Commit graph

4 commits

Author SHA1 Message Date
Rohit Mathew
c669f65359 refactor(sgi): move from "sgi" to "neoverse_rd"
Currently, reference design platforms such as RD-N2, RD-N1-Edge,
RD-V1-MC, RD-V1 and SGI-575 utilize "css/sgi" as the common source
directory. The "sgi" prefix originated from the System Guidance for
Infrastructure (SGI) and was initially associated with the SGI-575
platform. However, subsequent platforms released were under the Neoverse
Reference Design product name.

To align with the Neoverse Reference Design nomenclature, rename the
common source directory from "css/sgi" to "css/neoverse_rd" and update
all file prefixes from "sgi" to "nrd."

Signed-off-by: Rohit Mathew <Rohit.Mathew@arm.com>
Change-Id: I3dcbb31b9ab202e82caf25218ba33c520dcea4e4
2024-02-22 15:08:03 +05:30
Rohit Mathew
89d857780c fix(board): update spi_id max for sgi multichip platforms
According to GIC-600 TRM, 960 SPIs could be supported on a platform.
However, with the current configuration, platforms such as RD-V1-MC and
RD-N1-Edge Dual-Chip utilize a much smaller range. With commit
'a02a45dfe' gic600 driver is updated to get the max SPI id from the
GIC-600 and probe for the corresponding GIC instance for each SPI id.
Since RD-V1-MC and RD-N1-Edge Dual-Chip supports wider range, increase
SPI range for the chip 0 to max SPI range supported.

Signed-off-by: Rohit Mathew <Rohit.Mathew@arm.com>
Change-Id: Ia8c02c6d999033af33d8e7a0bedc7b73c6552ab4
2024-02-22 14:59:06 +05:30
Varun Wadekar
a02a45dfef fix(gicv3): workaround for NVIDIA erratum T241-FABRIC-4
The purpose of this patch is to address the T241 erratum T241-FABRIC-4,
which causes unexpected behavior in the GIC when multiple transactions
are received simultaneously from different sources. This hardware issue
impacts NVIDIA server platforms that use more than two T241 chips
interconnected. Each chip has support for 320 {E}SPIs.

This issue occurs when multiple packets from different GICs are
incorrectly interleaved at the target chip. The erratum text below
specifies exactly what can cause multiple transfer packets susceptible
to interleaving and GIC state corruption. GIC state corruption can
lead to a range of problems, including kernel panics, and unexpected
behavior.

Erratum documentation:
https://developer.nvidia.com/docs/t241-fabric-4/nvidia-t241-fabric-4-errata.pdf

The workaround is to ensure that MMIO accesses target the GIC on the
socket that holds the data, for example SPI ranges owned by the socket’s
GIC. This ensures that the GIC will not utilize the inter-socket AXI
Stream interface for servicing these GIC MMIO accesses.

This patch updates the functions that use the GICD_In{E} registers to
ensure that the accesses are directed to the chip that owns the SPI,
instead of using the global alias.

Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Change-Id: I04e33ba64eb306bd5fdabb56e63cbe273d8cd632
2023-03-23 23:06:07 +00:00
Aditya Angadi
90aecf1e0a plat/arm: rename rddanielxlr to rdv1mc
Reference Design platform RD-Daniel-ConfigXLR has been renamed to
RD-V1-MC. Correspondingly, remove all uses of 'rddanielxlr' and replace
it with 'rdv1mc' where appropriate.

Signed-off-by: Aditya Angadi <aditya.angadi@arm.com>
Change-Id: I5d91c69738397b19ced43949b4080c74678e604c
2021-01-11 12:29:51 +05:30
Renamed from plat/arm/board/rddanielxlr/rddanielxlr_plat.c (Browse further)