Re: [RFC PATCH v3 0/7] RDMA/rxe: Add RDMA FLUSH operation

From: Jason Gunthorpe <jgg@nvidia.com>
To: Li Zhijian <lizhijian@fujitsu.com>
Cc: linux-rdma@vger.kernel.org, zyjzyj2000@gmail.com,
	aharonl@nvidia.com, leon@kernel.org, tom@talpey.com,
	tomasz.gromadzki@intel.com, linux-kernel@vger.kernel.org,
	mbloch@nvidia.com, liangwenpeng@huawei.com, yangx.jy@fujitsu.com,
	y-goto@fujitsu.com, rpearsonhpe@gmail.com,
	dan.j.williams@intel.com
Subject: Re: [RFC PATCH v3 0/7] RDMA/rxe: Add RDMA FLUSH operation
Date: Mon, 4 Jul 2022 10:26:23 -0300	[thread overview]
Message-ID: <20220704132623.GA1420657@nvidia.com> (raw)
In-Reply-To: <20220315101845.4166983-1-lizhijian@fujitsu.com>

On Tue, Mar 15, 2022 at 06:18:38PM +0800, Li Zhijian wrote:
> Hey folks,
> 
> These patches are going to implement a *NEW* RDMA opcode "RDMA FLUSH".
> In IB SPEC 1.5[1], 2 new opcodes, ATOMIC WRITE and RDMA FLUSH were
> added in the MEMORY PLACEMENT EXTENSIONS section.
> 
> This patchset makes SoftRoCE support new RDMA FLUSH on RC and RD service.
> 
> You can verify the patchset by building and running the rdma_flush example[2].
> server:
> $ ./rdma_flush_server -s [server_address] -p [port_number]
> client:
> $ ./rdma_flush_client -s [server_address] -p [port_number]
> 
> - We introduce new packet format for FLUSH request.
> - We introduce FLUSH placement type attributes to HCA
> - We introduce FLUSH access flags that users are able to register with

So where are we on this? Are all the rxe regressions fixed now? It
doesn't apply so I'm dropping it from patchworks.

Thanks,
Jason