linux-kernel - Re: [PATCH net v4] nfc: nci: Fix uninit-value in nci_rx

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: <Zj9QRIjGLbVdd7MX@zeus>
Date: Sat, 11 May 2024 20:02:28 +0900
From: Ryosuke Yasuoka <ryasuoka@...hat.com>
To: Jakub Kicinski <kuba@...nel.org>
Cc: krzk@...nel.org, davem@...emloft.net, edumazet@...gle.com,
	pabeni@...hat.com, horms@...nel.org, netdev@...r.kernel.org,
	linux-kernel@...r.kernel.org, syoshida@...hat.com,
	syzbot+d7b4dc6cd50410152534@...kaller.appspotmail.com
Subject: Re: [PATCH net v4] nfc: nci: Fix uninit-value in nci_rx_work

Thank you for your review.

On Fri, May 10, 2024 at 07:06:13PM -0700, Jakub Kicinski wrote:
> On Thu,  9 May 2024 20:30:33 +0900 Ryosuke Yasuoka wrote:
> > -		if (!nci_plen(skb->data)) {
> > +		if (!skb->len) {
> >  			kfree_skb(skb);
> > -			kcov_remote_stop();
> > -			break;
> > +			continue;
> 
> the change from break to continue looks unrelated

OK. I'll leave this break in this patch. I'll send another patch about
it.

> >  		}
> 
> > -			nci_ntf_packet(ndev, skb);
> > +			if (nci_valid_size(skb, NCI_CTRL_HDR_SIZE))
> 
> > +			if (nci_valid_size(skb, NCI_DATA_HDR_SIZE))
> 
> 
> #define NCI_CTRL_HDR_SIZE                                       3
> #define NCI_DATA_HDR_SIZE                                       3
> 
> you can add a BUILD_BUG_ON(NCI_CTRL_HDR_SIZE == NCI_DATA_HDR_SIZE)
> and save all the code duplication.
> 

Sorry I don't get it. Do you mean I just insert
BUILD_BUG_ON(NCI_CTRL_HDR_SIZE != NCI_DATA_HDR_SIZE) or insert this and
clean up the code duplication like this? (It is just a draft. I just
share what I mean.) I can avoid to call nci_valid_size() repeatedly
inside the switch statement.

static void nci_rx_work(struct work_struct *work)
{
..
		if (!skb->len) {
			kfree_skb(skb);
			kcov_remote_stop();
			break;
		}

		BUILD_BUG_ON(NCI_CTRL_HDR_SIZE != NCI_DATA_HDR_SIZE);
		unsigned int hdr_size = NCI_CTRL_HDR_SIZE;

		if (!nci_valid_size(skb, hdr_size)) {
			kfree_skb(skb);
			continue;
		}

		/* Process frame */
		switch (nci_mt(skb->data)) {
		case NCI_MT_RSP_PKT:
			nci_rsp_packet(ndev, skb);
			break;

		case NCI_MT_NTF_PKT:
			nci_ntf_packet(ndev, skb);
			break;