linux-kernel - Re: [PATCH 2/2] PCI: uniphier: Add checking whether PERST# is deasserted

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:   Thu, 7 Nov 2019 12:46:17 +0000
From:   Andrew Murray <andrew.murray@....com>
To:     Kunihiko Hayashi <hayashi.kunihiko@...ionext.com>
Cc:     Lorenzo Pieralisi <lorenzo.pieralisi@....com>,
        Bjorn Helgaas <bhelgaas@...gle.com>, linux-pci@...r.kernel.org,
        linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
        Masami Hiramatsu <masami.hiramatsu@...aro.org>,
        Jassi Brar <jaswinder.singh@...aro.org>
Subject: Re: [PATCH 2/2] PCI: uniphier: Add checking whether PERST# is
 deasserted

On Thu, Nov 07, 2019 at 08:52:39PM +0900, Kunihiko Hayashi wrote:
> Hi Andrew,
> Thank you for your comments.
> 
> On Thu, 7 Nov 2019 10:02:08 +0000 <andrew.murray@....com> wrote:
> 
> > On Thu, Nov 07, 2019 at 01:58:15PM +0900, Kunihiko Hayashi wrote:
> > > When PERST# is asserted once, EP configuration will be initialized.
> > 
> > I don't quite understand this - does the EP/RC mode depend on how often
> > PERST# is toggled?
> 
> I think of connecting this RC controller and EP based on `Linux PCI
> endpoint framework' in another machine.
> 
> While this RC driver is probing, the EP driver might be also probing and
> configurating itself using configfs. If PERST# is toggled after the EP
> has done its configuration, this configuration will be lost.
> 
> I expect that the EP configurates after RC has toggled PERST#, however,
> there is no way to synchronize both of them.
> 

OK I understand where you are coming from now. Please ensure the commit
message gives this rationale.

However, If I understand correctly, doesn't your solution only work some
of the time? What happens if you boot both machines at the same time,
and PERST# isn't asserted prior to the kernel booting?

The only way you can ensure the EP is started after the RC is initialised
is to start the EP after the RC is initialised.

I'm not sure what the solution is here, but it feels like this approach
only partially solves it.

> 
> > 
> > > If PERST# has been already deasserted, it isn't necessary to assert
> > > here.
> > 
> > What is the motativation for this patch? Is it to avoid a delay during
> > boot, or to fix some bug? Isn't it nice to always reset the IP before
> > use anyway?
> 
> Since EP device usually works without its configuration after PERST#,
> deassering PERST# here is no problem. Since bus reset breaks configuration
> of EP as shown above, bus reset should be done before EP has probed.
> Maybe boot sequence in host machine will do.
> 
> 
> >
> > > 
> > > This checks whether PERST# is deasserted using PCL_PINMON register,
> > > and adds omit controlling PERST#.
> > 
> > Should this read 'and omits controlling PERST#'?
> 
> Yes, I'll fix it.
> 
> 
> > 
> > > 
> > > Signed-off-by: Kunihiko Hayashi <hayashi.kunihiko@...ionext.com>
> > > ---
> > >  drivers/pci/controller/dwc/pcie-uniphier.c | 8 ++++++++
> > >  1 file changed, 8 insertions(+)
> > > 
> > > diff --git a/drivers/pci/controller/dwc/pcie-uniphier.c b/drivers/pci/controller/dwc/pcie-uniphier.c
> > > index 8fd7bad..1ea4220 100644
> > > --- a/drivers/pci/controller/dwc/pcie-uniphier.c
> > > +++ b/drivers/pci/controller/dwc/pcie-uniphier.c
> > > @@ -22,6 +22,9 @@
> > >  
> > >  #include "pcie-designware.h"
> > >  
> > > +#define PCL_PINMON			0x0028
> > > +#define PCL_PINMON_PERST_OUT		BIT(16)
> > > +
> > >  #define PCL_PINCTRL0			0x002c
> > >  #define PCL_PERST_PLDN_REGEN		BIT(12)
> > >  #define PCL_PERST_NOE_REGEN		BIT(11)
> > > @@ -100,6 +103,11 @@ static void uniphier_pcie_init_rc(struct uniphier_pcie_priv *priv)
> > >  	val |= PCL_SYS_AUX_PWR_DET;
> > >  	writel(val, priv->base + PCL_APP_PM0);
> > >  
> > > +	/* return if PERST# is already deasserted */
> > 
> > This comment just describes what the following line does which may be
> > self-explanatory, perhaps a better comment would describe why we avoid
> > a reset.
> 
> Indeed. I'll write the reason here.
> 
Thanks,

Andrew Murray

> Thank you,
> 
> ---
> Best Regards,
> Kunihiko Hayashi
>