CACHE_MAX_SIZE)
>> - gfp_mask |= __GFP_COMP | __GFP_NOWARN | __GFP_NORETRY |
>> - __GFP_NOMEMALLOC;
>> + gfp_mask = (gfp_mask & ~__GFP_DIRECT_RECLAIM) | __GFP_COMP |
>> + __GFP_NOWARN | __GFP_NORETRY | __GFP_NOMEMALLOC;
>> page = alloc_pages_node(NUMA_NO_NODE, gfp_mask,
>> PAGE_FRAG_CACHE_MAX_ORDER);
>> nc->size = page ? PAGE_FRAG_CACHE_MAX_SIZE : PAGE_SIZE;
>> diff --git a/net/core/sock.c b/net/core/sock.c
>> index 88bf810394a5..8289a3d8c375 100644
>> --- a/net/core/sock.c
>> +++ b/net/core/sock.c
>> @@ -2919,7 +2919,7 @@ bool skb_page_frag_refill(unsigned int sz, struct page_frag *pfrag, gfp_t gfp)
>> /* Avoid direct reclaim but allow kswapd to wake */
>> pfrag->page = alloc_pages((gfp & ~__GFP_DIRECT_RECLAIM) |
>> __GFP_COMP | __GFP_NOWARN |
>> - __GFP_NORETRY,
>> + __GFP_NORETRY | __GFP_NOMEMALLOC,
>> SKB_FRAG_PAGE_ORDER);
>
> This will prevent memory reserve usage when allocating order 3 pages,
> but not when allocating a single page as a fallback. Still different
More accurately, the above ensures memory reserve is always not used
for order 3 pages, whether memory reserve is used for order 0 pages
depending on original 'gfp' flags, if 'gfp' does not have __GFP_NOMEMALLOC
bit set, memory reserve may still be used for order 0 pages.
> from the __page_frag_cache_refill() allocator - which never accesses
> the memory reserves.
I am not really sure I understand the above commemt.
The semantic is the same as skb_page_frag_refill() as explained above
as my understanding. Note that __page_frag_cache_refill() use 'gfp_mask'
for allocating order 3 pages and use the original 'gfp' for allocating
order 0 pages.
>
> I'm unsure we want to propagate the __page_frag_cache_refill behavior
> here, the current behavior could be required by some systems.
>
> It looks like this series still leave the skb_page_frag_refill()
> allocator alone, what about dropping this chunk, too?
As explained above, I would prefer to keep it as it is as it seems
to be quite obvious that we can avoid possible pressure for mm by
not using memory reserve for order 3 pages as we have the fallback
for order 0 pages.
Please let me know if there is anything obvious I missed.
>
> Thanks!
>
> Paolo
>
>
> .
>
next prev parent reply other threads:[~2024-02-02 2:10 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-30 11:37 [PATCH net-next v4 0/5] remove page frag implementation in vhost_net Yunsheng Lin
2024-01-30 11:37 ` [PATCH net-next v4 1/5] mm/page_alloc: modify page_frag_alloc_align() to accept align as an argument Yunsheng Lin
2024-01-30 11:37 ` [PATCH net-next v4 2/5] page_frag: unify gfp bits for order 3 page allocation Yunsheng Lin
2024-02-01 13:16 ` Paolo Abeni
2024-02-02 2:10 ` Yunsheng Lin [this message]
2024-02-02 8:36 ` Paolo Abeni
2024-02-02 12:26 ` Yunsheng Lin
2024-01-30 11:37 ` [PATCH net-next v4 3/5] net: introduce page_frag_cache_drain() Yunsheng Lin
2024-01-30 11:37 ` [PATCH net-next v4 4/5] vhost/net: remove vhost_net_page_frag_refill() Yunsheng Lin
2024-01-30 11:37 ` [PATCH net-next v4 5/5] tools: virtio: introduce vhost_net_test Yunsheng Lin
2024-02-02 4:05 ` Jason Wang
2024-02-02 12:24 ` Yunsheng Lin
2024-02-04 1:30 ` Jason Wang
2024-02-04 3:50 ` Yunsheng Lin
2024-02-05 1:43 ` Jason Wang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=2e8606b1-81c2-6f3f-622c-607db5e90253@huawei.com \
--to=linyunsheng@huawei.com \
--cc=akpm@linux-foundation.org \
--cc=alexander.duyck@gmail.com \
--cc=alexanderduyck@fb.com \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=jasowang@redhat.com \
--cc=kuba@kernel.org \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mst@redhat.com \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=virtualization@lists.linux.dev \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).