From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 6C671460D7; Tue, 21 Jan 2025 23:36:28 +0100 (CET) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 422B84060F; Tue, 21 Jan 2025 23:36:19 +0100 (CET) Received: from linux.microsoft.com (linux.microsoft.com [13.77.154.182]) by mails.dpdk.org (Postfix) with ESMTP id A46D2402D6 for ; Tue, 21 Jan 2025 23:36:15 +0100 (CET) Received: by linux.microsoft.com (Postfix, from userid 1213) id AB7F4205A9F0; Tue, 21 Jan 2025 14:36:14 -0800 (PST) DKIM-Filter: OpenDKIM Filter v2.11.0 linux.microsoft.com AB7F4205A9F0 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linux.microsoft.com; s=default; t=1737498974; bh=KkQCemj56ElOYryqxqgxiTmJs1wME0HB1+m5Fmwj9E8=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=mBtiVw+HSVZqOTJl5HaCt/Q/JScqqQFlL2Eg0suvUGF63AkZq3U/eL4PMQAU5YoR2 LJwWz9B3mM8wiL4nK0F736MjeQ58LPxtTdZLCb6lMTEcpbK0lrE0MAuD6k0EkmzKrr fQbBiQcpDXyVhxRij5tEwvV6lqkqtFRbz58Vfdys= From: Andre Muezerie To: andremue@linux.microsoft.com Cc: dev@dpdk.org, stephen@networkplumber.org, bruce.richardson@intel.com Subject: [PATCH v16 2/3] drivers/common: add diagnostics macros to make code portable Date: Tue, 21 Jan 2025 14:36:09 -0800 Message-Id: <1737498970-4631-3-git-send-email-andremue@linux.microsoft.com> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: <1737498970-4631-1-git-send-email-andremue@linux.microsoft.com> References: <1735263196-2809-1-git-send-email-andremue@linux.microsoft.com> <1737498970-4631-1-git-send-email-andremue@linux.microsoft.com> X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org It was a common pattern to have "GCC diagnostic ignored" pragmas sprinkled over the code and only activate these pragmas for certain compilers (gcc and clang). Clang supports GCC's pragma for compatibility with existing source code, so #pragma GCC diagnostic and #pragma clang diagnostic are synonyms for Clang (https://clang.llvm.org/docs/UsersManual.html). Now that effort is being made to make the code compatible with MSVC these expressions would become more complex. It makes sense to hide this complexity behind macros. This makes maintenance easier as these macros are defined in a single place. As a plus the code becomes more readable as well. Signed-off-by: Andre Muezerie --- drivers/common/idpf/idpf_common_rxtx_avx512.c | 72 +++++++++---------- 1 file changed, 34 insertions(+), 38 deletions(-) diff --git a/drivers/common/idpf/idpf_common_rxtx_avx512.c b/drivers/common/idpf/idpf_common_rxtx_avx512.c index b8450b03ae..81052e72c1 100644 --- a/drivers/common/idpf/idpf_common_rxtx_avx512.c +++ b/drivers/common/idpf/idpf_common_rxtx_avx512.c @@ -6,10 +6,6 @@ #include "idpf_common_device.h" #include "idpf_common_rxtx.h" -#ifndef __INTEL_COMPILER -#pragma GCC diagnostic ignored "-Wcast-qual" -#endif - #define IDPF_DESCS_PER_LOOP_AVX 8 #define PKTLEN_SHIFT 10 @@ -34,7 +30,7 @@ idpf_singleq_rearm_common(struct idpf_rx_queue *rxq) dma_addr0 = _mm_setzero_si128(); for (i = 0; i < IDPF_VPMD_DESCS_PER_LOOP; i++) { rxp[i] = &rxq->fake_mbuf; - _mm_store_si128((__m128i *)&rxdp[i].read, + _mm_store_si128(RTE_CAST_PTR(__m128i *, &rxdp[i].read), dma_addr0); } } @@ -108,8 +104,8 @@ idpf_singleq_rearm_common(struct idpf_rx_queue *rxq) dma_addr4_7 = _mm512_add_epi64(dma_addr4_7, hdr_room); /* flush desc with pa dma_addr */ - _mm512_store_si512((__m512i *)&rxdp->read, dma_addr0_3); - _mm512_store_si512((__m512i *)&(rxdp + 4)->read, dma_addr4_7); + _mm512_store_si512(RTE_CAST_PTR(__m512i *, &rxdp->read), dma_addr0_3); + _mm512_store_si512(RTE_CAST_PTR(__m512i *, &(rxdp + 4)->read), dma_addr4_7); } rxq->rxrearm_start += IDPF_RXQ_REARM_THRESH; @@ -164,8 +160,8 @@ idpf_singleq_rearm(struct idpf_rx_queue *rxq) dma_addr0 = _mm_setzero_si128(); for (i = 0; i < IDPF_VPMD_DESCS_PER_LOOP; i++) { rxp[i] = &rxq->fake_mbuf; - _mm_storeu_si128((__m128i *)&rxdp[i].read, - dma_addr0); + _mm_storeu_si128(RTE_CAST_PTR + (__m128i *, &rxdp[i].read), dma_addr0); } } rte_atomic_fetch_add_explicit(&rxq->rx_stats.mbuf_alloc_failed, @@ -216,10 +212,10 @@ idpf_singleq_rearm(struct idpf_rx_queue *rxq) iovas1); const __m512i desc6_7 = _mm512_bsrli_epi128(desc4_5, 8); - _mm512_storeu_si512((void *)rxdp, desc0_1); - _mm512_storeu_si512((void *)(rxdp + 2), desc2_3); - _mm512_storeu_si512((void *)(rxdp + 4), desc4_5); - _mm512_storeu_si512((void *)(rxdp + 6), desc6_7); + _mm512_storeu_si512(RTE_CAST_PTR(void *, rxdp), desc0_1); + _mm512_storeu_si512(RTE_CAST_PTR(void *, (rxdp + 2)), desc2_3); + _mm512_storeu_si512(RTE_CAST_PTR(void *, (rxdp + 4)), desc4_5); + _mm512_storeu_si512(RTE_CAST_PTR(void *, (rxdp + 6)), desc6_7); rxp += IDPF_DESCS_PER_LOOP_AVX; rxdp += IDPF_DESCS_PER_LOOP_AVX; @@ -337,28 +333,28 @@ _idpf_singleq_recv_raw_pkts_avx512(struct idpf_rx_queue *rxq, __m512i raw_desc0_3, raw_desc4_7; const __m128i raw_desc7 = - _mm_load_si128((void *)(rxdp + 7)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 7)); rte_compiler_barrier(); const __m128i raw_desc6 = - _mm_load_si128((void *)(rxdp + 6)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 6)); rte_compiler_barrier(); const __m128i raw_desc5 = - _mm_load_si128((void *)(rxdp + 5)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 5)); rte_compiler_barrier(); const __m128i raw_desc4 = - _mm_load_si128((void *)(rxdp + 4)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 4)); rte_compiler_barrier(); const __m128i raw_desc3 = - _mm_load_si128((void *)(rxdp + 3)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 3)); rte_compiler_barrier(); const __m128i raw_desc2 = - _mm_load_si128((void *)(rxdp + 2)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 2)); rte_compiler_barrier(); const __m128i raw_desc1 = - _mm_load_si128((void *)(rxdp + 1)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 1)); rte_compiler_barrier(); const __m128i raw_desc0 = - _mm_load_si128((void *)(rxdp + 0)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 0)); raw_desc4_7 = _mm512_broadcast_i32x4(raw_desc4); raw_desc4_7 = _mm512_inserti32x4(raw_desc4_7, raw_desc5, 1); @@ -560,7 +556,7 @@ idpf_splitq_rearm_common(struct idpf_rx_queue *rx_bufq) dma_addr0 = _mm_setzero_si128(); for (i = 0; i < IDPF_VPMD_DESCS_PER_LOOP; i++) { rxp[i] = &rx_bufq->fake_mbuf; - _mm_store_si128((__m128i *)&rxdp[i], + _mm_store_si128(RTE_CAST_PTR(__m128i *, &rxdp[i]), dma_addr0); } } @@ -634,7 +630,7 @@ idpf_splitq_rearm(struct idpf_rx_queue *rx_bufq) dma_addr0 = _mm_setzero_si128(); for (i = 0; i < IDPF_VPMD_DESCS_PER_LOOP; i++) { rxp[i] = &rx_bufq->fake_mbuf; - _mm_storeu_si128((__m128i *)&rxdp[i], + _mm_storeu_si128(RTE_CAST_PTR(__m128i *, &rxdp[i]), dma_addr0); } } @@ -798,28 +794,28 @@ _idpf_splitq_recv_raw_pkts_avx512(struct idpf_rx_queue *rxq, __m512i raw_desc0_3, raw_desc4_7; const __m128i raw_desc7 = - _mm_load_si128((void *)(rxdp + 7)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 7)); rte_compiler_barrier(); const __m128i raw_desc6 = - _mm_load_si128((void *)(rxdp + 6)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 6)); rte_compiler_barrier(); const __m128i raw_desc5 = - _mm_load_si128((void *)(rxdp + 5)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 5)); rte_compiler_barrier(); const __m128i raw_desc4 = - _mm_load_si128((void *)(rxdp + 4)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 4)); rte_compiler_barrier(); const __m128i raw_desc3 = - _mm_load_si128((void *)(rxdp + 3)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 3)); rte_compiler_barrier(); const __m128i raw_desc2 = - _mm_load_si128((void *)(rxdp + 2)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 2)); rte_compiler_barrier(); const __m128i raw_desc1 = - _mm_load_si128((void *)(rxdp + 1)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 1)); rte_compiler_barrier(); const __m128i raw_desc0 = - _mm_load_si128((void *)(rxdp + 0)); + _mm_load_si128(RTE_CAST_PTR(const __m128i *, rxdp + 0)); raw_desc4_7 = _mm512_broadcast_i32x4(raw_desc4); raw_desc4_7 = _mm512_inserti32x4(raw_desc4_7, raw_desc5, 1); @@ -1131,7 +1127,7 @@ idpf_singleq_vtx1(volatile struct idpf_base_tx_desc *txdp, __m128i descriptor = _mm_set_epi64x(high_qw, pkt->buf_iova + pkt->data_off); - _mm_storeu_si128((__m128i *)txdp, descriptor); + _mm_storeu_si128(RTE_CAST_PTR(__m128i *, txdp), descriptor); } #define IDPF_TX_LEN_MASK 0xAA @@ -1178,7 +1174,7 @@ idpf_singleq_vtx(volatile struct idpf_base_tx_desc *txdp, pkt[1]->buf_iova + pkt[1]->data_off, hi_qw0, pkt[0]->buf_iova + pkt[0]->data_off); - _mm512_storeu_si512((void *)txdp, desc0_3); + _mm512_storeu_si512(RTE_CAST_PTR(void *, txdp), desc0_3); } /* do any last ones */ @@ -1435,7 +1431,7 @@ idpf_splitq_vtx1(volatile struct idpf_flex_tx_sched_desc *txdp, __m128i descriptor = _mm_set_epi64x(high_qw, pkt->buf_iova + pkt->data_off); - _mm_storeu_si128((__m128i *)txdp, descriptor); + _mm_storeu_si128(RTE_CAST_PTR(__m128i *, txdp), descriptor); } static __rte_always_inline void @@ -1480,7 +1476,7 @@ idpf_splitq_vtx(volatile struct idpf_flex_tx_sched_desc *txdp, pkt[1]->buf_iova + pkt[1]->data_off, hi_qw0, pkt[0]->buf_iova + pkt[0]->data_off); - _mm512_storeu_si512((void *)txdp, desc0_3); + _mm512_storeu_si512(RTE_CAST_PTR(void *, txdp), desc0_3); } /* do any last ones */ @@ -1521,11 +1517,11 @@ idpf_splitq_xmit_fixed_burst_vec_avx512(void *tx_queue, struct rte_mbuf **tx_pkt if (nb_commit >= n) { tx_backlog_entry_avx512(txep, tx_pkts, n); - idpf_splitq_vtx((void *)txdp, tx_pkts, n - 1, cmd_dtype); + idpf_splitq_vtx(txdp, tx_pkts, n - 1, cmd_dtype); tx_pkts += (n - 1); txdp += (n - 1); - idpf_splitq_vtx1((void *)txdp, *tx_pkts++, cmd_dtype); + idpf_splitq_vtx1(txdp, *tx_pkts++, cmd_dtype); nb_commit = (uint16_t)(nb_commit - n); @@ -1540,7 +1536,7 @@ idpf_splitq_xmit_fixed_burst_vec_avx512(void *tx_queue, struct rte_mbuf **tx_pkt tx_backlog_entry_avx512(txep, tx_pkts, nb_commit); - idpf_splitq_vtx((void *)txdp, tx_pkts, nb_commit, cmd_dtype); + idpf_splitq_vtx(txdp, tx_pkts, nb_commit, cmd_dtype); tx_id = (uint16_t)(tx_id + nb_commit); if (tx_id > txq->next_rs) -- 2.47.2.vfs.0.1