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: <AM6PR04MB6341CDED2DB4CB01A49678A0E76A9@AM6PR04MB6341.eurprd04.prod.outlook.com>
Date:   Fri, 3 Dec 2021 01:54:20 +0000
From:   Ming Qian <ming.qian@....com>
To:     Hans Verkuil <hverkuil-cisco@...all.nl>,
        "mchehab@...nel.org" <mchehab@...nel.org>,
        "shawnguo@...nel.org" <shawnguo@...nel.org>,
        "robh+dt@...nel.org" <robh+dt@...nel.org>,
        "s.hauer@...gutronix.de" <s.hauer@...gutronix.de>
CC:     "kernel@...gutronix.de" <kernel@...gutronix.de>,
        "festevam@...il.com" <festevam@...il.com>,
        dl-linux-imx <linux-imx@....com>,
        Aisheng Dong <aisheng.dong@....com>,
        "linux-media@...r.kernel.org" <linux-media@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
        "linux-arm-kernel@...ts.infradead.org" 
        <linux-arm-kernel@...ts.infradead.org>
Subject: RE: [EXT] Re: [PATCH v13 06/13] media: amphion: add vpu v4l2 m2m
 support

> > +
> > +int vpu_v4l2_open(struct file *file, struct vpu_inst *inst)
> > +{
> > +     struct vpu_dev *vpu = video_drvdata(file);
> > +     struct vpu_func *func;
> > +     int ret = 0;
> > +
> > +     WARN_ON(!file || !inst || !inst->ops);
> > +
> > +     if (inst->type == VPU_CORE_TYPE_ENC)
> > +             func = &vpu->encoder;
> > +     else
> > +             func = &vpu->decoder;
> > +
> > +     atomic_set(&inst->ref_count, 0);
> > +     vpu_inst_get(inst);
> > +     inst->vpu = vpu;
> > +     inst->core = vpu_request_core(vpu, inst->type);
> > +     if (inst->core)
> > +             inst->dev = get_device(inst->core->dev);
> > +     mutex_init(&inst->lock);
> > +     INIT_LIST_HEAD(&inst->cmd_q);
> > +     inst->id = VPU_INST_NULL_ID;
> > +     inst->release = vpu_v4l2_release;
> > +     inst->pid = current->pid;
> > +     inst->tgid = current->tgid;
> > +     inst->min_buffer_cap = 2;
> > +     inst->min_buffer_out = 2;
> 
> Assuming this means the minimum number of buffers needed, why is
> min_buffers_needed set to 1 when initializing the vb2_queue structs?

In my opinion, the min_buffers_needed determine when vb2_start_streaming will be called,
Like the following code:
if (q->queued_count >= q->min_buffers_needed) {
        ... ...
        ret = vb2_start_streaming(q);
        ... ...
}
I hope driver starts a vpu instance when 1 frame is queued, so I set the min_buffers_needed to 1.
And the min_buffer_cap means the minimum vb2 buffer count, and it will changed according to the stream,
I just set default value to 2, it will be changed after vpu parsed the stream info.

> 
> > +     v4l2_fh_init(&inst->fh, func->vfd);
> > +     v4l2_fh_add(&inst->fh);
> > +
> > +     ret = call_vop(inst, ctrl_init);
> > +     if (ret)
> > +             goto error;
> > +
> > +     inst->fh.m2m_ctx = v4l2_m2m_ctx_init(func->m2m_dev,
> > +                                     inst, vpu_m2m_queue_init);
> > +     if (IS_ERR(inst->fh.m2m_ctx)) {
> > +             dev_err(vpu->dev, "v4l2_m2m_ctx_init fail\n");
> > +             ret = PTR_ERR(func->m2m_dev);
> > +             goto error;
> > +     }
> > +
> > +     inst->fh.ctrl_handler = &inst->ctrl_handler;
> > +     file->private_data = &inst->fh;
> > +     inst->state = VPU_CODEC_STATE_DEINIT;
> > +     inst->workqueue = alloc_workqueue("vpu_inst", WQ_UNBOUND |
> WQ_MEM_RECLAIM, 1);
> > +     if (inst->workqueue) {
> > +             INIT_WORK(&inst->msg_work, vpu_inst_run_work);
> > +             ret = kfifo_init(&inst->msg_fifo,
> > +                             inst->msg_buffer,
> > +
> roundup_pow_of_two(sizeof(inst->msg_buffer)));
> > +             if (ret) {
> > +                     destroy_workqueue(inst->workqueue);
> > +                     inst->workqueue = NULL;
> > +             }
> > +     }
> > +     vpu_trace(vpu->dev, "tgid = %d, pid = %d, type = %s, inst = %p\n",
> > +                     inst->tgid, inst->pid,
> vpu_core_type_desc(inst->type), inst);
> > +
> > +     return 0;
> > +error:
> > +     vpu_inst_put(inst);
> > +     return ret;
> > +}
> > +

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ