With the release of 2.4.25-pre7, stable 2.4 kernel [forum] maintainer Marcelo Tosatti [interview] noted that the 2.4 kernel will not go into maintenance mode until after 2.4.27 is released. The original plan when 2.4.23 was released [story] had been to only accept fixes for critical issues and security holes following the then future release of 2.4.24 [story]. However as it turned out 2.4.24 had to be quickly released to address a local vulnerability in the mremap() system call [story], thereby pushing back the freeze schedule.

Regarding 2.4.25-pre7, Marcelo says, "It contains a bunch of architecture-specific updates (ia64, ppc, mips), JFS and XFS updates, [a] bugfix for big (>128GB) FAT filesystem corruption, amongst others." Looking forward, he explains:

"There are a few bad problems I'm aware of which still need to be fixed (aic7xxx needs to be updated, modular IDE has some problems, etc). Those should get fixed during 2.4.26-pre. Please help testing! :)"

From: Marcelo Tosatti [email blocked]
To:  linux-kernel
Subject: Linux 2.4.25-pre7
Date: Fri, 23 Jan 2004 16:58:24 -0200 (BRST)


Hi,

Here goes -pre number 7 of 2.4.25 series.

It contains a bunch of architecture-specific updates (ia64, ppc, mips),
JFS and XFS updates, bugfix for big (>128GB) FAT filesystem corruption,
amongst others.

About 2.4 freeze:

The planned freeze during 2.4.26 can happen only for 2.4.27.

There are a few bad problems I'm aware of which still need to be fixed
(aic7xxx needs to be updated, modular IDE has some problems, etc). Those
should get fixed during 2.4.26-pre.

Please help testing! :)


Summary of changes from v2.4.25-pre6 to v2.4.25-pre7
============================================

<alex.williamson:hp.com>:
  o ia64: sba_iommu update
  o ia64: sba_iommu: use memparse, long double

<bjorn.helgaas:hp.com>:
  o ia64: work around a menuconfig bug
  o ia64: Fix system type selection to workaround menuconfig bug (select "HP", get "HP-simulator").
  o ia64: Fix broken merge (remove mmu_gathers[] defn)
  o ia64: Skip zero-length resources in PCI root bridge _CRS
  o ia64: sba_iommu: print note about reserving IOVA space for agpgart
  o ia64: Export acpi_hp_csr_space() for modular agpgart
  o ia64: Add acpi_register_irq() interface

<grundler:parisc-linux.org>:
  o obmouse driver for HP OB600 C/CT laptop

<jet:gyve.org>:
  o Fix hfs oops

<marcelo:logos.cnet>:
  o Changed EXTRAVERSION to -pre7

Adrian Bunk:
  o simplify PARPORT_PC_PCMCIA dependencies

Arun Sharma:
  o ia64: ia32 sigaltstack() fix

Atul Mukker:
  o megaraid2 update

Bjorn Helgaas:
  o ia64 HP iommu: add "sx1000" detection (no functional change)

Dave Kleikamp:
  o JFS: Avoid segfault when dirty inodes are written on readonly mount
  o JFS: Creating large xattr lists may cause BUG

Dean Roehrich:
  o [XFS] In xfs_bulkstat, we need to do the readahead loop always

Eric Sandeen:
  o [XFS] Fix for large allocation groups, so that extent sizes will not overflow pagebuf lengths.

Hirofumi Ogawa:
  o FAT: Support large partition (> 128GB)

Jack Steiner:
  o ia64: fix ia64_ctx.lock deadlock

Keith Owens:
  o ia64: fix deadlock in ia64_mca_cmc_int_caller()
  o ia64: Avoid double clear of CMC/CPE records

Manfred Spraul:
  o ldt optimization

Martin Hicks:
  o ia64: Move mmu_gathers[] to local_cpu_data on ia64 (only ia64-specific bits)

Matthew Wilcox:
  o ia64: Add generic RAID xor routines with prefetch

Petr Vandrovec:
  o Deep stack usage in ncpfs

Ralf Bächle:
  o MIPS updates
  o MIPS/DECstation video drivers update
  o Turbochannel driver updates

Rik van Riel:
  o some more fixes for fs/inode.c inode reclaiming changes

Seth Rohit:
  o ia64: hugetlb support (ia64-specific parts)

Stéphane Eranian:
  o ia64: Fix PFM_WRITE_PMCS failure in system-wide mode when PMC12 is zero

Tom Rini:
  o PPC32: Fix finding the MAC address on Motorola MBX860
  o PPC32: Fix the todc definitions for mc146818

Tony Luck:
  o ia64: enable recovery from TLB errors

• Archive of above thread
• KernelTrap interview with Marcelo Tosatti