From mboxrd@z Thu Jan 1 00:00:00 1970 Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.remote-ip=100.103.45.18 ARC-Seal: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; cv=none; t=1794564676; b=2yyjgyETabnpMZdKaKeWLK88pdheDVgGPW0ARPaABhrUjo9rTd5UgT9KPbzzZV94FSMy JW2JHASwn80WakHlz3xS4Ro03PinoPQx/r4B/j05Rt6DU2jzjUBM/7m8kA+9p6L/CYBI8 +uhj/1SypJH92sZvamYWaN7DXdXv+ZrLYMpq2etHFN5wvmQaxBj3lGr3VVCj/r+PIdk0Z lYJGFa+a7seck+YcnVlI5+UBHTfCJ5ZaTEGqpfXJZLss9A1W19I8kzSEEAk2xUZ4/d7rN EbdzO6a3ARn0kLHFby2kFZQCdiY/GNAG6MURfJ1Sg5FYWfFY/1AAQBiroCJLXlm1Ogg== ARC-Message-Signature: i=1; d=subspace.kernel.org; s=arc20261015; a=rsa-sha256; c=relaxed/relaxed; t=1794564676; h=Received:DKIM-Signature:From:To:Cc:Subject:Date:Message-ID:X-Mailer: MIME-Version:Content-Transfer-Encoding; bh=z5RL1Gs6qBWGEalLbjFuinE+mo9K4/ly08HuaMSPvVI=; b=BNoaXYnStqfhKsXDRkFg4nG8l15VOQF0qc3efFi6E8VQhhguxV15H/Ky5uRgtmy+JR+4 AaFhFMxIUAgLpxZP2OaswAz16BoOyQcO3HV0KI1j6ioeDaCKdTyrc0w4jWPqF4Ikcxp26 X7/vNXV8FbU+qyGVYGA6D1WUE2Ju+E3KX8hPZzFSa3i1TeLIykerKVrsiQZr4kdsiRx/I r9Di1WIJkguiGCe95gimQ+Vbqu/AnD81GoSRGZJTsLUkHXqfHp4t9imf4IJXDIRDRcU5Z HsjQIAjwZgm1zmMpNwqPgg3usmKUIUDxwDuApUb8dWLw/SmNuNAimZceLRF9WBeptFA== ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=axwNFDjK; arc=none smtp.remote-ip=100.103.45.18 Received: from smtp.kernel.org (quasi.space.kernel.org [100.103.45.18]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (prime256v1)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id DA4A8606A31C; Fri, 13 Nov 2026 10:11:16 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.a=rsa-sha256 header.s=k20260515 header.b=axwNFDjK Received: by smtp.kernel.org (Postfix) with ESMTPSA id 7DBE51F000C4; Fri, 13 Nov 2026 10:11:13 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel.org; s=k20260515; t=1794564676; bh=z5RL1Gs6qBWGEalLbjFuinE+mo9K4/ly08HuaMSPvVI=; h=From:To:Cc:Subject:Date; b=axwNFDjKra/n9fGQ6DitqiUJGJUTfTwNUpSzMjBM7UCepgbrgCqOsJYiF7/GeIVxy UZqGu0peJSmnZ6+Lw9UIWBt89YUmE3uySjtUdbt8rBY9ZELWMnYjo9Qkkznnq5jjL0 JMXi1uyP6kjxGwfWfVb+pdfrWazOF1Bt7C1X2YNM1OcBn2Y4jImDZCk0ihBilqsUgA lFqBJYenW5MRpc4fAmhG660z9GNs6L2OuuHqRtPBsVsagkBgTH+c/QsYn43tPFzoir 9NqcZ7Bhhq1Vq+i4hDEAeNCH/W945br1IDrcqI7ob8bmERm9F/u1vnHllz9qPEEbPp MXctHNnE0BuuQ== From: Arnd Bergmann To: "David S. Miller" , Eric Dumazet , Jakub Kicinski , Paolo Abeni , Aleksandr Loktionov , Stanislav Fomichev , Mina Almasry Cc: Arnd Bergmann , Simon Horman , Willem de Bruijn , Josef Bacik , Wang Zhan , Daniel Borkmann , netdev@vger.kernel.org, linux-kernel@vger.kernel.org Subject: [PATCH] net: kunit: move struct net_devmem_dmabuf_binding off stack Date: Fri, 13 Nov 2026 11:11:00 +0100 Message-ID: <20261113101109.3227815-1-arnd@kernel.org> X-Mailer: git-send-email 2.53.0 X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Post: List-Help: List-Subscribe: List-Unsubscribe: List-Owner: Precedence: list MIME-Version: 1.0 Content-Transfer-Encoding: 8bit From: Arnd Bergmann The net_devmem_dmabuf_binding structure is really way too big to live on the stack, and causes warnings on my randconfig build machine with a reasonable per-function limit: net/core/net_test.c: In function 'netmem_ref_test_tcp_devmem_recv': net/core/net_test.c:1720:1: error: the frame size of 1744 bytes is larger than 1536 bytes [-Werror=frame-larger-than=] 1720 | } | ^ net/core/net_test.c: In function 'netmem_ref_test_skb_segment': net/core/net_test.c:1477:1: error: the frame size of 1584 bytes is larger than 1536 bytes [-Werror=frame-larger-than=] Since the binding object is never used concurrently in multiple tests, avoid this by just having a global copy of it and reinitializing it for every user. Fixes: 2285346afc99 ("net: kunit: test netmem, page_pool, and skb frag refcounting") Signed-off-by: Arnd Bergmann --- I don't particularly like this workaround, as global variables are always hacky like this. If anyone has a better idea to avoid the last stack structure, let's do that instead. Without the patch, I run into this every few randconfig builds, which did not happen on linux-next before the merge window. Signed-off-by: Arnd Bergmann --- net/core/net_test.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/net/core/net_test.c b/net/core/net_test.c index 976d14896251..19f422300d9c 100644 --- a/net/core/net_test.c +++ b/net/core/net_test.c @@ -984,13 +984,14 @@ static void netmem_ref_test_pp_page(struct kunit *test) #endif #if defined(CONFIG_PAGE_POOL) && defined(CONFIG_NET_DEVMEM) +static struct net_devmem_dmabuf_binding binding; + static void dummy_percpu_ref_release(struct percpu_ref *ref) { } static void netmem_ref_test_net_iov(struct kunit *test) { - struct net_devmem_dmabuf_binding binding = {}; struct page_pool_params pp_params = { .order = 0, .pool_size = 4, @@ -1003,6 +1004,7 @@ static void netmem_ref_test_net_iov(struct kunit *test) long base_ref; int err; + memset(&binding, 0, sizeof(binding)); err = percpu_ref_init(&binding.ref, dummy_percpu_ref_release, PERCPU_REF_INIT_ATOMIC, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); @@ -1145,7 +1147,6 @@ static void netmem_ref_test_release_data(struct kunit *test) .nid = NUMA_NO_NODE, }; struct sk_buff *skb, *clone1, *clone2, *ext; - struct net_devmem_dmabuf_binding binding = {}; struct net_iov niov = {}; struct page_pool *pool; struct page *page; @@ -1154,6 +1155,7 @@ static void netmem_ref_test_release_data(struct kunit *test) int base_ref; int i; + memset(&binding, 0, sizeof(binding)); pool = page_pool_create(&pp_params); KUNIT_ASSERT_FALSE(test, IS_ERR(pool)); page = page_pool_dev_alloc_pages(pool); @@ -1335,7 +1337,6 @@ static void netmem_ref_test_skb_segment(struct kunit *test) .flags = SKBFL_ZEROCOPY_ENABLE, }; struct sk_buff *head, *list, *segs, *seg; - struct net_devmem_dmabuf_binding binding = {}; struct net_iov niov = {}; struct page *pp_page, *reg_page; netmem_ref pp_nm, niov_nm; @@ -1343,6 +1344,7 @@ static void netmem_ref_test_skb_segment(struct kunit *test) long base_pp_ref, base_iov_ref; int base_pp_page_ref, base_reg_ref; + memset(&binding, 0, sizeof(binding)); pool = page_pool_create(&pp_params); KUNIT_ASSERT_FALSE(test, IS_ERR(pool)); pp_page = page_pool_dev_alloc_pages(pool); @@ -1634,7 +1636,6 @@ static void netmem_ref_test_try_coalesce(struct kunit *test) .nid = NUMA_NO_NODE, }; struct sk_buff *to, *from, *clone; - struct net_devmem_dmabuf_binding binding = {}; struct net_iov niov = {}; struct page *pp_page, *reg_page; netmem_ref pp_nm, niov_nm; @@ -1644,6 +1645,7 @@ static void netmem_ref_test_try_coalesce(struct kunit *test) bool fragstolen; int delta; + memset(&binding, 0, sizeof(binding)); pool = page_pool_create(&pp_params); KUNIT_ASSERT_FALSE(test, IS_ERR(pool)); pp_page = page_pool_dev_alloc_pages(pool); @@ -1766,7 +1768,6 @@ static void netmem_ref_test_tcp_devmem_recv(struct kunit *test) .pool_size = 4, .nid = NUMA_NO_NODE, }; - struct net_devmem_dmabuf_binding binding = {}; struct msghdr msg = {}; struct dmabuf_token token; struct dmabuf_cmsg *cmsg; @@ -1780,6 +1781,7 @@ static void netmem_ref_test_tcp_devmem_recv(struct kunit *test) long base_iov_ref; int ret; + memset(&binding, 0, sizeof(binding)); KUNIT_ASSERT_EQ(test, sock_create_kern(&init_net, AF_INET, SOCK_STREAM, IPPROTO_TCP, &sock), -- 2.53.0