From: Takashi Iwai <tiwai@suse.de>
To: "Geoffrey D. Bennett" <g@b4.vu>
Cc: Takashi Iwai <tiwai@suse.de>, Jaroslav Kysela <perex@perex.cz>,
Takashi Iwai <tiwai@suse.com>,
linux-sound@vger.kernel.org
Subject: Re: [PATCH v7 1/2] ALSA: FCP: Add Focusrite Control Protocol driver
Date: Thu, 16 Jan 2025 16:03:48 +0100 [thread overview]
Message-ID: <878qraiywb.wl-tiwai@suse.de> (raw)
In-Reply-To: <7355f25d9b57386c280aef651b6ba2b96cf0d5a7.1737017960.git.g@b4.vu>
On Thu, 16 Jan 2025 10:03:35 +0100,
Geoffrey D. Bennett wrote:
> diff --git a/include/uapi/sound/tlv.h b/include/uapi/sound/tlv.h
> index b99a2414b53d..b6e260e35529 100644
> --- a/include/uapi/sound/tlv.h
> +++ b/include/uapi/sound/tlv.h
> @@ -18,6 +18,8 @@
> #define SNDRV_CTL_TLVT_CHMAP_VAR 0x102 /* channels freely swappable */
> #define SNDRV_CTL_TLVT_CHMAP_PAIRED 0x103 /* pair-wise swappable */
>
> +#define SNDRV_CTL_TLVT_LABELS 0x110 /* channel labels */
IMHO, the data definition looks too ambiguous here to be put as a
public API. It can be also a more specific name such as
*_CHANNEL_LABELS, too.
> +static int fcp_meter_ctl_get(struct snd_kcontrol *kctl,
> + struct snd_ctl_elem_value *ucontrol)
> +{
> + struct usb_mixer_elem_info *elem = kctl->private_data;
> + struct usb_mixer_interface *mixer = elem->head.mixer;
> + struct fcp_data *private = mixer->private_data;
> + int num_meter_slots, resp_size;
> + u32 *resp __free(kfree) = NULL;
> + int i, err = 0;
> +
> + struct {
> + __le16 pad;
> + __le16 num_meters;
> + __le32 magic;
> + } __packed req;
> +
> + guard(mutex)(&private->mutex);
> +
> + err = fcp_reinit(mixer);
> + if (err < 0)
> + return err;
> +
> + num_meter_slots = private->num_meter_slots;
> + resp_size = num_meter_slots * sizeof(u32);
> +
> + resp = kmalloc(resp_size, GFP_KERNEL);
> + if (!resp)
> + return -ENOMEM;
It depends on how often this kcontrol gets read, but if it's
frequently read, it might make sense to allocate this buffer
beforehand and keep using it instead of allocating / freeing at each
call. The call is mutual, hence a common buffer is fine.
> +static int fcp_meter_tlv_callback(struct snd_kcontrol *kctl,
> + int op_flag, unsigned int size,
> + unsigned int __user *tlv)
> +{
> + struct usb_mixer_elem_info *elem = kctl->private_data;
> + struct usb_mixer_interface *mixer = elem->head.mixer;
> + struct fcp_data *private = mixer->private_data;
> +
> + if (op_flag == SNDRV_CTL_TLV_OP_READ) {
> + if (private->meter_labels_tlv_size == 0)
> + return 0;
> +
> + if (size > private->meter_labels_tlv_size)
> + size = private->meter_labels_tlv_size;
> +
> + if (copy_to_user(tlv, private->meter_labels_tlv, size))
> + return -EFAULT;
The private->meter_label* stuff can be changed dynamically via ioctl,
hence this call should have the data protection, too.
I didn't check fully, but private->mutex can be taken in this code
path, I suppose.
> +/* FCP initialisation */
> +static int fcp_ioctl_init(struct usb_mixer_interface *mixer,
> + struct fcp_init __user *arg)
> +{
> + struct fcp_init init;
> + struct usb_device *dev = mixer->chip->dev;
> + struct fcp_data *private = mixer->private_data;
> + void *resp __free(kfree) = NULL;
> + void *step2_resp;
> + int err, buf_size;
> +
> + if (usb_pipe_type_check(dev, usb_sndctrlpipe(dev, 0)))
> + return -EINVAL;
> +
> + /* Get initialisation parameters */
> + if (copy_from_user(&init, arg, sizeof(init)))
> + return -EFAULT;
> +
> + /* Validate the response sizes */
> + if (init.step0_resp_size < 1 ||
> + init.step0_resp_size > 255 ||
> + init.step2_resp_size < 1 ||
> + init.step2_resp_size > 255)
> + return -EINVAL;
> +
> + private->step0_resp_size = init.step0_resp_size;
> + private->step2_resp_size = init.step2_resp_size;
> + private->init1_opcode = init.init1_opcode;
> + private->init2_opcode = init.init2_opcode;
> +
> + /* Allocate response buffer */
> + buf_size = private->step0_resp_size + private->step2_resp_size;
> +
> + resp = kmalloc(buf_size, GFP_KERNEL);
> + if (!resp)
> + return -ENOMEM;
I'd set private->step* and co after allocation of the buffer.
When they are set beforehand, even if the buffer allocation failed,
those values are still set.
> +/* Set the Level Meter labels */
> +static int fcp_ioctl_set_meter_labels(struct usb_mixer_interface *mixer,
> + struct fcp_meter_labels __user *arg)
> +{
(snip)
> + /* Calculate padded data size */
> + data_size = ((labels.labels_size + sizeof(unsigned int) - 1) &
> + ~(sizeof(unsigned int) - 1));
ALIGN() macro can be used nicely here.
thanks,
Takashi
next prev parent reply other threads:[~2025-01-16 15:03 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-01-16 9:02 [PATCH v7 0/2] ALSA: Add driver for big Scarlett 4th Gen interfaces Geoffrey D. Bennett
2025-01-16 9:03 ` [PATCH v7 1/2] ALSA: FCP: Add Focusrite Control Protocol driver Geoffrey D. Bennett
2025-01-16 15:03 ` Takashi Iwai [this message]
2025-01-16 16:43 ` Geoffrey D. Bennett
2025-01-16 16:48 ` Takashi Iwai
2025-01-16 9:03 ` [PATCH v7 2/2] ALSA: scarlett2: Add device_setup option to use FCP driver Geoffrey D. Bennett
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=878qraiywb.wl-tiwai@suse.de \
--to=tiwai@suse.de \
--cc=g@b4.vu \
--cc=linux-sound@vger.kernel.org \
--cc=perex@perex.cz \
--cc=tiwai@suse.com \
/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