| 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: <b614c56f007b2669f1a23bfe8a8bc6c273f81bba.camel@codeconstruct.com.au>
Date: Thu, 31 Oct 2024 09:28:12 +0800
From: Jeremy Kerr <jk@...econstruct.com.au>
To: admiyo@...amperecomputing.com, Matt Johnston
<matt@...econstruct.com.au>, "David S. Miller" <davem@...emloft.net>, Eric
Dumazet <edumazet@...gle.com>, Jakub Kicinski <kuba@...nel.org>, Paolo
Abeni <pabeni@...hat.com>
Cc: netdev@...r.kernel.org, linux-kernel@...r.kernel.org, Sudeep Holla
<sudeep.holla@....com>, Jonathan Cameron <Jonathan.Cameron@...wei.com>,
Huisong Li <lihuisong@...wei.com>
Subject: Re: [PATCH v6 2/2] mctp pcc: Implement MCTP over PCC Transport
Hi Adam,
> From: Adam Young <admiyo@...amperecomputing.com>
>
> Implementation of network driver for
> Management Control Transport Protocol(MCTP) over
> Platform Communication Channel(PCC)
>
> DMTF DSP:0292
> https://www.dmtf.org/sites/default/files/standards/documents/DSP0292_1.0.0WIP50.pdf
>
> MCTP devices are specified by entries in DSDT/SDST and
> reference channels specified in the PCCT.
>
> Communication with other devices use the PCC based
> doorbell mechanism.
Nice progress on these. A few things inline, mainly the query on device
addressing.
> diff --git a/drivers/net/mctp/Kconfig b/drivers/net/mctp/Kconfig
> index 15860d6ac39f..7e55db0fb7a0 100644
> --- a/drivers/net/mctp/Kconfig
> +++ b/drivers/net/mctp/Kconfig
> @@ -47,6 +47,19 @@ config MCTP_TRANSPORT_I3C
> A MCTP protocol network device is created for each I3C bus
> having a "mctp-controller" devicetree property.
>
> +config MCTP_TRANSPORT_PCC
> + tristate "MCTP PCC transport"
> + select ACPI
> + help
> + Provides a driver to access MCTP devices over PCC transport,
> + A MCTP protocol network device is created via ACPI for each
> + entry in the DST/SDST that matches the identifier. The Platform
> + commuinucation channels are selected from the corresponding
Minor typo: "communication"
> diff --git a/drivers/net/mctp/mctp-pcc.c b/drivers/net/mctp/mctp-pcc.c
> new file mode 100644
> index 000000000000..b21fdca69538
> --- /dev/null
> +++ b/drivers/net/mctp/mctp-pcc.c
> @@ -0,0 +1,332 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * mctp-pcc.c - Driver for MCTP over PCC.
> + * Copyright (c) 2024, Ampere Computing LLC
> + */
> +
> +/* Implelmentation of MCTP over PCC DMTF Specification 256
> + * https://www.dmtf.org/sites/default/files/standards/documents/DSP0256_2.0.0WIP50.pdf
> + */
> +
> +#include <linux/acpi.h>
> +#include <linux/if_arp.h>
> +#include <linux/init.h>
> +#include <linux/kernel.h>
> +#include <linux/module.h>
> +#include <linux/netdevice.h>
> +#include <linux/platform_device.h>
> +#include <linux/string.h>
> +
> +#include <acpi/acpi_bus.h>
> +#include <acpi/acpi_drivers.h>
> +#include <acpi/acrestyp.h>
> +#include <acpi/actbl.h>
> +#include <net/mctp.h>
> +#include <net/mctpdevice.h>
> +#include <acpi/pcc.h>
> +
> +#define MCTP_PAYLOAD_LENGTH 256
> +#define MCTP_CMD_LENGTH 4
> +#define MCTP_PCC_VERSION 0x1 /* DSP0253 defines a single version: 1 */
> +#define MCTP_SIGNATURE "MCTP"
> +#define MCTP_SIGNATURE_LENGTH (sizeof(MCTP_SIGNATURE) - 1)
> +#define MCTP_HEADER_LENGTH 12
> +#define MCTP_MIN_MTU 68
> +#define PCC_MAGIC 0x50434300
> +#define PCC_HEADER_FLAG_REQ_INT 0x1
> +#define PCC_HEADER_FLAGS PCC_HEADER_FLAG_REQ_INT
> +#define PCC_DWORD_TYPE 0x0c
> +
> +struct mctp_pcc_hdr {
> + u32 signature;
> + u32 flags;
> + u32 length;
> + char mctp_signature[MCTP_SIGNATURE_LENGTH];
> +};
These signature/flags/length still don't have the endian annotations
(nor conversions on access). This was raised on v2, but looks like that
got lost?
> +
> +struct mctp_pcc_mailbox {
> + u32 index;
> + struct pcc_mbox_chan *chan;
> + struct mbox_client client;
> + void __iomem *addr;
> +};
> +
> +struct mctp_pcc_hw_addr {
> + __be32 parent_id;
> + __be16 inbox_id;
> + __be16 outbox_id;
> +};
> +
> +/* The netdev structure. One of these per PCC adapter. */
> +struct mctp_pcc_ndev {
> + /* spinlock to serialize access to PCC outbox buffer and registers
> + * Note that what PCC calls registers are memory locations, not CPU
> + * Registers. They include the fields used to synchronize access
> + * between the OS and remote endpoints.
> + *
> + * Only the Outbox needs a spinlock, to prevent multiple
> + * sent packets triggering multiple attempts to over write
> + * the outbox. The Inbox buffer is controlled by the remote
> + * service and a spinlock would have no effect.
> + */
Nice!
> +static void
> +mctp_pcc_net_stats(struct net_device *net_dev,
> + struct rtnl_link_stats64 *stats)
> +{
> + stats->rx_errors = 0;
> + stats->rx_packets = net_dev->stats.rx_packets;
> + stats->tx_packets = net_dev->stats.tx_packets;
> + stats->rx_dropped = 0;
> + stats->tx_bytes = net_dev->stats.tx_bytes;
> + stats->rx_bytes = net_dev->stats.rx_bytes;
> +}
Is this missing the rx_dropped stat (which you're updating in
_rx_callback)?
If you like, there are some new tstats helpers available, meaning you
wouldn't need the ndo_get_stats64 op at all. Let me know if you're
interested in using those, and would like a hand doing so.
> +static int mctp_pcc_initialize_mailbox(struct device *dev,
> + struct mctp_pcc_mailbox *box, u32 index)
> +{
> + pr_info("index = %u", index);
Left over from debug?
> +static int mctp_pcc_driver_add(struct acpi_device *acpi_dev)
> +{
> + struct mctp_pcc_lookup_context context = {0, 0, 0};
> + struct mctp_pcc_hw_addr mctp_pcc_hw_addr;
> + struct mctp_pcc_ndev *mctp_pcc_ndev;
> + struct device *dev = &acpi_dev->dev;
> + struct net_device *ndev;
> + acpi_handle dev_handle;
> + acpi_status status;
> + int mctp_pcc_mtu;
> + char name[32];
> + int rc;
> +
> + dev_dbg(dev, "Adding mctp_pcc device for HID %s\n",
Super minor: double space before the %s here
> + acpi_device_hid(acpi_dev));
> + dev_handle = acpi_device_handle(acpi_dev);
> + status = acpi_walk_resources(dev_handle, "_CRS", lookup_pcct_indices,
> + &context);
> + if (!ACPI_SUCCESS(status)) {
> + dev_err(dev, "FAILURE to lookup PCC indexes from CRS");
+ trailing newline
> + return -EINVAL;
> + }
> +
> + //inbox initialization
> + snprintf(name, sizeof(name), "mctpipcc%d", context.inbox_index);
> + ndev = alloc_netdev(sizeof(struct mctp_pcc_ndev), name, NET_NAME_ENUM,
> + mctp_pcc_setup);
> + if (!ndev)
> + return -ENOMEM;
> +
> + mctp_pcc_ndev = netdev_priv(ndev);
> + rc = devm_add_action_or_reset(dev, mctp_cleanup_netdev, ndev);
> + if (rc)
> + goto cleanup_netdev;
> + spin_lock_init(&mctp_pcc_ndev->lock);
> +
> + rc = mctp_pcc_initialize_mailbox(dev, &mctp_pcc_ndev->inbox,
> + context.inbox_index);
> + if (rc)
> + goto cleanup_netdev;
> + mctp_pcc_ndev->inbox.client.rx_callback = mctp_pcc_client_rx_callback;
> +
> + //outbox initialization
> + rc = mctp_pcc_initialize_mailbox(dev, &mctp_pcc_ndev->outbox,
> + context.outbox_index);
> + if (rc)
> + goto cleanup_netdev;
> +
> + mctp_pcc_hw_addr.parent_id = cpu_to_be32(0);
> + mctp_pcc_hw_addr.inbox_id = cpu_to_be16(context.inbox_index);
> + mctp_pcc_hw_addr.outbox_id = cpu_to_be16(context.outbox_index);
> + ndev->addr_len = sizeof(mctp_pcc_hw_addr);
> + dev_addr_set(ndev, (const u8 *)&mctp_pcc_hw_addr);
I recall querying this in v1, not sure if there was a response, but:
Given there is no hardware addressing in the packet format, what is the
meaning of the physical address on the interface? It's a little strange
to define a hardware address here that isn't used for actual addressing.
For point-to-point links like this (and the serial transport), it's fine
to have no hw address on the device.
If this is purely local-machine-specific instance data, I suspect that
this belongs elsewhere. A read-only sysfs attribute could work?
Cheers,
Jeremy
Powered by blists - more mailing lists