[PATCH 0/7] fix unused but set variables

[PATCH 0/7] fix unused but set variables

[Conor Walsh]

When DPDK is compiled with clang 13, several warnings for unused but set
variables are present.

This patchset fixes these warnings, which were present within:
 - app/test
 - drivers/bus/fslmc
 - drivers/common/qat
 - drivers/event/sw
 - drivers/net
 - examples/performance-thread
 - lib/power

I split this patchset into 7 patches grouped broadly by subdir, as this
set touches code from a lot of different authors I thought it would make
it easier to review.

Bugzilla ID: 881

Conor Walsh (7):
  app/test: fix unused but set variables
  drivers/bus/fslmc: fix unused but set variables
  drivers/common/qat: fix unused but set variables
  drivers/event/sw: fix unused but set variables
  drivers/net: fix unused but set variables
  examples/performance-thread: fix unused but set variables
  lib/power: fix unused but set variables

 app/test/test_distributor_perf.c              |  3 ---
 app/test/test_thash_perf.c                    |  2 ++
 drivers/bus/fslmc/fslmc_bus.c                 |  2 --
 drivers/common/qat/qat_pf2vf.c                |  2 --
 drivers/event/sw/sw_evdev.c                   |  2 --
 drivers/net/bnxt/bnxt_rxr.c                   | 22 -------------------
 drivers/net/bnxt/tf_ulp/ulp_utils.c           |  3 ---
 drivers/net/hinic/base/hinic_pmd_mgmt.c       |  6 -----
 drivers/net/liquidio/lio_rxtx.c               |  2 --
 drivers/net/nfp/nfp_cpp_bridge.c              |  6 ++---
 drivers/net/octeontx/octeontx_rxtx.h          |  3 +--
 drivers/net/qede/base/ecore_vf.c              |  5 +----
 drivers/net/txgbe/txgbe_ethdev_vf.c           |  2 ++
 drivers/net/vmxnet3/vmxnet3_rxtx.c            |  8 +++++++
 .../performance-thread/l3fwd-thread/main.c    |  3 ---
 lib/power/rte_power_empty_poll.c              |  3 +--
 16 files changed, 17 insertions(+), 57 deletions(-)

-- 
2.25.1

[PATCH 1/7] app/test: fix unused but set variables

[Conor Walsh]

This patch fixes unused but set variables within the following test apps:
 - test distributor perf
 - test thash perf

Bugzilla ID: 881

Fixes: c0de0eb82e40 ("distributor: switch over to new API")
Fixes: 239fffe0402e ("test/thash: add performance tests for Toeplitz hash")

Signed-off-by: Conor Walsh <conor.walsh@intel.com>
Reported-by: Liang Longfeng <longfengx.liang@intel.com>
---
 app/test/test_distributor_perf.c | 3 ---
 app/test/test_thash_perf.c       | 2 ++
 2 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/app/test/test_distributor_perf.c b/app/test/test_distributor_perf.c
index fdbeae6d2f..92e330f194 100644
--- a/app/test/test_distributor_perf.c
+++ b/app/test/test_distributor_perf.c
@@ -108,7 +108,6 @@ static int
 handle_work(void *arg)
 {
 	struct rte_distributor *d = arg;
-	unsigned int count = 0;
 	unsigned int num = 0;
 	int i;
 	unsigned int id = __atomic_fetch_add(&worker_idx, 1, __ATOMIC_RELAXED);
@@ -120,11 +119,9 @@ handle_work(void *arg)
 	num = rte_distributor_get_pkt(d, id, buf, buf, num);
 	while (!quit) {
 		worker_stats[id].handled_packets += num;
-		count += num;
 		num = rte_distributor_get_pkt(d, id, buf, buf, num);
 	}
 	worker_stats[id].handled_packets += num;
-	count += num;
 	rte_distributor_return_pkt(d, id, buf, num);
 	return 0;
 }
diff --git a/app/test/test_thash_perf.c b/app/test/test_thash_perf.c
index 7aa9360120..c95785d8bb 100644
--- a/app/test/test_thash_perf.c
+++ b/app/test/test_thash_perf.c
@@ -61,6 +61,8 @@ run_rss_calc(uint32_t *tuples[BATCH_SZ], enum test_rss_type type, int len,
 	}
 	end_tsc = rte_rdtsc_precise();
 
+	RTE_SET_USED(hash);
+
 	return end_tsc - start_tsc;
 }
 
-- 
2.25.1

Re: [PATCH 1/7] app/test: fix unused but set variables

[Medvedkin, Vladimir]

On 12/11/2021 15:01, Conor Walsh wrote:
> This patch fixes unused but set variables within the following test apps:
>   - test distributor perf
>   - test thash perf
> 
> Bugzilla ID: 881
> 
> Fixes: c0de0eb82e40 ("distributor: switch over to new API")
> Fixes: 239fffe0402e ("test/thash: add performance tests for Toeplitz hash")
> 
> Signed-off-by: Conor Walsh <conor.walsh@intel.com>
> Reported-by: Liang Longfeng <longfengx.liang@intel.com>
> ---
>   app/test/test_distributor_perf.c | 3 ---
>   app/test/test_thash_perf.c       | 2 ++
>   2 files changed, 2 insertions(+), 3 deletions(-)
> 
> diff --git a/app/test/test_distributor_perf.c b/app/test/test_distributor_perf.c
> index fdbeae6d2f..92e330f194 100644
> --- a/app/test/test_distributor_perf.c
> +++ b/app/test/test_distributor_perf.c
> @@ -108,7 +108,6 @@ static int
>   handle_work(void *arg)
>   {
>   	struct rte_distributor *d = arg;
> -	unsigned int count = 0;
>   	unsigned int num = 0;
>   	int i;
>   	unsigned int id = __atomic_fetch_add(&worker_idx, 1, __ATOMIC_RELAXED);
> @@ -120,11 +119,9 @@ handle_work(void *arg)
>   	num = rte_distributor_get_pkt(d, id, buf, buf, num);
>   	while (!quit) {
>   		worker_stats[id].handled_packets += num;
> -		count += num;
>   		num = rte_distributor_get_pkt(d, id, buf, buf, num);
>   	}
>   	worker_stats[id].handled_packets += num;
> -	count += num;
>   	rte_distributor_return_pkt(d, id, buf, num);
>   	return 0;
>   }
> diff --git a/app/test/test_thash_perf.c b/app/test/test_thash_perf.c
> index 7aa9360120..c95785d8bb 100644
> --- a/app/test/test_thash_perf.c
> +++ b/app/test/test_thash_perf.c
> @@ -61,6 +61,8 @@ run_rss_calc(uint32_t *tuples[BATCH_SZ], enum test_rss_type type, int len,
>   	}
>   	end_tsc = rte_rdtsc_precise();
>   
> +	RTE_SET_USED(hash);
> +
>   	return end_tsc - start_tsc;
>   }
>   
> 

Acked-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>

-- 
Regards,
Vladimir

[PATCH 2/7] drivers/bus/fslmc: fix unused but set variables

[Conor Walsh]

This patch removes the device_count variable from the rte_fslmc_scan
function within the fslmc bus driver as it is an unused but set variable.

Bugzilla ID: 881

Fixes: 828d51d8fc3e ("bus/fslmc: refactor scan and probe functions")

Signed-off-by: Conor Walsh <conor.walsh@intel.com>
Reported-by: Liang Longfeng <longfengx.liang@intel.com>
---
 drivers/bus/fslmc/fslmc_bus.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/bus/fslmc/fslmc_bus.c b/drivers/bus/fslmc/fslmc_bus.c
index ac3cb4aa5a..a0ef24cdc8 100644
--- a/drivers/bus/fslmc/fslmc_bus.c
+++ b/drivers/bus/fslmc/fslmc_bus.c
@@ -315,7 +315,6 @@ static int
 rte_fslmc_scan(void)
 {
 	int ret;
-	int device_count = 0;
 	char fslmc_dirpath[PATH_MAX];
 	DIR *dir;
 	struct dirent *entry;
@@ -349,7 +348,6 @@ rte_fslmc_scan(void)
 			/* Error in parsing directory - exit gracefully */
 			goto scan_fail_cleanup;
 		}
-		device_count += 1;
 	}
 
 	closedir(dir);
-- 
2.25.1

[PATCH 3/7] drivers/common/qat: fix unused but set variables

[Conor Walsh]

This patch removes the us variable from the qat_pf2vf_exch_msg function
within the PF to VF section of the QAT driver as it is an unused but set
variable.

Bugzilla ID: 881

Fixes: b17d16fb47b4 ("common/qat: add PF to VF communication")

Signed-off-by: Conor Walsh <conor.walsh@intel.com>
Reported-by: Liang Longfeng <longfengx.liang@intel.com>
---
 drivers/common/qat/qat_pf2vf.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/common/qat/qat_pf2vf.c b/drivers/common/qat/qat_pf2vf.c
index 6327311199..4e9ffc72fc 100644
--- a/drivers/common/qat/qat_pf2vf.c
+++ b/drivers/common/qat/qat_pf2vf.c
@@ -54,14 +54,12 @@ int qat_pf2vf_exch_msg(struct qat_pci_device *qat_dev,
 		msg |= ADF_PFVF_INT | ADF_PFVF_MSGORIGIN_SYSTEM;
 
 		ADF_CSR_WR(pmisc_bar_addr, vf_csr_off, msg);
-		int us = 0;
 		/*
 		 * Wait for confirmation from remote that it received
 		 * the message
 		 */
 		do {
 			rte_delay_us_sleep(5);
-			us += 5;
 			val = ADF_CSR_RD(pmisc_bar_addr, vf_csr_off);
 		} while ((val & ADF_PFVF_INT) &&
 			(++count < ADF_IOV_MSG_ACK_MAX_RETRY));
-- 
2.25.1

[PATCH 4/7] drivers/event/sw: fix unused but set variables

[Conor Walsh]

This patch removes the inflights variable from the sw_dump function
within the software section of the event driver as it is an unused but
set variable.

Bugzilla ID: 881

Fixes: c66baa68e453 ("event/sw: add dump function for easier debugging")

Signed-off-by: Conor Walsh <conor.walsh@intel.com>
Reported-by: Liang Longfeng <longfengx.liang@intel.com>
---
 drivers/event/sw/sw_evdev.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/event/sw/sw_evdev.c b/drivers/event/sw/sw_evdev.c
index 070a4802e9..fa72ceb949 100644
--- a/drivers/event/sw/sw_evdev.c
+++ b/drivers/event/sw/sw_evdev.c
@@ -711,7 +711,6 @@ sw_dump(struct rte_eventdev *dev, FILE *f)
 			continue;
 		}
 		int affinities_per_port[SW_PORTS_MAX] = {0};
-		uint32_t inflights = 0;
 
 		fprintf(f, "  Queue %d (%s)\n", i, q_type_strings[qid->type]);
 		fprintf(f, "\trx   %"PRIu64"\tdrop %"PRIu64"\ttx   %"PRIu64"\n",
@@ -732,7 +731,6 @@ sw_dump(struct rte_eventdev *dev, FILE *f)
 		for (flow = 0; flow < RTE_DIM(qid->fids); flow++)
 			if (qid->fids[flow].cq != -1) {
 				affinities_per_port[qid->fids[flow].cq]++;
-				inflights += qid->fids[flow].pcount;
 			}
 
 		uint32_t port;
-- 
2.25.1

RE: [PATCH 4/7] drivers/event/sw: fix unused but set variables

[Van Haaren, Harry]

> -----Original Message-----
> From: Walsh, Conor <conor.walsh@intel.com>
> Sent: Friday, November 12, 2021 2:01 PM
> To: Yigit, Ferruh <ferruh.yigit@intel.com>; david.marchand@redhat.com; Liang,
> LongfengX <longfengx.liang@intel.com>; dev@dpdk.org
> Cc: Hunt, David <david.hunt@intel.com>; Wang, Yipeng1
> <yipeng1.wang@intel.com>; Gobriel, Sameh <sameh.gobriel@intel.com>;
> Richardson, Bruce <bruce.richardson@intel.com>; Medvedkin, Vladimir
> <vladimir.medvedkin@intel.com>; hemant.agrawal@nxp.com;
> sachin.saxena@oss.nxp.com; Trahe, Fiona <fiona.trahe@intel.com>; Griffin, John
> <john.griffin@intel.com>; Jain, Deepak K <deepak.k.jain@intel.com>; Van
> Haaren, Harry <harry.van.haaren@intel.com>; ajit.khaparde@broadcom.com;
> somnath.kotur@broadcom.com; xuanziyang2@huawei.com;
> cloud.wangxiaoyun@huawei.com; zhouguoyang@huawei.com;
> sthotton@marvell.com; srinivasan@marvell.com; heinrich.kuhn@corigine.com;
> hkalra@marvell.com; rmody@marvell.com; dsinghrawat@marvell.com;
> jiawenwu@trustnetic.com; Wang, Yong <yongwang@vmware.com>; Mcnamara,
> John <john.mcnamara@intel.com>; shreyansh.jain@nxp.com;
> santosh.shukla@caviumnetworks.com; gakhil@marvell.com; Zhang, Roy Fan
> <roy.fan.zhang@intel.com>; Kusztal, ArkadiuszX
> <arkadiuszx.kusztal@intel.com>; michael.baucom@broadcom.com;
> venkatkumar.duvvuru@broadcom.com; lance.richardson@broadcom.com;
> kishore.padmanabha@broadcom.com; jerin.jacob@caviumnetworks.com;
> derek.chickles@caviumnetworks.com; ssrinivasan@caviumnetworks.com;
> mjatharakonda@oneconvergence.com; simon.horman@corigine.com;
> david.george@sophos.com; sony.chacko@qlogic.com; harish.patil@qlogic.com;
> stephen@networkplumber.org; tomaszx.kulasek@intel.com;
> ian.betts@intel.com; liang.j.ma@intel.com; lei.a.yao@intel.com; Walsh, Conor
> <conor.walsh@intel.com>
> Subject: [PATCH 4/7] drivers/event/sw: fix unused but set variables

-CC to just to people involved & list to avoid spamming too wide.

> This patch removes the inflights variable from the sw_dump function
> within the software section of the event driver as it is an unused but
> set variable.
> 
> Bugzilla ID: 881
> 
> Fixes: c66baa68e453 ("event/sw: add dump function for easier debugging")
> 
> Signed-off-by: Conor Walsh <conor.walsh@intel.com>
> Reported-by: Liang Longfeng <longfengx.liang@intel.com>

Acked by: Harry van Haaren <harry.van.haaren@intel.com>

[PATCH 5/7] drivers/net: fix unused but set variables

[Conor Walsh]

This patch fixes unused but set variables in the following net drivers:
 - bnxt
 - hinic
 - liquidio
 - nfp
 - octeontx
 - qede
 - txgbe
 - vmxnet3

This patch also fixes an unused function warning for vmxnet3_unmap_pkt
from within vmxnet3

Bugzilla ID: 881

Fixes: b87abb2e55cb ("net/bnxt: support marking packet")
Fixes: 88badb3aef94 ("net/bnxt: add helper functions for blob/regfile ops")
Fixes: a4957d87e05a ("net/hinic/base: add mgmt module")
Fixes: 5ee7640f95a0 ("net/liquidio: add API to flush IQ")
Fixes: 19af5a38f7b6 ("net/nfp: move CPP bridge to separate file")
Fixes: 9eb5cb3b11cc ("net/octeontx: fix access to indirect buffers")
Fixes: 86a2265e59d7 ("qede: add SRIOV support")
Fixes: b7311360fb67 ("net/txgbe: support VF interrupt")
Fixes: c3ecdbb376da ("vmxnet3: support TSO")

Signed-off-by: Conor Walsh <conor.walsh@intel.com>
Reported-by: Liang Longfeng <longfengx.liang@intel.com>
---
 drivers/net/bnxt/bnxt_rxr.c             | 22 ----------------------
 drivers/net/bnxt/tf_ulp/ulp_utils.c     |  3 ---
 drivers/net/hinic/base/hinic_pmd_mgmt.c |  6 ------
 drivers/net/liquidio/lio_rxtx.c         |  2 --
 drivers/net/nfp/nfp_cpp_bridge.c        |  6 ++----
 drivers/net/octeontx/octeontx_rxtx.h    |  3 +--
 drivers/net/qede/base/ecore_vf.c        |  5 +----
 drivers/net/txgbe/txgbe_ethdev_vf.c     |  2 ++
 drivers/net/vmxnet3/vmxnet3_rxtx.c      |  8 ++++++++
 9 files changed, 14 insertions(+), 43 deletions(-)

diff --git a/drivers/net/bnxt/bnxt_rxr.c b/drivers/net/bnxt/bnxt_rxr.c
index 8bc8ddc353..157297bc64 100644
--- a/drivers/net/bnxt/bnxt_rxr.c
+++ b/drivers/net/bnxt/bnxt_rxr.c
@@ -823,9 +823,6 @@ void bnxt_set_mark_in_mbuf(struct bnxt *bp,
 			   struct rte_mbuf *mbuf)
 {
 	uint32_t cfa_code = 0;
-	uint8_t meta_fmt = 0;
-	uint16_t flags2 = 0;
-	uint32_t meta =  0;
 
 	cfa_code = rte_le_to_cpu_16(rxcmp1->cfa_code);
 	if (!cfa_code)
@@ -834,25 +831,6 @@ void bnxt_set_mark_in_mbuf(struct bnxt *bp,
 	if (cfa_code && !bp->mark_table[cfa_code].valid)
 		return;
 
-	flags2 = rte_le_to_cpu_16(rxcmp1->flags2);
-	meta = rte_le_to_cpu_32(rxcmp1->metadata);
-	if (meta) {
-		meta >>= BNXT_RX_META_CFA_CODE_SHIFT;
-
-		/* The flags field holds extra bits of info from [6:4]
-		 * which indicate if the flow is in TCAM or EM or EEM
-		 */
-		meta_fmt = (flags2 & BNXT_CFA_META_FMT_MASK) >>
-			   BNXT_CFA_META_FMT_SHFT;
-
-		/* meta_fmt == 4 => 'b100 => 'b10x => EM.
-		 * meta_fmt == 5 => 'b101 => 'b10x => EM + VLAN
-		 * meta_fmt == 6 => 'b110 => 'b11x => EEM
-		 * meta_fmt == 7 => 'b111 => 'b11x => EEM + VLAN.
-		 */
-		meta_fmt >>= BNXT_CFA_META_FMT_EM_EEM_SHFT;
-	}
-
 	mbuf->hash.fdir.hi = bp->mark_table[cfa_code].mark_id;
 	mbuf->ol_flags |= RTE_MBUF_F_RX_FDIR | RTE_MBUF_F_RX_FDIR_ID;
 }
diff --git a/drivers/net/bnxt/tf_ulp/ulp_utils.c b/drivers/net/bnxt/tf_ulp/ulp_utils.c
index df3afaa6fd..c60d81d14a 100644
--- a/drivers/net/bnxt/tf_ulp/ulp_utils.c
+++ b/drivers/net/bnxt/tf_ulp/ulp_utils.c
@@ -200,7 +200,6 @@ ulp_bs_push_msb(uint8_t *bs, uint16_t pos, uint8_t len, uint8_t *val)
 {
 	int i;
 	int cnt = (len + 7) / 8;
-	int tlen = len;
 
 	/* Handle any remainder bits */
 	int tmp = len % 8;
@@ -211,12 +210,10 @@ ulp_bs_push_msb(uint8_t *bs, uint16_t pos, uint8_t len, uint8_t *val)
 	ulp_bs_put_msb(bs, pos, tmp, val[0]);
 
 	pos += tmp;
-	tlen -= tmp;
 
 	for (i = 1; i < cnt; i++) {
 		ulp_bs_put_msb(bs, pos, 8, val[i]);
 		pos += 8;
-		tlen -= 8;
 	}
 
 	return len;
diff --git a/drivers/net/hinic/base/hinic_pmd_mgmt.c b/drivers/net/hinic/base/hinic_pmd_mgmt.c
index 9b399502de..6ecca407a1 100644
--- a/drivers/net/hinic/base/hinic_pmd_mgmt.c
+++ b/drivers/net/hinic/base/hinic_pmd_mgmt.c
@@ -133,16 +133,12 @@ static void prepare_header(struct hinic_msg_pf_to_mgmt *pf_to_mgmt,
 static void prepare_mgmt_cmd(u8 *mgmt_cmd, u64 *header, void *msg,
 			     int msg_len)
 {
-	u32 cmd_buf_max = MAX_PF_MGMT_BUF_SIZE;
-
 	memset(mgmt_cmd, 0, MGMT_MSG_RSVD_FOR_DEV);
 
 	mgmt_cmd += MGMT_MSG_RSVD_FOR_DEV;
-	cmd_buf_max -= MGMT_MSG_RSVD_FOR_DEV;
 	memcpy(mgmt_cmd, header, sizeof(*header));
 
 	mgmt_cmd += sizeof(*header);
-	cmd_buf_max -= sizeof(*header);
 	memcpy(mgmt_cmd, msg, msg_len);
 }
 
@@ -615,7 +611,6 @@ static int recv_mgmt_msg_handler(struct hinic_msg_pf_to_mgmt *pf_to_mgmt,
 	void *msg_body = header + sizeof(msg_header);
 	u8 *dest_msg;
 	u8 seq_id, seq_len;
-	u32 msg_buf_max = MAX_PF_MGMT_BUF_SIZE;
 	u8 front_id;
 	u16 msg_id;
 
@@ -635,7 +630,6 @@ static int recv_mgmt_msg_handler(struct hinic_msg_pf_to_mgmt *pf_to_mgmt,
 	}
 
 	dest_msg = (u8 *)recv_msg->msg + seq_id * HINIC_MSG_SEG_LEN;
-	msg_buf_max -= seq_id * HINIC_MSG_SEG_LEN;
 	memcpy(dest_msg, msg_body, seq_len);
 
 	if (!HINIC_MSG_HEADER_GET(msg_header, LAST))
diff --git a/drivers/net/liquidio/lio_rxtx.c b/drivers/net/liquidio/lio_rxtx.c
index ef127f26c4..e09798ddd7 100644
--- a/drivers/net/liquidio/lio_rxtx.c
+++ b/drivers/net/liquidio/lio_rxtx.c
@@ -1050,7 +1050,6 @@ lio_update_read_index(struct lio_instr_queue *iq)
 int
 lio_flush_iq(struct lio_device *lio_dev, struct lio_instr_queue *iq)
 {
-	uint32_t tot_inst_processed = 0;
 	uint32_t inst_processed = 0;
 	int tx_done = 1;
 
@@ -1073,7 +1072,6 @@ lio_flush_iq(struct lio_device *lio_dev, struct lio_instr_queue *iq)
 			iq->stats.instr_processed += inst_processed;
 		}
 
-		tot_inst_processed += inst_processed;
 		inst_processed = 0;
 
 	} while (1);
diff --git a/drivers/net/nfp/nfp_cpp_bridge.c b/drivers/net/nfp/nfp_cpp_bridge.c
index 74a0eacb3f..0922ea9cf9 100644
--- a/drivers/net/nfp/nfp_cpp_bridge.c
+++ b/drivers/net/nfp/nfp_cpp_bridge.c
@@ -58,7 +58,7 @@ nfp_cpp_bridge_serve_write(int sockfd, struct nfp_cpp *cpp)
 	off_t offset, nfp_offset;
 	uint32_t cpp_id, pos, len;
 	uint32_t tmpbuf[16];
-	size_t count, curlen, totlen = 0;
+	size_t count, curlen;
 	int err = 0;
 
 	PMD_CPP_LOG(DEBUG, "%s: offset size %zu, count_size: %zu\n", __func__,
@@ -135,7 +135,6 @@ nfp_cpp_bridge_serve_write(int sockfd, struct nfp_cpp *cpp)
 		}
 
 		nfp_offset += pos;
-		totlen += pos;
 		nfp_cpp_area_release(area);
 		nfp_cpp_area_free(area);
 
@@ -160,7 +159,7 @@ nfp_cpp_bridge_serve_read(int sockfd, struct nfp_cpp *cpp)
 	off_t offset, nfp_offset;
 	uint32_t cpp_id, pos, len;
 	uint32_t tmpbuf[16];
-	size_t count, curlen, totlen = 0;
+	size_t count, curlen;
 	int err = 0;
 
 	PMD_CPP_LOG(DEBUG, "%s: offset size %zu, count_size: %zu\n", __func__,
@@ -236,7 +235,6 @@ nfp_cpp_bridge_serve_read(int sockfd, struct nfp_cpp *cpp)
 		}
 
 		nfp_offset += pos;
-		totlen += pos;
 		nfp_cpp_area_release(area);
 		nfp_cpp_area_free(area);
 
diff --git a/drivers/net/octeontx/octeontx_rxtx.h b/drivers/net/octeontx/octeontx_rxtx.h
index 56f11d3dd5..b63a5edfb3 100644
--- a/drivers/net/octeontx/octeontx_rxtx.h
+++ b/drivers/net/octeontx/octeontx_rxtx.h
@@ -365,7 +365,7 @@ __octeontx_xmit_mseg_prepare(struct rte_mbuf *tx_pkt, uint64_t *cmd_buf,
 			const uint16_t flag)
 {
 	uint16_t nb_segs, nb_desc = 0;
-	uint16_t gaura_id, len = 0;
+	uint16_t gaura_id;
 	struct rte_mbuf *m_next = NULL, *m_tofree;
 	rte_iova_t iova;
 	uint16_t data_len;
@@ -425,7 +425,6 @@ __octeontx_xmit_mseg_prepare(struct rte_mbuf *tx_pkt, uint64_t *cmd_buf,
 		cmd_buf[nb_desc++] = iova;
 
 		nb_segs--;
-		len += data_len;
 		tx_pkt = m_next;
 	} while (nb_segs);
 
diff --git a/drivers/net/qede/base/ecore_vf.c b/drivers/net/qede/base/ecore_vf.c
index db03bc494f..a36ae47c2b 100644
--- a/drivers/net/qede/base/ecore_vf.c
+++ b/drivers/net/qede/base/ecore_vf.c
@@ -73,7 +73,7 @@ static void ecore_vf_pf_req_end(struct ecore_hwfn *p_hwfn,
 #endif
 static enum _ecore_status_t
 ecore_send_msg2pf(struct ecore_hwfn *p_hwfn,
-		  u8 *done, u32 resp_size)
+		  u8 *done, __rte_unused u32 resp_size)
 {
 	union vfpf_tlvs *p_req = p_hwfn->vf_iov_info->vf2pf_request;
 	struct ustorm_trigger_vf_zone trigger;
@@ -86,9 +86,6 @@ ecore_send_msg2pf(struct ecore_hwfn *p_hwfn,
 	/* output tlvs list */
 	ecore_dp_tlv_list(p_hwfn, p_req);
 
-	/* need to add the END TLV to the message size */
-	resp_size += sizeof(struct channel_list_end_tlv);
-
 	/* Send TLVs over HW channel */
 	OSAL_MEMSET(&trigger, 0, sizeof(struct ustorm_trigger_vf_zone));
 	trigger.vf_pf_msg_valid = 1;
diff --git a/drivers/net/txgbe/txgbe_ethdev_vf.c b/drivers/net/txgbe/txgbe_ethdev_vf.c
index 67ae69dec3..be460ab733 100644
--- a/drivers/net/txgbe/txgbe_ethdev_vf.c
+++ b/drivers/net/txgbe/txgbe_ethdev_vf.c
@@ -1291,6 +1291,8 @@ txgbevf_dev_interrupt_get_status(struct rte_eth_dev *dev)
 	/* Workround for ICR lost */
 	intr->flags |= TXGBE_FLAG_MAILBOX;
 
+	RTE_SET_USED(eicr);
+
 	return 0;
 }
 
diff --git a/drivers/net/vmxnet3/vmxnet3_rxtx.c b/drivers/net/vmxnet3/vmxnet3_rxtx.c
index deba64be6a..642c64dc2e 100644
--- a/drivers/net/vmxnet3/vmxnet3_rxtx.c
+++ b/drivers/net/vmxnet3/vmxnet3_rxtx.c
@@ -294,6 +294,7 @@ vmxnet3_dev_clear_queues(struct rte_eth_dev *dev)
 	}
 }
 
+#ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
 static int
 vmxnet3_unmap_pkt(uint16_t eop_idx, vmxnet3_tx_queue_t *txq)
 {
@@ -322,24 +323,31 @@ vmxnet3_unmap_pkt(uint16_t eop_idx, vmxnet3_tx_queue_t *txq)
 
 	return completed + 1;
 }
+#endif
 
 static void
 vmxnet3_tq_tx_complete(vmxnet3_tx_queue_t *txq)
 {
+	#ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
 	int completed = 0;
+	#endif
 	vmxnet3_comp_ring_t *comp_ring = &txq->comp_ring;
 	struct Vmxnet3_TxCompDesc *tcd = (struct Vmxnet3_TxCompDesc *)
 		(comp_ring->base + comp_ring->next2proc);
 
 	while (tcd->gen == comp_ring->gen) {
+		#ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
 		completed += vmxnet3_unmap_pkt(tcd->txdIdx, txq);
+		#endif
 
 		vmxnet3_comp_ring_adv_next2proc(comp_ring);
 		tcd = (struct Vmxnet3_TxCompDesc *)(comp_ring->base +
 						    comp_ring->next2proc);
 	}
 
+	#ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
 	PMD_TX_LOG(DEBUG, "Processed %d tx comps & command descs.", completed);
+	#endif
 }
 
 uint16_t
-- 
2.25.1

Re: [PATCH 5/7] drivers/net: fix unused but set variables

[Ferruh Yigit]

On 11/12/2021 2:01 PM, Conor Walsh wrote:
>   static void
>   vmxnet3_tq_tx_complete(vmxnet3_tx_queue_t *txq)
>   {
> +	#ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
>   	int completed = 0;
> +	#endif
>   	vmxnet3_comp_ring_t *comp_ring = &txq->comp_ring;
>   	struct Vmxnet3_TxCompDesc *tcd = (struct Vmxnet3_TxCompDesc *)
>   		(comp_ring->base + comp_ring->next2proc);
>   
>   	while (tcd->gen == comp_ring->gen) {
> +		#ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
>   		completed += vmxnet3_unmap_pkt(tcd->txdIdx, txq);
> +		#endif
>   
>   		vmxnet3_comp_ring_adv_next2proc(comp_ring);
>   		tcd = (struct Vmxnet3_TxCompDesc *)(comp_ring->base +
>   						    comp_ring->next2proc);
>   	}
>   
> +	#ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
>   	PMD_TX_LOG(DEBUG, "Processed %d tx comps & command descs.", completed);
> +	#endif
>   }

'PMD_TX_LOG' doesn't require 'RTE_LIBRTE_VMXNET3_DEBUG_TX' wrap.

Re: [PATCH 5/7] drivers/net: fix unused but set variables

[David Marchand]

On Fri, Nov 12, 2021 at 5:18 PM Ferruh Yigit <ferruh.yigit@intel.com> wrote:
>
> On 11/12/2021 2:01 PM, Conor Walsh wrote:
> >   static void
> >   vmxnet3_tq_tx_complete(vmxnet3_tx_queue_t *txq)
> >   {
> > +     #ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
> >       int completed = 0;
> > +     #endif
> >       vmxnet3_comp_ring_t *comp_ring = &txq->comp_ring;
> >       struct Vmxnet3_TxCompDesc *tcd = (struct Vmxnet3_TxCompDesc *)
> >               (comp_ring->base + comp_ring->next2proc);
> >
> >       while (tcd->gen == comp_ring->gen) {
> > +             #ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
> >               completed += vmxnet3_unmap_pkt(tcd->txdIdx, txq);
> > +             #endif

And this vmxnet_unmap_pkt does useful work.
We should probably still call it regardless of debug logs.


> >
> >               vmxnet3_comp_ring_adv_next2proc(comp_ring);
> >               tcd = (struct Vmxnet3_TxCompDesc *)(comp_ring->base +
> >                                                   comp_ring->next2proc);
> >       }
> >
> > +     #ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
> >       PMD_TX_LOG(DEBUG, "Processed %d tx comps & command descs.", completed);
> > +     #endif
> >   }
>
> 'PMD_TX_LOG' doesn't require 'RTE_LIBRTE_VMXNET3_DEBUG_TX' wrap.
>


-- 
David Marchand

RE: [PATCH 5/7] drivers/net: fix unused but set variables

[Walsh, Conor]

> Subject: Re: [PATCH 5/7] drivers/net: fix unused but set variables
> 
> On Fri, Nov 12, 2021 at 5:18 PM Ferruh Yigit <ferruh.yigit@intel.com> wrote:
> >
> > On 11/12/2021 2:01 PM, Conor Walsh wrote:
> > >   static void
> > >   vmxnet3_tq_tx_complete(vmxnet3_tx_queue_t *txq)
> > >   {
> > > +     #ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
> > >       int completed = 0;
> > > +     #endif
> > >       vmxnet3_comp_ring_t *comp_ring = &txq->comp_ring;
> > >       struct Vmxnet3_TxCompDesc *tcd = (struct Vmxnet3_TxCompDesc *)
> > >               (comp_ring->base + comp_ring->next2proc);
> > >
> > >       while (tcd->gen == comp_ring->gen) {
> > > +             #ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
> > >               completed += vmxnet3_unmap_pkt(tcd->txdIdx, txq);
> > > +             #endif
> 
> And this vmxnet_unmap_pkt does useful work.
> We should probably still call it regardless of debug logs.

Would something along the lines of this make more sense?

#ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
completed += vmxnet3_unmap_pkt(tcd->txdIdx, txq);
#else
vmxnet3_unmap_pkt(tcd->txdIdx, txq);
#endif

Thanks,
Conor.

> 
> 
> > >
> > >               vmxnet3_comp_ring_adv_next2proc(comp_ring);
> > >               tcd = (struct Vmxnet3_TxCompDesc *)(comp_ring->base +
> > >                                                   comp_ring->next2proc);
> > >       }
> > >
> > > +     #ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
> > >       PMD_TX_LOG(DEBUG, "Processed %d tx comps & command descs.",
> completed);
> > > +     #endif
> > >   }
> >
> > 'PMD_TX_LOG' doesn't require 'RTE_LIBRTE_VMXNET3_DEBUG_TX' wrap.
> >
> 
> 
> --
> David Marchand

Re: [PATCH 5/7] drivers/net: fix unused but set variables

[Stephen Hemminger]

On Fri, 12 Nov 2021 16:38:37 +0000
"Walsh, Conor" <conor.walsh@intel.com> wrote:

> Would something along the lines of this make more sense?
> 
> #ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
> completed += vmxnet3_unmap_pkt(tcd->txdIdx, txq);
> #else
> vmxnet3_unmap_pkt(tcd->txdIdx, txq);
> #endif
> 
> Thanks,
> Conor.

Why not just
	RTE_USED(completed);

#ifdef's introduce more testing issues.

RE: [PATCH 5/7] drivers/net: fix unused but set variables

[Walsh, Conor]

> On Fri, 12 Nov 2021 16:38:37 +0000
> "Walsh, Conor" <conor.walsh@intel.com> wrote:
> 
> > Would something along the lines of this make more sense?
> >
> > #ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
> > completed += vmxnet3_unmap_pkt(tcd->txdIdx, txq);
> > #else
> > vmxnet3_unmap_pkt(tcd->txdIdx, txq);
> > #endif
> >
> > Thanks,
> > Conor.
> 
> Why not just
> 	RTE_USED(completed);
> 
> #ifdef's introduce more testing issues.

That makes more sense, I'll change it to that in v2.

Thanks,
Conor.

RE: [PATCH 5/7] drivers/net: fix unused but set variables

[Walsh, Conor]

> On 11/12/2021 2:01 PM, Conor Walsh wrote:
> >   static void
> >   vmxnet3_tq_tx_complete(vmxnet3_tx_queue_t *txq)
> >   {
> > +	#ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
> >   	int completed = 0;
> > +	#endif
> >   	vmxnet3_comp_ring_t *comp_ring = &txq->comp_ring;
> >   	struct Vmxnet3_TxCompDesc *tcd = (struct Vmxnet3_TxCompDesc
> *)
> >   		(comp_ring->base + comp_ring->next2proc);
> >
> >   	while (tcd->gen == comp_ring->gen) {
> > +		#ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
> >   		completed += vmxnet3_unmap_pkt(tcd->txdIdx, txq);
> > +		#endif
> >
> >   		vmxnet3_comp_ring_adv_next2proc(comp_ring);
> >   		tcd = (struct Vmxnet3_TxCompDesc *)(comp_ring->base +
> >   						    comp_ring->next2proc);
> >   	}
> >
> > +	#ifdef RTE_LIBRTE_VMXNET3_DEBUG_TX
> >   	PMD_TX_LOG(DEBUG, "Processed %d tx comps & command descs.",
> completed);
> > +	#endif
> >   }
> 
> 'PMD_TX_LOG' doesn't require 'RTE_LIBRTE_VMXNET3_DEBUG_TX' wrap.

Hi Ferruh,

I will rework in v2.

Thanks,
Conor.

Re: [PATCH 5/7] drivers/net: fix unused but set variables

[David Marchand]

Hello Ajit, Somnath,

On Fri, Nov 12, 2021 at 3:02 PM Conor Walsh <conor.walsh@intel.com> wrote:
>
> This patch fixes unused but set variables in the following net drivers:
>  - bnxt
>  - hinic
>  - liquidio
>  - nfp
>  - octeontx
>  - qede
>  - txgbe
>  - vmxnet3
>
> This patch also fixes an unused function warning for vmxnet3_unmap_pkt
> from within vmxnet3
>
> Bugzilla ID: 881
>
> Fixes: b87abb2e55cb ("net/bnxt: support marking packet")
> Fixes: 88badb3aef94 ("net/bnxt: add helper functions for blob/regfile ops")
> Fixes: a4957d87e05a ("net/hinic/base: add mgmt module")
> Fixes: 5ee7640f95a0 ("net/liquidio: add API to flush IQ")
> Fixes: 19af5a38f7b6 ("net/nfp: move CPP bridge to separate file")
> Fixes: 9eb5cb3b11cc ("net/octeontx: fix access to indirect buffers")
> Fixes: 86a2265e59d7 ("qede: add SRIOV support")
> Fixes: b7311360fb67 ("net/txgbe: support VF interrupt")
> Fixes: c3ecdbb376da ("vmxnet3: support TSO")
>
> Signed-off-by: Conor Walsh <conor.walsh@intel.com>
> Reported-by: Liang Longfeng <longfengx.liang@intel.com>
> ---
>  drivers/net/bnxt/bnxt_rxr.c             | 22 ----------------------
>  drivers/net/bnxt/tf_ulp/ulp_utils.c     |  3 ---
>  drivers/net/hinic/base/hinic_pmd_mgmt.c |  6 ------
>  drivers/net/liquidio/lio_rxtx.c         |  2 --
>  drivers/net/nfp/nfp_cpp_bridge.c        |  6 ++----
>  drivers/net/octeontx/octeontx_rxtx.h    |  3 +--
>  drivers/net/qede/base/ecore_vf.c        |  5 +----
>  drivers/net/txgbe/txgbe_ethdev_vf.c     |  2 ++
>  drivers/net/vmxnet3/vmxnet3_rxtx.c      |  8 ++++++++
>  9 files changed, 14 insertions(+), 43 deletions(-)
>
> diff --git a/drivers/net/bnxt/bnxt_rxr.c b/drivers/net/bnxt/bnxt_rxr.c
> index 8bc8ddc353..157297bc64 100644
> --- a/drivers/net/bnxt/bnxt_rxr.c
> +++ b/drivers/net/bnxt/bnxt_rxr.c
> @@ -823,9 +823,6 @@ void bnxt_set_mark_in_mbuf(struct bnxt *bp,
>                            struct rte_mbuf *mbuf)
>  {
>         uint32_t cfa_code = 0;
> -       uint8_t meta_fmt = 0;
> -       uint16_t flags2 = 0;
> -       uint32_t meta =  0;
>
>         cfa_code = rte_le_to_cpu_16(rxcmp1->cfa_code);
>         if (!cfa_code)
> @@ -834,25 +831,6 @@ void bnxt_set_mark_in_mbuf(struct bnxt *bp,
>         if (cfa_code && !bp->mark_table[cfa_code].valid)
>                 return;
>
> -       flags2 = rte_le_to_cpu_16(rxcmp1->flags2);
> -       meta = rte_le_to_cpu_32(rxcmp1->metadata);
> -       if (meta) {
> -               meta >>= BNXT_RX_META_CFA_CODE_SHIFT;
> -
> -               /* The flags field holds extra bits of info from [6:4]
> -                * which indicate if the flow is in TCAM or EM or EEM
> -                */
> -               meta_fmt = (flags2 & BNXT_CFA_META_FMT_MASK) >>
> -                          BNXT_CFA_META_FMT_SHFT;
> -
> -               /* meta_fmt == 4 => 'b100 => 'b10x => EM.
> -                * meta_fmt == 5 => 'b101 => 'b10x => EM + VLAN
> -                * meta_fmt == 6 => 'b110 => 'b11x => EEM
> -                * meta_fmt == 7 => 'b111 => 'b11x => EEM + VLAN.
> -                */
> -               meta_fmt >>= BNXT_CFA_META_FMT_EM_EEM_SHFT;
> -       }
> -
>         mbuf->hash.fdir.hi = bp->mark_table[cfa_code].mark_id;
>         mbuf->ol_flags |= RTE_MBUF_F_RX_FDIR | RTE_MBUF_F_RX_FDIR_ID;
>  }
> diff --git a/drivers/net/bnxt/tf_ulp/ulp_utils.c b/drivers/net/bnxt/tf_ulp/ulp_utils.c
> index df3afaa6fd..c60d81d14a 100644
> --- a/drivers/net/bnxt/tf_ulp/ulp_utils.c
> +++ b/drivers/net/bnxt/tf_ulp/ulp_utils.c
> @@ -200,7 +200,6 @@ ulp_bs_push_msb(uint8_t *bs, uint16_t pos, uint8_t len, uint8_t *val)
>  {
>         int i;
>         int cnt = (len + 7) / 8;
> -       int tlen = len;
>
>         /* Handle any remainder bits */
>         int tmp = len % 8;
> @@ -211,12 +210,10 @@ ulp_bs_push_msb(uint8_t *bs, uint16_t pos, uint8_t len, uint8_t *val)
>         ulp_bs_put_msb(bs, pos, tmp, val[0]);
>
>         pos += tmp;
> -       tlen -= tmp;
>
>         for (i = 1; i < cnt; i++) {
>                 ulp_bs_put_msb(bs, pos, 8, val[i]);
>                 pos += 8;
> -               tlen -= 8;
>         }
>
>         return len;

Review please.


-- 
David Marchand

Re: [PATCH 5/7] drivers/net: fix unused but set variables

[Somnath Kotur]

On Mon, Nov 15, 2021 at 2:58 PM David Marchand
<david.marchand@redhat.com> wrote:
>
> Hello Ajit, Somnath,
>
> On Fri, Nov 12, 2021 at 3:02 PM Conor Walsh <conor.walsh@intel.com> wrote:
> >
> > This patch fixes unused but set variables in the following net drivers:
> >  - bnxt
> >  - hinic
> >  - liquidio
> >  - nfp
> >  - octeontx
> >  - qede
> >  - txgbe
> >  - vmxnet3
> >
> > This patch also fixes an unused function warning for vmxnet3_unmap_pkt
> > from within vmxnet3
> >
> > Bugzilla ID: 881
> >
> > Fixes: b87abb2e55cb ("net/bnxt: support marking packet")
> > Fixes: 88badb3aef94 ("net/bnxt: add helper functions for blob/regfile ops")
> > Fixes: a4957d87e05a ("net/hinic/base: add mgmt module")
> > Fixes: 5ee7640f95a0 ("net/liquidio: add API to flush IQ")
> > Fixes: 19af5a38f7b6 ("net/nfp: move CPP bridge to separate file")
> > Fixes: 9eb5cb3b11cc ("net/octeontx: fix access to indirect buffers")
> > Fixes: 86a2265e59d7 ("qede: add SRIOV support")
> > Fixes: b7311360fb67 ("net/txgbe: support VF interrupt")
> > Fixes: c3ecdbb376da ("vmxnet3: support TSO")
> >
> > Signed-off-by: Conor Walsh <conor.walsh@intel.com>
> > Reported-by: Liang Longfeng <longfengx.liang@intel.com>
> > ---
> >  drivers/net/bnxt/bnxt_rxr.c             | 22 ----------------------
> >  drivers/net/bnxt/tf_ulp/ulp_utils.c     |  3 ---
> >  drivers/net/hinic/base/hinic_pmd_mgmt.c |  6 ------
> >  drivers/net/liquidio/lio_rxtx.c         |  2 --
> >  drivers/net/nfp/nfp_cpp_bridge.c        |  6 ++----
> >  drivers/net/octeontx/octeontx_rxtx.h    |  3 +--
> >  drivers/net/qede/base/ecore_vf.c        |  5 +----
> >  drivers/net/txgbe/txgbe_ethdev_vf.c     |  2 ++
> >  drivers/net/vmxnet3/vmxnet3_rxtx.c      |  8 ++++++++
> >  9 files changed, 14 insertions(+), 43 deletions(-)
> >
> > diff --git a/drivers/net/bnxt/bnxt_rxr.c b/drivers/net/bnxt/bnxt_rxr.c
> > index 8bc8ddc353..157297bc64 100644
> > --- a/drivers/net/bnxt/bnxt_rxr.c
> > +++ b/drivers/net/bnxt/bnxt_rxr.c
> > @@ -823,9 +823,6 @@ void bnxt_set_mark_in_mbuf(struct bnxt *bp,
> >                            struct rte_mbuf *mbuf)
> >  {
> >         uint32_t cfa_code = 0;
> > -       uint8_t meta_fmt = 0;
> > -       uint16_t flags2 = 0;
> > -       uint32_t meta =  0;
> >
> >         cfa_code = rte_le_to_cpu_16(rxcmp1->cfa_code);
> >         if (!cfa_code)
> > @@ -834,25 +831,6 @@ void bnxt_set_mark_in_mbuf(struct bnxt *bp,
> >         if (cfa_code && !bp->mark_table[cfa_code].valid)
> >                 return;
> >
> > -       flags2 = rte_le_to_cpu_16(rxcmp1->flags2);
> > -       meta = rte_le_to_cpu_32(rxcmp1->metadata);
> > -       if (meta) {
> > -               meta >>= BNXT_RX_META_CFA_CODE_SHIFT;
> > -
> > -               /* The flags field holds extra bits of info from [6:4]
> > -                * which indicate if the flow is in TCAM or EM or EEM
> > -                */
> > -               meta_fmt = (flags2 & BNXT_CFA_META_FMT_MASK) >>
> > -                          BNXT_CFA_META_FMT_SHFT;
> > -
> > -               /* meta_fmt == 4 => 'b100 => 'b10x => EM.
> > -                * meta_fmt == 5 => 'b101 => 'b10x => EM + VLAN
> > -                * meta_fmt == 6 => 'b110 => 'b11x => EEM
> > -                * meta_fmt == 7 => 'b111 => 'b11x => EEM + VLAN.
> > -                */
> > -               meta_fmt >>= BNXT_CFA_META_FMT_EM_EEM_SHFT;
> > -       }
> > -
> >         mbuf->hash.fdir.hi = bp->mark_table[cfa_code].mark_id;
> >         mbuf->ol_flags |= RTE_MBUF_F_RX_FDIR | RTE_MBUF_F_RX_FDIR_ID;
> >  }
> > diff --git a/drivers/net/bnxt/tf_ulp/ulp_utils.c b/drivers/net/bnxt/tf_ulp/ulp_utils.c
> > index df3afaa6fd..c60d81d14a 100644
> > --- a/drivers/net/bnxt/tf_ulp/ulp_utils.c
> > +++ b/drivers/net/bnxt/tf_ulp/ulp_utils.c
> > @@ -200,7 +200,6 @@ ulp_bs_push_msb(uint8_t *bs, uint16_t pos, uint8_t len, uint8_t *val)
> >  {
> >         int i;
> >         int cnt = (len + 7) / 8;
> > -       int tlen = len;
> >
> >         /* Handle any remainder bits */
> >         int tmp = len % 8;
> > @@ -211,12 +210,10 @@ ulp_bs_push_msb(uint8_t *bs, uint16_t pos, uint8_t len, uint8_t *val)
> >         ulp_bs_put_msb(bs, pos, tmp, val[0]);
> >
> >         pos += tmp;
> > -       tlen -= tmp;
> >
> >         for (i = 1; i < cnt; i++) {
> >                 ulp_bs_put_msb(bs, pos, 8, val[i]);
> >                 pos += 8;
> > -               tlen -= 8;
> >         }
> >
> >         return len;
>
> Review please.
Acked-by: Somnath Kotur <somnath.kotur@broadcom.com>
>
>
> --
> David Marchand
>

Re: [PATCH 5/7] drivers/net: fix unused but set variables

[Ajit Khaparde]

On Mon, Nov 15, 2021 at 8:58 PM Somnath Kotur
<somnath.kotur@broadcom.com> wrote:
>
> On Mon, Nov 15, 2021 at 2:58 PM David Marchand
> <david.marchand@redhat.com> wrote:
> >
> > Hello Ajit, Somnath,
> >
> > On Fri, Nov 12, 2021 at 3:02 PM Conor Walsh <conor.walsh@intel.com> wrote:
> > >
> > > This patch fixes unused but set variables in the following net drivers:
> > >  - bnxt
> > >  - hinic
> > >  - liquidio
> > >  - nfp
> > >  - octeontx
> > >  - qede
> > >  - txgbe
> > >  - vmxnet3
> > >
> > > This patch also fixes an unused function warning for vmxnet3_unmap_pkt
> > > from within vmxnet3
> > >
> > > Bugzilla ID: 881
> > >
> > > Fixes: b87abb2e55cb ("net/bnxt: support marking packet")
> > > Fixes: 88badb3aef94 ("net/bnxt: add helper functions for blob/regfile ops")
> > > Fixes: a4957d87e05a ("net/hinic/base: add mgmt module")
> > > Fixes: 5ee7640f95a0 ("net/liquidio: add API to flush IQ")
> > > Fixes: 19af5a38f7b6 ("net/nfp: move CPP bridge to separate file")
> > > Fixes: 9eb5cb3b11cc ("net/octeontx: fix access to indirect buffers")
> > > Fixes: 86a2265e59d7 ("qede: add SRIOV support")
> > > Fixes: b7311360fb67 ("net/txgbe: support VF interrupt")
> > > Fixes: c3ecdbb376da ("vmxnet3: support TSO")
> > >
> > > Signed-off-by: Conor Walsh <conor.walsh@intel.com>
> > > Reported-by: Liang Longfeng <longfengx.liang@intel.com>
> > > ---
> > >  drivers/net/bnxt/bnxt_rxr.c             | 22 ----------------------
> > >  drivers/net/bnxt/tf_ulp/ulp_utils.c     |  3 ---
> > >  drivers/net/hinic/base/hinic_pmd_mgmt.c |  6 ------
> > >  drivers/net/liquidio/lio_rxtx.c         |  2 --
> > >  drivers/net/nfp/nfp_cpp_bridge.c        |  6 ++----
> > >  drivers/net/octeontx/octeontx_rxtx.h    |  3 +--
> > >  drivers/net/qede/base/ecore_vf.c        |  5 +----
> > >  drivers/net/txgbe/txgbe_ethdev_vf.c     |  2 ++
> > >  drivers/net/vmxnet3/vmxnet3_rxtx.c      |  8 ++++++++
> > >  9 files changed, 14 insertions(+), 43 deletions(-)
> > >
> > > diff --git a/drivers/net/bnxt/bnxt_rxr.c b/drivers/net/bnxt/bnxt_rxr.c
> > > index 8bc8ddc353..157297bc64 100644
> > > --- a/drivers/net/bnxt/bnxt_rxr.c
> > > +++ b/drivers/net/bnxt/bnxt_rxr.c
> > > @@ -823,9 +823,6 @@ void bnxt_set_mark_in_mbuf(struct bnxt *bp,
> > >                            struct rte_mbuf *mbuf)
> > >  {
> > >         uint32_t cfa_code = 0;
> > > -       uint8_t meta_fmt = 0;
> > > -       uint16_t flags2 = 0;
> > > -       uint32_t meta =  0;
> > >
> > >         cfa_code = rte_le_to_cpu_16(rxcmp1->cfa_code);
> > >         if (!cfa_code)
> > > @@ -834,25 +831,6 @@ void bnxt_set_mark_in_mbuf(struct bnxt *bp,
> > >         if (cfa_code && !bp->mark_table[cfa_code].valid)
> > >                 return;
> > >
> > > -       flags2 = rte_le_to_cpu_16(rxcmp1->flags2);
> > > -       meta = rte_le_to_cpu_32(rxcmp1->metadata);
> > > -       if (meta) {
> > > -               meta >>= BNXT_RX_META_CFA_CODE_SHIFT;
> > > -
> > > -               /* The flags field holds extra bits of info from [6:4]
> > > -                * which indicate if the flow is in TCAM or EM or EEM
> > > -                */
> > > -               meta_fmt = (flags2 & BNXT_CFA_META_FMT_MASK) >>
> > > -                          BNXT_CFA_META_FMT_SHFT;
> > > -
> > > -               /* meta_fmt == 4 => 'b100 => 'b10x => EM.
> > > -                * meta_fmt == 5 => 'b101 => 'b10x => EM + VLAN
> > > -                * meta_fmt == 6 => 'b110 => 'b11x => EEM
> > > -                * meta_fmt == 7 => 'b111 => 'b11x => EEM + VLAN.
> > > -                */
> > > -               meta_fmt >>= BNXT_CFA_META_FMT_EM_EEM_SHFT;
> > > -       }
> > > -
> > >         mbuf->hash.fdir.hi = bp->mark_table[cfa_code].mark_id;
> > >         mbuf->ol_flags |= RTE_MBUF_F_RX_FDIR | RTE_MBUF_F_RX_FDIR_ID;
> > >  }
> > > diff --git a/drivers/net/bnxt/tf_ulp/ulp_utils.c b/drivers/net/bnxt/tf_ulp/ulp_utils.c
> > > index df3afaa6fd..c60d81d14a 100644
> > > --- a/drivers/net/bnxt/tf_ulp/ulp_utils.c
> > > +++ b/drivers/net/bnxt/tf_ulp/ulp_utils.c
> > > @@ -200,7 +200,6 @@ ulp_bs_push_msb(uint8_t *bs, uint16_t pos, uint8_t len, uint8_t *val)
> > >  {
> > >         int i;
> > >         int cnt = (len + 7) / 8;
> > > -       int tlen = len;
> > >
> > >         /* Handle any remainder bits */
> > >         int tmp = len % 8;
> > > @@ -211,12 +210,10 @@ ulp_bs_push_msb(uint8_t *bs, uint16_t pos, uint8_t len, uint8_t *val)
> > >         ulp_bs_put_msb(bs, pos, tmp, val[0]);
> > >
> > >         pos += tmp;
> > > -       tlen -= tmp;
> > >
> > >         for (i = 1; i < cnt; i++) {
> > >                 ulp_bs_put_msb(bs, pos, 8, val[i]);
> > >                 pos += 8;
> > > -               tlen -= 8;
> > >         }
> > >
> > >         return len;
> >
> > Review please.
Thanks for the reminder.

> Acked-by: Somnath Kotur <somnath.kotur@broadcom.com>

Acked-by: Ajit Khaparde <ajit.khaparde@broadcom.com>

> >
> >
> > --
> > David Marchand
> >

[PATCH 6/7] examples/performance-thread: fix unused but set variables

[Conor Walsh]

This patch removes the hits variable from the cpu_load_collector function
within the performance thread example app as it is an unused but set
variable.

Bugzilla ID: 881

Fixes: d48415e1fee3 ("examples/performance-thread: add l3fwd-thread app")

Signed-off-by: Conor Walsh <conor.walsh@intel.com>
Reported-by: Liang Longfeng <longfengx.liang@intel.com>
---
 examples/performance-thread/l3fwd-thread/main.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/examples/performance-thread/l3fwd-thread/main.c b/examples/performance-thread/l3fwd-thread/main.c
index 3a6a33bda3..8a35040597 100644
--- a/examples/performance-thread/l3fwd-thread/main.c
+++ b/examples/performance-thread/l3fwd-thread/main.c
@@ -1885,7 +1885,6 @@ process_burst(struct rte_mbuf *pkts_burst[MAX_PKT_BURST], int nb_rx,
 static int __rte_noreturn
 cpu_load_collector(__rte_unused void *arg) {
 	unsigned i, j, k;
-	uint64_t hits;
 	uint64_t prev_tsc, diff_tsc, cur_tsc;
 	uint64_t total[MAX_CPU] = { 0 };
 	unsigned min_cpu = MAX_CPU;
@@ -1972,12 +1971,10 @@ cpu_load_collector(__rte_unused void *arg) {
 			printf("cpu#     proc%%  poll%%  overhead%%\n\n");
 
 			for (i = min_cpu; i <= max_cpu; i++) {
-				hits = 0;
 				printf("CPU %d:", i);
 				for (j = 0; j < MAX_CPU_COUNTER; j++) {
 					printf("%7" PRIu64 "",
 							cpu_load.hits[j][i] * 100 / cpu_load.counter);
-					hits += cpu_load.hits[j][i];
 					cpu_load.hits[j][i] = 0;
 				}
 				printf("%7" PRIu64 "\n",
-- 
2.25.1

[PATCH 7/7] lib/power: fix unused but set variables

[Conor Walsh]

This patch removes the tot_ppi variable from the update_stats function
within empty poll in the power library as it is an unused but set
variable.

Bugzilla ID: 881

Fixes: 450f0791312c ("power: add traffic pattern aware power control")

Signed-off-by: Conor Walsh <conor.walsh@intel.com>
Reported-by: Liang Longfeng <longfengx.liang@intel.com>
---
 lib/power/rte_power_empty_poll.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/lib/power/rte_power_empty_poll.c b/lib/power/rte_power_empty_poll.c
index 975aa92997..8a2d60c576 100644
--- a/lib/power/rte_power_empty_poll.c
+++ b/lib/power/rte_power_empty_poll.c
@@ -207,7 +207,7 @@ update_training_stats(struct priority_worker *poll_stats,
 static __rte_always_inline uint32_t
 update_stats(struct priority_worker *poll_stats)
 {
-	uint64_t tot_edpi = 0, tot_ppi = 0;
+	uint64_t tot_edpi = 0;
 	uint32_t j, percent;
 
 	struct priority_worker *s = poll_stats;
@@ -237,7 +237,6 @@ update_stats(struct priority_worker *poll_stats)
 
 	for (j = 0; j < BINS_AV; j++) {
 		tot_edpi += s->edpi_av[j];
-		tot_ppi += s->ppi_av[j];
 	}
 
 	tot_edpi = tot_edpi / BINS_AV;
-- 
2.25.1

Re: [PATCH 7/7] lib/power: fix unused but set variables

[David Marchand]

Hello Conor,


On Fri, Nov 12, 2021 at 3:02 PM Conor Walsh <conor.walsh@intel.com> wrote:
>
> This patch removes the tot_ppi variable from the update_stats function
> within empty poll in the power library as it is an unused but set
> variable.

About the commitlog, commenting on this patch, but it applies to other
of the series.
>
> Bugzilla ID: 881

No need for an empty line between bugzilla and fixes lines.

>
> Fixes: 450f0791312c ("power: add traffic pattern aware power control")

We will want those fixes in stable branches.
Cc: stable@dpdk.org

>
> Signed-off-by: Conor Walsh <conor.walsh@intel.com>
> Reported-by: Liang Longfeng <longfengx.liang@intel.com>

Reports come before signing off a fix :-)


Now, on this specific patch, Jim already provided a fix for this
issue, you can drop it from your series.
https://patchwork.dpdk.org/project/dpdk/patch/20211105155351.350403-1-james.r.harris@intel.com/


-- 
David Marchand

RE: [PATCH 7/7] lib/power: fix unused but set variables

[Walsh, Conor]

Hi David,

I will make those changes to the commit logs and respin them.
I'll also drop this patch from the set.

Thanks,
Conor.

> Hello Conor,
> 
> 
> On Fri, Nov 12, 2021 at 3:02 PM Conor Walsh <conor.walsh@intel.com>
> wrote:
> >
> > This patch removes the tot_ppi variable from the update_stats function
> > within empty poll in the power library as it is an unused but set
> > variable.
> 
> About the commitlog, commenting on this patch, but it applies to other
> of the series.
> >
> > Bugzilla ID: 881
> 
> No need for an empty line between bugzilla and fixes lines.
> 
> >
> > Fixes: 450f0791312c ("power: add traffic pattern aware power control")
> 
> We will want those fixes in stable branches.
> Cc: stable@dpdk.org
> 
> >
> > Signed-off-by: Conor Walsh <conor.walsh@intel.com>
> > Reported-by: Liang Longfeng <longfengx.liang@intel.com>
> 
> Reports come before signing off a fix :-)
> 
> 
> Now, on this specific patch, Jim already provided a fix for this
> issue, you can drop it from your series.
> https://patchwork.dpdk.org/project/dpdk/patch/20211105155351.350403-1-
> james.r.harris@intel.com/
> 
> 
> --
> David Marchand

Re: [PATCH 7/7] lib/power: fix unused but set variables

[David Marchand]

On Fri, Nov 12, 2021 at 3:17 PM Walsh, Conor <conor.walsh@intel.com> wrote:
>
> Hi David,
>
> I will make those changes to the commit logs and respin them.
> I'll also drop this patch from the set.

These comments are trivial.

We might get more substantial comments from drivers maintainers.
I recommend waiting a bit before respinning.

Thanks.


-- 
David Marchand

[PATCH 0/6] fix unused but set variables

[Conor Walsh]

When DPDK is compiled with clang 13, several warnings for unused but set
variables are present.

This patchset fixes these warnings, which were present within:
 - app/test
 - drivers/bus/fslmc
 - drivers/common/qat
 - drivers/event/sw
 - drivers/net
 - examples/performance-thread

This patchset is split into 6 patches grouped broadly by subdir, as this
set touches code from a lot of different authors this should make it
easier to review.

Bugzilla ID: 881

---

v2:
 - Drop lib/power patch as already covered by 0353121c3384 ("power: fix
   build with clang 13").
 - Rework changes to vmxnet3 to ensure vmxnet_unmap_pkt is run.
 - Minor changes to commit logs.
 - Dropped all addresses from CC that bounced back.

Conor Walsh (6):
  app/test: fix unused but set variables
  drivers/bus/fslmc: fix unused but set variables
  drivers/common/qat: fix unused but set variables
  drivers/event/sw: fix unused but set variables
  drivers/net: fix unused but set variables
  examples/performance-thread: fix unused but set variables

 app/test/test_distributor_perf.c              |  3 ---
 app/test/test_thash_perf.c                    |  3 +++
 drivers/bus/fslmc/fslmc_bus.c                 |  2 --
 drivers/common/qat/qat_pf2vf.c                |  2 --
 drivers/event/sw/sw_evdev.c                   |  2 --
 drivers/net/bnxt/bnxt_rxr.c                   | 22 -------------------
 drivers/net/bnxt/tf_ulp/ulp_utils.c           |  3 ---
 drivers/net/hinic/base/hinic_pmd_mgmt.c       |  6 -----
 drivers/net/liquidio/lio_rxtx.c               |  2 --
 drivers/net/nfp/nfp_cpp_bridge.c              |  6 ++---
 drivers/net/octeontx/octeontx_rxtx.h          |  3 +--
 drivers/net/qede/base/ecore_vf.c              |  5 +----
 drivers/net/txgbe/txgbe_ethdev_vf.c           |  3 +++
 drivers/net/vmxnet3/vmxnet3_rxtx.c            |  3 +++
 .../performance-thread/l3fwd-thread/main.c    |  3 ---
 15 files changed, 13 insertions(+), 55 deletions(-)

-- 
2.25.1

[PATCH 1/6] app/test: fix unused but set variables

[Conor Walsh]

This patch fixes unused but set variables within the following test apps:
 - test distributor perf
 - test thash perf

Bugzilla ID: 881
Fixes: c0de0eb82e40 ("distributor: switch over to new API")
Fixes: 239fffe0402e ("test/thash: add performance tests for Toeplitz hash")
Cc: stable@dpdk.org

Reported-by: Liang Longfeng <longfengx.liang@intel.com>
Signed-off-by: Conor Walsh <conor.walsh@intel.com>
Acked-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>
---
 app/test/test_distributor_perf.c | 3 ---
 app/test/test_thash_perf.c       | 3 +++
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/app/test/test_distributor_perf.c b/app/test/test_distributor_perf.c
index fdbeae6d2f..92e330f194 100644
--- a/app/test/test_distributor_perf.c
+++ b/app/test/test_distributor_perf.c
@@ -108,7 +108,6 @@ static int
 handle_work(void *arg)
 {
 	struct rte_distributor *d = arg;
-	unsigned int count = 0;
 	unsigned int num = 0;
 	int i;
 	unsigned int id = __atomic_fetch_add(&worker_idx, 1, __ATOMIC_RELAXED);
@@ -120,11 +119,9 @@ handle_work(void *arg)
 	num = rte_distributor_get_pkt(d, id, buf, buf, num);
 	while (!quit) {
 		worker_stats[id].handled_packets += num;
-		count += num;
 		num = rte_distributor_get_pkt(d, id, buf, buf, num);
 	}
 	worker_stats[id].handled_packets += num;
-	count += num;
 	rte_distributor_return_pkt(d, id, buf, num);
 	return 0;
 }
diff --git a/app/test/test_thash_perf.c b/app/test/test_thash_perf.c
index 7aa9360120..687582aa32 100644
--- a/app/test/test_thash_perf.c
+++ b/app/test/test_thash_perf.c
@@ -61,6 +61,9 @@ run_rss_calc(uint32_t *tuples[BATCH_SZ], enum test_rss_type type, int len,
 	}
 	end_tsc = rte_rdtsc_precise();
 
+	/* To avoid compiler warnings set hash to used. */
+	RTE_SET_USED(hash);
+
 	return end_tsc - start_tsc;
 }
 
-- 
2.25.1

[PATCH 2/6] drivers/bus/fslmc: fix unused but set variables

[Conor Walsh]

This patch removes the device_count variable from the rte_fslmc_scan
function within the fslmc bus driver as it is an unused but set variable.

Bugzilla ID: 881
Fixes: 828d51d8fc3e ("bus/fslmc: refactor scan and probe functions")
Cc: stable@dpdk.org

Reported-by: Liang Longfeng <longfengx.liang@intel.com>
Signed-off-by: Conor Walsh <conor.walsh@intel.com>
---
 drivers/bus/fslmc/fslmc_bus.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/bus/fslmc/fslmc_bus.c b/drivers/bus/fslmc/fslmc_bus.c
index ac3cb4aa5a..a0ef24cdc8 100644
--- a/drivers/bus/fslmc/fslmc_bus.c
+++ b/drivers/bus/fslmc/fslmc_bus.c
@@ -315,7 +315,6 @@ static int
 rte_fslmc_scan(void)
 {
 	int ret;
-	int device_count = 0;
 	char fslmc_dirpath[PATH_MAX];
 	DIR *dir;
 	struct dirent *entry;
@@ -349,7 +348,6 @@ rte_fslmc_scan(void)
 			/* Error in parsing directory - exit gracefully */
 			goto scan_fail_cleanup;
 		}
-		device_count += 1;
 	}
 
 	closedir(dir);
-- 
2.25.1

[PATCH 3/6] drivers/common/qat: fix unused but set variables

[Conor Walsh]

This patch removes the us variable from the qat_pf2vf_exch_msg function
within the PF to VF section of the QAT driver as it is an unused but set
variable.

Bugzilla ID: 881
Fixes: b17d16fb47b4 ("common/qat: add PF to VF communication")
Cc: stable@dpdk.org

Reported-by: Liang Longfeng <longfengx.liang@intel.com>
Signed-off-by: Conor Walsh <conor.walsh@intel.com>
---
 drivers/common/qat/qat_pf2vf.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/common/qat/qat_pf2vf.c b/drivers/common/qat/qat_pf2vf.c
index 6327311199..4e9ffc72fc 100644
--- a/drivers/common/qat/qat_pf2vf.c
+++ b/drivers/common/qat/qat_pf2vf.c
@@ -54,14 +54,12 @@ int qat_pf2vf_exch_msg(struct qat_pci_device *qat_dev,
 		msg |= ADF_PFVF_INT | ADF_PFVF_MSGORIGIN_SYSTEM;
 
 		ADF_CSR_WR(pmisc_bar_addr, vf_csr_off, msg);
-		int us = 0;
 		/*
 		 * Wait for confirmation from remote that it received
 		 * the message
 		 */
 		do {
 			rte_delay_us_sleep(5);
-			us += 5;
 			val = ADF_CSR_RD(pmisc_bar_addr, vf_csr_off);
 		} while ((val & ADF_PFVF_INT) &&
 			(++count < ADF_IOV_MSG_ACK_MAX_RETRY));
-- 
2.25.1

[PATCH 4/6] drivers/event/sw: fix unused but set variables

[Conor Walsh]

This patch removes the inflights variable from the sw_dump function
within the software section of the event driver as it is an unused but
set variable.

Bugzilla ID: 881
Fixes: c66baa68e453 ("event/sw: add dump function for easier debugging")
Cc: stable@dpdk.org

Reported-by: Liang Longfeng <longfengx.liang@intel.com>
Signed-off-by: Conor Walsh <conor.walsh@intel.com>
Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
---
 drivers/event/sw/sw_evdev.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/event/sw/sw_evdev.c b/drivers/event/sw/sw_evdev.c
index 070a4802e9..fa72ceb949 100644
--- a/drivers/event/sw/sw_evdev.c
+++ b/drivers/event/sw/sw_evdev.c
@@ -711,7 +711,6 @@ sw_dump(struct rte_eventdev *dev, FILE *f)
 			continue;
 		}
 		int affinities_per_port[SW_PORTS_MAX] = {0};
-		uint32_t inflights = 0;
 
 		fprintf(f, "  Queue %d (%s)\n", i, q_type_strings[qid->type]);
 		fprintf(f, "\trx   %"PRIu64"\tdrop %"PRIu64"\ttx   %"PRIu64"\n",
@@ -732,7 +731,6 @@ sw_dump(struct rte_eventdev *dev, FILE *f)
 		for (flow = 0; flow < RTE_DIM(qid->fids); flow++)
 			if (qid->fids[flow].cq != -1) {
 				affinities_per_port[qid->fids[flow].cq]++;
-				inflights += qid->fids[flow].pcount;
 			}
 
 		uint32_t port;
-- 
2.25.1

[PATCH 5/6] drivers/net: fix unused but set variables

[Conor Walsh]

This patch fixes unused but set variables in the following net drivers:
 - bnxt
 - hinic
 - liquidio
 - nfp
 - octeontx
 - qede
 - txgbe
 - vmxnet3

Bugzilla ID: 881
Fixes: b87abb2e55cb ("net/bnxt: support marking packet")
Fixes: 88badb3aef94 ("net/bnxt: add helper functions for blob/regfile ops")
Fixes: a4957d87e05a ("net/hinic/base: add mgmt module")
Fixes: 5ee7640f95a0 ("net/liquidio: add API to flush IQ")
Fixes: 19af5a38f7b6 ("net/nfp: move CPP bridge to separate file")
Fixes: 9eb5cb3b11cc ("net/octeontx: fix access to indirect buffers")
Fixes: 86a2265e59d7 ("qede: add SRIOV support")
Fixes: b7311360fb67 ("net/txgbe: support VF interrupt")
Fixes: c3ecdbb376da ("vmxnet3: support TSO")
Cc: stable@dpdk.org

Reported-by: Liang Longfeng <longfengx.liang@intel.com>
Signed-off-by: Conor Walsh <conor.walsh@intel.com>
---
 drivers/net/bnxt/bnxt_rxr.c             | 22 ----------------------
 drivers/net/bnxt/tf_ulp/ulp_utils.c     |  3 ---
 drivers/net/hinic/base/hinic_pmd_mgmt.c |  6 ------
 drivers/net/liquidio/lio_rxtx.c         |  2 --
 drivers/net/nfp/nfp_cpp_bridge.c        |  6 ++----
 drivers/net/octeontx/octeontx_rxtx.h    |  3 +--
 drivers/net/qede/base/ecore_vf.c        |  5 +----
 drivers/net/txgbe/txgbe_ethdev_vf.c     |  3 +++
 drivers/net/vmxnet3/vmxnet3_rxtx.c      |  3 +++
 9 files changed, 10 insertions(+), 43 deletions(-)

diff --git a/drivers/net/bnxt/bnxt_rxr.c b/drivers/net/bnxt/bnxt_rxr.c
index 8bc8ddc353..157297bc64 100644
--- a/drivers/net/bnxt/bnxt_rxr.c
+++ b/drivers/net/bnxt/bnxt_rxr.c
@@ -823,9 +823,6 @@ void bnxt_set_mark_in_mbuf(struct bnxt *bp,
 			   struct rte_mbuf *mbuf)
 {
 	uint32_t cfa_code = 0;
-	uint8_t meta_fmt = 0;
-	uint16_t flags2 = 0;
-	uint32_t meta =  0;
 
 	cfa_code = rte_le_to_cpu_16(rxcmp1->cfa_code);
 	if (!cfa_code)
@@ -834,25 +831,6 @@ void bnxt_set_mark_in_mbuf(struct bnxt *bp,
 	if (cfa_code && !bp->mark_table[cfa_code].valid)
 		return;
 
-	flags2 = rte_le_to_cpu_16(rxcmp1->flags2);
-	meta = rte_le_to_cpu_32(rxcmp1->metadata);
-	if (meta) {
-		meta >>= BNXT_RX_META_CFA_CODE_SHIFT;
-
-		/* The flags field holds extra bits of info from [6:4]
-		 * which indicate if the flow is in TCAM or EM or EEM
-		 */
-		meta_fmt = (flags2 & BNXT_CFA_META_FMT_MASK) >>
-			   BNXT_CFA_META_FMT_SHFT;
-
-		/* meta_fmt == 4 => 'b100 => 'b10x => EM.
-		 * meta_fmt == 5 => 'b101 => 'b10x => EM + VLAN
-		 * meta_fmt == 6 => 'b110 => 'b11x => EEM
-		 * meta_fmt == 7 => 'b111 => 'b11x => EEM + VLAN.
-		 */
-		meta_fmt >>= BNXT_CFA_META_FMT_EM_EEM_SHFT;
-	}
-
 	mbuf->hash.fdir.hi = bp->mark_table[cfa_code].mark_id;
 	mbuf->ol_flags |= RTE_MBUF_F_RX_FDIR | RTE_MBUF_F_RX_FDIR_ID;
 }
diff --git a/drivers/net/bnxt/tf_ulp/ulp_utils.c b/drivers/net/bnxt/tf_ulp/ulp_utils.c
index df3afaa6fd..c60d81d14a 100644
--- a/drivers/net/bnxt/tf_ulp/ulp_utils.c
+++ b/drivers/net/bnxt/tf_ulp/ulp_utils.c
@@ -200,7 +200,6 @@ ulp_bs_push_msb(uint8_t *bs, uint16_t pos, uint8_t len, uint8_t *val)
 {
 	int i;
 	int cnt = (len + 7) / 8;
-	int tlen = len;
 
 	/* Handle any remainder bits */
 	int tmp = len % 8;
@@ -211,12 +210,10 @@ ulp_bs_push_msb(uint8_t *bs, uint16_t pos, uint8_t len, uint8_t *val)
 	ulp_bs_put_msb(bs, pos, tmp, val[0]);
 
 	pos += tmp;
-	tlen -= tmp;
 
 	for (i = 1; i < cnt; i++) {
 		ulp_bs_put_msb(bs, pos, 8, val[i]);
 		pos += 8;
-		tlen -= 8;
 	}
 
 	return len;
diff --git a/drivers/net/hinic/base/hinic_pmd_mgmt.c b/drivers/net/hinic/base/hinic_pmd_mgmt.c
index 9b399502de..6ecca407a1 100644
--- a/drivers/net/hinic/base/hinic_pmd_mgmt.c
+++ b/drivers/net/hinic/base/hinic_pmd_mgmt.c
@@ -133,16 +133,12 @@ static void prepare_header(struct hinic_msg_pf_to_mgmt *pf_to_mgmt,
 static void prepare_mgmt_cmd(u8 *mgmt_cmd, u64 *header, void *msg,
 			     int msg_len)
 {
-	u32 cmd_buf_max = MAX_PF_MGMT_BUF_SIZE;
-
 	memset(mgmt_cmd, 0, MGMT_MSG_RSVD_FOR_DEV);
 
 	mgmt_cmd += MGMT_MSG_RSVD_FOR_DEV;
-	cmd_buf_max -= MGMT_MSG_RSVD_FOR_DEV;
 	memcpy(mgmt_cmd, header, sizeof(*header));
 
 	mgmt_cmd += sizeof(*header);
-	cmd_buf_max -= sizeof(*header);
 	memcpy(mgmt_cmd, msg, msg_len);
 }
 
@@ -615,7 +611,6 @@ static int recv_mgmt_msg_handler(struct hinic_msg_pf_to_mgmt *pf_to_mgmt,
 	void *msg_body = header + sizeof(msg_header);
 	u8 *dest_msg;
 	u8 seq_id, seq_len;
-	u32 msg_buf_max = MAX_PF_MGMT_BUF_SIZE;
 	u8 front_id;
 	u16 msg_id;
 
@@ -635,7 +630,6 @@ static int recv_mgmt_msg_handler(struct hinic_msg_pf_to_mgmt *pf_to_mgmt,
 	}
 
 	dest_msg = (u8 *)recv_msg->msg + seq_id * HINIC_MSG_SEG_LEN;
-	msg_buf_max -= seq_id * HINIC_MSG_SEG_LEN;
 	memcpy(dest_msg, msg_body, seq_len);
 
 	if (!HINIC_MSG_HEADER_GET(msg_header, LAST))
diff --git a/drivers/net/liquidio/lio_rxtx.c b/drivers/net/liquidio/lio_rxtx.c
index ef127f26c4..e09798ddd7 100644
--- a/drivers/net/liquidio/lio_rxtx.c
+++ b/drivers/net/liquidio/lio_rxtx.c
@@ -1050,7 +1050,6 @@ lio_update_read_index(struct lio_instr_queue *iq)
 int
 lio_flush_iq(struct lio_device *lio_dev, struct lio_instr_queue *iq)
 {
-	uint32_t tot_inst_processed = 0;
 	uint32_t inst_processed = 0;
 	int tx_done = 1;
 
@@ -1073,7 +1072,6 @@ lio_flush_iq(struct lio_device *lio_dev, struct lio_instr_queue *iq)
 			iq->stats.instr_processed += inst_processed;
 		}
 
-		tot_inst_processed += inst_processed;
 		inst_processed = 0;
 
 	} while (1);
diff --git a/drivers/net/nfp/nfp_cpp_bridge.c b/drivers/net/nfp/nfp_cpp_bridge.c
index 74a0eacb3f..0922ea9cf9 100644
--- a/drivers/net/nfp/nfp_cpp_bridge.c
+++ b/drivers/net/nfp/nfp_cpp_bridge.c
@@ -58,7 +58,7 @@ nfp_cpp_bridge_serve_write(int sockfd, struct nfp_cpp *cpp)
 	off_t offset, nfp_offset;
 	uint32_t cpp_id, pos, len;
 	uint32_t tmpbuf[16];
-	size_t count, curlen, totlen = 0;
+	size_t count, curlen;
 	int err = 0;
 
 	PMD_CPP_LOG(DEBUG, "%s: offset size %zu, count_size: %zu\n", __func__,
@@ -135,7 +135,6 @@ nfp_cpp_bridge_serve_write(int sockfd, struct nfp_cpp *cpp)
 		}
 
 		nfp_offset += pos;
-		totlen += pos;
 		nfp_cpp_area_release(area);
 		nfp_cpp_area_free(area);
 
@@ -160,7 +159,7 @@ nfp_cpp_bridge_serve_read(int sockfd, struct nfp_cpp *cpp)
 	off_t offset, nfp_offset;
 	uint32_t cpp_id, pos, len;
 	uint32_t tmpbuf[16];
-	size_t count, curlen, totlen = 0;
+	size_t count, curlen;
 	int err = 0;
 
 	PMD_CPP_LOG(DEBUG, "%s: offset size %zu, count_size: %zu\n", __func__,
@@ -236,7 +235,6 @@ nfp_cpp_bridge_serve_read(int sockfd, struct nfp_cpp *cpp)
 		}
 
 		nfp_offset += pos;
-		totlen += pos;
 		nfp_cpp_area_release(area);
 		nfp_cpp_area_free(area);
 
diff --git a/drivers/net/octeontx/octeontx_rxtx.h b/drivers/net/octeontx/octeontx_rxtx.h
index 56f11d3dd5..b63a5edfb3 100644
--- a/drivers/net/octeontx/octeontx_rxtx.h
+++ b/drivers/net/octeontx/octeontx_rxtx.h
@@ -365,7 +365,7 @@ __octeontx_xmit_mseg_prepare(struct rte_mbuf *tx_pkt, uint64_t *cmd_buf,
 			const uint16_t flag)
 {
 	uint16_t nb_segs, nb_desc = 0;
-	uint16_t gaura_id, len = 0;
+	uint16_t gaura_id;
 	struct rte_mbuf *m_next = NULL, *m_tofree;
 	rte_iova_t iova;
 	uint16_t data_len;
@@ -425,7 +425,6 @@ __octeontx_xmit_mseg_prepare(struct rte_mbuf *tx_pkt, uint64_t *cmd_buf,
 		cmd_buf[nb_desc++] = iova;
 
 		nb_segs--;
-		len += data_len;
 		tx_pkt = m_next;
 	} while (nb_segs);
 
diff --git a/drivers/net/qede/base/ecore_vf.c b/drivers/net/qede/base/ecore_vf.c
index db03bc494f..a36ae47c2b 100644
--- a/drivers/net/qede/base/ecore_vf.c
+++ b/drivers/net/qede/base/ecore_vf.c
@@ -73,7 +73,7 @@ static void ecore_vf_pf_req_end(struct ecore_hwfn *p_hwfn,
 #endif
 static enum _ecore_status_t
 ecore_send_msg2pf(struct ecore_hwfn *p_hwfn,
-		  u8 *done, u32 resp_size)
+		  u8 *done, __rte_unused u32 resp_size)
 {
 	union vfpf_tlvs *p_req = p_hwfn->vf_iov_info->vf2pf_request;
 	struct ustorm_trigger_vf_zone trigger;
@@ -86,9 +86,6 @@ ecore_send_msg2pf(struct ecore_hwfn *p_hwfn,
 	/* output tlvs list */
 	ecore_dp_tlv_list(p_hwfn, p_req);
 
-	/* need to add the END TLV to the message size */
-	resp_size += sizeof(struct channel_list_end_tlv);
-
 	/* Send TLVs over HW channel */
 	OSAL_MEMSET(&trigger, 0, sizeof(struct ustorm_trigger_vf_zone));
 	trigger.vf_pf_msg_valid = 1;
diff --git a/drivers/net/txgbe/txgbe_ethdev_vf.c b/drivers/net/txgbe/txgbe_ethdev_vf.c
index 67ae69dec3..8eb6fdd3c9 100644
--- a/drivers/net/txgbe/txgbe_ethdev_vf.c
+++ b/drivers/net/txgbe/txgbe_ethdev_vf.c
@@ -1291,6 +1291,9 @@ txgbevf_dev_interrupt_get_status(struct rte_eth_dev *dev)
 	/* Workround for ICR lost */
 	intr->flags |= TXGBE_FLAG_MAILBOX;
 
+	/* To avoid compiler warnings set eicr to used. */
+	RTE_SET_USED(eicr);
+
 	return 0;
 }
 
diff --git a/drivers/net/vmxnet3/vmxnet3_rxtx.c b/drivers/net/vmxnet3/vmxnet3_rxtx.c
index deba64be6a..77a2917137 100644
--- a/drivers/net/vmxnet3/vmxnet3_rxtx.c
+++ b/drivers/net/vmxnet3/vmxnet3_rxtx.c
@@ -340,6 +340,9 @@ vmxnet3_tq_tx_complete(vmxnet3_tx_queue_t *txq)
 	}
 
 	PMD_TX_LOG(DEBUG, "Processed %d tx comps & command descs.", completed);
+
+	/* To avoid compiler warnings when not in DEBUG mode set completed to used. */
+	RTE_SET_USED(completed);
 }
 
 uint16_t
-- 
2.25.1

Re: [PATCH 5/6] drivers/net: fix unused but set variables

[David Marchand]

On Mon, Nov 15, 2021 at 6:59 PM Conor Walsh <conor.walsh@intel.com> wrote:
>
> This patch fixes unused but set variables in the following net drivers:
>  - bnxt
>  - hinic
>  - liquidio
>  - nfp
>  - octeontx
>  - qede
>  - txgbe
>  - vmxnet3
>
> Bugzilla ID: 881
> Fixes: b87abb2e55cb ("net/bnxt: support marking packet")
> Fixes: 88badb3aef94 ("net/bnxt: add helper functions for blob/regfile ops")
> Fixes: a4957d87e05a ("net/hinic/base: add mgmt module")
> Fixes: 5ee7640f95a0 ("net/liquidio: add API to flush IQ")
> Fixes: 19af5a38f7b6 ("net/nfp: move CPP bridge to separate file")

This change only moves code, it should be:
Fixes: 29a62d1476b6 ("net/nfp: add CPP bridge as service")

I will fix while applying.


> Fixes: 9eb5cb3b11cc ("net/octeontx: fix access to indirect buffers")
> Fixes: 86a2265e59d7 ("qede: add SRIOV support")
> Fixes: b7311360fb67 ("net/txgbe: support VF interrupt")
> Fixes: c3ecdbb376da ("vmxnet3: support TSO")

Putting everything in a single patch will make life a bit harder for
backports: we have candidates for backports in 19.11 and 20.11, and
some patches are not relevant to backport.
I'll split this patch per driver since the Fixes: tags for each driver
only affect one version.


> Cc: stable@dpdk.org
>
> Reported-by: Liang Longfeng <longfengx.liang@intel.com>
> Signed-off-by: Conor Walsh <conor.walsh@intel.com>

Reviewed-by: David Marchand <david.marchand@redhat.com>


-- 
David Marchand

[PATCH 6/6] examples/performance-thread: fix unused but set variables

[Conor Walsh]

This patch removes the hits variable from the cpu_load_collector function
within the performance thread example app as it is an unused but set
variable.

Bugzilla ID: 881
Fixes: d48415e1fee3 ("examples/performance-thread: add l3fwd-thread app")
Cc: stable@dpdk.org

Reported-by: Liang Longfeng <longfengx.liang@intel.com>
Signed-off-by: Conor Walsh <conor.walsh@intel.com>
---
 examples/performance-thread/l3fwd-thread/main.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/examples/performance-thread/l3fwd-thread/main.c b/examples/performance-thread/l3fwd-thread/main.c
index 3a6a33bda3..8a35040597 100644
--- a/examples/performance-thread/l3fwd-thread/main.c
+++ b/examples/performance-thread/l3fwd-thread/main.c
@@ -1885,7 +1885,6 @@ process_burst(struct rte_mbuf *pkts_burst[MAX_PKT_BURST], int nb_rx,
 static int __rte_noreturn
 cpu_load_collector(__rte_unused void *arg) {
 	unsigned i, j, k;
-	uint64_t hits;
 	uint64_t prev_tsc, diff_tsc, cur_tsc;
 	uint64_t total[MAX_CPU] = { 0 };
 	unsigned min_cpu = MAX_CPU;
@@ -1972,12 +1971,10 @@ cpu_load_collector(__rte_unused void *arg) {
 			printf("cpu#     proc%%  poll%%  overhead%%\n\n");
 
 			for (i = min_cpu; i <= max_cpu; i++) {
-				hits = 0;
 				printf("CPU %d:", i);
 				for (j = 0; j < MAX_CPU_COUNTER; j++) {
 					printf("%7" PRIu64 "",
 							cpu_load.hits[j][i] * 100 / cpu_load.counter);
-					hits += cpu_load.hits[j][i];
 					cpu_load.hits[j][i] = 0;
 				}
 				printf("%7" PRIu64 "\n",
-- 
2.25.1

Re: [PATCH 0/6] fix unused but set variables

[David Marchand]

n Mon, Nov 15, 2021 at 6:59 PM Conor Walsh <conor.walsh@intel.com> wrote:
>
> When DPDK is compiled with clang 13, several warnings for unused but set
> variables are present.
>
> This patchset fixes these warnings, which were present within:
>  - app/test
>  - drivers/bus/fslmc
>  - drivers/common/qat
>  - drivers/event/sw
>  - drivers/net
>  - examples/performance-thread
>
> This patchset is split into 6 patches grouped broadly by subdir, as this
> set touches code from a lot of different authors this should make it
> easier to review.

For the series,
Reviewed-by: David Marchand <david.marchand@redhat.com>

I updated titles and did some patch splitting to simplify backports.

This has been tested with a F35 container in GHA (I'll send the patch for this):
https://github.com/david-marchand/dpdk/runs/4224438074

Series applied, thanks Conor.


-- 
David Marchand