Re: [PATCH net-next 1/6] qed: Add doorbell overflow recovery mechanism

From: kbuild test robot <lkp@intel.com>
To: Ariel Elior <Ariel.Elior@cavium.com>
Cc: kbuild-all@01.org, davem@davemloft.net, netdev@vger.kernel.org,
	Ariel Elior <Ariel.Elior@cavium.com>,
	Michal Kalderon <Michal.Kalderon@cavium.com>,
	Tomer Tayar <Tomer.Tayar@cavium.com>
Subject: Re: [PATCH net-next 1/6] qed: Add doorbell overflow recovery mechanism
Date: Tue, 23 Oct 2018 13:15:10 +0800	[thread overview]
Message-ID: <201810231303.wGAsmMkZ%fengguang.wu@intel.com> (raw)
In-Reply-To: <20181022122743.20384-2-Ariel.Elior@cavium.com>

[-- Attachment #1: Type: text/plain, Size: 2837 bytes --]

Hi Ariel,

I love your patch! Yet something to improve:

[auto build test ERROR on net-next/master]

url:    https://github.com/0day-ci/linux/commits/Ariel-Elior/qed-Doorbell-overflow-recovery/20181022-212749
config: i386-allyesconfig (attached as .config)
compiler: gcc-7 (Debian 7.3.0-1) 7.3.0
reproduce:
        # save the attached .config to linux build tree
        make ARCH=i386 

All errors (new ones prefixed by >>):

   drivers/net/ethernet/qlogic/qed/qed_dev.c: In function 'qed_db_recovery_ring':
>> drivers/net/ethernet/qlogic/qed/qed_dev.c:342:4: error: implicit declaration of function 'writeq'; did you mean 'writel'? [-Werror=implicit-function-declaration]
       DIRECT_REG_WR64(db_entry->db_addr,
       ^~~~~~
       writel
   cc1: some warnings being treated as errors

vim +342 drivers/net/ethernet/qlogic/qed/qed_dev.c

   300	
   301	/* Ring the doorbell of a single doorbell recovery entry */
   302	static void qed_db_recovery_ring(struct qed_hwfn *p_hwfn,
   303					 struct qed_db_recovery_entry *db_entry,
   304					 enum qed_db_rec_exec db_exec)
   305	{
   306		if (db_exec != DB_REC_ONCE) {
   307			/* Print according to width */
   308			if (db_entry->db_width == DB_REC_WIDTH_32B) {
   309				DP_VERBOSE(p_hwfn, QED_MSG_SPQ,
   310					   "%s doorbell address %p data %x\n",
   311					   db_exec == DB_REC_DRY_RUN ?
   312					   "would have rung" : "ringing",
   313					   db_entry->db_addr,
   314					   *(u32 *)db_entry->db_data);
   315			} else {
   316				DP_VERBOSE(p_hwfn, QED_MSG_SPQ,
   317					   "%s doorbell address %p data %llx\n",
   318					   db_exec == DB_REC_DRY_RUN ?
   319					   "would have rung" : "ringing",
   320					   db_entry->db_addr,
   321					   *(u64 *)(db_entry->db_data));
   322			}
   323		}
   324	
   325		/* Sanity */
   326		if (!qed_db_rec_sanity(p_hwfn->cdev, db_entry->db_addr,
   327				       db_entry->db_data))
   328			return;
   329	
   330		/* Flush the write combined buffer. Since there are multiple doorbelling
   331		 * entities using the same address, if we don't flush, a transaction
   332		 * could be lost.
   333		 */
   334		wmb();
   335	
   336		/* Ring the doorbell */
   337		if (db_exec == DB_REC_REAL_DEAL || db_exec == DB_REC_ONCE) {
   338			if (db_entry->db_width == DB_REC_WIDTH_32B)
   339				DIRECT_REG_WR(db_entry->db_addr,
   340					      *(u32 *)(db_entry->db_data));
   341			else
 > 342				DIRECT_REG_WR64(db_entry->db_addr,
   343						*(u64 *)(db_entry->db_data));
   344		}
   345	
   346		/* Flush the write combined buffer. Next doorbell may come from a
   347		 * different entity to the same address...
   348		 */
   349		wmb();
   350	}
   351	

---
0-DAY kernel test infrastructure                Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all                   Intel Corporation

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 64635 bytes --]