netdev - Re: [PATCH net-next] sfc: complete the next packet when we receive a timestamp

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <202002222146.hHWMuzDx%lkp@intel.com>
Date:   Sat, 22 Feb 2020 21:36:11 +0800
From:   kbuild test robot <lkp@...el.com>
To:     Tom Zhao <tzhao@...arflare.com>
Cc:     kbuild-all@...ts.01.org, davem@...emloft.net,
        netdev@...r.kernel.org, linux-net-drivers@...arflare.com
Subject: Re: [PATCH net-next] sfc: complete the next packet when we receive a
 timestamp

Hi Tom,

Thank you for the patch! Yet something to improve:

[auto build test ERROR on v5.5]
[cannot apply to net-next/master linus/master sparc-next/master ipvs/master v5.6-rc2 v5.6-rc1 next-20200221]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system. BTW, we also suggest to use '--base' option to specify the
base tree in git format-patch, please see https://stackoverflow.com/a/37406982]

url:    https://github.com/0day-ci/linux/commits/Tom-Zhao/sfc-complete-the-next-packet-when-we-receive-a-timestamp/20200221-083821
base:    d5226fa6dbae0569ee43ecfc08bdcd6770fc4755
config: arm64-allyesconfig (attached as .config)
compiler: aarch64-linux-gcc (GCC) 7.5.0
reproduce:
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # save the attached .config to linux build tree
        GCC_VERSION=7.5.0 make.cross ARCH=arm64 

If you fix the issue, kindly add following tag
Reported-by: kbuild test robot <lkp@...el.com>

All errors (new ones prefixed by >>):

   drivers/net/ethernet/sfc/tx.c: In function 'efx_xmit_done_check_empty':
>> drivers/net/ethernet/sfc/tx.c:845:31: error: implicit declaration of function 'ACCESS_ONCE'; did you mean 'READ_ONCE'? [-Werror=implicit-function-declaration]
      tx_queue->old_write_count = ACCESS_ONCE(tx_queue->write_count);
                                  ^~~~~~~~~~~
                                  READ_ONCE
   drivers/net/ethernet/sfc/tx.c: In function 'efx_xmit_done_single':
>> drivers/net/ethernet/sfc/tx.c:905:19: error: 'struct efx_nic' has no member named 'errors'
       atomic_inc(&efx->errors.spurious_tx);
                      ^~
   cc1: some warnings being treated as errors

vim +845 drivers/net/ethernet/sfc/tx.c

   841	
   842	static void efx_xmit_done_check_empty(struct efx_tx_queue *tx_queue)
   843	{
   844		if ((int)(tx_queue->read_count - tx_queue->old_write_count) >= 0) {
 > 845			tx_queue->old_write_count = ACCESS_ONCE(tx_queue->write_count);
   846			if (tx_queue->read_count == tx_queue->old_write_count) {
   847				smp_mb();
   848				tx_queue->empty_read_count =
   849					tx_queue->read_count | EFX_EMPTY_COUNT_VALID;
   850			}
   851		}
   852	}
   853	
   854	void efx_xmit_done(struct efx_tx_queue *tx_queue, unsigned int index)
   855	{
   856		unsigned fill_level;
   857		struct efx_nic *efx = tx_queue->efx;
   858		struct efx_tx_queue *txq2;
   859		unsigned int pkts_compl = 0, bytes_compl = 0;
   860	
   861		EFX_WARN_ON_ONCE_PARANOID(index > tx_queue->ptr_mask);
   862	
   863		efx_dequeue_buffers(tx_queue, index, &pkts_compl, &bytes_compl);
   864		tx_queue->pkts_compl += pkts_compl;
   865		tx_queue->bytes_compl += bytes_compl;
   866	
   867		if (pkts_compl > 1)
   868			++tx_queue->merge_events;
   869	
   870		/* See if we need to restart the netif queue.  This memory
   871		 * barrier ensures that we write read_count (inside
   872		 * efx_dequeue_buffers()) before reading the queue status.
   873		 */
   874		smp_mb();
   875		if (unlikely(netif_tx_queue_stopped(tx_queue->core_txq)) &&
   876		    likely(efx->port_enabled) &&
   877		    likely(netif_device_present(efx->net_dev))) {
   878			txq2 = efx_tx_queue_partner(tx_queue);
   879			fill_level = max(tx_queue->insert_count - tx_queue->read_count,
   880					 txq2->insert_count - txq2->read_count);
   881			if (fill_level <= efx->txq_wake_thresh)
   882				netif_tx_wake_queue(tx_queue->core_txq);
   883		}
   884	
   885		efx_xmit_done_check_empty(tx_queue);
   886	}
   887	
   888	void efx_xmit_done_single(struct efx_tx_queue *tx_queue)
   889	{
   890		unsigned int pkts_compl = 0, bytes_compl = 0;
   891		unsigned int read_ptr;
   892		bool finished = false;
   893	
   894		read_ptr = tx_queue->read_count & tx_queue->ptr_mask;
   895	
   896		while (!finished) {
   897			struct efx_tx_buffer *buffer = &tx_queue->buffer[read_ptr];
   898	
   899			if (!efx_tx_buffer_in_use(buffer)) {
   900				struct efx_nic *efx = tx_queue->efx;
   901	
   902				netif_err(efx, hw, efx->net_dev,
   903					  "TX queue %d spurious single TX completion\n",
   904					  tx_queue->queue);
 > 905				atomic_inc(&efx->errors.spurious_tx);
   906				efx_schedule_reset(efx, RESET_TYPE_TX_SKIP);
   907				return;
   908			}
   909	
   910			/* Need to check the flag before dequeueing. */
   911			if (buffer->flags & EFX_TX_BUF_SKB)
   912				finished = true;
   913			efx_dequeue_buffer(tx_queue, buffer, &pkts_compl, &bytes_compl);
   914	
   915			++tx_queue->read_count;
   916			read_ptr = tx_queue->read_count & tx_queue->ptr_mask;
   917		}
   918	
   919		tx_queue->pkts_compl += pkts_compl;
   920		tx_queue->bytes_compl += bytes_compl;
   921	
   922		EFX_WARN_ON_PARANOID(pkts_compl != 1);
   923	
   924		efx_xmit_done_check_empty(tx_queue);
   925	}
   926	

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

Download attachment ".config.gz" of type "application/gzip" (69423 bytes)