From: Vadim Fedorenko <vadim.fedorenko@linux.dev>
To: Lee Trager <lee@trager.us>,
Alexander Duyck <alexanderduyck@fb.com>,
Jakub Kicinski <kuba@kernel.org>,
kernel-team@meta.com, "David S. Miller" <davem@davemloft.net>,
Eric Dumazet <edumazet@google.com>,
Paolo Abeni <pabeni@redhat.com>, Jonathan Corbet <corbet@lwn.net>,
Mohsin Bashir <mohsin.bashr@gmail.com>,
Simon Horman <horms@kernel.org>,
Sanman Pradhan <sanmanpradhan@meta.com>,
Al Viro <viro@zeniv.linux.org.uk>
Cc: netdev@vger.kernel.org, linux-doc@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH net-next 2/2] eth: fbnic: Add devlink dev flash support
Date: Mon, 14 Oct 2024 12:18:36 +0100 [thread overview]
Message-ID: <8502a496-f83d-470c-a84d-081a7c7e2cae@linux.dev> (raw)
In-Reply-To: <20241012023646.3124717-3-lee@trager.us>
On 12/10/2024 03:34, Lee Trager wrote:
> fbnic supports updating firmware using a PLDM image signed and distributed
> by Meta. PLDM images are written into stored flashed. Flashing does not
> interrupt operation.
>
> On host reboot the newly flashed UEFI driver will be used. To run new
> control or cmrt firmware the NIC must be power cycled.
>
> Signed-off-by: Lee Trager <lee@trager.us>
> ---
> .../device_drivers/ethernet/meta/fbnic.rst | 11 +
> drivers/net/ethernet/meta/Kconfig | 1 +
> .../net/ethernet/meta/fbnic/fbnic_devlink.c | 270 +++++++++++++++++-
> 3 files changed, 281 insertions(+), 1 deletion(-)
>
> diff --git a/Documentation/networking/device_drivers/ethernet/meta/fbnic.rst b/Documentation/networking/device_drivers/ethernet/meta/fbnic.rst
> index 32ff114f5c26..d6726c254818 100644
> --- a/Documentation/networking/device_drivers/ethernet/meta/fbnic.rst
> +++ b/Documentation/networking/device_drivers/ethernet/meta/fbnic.rst
> @@ -27,3 +27,14 @@ driver takes over.
> devlink dev info provides version information for all three components. In
> addition to the version the hg commit hash of the build is included as a
> separate entry.
> +
> +Upgrading Firmware
> +------------------
> +
> +fbnic supports upgrading firmware using devlink dev flash. Firmware images
> +are signed and distributed by Meta. All firmware is bundled into a single
> +PLDM image which is written into stored flash. Flashing firmware does not
> +interrupt operation.
> +
> +On host reboot the newly flashed UEFI driver will be used. To run new control
> +or cmrt firmware the NIC must be power cycled.
> diff --git a/drivers/net/ethernet/meta/Kconfig b/drivers/net/ethernet/meta/Kconfig
> index 85519690b837..f5d2c6b6399f 100644
> --- a/drivers/net/ethernet/meta/Kconfig
> +++ b/drivers/net/ethernet/meta/Kconfig
> @@ -26,6 +26,7 @@ config FBNIC
> select NET_DEVLINK
> select PAGE_POOL
> select PHYLINK
> + select PLDMFW
> help
> This driver supports Meta Platforms Host Network Interface.
>
> diff --git a/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c b/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c
> index 0072d612215e..d487ae7f1126 100644
> --- a/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c
> +++ b/drivers/net/ethernet/meta/fbnic/fbnic_devlink.c
> @@ -3,6 +3,7 @@
>
> #include <linux/unaligned.h>
> #include <linux/pci.h>
> +#include <linux/pldmfw.h>
> #include <linux/types.h>
> #include <net/devlink.h>
>
> @@ -109,8 +110,275 @@ static int fbnic_devlink_info_get(struct devlink *devlink,
> return 0;
> }
>
> +/**
> + * fbnic_send_package_data - Send record package data to firmware
> + * @context: PLDM FW update structure
> + * @data: pointer to the package data
> + * @length: length of the package data
> + *
> + * Send a copy of the package data associated with the PLDM record matching
> + * this device to the firmware.
> + *
> + * Return: zero on success
> + * negative error code on failure
> + */
> +static int fbnic_send_package_data(struct pldmfw *context, const u8 *data,
> + u16 length)
> +{
> + struct device *dev = context->dev;
> +
> + /* Temp placeholder required by devlink */
> + dev_info(dev,
> + "Sending %u bytes of PLDM record package data to firmware\n",
> + length);
> +
> + return 0;
> +}
> +
> +/**
> + * fbnic_send_component_table - Send PLDM component table to the firmware
> + * @context: PLDM FW update structure
> + * @component: The component to send
> + * @transfer_flag: Flag indication location in component tables
> + *
> + * Read relevant data from component table and forward it to the firmware.
> + * Check response to verify if the firmware indicates that it wishes to
> + * proceed with the update.
> + *
> + * Return: zero on success
> + * negative error code on failure
> + */
> +static int fbnic_send_component_table(struct pldmfw *context,
> + struct pldmfw_component *component,
> + u8 transfer_flag)
> +{
> + struct device *dev = context->dev;
> + u16 id = component->identifier;
> + u8 test_string[80];
> +
> + switch (id) {
> + case QSPI_SECTION_CMRT:
> + case QSPI_SECTION_CONTROL_FW:
> + case QSPI_SECTION_OPTION_ROM:
> + break;
> + default:
> + dev_err(dev, "Unknown component ID %u\n", id);
> + return -EINVAL;
> + }
> +
> + dev_dbg(dev, "Sending PLDM component table to firmware\n");
> +
> + /* Temp placeholder */
> + memcpy(test_string, component->version_string,
> + min_t(u8, component->version_len, 79));
> + test_string[min_t(u8, component->version_len, 79)] = 0;
Looks like this construction can be replaced with strscpy().
There were several patchsets in the tree to use strscpy(), let's follow
the pattern.
> + dev_info(dev, "PLDMFW: Component ID: %u version %s\n",
> + id, test_string);
> +
> + return 0;
> +}
> +
> +/**
> + * fbnic_flash_component - Flash a component of the QSPI
> + * @context: PLDM FW update structure
> + * @component: The component table to send to FW
> + *
> + * Map contents of component and make it available for FW to download
> + * so that it can update the contents of the QSPI Flash.
> + *
> + * Return: zero on success
> + * negative error code on failure
> + */
> +static int fbnic_flash_component(struct pldmfw *context,
> + struct pldmfw_component *component)
> +{
> + const u8 *data = component->component_data;
> + u32 size = component->component_size;
> + struct fbnic_fw_completion *fw_cmpl;
> + struct device *dev = context->dev;
> + struct pci_dev *pdev = to_pci_dev(dev);
> + u16 id = component->identifier;
> + const char *component_name;
> + int retries = 2;
> + int err;
> +
> + struct devlink *devlink;
> + struct fbnic_dev *fbd;
> +
> + switch (id) {
> + case QSPI_SECTION_CMRT:
> + component_name = "boot1";
> + break;
> + case QSPI_SECTION_CONTROL_FW:
> + component_name = "boot2";
> + break;
> + case QSPI_SECTION_OPTION_ROM:
> + component_name = "option-rom";
> + break;
> + default:
> + dev_err(dev, "Unknown component ID %u\n", id);
> + return -EINVAL;
> + }
> +
> + fw_cmpl = kzalloc(sizeof(*fw_cmpl), GFP_KERNEL);
> + if (!fw_cmpl)
> + return -ENOMEM;
> +
> + pdev = to_pci_dev(dev);
> + fbd = pci_get_drvdata(pdev);
> + devlink = priv_to_devlink(fbd);
> +
> + /* Initialize completion and queue it for FW to process */
> + fw_cmpl->msg_type = FBNIC_TLV_MSG_ID_FW_WRITE_CHUNK_REQ;
> + init_completion(&fw_cmpl->done);
> +
> + fw_cmpl->fw_update.last_offset = 0;
> + fw_cmpl->fw_update.data = data;
> + fw_cmpl->fw_update.size = size;
> +
> + err = fbnic_fw_xmit_fw_start_upgrade(fbd, fw_cmpl, id, size);
> + if (err)
> + goto cmpl_free;
> +
> + /* Monitor completions and report status of update */
> + while (fw_cmpl->fw_update.data) {
> + u32 offset = fw_cmpl->fw_update.last_offset;
> +
> + devlink_flash_update_status_notify(devlink, "Flashing",
> + component_name, offset,
> + size);
> +
> + /* Allow 5 seconds for reply, resend and try up to 2 times */
> + if (wait_for_completion_timeout(&fw_cmpl->done, 5 * HZ)) {
> + reinit_completion(&fw_cmpl->done);
> + /* If we receive a reply, reinit our retry counter */
> + retries = 2;
> + } else if (--retries == 0) {
> + dev_err(fbd->dev, "Timed out waiting on update\n");
> + err = -ETIMEDOUT;
> + goto cmpl_cleanup;
> + }
> + }
> +
> + err = fw_cmpl->result;
> + if (err)
> + goto cmpl_cleanup;
> +
> + devlink_flash_update_status_notify(devlink, "Flashing",
> + component_name, size, size);
> +
> +cmpl_cleanup:
> + fbd->cmpl_data = NULL;
> +cmpl_free:
> + kfree(fw_cmpl);
> +
> + return err;
> +}
> +
[ strip ]
next prev parent reply other threads:[~2024-10-14 11:18 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-12 2:34 [PATCH net-next 0/2] eth: fbnic: Add devlink dev flash support Lee Trager
2024-10-12 2:34 ` [PATCH net-next 1/2] eth: fbnic: Add mailbox support for PLDM updates Lee Trager
2024-10-14 11:05 ` Vadim Fedorenko
2024-10-12 2:34 ` [PATCH net-next 2/2] eth: fbnic: Add devlink dev flash support Lee Trager
2024-10-14 11:18 ` Vadim Fedorenko [this message]
2024-10-15 10:43 ` Simon Horman
2024-10-18 22:48 ` Lee Trager
2024-10-19 9:34 ` Simon Horman
2024-10-22 1:37 ` [PATCH net-next v2 0/2] " Lee Trager
2024-10-22 1:37 ` [PATCH net-next v2 1/2] eth: fbnic: Add mailbox support for PLDM updates Lee Trager
2024-10-26 15:12 ` Simon Horman
2024-10-22 1:42 ` [PATCH net-next v2 2/2] eth: fbnic: Add devlink dev flash support Lee Trager
2024-10-24 9:10 ` Simon Horman
2024-10-25 22:32 ` Lee Trager
2024-10-26 15:12 ` Simon Horman
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=8502a496-f83d-470c-a84d-081a7c7e2cae@linux.dev \
--to=vadim.fedorenko@linux.dev \
--cc=alexanderduyck@fb.com \
--cc=corbet@lwn.net \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=horms@kernel.org \
--cc=kernel-team@meta.com \
--cc=kuba@kernel.org \
--cc=lee@trager.us \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mohsin.bashr@gmail.com \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=sanmanpradhan@meta.com \
--cc=viro@zeniv.linux.org.uk \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).