From: Lukas Wunner <lukas@wunner.de>
To: Peter Wu <peter@lekensteyn.nl>
Cc: Emil Velikov <emil.l.velikov@gmail.com>,
ML nouveau <nouveau@lists.freedesktop.org>,
ML dri-devel <dri-devel@lists.freedesktop.org>,
Dave Airlie <airlied@redhat.com>,
linux-pm@vger.kernel.org, Bjorn Helgaas <helgaas@kernel.org>,
Mika Westerberg <mika.westerberg@linux.intel.com>,
linux-pci@vger.kernel.org
Subject: Re: [Nouveau] [PATCH 4/4] drm/nouveau/acpi: fix lockup with PCIe runtime PM
Date: Sat, 28 May 2016 14:27:01 +0200 [thread overview]
Message-ID: <20160528122701.GA11464@wunner.de> (raw)
In-Reply-To: <20160527213123.GB1777@al>
Hi Peter,
On Fri, May 27, 2016 at 11:31:23PM +0200, Peter Wu wrote:
> On Fri, May 27, 2016 at 02:01:39PM +0100, Emil Velikov wrote:
> > On 24 May 2016 at 23:53, Peter Wu <peter@lekensteyn.nl> wrote:
[snip]
> > > @@ -273,14 +296,14 @@ static bool nouveau_dsm_detect(void)
> > > vga_count++;
> > >
> > > nouveau_dsm_pci_probe(pdev, &has_mux, &has_optimus,
> > > - &has_optimus_flags);
> > > + &has_optimus_flags, &has_power_resources);
> > > }
> > >
> > > while ((pdev = pci_get_class(PCI_CLASS_DISPLAY_3D << 8, pdev)) != NULL) {
> > > vga_count++;
> > >
> > > nouveau_dsm_pci_probe(pdev, &has_mux, &has_optimus,
> > > - &has_optimus_flags);
> > > + &has_optimus_flags, &has_power_resources);
> > > }
> > >
> > This and earlier patch break things in a subtle way.
> >
> > Namely: upon the second (and any later) call into the
> > nouveau_dsm_pci_probe() function, the had_foo flags are reset. Thus
> > only the specifics of the _final_ device are being used (at a later
> > stage). IMHO one should change that to "_any_ device", which will
> > match the original code and the actual intent further down in the
> > file.
>
> The flags are only reset if any of the MUX or Optimus handles are found.
> If both are missing, the flags are not overridden. This is from patch 1:
>
> + /* Does not look like a Nvidia device. */
> + if (!supports_mux && !supports_opt)
> + return;
>
> The reason why later calls override early ones is because some Optimus
> laptops have the _DSM method on both the Intel GPU (00:02.0) and the
> Nvidia one (01:00.0).
Sounds like you may want to check for pdev->vendor == PCI_VENDOR_ID_NVIDIA
or export pci_get_dev_by_id() and use that to match for class and vendor.
Best regards,
Lukas
WARNING: multiple messages have this Message-ID (diff)
From: Lukas Wunner <lukas-JFq808J9C/izQB+pC5nmwQ@public.gmane.org>
To: Peter Wu <peter-VTkQYDcBqhK7DlmcbJSQ7g@public.gmane.org>
Cc: linux-pm-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
ML nouveau
<nouveau-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org>,
Emil Velikov
<emil.l.velikov-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>,
ML dri-devel
<dri-devel-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org>,
Bjorn Helgaas <helgaas-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
linux-pci-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Dave Airlie <airlied-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
Mika Westerberg
<mika.westerberg-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
Subject: Re: [PATCH 4/4] drm/nouveau/acpi: fix lockup with PCIe runtime PM
Date: Sat, 28 May 2016 14:27:01 +0200 [thread overview]
Message-ID: <20160528122701.GA11464@wunner.de> (raw)
In-Reply-To: <20160527213123.GB1777@al>
Hi Peter,
On Fri, May 27, 2016 at 11:31:23PM +0200, Peter Wu wrote:
> On Fri, May 27, 2016 at 02:01:39PM +0100, Emil Velikov wrote:
> > On 24 May 2016 at 23:53, Peter Wu <peter@lekensteyn.nl> wrote:
[snip]
> > > @@ -273,14 +296,14 @@ static bool nouveau_dsm_detect(void)
> > > vga_count++;
> > >
> > > nouveau_dsm_pci_probe(pdev, &has_mux, &has_optimus,
> > > - &has_optimus_flags);
> > > + &has_optimus_flags, &has_power_resources);
> > > }
> > >
> > > while ((pdev = pci_get_class(PCI_CLASS_DISPLAY_3D << 8, pdev)) != NULL) {
> > > vga_count++;
> > >
> > > nouveau_dsm_pci_probe(pdev, &has_mux, &has_optimus,
> > > - &has_optimus_flags);
> > > + &has_optimus_flags, &has_power_resources);
> > > }
> > >
> > This and earlier patch break things in a subtle way.
> >
> > Namely: upon the second (and any later) call into the
> > nouveau_dsm_pci_probe() function, the had_foo flags are reset. Thus
> > only the specifics of the _final_ device are being used (at a later
> > stage). IMHO one should change that to "_any_ device", which will
> > match the original code and the actual intent further down in the
> > file.
>
> The flags are only reset if any of the MUX or Optimus handles are found.
> If both are missing, the flags are not overridden. This is from patch 1:
>
> + /* Does not look like a Nvidia device. */
> + if (!supports_mux && !supports_opt)
> + return;
>
> The reason why later calls override early ones is because some Optimus
> laptops have the _DSM method on both the Intel GPU (00:02.0) and the
> Nvidia one (01:00.0).
Sounds like you may want to check for pdev->vendor == PCI_VENDOR_ID_NVIDIA
or export pci_get_dev_by_id() and use that to match for class and vendor.
Best regards,
Lukas
_______________________________________________
Nouveau mailing list
Nouveau@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/nouveau
next prev parent reply other threads:[~2016-05-28 12:27 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-24 22:52 [PATCH 0/4] nouveau fixes for RPM/Optimus-related hangs Peter Wu
[not found] ` <1464130381-4797-1-git-send-email-peter-VTkQYDcBqhK7DlmcbJSQ7g@public.gmane.org>
2016-05-24 22:52 ` [PATCH 1/4] drm/nouveau/acpi: ensure matching ACPI handle and supported functions Peter Wu
2016-05-24 22:52 ` [PATCH 2/4] drm/nouveau/acpi: return supported DSM functions Peter Wu
2016-05-24 22:53 ` [PATCH 3/4] drm/nouveau/acpi: check for function 0x1B before using it Peter Wu
2016-05-24 22:53 ` [PATCH 4/4] drm/nouveau/acpi: fix lockup with PCIe runtime PM Peter Wu
2016-05-24 22:53 ` Peter Wu
2016-05-25 13:55 ` Mika Westerberg
2016-05-27 11:10 ` Peter Wu
2016-05-27 11:10 ` Peter Wu
2016-05-27 11:55 ` [Nouveau] " Hans de Goede
2016-05-30 9:57 ` Mika Westerberg
2016-05-30 9:57 ` Mika Westerberg
2016-05-30 12:20 ` Peter Wu
2016-05-30 12:20 ` Peter Wu
2016-05-30 13:09 ` Mika Westerberg
2016-05-30 13:09 ` Mika Westerberg
2016-05-30 16:13 ` Peter Wu
2016-05-30 16:13 ` Peter Wu
2016-05-31 8:43 ` Mika Westerberg
2016-05-31 11:02 ` Peter Wu
2016-05-31 11:02 ` Peter Wu
2016-06-01 9:28 ` Mika Westerberg
2016-06-01 17:21 ` Peter Wu
2016-06-01 17:21 ` Peter Wu
2016-05-31 12:20 ` [Nouveau] " Lukas Wunner
2016-05-31 12:20 ` Lukas Wunner
2016-06-01 16:51 ` [Nouveau] " Peter Wu
2016-06-01 16:51 ` Peter Wu
2016-06-01 17:40 ` [Nouveau] " Lukas Wunner
2016-05-27 13:01 ` Emil Velikov
2016-05-27 21:31 ` Peter Wu
2016-05-27 21:31 ` Peter Wu
2016-05-28 12:27 ` Lukas Wunner [this message]
2016-05-28 12:27 ` Lukas Wunner
2016-05-30 10:48 ` [Nouveau] " Emil Velikov
2016-05-30 11:23 ` Peter Wu
2016-05-30 11:23 ` Peter Wu
2016-05-30 12:41 ` [Nouveau] " Emil Velikov
2016-05-30 12:41 ` Emil Velikov
2016-05-25 9:08 ` [Nouveau] [PATCH 0/4] nouveau fixes for RPM/Optimus-related hangs Hans de Goede
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20160528122701.GA11464@wunner.de \
--to=lukas@wunner.de \
--cc=airlied@redhat.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=emil.l.velikov@gmail.com \
--cc=helgaas@kernel.org \
--cc=linux-pci@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=mika.westerberg@linux.intel.com \
--cc=nouveau@lists.freedesktop.org \
--cc=peter@lekensteyn.nl \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.