netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH net-next v3 0/2] net: sfp: skip PHY probe on bridge-less OEM SFP-10G-T
@ 2026-11-20  6:37 Daniel Pawlik
  2026-11-20  6:37 ` [PATCH net-next v3 1/2] net: sfp: move sfp_strlen() above the quirk fixups Daniel Pawlik
  2026-11-20  6:37 ` [PATCH net-next v3 2/2] net: sfp: skip PHY probe on bridge-less OEM SFP-10G-T revisions A and B Daniel Pawlik
  0 siblings, 2 replies; 5+ messages in thread
From: Daniel Pawlik @ 2026-11-20  6:37 UTC (permalink / raw)
  To: andrew, hkallweit1, linux
  Cc: netdev, andrew+netdev, davem, edumazet, kuba, pabeni, horms,
	kabel, petr.wozniak, maxime.chevallier, olek2, Daniel Pawlik

Some "OEM SFP-10G-T" copper modules have no RollBall I2C-to-MDIO
bridge, but the SFP-10G-T quirk forces RollBall on all of them. The PHY
probe then retries for over nine minutes under rtnl_lock before the
module links up without a PHY. This series skips the PHY probe for the
bridge-less revisions A and B, and leaves other revisions, which do
have a bridge, unchanged.

Patch 1 moves sfp_strlen() so the fixup can use it, with no functional
change. Patch 2 adds the fixup.

This does not re-introduce the generic RollBall bridge probe reverted
in commit b521003c27eb ("Revert "net: phy: sfp: probe for RollBall
I2C-to-MDIO bridge in mdio-i2c""): the probe path, its timeouts and the
MDIO bus creation for genuine RollBall modules are untouched. Further
bridge-less revisions of this part number may exist; they can be added
once their lack of a bridge is confirmed on hardware.

Changes in v3:
- Target net-next and drop the Fixes: tag: rev A never worked without
  the stall, so this is not a regression (Andrew Lunn).
- Split the sfp_strlen() move into its own patch (Andrew Lunn).

Changes in v2:
- Match the vendor revision with sfp_strlen(), so NUL-padded revisions
  are accepted like the rest of the quirk matching.
- Spell out in the commit message how the rev B module reaches this
  quirk entry (its part number depends on the peer speed).

v2: https://lore.kernel.org/netdev/20261118221450.1764790-1-pawlik.dan@gmail.com/
v1: https://lore.kernel.org/netdev/20261116093605.294365-1-pawlik.dan@gmail.com/

Daniel Pawlik (2):
  net: sfp: move sfp_strlen() above the quirk fixups
  net: sfp: skip PHY probe on bridge-less OEM SFP-10G-T revisions A and
    B

 drivers/net/phy/sfp.c | 54 +++++++++++++++++++++++++++++++------------
 1 file changed, 39 insertions(+), 15 deletions(-)

-- 
2.56.1


^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2026-11-20 12:56 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-20  6:37 [PATCH net-next v3 0/2] net: sfp: skip PHY probe on bridge-less OEM SFP-10G-T Daniel Pawlik
2026-11-20  6:37 ` [PATCH net-next v3 1/2] net: sfp: move sfp_strlen() above the quirk fixups Daniel Pawlik
2026-11-20 12:52   ` Andrew Lunn
2026-11-20  6:37 ` [PATCH net-next v3 2/2] net: sfp: skip PHY probe on bridge-less OEM SFP-10G-T revisions A and B Daniel Pawlik
2026-11-20 12:56   ` Andrew Lunn

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).