linux-kernel - Re: [PATCH 1/1] PCI/bwctrl: Remove IRQF

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <ec7f9169-26dc-cc0e-e321-b66ca9d3f40e@linux.intel.com>
Date: Sat, 16 Nov 2024 16:40:51 +0200 (EET)
From: Ilpo Järvinen <ilpo.jarvinen@...ux.intel.com>
To: Bjorn Helgaas <helgaas@...nel.org>
cc: Stefan Wahren <wahrenst@....net>, 
    Florian Fainelli <florian.fainelli@...adcom.com>, 
    Lukas Wunner <lukas@...ner.de>, Bjorn Helgaas <bhelgaas@...gle.com>, 
    Jonathan Cameron <Jonathan.Cameron@...wei.com>, linux-pci@...r.kernel.org, 
    LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 1/1] PCI/bwctrl: Remove IRQF_ONESHOT and handle hardirqs
 instead

On Fri, 15 Nov 2024, Bjorn Helgaas wrote:

> On Fri, Nov 15, 2024 at 06:57:17PM +0200, Ilpo Järvinen wrote:
> > bwctrl cannot use IRQF_ONESHOT because it shares interrupt with other
> > service drivers that are not using IRQF_ONESHOT nor compatible with it.
> > 
> > Remove IRQF_ONESHOT from bwctrl and convert the irq thread to hardirq
> > handler. Rename the handler to pcie_bwnotif_irq() to indicate its new
> > purpose.
> > 
> > The IRQ handler is simple enough to not require not require other
> > changes.
> > 
> > Fixes: 058a4cb11620 ("PCI/bwctrl: Re-add BW notification portdrv as PCIe BW controller")
> > Reported-by: Stefan Wahren <wahrenst@....net>
> > Link: https://lore.kernel.org/linux-pci/dcd660fd-a265-4f47-8696-776a85e097a0@gmx.net/
> > Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@...ux.intel.com>
> 
> Squashed into 058a4cb11620, thanks!
> 
> Also added your tested-by, Stefan, thanks very much for doing that!

Hi Bjorn,

You might want to also remove "3) ..." part from the commit message as it 
still refers to threaded IRQ and IRQF_ONESHOT so it won't confuse anybody 
when looking at this years from now :-).

-- 
 i.

> > ---
> >  drivers/pci/pcie/bwctrl.c | 8 +++-----
> >  1 file changed, 3 insertions(+), 5 deletions(-)
> > 
> > diff --git a/drivers/pci/pcie/bwctrl.c b/drivers/pci/pcie/bwctrl.c
> > index ff5d12e01f9c..a6c65bbe3735 100644
> > --- a/drivers/pci/pcie/bwctrl.c
> > +++ b/drivers/pci/pcie/bwctrl.c
> > @@ -230,7 +230,7 @@ static void pcie_bwnotif_disable(struct pci_dev *port)
> >  				   PCI_EXP_LNKCTL_LBMIE | PCI_EXP_LNKCTL_LABIE);
> >  }
> >  
> > -static irqreturn_t pcie_bwnotif_irq_thread(int irq, void *context)
> > +static irqreturn_t pcie_bwnotif_irq(int irq, void *context)
> >  {
> >  	struct pcie_device *srv = context;
> >  	struct pcie_bwctrl_data *data = srv->port->link_bwctrl;
> > @@ -302,10 +302,8 @@ static int pcie_bwnotif_probe(struct pcie_device *srv)
> >  	if (ret)
> >  		return ret;
> >  
> > -	ret = devm_request_threaded_irq(&srv->device, srv->irq, NULL,
> > -					pcie_bwnotif_irq_thread,
> > -					IRQF_SHARED | IRQF_ONESHOT,
> > -					"PCIe bwctrl", srv);
> > +	ret = devm_request_irq(&srv->device, srv->irq, pcie_bwnotif_irq,
> > +			       IRQF_SHARED, "PCIe bwctrl", srv);
> >  	if (ret)
> >  		return ret;
> >  
> > -- 
> > 2.39.5
> > 
>