| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <2vvet4ai3uihb2skzyfiym2qh6g26knb7ymjp73eejoiywqnkm@2rxxv6zqvi33> Date: Tue, 6 Aug 2024 01:43:15 +0300 From: Serge Semin <fancer.lancer@...il.com> To: Jitendra Vegiraju <jitendra.vegiraju@...adcom.com> Cc: netdev@...r.kernel.org, alexandre.torgue@...s.st.com, joabreu@...opsys.com, davem@...emloft.net, edumazet@...gle.com, kuba@...nel.org, pabeni@...hat.com, mcoquelin.stm32@...il.com, bcm-kernel-feedback-list@...adcom.com, richardcochran@...il.com, ast@...nel.org, daniel@...earbox.net, hawk@...nel.org, john.fastabend@...il.com, linux-kernel@...r.kernel.org, linux-stm32@...md-mailman.stormreply.com, linux-arm-kernel@...ts.infradead.org, bpf@...r.kernel.org, andrew@...n.ch, linux@...linux.org.uk, horms@...nel.org, florian.fainelli@...adcom.com Subject: Re: [PATCH net-next v3 0/3] net: stmmac: Add PCI driver support for BCM8958x On Fri, Aug 02, 2024 at 03:06:05PM -0700, Jitendra Vegiraju wrote: > On Fri, Aug 2, 2024 at 3:02 AM Serge Semin <fancer.lancer@...il.com> wrote: > > > > Hi Jitendra > > > > On Thu, Aug 01, 2024 at 08:18:19PM -0700, jitendra.vegiraju@...adcom.com wrote: > > > From: Jitendra Vegiraju <jitendra.vegiraju@...adcom.com> > > > > > > This patchset adds basic PCI ethernet device driver support for Broadcom > > > BCM8958x Automotive Ethernet switch SoC devices. > > > > > > This SoC device has PCIe ethernet MAC attached to an integrated ethernet > > > switch using XGMII interface. The PCIe ethernet controller is presented to > > > the Linux host as PCI network device. > > > > > > The following block diagram gives an overview of the application. > > > +=================================+ > > > | Host CPU/Linux | > > > +=================================+ > > > || PCIe > > > || > > > +==========================================+ > > > | +--------------+ | > > > | | PCIE Endpoint| | > > > | | Ethernet | | > > > | | Controller | | > > > | | DMA | | > > > | +--------------+ | > > > | | MAC | BCM8958X | > > > | +--------------+ SoC | > > > | || XGMII | > > > | || | > > > | +--------------+ | > > > | | Ethernet | | > > > | | switch | | > > > | +--------------+ | > > > | || || || || | > > > +==========================================+ > > > || || || || More external interfaces > > > > > > The MAC block on BCM8958x is based on Synopsis XGMAC 4.00a core. This > > > driver uses common dwxgmac2 code where applicable. > > > > Thanks for submitting the series. > > > > I am curious how come Broadcom got to use an IP-core which hasn't > > been even announced by Synopsys. AFAICS the most modern DW XGMAC > > IP-core is of v3.xxa version: > > > > https://www.synopsys.com/dw/ipdir.php?ds=dwc_ether_xgmac > > > I am not sure why the 4.00a IP-code is not announced for general > availability yet. > The Synopsis documentation for this IP mentions 3.xx IP as reference > for this design and lists > new features for 4.00a. > > > Are you sure that your device isn't equipped with some another DW MAC > > IP-core, like DW 25G Ethernet MAC? (which BTW is equipped with a new > > Hyper DMA engine with a capability to have up to 128/256 channels with > > likely indirect addressing.) Do I miss something? > > > Yes, I briefly mentioned the new DMA architecture in the commit log > for patch 1/3. > You are correct, the name for the new DMA engine is Hyper DMA. It > probably started with some 3.xx IP-Core. > This DW-MAC is capable of 25G, but this SOC is not using 25G support. Then what you have is likely the DW 25GMAC since just DW XGMAC hasn't been announced to have neither 25G speed nor the Hyper-DMA with the virtualization channels capabilities. Meanwhile the former IP-core does have these features: https://www.synopsys.com/dw/ipdir.php?ds=dwc_25g_ethernet_mac_ip Alas I don't have the DW 25GMAC IP-core databook to say for sure, but that's the only explanation of why you have the 0x40 Synopsys ID and the IP-core version of v4.00a, and the 25G capability of the MAC. Seeing Synopsys tends to re-use the CSRs mapping even across the major IP-core releases it isn't that surprising that the DW XGMAC 3.xx IP-core was referenced in the doc. (See the driver, DW XLGMAC is almost fully compatible with the DW XGMAC CSRs mapping.) Moreover the most DMA-capable device currently supported by the STMMAC-driver is DW XGMAC/XLGMAC and it can't have more than 16 DMA-channels. That allows to directly map all the channels CSRs to the system memory. But your case is different. The DW 25GMAC IP-core is announced to support virtualization up to 128/256 channels, for which the direct CSRs mapping could require 16-times more memory. That's likely why the indirect addressing was implemented to access the settings of all the possible channels. That's also implicitly proofs that you have the DW 25GMAC IP-core. -Serge(y) > > > * I'll join the patch set review after the weekend, sometime on the > > next week. > > > > -Serge(y) > > > > > Driver functionality specific to this MAC is implemented in dwxgmac4.c. > > > Management of integrated ethernet switch on this SoC is not handled by > > > the PCIe interface. > > > This SoC device has PCIe ethernet MAC directly attached to an integrated > > > ethernet switch using XGMII interface. > > > > > > v2->v3: > > > Addressed v2 comments from Andrew, Jakub, Russel and Simon. > > > Based on suggestion by Russel and Andrew, added software node to create > > > phylink in fixed-link mode. > > > Moved dwxgmac4 specific functions to new files dwxgmac4.c and dwxgmac4.h > > > in stmmac core module. > > > Reorganized the code to use the existing glue logic support for xgmac in > > > hwif.c and override ops functions for dwxgmac4 specific functions. > > > The patch is split into three parts. > > > Patch#1 Adds dma_ops for dwxgmac4 in stmmac core > > > Patch#2 Hooks in the hardware interface handling for dwxgmac4 > > > Patch#3 Adds PCI driver for BCM8958x device > > > > > > v1->v2: > > > Minor fixes to address coding style issues. > > > Sent v2 too soon by mistake, without waiting for review comments. > > > Received feedback on this version. > > > https://lore.kernel.org/netdev/20240511015924.41457-1-jitendra.vegiraju@broadcom.com/ > > > > > > v1: > > > https://lore.kernel.org/netdev/20240510000331.154486-1-jitendra.vegiraju@broadcom.com/ > > > > > > Jitendra Vegiraju (3): > > > Add basic dwxgmac4 support to stmmac core > > > Integrate dwxgmac4 into stmmac hwif handling > > > Add PCI driver support for BCM8958x > > > > > > MAINTAINERS | 8 + > > > drivers/net/ethernet/stmicro/stmmac/Kconfig | 11 + > > > drivers/net/ethernet/stmicro/stmmac/Makefile | 3 +- > > > drivers/net/ethernet/stmicro/stmmac/common.h | 4 + > > > .../net/ethernet/stmicro/stmmac/dwmac-brcm.c | 517 ++++++++++++++++++ > > > .../ethernet/stmicro/stmmac/dwxgmac2_dma.c | 31 ++ > > > .../net/ethernet/stmicro/stmmac/dwxgmac4.c | 142 +++++ > > > .../net/ethernet/stmicro/stmmac/dwxgmac4.h | 84 +++ > > > drivers/net/ethernet/stmicro/stmmac/hwif.c | 26 +- > > > drivers/net/ethernet/stmicro/stmmac/hwif.h | 1 + > > > 10 files changed, 825 insertions(+), 2 deletions(-) > > > create mode 100644 drivers/net/ethernet/stmicro/stmmac/dwmac-brcm.c > > > create mode 100644 drivers/net/ethernet/stmicro/stmmac/dwxgmac4.c > > > create mode 100644 drivers/net/ethernet/stmicro/stmmac/dwxgmac4.h > > > > > > -- > > > 2.34.1 > > > > > >
Powered by blists - more mailing lists