[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <df4c25ac-8929-434a-9363-2d3761f8fafa@redhat.com>
Date: Wed, 19 Mar 2025 08:50:44 +0100
From: Paolo Abeni <pabeni@...hat.com>
To: edward.cree@....com, linux-net-drivers@....com, davem@...emloft.net,
kuba@...nel.org, edumazet@...gle.com, horms@...nel.org, andrew+netdev@...n.ch
Cc: Edward Cree <ecree.xilinx@...il.com>, netdev@...r.kernel.org
Subject: Re: [PATCH net-next 3/3] sfc: support X4 devlink flash
On 3/17/25 4:00 PM, edward.cree@....com wrote:
> diff --git a/drivers/net/ethernet/sfc/efx_reflash.c b/drivers/net/ethernet/sfc/efx_reflash.c
> index ddc53740f098..bc96dd29b675 100644
> --- a/drivers/net/ethernet/sfc/efx_reflash.c
> +++ b/drivers/net/ethernet/sfc/efx_reflash.c
> @@ -407,31 +407,40 @@ int efx_reflash_flash_firmware(struct efx_nic *efx, const struct firmware *fw,
> return -EOPNOTSUPP;
> }
>
> - devlink_flash_update_status_notify(devlink, "Checking update", NULL, 0, 0);
> + mutex_lock(&efx->reflash_mutex);
>
> - rc = efx_reflash_parse_firmware_data(fw, &type, &data_subtype, &data,
> - &data_size);
> - if (rc) {
> - NL_SET_ERR_MSG_MOD(extack,
> - "Firmware image validation check failed");
> - goto out;
> - }
> + devlink_flash_update_status_notify(devlink, "Checking update", NULL, 0, 0);
>
> - mutex_lock(&efx->reflash_mutex);
> + if (efx->type->flash_auto_partition) {
> + /* NIC wants entire FW file including headers;
> + * FW will validate 'subtype' if there is one
> + */
> + type = NVRAM_PARTITION_TYPE_AUTO;
> + data = fw->data;
> + data_size = fw->size;
> + } else {
> + rc = efx_reflash_parse_firmware_data(fw, &type, &data_subtype, &data,
> + &data_size);
> + if (rc) {
> + NL_SET_ERR_MSG_MOD(extack,
> + "Firmware image validation check failed");
Not a full code review, but on this error path the reflash_mutex is
apparently not released.
/P
Powered by blists - more mailing lists