Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Feedback-ID: ic081425d:Fastmail
From:   John Thomson <git@johnthomson.fastmail.com.au>
To:     Sergio Paracuellos <sergio.paracuellos@gmail.com>,
        Thomas Bogendoerfer <tsbogend@alpha.franken.de>,
        John Crispin <john@phrozen.org>,
        =?UTF-8?q?Ar=C4=B1n=C3=A7=20=C3=9CNAL?= <arinc.unal@arinc9.com>
Cc:     linux-mips@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH 0/3] mips: ralink: mt7621: fix kzalloc too early
Date:   Mon, 14 Nov 2022 11:56:55 +1000
Message-Id: <20221114015658.2873120-1-git@johnthomson.fastmail.com.au>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

ralink mt7621 attempts to use kzalloc before normal memory management is
available.
Before v6.1-rc1, mt7621.c soc_dev_init silently failed to kzalloc and
returned immediately afterwards, continuing to boot without the
SOC_BUS device registered.
After, (with CONFIG_SLUB) kernel crashes before it outputs any console
messages

This was bisected to an mm/slub change (detailed in patch 3)

This change exposed oops in the quirks tables in the mt7621 phy-pci &
pci-controller drivers. The phy-pci quirk fix got into -rc5,
but the pci-controller quirk fix has not been picked up yet.
https://lore.kernel.org/lkml/20221114012036.2870067-1-git@johnthomson.fastmail.com.au/T/#u
This is mentioned in patch 3 as:
not yet applied ("PCI: mt7621: add sentinel to quirks table")

Changes from RFC
https://lore.kernel.org/lkml/20221103050538.1930758-1-git@johnthomson.fastmail.com.au/T/#u
to v1:

1/3:
cut the CPHYSADDR to match the line moved to the header file
2/3:
No useless const unsigned {int,bool} function declarations (thanks
kernel test robot)
3/3:
static struct *soc_info_ptr
Mention fixed phy-pci & (unapplied) pci-controller quirks sentinels
Add fixes tag

Cheers,
John