[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240328154144.272275-1-gbayer@linux.ibm.com>
Date: Thu, 28 Mar 2024 16:41:43 +0100
From: Gerd Bayer <gbayer@...ux.ibm.com>
To: Wenjia Zhang <wenjia@...ux.ibm.com>, Wen Gu <guwen@...ux.alibaba.com>,
Heiko Carstens <hca@...ux.ibm.com>, pasic@...ux.ibm.com,
schnelle@...ux.ibm.com
Cc: linux-s390@...r.kernel.org, netdev@...r.kernel.org,
Alexandra Winter <wintera@...ux.ibm.com>,
Thorsten Winkler <twinkler@...ux.ibm.com>,
Vasily Gorbik <gor@...ux.ibm.com>,
Alexander Gordeev <agordeev@...ux.ibm.com>,
Christian Borntraeger <borntraeger@...ux.ibm.com>,
Sven Schnelle <svens@...ux.ibm.com>, Gerd Bayer <gbayer@...ux.ibm.com>
Subject: [PATCH net 0/1] s390/ism: Fix splice for SMC-D
Hi all,
due to a change in the DMA API - to no longer provide compound
pages with dma_alloc_coherent - splice broke for the SMC-D protocol.
Here, I'm proposing a fix for that.
I'm aware that this is a rather coarse fix attempt and a proper
solution would be to rework the SMC-D protocol to drop the requirement
for compound pages. That work might take some more time.
Meanwhile, I'd like to probe if this change in how DMA buffers for ISM
devices get allocated is acceptable as an interim solution.
With this change applied on top of current master, our test-cases for
SMC-D splice complete successfully again.
Gerd Bayer (1):
s390/ism: fix receive message buffer allocation
drivers/s390/net/ism_drv.c | 35 ++++++++++++++++++++++++++---------
1 file changed, 26 insertions(+), 9 deletions(-)
--
2.44.0
Powered by blists - more mailing lists