[Intel-gfx] [PATCH 3/8] drm/i915/huc: Add HuC fw loading support

[Tvrtko Ursulin]

Hi all,

Executive decision required below:

On 08/12/2016 23:02, anushasr wrote:

[snip]

> +
> +/**
> + * DOC: HuC Firmware
> + *
> + * Motivation:
> + * GEN9 introduces a new dedicated firmware for usage in media HEVC (High
> + * Efficiency Video Coding) operations. Userspace can use the firmware
> + * capabilities by adding HuC specific commands to batch buffers.
> + *
> + * Implementation:
> + * The same firmware loader is used as the GuC. However, the actual
> + * loading to HW is deferred until GEM initialization is done.
> + *
> + * Note that HuC firmware loading must be done before GuC loading.
> + */
> +
> +#define SKL_FW_MAJOR 01
> +#define SKL_FW_MINOR 07
> +#define SKL_BLD_NUM 1398
> +
> +#define HUC_FW_PATH(platform, major, minor, bld_num) \
> +	"i915/" __stringify(platform) "_huc_ver" __stringify(major) "_" \
> +	__stringify(minor) "_" __stringify(bld_num) ".bin"
> +
> +#define I915_SKL_HUC_UCODE HUC_FW_PATH(skl, SKL_FW_MAJOR, \
> +	SKL_FW_MINOR, SKL_BLD_NUM)
> +MODULE_FIRMWARE(I915_SKL_HUC_UCODE);

Daniel & Jani, I understand in the GuC firmware discussions you were 
very much in the favour of encoding the full name (major and minor 
included) as the fw firmware?

Argument was that we want to in effect claim support only for one 
validated firmware binary with one version of i915.

In the case of the HuC we have a very similar situation with two key 
differences.

First, there is a build number in the file name as provided by the 
firmware team. We know that it will happen that only the build number 
changes with some fixes and the minor stays the same. And we know that 
the major indicates the interface compatibility.

Secondly, from all I can see, there are no interactions between the 
driver and the HuC firmware apart from driver loading it and thats it.

In the light of that I was advocating only using the major in the driver 
request fw name in order to improve usability both for developers 
(easier to test with different firmwares), and for users (be it 
distributions or end users - easier to upgrade the HuC firmware in case 
of codec issues by not having to patch and recompile the kernel).

Since I understand this topic has been beaten to death in the past and 
there are strong opinions on it, could you just okay (or not) the 
current proposal (as in posted patches) which encodes major, minor and 
build number in the fw name?

Regards,

Tvrtko