linux-kernel - Re: [PATCH 2/2] PCI/DPC: Use defines with register fields

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]

Date:   Mon, 16 Oct 2023 15:53:11 +0300 (EEST)
From:   Ilpo Järvinen <ilpo.jarvinen@...ux.intel.com>
To:     Bjorn Helgaas <helgaas@...nel.org>
cc:     linux-pci@...r.kernel.org, Bjorn Helgaas <bhelgaas@...gle.com>,
        Lorenzo Pieralisi <lorenzo.pieralisi@....com>,
        Krzysztof Wilczyński <kw@...ux.com>,
        Jonathan Cameron <Jonathan.Cameron@...wei.com>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 2/2] PCI/DPC: Use defines with register fields

On Fri, 13 Oct 2023, Bjorn Helgaas wrote:

> On Fri, Oct 13, 2023 at 02:20:04PM +0300, Ilpo Järvinen wrote:
> > Use defines instead of literals and replace custom masking and shifts
> > with FIELD_GET() where it makes sense.
> > ...
> 
> >  	pci_read_config_word(pdev, pdev->dpc_cap + PCI_EXP_DPC_CAP, &cap);
> > -	pci_read_config_word(pdev, pdev->dpc_cap + PCI_EXP_DPC_CTL, &ctl);
> >  
> > -	ctl = (ctl & 0xfff4) | PCI_EXP_DPC_CTL_EN_FATAL | PCI_EXP_DPC_CTL_INT_EN;
> > +	pci_read_config_word(pdev, pdev->dpc_cap + PCI_EXP_DPC_CTL, &ctl);
> > +	ctl |= PCI_EXP_DPC_CTL_EN_FATAL | PCI_EXP_DPC_CTL_INT_EN;
> > +	ctl &= ~PCI_EXP_DPC_CTL_EN_NONFATAL;
> 
> This has been a little obtuse from the beginning.
> 
> The original clears bits 0, 1, 3, then sets bits 0 and 3.
> The new code sets bits 0, 3, then clears bit 1.
> 
> These are equivalent, but it's definitely some work to verify it.
>
> I think the point is to enable DPC on ERR_FATAL (but not ERR_NONFATAL)
> and to enable DPC interrupts.  What about something like this?
>   #define PCI_EXP_DPC_CTL_EN_MASK  0x0003
> 
>   ctl &= ~PCI_EXP_DPC_CTL_EN_MASK;
>   ctl |= PCI_EXP_DPC_CTL_EN_FATAL | PCI_EXP_DPC_CTL_INT_EN;

Thanks for the suggestion, it looks cleaner, yes. I realized the bit 
changes weren't as obvious as they could be but I failed to see I could 
add such a combined mask to make the intent dead obvious.

With a small change replacing the 0x0003 literal with the combined defines 
of the actual bits it seems very clear what's going on. I'll go for that. 

I'll place my non-FIELD_GET() changes on DPC top of your 
FIELD_GET/PREP()-only change so each patch can be smaller and more 
focused.

-- 
 i.


> >  	pci_write_config_word(pdev, pdev->dpc_cap + PCI_EXP_DPC_CTL, ctl);
> > -	pci_info(pdev, "enabled with IRQ %d\n", dev->irq);
> >  
> > +	pci_info(pdev, "enabled with IRQ %d\n", dev->irq);
> >  	pci_info(pdev, "error containment capabilities: Int Msg #%d, RPExt%c PoisonedTLP%c SwTrigger%c RP PIO Log %d, DL_ActiveErr%c\n",
> >  		 cap & PCI_EXP_DPC_IRQ, FLAG(cap, PCI_EXP_DPC_CAP_RP_EXT),
> >  		 FLAG(cap, PCI_EXP_DPC_CAP_POISONED_TLP),
> > -- 
> > 2.30.2
> > 
>