From 6ae520c1ba4d2f78d4cbd1c099e178503dbcef3b Mon Sep 17 00:00:00 2001 From: Simone Date: Tue, 28 Jul 2026 23:09:53 +0200 Subject: [PATCH] i2c-imc-skylake: the i2cdetect example shows what the driver does now The documentation's sample scan predates SPD instantiation at probe. It shows 0x36, 0x50 and 0x52 as detected addresses, but the driver claims them itself now, so a real run reports UU there and 0x37 as well. The example printed output the driver can no longer produce. Measured on the X299 system, both channels identical: 0 1 2 3 4 5 6 7 8 9 a b c d e f 20: -- -- -- -- -- -- -- 27 -- -- -- -- -- -- -- -- 30: 30 31 -- -- 34 35 UU UU -- -- -- -- -- -- -- -- 50: UU -- UU -- -- -- -- -- -- -- -- -- -- -- -- -- Say what the UU entries are, since the section that explains instantiation comes further down and a reader scanning the bus meets them first. --- docs/submission/i2c-imc-skylake.rst | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/docs/submission/i2c-imc-skylake.rst b/docs/submission/i2c-imc-skylake.rst index a1babbb..b26cba6 100644 --- a/docs/submission/i2c-imc-skylake.rst +++ b/docs/submission/i2c-imc-skylake.rst @@ -112,8 +112,13 @@ read mode:: $ i2cdetect -y -r 6 0 1 2 3 4 5 6 7 8 9 a b c d e f 20: -- -- -- -- -- -- -- 27 -- -- -- -- -- -- -- -- - 30: 30 31 -- -- 34 35 36 -- -- -- -- -- -- -- -- -- - 50: 50 -- 52 -- -- -- -- -- -- -- -- -- -- -- -- -- + 30: 30 31 -- -- 34 35 UU UU -- -- -- -- -- -- -- -- + 50: UU -- UU -- -- -- -- -- -- -- -- -- -- -- -- -- + +The ``UU`` entries are the addresses the driver claims at probe: the SPD +EEPROMs at 0x50 and 0x52, and the page-select pair at 0x36 and 0x37 that +``i2c_register_spd_write_disable()`` reserves so that nothing can write them. +See "Instantiating clients" below. Not advertising Quick is deliberate as well as accurate. A write to the 0x30-0x37 range is how an EE1004 SPD is write-protected, permanently on many