From: Ping-Ke Shih <pkshih@realtek.com>
To: Bitterblue Smith <rtl8821cerfe2@gmail.com>,
"linux-wireless@vger.kernel.org" <linux-wireless@vger.kernel.org>
Cc: "i.mafifi17@gmail.com" <i.mafifi17@gmail.com>
Subject: RE: [PATCH rtw-next v3] wifi: rtw88: usb: Download the reserved page synchronously
Date: Wed, 16 Sep 2026 08:18:10 +0000 [thread overview]
Message-ID: <40bf5c51a1c24677bbd5e1ede8cbc5ce@realtek.com> (raw)
In-Reply-To: <5e0ce93e-0615-405e-956c-0f8dd8a543d9@gmail.com>
Bitterblue Smith <rtl8821cerfe2@gmail.com> wrote:
> @@ -499,16 +506,21 @@ static void rtw_usb_tx_queue_purge(struct rtw_usb *rtwusb)
>
> static void rtw_usb_write_port_complete(struct urb *urb)
> {
> - struct sk_buff *skb = urb->context;
> + struct rtw_usb_write_data_cb *cb = urb->context;
>
I think here is a racing with timeout case of wait_for_completion_timeout().
CPU0 CPU1
(rtw_usb_write_port_complete) (rtw_usb_write_data)
cb = urb->context
wait_for_completion_timeout() timeout
kfree(cb);
cb->status = urb->status;
(cb is free-after-use)
rtw89 handles this kind of problem by RCU. You can reference
rtw89_wait_for_cond_eval() to know how it does.
> - dev_kfree_skb_any(skb);
> + cb->status = urb->status;
> + complete(&cb->done);
> + dev_kfree_skb_any(cb->skb);
> + if (!cb->sync)
> + kfree(cb);
> }
>
> static int rtw_usb_write_data(struct rtw_dev *rtwdev,
> struct rtw_tx_pkt_info *pkt_info,
> - u8 *buf)
> + u8 *buf, bool sync)
> {
> const struct rtw_chip_info *chip = rtwdev->chip;
> + struct rtw_usb_write_data_cb *cb;
> struct sk_buff *skb;
> unsigned int size;
> u8 qsel;
> @@ -528,11 +540,29 @@ static int rtw_usb_write_data(struct rtw_dev *rtwdev,
> rtw_tx_fill_tx_desc(rtwdev, pkt_info, skb);
> rtw_tx_fill_txdesc_checksum(rtwdev, pkt_info, skb->data);
>
> + cb = kmalloc_obj(*cb, GFP_KERNEL);
> + if (!cb) {
> + dev_kfree_skb_any(skb);
> + return -ENOMEM;
> + }
> +
> + init_completion(&cb->done);
> + cb->status = -EINPROGRESS;
> + cb->skb = skb;
> + cb->sync = sync;
> +
> ret = rtw_usb_write_port(rtwdev, qsel, skb,
> - rtw_usb_write_port_complete, skb);
> + rtw_usb_write_port_complete, cb);
> if (unlikely(ret)) {
> rtw_err(rtwdev, "failed to do USB write, ret=%d\n", ret);
> dev_kfree_skb_any(skb);
> + kfree(cb);
> + } else if (sync) {
> + if (!wait_for_completion_timeout(&cb->done, secs_to_jiffies(5)))
> + ret = -ETIMEDOUT;
> + else
> + ret = cb->status;
> + kfree(cb);
> }
>
> return ret;
next prev parent reply other threads:[~2026-09-16 8:18 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-14 20:18 [PATCH rtw-next v3] wifi: rtw88: usb: Download the reserved page synchronously Bitterblue Smith
2026-09-16 8:18 ` Ping-Ke Shih [this message]
2026-09-16 9:00 ` Ping-Ke Shih
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=40bf5c51a1c24677bbd5e1ede8cbc5ce@realtek.com \
--to=pkshih@realtek.com \
--cc=i.mafifi17@gmail.com \
--cc=linux-wireless@vger.kernel.org \
--cc=rtl8821cerfe2@gmail.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