mbox series

[v3,0/2] fw_devlink overlay fix

Message ID 20240411235623.1260061-1-saravanak@google.com
Headers show
Series fw_devlink overlay fix | expand

Message

Saravana Kannan April 11, 2024, 11:56 p.m. UTC 
Overlays don't work correctly with fw_devlink. This patch series fixes
it. This series is now ready for review and merging once Geert and Herve
give they Tested-by.

Geert and Herve,

This patch series should hopefully fix both of your use cases [1][2][3].
Can you please check to make sure the device links created to/from the
overlay devices are to/from the right ones?

Thanks,
Saravana

Saravana Kannan (2):
  Revert "treewide: Fix probing of devices in DT overlays"
  of: dynamic: Fix overlayed devices not probing because of fw_devlink

 drivers/base/core.c       | 76 ++++++++++++++++++++++++++++++++++-----
 drivers/bus/imx-weim.c    |  6 ----
 drivers/i2c/i2c-core-of.c |  5 ---
 drivers/of/dynamic.c      |  1 -
 drivers/of/overlay.c      | 15 ++++++++
 drivers/of/platform.c     |  5 ---
 drivers/spi/spi.c         |  5 ---
 include/linux/fwnode.h    |  1 +
 8 files changed, 83 insertions(+), 31 deletions(-)

Comments

Herve Codina July 12, 2024, 1:34 p.m. UTC | #1 
Hi Saravana,

On Tue, 23 Apr 2024 10:39:24 +0200
Herve Codina <herve.codina@bootlin.com> wrote:

> Hi Saravana,
> 
> On Thu, 11 Apr 2024 16:56:20 -0700
> Saravana Kannan <saravanak@google.com> wrote:
> 
> > Overlays don't work correctly with fw_devlink. This patch series fixes
> > it. This series is now ready for review and merging once Geert and Herve
> > give they Tested-by.
> > 
> > Geert and Herve,
> > 
> > This patch series should hopefully fix both of your use cases [1][2][3].
> > Can you please check to make sure the device links created to/from the
> > overlay devices are to/from the right ones?
> > 
> > Thanks,
> > Saravana
> >   
> 
> I tested the series.
> 
> On my Microchip use case (i.e. DT overlay on a PCIe device), I observed that
> some driver removal were done in a wrong order. For instance, the onboard
> PCIe device interrupt controller (oic@e00c0120) was removed before its
> consumers.
> 
> I enabled debug traces in core.c and observed that many links were dropped.
> These links are related to pinctrl, clock, reset, interrupts, ...
> I have the feeling that these links should not be dropped.
> 

Have you made any progress on this topic ?

I haven't seen any updates.
Maybe I missed something.

Best regards,
Hervé