linux-kernel - Re: [PATCH v2 6/7] accel/rocket: Add job submission IOCTL

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

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <4d617199-191c-47d9-a0f1-5bf9c149d09d@oss.qualcomm.com>
Date: Fri, 21 Mar 2025 10:09:05 -0600
From: Jeff Hugo <jeff.hugo@....qualcomm.com>
To: Tomeu Vizoso <tomeu@...euvizoso.net>, Rob Herring <robh@...nel.org>,
        Krzysztof Kozlowski <krzk+dt@...nel.org>,
        Conor Dooley
 <conor+dt@...nel.org>, Heiko Stuebner <heiko@...ech.de>,
        Oded Gabbay <ogabbay@...nel.org>, Jonathan Corbet <corbet@....net>,
        Maarten Lankhorst <maarten.lankhorst@...ux.intel.com>,
        Maxime Ripard <mripard@...nel.org>,
        Thomas Zimmermann <tzimmermann@...e.de>,
        David Airlie <airlied@...il.com>, Simona Vetter <simona@...ll.ch>,
        Sumit Semwal <sumit.semwal@...aro.org>,
        Christian König <christian.koenig@....com>,
        Sebastian Reichel <sebastian.reichel@...labora.com>
Cc: devicetree@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
        linux-rockchip@...ts.infradead.org, linux-kernel@...r.kernel.org,
        dri-devel@...ts.freedesktop.org, linux-doc@...r.kernel.org,
        linux-media@...r.kernel.org, linaro-mm-sig@...ts.linaro.org
Subject: Re: [PATCH v2 6/7] accel/rocket: Add job submission IOCTL

On 2/25/2025 12:55 AM, Tomeu Vizoso wrote:
> +/**
> + * struct drm_rocket_task - A task to be run on the NPU
> + *
> + * A task is the smallest unit of work that can be run on the NPU.
> + */
> +struct drm_rocket_task {
> +	/** DMA address to NPU mapping of register command buffer */
> +	__u64 regcmd;
> +
> +	/** Number of commands in the register command buffer */
> +	__u32 regcmd_count;
> +};
> +
> +/**
> + * struct drm_rocket_job - A job to be run on the NPU
> + *
> + * The kernel will schedule the execution of this job taking into account its
> + * dependencies with other jobs. All tasks in the same job will be executed
> + * sequentially on the same core, to benefit from memory residency in SRAM.
> + */
> +struct drm_rocket_job {
> +	/** Pointer to an array of struct drm_rocket_task. */
> +	__u64 tasks;
> +
> +	/** Pointer to a u32 array of the BOs that are read by the job. */
> +	__u64 in_bo_handles;
> +
> +	/** Pointer to a u32 array of the BOs that are written to by the job. */
> +	__u64 out_bo_handles;
> +
> +	/** Number of tasks passed in. */
> +	__u32 task_count;
> +
> +	/** Number of input BO handles passed in (size is that times 4). */
> +	__u32 in_bo_handle_count;
> +
> +	/** Number of output BO handles passed in (size is that times 4). */
> +	__u32 out_bo_handle_count;
> +};
> +
> +/**
> + * struct drm_rocket_submit - ioctl argument for submitting commands to the NPU.
> + *
> + * The kernel will schedule the execution of these jobs in dependency order.
> + */
> +struct drm_rocket_submit {
> +	/** Pointer to an array of struct drm_rocket_job. */
> +	__u64 jobs;
> +
> +	/** Number of jobs passed in. */
> +	__u32 job_count;
> +};

These 3 structs will be different sizes in 32-bit env vs 64-bit env. Yes 
the driver depends on ARM64, but compat (32-bit userspace with 64-bit 
kernel) is still possible. They should all be padded out to 64-bit 
alignment.  When you do that, you should specify that the padding must 
be zero, and check for that in the driver so that you have the option to 
use the padding in the future.