[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20250319-topic-nonposted_mmio-v1-0-dfb886fbd15f@oss.qualcomm.com>
Date: Wed, 19 Mar 2025 15:25:56 +0100
From: Konrad Dybcio <konradybcio@...nel.org>
To: Rob Herring <robh@...nel.org>, Saravana Kannan <saravanak@...gle.com>
Cc: Bjorn Andersson <bjorn.andersson@....qualcomm.com>,
Melody Olvera <quic_molvera@...cinc.com>,
Marijn Suijten <marijn.suijten@...ainline.org>, devicetree@...r.kernel.org,
linux-kernel@...r.kernel.org,
Konrad Dybcio <konrad.dybcio@....qualcomm.com>
Subject: [PATCH 0/2] Allow 'nonposted-mmio' on single devices
For a plethora of reasons, usually not very pretty, a single device on
a given bus/platform may require a different type of memory mapping.
With nonposted-mmio specifically, we want to limit the scope of setting
it where it's not necessary, so as not to shoot ourselves in the foot
in the performance department.
This series allows mapping a single device as nE on arm64.
A user will be sent separately, as a new node for a QC platform.
The dt-bindings part is handled in the checker itself, see the following:
https://github.com/devicetree-org/dt-schema/pull/156
Signed-off-by: Konrad Dybcio <konrad.dybcio@....qualcomm.com>
---
Konrad Dybcio (2):
of: address: Expand nonposted-mmio to non-Apple Silicon platforms
of: address: Allow to specify nonposted-mmio per-device
drivers/of/address.c | 13 ++++---------
1 file changed, 4 insertions(+), 9 deletions(-)
---
base-commit: c4d4884b67802c41fd67399747165d65c770621a
change-id: 20250315-topic-nonposted_mmio-5393a3de0d35
Best regards,
--
Konrad Dybcio <konrad.dybcio@....qualcomm.com>
Powered by blists - more mailing lists