Hi,
On 11/3/23 20:07, Mario Limonciello wrote:
> The `is_thunderbolt` bit has been used to indicate that a PCIe device
> contained the Intel VSEC which is used by various parts of the kernel
> to change behavior. To later allow usage with USB4 controllers as well,
> rename this to `is_tunneled`.
On Fri, Nov 03, 2023 at 02:07:49PM -0500, Mario Limonciello wrote:
> Downstream drivers are getting the wrong values from
> pcie_bandwidth_available() which is causing problems for performance
> of eGPUs.
>
> This series overhauls Thunderbolt related device detection and uses
> the changes to
The PCI root port used for tunneling USB4 traffic on Tiger Lake is
is not marked as tunneling but has the same limitations as other
PCIe root ports used for tunneling.
This causes pcie_bandwidth_available() to treat it as the limiting
device in the PCI hierarchy and downstream driver to program
The USB4 spec specifies that PCIe ports that are used for tunneling
PCIe traffic over USB4 fabric will be hardcoded to advertise 2.5GT/s and
behave as a PCIe Gen1 device. The actual performance of these ports is
controlled by the fabric implementation.
Downstream drivers such as amdgpu which
USB4 routers support a feature called "PCIe tunneling". This
allows PCIe traffic to be transmitted over USB4 fabric.
PCIe root ports that are used in this fashion can be discovered
by device specific data that specifies the USB4 router they are
connected to. For the PCI core, the specific
The `is_thunderbolt` bit has been used to indicate that a PCIe device
contained the Intel VSEC which is used by various parts of the kernel
to change behavior. To later allow usage with USB4 controllers as well,
rename this to `is_tunneled`.
Signed-off-by: Mario Limonciello
---
commit 493fb50e958c ("PCI: pciehp: Assume NoCompl+ for Thunderbolt
ports") added a check into pciehp code to explicitly set NoCompl+
for all Intel Thunderbolt controllers, including those that don't
need it.
This overloaded the purpose of the `is_thunderbolt` member of
`struct pci_device` because
All callers have switched to dev_is_removable() for detecting
hotpluggable PCIe devices.
Signed-off-by: Mario Limonciello
---
include/linux/pci.h | 22 --
1 file changed, 22 deletions(-)
diff --git a/include/linux/pci.h b/include/linux/pci.h
index b56417276042..530b0a360514
`PCI_CLASS_SERIAL_USB_USB4` may be used by code outside of thunderbolt.
Move the declaration into the common pci_ids.h header.
Acked-by: Mika Westerberberg
Signed-off-by: Mario Limonciello
---
drivers/thunderbolt/nhi.h | 2 --
include/linux/pci_ids.h | 1 +
2 files changed, 1 insertion(+), 2
pci_is_thunderbolt_attached() only works for Intel TBT devices. Switch to
using dev_is_removable() to be able to detect USB4 devices as well.
Signed-off-by: Mario Limonciello
---
drivers/gpu/drm/nouveau/nouveau_vga.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git
pci_is_thunderbolt_attached() only works for Intel TBT devices. Switch to
using dev_is_removable() to be able to detect USB4 devices as well.
Signed-off-by: Mario Limonciello
---
drivers/gpu/drm/radeon/radeon_device.c | 4 ++--
drivers/gpu/drm/radeon/radeon_kms.c| 2 +-
2 files changed, 3
Downstream drivers are getting the wrong values from
pcie_bandwidth_available() which is causing problems for performance
of eGPUs.
This series overhauls Thunderbolt related device detection and uses
the changes to change the behavior of pcie_bandwidth_available().
NOTE: This series is currently
On 10/3/23 10:44, Timothy Madden wrote:
Hello,
Whenever I try to run a graphical application with my RTX 3090 Strix I
get the following error in the console output:
nouveau: kernel rejected pushbuf: No such device
nouveau: ch23: krec 0 pushes 1 bufs 21 relocs 0
Is there maybe a
On 11/3/23 15:04, Christian König wrote:
Am 03.11.23 um 14:14 schrieb Danilo Krummrich:
On Fri, Nov 03, 2023 at 08:18:35AM +0100, Christian König wrote:
Am 02.11.23 um 00:31 schrieb Danilo Krummrich:
Implement reference counting for struct drm_gpuvm.
From the design point of view what is
Am 03.11.23 um 14:14 schrieb Danilo Krummrich:
On Fri, Nov 03, 2023 at 08:18:35AM +0100, Christian König wrote:
Am 02.11.23 um 00:31 schrieb Danilo Krummrich:
Implement reference counting for struct drm_gpuvm.
From the design point of view what is that good for?
It was discussed in this
On Fri, Nov 03, 2023 at 08:18:35AM +0100, Christian König wrote:
> Am 02.11.23 um 00:31 schrieb Danilo Krummrich:
> > Implement reference counting for struct drm_gpuvm.
>
> From the design point of view what is that good for?
It was discussed in this thread [1].
Essentially, the idea is to make
Am 02.11.23 um 00:31 schrieb Danilo Krummrich:
Implement reference counting for struct drm_gpuvm.
From the design point of view what is that good for?
Background is that the most common use case I see is that this object is
embedded into something else and a reference count is then not
17 matches
Mail list logo