Marcelo Tosatti [interview] released 2.4.25-rc1 noting, "it contains mostly networking updates, XFS update, amongst others." He goes on to highlight a specific issue:
"This release contains a fix for excessive inode memory pressure with highmem boxes. Help is specially wanted with testing this on heavy-load
highmem machines."
Read on for the full changelog.
From: Marcelo Tosatti [email blocked] To: linux-kernel Subject: Linux 2.4.25-rc1 Date: Thu, 5 Feb 2004 10:44:31 -0200 (BRST) Here goes the first release candidate. It contains mostly networking updates, XFS update, amongst others. This release contains a fix for excessive inode memory pressure with highmem boxes. Help is specially wanted with testing this on heavy-load highmem machines. Summary of changes from v2.4.25-pre8 to v2.4.25-rc1 ============================================ <chrisw:osdl.org>: o Verify interpreter arch <davej:redhat.com>: o fix agpgart warning <davem:nuts.davemloft.net>: o [IRDA]: Mark driver init/exit funcs static where possible o [SPARC64]: Fix TUNSETIFF ioctl compat, it takes an ifreq ptr not an int o [TG3]: Bump version and reldate o [SPARC64}: Fix ultra-III and later support of new-style SILO booting <grundler:parisc-linux.org>: o [TG3]: Fix DMA test failures o [TG3]: Only fetch NVRAM_CMD reg if TG3_FLAG_NVRAM <jlcooke:certainkey.com>: o [CRYPTO]: Help gcc optimize sha256/sha512 better <jmorris:redhat.com>: o [CRYPTO]: Make padding[] array static in sha{256,512}_final() o Zero last byte of mount option page <marcelo:logos.cnet>: o Changed EXTRAVERSION to -rc1 <rml:ximian.com>: o Fix sys_readahead(): Count free pages on maximum ra size calculation Adrian Bunk: o fix a compile warning in amd76x_pm.c o fix a compile warning in tipar.c o fix a compile warning in InterMezzo file.c Ben Collins: o [SPARC64]: Add comment for HdrS ver 0x201 o [SPARC64]: Add header for section boundary references o [SPARC64]: Changes to accomodate booting from non-phys_base memory Chas Williams: o [ATM]: [idt77252] fix dma_addr_t type error with CONFIG_HIGHMEM64G=y (by "Randy.Dunlap" [email blocked]) o [ATM]: [clip] check return code from kmem_cache_create (by "Randy.Dunlap" [email blocked]) Christoph Hellwig: o [XFS] Small ktrace fixes o [XFS] Don't fail pagebuf allocations David Brownell: o usb/gadget/file_storage.c doesn't compile with gcc 2.95 David S. Miller: o [DECNET]: Fix filling in of header length field o [CREDITS]: Update Bjorn Ekwall's address David Stevens: o [IPV4]: Add per-device sysctl to force IGMP version o [IPV4]: Fix IGMP device reference counting Harald Welte: o [NETFILTER]: Fix ipt_conntrack/ipt_state module refcounting o [NETFILTER]: Really fix ipt_state/ipt_conntrack refcounting Herbert Xu: o invalid kfree in ntfs_printcb Luca Tettamanti: o Fix ac97_plugin_ad1980.c compilation warning o Fix aha1542.c compilation warning o Fix cpqfcTSi2c.c compilation warning o IEEE1394(r1123): Fix compile warning o Fix amd7930_fn.h compilation warning o Fix drivers/net/wan/8253x/crc32.c compilation warning o Fix vac-serial.c compilation warning Mirko Lindner: o sk98lin: Reset Xmac when stopping the port Nathan Scott: o [XFS] Remove xfsidbg debugger interfaces, not useful without kdb o [XFS] Fix a warning from some gcc variants after recent flags botch o [XFS] Add the security extended attributes namespace o [XFS] Remove no-longer-needed debug symbol exports Patrick McHardy: o [NET_SCHED]: Add HFSC packet scheduler Russell Cattelan: o [XFS] Christoph has signed over copyrights o [XFS] Move bits around to better manage common code. No functional change o [XFS] Remove non 2.4 ifdefs from the linux-2.4 dir Rusty Russell: o [NETFILTER]: ipt_limit fix for HZ=1000 Scott Feldman: o e100 sync with 2.6