netdev - RE: [PATCH net v2 2/2] net/txgbe: add extra handle for MSI/INTx into thread irq handle

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] [day] [month] [year] [list]

Message-ID: <053701dac901$2e068360$8a138a20$@trustnetic.com>
Date: Fri, 28 Jun 2024 10:16:26 +0800
From: Jiawen Wu <jiawenwu@...stnetic.com>
To: "'Jakub Kicinski'" <kuba@...nel.org>
Cc: <davem@...emloft.net>,
	<edumazet@...gle.com>,
	<pabeni@...hat.com>,
	<horms@...nel.org>,
	<andrew@...n.ch>,
	<netdev@...r.kernel.org>,
	<przemyslaw.kitszel@...el.com>,
	<mengyuanlou@...-swift.com>,
	<duanqiangwen@...-swift.com>
Subject: RE: [PATCH net v2 2/2] net/txgbe: add extra handle for MSI/INTx into thread irq handle

On Fri, Jun 28, 2024 7:44 AM, Jakub Kicinski wrote:
> On Wed, 26 Jun 2024 14:07:03 +0800 Jiawen Wu wrote:
> > Moreover, do not free isb resources in .ndo_stop, to avoid reading
> > memory by a null pointer.
> 
> Please provide more detail on the sequence of events leading to the
> null-defer.
> 
> >  	pdev->irq = pci_irq_vector(pdev, 0);
> > +	wx->num_q_vectors = 1;
> 
> this doesn't seem obviously related

Umm, this is related another fix for MSI/INTx. I'll split it into a separate
commit.

> 
> >
> >  	return 0;
> >  }
> > @@ -2027,6 +2028,9 @@ int wx_setup_isb_resources(struct wx *wx)
> >  {
> >  	struct pci_dev *pdev = wx->pdev;
> >
> > +	if (wx->isb_mem)
> > +		return 0;
> > +
> >  	wx->isb_mem = dma_alloc_coherent(&pdev->dev,
> >  					 sizeof(u32) * 4,
> >  					 &wx->isb_dma,
> > @@ -2050,6 +2054,9 @@ void wx_free_isb_resources(struct wx *wx)
> >  {
> >  	struct pci_dev *pdev = wx->pdev;
> >
> > +	if (!wx->isb_mem)
> > +		return;
> > +
> 
> And neither does this. Why do you need to make these function
> idempotent?

This is also to implement txgbe changes without changing the flow of ngbe.