Patch Detail
get:
Show a patch.
patch:
Update a patch.
put:
Update a patch.
GET /api/patches/53996/?format=api
{ "id": 53996, "url": "http://patches.dpdk.org/api/patches/53996/?format=api", "web_url": "http://patches.dpdk.org/project/dpdk/patch/20190601014905.45531-26-jerinj@marvell.com/", "project": { "id": 1, "url": "http://patches.dpdk.org/api/projects/1/?format=api", "name": "DPDK", "link_name": "dpdk", "list_id": "dev.dpdk.org", "list_email": "dev@dpdk.org", "web_url": "http://core.dpdk.org", "scm_url": "git://dpdk.org/dpdk", "webscm_url": "http://git.dpdk.org/dpdk", "list_archive_url": "https://inbox.dpdk.org/dev", "list_archive_url_format": "https://inbox.dpdk.org/dev/{}", "commit_url_format": "" }, "msgid": "<20190601014905.45531-26-jerinj@marvell.com>", "list_archive_url": "https://inbox.dpdk.org/dev/20190601014905.45531-26-jerinj@marvell.com", "date": "2019-06-01T01:49:03", "name": "[v2,25/27] mempool/octeontx2: add optimized dequeue operation for arm64", "commit_ref": null, "pull_url": null, "state": "superseded", "archived": true, "hash": "9baf874475cb68dda5ed04e0cdfe21819e1f5a76", "submitter": { "id": 1188, "url": "http://patches.dpdk.org/api/people/1188/?format=api", "name": "Jerin Jacob Kollanukkaran", "email": "jerinj@marvell.com" }, "delegate": { "id": 1, "url": "http://patches.dpdk.org/api/users/1/?format=api", "username": "tmonjalo", "first_name": "Thomas", "last_name": "Monjalon", "email": "thomas@monjalon.net" }, "mbox": "http://patches.dpdk.org/project/dpdk/patch/20190601014905.45531-26-jerinj@marvell.com/mbox/", "series": [ { "id": 4844, "url": "http://patches.dpdk.org/api/series/4844/?format=api", "web_url": "http://patches.dpdk.org/project/dpdk/list/?series=4844", "date": "2019-06-01T01:48:38", "name": "OCTEON TX2 common and mempool driver", "version": 2, "mbox": "http://patches.dpdk.org/series/4844/mbox/" } ], "comments": "http://patches.dpdk.org/api/patches/53996/comments/", "check": "fail", "checks": "http://patches.dpdk.org/api/patches/53996/checks/", "tags": {}, "related": [], "headers": { "Return-Path": "<dev-bounces@dpdk.org>", "X-Original-To": "patchwork@dpdk.org", "Delivered-To": "patchwork@dpdk.org", "Received": [ "from [92.243.14.124] (localhost [127.0.0.1])\n\tby dpdk.org (Postfix) with ESMTP id 324EA1B95A;\n\tSat, 1 Jun 2019 03:50:35 +0200 (CEST)", "from mx0b-0016f401.pphosted.com (mx0b-0016f401.pphosted.com\n\t[67.231.156.173]) by dpdk.org (Postfix) with ESMTP id 147561B94D\n\tfor <dev@dpdk.org>; Sat, 1 Jun 2019 03:50:32 +0200 (CEST)", "from pps.filterd (m0045851.ppops.net [127.0.0.1])\n\tby mx0b-0016f401.pphosted.com (8.16.0.27/8.16.0.27) with SMTP id\n\tx511lJt1002642; Fri, 31 May 2019 18:50:31 -0700", "from sc-exch02.marvell.com ([199.233.58.182])\n\tby mx0b-0016f401.pphosted.com with ESMTP id 2su5xh2bxc-1\n\t(version=TLSv1.2 cipher=ECDHE-RSA-AES256-SHA384 bits=256 verify=NOT); \n\tFri, 31 May 2019 18:50:31 -0700", "from SC-EXCH03.marvell.com (10.93.176.83) by SC-EXCH02.marvell.com\n\t(10.93.176.82) with Microsoft SMTP Server (TLS) id 15.0.1367.3;\n\tFri, 31 May 2019 18:50:30 -0700", "from maili.marvell.com (10.93.176.43) by SC-EXCH03.marvell.com\n\t(10.93.176.83) with Microsoft SMTP Server id 15.0.1367.3 via Frontend\n\tTransport; Fri, 31 May 2019 18:50:30 -0700", "from jerin-lab.marvell.com (jerin-lab.marvell.com [10.28.34.14])\n\tby maili.marvell.com (Postfix) with ESMTP id 6ED1C3F703F;\n\tFri, 31 May 2019 18:50:28 -0700 (PDT)" ], "DKIM-Signature": "v=1; a=rsa-sha256; c=relaxed/relaxed; d=marvell.com;\n\th=from : to : cc :\n\tsubject : date : message-id : in-reply-to : references : mime-version\n\t: content-transfer-encoding : content-type; s=pfpt0818;\n\tbh=xTsuVHVMiZ8nUreHYyjSv/V/yfcDS7VpsKZIxG5/VV4=;\n\tb=Ukf7KEcjvLl1q4NujhqNk7hKHM6868mg9OKdwd71AU+HBYKTshDMbbzYNz6J5m9HR2LG\n\tz26U8RFvXD6wKor/zHGoyTWUXWZsTz51zkLBoqlTwlLCfr2icqnhgLLyLTdhTYX34XN5\n\t7Xly3/N1baoO8buq0Vi4gYU31954SLcRRUAsZ7BZVWrb7J0v63jubhIDqTSrWXd3yh53\n\tUmitBVXJSJpXpHkOKlv5acY4mk6R4m22PuncEtoW4u2G2R1OaBVkyzMciGaLICYIIegL\n\tY809P8AOMzUqDl4UW3YEk1IZ9pn9Hzr/OkUgBGL3W3OC1EYPc2wwz8m9e+UlJDAo4e/f\n\tuA== ", "From": "<jerinj@marvell.com>", "To": "<dev@dpdk.org>, Jerin Jacob <jerinj@marvell.com>, Nithin Dabilpuram\n\t<ndabilpuram@marvell.com>, Vamsi Attunuru <vattunuru@marvell.com>", "CC": "Pavan Nikhilesh <pbhagavatula@marvell.com>, Olivier Matz\n\t<olivier.matz@6wind.com>, Aaron Conole <aconole@redhat.com>", "Date": "Sat, 1 Jun 2019 07:19:03 +0530", "Message-ID": "<20190601014905.45531-26-jerinj@marvell.com>", "X-Mailer": "git-send-email 2.21.0", "In-Reply-To": "<20190601014905.45531-1-jerinj@marvell.com>", "References": "<20190523081339.56348-1-jerinj@marvell.com>\n\t<20190601014905.45531-1-jerinj@marvell.com>", "MIME-Version": "1.0", "Content-Transfer-Encoding": "8bit", "Content-Type": "text/plain", "X-Proofpoint-Virus-Version": "vendor=fsecure engine=2.50.10434:, ,\n\tdefinitions=2019-06-01_02:, , signatures=0", "Subject": "[dpdk-dev] [PATCH v2 25/27] mempool/octeontx2: add optimized\n\tdequeue operation for arm64", "X-BeenThere": "dev@dpdk.org", "X-Mailman-Version": "2.1.15", "Precedence": "list", "List-Id": "DPDK patches and discussions <dev.dpdk.org>", "List-Unsubscribe": "<https://mails.dpdk.org/options/dev>,\n\t<mailto:dev-request@dpdk.org?subject=unsubscribe>", "List-Archive": "<http://mails.dpdk.org/archives/dev/>", "List-Post": "<mailto:dev@dpdk.org>", "List-Help": "<mailto:dev-request@dpdk.org?subject=help>", "List-Subscribe": "<https://mails.dpdk.org/listinfo/dev>,\n\t<mailto:dev-request@dpdk.org?subject=subscribe>", "Errors-To": "dev-bounces@dpdk.org", "Sender": "\"dev\" <dev-bounces@dpdk.org>" }, "content": "From: Pavan Nikhilesh <pbhagavatula@marvell.com>\n\nThis patch adds an optimized arm64 instruction based routine to leverage\nCPU pipeline characteristics of octeontx2. The theme is to fill the\npipeline with CASP operations as much HW can do so that HW can do alloc()\nHW ops in full throttle.\n\nCc: Olivier Matz <olivier.matz@6wind.com>\nCc: Aaron Conole <aconole@redhat.com>\n\nSigned-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>\nSigned-off-by: Jerin Jacob <jerinj@marvell.com>\nSigned-off-by: Vamsi Attunuru <vattunuru@marvell.com>\n---\n drivers/mempool/octeontx2/otx2_mempool_ops.c | 291 +++++++++++++++++++\n 1 file changed, 291 insertions(+)", "diff": "diff --git a/drivers/mempool/octeontx2/otx2_mempool_ops.c b/drivers/mempool/octeontx2/otx2_mempool_ops.c\nindex c59bd73c0..e6737abda 100644\n--- a/drivers/mempool/octeontx2/otx2_mempool_ops.c\n+++ b/drivers/mempool/octeontx2/otx2_mempool_ops.c\n@@ -37,6 +37,293 @@ npa_lf_aura_op_alloc_one(const int64_t wdata, int64_t * const addr,\n \treturn -ENOENT;\n }\n \n+#if defined(RTE_ARCH_ARM64)\n+static __rte_noinline int\n+npa_lf_aura_op_search_alloc(const int64_t wdata, int64_t * const addr,\n+\t\tvoid **obj_table, unsigned int n)\n+{\n+\tuint8_t i;\n+\n+\tfor (i = 0; i < n; i++) {\n+\t\tif (obj_table[i] != NULL)\n+\t\t\tcontinue;\n+\t\tif (npa_lf_aura_op_alloc_one(wdata, addr, obj_table, i))\n+\t\t\treturn -ENOENT;\n+\t}\n+\n+\treturn 0;\n+}\n+\n+static __attribute__((optimize(\"-O3\"))) __rte_noinline int __hot\n+npa_lf_aura_op_alloc_bulk(const int64_t wdata, int64_t * const addr,\n+\t\t\t unsigned int n, void **obj_table)\n+{\n+\tconst __uint128_t wdata128 = ((__uint128_t)wdata << 64) | wdata;\n+\tuint64x2_t failed = vdupq_n_u64(~0);\n+\n+\tswitch (n) {\n+\tcase 32:\n+\t{\n+\t\t__uint128_t t0, t1, t2, t3, t4, t5, t6, t7, t8, t9;\n+\t\t__uint128_t t10, t11;\n+\n+\t\tasm volatile (\n+\t\t\".cpu generic+lse\\n\"\n+\t\t\"casp %[t0], %H[t0], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t1], %H[t1], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t2], %H[t2], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t3], %H[t3], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t4], %H[t4], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t5], %H[t5], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t6], %H[t6], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t7], %H[t7], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t8], %H[t8], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t9], %H[t9], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t10], %H[t10], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t11], %H[t11], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"fmov d16, %[t0]\\n\"\n+\t\t\"fmov v16.D[1], %H[t0]\\n\"\n+\t\t\"casp %[t0], %H[t0], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"fmov d17, %[t1]\\n\"\n+\t\t\"fmov v17.D[1], %H[t1]\\n\"\n+\t\t\"casp %[t1], %H[t1], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"fmov d18, %[t2]\\n\"\n+\t\t\"fmov v18.D[1], %H[t2]\\n\"\n+\t\t\"casp %[t2], %H[t2], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"fmov d19, %[t3]\\n\"\n+\t\t\"fmov v19.D[1], %H[t3]\\n\"\n+\t\t\"casp %[t3], %H[t3], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v16.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v17.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v18.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v19.16B\\n\"\n+\t\t\"fmov d20, %[t4]\\n\"\n+\t\t\"fmov v20.D[1], %H[t4]\\n\"\n+\t\t\"fmov d21, %[t5]\\n\"\n+\t\t\"fmov v21.D[1], %H[t5]\\n\"\n+\t\t\"fmov d22, %[t6]\\n\"\n+\t\t\"fmov v22.D[1], %H[t6]\\n\"\n+\t\t\"fmov d23, %[t7]\\n\"\n+\t\t\"fmov v23.D[1], %H[t7]\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v20.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v21.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v22.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v23.16B\\n\"\n+\t\t\"st1 { v16.2d, v17.2d, v18.2d, v19.2d}, [%[dst]], 64\\n\"\n+\t\t\"st1 { v20.2d, v21.2d, v22.2d, v23.2d}, [%[dst]], 64\\n\"\n+\t\t\"fmov d16, %[t8]\\n\"\n+\t\t\"fmov v16.D[1], %H[t8]\\n\"\n+\t\t\"fmov d17, %[t9]\\n\"\n+\t\t\"fmov v17.D[1], %H[t9]\\n\"\n+\t\t\"fmov d18, %[t10]\\n\"\n+\t\t\"fmov v18.D[1], %H[t10]\\n\"\n+\t\t\"fmov d19, %[t11]\\n\"\n+\t\t\"fmov v19.D[1], %H[t11]\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v16.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v17.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v18.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v19.16B\\n\"\n+\t\t\"fmov d20, %[t0]\\n\"\n+\t\t\"fmov v20.D[1], %H[t0]\\n\"\n+\t\t\"fmov d21, %[t1]\\n\"\n+\t\t\"fmov v21.D[1], %H[t1]\\n\"\n+\t\t\"fmov d22, %[t2]\\n\"\n+\t\t\"fmov v22.D[1], %H[t2]\\n\"\n+\t\t\"fmov d23, %[t3]\\n\"\n+\t\t\"fmov v23.D[1], %H[t3]\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v20.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v21.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v22.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v23.16B\\n\"\n+\t\t\"st1 { v16.2d, v17.2d, v18.2d, v19.2d}, [%[dst]], 64\\n\"\n+\t\t\"st1 { v20.2d, v21.2d, v22.2d, v23.2d}, [%[dst]], 64\\n\"\n+\t\t: \"+Q\" (*addr), [failed] \"=&w\" (failed),\n+\t\t[t0] \"=&r\" (t0), [t1] \"=&r\" (t1), [t2] \"=&r\" (t2),\n+\t\t[t3] \"=&r\" (t3), [t4] \"=&r\" (t4), [t5] \"=&r\" (t5),\n+\t\t[t6] \"=&r\" (t6), [t7] \"=&r\" (t7), [t8] \"=&r\" (t8),\n+\t\t[t9] \"=&r\" (t9), [t10] \"=&r\" (t10), [t11] \"=&r\" (t11)\n+\t\t: [wdata] \"r\" (wdata128), [dst] \"r\" (obj_table),\n+\t\t[loc] \"r\" (addr)\n+\t\t: \"memory\", \"v16\", \"v17\", \"v18\",\n+\t\t\"v19\", \"v20\", \"v21\", \"v22\", \"v23\"\n+\t\t);\n+\t\tbreak;\n+\t}\n+\tcase 16:\n+\t{\n+\t\t__uint128_t t0, t1, t2, t3, t4, t5, t6, t7;\n+\n+\t\tasm volatile (\n+\t\t\".cpu generic+lse\\n\"\n+\t\t\"casp %[t0], %H[t0], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t1], %H[t1], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t2], %H[t2], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t3], %H[t3], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t4], %H[t4], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t5], %H[t5], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t6], %H[t6], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t7], %H[t7], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"fmov d16, %[t0]\\n\"\n+\t\t\"fmov v16.D[1], %H[t0]\\n\"\n+\t\t\"fmov d17, %[t1]\\n\"\n+\t\t\"fmov v17.D[1], %H[t1]\\n\"\n+\t\t\"fmov d18, %[t2]\\n\"\n+\t\t\"fmov v18.D[1], %H[t2]\\n\"\n+\t\t\"fmov d19, %[t3]\\n\"\n+\t\t\"fmov v19.D[1], %H[t3]\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v16.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v17.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v18.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v19.16B\\n\"\n+\t\t\"fmov d20, %[t4]\\n\"\n+\t\t\"fmov v20.D[1], %H[t4]\\n\"\n+\t\t\"fmov d21, %[t5]\\n\"\n+\t\t\"fmov v21.D[1], %H[t5]\\n\"\n+\t\t\"fmov d22, %[t6]\\n\"\n+\t\t\"fmov v22.D[1], %H[t6]\\n\"\n+\t\t\"fmov d23, %[t7]\\n\"\n+\t\t\"fmov v23.D[1], %H[t7]\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v20.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v21.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v22.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v23.16B\\n\"\n+\t\t\"st1 { v16.2d, v17.2d, v18.2d, v19.2d}, [%[dst]], 64\\n\"\n+\t\t\"st1 { v20.2d, v21.2d, v22.2d, v23.2d}, [%[dst]], 64\\n\"\n+\t\t: \"+Q\" (*addr), [failed] \"=&w\" (failed),\n+\t\t[t0] \"=&r\" (t0), [t1] \"=&r\" (t1), [t2] \"=&r\" (t2),\n+\t\t[t3] \"=&r\" (t3), [t4] \"=&r\" (t4), [t5] \"=&r\" (t5),\n+\t\t[t6] \"=&r\" (t6), [t7] \"=&r\" (t7)\n+\t\t: [wdata] \"r\" (wdata128), [dst] \"r\" (obj_table),\n+\t\t[loc] \"r\" (addr)\n+\t\t: \"memory\", \"v16\", \"v17\", \"v18\", \"v19\",\n+\t\t \"v20\", \"v21\", \"v22\", \"v23\"\n+\t\t);\n+\t\tbreak;\n+\t}\n+\tcase 8:\n+\t{\n+\t\t__uint128_t t0, t1, t2, t3;\n+\n+\t\tasm volatile (\n+\t\t\".cpu generic+lse\\n\"\n+\t\t\"casp %[t0], %H[t0], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t1], %H[t1], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t2], %H[t2], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t3], %H[t3], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"fmov d16, %[t0]\\n\"\n+\t\t\"fmov v16.D[1], %H[t0]\\n\"\n+\t\t\"fmov d17, %[t1]\\n\"\n+\t\t\"fmov v17.D[1], %H[t1]\\n\"\n+\t\t\"fmov d18, %[t2]\\n\"\n+\t\t\"fmov v18.D[1], %H[t2]\\n\"\n+\t\t\"fmov d19, %[t3]\\n\"\n+\t\t\"fmov v19.D[1], %H[t3]\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v16.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v17.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v18.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v19.16B\\n\"\n+\t\t\"st1 { v16.2d, v17.2d, v18.2d, v19.2d}, [%[dst]], 64\\n\"\n+\t\t: \"+Q\" (*addr), [failed] \"=&w\" (failed),\n+\t\t[t0] \"=&r\" (t0), [t1] \"=&r\" (t1), [t2] \"=&r\" (t2),\n+\t\t[t3] \"=&r\" (t3)\n+\t\t: [wdata] \"r\" (wdata128), [dst] \"r\" (obj_table),\n+\t\t[loc] \"r\" (addr)\n+\t\t: \"memory\", \"v16\", \"v17\", \"v18\", \"v19\"\n+\t\t);\n+\t\tbreak;\n+\t}\n+\tcase 4:\n+\t{\n+\t\t__uint128_t t0, t1;\n+\n+\t\tasm volatile (\n+\t\t\".cpu generic+lse\\n\"\n+\t\t\"casp %[t0], %H[t0], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"casp %[t1], %H[t1], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"fmov d16, %[t0]\\n\"\n+\t\t\"fmov v16.D[1], %H[t0]\\n\"\n+\t\t\"fmov d17, %[t1]\\n\"\n+\t\t\"fmov v17.D[1], %H[t1]\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v16.16B\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v17.16B\\n\"\n+\t\t\"st1 { v16.2d, v17.2d}, [%[dst]], 32\\n\"\n+\t\t: \"+Q\" (*addr), [failed] \"=&w\" (failed),\n+\t\t[t0] \"=&r\" (t0), [t1] \"=&r\" (t1)\n+\t\t: [wdata] \"r\" (wdata128), [dst] \"r\" (obj_table),\n+\t\t[loc] \"r\" (addr)\n+\t\t: \"memory\", \"v16\", \"v17\"\n+\t\t);\n+\t\tbreak;\n+\t}\n+\tcase 2:\n+\t{\n+\t\t__uint128_t t0;\n+\n+\t\tasm volatile (\n+\t\t\".cpu generic+lse\\n\"\n+\t\t\"casp %[t0], %H[t0], %[wdata], %H[wdata], [%[loc]]\\n\"\n+\t\t\"fmov d16, %[t0]\\n\"\n+\t\t\"fmov v16.D[1], %H[t0]\\n\"\n+\t\t\"and %[failed].16B, %[failed].16B, v16.16B\\n\"\n+\t\t\"st1 { v16.2d}, [%[dst]], 16\\n\"\n+\t\t: \"+Q\" (*addr), [failed] \"=&w\" (failed),\n+\t\t[t0] \"=&r\" (t0)\n+\t\t: [wdata] \"r\" (wdata128), [dst] \"r\" (obj_table),\n+\t\t[loc] \"r\" (addr)\n+\t\t: \"memory\", \"v16\"\n+\t\t);\n+\t\tbreak;\n+\t}\n+\tcase 1:\n+\t\treturn npa_lf_aura_op_alloc_one(wdata, addr, obj_table, 0);\n+\t}\n+\n+\tif (unlikely(!(vgetq_lane_u64(failed, 0) & vgetq_lane_u64(failed, 1))))\n+\t\treturn npa_lf_aura_op_search_alloc(wdata, addr, (void **)\n+\t\t\t((char *)obj_table - (sizeof(uint64_t) * n)), n);\n+\n+\treturn 0;\n+}\n+\n+static __rte_noinline void\n+otx2_npa_clear_alloc(struct rte_mempool *mp, void **obj_table, unsigned int n)\n+{\n+\tunsigned int i;\n+\n+\tfor (i = 0; i < n; i++) {\n+\t\tif (obj_table[i] != NULL) {\n+\t\t\totx2_npa_enq(mp, &obj_table[i], 1);\n+\t\t\tobj_table[i] = NULL;\n+\t\t}\n+\t}\n+}\n+\n+static inline int __hot\n+otx2_npa_deq_arm64(struct rte_mempool *mp, void **obj_table, unsigned int n)\n+{\n+\tconst int64_t wdata = npa_lf_aura_handle_to_aura(mp->pool_id);\n+\tvoid **obj_table_bak = obj_table;\n+\tconst unsigned int nfree = n;\n+\tunsigned int parts;\n+\n+\tint64_t * const addr = (int64_t * const)\n+\t\t\t(npa_lf_aura_handle_to_base(mp->pool_id) +\n+\t\t\t\tNPA_LF_AURA_OP_ALLOCX(0));\n+\twhile (n) {\n+\t\tparts = n > 31 ? 32 : rte_align32prevpow2(n);\n+\t\tn -= parts;\n+\t\tif (unlikely(npa_lf_aura_op_alloc_bulk(wdata, addr,\n+\t\t\t\tparts, obj_table))) {\n+\t\t\totx2_npa_clear_alloc(mp, obj_table_bak, nfree - n);\n+\t\t\treturn -ENOENT;\n+\t\t}\n+\t\tobj_table += parts;\n+\t}\n+\n+\treturn 0;\n+}\n+#endif\n+\n static inline int __hot\n otx2_npa_deq(struct rte_mempool *mp, void **obj_table, unsigned int n)\n {\n@@ -463,7 +750,11 @@ static struct rte_mempool_ops otx2_npa_ops = {\n \t.get_count = otx2_npa_get_count,\n \t.calc_mem_size = otx2_npa_calc_mem_size,\n \t.populate = otx2_npa_populate,\n+#if defined(RTE_ARCH_ARM64)\n+\t.dequeue = otx2_npa_deq_arm64,\n+#else\n \t.dequeue = otx2_npa_deq,\n+#endif\n };\n \n MEMPOOL_REGISTER_OPS(otx2_npa_ops);\n", "prefixes": [ "v2", "25/27" ] }