linux-kernel - Re: [PATCH] PCI: fu740: do not use clock name when requesting clock

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]

Date:   Mon, 12 Sep 2022 09:38:28 +0800
From:   Icenowy Zheng <uwu@...nowy.me>
To:     Conor.Dooley@...rochip.com, paul.walmsley@...ive.com,
        greentime.hu@...ive.com, lpieralisi@...nel.org, robh@...nel.org,
        kw@...ux.com, bhelgaas@...gle.com
Cc:     linux-pci@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-riscv@...ts.infradead.org
Subject: Re: [PATCH] PCI: fu740: do not use clock name when requesting clock

在 2022-09-08星期四的 18:14 +0000，Conor.Dooley@...rochip.com写道：
> On 07/09/2022 06:40, Icenowy Zheng wrote:
> > EXTERNAL EMAIL: Do not click links or open attachments unless you
> > know the content is safe
> > 
> > The DT binding of FU740 PCIe does not enforce a clock-names property,
> > and there exist some device tree that has a clock name that does not
> > stick to the one used by Linux DT (e.g. the one shipped with current
> > U-Boot mainline).
> 
> I recently added the missing enforcement:
> https://git.kernel.org/pub/scm/linux/kernel/git/lpieralisi/pci.git/commit/?h=pci/dt&id=b408fad61d34c765c3e01895286332af2d50402a

Unfortunately binding w/o clock-names enforcement has already entered a
stable release (5.19), and the real clock name "pcie_aux" is never
enforced before (there's a DT in U-Boot that uses "pcieaux" instead),
should this be considered as breakage to stable DT binding?

Anyway, I had sent out a patch that synchorizes all FU740-related DT
files to U-Boot, see [1].

[1]
https://lore.kernel.org/all/20220825081119.1694007-2-uwu@icenowy.me/

> 
> Since there's only one clock though, I'd imagine it makes little to no
> real difference if the check here is relaxed.
> 
> Reviewed-by: Conor Dooley <conor.dooley@...rochip.com>
> 
> > 
> > Drop the name in the clock request, instead just pass NULL (because
> > this device should have only a single clock).
> > 
> > Signed-off-by: Icenowy Zheng <uwu@...nowy.me>
> > ---
> >  drivers/pci/controller/dwc/pcie-fu740.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/drivers/pci/controller/dwc/pcie-fu740.c
> > b/drivers/pci/controller/dwc/pcie-fu740.c
> > index 0c90583c078b..edb218a37a4f 100644
> > --- a/drivers/pci/controller/dwc/pcie-fu740.c
> > +++ b/drivers/pci/controller/dwc/pcie-fu740.c
> > @@ -315,7 +315,7 @@ static int fu740_pcie_probe(struct
> > platform_device *pdev)
> >                 return dev_err_probe(dev, PTR_ERR(afp->pwren),
> > "unable to get pwren-gpios\n");
> > 
> >         /* Fetch clocks */
> > -       afp->pcie_aux = devm_clk_get(dev, "pcie_aux");
> > +       afp->pcie_aux = devm_clk_get(dev, NULL);
> >         if (IS_ERR(afp->pcie_aux))
> >                 return dev_err_probe(dev, PTR_ERR(afp->pcie_aux),
> >                                              "pcie_aux clock source
> > missing or invalid\n");
> > --
> > 2.37.1
> > 
>