Summary of changes from v2.6.1-rc2 to v2.6.1-rc3 ============================================ [IPV6]: Kill obsolete functions (ip6_frag_xmit() and ip6_build_xmit()). [NET]: Do type checking in {udp,inet6,raw6,inet}_sk(). [AGPGART] printk level changes for amd64 [SPARC]: Add missing MODULE_LICENSE tags to various Sparc driver. [SUNZILOG]: Register the correct number of ports, ignore keyb/mouse lines. [SPARC64]: Fix sun_uflash MTD driver build. [AF_PACKET]: Convert to seq_file. [DECNET]: Better way to prevent decnet module unload. [NET]: Fix multiple eth0 mixed PCI/ISA init. This patch for 2.6 fixes the problem found by Zoltan Farkas with mixed PCI/ISA and a non-modular config. The problem is the old_netdev ISA probing isn't skipping eth0 which already got assigned by the PCI initialization. [SPARC64]: Fix CONFIG_DRM_FFB=y build. [SPARC]: Add placeholder asm-{sparc,sparc64}/setup.h so MTD builds. [SPARC64]: Use drivers/block/Kconfig. [SPARC64]: Export sys_close for solaris emulation. [NET]: In dev_kfree_skb_any() use dev_kfree_skb_irq() if irqs_disabled(). With help from Jeff Garzik and others. [IRDA]: Fix locking in the ircomm-shutdown path. [SPARC32]: Fix BUG on swapout on srmmu systems. This fixes BUG-on-swapout for srmmu-based systems. The problem is caused by kmap_atomic_to_page being fed an aliased (pagetable) address and returning bogons. This also adjusts the pkmap and fixmap base addresses so they cannot overlap. [NET]: Un-deprecate skb_linearize(), we can re-deprecate in 2.7.x [PATCH] Fix interrupt routing problem on x86-64 The MSI port to x86-64 added an interrupt routing bug that makes the kernel not boot anymore on some machines. Fix that. [PATCH] cmd640.c: fix PCI type1 access Revert wrong changes introduced in 2.4.21. From: Stefan Talpalaru , Ingo Kilian . [PATCH] ide-tape.c: stop abusing rq->flags Use rq->cmd[0] instead of rq->flags for storing special request flags. Per Jens' suggestion. Tested by Stef van der Made . [PATCH] remove dead and broken DISK_RECOVERY_TIME support It was also removed in 2.4.23. [PATCH] fix oopses on rmmod in some OSS drivers From Jakub Bogusz Fix for oops on rmmod caused by *_remove() function marked as __devinit (and thus discarded after module initialization - if CONFIG_MODULES=y and CONFIG_HOTPLUG is not set). This patch changes __devinit to __devexit and adds __devexit_p() where pointer to such function is used. The only exception is au1000, where au1000_remove() is called from cleanup_au1000() function - __devinit is jest removed there. [PATCH] generate an error if writing of kernel config failed generate an error if writing of kernel config failed [PATCH] fix gconf segfault problem fix gconf segfault problem (by Romain Lievin ) [PATCH] gconf compile warning fixes gconf compile warning fixes (by Buddy Lucas ) [PATCH] gconf startup fixes Let gconf find it's glade file, even it's started with an absolute path and check srcdir so it also works if it's compiled outside of the tree. [PATCH] qconf fix Sometimes a menuconfig item is shown twice, so hide the other item. [PATCH] Fix PnP BIOS call The recent escd fix I have made corrects a thinko in the PnPBIOS code and it turns out that faults from calling /proc/pnp/bus/escd were probably not caused by BIOS bugs. [NETFILTER]: Add SELINUX priority values for ipv4/ipv6, approved by Harald Welte. [NET]: Make sure that class selected by priority is a leaf in HTB scheduler. Fix my PGP fingerprint in the CREDITS file. [AGPGART] Add support for Radeon IGP345M to ATI GART driver. Also consolidate a bunch of convoluted if's into a single "is this an r200 or r300" function. [PATCH] MSI build fixes Add missing NR_VECTORS definition to visws and pc9800. Also, make MSI support dependent on CONFIG_X86_IO_APIC, as it won't build without IOAPIC support. [PATCH] fix sysfs oops From: Greg KH This fixes an oops when a kobject is unregistered before it's child is. The usb-serial devices show this bug very easily (yank out a device while its port is opened...) Patch was originally written by Mike Gorse [PATCH] JFS fix for NFS on little-endian systems From: Dave Kleikamp After Jose debugged the problem down to the routine jfs_get_parent, we were able to find the problem. I believe it only affects users of NFS-exported JFS file systems on big-endian hardware. The problem was a missing le32_to_cpu macro. The patch also fixes a return code to be more consistent other implementations of get_parent. [PATCH] dvb: firmware fixes From: Michael Hunold - move around sp887x firmware file entry to be close to sp887x selection - fix a bunch of filenames to point to /usr/lib/hotplug/firmware/ rather then /etc/dvb - fix the av7110 firmware config entry for "make allyesconfig" [PATCH] ia32 sched_clock() deadlock fix From: Ingo Molnar , Voyager is getting odd deadlocks due to the taking of xtime_lock() in sched_clock()->get_jiffies_64(). I had this patch queued up to fix a different deadlock, which occurs when we relax the requirement that TSC's be synchronised across CPUs. But it will fix James' deadlock too. [PATCH] don't clear i_sb From: Christoph Hellwig JFS currently clears i_sb in some error pathes which can make the core kernel OOPS because it may never be NULL. Noticed because some IBM people try to "fix" the core kernel for it now.. [SPARC32]: Use drivers/block/Kconfig [COMPAT]: Handle SO_TIMESTAMP cmsgs. [IPV6]: Autoconfig link-local address on ip6-ip6 tunnel device. Fix ttpci bogus use of floating point by casting the constant expression properly. Fix silly mremap test. Get off the drugs, Linus. Don't relocate non-allocated regions in modules. This fixes loading of modules compiled with debugging on some platforms. From Rusty. [PATCH] /proc/ppc64 and /proc/iSeries fixes from Linas Vepstas From: Anton Blanchard Linas Vepstas has audited the ppc64 proc code and found a number of issues. [PATCH] ppc64: Add missing section definition From: Anton Blanchard For a laugh we moved the BSS and our world exploded. Turns out we zero from __bss_start to _end. Add __bss_stop to our section definitions so we can use it instead. Linux 2.6.1-rc3