linux-kernel - Re: [PATCH v2] Bluetooth: btqca: sequential validation

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: <42E81D62-346C-4AC0-9B97-A54D67482331@holtmann.org>
Date:   Thu, 9 Dec 2021 11:45:22 +0100
From:   Marcel Holtmann <marcel@...tmann.org>
To:     Sai Teja Aluvala <quic_saluvala@...cinc.com>
Cc:     Johan Hedberg <johan.hedberg@...il.com>,
        Matthias Kaehlcke <mka@...omium.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        linux-bluetooth <linux-bluetooth@...r.kernel.org>,
        quic_hemantg@...cinc.com, MSM <linux-arm-msm@...r.kernel.org>,
        quic_bgodavar@...cinc.com, Rocky Liao <rjliao@...eaurora.org>,
        hbandi@...eaurora.org,
        Abhishek Pandit-Subedi <abhishekpandit@...omium.org>,
        Miao-chen Chou <mcchou@...omium.org>, quic_pharish@...cinc.com
Subject: Re: [PATCH v2] Bluetooth: btqca: sequential validation

Hi Sai,

> This change will have sequential validation support
> & patch config command is added
> 
> Signed-off-by: Sai Teja Aluvala <quic_saluvala@...cinc.com>
> ---
> drivers/bluetooth/btqca.c | 50 +++++++++++++++++++++++++++++++++++++++++++++++
> drivers/bluetooth/btqca.h |  2 ++
> 2 files changed, 52 insertions(+)
> 
> diff --git a/drivers/bluetooth/btqca.c b/drivers/bluetooth/btqca.c
> index be04d74..633a24c 100644
> --- a/drivers/bluetooth/btqca.c
> +++ b/drivers/bluetooth/btqca.c
> @@ -141,6 +141,53 @@ static int qca_read_fw_build_info(struct hci_dev *hdev)
> 	return err;
> }
> 
> +static int qca_send_patch_config_cmd(struct hci_dev *hdev)
> +{
> +	struct sk_buff *skb;
> +	int err = 0;
> +	u8 cmd[] = {EDL_PATCH_CONFIG_CMD, 0x01, 0, 0, 0};

is there are reason not to address review comments?

	const u8 cmd[] = { EDL_PATCH_CONFIG_CMD, 0x01, 0, 0, 0 };

> +	u8 rlen = 0x02;
> +	struct edl_event_hdr *edl;
> +	u8 rtype = EDL_PATCH_CONFIG_CMD;

I missed these two and must have assumed there are used somewhere, rlen and rtype are a waste to be declared.

> +
> +	bt_dev_dbg(hdev, "QCA Patch config");
> +
> +	skb = __hci_cmd_sync_ev(hdev, EDL_PATCH_CMD_OPCODE, sizeof(cmd),
> +				cmd, HCI_EV_VENDOR, HCI_INIT_TIMEOUT);
> +	if (IS_ERR(skb)) {
> +		err = PTR_ERR(skb);
> +		bt_dev_err(hdev, "Sending QCA Patch config failed (%d)", err);
> +		return err;
> +	}
> +
> +	if (skb->len != rlen) {

	if (skb->len != 2) {

> +		bt_dev_err(hdev, "QCA Patch config cmd size mismatch len %d", skb->len);
> +		err = -EILSEQ;
> +		goto out;
> +	}
> +
> +	edl = (struct edl_event_hdr *)(skb->data);
> +	if (!edl) {
> +		bt_dev_err(hdev, "QCA Patch config with no header");
> +		err = -EILSEQ;
> +		goto out;
> +	}
> +
> +	if (edl->cresp != EDL_PATCH_CONFIG_RES_EVT || edl->rtype != rtype) {

	     edl->rtype != EDL_PATCH_CONFIG_CMD) {

> +		bt_dev_err(hdev, "QCA Wrong packet received %d %d", edl->cresp,
> +			   edl->rtype);
> +		err = -EIO;
> +		goto out;
> +	}
> +
> +out:
> +	kfree(skb);
> +	if (err)
> +		bt_dev_err(hdev, "QCA Patch config cmd failed (%d)", err);
> +
> +	return err;
> +}
> +

Regards

Marcel