From: Sean Young <sean@mess.org>
To: Dan Carpenter <error27@gmail.com>
Cc: Oliver Neukum <oneukum@suse.com>, linux-media@vger.kernel.org
Subject: Re: [bug report] media: rc: ttusbir: respect DMA coherency rules
Date: Fri, 10 Apr 2026 20:59:34 +0100 [thread overview]
Message-ID: <adlWpq7_W0U0dgom@gofer.mess.org> (raw)
In-Reply-To: <adjNDQ6Y3HrG4Xi-@stanley.mountain>
On Fri, Apr 10, 2026 at 01:12:29PM +0300, Dan Carpenter wrote:
> Hello Oliver Neukum,
>
> Commit 50acaad3d202 ("media: rc: ttusbir: respect DMA coherency
> rules") from Feb 11, 2026 (linux-next), leads to the following Smatch
> static checker warning:
>
> drivers/media/rc/ttusbir.c:281 ttusbir_probe()
> error: we previously assumed 'tt->bulk_buffer' could be null (see line 198)
>
> drivers/media/rc/ttusbir.c
> 180 static int ttusbir_probe(struct usb_interface *intf,
> 181 const struct usb_device_id *id)
> 182 {
> 183 struct ttusbir *tt;
> 184 struct usb_interface_descriptor *idesc;
> 185 struct usb_endpoint_descriptor *desc;
> 186 struct rc_dev *rc;
> 187 int i, j, ret;
> 188 int altsetting = -1;
> 189 u8 *buffer;
> 190
> 191 tt = kzalloc_obj(*tt);
> 192 buffer = kzalloc(5, GFP_KERNEL);
> 193 rc = rc_allocate_device(RC_DRIVER_IR_RAW);
> 194 if (!tt || !rc || buffer) {
>
> This ! in !buffer is missing.
Yep, that's wrong.
Thank you for reporting, this obviously breaks the driver.
Sean
>
> 195 ret = -ENOMEM;
> 196 goto out;
> 197 }
> 198 tt->bulk_buffer = buffer;
> 199
> 200 /* find the correct alt setting */
> 201 for (i = 0; i < intf->num_altsetting && altsetting == -1; i++) {
> 202 int max_packet, bulk_out_endp = -1, iso_in_endp = -1;
> 203
> 204 idesc = &intf->altsetting[i].desc;
> 205
> 206 for (j = 0; j < idesc->bNumEndpoints; j++) {
> 207 desc = &intf->altsetting[i].endpoint[j].desc;
> 208 max_packet = le16_to_cpu(desc->wMaxPacketSize);
> 209 if (usb_endpoint_dir_in(desc) &&
> 210 usb_endpoint_xfer_isoc(desc) &&
> 211 max_packet == 0x10)
> 212 iso_in_endp = j;
> 213 else if (usb_endpoint_dir_out(desc) &&
> 214 usb_endpoint_xfer_bulk(desc) &&
> 215 max_packet == 0x20)
> 216 bulk_out_endp = j;
> 217
> 218 if (bulk_out_endp != -1 && iso_in_endp != -1) {
> 219 tt->bulk_out_endp = bulk_out_endp;
> 220 tt->iso_in_endp = iso_in_endp;
> 221 altsetting = i;
> 222 break;
> 223 }
> 224 }
> 225 }
> 226
> 227 if (altsetting == -1) {
> 228 dev_err(&intf->dev, "cannot find expected altsetting\n");
> 229 ret = -ENODEV;
> 230 goto out;
> 231 }
> 232
> 233 tt->dev = &intf->dev;
> 234 tt->udev = interface_to_usbdev(intf);
> 235 tt->rc = rc;
> 236
> 237 ret = usb_set_interface(tt->udev, 0, altsetting);
> 238 if (ret)
> 239 goto out;
> 240
> 241 for (i = 0; i < NUM_URBS; i++) {
> 242 struct urb *urb = usb_alloc_urb(8, GFP_KERNEL);
> 243 void *buffer;
> 244
> 245 if (!urb) {
> 246 ret = -ENOMEM;
> 247 goto out;
> 248 }
> 249
> 250 urb->dev = tt->udev;
> 251 urb->context = tt;
> 252 urb->pipe = usb_rcvisocpipe(tt->udev, tt->iso_in_endp);
> 253 urb->interval = 1;
> 254 buffer = usb_alloc_coherent(tt->udev, 128, GFP_KERNEL,
> 255 &urb->transfer_dma);
> 256 if (!buffer) {
> 257 usb_free_urb(urb);
> 258 ret = -ENOMEM;
> 259 goto out;
> 260 }
> 261 urb->transfer_flags = URB_NO_TRANSFER_DMA_MAP | URB_ISO_ASAP;
> 262 urb->transfer_buffer = buffer;
> 263 urb->complete = ttusbir_urb_complete;
> 264 urb->number_of_packets = 8;
> 265 urb->transfer_buffer_length = 128;
> 266
> 267 for (j = 0; j < 8; j++) {
> 268 urb->iso_frame_desc[j].offset = j * 16;
> 269 urb->iso_frame_desc[j].length = 16;
> 270 }
> 271
> 272 tt->urb[i] = urb;
> 273 }
> 274
> 275 tt->bulk_urb = usb_alloc_urb(0, GFP_KERNEL);
> 276 if (!tt->bulk_urb) {
> 277 ret = -ENOMEM;
> 278 goto out;
> 279 }
> 280
> --> 281 tt->bulk_buffer[0] = 0xaa;
> ^^^^^^^^^^^^^^^^^^
>
> 282 tt->bulk_buffer[1] = 0x01;
>
> This email is a free service from the Smatch-CI project [smatch.sf.net].
>
> regards,
> dan carpenter
prev parent reply other threads:[~2026-04-10 20:06 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-04-10 10:12 [bug report] media: rc: ttusbir: respect DMA coherency rules Dan Carpenter
2026-04-10 19:59 ` Sean Young [this message]
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=adlWpq7_W0U0dgom@gofer.mess.org \
--to=sean@mess.org \
--cc=error27@gmail.com \
--cc=linux-media@vger.kernel.org \
--cc=oneukum@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