Cover Detail
Show a cover letter.
GET /api/covers/88176/?format=api
{ "id": 88176, "url": "http://patchwork.dpdk.org/api/covers/88176/?format=api", "web_url": "http://patchwork.dpdk.org/project/dpdk/cover/20210224212018.17576-1-honnappa.nagarahalli@arm.com/", "project": { "id": 1, "url": "http://patchwork.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": "<20210224212018.17576-1-honnappa.nagarahalli@arm.com>", "list_archive_url": "https://inbox.dpdk.org/dev/20210224212018.17576-1-honnappa.nagarahalli@arm.com", "date": "2021-02-24T21:20:13", "name": "[RFC,0/5] Use correct memory ordering in eal functions", "submitter": { "id": 1045, "url": "http://patchwork.dpdk.org/api/people/1045/?format=api", "name": "Honnappa Nagarahalli", "email": "honnappa.nagarahalli@arm.com" }, "mbox": "http://patchwork.dpdk.org/project/dpdk/cover/20210224212018.17576-1-honnappa.nagarahalli@arm.com/mbox/", "series": [ { "id": 15367, "url": "http://patchwork.dpdk.org/api/series/15367/?format=api", "web_url": "http://patchwork.dpdk.org/project/dpdk/list/?series=15367", "date": "2021-02-24T21:20:13", "name": "Use correct memory ordering in eal functions", "version": 1, "mbox": "http://patchwork.dpdk.org/series/15367/mbox/" } ], "comments": "http://patchwork.dpdk.org/api/covers/88176/comments/", "headers": { "Return-Path": "<dev-bounces@dpdk.org>", "X-Original-To": "patchwork@inbox.dpdk.org", "Delivered-To": "patchwork@inbox.dpdk.org", "Received": [ "from mails.dpdk.org (mails.dpdk.org [217.70.189.124])\n\tby inbox.dpdk.org (Postfix) with ESMTP id 32F44A034F;\n\tWed, 24 Feb 2021 22:20:30 +0100 (CET)", "from [217.70.189.124] (localhost [127.0.0.1])\n\tby mails.dpdk.org (Postfix) with ESMTP id E732C40042;\n\tWed, 24 Feb 2021 22:20:29 +0100 (CET)", "from foss.arm.com (foss.arm.com [217.140.110.172])\n by mails.dpdk.org (Postfix) with ESMTP id 195E840040\n for <dev@dpdk.org>; Wed, 24 Feb 2021 22:20:28 +0100 (CET)", "from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14])\n by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 4F997ED1;\n Wed, 24 Feb 2021 13:20:27 -0800 (PST)", "from qc2400f-1.austin.arm.com (qc2400f-1.austin.arm.com\n [10.118.12.44])\n by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id 45DAC3F73D;\n Wed, 24 Feb 2021 13:20:27 -0800 (PST)" ], "From": "Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>", "To": "", "Cc": "dev@dpdk.org, honnappa.nagarahalli@arm.com, ruifeng.wang@arm.com,\n feifei.wang@arm.com, nd@arm.com", "Date": "Wed, 24 Feb 2021 15:20:13 -0600", "Message-Id": "<20210224212018.17576-1-honnappa.nagarahalli@arm.com>", "X-Mailer": "git-send-email 2.17.1", "Subject": "[dpdk-dev] [RFC 0/5] Use correct memory ordering in eal functions", "X-BeenThere": "dev@dpdk.org", "X-Mailman-Version": "2.1.29", "Precedence": "list", "List-Id": "DPDK patches and discussions <dev.dpdk.org>", "List-Unsubscribe": "<https://mails.dpdk.org/options/dev>,\n <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 <mailto:dev-request@dpdk.org?subject=subscribe>", "Errors-To": "dev-bounces@dpdk.org", "Sender": "\"dev\" <dev-bounces@dpdk.org>" }, "content": "rte_eal_remote_launch and rte_eal_wait_lcore need to provide\ncorrect memory ordering to address the data communication from\nmain core to worker core and vice versa.\n\nThere are 2 use cases:\n1) All the store operations (meant for worker) by main core\nshould be visible to worker core before the worker core runs the\nassigned function\n\n2) All the store operations by the worker core should be visible\nto the main core after rte_eal_wait_lcore returns.\n\nFor the data that needs to be communicated to the worker after\nthe rte_eal_remote_launch returns, load-acquire and store-release\nsemantics should be used (just like any other writer-reader use case).\n\nFor the main to worker communication, the pointer to function\nto execute is used as the guard variable. Hence, resetting of\nthe function pointer is important.\n\nFor the worker to main communication, the existing code uses the\nlcore state as the guard variable. However, it looks like\nthe FINISHED state is not really required. Hence the FINISHED state\nis removed before using the state as the guard variable. I would like\nsome feedback on why the FINISHED state is required. I have not\npaid attention to what it means for backward compatibility.\nIf it is decided to remove this state, documentation changes are\nrequired.\n\n\nHonnappa Nagarahalli (5):\n eal: reset lcore function pointer and argument\n eal: ensure memory operations are visible to worker\n eal: lcore state FINISHED is not required\n eal: ensure memory operations are visible to main\n test/ring: use relaxed barriers for ring stress test\n\n app/test/test_ring_stress_impl.h | 18 ++++-----\n drivers/event/dpaa2/dpaa2_eventdev_selftest.c | 2 +-\n drivers/event/octeontx/ssovf_evdev_selftest.c | 2 +-\n drivers/event/sw/sw_evdev_selftest.c | 4 +-\n examples/l2fwd-keepalive/main.c | 2 +-\n lib/librte_eal/common/eal_common_launch.c | 13 +++---\n lib/librte_eal/freebsd/eal_thread.c | 31 +++++++++++---\n lib/librte_eal/linux/eal_thread.c | 40 +++++++++++++------\n lib/librte_eal/windows/eal_thread.c | 34 +++++++++++-----\n 9 files changed, 94 insertions(+), 52 deletions(-)" }