[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <alpine.DEB.2.21.2510211901530.8377@angie.orcam.me.uk>
Date: Tue, 21 Oct 2025 20:38:15 +0100 (BST)
From: "Maciej W. Rozycki" <macro@...am.me.uk>
To: Bjorn Helgaas <helgaas@...nel.org>,
Thomas Bogendoerfer <tsbogend@...ha.franken.de>
cc: Ilpo Järvinen <ilpo.jarvinen@...ux.intel.com>,
Bjorn Helgaas <bhelgaas@...gle.com>, linux-mips@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: [PATCH 0/2] MIPS: Malta: Fix PCI southbridge legacy resource
clashes
Hi,
This mini patch series sorts out issues with southbridge legacy resource
management on the MIPS Malta platform. Two changes turned out required,
because merely removing the clash would regress the PS/2 interfaces, fixed
by accident with the PCIBIOS_MIN_IO fix.
This does prove nobody has used these interfaces since forever, or most
likely since the move to the new serio driver. Things most likely worked
fine with 2.4 and I still have such old Malta kernel builds lying around
(though sadly no hardware to try with), although I do know I've never used
the PS/2 stuff with this platform, e.g. quoting an arbitrarily picked
2.4.19-rc1 bootstrap log:
CPU revision is: 00018101
Primary instruction cache 16kb, linesize 32 bytes(4 ways)
Primary data cache 16kb, linesize 32 bytes (4 ways)
Linux version 2.4.19-rc1 (macro@...ro.ds2.pg.gda.pl) (gcc version 2.95.4 20010319 (prerelease)) #1 Fri Aug 23 02:55:02 CEST 2002
[...]
parport0: PC-style at 0x378 [PCSPP,EPP]
initialize_kbd: Keyboard reset failed, no ACK
Detected PS/2 Mouse Port.
pty: 256 Unix98 ptys configured
keyboard: Timeout - AT keyboard not present?(ed)
keyboard: Timeout - AT keyboard not present?(f4)
Serial driver version 5.05c (2001-07-08) with MANY_PORTS SHARE_IRQ DETECT_IRQ SERIAL_PCI enabled
[...]
However to prevent the PS/2 interfaces from getting fixed and then broken
again with backports in a random fashion I have marked both changes for
backporting as appropriate.
Bjorn, may I request that these changes be placed, with Thomas's ack of
course (hopefully a formality), ahead of Ilpo's commit 16fbaba2b78f
("MIPS: Malta: Use pcibios_align_resource() to block io range") (or
whatever the latest version is, as said commit seems to be missing tags
updates you mentioned), and then merged via your tree? That will prevent
things from breaking just to be fixed again shortly, and overall getting
out of sync.
See individual commit descriptions for details.
Maciej
Powered by blists - more mailing lists