linux-stable.git/drivers/thunderbolt, branch v6.3.7

thunderbolt: Clear registers properly when auto clear isn't in use

2023-05-24T16:30:22+00:00

commit c4af8e3fecd03b0aedcd38145955605cfebe7e3a upstream.

When `QUIRK_AUTO_CLEAR_INT` isn't set, interrupt masking should be
cleared by writing to Interrupt Mask Clear (IMR) and interrupt
status should be cleared properly at shutdown/init.

This fixes an error where interrupts are left enabled during resume
from hibernation with `CONFIG_USB4=y`.

Fixes: 468c49f44759 ("thunderbolt: Disable interrupt auto clear for rings")
Cc: stable@vger.kernel.org # v6.3
Reported-by: Takashi Iwai 
Link: https://bugzilla.kernel.org/show_bug.cgi?id=217343
Signed-off-by: Mario Limonciello 
Signed-off-by: Mika Westerberg 
Signed-off-by: Greg Kroah-Hartman

thunderbolt: Rename shadowed variables bit to interrupt_bit and auto_clear_bit

2023-03-20T17:00:58+00:00

cppcheck reports
drivers/thunderbolt/nhi.c:74:7: style: Local variable 'bit' shadows outer variable [shadowVariable]
  int bit;
      ^
drivers/thunderbolt/nhi.c:66:6: note: Shadowed declaration
 int bit = ring_interrupt_index(ring) & 31;
     ^
drivers/thunderbolt/nhi.c:74:7: note: Shadow variable
  int bit;
      ^
For readablity rename the outer to interrupt_bit and the innner
to auto_clear_bit.

Fixes: 468c49f44759 ("thunderbolt: Disable interrupt auto clear for ring")
Cc: stable@vger.kernel.org
Signed-off-by: Tom Rix 
Signed-off-by: Mika Westerberg

thunderbolt: Disable interrupt auto clear for rings

2023-03-20T07:38:40+00:00

When interrupt auto clear is programmed, any read to the interrupt
status register will clear all interrupts.  If two interrupts have
come in before one can be serviced then this will cause lost interrupts.

On AMD USB4 routers this has manifested in odd problems particularly
with long strings of control tranfers such as reading the DROM via bit
banging.

Instead of clearing interrupts automatically, clear the bit corresponding
to the given ring's interrupt in the ISR.

Fixes: 7a1808f82a37 ("thunderbolt: Handle ring interrupt by reading interrupt status register")
Cc: Sanju Mehta 
Cc: stable@vger.kernel.org
Tested-by: Anson Tsao 
Signed-off-by: Mario Limonciello 
Signed-off-by: Mika Westerberg

thunderbolt: Use const qualifier for `ring_interrupt_index`

2023-03-20T07:38:07+00:00

`ring_interrupt_index` doesn't change the data for `ring` so mark it as
const. This is needed by the following patch that disables interrupt
auto clear for rings.

Cc: Sanju Mehta 
Cc: stable@vger.kernel.org
Signed-off-by: Mario Limonciello 
Signed-off-by: Mika Westerberg

thunderbolt: Use scale field when allocating USB3 bandwidth

2023-03-13T09:54:41+00:00

When tunneling aggregated USB3 (20 Gb/s) the bandwidth values that are
programmed to the ADP_USB3_CS_2 go higher than 4096 and that does not
fit anymore to the 12-bit field. Fix this by scaling the value using
the scale field accordingly.

Fixes: 3b1d8d577ca8 ("thunderbolt: Implement USB3 bandwidth negotiation routines")
Cc: stable@vger.kernel.org
Signed-off-by: Mika Westerberg

thunderbolt: Limit USB3 bandwidth of certain Intel USB4 host routers

2023-03-13T09:54:31+00:00

Current Intel USB4 host routers have hardware limitation that the USB3
bandwidth cannot go higher than 16376 Mb/s. Work this around by adding a
new quirk that limits the bandwidth for the affected host routers.

Cc: stable@vger.kernel.org
Signed-off-by: Gil Fine 
Signed-off-by: Mika Westerberg

thunderbolt: Call tb_check_quirks() after initializing adapters

2023-03-13T09:54:25+00:00

In order to apply quirks based on certain adapter types move call to
tb_check_quirks() happen after the adapters are initialized. This should
not affect the existing quirks.

Cc: stable@vger.kernel.org
Signed-off-by: Mika Westerberg

thunderbolt: Add missing UNSET_INBOUND_SBTX for retimer access

2023-03-13T09:54:20+00:00

According to USB4 retimer specification, the process of firmware update
sequence requires issuing a SET_INBOUND_SBTX port operation that later
shall be followed by UNSET_INBOUND_SBTX port operation. This last step
is not currently issued by the driver but it is necessary to make sure
the retimers are put back to passthrough mode even during enumeration.

If this step is missing the link may not come up properly after
soft-reboot for example.

For this reason issue UNSET_INBOUND_SBTX after SET_INBOUND_SBTX for
enumeration and also when the NVM upgrade is run.

Reported-by: Christian Schaubschläger 
Link: https://lore.kernel.org/linux-usb/b556f5ed-5ee8-9990-9910-afd60db93310@gmx.at/
Cc: stable@vger.kernel.org
Signed-off-by: Gil Fine 
Signed-off-by: Mika Westerberg

thunderbolt: Fix memory leak in margining

2023-03-13T09:54:14+00:00

Memory for the usb4->margining needs to be relased for the upstream port
of the router as well, even though the debugfs directory gets released
with the router device removal. Fix this.

Fixes: d0f1e0c2a699 ("thunderbolt: Add support for receiver lane margining")
Cc: stable@vger.kernel.org
Signed-off-by: Mika Westerberg

thunderbolt: Add quirk to disable CLx

2023-03-06T09:54:44+00:00

Add QUIRK_NO_CLX to disable the CLx state for hardware which
doesn't supports it.

AMD Yellow Carp and Pink Sardine don't support CLx state,
hence disabling it using QUIRK_NO_CLX.

Cc: stable@vger.kernel.org
Signed-off-by: Sanjay R Mehta 
Signed-off-by: Basavaraj Natikar 
[mw: added debug log when the quirk is run]
Signed-off-by: Mika Westerberg