ChangeSet@1.1292, 2003-06-07 23:57:36-07:00, sam@ravnborg.org [PATCH] all archs: Replace O_TARGET with lib-y lib-y is the new way to define what objects belongs to a library. The implementation was not made backwards compatible and therefore an update to all architectures are needed. This is a simple replacement of obj-* to lib-* and deletion of L_TARGET. The new mechanish where lib.a can be mixed with built-in.o is not utilised. ChangeSet@1.1291, 2003-06-07 17:38:01-07:00, B.Zolnierkiewicz@elka.pw.edu.pl [PATCH] switch ide to taskfile IO - rewrite taskfile PIO handlers (they now comply with ide state machine and use bio walking) - switch ide-disk.c to use *only* taskfile IO - swicth pdc4030.c to use *only* taskfile IO (untested) - remove old cruft (>600 lines) ChangeSet@1.1290, 2003-06-07 16:52:55-07:00, akpm@digeo.com [PATCH] eventpoll: fix possible use-after-free From: Davide Libenzi After the ep_remove() the "epi" is given back to the cache, so "epi->ep" might become invalid. It was not cought by my tests because the element wasn't immediately reused (and because I was using a single epoll fd, so the "ep" item remained the same). ChangeSet@1.1289, 2003-06-07 16:52:49-07:00, akpm@digeo.com [PATCH] add bootmem failure warning From: Dave Hansen __alloc_bootmem_core() has a couple of BUG_ON()'s. Since the handlers aren't set up this early, if you hit it, you just get along stream of "Unknown Interrupt" messages. It would be very nice to have a little bit more information when something has decided to BUG() out this early. ChangeSet@1.1288, 2003-06-07 16:52:41-07:00, akpm@digeo.com [PATCH] rocket.c: devfs fix Christoph says this undef is not correct. ChangeSet@1.1287, 2003-06-07 16:52:33-07:00, akpm@digeo.com [PATCH] provide bus to node mapping for Summit From: Matt Dobson, via Martin Bligh This parses the machine's BIOS tables to populate the mp_bus_id_to_node[bus] array. Only affects Summit machines, safe, boring. Has been in -mjb tree for ages, and works fine. ChangeSet@1.1286, 2003-06-07 16:52:26-07:00, akpm@digeo.com [PATCH] cleanup conditionals in summit subarch From: "Martin J. Bligh" The "magic" switching in subarch was ugly when I put it there, and nobody liked it then (including me). It hasn't got any prettier since. Andi's generic arch stuff is a cleaner solution for now, so we can remove the old hacky stuff, and significantly simplify the code. All this does is replace "(x86_summit ? A : B)" with "A" everywhere. ChangeSet@1.1285, 2003-06-07 16:52:18-07:00, akpm@digeo.com [PATCH] fix apic handling for NUMA-Q From: "Martin J. Bligh" All this fancy stuff in cpu_mask_to_apicid doesn't work for NUMA-Q, because it's based on logical apicids, and we use physical. Drop back to just always returning 0xF instead, which is the broadcast physical ID, and has been working fine since the dawn of time. ChangeSet@1.1284, 2003-06-07 16:52:12-07:00, akpm@digeo.com [PATCH] JFFS_PROC_FS must depend on JFFS_FS From: Adrian Bunk Compilation fails if JFFS_PROC_FS and !PROC. The following dependency in the Kconfig file is needed. ChangeSet@1.1283, 2003-06-07 16:52:03-07:00, akpm@digeo.com [PATCH] remove_proc_entry() fix From Bartlomiej Zolnierkiewicz With !CONFIG_PROC_FS, ieee1394_core fails to compile because the argument to this inline is still evaluated. But it doesn't exist. A general fix is to not evaluate the arg at all. ChangeSet@1.1282, 2003-06-07 16:51:58-07:00, akpm@digeo.com [PATCH] x86_64: use common ioctl code From: Pavel Machek Convert x86_64 to use the common ioctl code. ChangeSet@1.1281, 2003-06-07 16:51:50-07:00, akpm@digeo.com [PATCH] ioctl32 cleanup: sparc64 From: Pavel Machek Make sparc64 use generic ioctl32 code. ChangeSet@1.1280, 2003-06-07 16:51:44-07:00, akpm@digeo.com [PATCH] common 32-bit ioctl code From: Pavel Machek Various 64-bit architectures are duplicating a ton of 32-bit compat code. Pavel's patch creates a generic 32-bit ioctl file in fs/compat_ioctl.c which architectures will #include from within their arch/ layer. Has been reviewed by everyone and tested on sparc64, x86_64, ppc64 and ia32. ChangeSet@1.1279, 2003-06-07 16:51:37-07:00, akpm@digeo.com [PATCH] Fix the build with !CONFIG_PROC_FS From: Adrian Bunk ChangeSet@1.1278, 2003-06-07 16:51:27-07:00, akpm@digeo.com [PATCH] ppc64: fixup for family/sk_family rename Fix the ppc64 build for the great socket member renaming. ChangeSet@1.1277, 2003-06-07 16:51:20-07:00, akpm@digeo.com [PATCH] Fix build for CONFIG_KALLSYMS=n From: "David S. Miller" add_kallsyms() doesn't exist if !CONFIG_KALLSYMS. ChangeSet@1.1276, 2003-06-07 14:54:35-07:00, sam@ravnborg.org [PATCH] be more flexible about creating library archives New makefile variable introduced: lib-y The lib-y syntax allows you to do the usual tricks such as: lib-$(CONFIG_SMP) += percpu_counter.o A built-in.o is always present in a directory that list .o files in either obj-* or lib-*. In contrast, lib.a is made only when lib-y is defined. I also updated lib/Makefile, so that crc32.o is now always built-in if selected. ChangeSet@1.1275, 2003-06-07 18:18:34-03:00, azarah@gentoo.org o ethertap: fix struct sock cleanup leftover ChangeSet@1.1274, 2003-06-07 13:14:28-07:00, ak@suse.de [PATCH] x86-64 merge This brings the x86-64 port uptodate for 2.5.70. Just various bugfixes and a few merges with other people. Only changes architecture specific files. - Fix compiling with CONFIG_IA32_EMULATION on - Readd lost apic power management patch from Pavel (fixes oprofile too) - Increase max IOAPICs to 16 - Fix compiling with CONFIG_IA32_EMULATION off - Compile fix for suspend (Pavel Machek) - Support boxes with APIC disabled - Remove code to forcibly enable APIC - Small fix for APIC timer calibration. - Fix deadlock in SMP reboot - Some warning fixes - Save edid info at boot (Bryan O'Sullivan) - Add better locking to oops printing and support it for page faults. - Don't printk handled signals. - Update defconfig - Add copy_in_user ChangeSet@1.1273, 2003-06-07 13:14:18-07:00, ak@suse.de [PATCH] Make spinlock debugging compile on x86-64 cpu_relax is on i386 and x86-64 in processor.h, not system.h This makes CONFIG_DEBUG_SPINLOCK compile for x86-64 ChangeSet@1.1272, 2003-06-07 13:14:11-07:00, paulus@samba.org [PATCH] Move BUG/BUG_ON/WARN_ON to asm headers This patch moves the definitions of BUG, BUG_ON and WARN_ON from to (which includes), and supplies a new implementation for PPC which uses a conditional trap instruction for BUG_ON and WARN_ON, thus avoiding a conditional branch. This patch trims over 50kB from the size of the kernel that I use on powermacs. With this patch, on PPC we have a __bug_table section in the vmlinux binary, and also in modules if they use BUG, BUG_ON or WARN_ON. The __bug_table section has one entry for each BUG/BUG_ON/WARN_ON, giving the address of the trap instruction and the corresponding line number, filename and function name. This information is used in the exception handler for the exception that the trap instruction produces. The arch-specific module code handles the __bug_table section so that BUG/BUG_ON/WARN_ON work correctly in modules. Several architecture maintainers have acked this change. It should be completely benign for all of the other architectures (though they may decide to do something similar if they have a conditional trap instruction available). ChangeSet@1.1271, 2003-06-07 13:06:36-07:00, paulus@samba.org [PATCH] Fix check warnings in PPP code This patch removes the warnings that the check program came up with in the PPP code: ppp_async.c, ppp_deflate.c and ppp_generic.c. This involved adding __user and converting K&R-style function definitions to ANSI-style. I also took the time to add some extra comments to ppp_deflate.c explaining in more detail what each function does and what its arguments are. ChangeSet@1.1270, 2003-06-07 13:06:29-07:00, paulus@samba.org [PATCH] fix check warnings in drivers/macintosh This patch removes the warnings that the `check' program came up with in drivers/macintosh. This involves adding __user in various places and fixing some non-ANSI function definitions for functions that take no arguments. ChangeSet@1.1269, 2003-06-07 12:55:41-07:00, elenstev@mesatop.com [PATCH] Another final K&R to ANSI C cleanup of zlib Here is another final K&R to ANSI C cleanup patch for zlib. ChangeSet@1.1268, 2003-06-07 12:05:54-07:00, l.s.r@web.de [PATCH] Some more stuff missed during the struct sock cleanup The members of struct sock got a prefix of 'sk_' recently. This patch updates smbfs to match the new world order. ChangeSet@1.1265, 2003-06-07 10:41:04-07:00, ak@suse.de [PATCH] Work around gcc 3.3 bug on amd64 in binfmt_elf.c This patch works around a gcc 3.3 bug on AMD64. On AMD64 the get_current() function expands to a switch on sizeof which has to be optimized away by the compiler, finally yielding an inline assembler statement. In some cases it seems to get that wrong and forgets to reference and use the argument. I have only seen it happen in binfmt_elf so far. Work around by just computing "current" once. ChangeSet@1.1264, 2003-06-07 10:40:57-07:00, l.s.r@web.de [PATCH] hugetlbfs: fix error reporting in case of invalid mount hugetlbfs was unnecessarily verbose, and didn't even print out the right thing if given invalid mount options. Just return EINVAL. ChangeSet@1.1243.51.1, 2003-06-07 14:22:10-03:00, acme@conectiva.com.br o list.h: improve hlist This changeset: 1. Implements hlist_add_after 2. uses prefetch in hlist_for_each, using a trick that ends up being equivalent to having the prefetch instruction in the first block of the hlist_for_each for block, the compiler optimizes the second "test" away, as its result is constant 3. implements hlist_for_each_entry and hlist_for_each_entry safe, using a struct hlist_node as iterator to avoid the extra branches a similar implementation to list_for_each_entry would have if used a typed iterator, but while avoiding having to have the explicit hlist_entry as in hlist_for_each. 4. Converts the hlist_for_each users that had explicit prefetches, i.e. removed the explicit prefetch 5. fix a harmless list_entry use in a hlist_for_each in inode.c ChangeSet@1.1243.42.6, 2003-06-07 14:13:44+01:00, davej@codemonkey.org.uk [CPUFREQ] missing export powernow-k7 needs dmi_broken ChangeSet@1.1243.1.84, 2003-06-07 03:02:45-07:00, davem@nuts.ninka.net [SCTP]: Kill unused local variable in init_sctp_mibs. ChangeSet@1.1243.1.83, 2003-06-07 03:00:42-07:00, shemminger@osdl.org [DECNET]: Fix build warnings. ChangeSet@1.1243.1.82, 2003-06-07 02:55:07-07:00, sri@us.ibm.com [IPV6]: Allow ipv6 fragmentation via ip6_xmit() when ipfragok is set. ChangeSet@1.1243.1.81, 2003-06-07 02:23:53-07:00, mk@linux-ipv6.org [IPV6]: Fix esp6 extension headers handling. ChangeSet@1.1243.1.80, 2003-06-07 02:20:40-07:00, rddunlap@osdl.org [IPV6]: Fix spelling/typos. ChangeSet@1.1261, 2003-06-07 19:20:29+10:00, paulus@samba.org PPC32: Convert some K&R-style functions to ANSI-style. From Steven Cole. ChangeSet@1.1260, 2003-06-07 19:18:09+10:00, paulus@samba.org PPC32: Fix various minor problems pointed out by Linus' check program. ChangeSet@1.1243.1.79, 2003-06-07 01:55:36-07:00, herbert@gondor.apana.org.au [XFRM_USER]: Allow del policy by id and get policy by selector. ChangeSet@1.1259, 2003-06-07 18:31:17+10:00, paulus@samba.org PPC32: Fix irq_desc initialization. ChangeSet@1.1243.1.78, 2003-06-07 01:30:45-07:00, hch@lst.de [NET]: Fix accidental revert of init_etherdev killing in PPC net drivers. ChangeSet@1.1258, 2003-06-07 18:24:33+10:00, paulus@samba.org PPC32: Start adding __user to mark pointers from userspace. This reduces the number of warnings from Linus' `check' program for stuff in arch/ppc. ChangeSet@1.1243.1.77, 2003-06-06 21:15:24-07:00, anton@samba.org [PATCH] update ppc64 MAINTAINERS entry An update to the ppc64 MAINTAINERS entry. ChangeSet@1.1243.1.76, 2003-06-06 17:30:11-07:00, rusty@rustcorp.com.au [PATCH] Move cpu notifiers et al to cpu.h Trivial patch: when these were introduced cpu.h didn't exist. ChangeSet@1.1243.1.75, 2003-06-06 17:26:23-07:00, rusty@rustcorp.com.au [PATCH] kallsyms in proc This adds a /proc/kallsyms if you have CONFIG_KALLSYMS in your kernel. The output is nm-like, with symbols in caps (global) if exported using EXPORT_SYMBOL, rather than the normal static vs. non-static differentiation. This is useful for things like performance monitoring tools (profiling etc) that want to match addresses to names in user space. ChangeSet@1.1243.1.74, 2003-06-06 17:22:34-07:00, torvalds@home.transmeta.com Merge bk://kernel.bkbits.net/davem/net-2.5 into home.transmeta.com:/home/torvalds/v2.5/linux ChangeSet@1.1243.50.22, 2003-06-06 10:33:36-07:00, torvalds@home.transmeta.com Remove half-deleted zero-sized sound file. ChangeSet@1.1243.50.21, 2003-06-06 08:55:47-07:00, akpm@digeo.com [PATCH] loop: remove the balance_dirty_pages() call The loop thread is getting permanently stuck in balance_dirty_pages() (nr_writeback is exceeded) because the loop thread itself is responsible for completing writeback on behalf of higher layers. So we need to take that out: don't throttle the loop thread. Throttle the tasks which are generating all the dirty data instead. ChangeSet@1.1243.50.20, 2003-06-06 08:55:40-07:00, akpm@digeo.com [PATCH] /proc/sys/vm/min_free_kbytes From: Matthew Dobson This resurrects the old /proc/sys/vm/free_pages functionality: the ability to tell page reclaim how much free memory to maintain. This may be needed for specialised networking applications, and it provides an interesting way to stress the kernel: set it very low so atomic allocations can easily fail. Also, a 16G ppc64 box currently cruises along at 1M free memory, which is surely too little to supporthigh-speed networking. We have not changed that setting here, but it is now possible to do so. The patch also reduces the amount of free memory which the VM will maintain in ZONE_HIGHMEM, as it is almost always wasted memory. ChangeSet@1.1243.50.19, 2003-06-06 08:55:34-07:00, akpm@digeo.com [PATCH] DAC960 fix for fibre channel transfer rate From: Dave Olien The change makes the transfer rate numbers come out right for the fibre channel version of this controller. For 1G FC, the NegotiatedSynchronousMegaTransfers is 1000, and the NegotiatedDataWidthBIts is 1. The old code assumed NegotiatedDataWidthBits was always either 8 or 16. The new code is simpler, and does the calculation correctly for all cases. ChangeSet@1.1243.50.18, 2003-06-06 08:55:28-07:00, akpm@digeo.com [PATCH] Don't let processes be scheduled on CPU-less nodes (3/3) From: Matthew Dobson This patch implements a generic version of the nr_cpus_node(node) macro implemented for ppc64 by the previous patch. The generic version simply computes an hweight of the bitmask returned by node_to_cpumask(node) topology macro. This patch also adds a generic_hweight64() function and an hweight_long() function which are used as helpers for the generic nr_cpus_node() macro. This patch also adds a for_each_node_with_cpus() macro, which is used in sched_best_cpu() in kernel/sched.c to fix the original problem of scheduling processes on CPU-less nodes. This macro should also be used in the future to avoid similar problems. Test compiled and booted by Andrew Theurer (habanero@us.ibm.com) on both x440 and ppc64. ChangeSet@1.1243.50.17, 2003-06-06 08:55:18-07:00, akpm@digeo.com [PATCH] Don't let processes be scheduled on CPU-less nodes (2/3) From: Matthew Dobson This patch defines a topology macro for ppc64, nr_cpus_node(node) which returns the number of CPUs on 'node'. This patch also adds code to compute and store these values in an array for quick lookup. Test compiled and booted by Andrew Theurer (habanero@us.ibm.com) on both x440 and ppc64. ChangeSet@1.1243.50.16, 2003-06-06 08:55:12-07:00, akpm@digeo.com [PATCH] Don't let processes be scheduled on CPU-less nodes (1/3) From: Matthew Dobson sched_best_cpu schedules processes on nodes based on node_nr_running. For CPU-less nodes, this is always 0, and thus sched_best_cpu tends to migrate tasks to these nodes, which eventually get remigrated elsewhere. This patch adds include/linux/topology.h, and modifies all includes of asm/topology.h to linux/topology.h. A subsequent patch in this series adds helper functions to linux/topology.h to ensure processes are only migrated to nodes with CPUs. Test compiled and booted by Andrew Theurer (habanero@us.ibm.com) on both x440 and ppc64. ChangeSet@1.1243.50.15, 2003-06-06 08:55:04-07:00, akpm@digeo.com [PATCH] remove triggerable BUG() from de_thread From: Ingo Molnar Apparently this BUG is triggerable due to correct and expected events. ChangeSet@1.1243.50.14, 2003-06-06 08:54:58-07:00, akpm@digeo.com [PATCH] remove get_current_user() As "Dmitry A. Fedorov" points out, get_current_user() has a local variable __user which conflicts with the sparse tagging. But get_current_user() has no callers. ChangeSet@1.1243.50.13, 2003-06-06 08:54:53-07:00, akpm@digeo.com [PATCH] cs423x fixes From: Adam Belay - cs4236 doesn't check if the memory for the resource table was successfully allocated. ChangeSet@1.1243.50.12, 2003-06-06 08:54:43-07:00, akpm@digeo.com [PATCH] misc fixes - Add comment about slab ctor behaviour (Ingo Oeser) - mm/slab.c:fprob() shows up in profiles a lot. Rename it to something more meaningful. - fatfs printk warning fix (Randy Dunlap) - give the the time interpolator list and lock file-static scope (hch) ChangeSet@1.1243.50.11, 2003-06-06 08:54:35-07:00, akpm@digeo.com [PATCH] Fix tty devfs mess From: Christoph Hellwig Currently the tty code abuses tty_driver.name as the prefix for the devfs names of the ttys. This is a very bad idea because it means the tty name changes depending on whether devfs is enabled or not, leading to different names in /proc/tty/ depending on whether we have devfs or not (and not whether it actually is mounted!) and a huge amount of ifdefs. The patch below adds a .devfs_name member instead, similar to the block device changes a few weeks ago. ChangeSet@1.1243.50.10, 2003-06-06 08:54:27-07:00, akpm@digeo.com [PATCH] Console privacy for braille users From: Samuel Thibault Still working on kernel facilities for braille devices, the need for being able to force blanking and unblanking raised: even when a key is pressed, the screen must remain blank, for privacy of the blind user who is typing on the keyboard and reading on its braille terminal. I merely added an ignore_poke variable which is set, and the screen blanked. Then, poke_blanked_console returns immediatly. Upon real unblank (because of an Oops or an explicit tioclinux), ignore_poke is reset to get back to normal operation mode. I had to remove the (unnecessary ?) call to unblank_screen from set_selection to prevent mouse selection unblanking the screen. I also added a way for processes to know whether the screen is blanked (the blind user might hence know whether people can read the screen). ChangeSet@1.1243.50.9, 2003-06-06 08:54:20-07:00, akpm@digeo.com [PATCH] Console blanking fix From: Samuel Thibault Some fixes for console blanking: on some laptops, doing VESA blanking after the bios did an apm blanking because of a screen closure thrashes the recovery (the video board doesn't seem to have synchronisation registers correctly initialized, since the LCD panel progressively turns white, maybe damaging it ?). I hence moved the schedule for vesa powerdown after the apm blank hook call, so that if it succeeds, it won't be called. I also moved the apm unblank & palette restoration after the vesa unblank, to have a more lifo scheme (also required, or the screen remains black). Btw, why del_timer_sync was called twice in timer_do_blank_screen when vesa_off_interval==0 ? ChangeSet@1.1243.50.8, 2003-06-06 08:54:14-07:00, akpm@digeo.com [PATCH] fix wobbly /proc/stat:btime From: john stultz Since jiffies didn't necessarily start incrementing at a second boundary, jiffies/HZ doesn't increment at the same moment as xtime.tv_sec. This causes one second wobbles in the calculation of btime (xtime.tv_sec - jiffies/HZ). This fix increases the precision of the calculation so the usec component of xtime is used as well. Additionally it fixes some of the non-atomic reading of time values. ChangeSet@1.1243.50.7, 2003-06-06 08:54:08-07:00, akpm@digeo.com [PATCH] force_successful_syscall_return() From: David Mosberger , Christoph Hellwig I believe this is the last outstanding piece that prevents ia64 from being fully in sync with Linus' tree (yes, there are some minor ACPI changes outstanding and a toolchain bug that's left to fix, but other than that, I think we're clean). Many architectures (alpha, ia64, ppc, ppc64, sparc, and sparc64 at least) use a syscall convention which provides for a return value and a separate error flag. On those architectures, it can be beneficial if the kernel provides a mechanism to signal that a syscall call has completed successfully, even when the returned value is potentially a (small) negative number. The patch below provides a hook for such a mechanism via a macro called force_successful_syscall_return(). On x86, this would be simply a no-op (because on x86, user-level has to be hacked to handle such cases). On Alpha, it would be something along the lines of: #define force_successful_syscall_return() ptregs->r0 = 0 where "ptregs" is a pointer to the user's ptregs structure of the current task. On ia64, we have been using this for a long time: static inline void force_successful_syscall_return (void) { ia64_task_regs(current)->r8 = 0; } The other architectures (ppc, ppc64, sparc, and sparc64) currently have no mechanism to force a syscall return to be successful. But since the syscall convention already provide for a separate error flag, the arch maintainers could change this if they wanted to. There are only 3 places in the platform-independent portion of the kernel that need this macro: - memory_lseek() in drivers/char/mem.c - fs/fcntl.c for F_GETOWN - lseek for /proc/mem in fs/proc/array.c Ideally, there are a couple of other places that could benefit from this macro: - sys_getpriority() - sys_shmat() - sys_brk() - do_mmap2() - do_mremap() but these are not so critical, because the can be worked around in platform-specific code (e.g., see arch/ia64/kernel/sys_ia64.c). Note that for the above 3 cases, handling them in user level is rather suboptimal: - it would affect all lseek() syscalls, even though only /proc/mem and /dev/mem need the special treatment (at least until there are filesystems that can handle files >= 2^63 bytes) - all fcntl() calls would be affected, even though only F_GETOWN needs the special treatment so I think handling these in the kernel for the platforms that can makes tons of sense. The only limitation of force_successful_syscall_return() is that it doesn't help with system calls performed by the kernel. But the kernel does that so rarely and for such a limited set of syscalls that this is not a real problem. ChangeSet@1.1243.50.6, 2003-06-06 08:53:57-07:00, akpm@digeo.com [PATCH] IRQs: fix up irq_desc initialisation for non-ia32 The addition of more fields to irq_desc_t may have broken compilation of other architectures. Go through and C99ify them (was needed anyway). ChangeSet@1.1243.50.5, 2003-06-06 08:53:50-07:00, akpm@digeo.com [PATCH] IRQs: handle bad return values from handlers Attempt to do something intelligent with IRQ handlers which don't return IRQ_HANDLED. - If they return neither IRQ_HANDLED nor IRQ_NONE, complain. - If they return IRQ_NONE more than 99900 times in 100000 interrupts, complain and disable the IRQ. I did have it at 750-in-1000, but someone had an otherwise-functioning system which triggered it. The 99.9% ratio is designed to address the problem wherein the babbling device shares an IRQ with a good device. We don't want the good device's trickle of IRQ_HANDLED callouts to defeat the lockup detector. (fat chance os this working right). - Add a kernel boot parameter `noirqdebug' to turn the whole thing off. ChangeSet@1.1243.50.4, 2003-06-06 08:53:43-07:00, akpm@digeo.com [PATCH] per-cpu support inside modules (minimal) From: Rusty Russell OK, this does the *minimum* required to support DEFINE_PER_CPU inside modules. If we decide to change kmalloc_percpu later, great, we can turf this out. Basically, overallocates the amount of per-cpu data at boot to at least PERCPU_ENOUGH_ROOM if CONFIG_MODULES=y (arch-specific by default 32k: I have only 7744 bytes of percpu data in my kernel here, so makes sense), and a special allocator in module.c dishes it out. ChangeSet@1.1243.50.3, 2003-06-06 08:53:37-07:00, akpm@digeo.com [PATCH] kmalloc_percpu: interface change From: Rusty Russell Several tweaks to the kmalloc_percpu()/kfree_percpu() interface, to allow future implementations to be more flexible, and make easier to use now we can see how it's actually being used. 1) No flags argument: GFP_ATOMIC doesn't make much sense, 2) Explicit alignment argument, so we don't have to give SMP_CACHE_BYTES alignment always, 3) Zeros memory, since most callers want that and it's not entirely trivial, 4) Convenient type-safe wrapper which takes a typename, and 5) Rename to alloc_percpu/__alloc_percpu, since usage no longer matches kmalloc. ChangeSet@1.1243.50.2, 2003-06-06 08:18:53-07:00, zippel@linux-m68k.org [PATCH] ignore attempts to change unchangable symbols This fixes a problem which can show up with the new select facility, e.g. a symbol is forced to 'y', so we should never even try to change such symbols. ChangeSet@1.1243.50.1, 2003-06-06 08:18:47-07:00, zippel@linux-m68k.org [PATCH] boolean symbol state fix This is an important fix to allow changing boolean symbols, whose dependency is 'm'. All internal symbol states must be converted from the tristate into boolean the state. I missed this change while adding expression support for defaults, please apply. ChangeSet@1.1243.37.36, 2003-06-06 07:22:42-07:00, davem@nuts.ninka.net [NET]: Missing __KERNEL__ ifdefs in linux/{tcp,udp}.h ChangeSet@1.1243.49.1, 2003-06-06 06:11:42-07:00, davem@kernel.bkbits.net Merge davem@nuts.ninka.net:/home/davem/src/BK/net-2.5 into kernel.bkbits.net:/home/davem/net-2.5 ChangeSet@1.1243.37.35, 2003-06-06 04:41:13-07:00, davem@nuts.ninka.net [NET]: Some stuff missed during acme's struct sock cleanup. ChangeSet@1.1243.37.34, 2003-06-06 03:24:56-07:00, davem@nuts.ninka.net Merge bk://kernel.bkbits.net/acme/sock-2.5 into nuts.ninka.net:/home/davem/src/BK/sock-2.5 ChangeSet@1.1243.1.72, 2003-06-06 11:59:11+02:00, perex@suse.cz ALSA update - added AZT3328 driver - added Terratec Aureon support to ICE1724 driver - fixed possible PCI posting problems - ENS1370, ENS1371, FM801, ICE1712, ICE1724, VIA82xx - AC'97 code - added new IDs - fixed typo in S/PDIF code - C-Media related fixes - USB driver - added nrpacks module option - added hack for AudioTrak Optoplay - removed OLD_USB stuff ChangeSet@1.1243.37.33, 2003-06-06 02:54:36-07:00, davem@nuts.ninka.net [IPSEC]: Implement xfrm type module autoloading. ChangeSet@1.1243.37.32, 2003-06-06 02:13:06-07:00, kuznet@oops.inr.ac.ru tcp_output.c, tcp.c, tcp.h: reconciling TCP_CORK and TCP_NODELAY ChangeSet@1.1243.37.31, 2003-06-06 02:05:57-07:00, davem@nuts.ninka.net [NET]: Kill drivers/net/setup.c, it no longer does anything. ChangeSet@1.1243.37.30, 2003-06-06 00:41:05-07:00, hch@lst.de [NET]: Convert ppc early-init drivers to initcalls. Ack'd by Paul Mackerras. ChangeSet@1.1243.37.29, 2003-06-06 00:15:37-07:00, davem@nuts.ninka.net Merge nuts.ninka.net:/home/davem/src/BK/network-2.5 into nuts.ninka.net:/home/davem/src/BK/net-2.5 ChangeSet@1.1243.1.71, 2003-06-06 01:28:45-04:00, jgarzik@redhat.com Merge redhat.com:/garz/repo/linus-2.5 into redhat.com:/garz/repo/net-drivers-2.5 ChangeSet@1.1243.1.70, 2003-06-06 01:13:35-04:00, yoshfuji@linux-ipv6.org [netdrvr] C99 initializers for arcnet ChangeSet@1.1243.1.69, 2003-06-06 01:08:11-04:00, scott.feldman@intel.com [PATCH] remove ethtool privileged references dev_ioctl already checks capable(CAP_NET_ADMIN) for SOICETHTOOL, so privileged reference are not necessary. ChangeSet@1.1243.1.68, 2003-06-06 01:07:46-04:00, scott.feldman@intel.com [PATCH] 10GbE ethtool support Add 10GbE support for ethtool. ChangeSet@1.1243.1.67, 2003-06-06 01:07:21-04:00, zwane@linuxpower.ca [PATCH] cli/sti cleanup for fmvj18x This one should be safe as we're protected by the xmit_lock in all instances ChangeSet@1.1243.1.66, 2003-06-06 01:04:47-04:00, jgarzik@redhat.com [netdrvr] add MAINTAINERS entry for atmel wireless driver ChangeSet@1.1243.1.65, 2003-06-06 01:03:23-04:00, srk@thekelleys.org.uk [netdrvr] add atmel[_cs], new wireless driver Attached is a driver for Atmel at76c50x WiFi cards. This code started out as a GPL release from Atmel of pretty horrible quality and I've extensively re-worked it with the aim of making it acceptable in the kernel. Please could you take a look and either pass it into the patch stream or let me know what's wrong with it? The code has been tested on at least three different brand cards by different people. Jean Tourrilhes took a look at an earlier version an was positive. He's put incorporating this into 2.6 as a priority 1. The patch works fine on 2.5.70. The firmware issue has been addressed now. The only firmware in the driver is a small stub which reads the MAC address from NVRAM on the card. The source for that is included so there are no GPL issues. The main firmware is loaded from userspace using Manuel Estrada Sainz's sysfs firmware class. I know that the patch for that has been accepted but it hasn't turned up anywhere I can see yet. The driver compiles fine even without the firmware class. I've made a package of the firmware images which is available from my website. The remaining issues with the driver are migrating PCMCIA to the new driver model and PCI support. I'm happy to produce followup patches as the PCMCIA system gets evolved to the new driver model: the timing on that is controlled by others. This set of chips includes a PCI version and the driver should support that, but AFAIK there is no PCI hardware available anywhere. If Atmel can provide me with some it will be simple to add PCI support. The driver uses the CRC32 library module and the firmware loader. I've not put in dependencies on those, but when the lastest set of patches go into Kconfig I'll set it up so that selecting the Atmel driver selects CRC32 and FW_LOADER too. ChangeSet@1.1243.1.64, 2003-06-06 00:52:32-04:00, shemminger@osdl.org [PATCH] sb1000 driver bugs Inspecting the sb1000 driver showed some interesting bugs: - net device pointer is used before the device is allocated; gcc does catch this. - unregister is called even though device not registered successfully - net device is not freed on remove. Compiles but don't have hardware to test. Don't know how it ever worked though. ChangeSet@1.1256, 2003-06-06 11:22:26+10:00, paulus@samba.org PPC32: Cleanups from Christoph Hellwig and Tom Rini. ChangeSet@1.1255, 2003-06-06 11:06:35+10:00, porter@cox.net PPC32: Resolve colliding includes in boot wrappers. ChangeSet@1.1254, 2003-06-06 11:04:38+10:00, paulus@samba.org PPC32: Compile fix for array initialization from Christoph Hellwig. ChangeSet@1.1243.37.28, 2003-06-05 17:36:56-07:00, hollisb@us.ibm.com [PATCH] awe_wave.c user pointer dereference Two ioctl functions in sound/oss/awe_wave.c were directly dereferencing a user-supplied pointer in a few places. ChangeSet@1.1243.37.27, 2003-06-05 16:30:39-07:00, torvalds@home.transmeta.com zlib cleanup: final fixups Jörn missed a few places of FAR conversion in inflate ChangeSet@1.1243.37.26, 2003-06-05 16:13:43-07:00, joern@wohnheim.fh-wedel.de [PATCH] zlib cleanup: OF Remove the stale support for K&R function declarations through the OF() macro. This is the last patch to clean up zconf.h, at least for now. ChangeSet@1.1243.37.25, 2003-06-05 16:13:36-07:00, joern@wohnheim.fh-wedel.de [PATCH] zlib cleanup: z_off_t This nice macro must have been one of the good intentions on the road to hell. Completely unused. :) ChangeSet@1.1243.37.24, 2003-06-05 16:13:30-07:00, joern@wohnheim.fh-wedel.de [PATCH] zlib cleanup: ZEXPORT Just a simple s/ZEXPORT//. ChangeSet@1.1243.37.23, 2003-06-05 16:11:39-07:00, joern@wohnheim.fh-wedel.de [PATCH] zlib cleanup: ZEXTERN This one was just simple s/ZEXTERN/extern/g. ChangeSet@1.1243.37.22, 2003-06-05 16:02:37-07:00, joern@wohnheim.fh-wedel.de [PATCH] zlib cleanup: __32BIT__ and STDC Remove unused __32BIT__ and STDC macros ChangeSet@1.1243.37.21, 2003-06-05 16:02:29-07:00, joern@wohnheim.fh-wedel.de [PATCH] zlib cleanup: remove FAR macro This removes FAR, the typedefs using FAR (Bytef and friends) and the function prototypes for zalloc and zfree that should have gone earlier already. ChangeSet@1.1243.46.3, 2003-06-05 15:19:41-07:00, greg@kroah.com TTY: release function should be set in the class, not the class_device. ChangeSet@1.1243.48.2, 2003-06-05 15:16:20-07:00, mochel@osdl.org [driver model] Remove extraneous class device release method. What happens when you get a patch that does something an applied patch already does, but a little better, and merge it sloppily: You end up calling the wrong function because you've defined equivalent methods in two places. Bad Pat, Bad. ChangeSet@1.1243.48.1, 2003-06-05 14:48:21-07:00, mochel@osdl.org Merge osdl.org:/home/mochel/src/kernel/devel/linux-2.5-virgin into osdl.org:/home/mochel/src/kernel/devel/linux-2.5-core ChangeSet@1.1243.46.2, 2003-06-05 14:21:57-07:00, greg@kroah.com TTY: add a release function for tty_class devices. This fixes a race with looking at files in /sys/class/tty/ and removing a tty device. ChangeSet@1.1243.43.4, 2003-06-05 13:48:47-07:00, mochel@osdl.org [kobject] Add warning + back trace if kobject_get() is called with 0 refcount. ChangeSet@1.1243.43.3, 2003-06-05 13:25:00-07:00, mochel@osdl.org [driver model] Add device_for_each_child iterator. From Mike Anderson: I have been using it on an outstanding patch for scsi_set_host_offline. It appears to work fine in my testing. ChangeSet@1.1243.45.1, 2003-06-05 13:22:30-07:00, Kurt.Robideau@comtrol.com [PATCH] Rocketport driver against 2.5.70-bk7 Update the Rocketport driver ChangeSet@1.1243.25.4, 2003-06-05 13:14:37-07:00, greg@kroah.com [PATCH] I2C: sync i2c-id.h with cvs version. ChangeSet@1.1243.44.2, 2003-06-05 12:52:22-07:00, shemminger@osdl.org [PATCH] typo in new class_device_release There is a typo in the current 2.5.70 bk version of class_device_release that was not there in my original patch. By confusing the class and the class_device, the release function oops. cd->release is always the function itself (class_device_release), cls->release is the one setup for the class (net_class in my case). ChangeSet@1.1243.25.3, 2003-06-05 12:45:18-07:00, aschultz@warp10.net [PATCH] I2C: fix unsafe usage of list_for_each in i2c-core i2c-core.c contains 2 loops that iterate over the list of the clients attached to an adapter and detaches them. Detaching the clients will actually remove them from the list the loop is iterating over. Therefore the list_for_each_safe() method has to be used. ChangeSet@1.1243.25.2, 2003-06-05 12:45:04-07:00, mhoffman@lightlink.com [PATCH] I2C: more w83781d fixes This patch fixes the various return values in the w83781d_detect() error paths. It also cleans up some formatting here and there. It should be applied on top of the previous one. It works for me; same caveat as above w.r.t. ISA. ChangeSet@1.1243.37.15, 2003-06-05 12:04:44-07:00, greg@kroah.com [PATCH] PCI: remove EXPORT_SYMBOL(pci_devices) Now the only users of this directly should be the pci core and arch specific pci core code. ChangeSet@1.1243.37.14, 2003-06-05 12:04:33-07:00, greg@kroah.com [PATCH] PCI: move pci_present() into drivers/pci/search.c This will let not have to export the pci_devices variable. ChangeSet@1.1243.37.13, 2003-06-05 12:04:22-07:00, greg@kroah.com [PATCH] PCI: remove pci_for_each_dev_reverse() now that all users of it are gone. ChangeSet@1.1243.37.12, 2003-06-05 12:04:12-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev_reverse() in ChangeSet@1.1243.37.11, 2003-06-05 12:04:02-07:00, greg@kroah.com [PATCH] PCI: add pci_find_device_reverse() for users of pci_find_each_dev_reverse() to use ChangeSet@1.1243.37.10, 2003-06-05 12:03:52-07:00, greg@kroah.com [PATCH] PCI: fix up previous fusion driver pci changes This makes the driver build properly now, and removes a direct access of the pci_devices variable. ChangeSet@1.1243.37.9, 2003-06-05 12:03:40-07:00, greg@kroah.com [PATCH] PCI: remove direct access of pci_devices from drivers/macintosh/via-pmu68k.c Yeah, this is commented out code, but just trying to be complete... ChangeSet@1.1243.37.8, 2003-06-05 12:03:29-07:00, greg@kroah.com [PATCH] PCI: remove direct access of pci_devices from arch/m68k/atari/hades-pci.c ChangeSet@1.1243.42.5, 2003-06-05 18:06:10+01:00, davej@codemonkey.org.uk [CPUFREQ] Fix ACPI P-State driver. from Dmitry Torokhov. I have the following problems with ACPI P-States driver: - It crashes because it tries to switch CPU speed without registering cpufreq driver first but acpi_processor_set_performance calls cpufreq_notify_transition. - When testing for capable CPUs it skips all online ones so for my single CPU notebook it can't activate at all. - If a processor does not support throttling then it will say that "limit" interface is not supported even after activating performance control. The patch below should fix these issues. It also adds some info messages since /proc/acpi/processor/*/performance interface is marked obsolete but i still would like to see if P-states were recognized during boot. ChangeSet@1.1243.42.4, 2003-06-05 17:26:43+01:00, davej@codemonkey.org.uk [CPUFREQ] CodingStyle fixes ChangeSet@1.1243.43.2, 2003-06-05 09:23:44-07:00, mochel@osdl.org [kobject] Remove kobj_lock and use lockless refcounting. The only thing preventing this from happening earlier was the circular sysfs registration dependency - it would need to be initialized before it was registered, but needed to be registered before it was initialized. With kobjects gone from struct filesystem_type, the dependency no longer exists and we don't have to special-case the possibility that a kobject will be passed to kobject_get with a refcount == 0. Note that a kobject with a count of 0 in that function is still a bug, but one in the subsystem making the call. We should add a debugging hook to dump the stack if it does happen. ChangeSet@1.1243.42.3, 2003-06-05 17:21:57+01:00, davej@codemonkey.org.uk [CPUFREQ] Kill unused variables. ChangeSet@1.1243.42.2, 2003-06-05 17:10:13+01:00, davej@codemonkey.org.uk [CPUFREQ] kill cpufreq_driver export. From Dominik Brodowski. This removes the special export of cpufreq_driver for proc_intf.c. Instead, the behaviour of /proc/cpufreq previous of Greg's class re-write is back: the check whether cpufreq_driver is loaded is done within cpufreq_cpu_get ChangeSet@1.1243.43.1, 2003-06-05 08:55:13-07:00, mochel@osdl.org [fs] Remove kobject support for filesystems It was initially added for the immediate gain of being able to see what filesystems were registered in the system. However, it was done without proper analysis of the lifetime rules associated with them. Someday, we should convert struct filesystem_type and struct super_block to use kobjects, though it will take sufficent time and effort to get it right. Until then, we go without.. ChangeSet@1.1243.37.7, 2003-06-05 08:47:10-07:00, paulus@samba.org [PATCH] get rid of CONFIG_ALL_PPC This patch gets rid of CONFIG_ALL_PPC, which was a very confusing option, since it didn't actually mean "ALL" at all, it was more a "common set" thing. The primary replacement for CONFIG_ALL_PPC is CONFIG_PPC_MULTIPLATFORM. I have also defined CONFIG_PPC_PMAC, CONFIG_PPC_PREP and CONFIG_PPC_CHRP for selecting code which is only needed for one of the three platforms that CONFIG_ALL_PPC represented. This is something that we (the PPC community) have been talking about doing for some time. There is also a CONFIG_PPC_OF which is for PPC machines with Open Firmware, which is currently powermacs and CHRP machines. At the moment, CONFIG_PPC_{PMAC,PREP,CHRP,OF} get unconditionally defined if CONFIG_PPC_MULTIPLATFORM is selected, but in future this split will let us have more control over what gets included, so that for example we don't necessarily have to include powermac bits in a kernel for a PReP machine. I have gone through the uses of CONFIG_ALL_PPC one by one and decided which of the new symbols best represents the set of machines that need the code in question. In fact most of the uses of CONFIG_ALL_PPC in the drivers have been replaced by CONFIG_PPC_PMAC. The other symbols are mostly confined to the PPC architecture code. ChangeSet@1.1243.42.1, 2003-06-05 16:43:33+01:00, davej@codemonkey.org.uk [CPUFREQ] Make powernow-k7 leap big buildings^Wranges. When scaling upwards, we need to change the voltage before the frequency. When scaling downwards, the opposite. This may fix some of the hangs people have been seeing when jumping from a low frequency to a very high frequency. ChangeSet@1.1243.41.2, 2003-06-05 08:31:38-07:00, axboe@suse.de [PATCH] kill old stuff and fix the start/stop thing as well. I think this is all of them. ChangeSet@1.1243.40.4, 2003-06-05 01:27:46-07:00, oliver@neukum.org [PATCH] USB: usb_set_configuration in empeg.c you should not drop errors. - proper error detection during initialisation ChangeSet@1.1243.40.3, 2003-06-05 01:27:33-07:00, oliver@neukum.org [PATCH] USB: cut usb_set_config from hpusbscsi this cuts out old cruft. ChangeSet@1.1243.40.2, 2003-06-05 01:27:20-07:00, david-b@pacbell.net [PATCH] USB: kerneldoc for gadget API Here's the non-inlined doc for the gadget API. ChangeSet@1.1243.40.1, 2003-06-05 01:12:30-07:00, bcollins@debian.org [PATCH] USB: fix keyboard leds > Ben, it looks like your patch broke something for USB keyboards, any > idea? Yep, my patch killed hid-input from scanning HID_OUTPUT_REPORT's. Fixed with this patch for 2.5.70+bk. I'll send one for 2.4.x in a few minutes. ChangeSet@1.1243.38.25, 2003-06-05 01:08:41-07:00, elenstev@mesatop.com [SPARC]: Fix non-ansi parameter lists. ChangeSet@1.1243.38.24, 2003-06-05 01:03:38-07:00, bcollins@debian.org [SPARC64]: Final image strip not to strip too much. ChangeSet@1.1243.39.3, 2003-06-05 01:01:48-07:00, steve@gw.chygwyn.com [AX25]: Sanitize ax25 netdevice private handling. ChangeSet@1.1243.38.23, 2003-06-05 01:00:17-07:00, bde@nwlink.com [SPARC64]: Fix transmit handling in sunsab.c serial driver. ChangeSet@1.1243.39.2, 2003-06-05 00:50:45-07:00, viro@parcelfarce.linux.theplanet.co.uk [NET]: Convert USB drivers away from init_etherdev(). ChangeSet@1.1243.39.1, 2003-06-05 00:44:33-07:00, davem@nuts.ninka.net Merge nuts.ninka.net:/home/davem/src/BK/network-2.5 into nuts.ninka.net:/home/davem/src/BK/net-2.5 ChangeSet@1.1243.37.5, 2003-06-04 23:24:39-07:00, davem@kernel.bkbits.net Merge davem@nuts.ninka.net:/home/davem/src/BK/net-2.5 into kernel.bkbits.net:/home/davem/net-2.5 ChangeSet@1.1243.37.4, 2003-06-04 23:50:39-07:00, shemminger@osdl.org [NET]: Fix device unregister in TUN driver. ChangeSet@1.1243.38.22, 2003-06-04 23:33:10-07:00, zippel@linux-m68k.org [PATCH] choice handling fixes A few choice handling fixes: - only visible choice values define the new state of the complete choice - improve handling of choices without visible value - two new warnings ChangeSet@1.1243.38.21, 2003-06-04 23:30:14-07:00, neilb@cse.unsw.edu.au [PATCH] Fix bug in /proc/mdstat If /proc/mdstat is large, or reads are for a small size, then the last line of /proc/mdstat is repeated infinitely. This patch will fix it. ChangeSet@1.1243.38.20, 2003-06-04 23:30:06-07:00, neilb@cse.unsw.edu.au [PATCH] Fix up freeing of kmalloc structures Some paths free things twice, others free un-initialised values :-( Not any more. ChangeSet@1.1243.38.19, 2003-06-04 23:29:59-07:00, neilb@cse.unsw.edu.au [PATCH] Fix raid1 handling of writing to multiple devices. When raid1 writes, it needs to schedule writes to some number of devices, and when all writes have completed, the r1_bio structure that holds it all together must be freed. However we must make sure not to free it before all devices have been considered for submitting writes to. This happens in two places: when submitting a normal write request and when submiting a write as part of resync. This patch makes both these places: the same simpler more correct. ChangeSet@1.1243.38.18, 2003-06-04 23:29:49-07:00, neilb@cse.unsw.edu.au [PATCH] Fix raid5 bug where wrong 'dev' is used. ChangeSet@1.1243.37.3, 2003-06-04 23:10:25-07:00, davem@nuts.ninka.net [PCI]: Move pci_remove_bus_device back to include/linux/pci.h, discussed with greg@kroah.com ChangeSet@1.1243.38.17, 2003-06-04 22:35:36-07:00, davidm@napali.hpl.hp.com [PATCH] allow thread_info to be allocated as part of task_struct This re-organizes the thread info allocations a bit, so that architectures like ia64 that want to allocate the thread_info structure as part of the task structure allocation can do so. The bulk of the patch is adding the "tsk" information to the thread info allocator (ignored by all non-ia64 architectures), and switching the order of the allocators to make this all possible. ChangeSet@1.1243.38.16, 2003-06-04 22:34:09-07:00, akpm@digeo.com [PATCH] ext3: fix deadlock in journal_create() From: Mark Fasheh journal_create() is called under down_write(s_umount) on the mount path. But it calls fsync_bdev(), which wants down_write(s_umount). Change journal_create() to run sync_blockdev() instead. Just to write out the new journal's blocks. ChangeSet@1.1243.38.15, 2003-06-04 22:34:02-07:00, akpm@digeo.com [PATCH] DEADLINE: hash removal fix From Jens: It fixes the occasional BUG_ON(list_empty(&req->queuelist)); which people have been hitting in include/linux/blkdev.h: blkdev_dequeue_request() ChangeSet@1.1243.38.14, 2003-06-04 22:33:54-07:00, akpm@digeo.com [PATCH] hugetlbfs: mount options and permissions From: "Seth, Rohit" This patch enables system administrators to more flexibly assign read/write permissions to files in hugetlbfs. Mount options are added which will specify the uid, gid and mode of the filesystem's root directory. This patch also enables super users to do chown/chgrp on files in hugetlbfs. Also included in this patch is the support for uid, gid and mode command line options for mount command for hugetlbfs. The following mount options are added to hugetlbfs: uid=value gid=value mode=value The values may be specified in hex (0xNNN), decimal or octal (0NNN). ChangeSet@1.1243.38.13, 2003-06-04 22:33:41-07:00, akpm@digeo.com [PATCH] reiserfs support for blocksizes other than 4096 bytes From: Oleg Drokin This patch allows reiserfs to support blocksizes from 1024 bytes and up to PAGE_CACHE_SIZE. Also it fixes two glitches that prevent reiserfs from working correctly in case if PAGE_CACHE_SIZE is bigger than blocksize. Originally this patch was created by Edward Shushkin and Vladimir Saveliev, and then it was adapted to modern 2.4 and 2.5 by me. Also people should be aware that 1024 bytes blocksize is not very good thing as tree grows very fast. mkreiserfs is able to create filesystems with different blocksizes for quite a while already (-b switch). ChangeSet@1.1243.38.12, 2003-06-04 22:33:33-07:00, akpm@digeo.com [PATCH] reiserfs option parser fix and ability to pass From: Oleg Drokin This patch adds support for remounting taking into account all extra options you may want to pass it. (by Jeff Mahooney). Also it reworks the parser to correctly deal with mutually exclusive options. Now whatever option was specified last will take an effect. - If you pass more than one jdev= option, the error will be reported and mount/remount refused. - If you pass incorrect alloc= suboptions, the mount/remount will fail. (it did not before). - nolargeio now actually looks at its argument. Argument is expected to be numeric. If it is zero, default io size is set to 128k (default setting), if it is non-zero, default io size is set to PAGE_SIZE. ChangeSet@1.1243.38.11, 2003-06-04 22:33:24-07:00, akpm@digeo.com [PATCH] Fix generic_file_write() again. From: "Milton D. Miller II" The code at present has a small problem: when a fault is encountered we will run commit_write() to cover the amount of data which was successfully copied in from userspace. But filemap_copy_from_user() may have zeroed out some more of the page. So pagecache now has zeroes and the buffer_head which represents those zeroes is not dirtied. So a subsequent eviction and re-read of the file in the window beyond the faulting offset will return the file's old contents and not the zeroes. So we change filemap_copy_from_user_iovec() to have the same behaviour as the non-iovec filemap_copy_from_user(), and ensure that the commit_write() covers the parts of the page which copy_from_user() zeroed out. ChangeSet@1.1243.38.10, 2003-06-04 22:33:15-07:00, akpm@digeo.com [PATCH] Graceful failure in devfs_remove() From: Pavel Roskin , via Christoph Hellwig It's already the second time that I encounter a kernel panic in the same place. When devfs_remove() is called on a non-existent file entry, the kernel panics and I have to reboot the system. First time it was unregistering of pseudoterminals. This time it's ide-floppy module that doesn't register devfs entries if the media is absent but still tries to unregister them. The bug in ide-floppy will be reported separately. The point of this message is that the failure in devfs_remove() is possible, especially with rarely used drivers. Secondly, is not fatal enough to justify an immediate panic and reboot. Thirdly, devfs misses a chance to tell the user what's going wrong. ChangeSet@1.1243.38.9, 2003-06-04 22:33:02-07:00, akpm@digeo.com [PATCH] fix bw-qcam.c bad copy_to_user From: Hollis Blanchard Like radio-cadet.c, bw-qcam.c is calling copy_to_user() where it shouldn't. The user buffer is copied to/from kernel space by drivers/media/video/videodev.c:video_usercopy(). ChangeSet@1.1243.38.8, 2003-06-04 22:32:54-07:00, akpm@digeo.com [PATCH] pnpbios dereferencing user pointer From: Hollis Blanchard Another simple case of a memcpy that should be copy_from_user... Also fix some error-path memory leaks. ChangeSet@1.1243.38.7, 2003-06-04 22:32:46-07:00, akpm@digeo.com [PATCH] remove unsafe BUG_ON() From: Hugh Dickins PageDirty BUG_ON in __remove_inode_page is, and always has been, unsafe for SMP or preemption: truncation may be racing against unmapping's set_page_dirty in zap_pte_range (amongst a few other possibilities). !PageUptodate error in __set_page_dirty_buffers is unsafe then too. ChangeSet@1.1243.38.6, 2003-06-04 22:32:38-07:00, akpm@digeo.com [PATCH] zr36120: fix improper access to userspace From: Hollis Blanchard Fix a direct userspace access, found by the Stanford checker. ChangeSet@1.1243.38.5, 2003-06-04 22:32:16-07:00, akpm@digeo.com [PATCH] cmpci: fix improper access to userspace From: Hollis Blanchard Fix a direct userspace access, found by the Stanford checker. ChangeSet@1.1243.38.4, 2003-06-04 22:31:42-07:00, akpm@digeo.com [PATCH] radio-cadet.c: remove unnecessary copy_to_user() From: Hollis Blanchard As pointed out by the Stanford checker, 'v' is not tainted. The driver shouldn't be using copy_to_user() in cadet_do_ioctl() at all: cadet_do_ioctl() is being called by drivers/media/video/videodev.c: video_usercopy(), which has already copied the buffer 'arg' (aka 'v') into kernel space, and will copy it back after cadet_do_ioctl() returns. So all the direct 'v' accesses are correct. ChangeSet@1.1243.38.3, 2003-06-04 22:31:27-07:00, akpm@digeo.com [PATCH] clean up timer interpolation code From: Christoph Hellwig - don't add one level of indentation when taking a lock - remove useless ti_global struct ChangeSet@1.1243.38.2, 2003-06-04 22:31:03-07:00, akpm@digeo.com [PATCH] aio: small cleanups - comment fixes - duplicated assignments - Remove a prototype which is in aio.h already - Some __user annotation - use the existing symbolic names, not magic numbers. ChangeSet@1.1243.37.2, 2003-06-04 22:26:40-07:00, akpm@digeo.com [NET]: Fix broken cleanups in net/core/iovec.c ChangeSet@1.1243.38.1, 2003-06-04 22:25:07-07:00, akpm@digeo.com [PATCH] fix broken networking Fix broken "cleanup" ChangeSet@1.1243.1.63, 2003-06-05 00:38:08-04:00, reeja.john@amd.com [netdrvr amd8111e] link against mii lib ChangeSet@1.1243.1.62, 2003-06-05 00:36:38-04:00, jgarzik@redhat.com [netdrvr skge] add ULL modifier to 64-bit constant ChangeSet@1.1243.1.61, 2003-06-05 00:01:42-04:00, jgarzik@redhat.com [netdrvr] gcc 3.3 cleanups Mostly adding 'ULL' modifier to 64-bit constants. ChangeSet@1.1243.35.2, 2003-06-04 18:57:45-07:00, rth@kanga.twiddle.net [ALPHA] Add posix timer and clock syscalls. ChangeSet@1.1243.35.1, 2003-06-04 18:28:50-07:00, Jeff.Wiedemeier@hp.com [PATCH] compile fix for agp_memory struct definitition change In 2.5.70, the agp_memory struct changed from: typedef struct agp_memory to: struct agp_memory This patch propagates that change into include/asm-alpha/agp_backend.h and arch/alpha/kernel/core_{titan,marvel}.c /jeff ChangeSet@1.1243.34.1, 2003-06-04 22:00:32-03:00, acme@conectiva.com.br o net: create struct sock_common and use in struct sock & tcp_tw_bucket With this the data dependency is reduced to just making sure that the first member of both struct sock and struct tcp_tw_bucket are a struct sock_common. Also makes it easier to grep for struct sock and struct tcp_tw_bucket usage in the tree as all the members in those structs are prefixed, respectively, with sk_ and tw_, like struct inode (i_), struct block_device (bd_), etc. Checked namespace with make tags/ctags, just one colision with the macros for the struct sock members, with a wanrouter struct, fixed that s/sk_state/state_sk/g in the wanrouter struct. Checked as well if the names of the members in both structs collided with some macro, none found. ChangeSet@1.1243.1.59, 2003-06-04 17:51:51-07:00, greg@kroah.com PCI: remove usage of pci_for_each_dev() in arch/ppc64/kernel/pci.c ChangeSet@1.1243.1.58, 2003-06-04 17:42:57-07:00, greg@kroah.com [PATCH] PCI: finally remove pci_for_each_dev() now that all users of it are gone. ChangeSet@1.1243.1.57, 2003-06-04 17:42:47-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/x86_64/pci/x86-64.c ChangeSet@1.1243.1.56, 2003-06-04 17:42:37-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/x86_64/pci/irq.c ChangeSet@1.1243.1.55, 2003-06-04 17:42:26-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/x86_64/kernel/pci-gart.c ChangeSet@1.1243.1.54, 2003-06-04 17:42:16-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/x86_64/kernel/bluesmoke.c ChangeSet@1.1243.1.53, 2003-06-04 17:42:05-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/v850/kernel/rte_mb_a_pci.c ChangeSet@1.1243.1.52, 2003-06-04 17:41:55-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/sh/kernel/pci-sh7751.c ChangeSet@1.1243.1.51, 2003-06-04 17:41:45-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/ppc64/kernel/pci_dma.c ChangeSet@1.1243.1.50, 2003-06-04 17:41:34-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/ppc64/kernel/pSeries_pci.c ChangeSet@1.1243.1.49, 2003-06-04 17:41:24-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/ppc64/kernel/iSeries_pci.c ChangeSet@1.1243.1.48, 2003-06-04 17:41:14-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/ppc/platforms/prep_pci.c ChangeSet@1.1243.1.47, 2003-06-04 17:41:03-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/ppc/platforms/pmac_pci.c ChangeSet@1.1243.1.46, 2003-06-04 17:40:52-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/ppc/platforms/gemini_pci.c ChangeSet@1.1243.1.45, 2003-06-04 17:40:42-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/ppc/platforms/chrp_pci.c ChangeSet@1.1243.1.44, 2003-06-04 17:40:32-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/ppc/kernel/pci.c ChangeSet@1.1243.1.43, 2003-06-04 17:40:21-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/mips64/sgi-ip32/ip32-pci.c ChangeSet@1.1243.1.42, 2003-06-04 17:40:11-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/mips64/mips-boards/generic/pci.c ChangeSet@1.1243.1.41, 2003-06-04 17:40:01-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/mips/sni/pci.c ChangeSet@1.1243.1.40, 2003-06-04 17:39:50-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/mips/mips-boards/generic/pci.c ChangeSet@1.1243.1.39, 2003-06-04 17:39:40-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/mips/ddb5xxx/ddb5477/pci.c ChangeSet@1.1243.1.38, 2003-06-04 17:39:30-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/mips/ddb5476/pci.c ChangeSet@1.1243.1.37, 2003-06-04 17:39:20-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/mips/ddb5074/pci.c ChangeSet@1.1243.1.36, 2003-06-04 17:39:09-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/ia64/sn/io/sn2/pci_bus_cvlink.c ChangeSet@1.1243.1.35, 2003-06-04 17:39:00-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/ia64/sn/io/pciba.c ChangeSet@1.1243.1.34, 2003-06-04 17:38:49-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/ia64/sn/io/pci_bus_cvlink.c ChangeSet@1.1243.1.33, 2003-06-04 17:38:38-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/ia64/hp/common/sba_iommu.c ChangeSet@1.1243.1.32, 2003-06-04 17:37:54-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/arm/kernel/bios32.c ChangeSet@1.1243.1.31, 2003-06-04 17:37:31-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/alpha/kernel/sys_sio.c ChangeSet@1.1243.1.30, 2003-06-04 17:29:10-07:00, greg@kroah.com Merge gregkh@kernel.bkbits.net:/home/gregkh/linux/pci-good-2.5 into kroah.com:/home/greg/linux/BK/p-2.5 ChangeSet@1.1243.33.4, 2003-06-04 23:55:46+01:00, rmk@flint.arm.linux.org.uk [ARM] Tidy up Integrator core support. This merges arch.c, irq.c and mm.c into one core file for this platform; it's pointless keeping these separate. ChangeSet@1.1243.33.3, 2003-06-04 23:18:21+01:00, rmk@flint.arm.linux.org.uk [ARM] Convert platform devices to use platform_device Since struct platform_device now has the ability to pass resources, defined by the platform to the device driver, we can now use this to handle platform specific devices. One such instance is the StrongARM SA1111 companion chip, which can appear in various address spaces, and connected to different IRQ lines depending on how many cups of coffee the hardware designer had, the direction of the wind outside the designers office that day. We also convert some of the other StrongARM peripheral on-chip devices to use struct platform_device. ARM also provides a platform_add_devices() function which can be used by platform code to bulk-register a tabular set of platform devices. ChangeSet@1.1243.9.45, 2003-06-04 13:34:19-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/i386/pci/i386.c ChangeSet@1.1243.9.44, 2003-06-04 13:34:03-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/i386/kernel/cpu/cpufreq/gx-suspmod.c ChangeSet@1.1243.9.43, 2003-06-04 13:33:49-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in arch/i386/pci/irq.c ChangeSet@1.1243.9.42, 2003-06-04 12:34:55-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/video/sis/sis_main.c ChangeSet@1.1243.9.41, 2003-06-04 12:34:45-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/video/pm2fb.c ChangeSet@1.1243.9.40, 2003-06-04 12:34:35-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/pnp/resource.c ChangeSet@1.1243.9.39, 2003-06-04 12:34:25-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/pci/setup-irq.c ChangeSet@1.1243.9.38, 2003-06-04 12:34:15-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/pci/search.c ChangeSet@1.1243.9.37, 2003-06-04 12:34:05-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/pci/proc.c ChangeSet@1.1243.9.36, 2003-06-04 12:33:55-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/pci/pci.c ChangeSet@1.1243.9.35, 2003-06-04 12:33:45-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/pci/hotplug/ibmphp_core.c ChangeSet@1.1243.9.34, 2003-06-04 12:33:35-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/parport/parport_pc.c ChangeSet@1.1243.9.33, 2003-06-04 12:33:25-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/net/ixgb/ixgb_main.c ChangeSet@1.1243.9.32, 2003-06-04 12:33:15-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/net/e1000/e1000_main.c ChangeSet@1.1243.9.31, 2003-06-04 12:33:05-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/net/e100/e100_main.c ChangeSet@1.1243.9.30, 2003-06-04 12:32:55-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/message/i2o/i2o_core.c ChangeSet@1.1243.9.29, 2003-06-04 12:32:45-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/message/fusion/mptbase.c ChangeSet@1.1243.9.28, 2003-06-04 12:32:35-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/macintosh/via-pmu.c ChangeSet@1.1243.9.27, 2003-06-04 12:32:25-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/ide/setup-pci.c ChangeSet@1.1243.9.26, 2003-06-04 12:32:15-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/ide/pci/pdc202xx_new.c ChangeSet@1.1243.9.25, 2003-06-04 12:32:05-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/ide/pci/hp2366.c ChangeSet@1.1243.9.24, 2003-06-04 12:31:55-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/ide/pci/cs5530.c ChangeSet@1.1243.9.23, 2003-06-04 12:31:45-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/char/watchdog/i810-tco.c ChangeSet@1.1243.9.22, 2003-06-04 12:31:35-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/char/watchdog/amd7xx_tco.c ChangeSet@1.1243.9.21, 2003-06-04 12:31:25-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/char/hw_random.c ChangeSet@1.1243.9.20, 2003-06-04 12:31:15-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/char/agp/isoch.c ChangeSet@1.1243.9.19, 2003-06-04 12:31:05-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/char/agp/generic.c ChangeSet@1.1243.9.18, 2003-06-04 12:30:55-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/char/agp/amd-k8-agp.c ChangeSet@1.1243.9.17, 2003-06-04 12:30:45-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in drivers/acpi/pci_irq.c ChangeSet@1.1243.9.16, 2003-06-04 12:30:35-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in sound/pci/rme9652/hammerfall_mem.c ChangeSet@1.1243.9.15, 2003-06-04 12:30:25-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in sound/oss/via82cxxx_audio.c ChangeSet@1.1243.9.14, 2003-06-04 12:30:15-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in sound/oss/maestro.c ChangeSet@1.1243.9.13, 2003-06-04 12:30:06-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in sound/oss/esssolo1.c ChangeSet@1.1243.9.12, 2003-06-04 12:29:53-07:00, greg@kroah.com [PATCH] PCI: remove usage of pci_for_each_dev() in sound/core/memalloc.c ChangeSet@1.1243.9.11, 2003-06-04 10:29:42-07:00, greg@kroah.com [PATCH] PCI: Grab reference count on pci_dev if the pci driver binds to the device. And remember to decrement the count after remove() is called. ChangeSet@1.1243.33.2, 2003-06-04 16:11:38+01:00, rmk@flint.arm.linux.org.uk [ARM] Move dma_alloc_coherent() to consistent.c ChangeSet@1.1243.32.32, 2003-06-04 04:17:33-07:00, thomas@osterried.de [AX25]: AX.25 bug fixes. - Flxnet CRC handling fix for mkiss.c - Use after free bug in ax25_ip.c ChangeSet@1.1243.32.31, 2003-06-04 04:12:30-07:00, davem@nuts.ninka.net [NET]: Fix thinkos in PPC 8260_io/fcc_enet.c changes. ChangeSet@1.1243.32.30, 2003-06-04 04:04:45-07:00, davem@nuts.ninka.net [NET]: Actually apply Al's sunqe.c changes. ChangeSet@1.1243.32.29, 2003-06-04 03:59:39-07:00, davem@nuts.ninka.net [NET]: Convert PPC 8xx_io/fec.c away from init_etherdev(). ChangeSet@1.1243.32.28, 2003-06-04 03:55:43-07:00, davem@nuts.ninka.net [NET]: Convert PPC 8xx_io/enet.c away from init_etherdev(). ChangeSet@1.1243.32.27, 2003-06-04 03:51:42-07:00, davem@nuts.ninka.net [NET]: Convert PPC 8260_io/fcc_enet.c away from init_etherdev(). ChangeSet@1.1243.32.26, 2003-06-04 03:49:17-07:00, davem@nuts.ninka.net [NET]: Convert PPC 8260_io/enet.c away from init_etherdev(). ChangeSet@1.1243.32.25, 2003-06-04 03:39:01-07:00, davem@nuts.ninka.net [NET]: Convert ia64 simeth.c away from init_etherdev(). ChangeSet@1.1243.32.24, 2003-06-04 03:34:22-07:00, viro@parcelfarce.linux.theplanet.co.uk [NET]: Move bmac.c away from init_etherdev(). ChangeSet@1.1243.32.23, 2003-06-04 03:30:15-07:00, davem@nuts.ninka.net [NET]: Move arch/cris drivers away from init_etherdev(). ChangeSet@1.1243.32.22, 2003-06-04 03:11:52-07:00, viro@parcelfarce.linux.theplanet.co.uk [NET]: Move sunqe.c driver away from init_etherdev(). ChangeSet@1.1243.32.21, 2003-06-04 02:51:25-07:00, herbert@gondor.apana.org.au [IPSEC]: Include linux/slab.h where necessary. ChangeSet@1.1243.33.1, 2003-06-04 10:35:15+01:00, rmk@flint.arm.linux.org.uk [ARM] Fix more missing irqreturn_t and remove a static no_action func. ChangeSet@1.1243.32.20, 2003-06-04 02:24:19-07:00, viro@parcelfarce.linux.theplanet.co.uk [NET]: Move 3c509 driver away from init_etherdev(). ChangeSet@1.1243.32.19, 2003-06-04 02:22:31-07:00, viro@parcelfarce.linux.theplanet.co.uk [NET]: Move sk98lin driver away from init_etherdev(). ChangeSet@1.1243.32.18, 2003-06-04 01:17:04-07:00, hch@lst.de [NET]: Fix coding style in net/core/iovec.c ChangeSet@1.1243.32.17, 2003-06-04 01:15:42-07:00, hch@lst.de [NET]: Fix coding style in net/core/filter.c ChangeSet@1.1243.32.16, 2003-06-04 01:14:38-07:00, hch@lst.de [NET]: net/core/dst.c typo. ChangeSet@1.1243.32.15, 2003-06-04 00:59:28-07:00, hch@lst.de [NET]: Kill useless/wrong Version line from net/core/dv.c ChangeSet@1.1243.32.14, 2003-06-04 00:48:08-07:00, yoshfuji@linux-ipv6.org [IPV6]: typo, unrequired #undef and bad operator precedence. - no need to #undef CONFIG_IPV6_SUBTREE - use parentheses around "&" and "|". - fib_repair_tree() is typo. ChangeSet@1.1243.9.10, 2003-06-04 00:02:59-07:00, greg@kroah.com [PATCH] PCI: Move more functions out of include/linux/pci.h that don't need to be there. ChangeSet@1.1243.9.9, 2003-06-03 23:33:47-07:00, greg@kroah.com [PATCH] PCI: remove CONFIG_PROC_FS checks in .c files. ChangeSet@1.1243.32.13, 2003-06-03 22:50:13-07:00, jmorris@intercode.com.au [CRYPTO]: Use "select" kconfig facility instead of fragile defaults. ChangeSet@1.1243.9.8, 2003-06-03 22:48:45-07:00, greg@kroah.com [PATCH] PCI: Remove a lot of PCI core only functions from include/linux/pci.h ChangeSet@1.1243.32.12, 2003-06-03 22:45:40-07:00, yoshfuji@linux-ipv6.org [IPV6]: Fix several errors in udpv6_connect(). - Pointer within an automatic storage class variable fl was illegally cached using ip6_dst_store() - uninitialized saddr was copied to fl.fl6_src - dont cache if ipv6_saddr_get() failed. Based upon patch from Ville Nuorvala (vnuorval@tcs.hut.fi) ChangeSet@1.1243.32.11, 2003-06-03 22:29:21-07:00, hch@lst.de [NET]: Fix non-modular sdla.c build. ChangeSet@1.1243.32.10, 2003-06-03 22:18:50-07:00, davem@nuts.ninka.net [IPSEC]: {esp,ah} --> {esp4,ah4}. ChangeSet@1.1243.32.9, 2003-06-03 22:15:30-07:00, yoshfuji@linux-ipv6.org [IPV6]: Set dead flag on idev if snmp6_register_dev() fails. ChangeSet@1.1243.32.8, 2003-06-03 22:14:10-07:00, hch@lst.de [NET]: Move dmascc away from setup.c ChangeSet@1.1243.32.7, 2003-06-03 22:07:05-07:00, yoshfuji@linux-ipv6.org [NET]: Add ip-sysctl.txt entries for missing ip{,6}frag_* sysctls. ChangeSet@1.1243.32.6, 2003-06-03 22:05:20-07:00, rddunlap@osdl.org [IPV6]: Add IPv6 routing table statistic: fib_discarded_routes. ChangeSet@1.1243.32.5, 2003-06-03 22:00:27-07:00, rddunlap@osdl.org [NET]: Typo corrections only. ChangeSet@1.1243.32.4, 2003-06-03 21:56:35-07:00, rddunlap@osdl.org [NET]: add RFC references for Linux SNMP MIBs. ChangeSet@1.1243.32.3, 2003-06-03 21:54:50-07:00, davem@nuts.ninka.net [NET]: Use INIT_LIST_HEAD correctly in arch/um/drivers/net_kern.c ChangeSet@1.1243.32.2, 2003-06-03 21:53:51-07:00, chas@cmf.nrl.navy.mil [ATM]: more cleanup in HE driver ChangeSet@1.1243.9.7, 2003-06-03 20:54:30-07:00, greg@kroah.com PCI Hotplug: move drivers/hotplug/* to drivers/pci/hotplug/* This will let include/linux/pci.h get smaller, and is what I should have done in the first place 2 years ago... ChangeSet@1.1243.9.6, 2003-06-03 19:36:54-07:00, greg@kroah.com [PATCH] IBM PCI hotplug: remove direct access of pci_devices variable. ChangeSet@1.1243.23.22, 2003-06-03 17:37:54-07:00, ahaas@airmail.net [PATCH] C99 struct initializers for kernel files ChangeSet@1.1243.23.21, 2003-06-03 17:37:46-07:00, ahaas@airmail.net [PATCH] C99 patches for fs/ Add C99 initializers to fs/bio.c and fs/dquot.c. And fs/libfs.c had an obsolete GCC style initialzers that is converted to C99 style. ChangeSet@1.1243.1.26, 2003-06-03 17:24:15-07:00, mochel@osdl.org Merge bk://ldm@bkbits.net/linux-2.5-core into osdl.org:/home/mochel/src/kernel/devel/linux-2.5-core ChangeSet@1.1243.23.20, 2003-06-03 17:16:47-07:00, hawkes@oss.sgi.com [PATCH] 2.5.70 remove smp_send_reschedule() cruft smp_send_reschedule_all() is unused in 2.5 and can be eliminated. ChangeSet@1.1243.23.19, 2003-06-03 17:10:32-07:00, B.Zolnierkiewicz@elka.pw.edu.pl [PATCH] kill recreate_proc_ide_device() It is unused and not needed ChangeSet@1.1243.23.18, 2003-06-03 17:10:24-07:00, B.Zolnierkiewicz@elka.pw.edu.pl [PATCH] create /proc/ide/hdX/capacity only once In ide_register_subdriver() create drive->proc entries only if driver is not idedefault_driver. [ There won't be /proc/ide/hdX/capacity for devices attached to idedefault_driver now, it reported 0x7fffffff previously. ] Do not create drive->proc entries in create_proc_ide_drives(), they are added later in ide_register_subdriver(). ChangeSet@1.1243.9.5, 2003-06-03 17:08:26-07:00, greg@kroah.com [PATCH] IBM PCI hotplug: remove hand made pci_find_bus function. ChangeSet@1.1243.9.4, 2003-06-03 17:08:13-07:00, greg@kroah.com [PATCH] ACPI PCI Hotplug: remove hand made pci_find_bus function ChangeSet@1.1243.9.3, 2003-06-03 16:41:35-07:00, greg@kroah.com [PATCH] PCI: make pools_lock and pci_lock static. No one is using them outside the pci core. ChangeSet@1.1243.9.2, 2003-06-03 16:41:07-07:00, greg@kroah.com [PATCH] PCI: make pci_setup_device(), pci_alloc_primary_bus() and pci_alloc_primary_bus_parented() static No one is calling these functions. ChangeSet@1.1243.31.1, 2003-06-03 16:29:24-07:00, ldm.adm@hostme.bitkeeper.com Merge hostme.bitkeeper.com:/ua/repos/l/ldm/linux-2.5 into hostme.bitkeeper.com:/ua/repos/l/ldm/linux-2.5-core ChangeSet@1.1243.1.24, 2003-06-03 16:22:11-07:00, mochel@osdl.org [driver model] fix comment in device.h ChangeSet@1.1243.1.23, 2003-06-03 16:19:54-07:00, mochel@osdl.org [driver model] Update copyrights and license statements. It's a slow afternoon. ChangeSet@1.1243.1.22, 2003-06-03 16:15:34-07:00, mochel@osdl.org [kobject] Update Documentation and licenses. - Make it very explicit that supplying an object desctructor is imperative if using the interface for object reference counting. - Add GPL statement to the source files. - Add GFL (http://www.fsf.org/licenses/fdl.html) statement to documentation. ChangeSet@1.1243.1.21, 2003-06-03 15:41:46-07:00, mochel@osdl.org [driver model] Clean up class release handling. Based on a patch by Stephen Hemminger. ChangeSet@1.1243.1.20, 2003-06-03 15:01:15-07:00, mochel@osdl.org Driver Class: don't call put_device() when we never called get_device() From Greg: This fixes a oops when unplugging pci network devices. ChangeSet@1.1243.30.3, 2003-06-03 23:27:23+02:00, sam@mars.ravnborg.org kbuild: CROSS_COMPILE and ARCH definitions Patch originally by Jesse Barnes Previously the user were required to supply CROSS_COMPILE and ARCH on the commandline to make, alternatively they patched the Makefile direct. The following patch allows the user to specify the value of these in a variable assigned during init or similar. ChangeSet@1.1243.1.19, 2003-06-03 14:24:00-07:00, mochel@osdl.org [driver model] Clean up CPU unregistration. From Dave Jones. ChangeSet@1.1243.30.2, 2003-06-03 23:15:07+02:00, sam@mars.ravnborg.org kbuild: Silence kbuild with make V=0 With make version 3.80 kbuild echo'ed the fixdep command executed each time a c file was compiled. This has been tracked down to a bug in version 3.80 of make. Avoiding newlines in canned command sequences avoid this problem. At the same time consolidated similar code in Makefile.build, and avoiding a few ifdef/endif pairs resulting in a more readable makefile. ChangeSet@1.1243.30.1, 2003-06-03 23:06:54+02:00, sam@mars.ravnborg.org kbuild: Updated make help Patches originally by Adrian Bunk and Rudmer van Dijk. Included "make V=0|1" and "make C=1" ChangeSet@1.1243.23.16, 2003-06-03 11:01:07-07:00, markh@osdl.org [PATCH] megaraid driver fix for 2.5.70 A recent change to the megaraid driver to fix some memset calls resulted in overflowing the arrays being cleared and causing a system panic. This patch fixes the problem by making sure that the arrays being cleared are dimensioned to the correct size. The patch has been tested on osdl's stp machines that have megaraid controllers. ChangeSet@1.1243.23.15, 2003-06-03 09:10:49-07:00, anton@samba.org [PATCH] fix matroxfb compile on ppc64] Here we really want CONFIG_ALL_PPC, since these headers are only valid for PPC Mac machines (barf: badly named config option). ChangeSet@1.1243.23.14, 2003-06-03 09:07:00-07:00, davidm@napali.hpl.hp.com [PATCH] fix TCP roundtrip time update code One of those very-hard-to-track-down, trivial-to-fix kind of problems: without this patch, TCP roundtrip time measurements will corrupt the routing cache's RTT estimates under heavy network load (the bug causes RTAX_RTT to go negative, but since its type is u32, you end up with a huge positive value...). From there on, later TCP connections quickly will go south. The typo was introduced 8 months ago in v1.29 of the file by the patch entitled "Cleanup DST metrics and abstrct MSS/PMTU further". ChangeSet@1.1243.29.4, 2003-06-03 08:56:02-07:00, hch@lst.de [PATCH] fix signal.h compilation on PPC signal.h uses spinlock_t now so it needs to include spinlock.h. Without this compilation failes on PPC. ChangeSet@1.1243.29.3, 2003-06-03 08:55:52-07:00, herbert@gondor.apana.org.au [PATCH] Fix ide-mod unload crash This patch fixes an unload crash when no PCI drivers are loaded. ChangeSet@1.1243.28.12, 2003-06-03 03:20:36-07:00, davem@nuts.ninka.net [NET]: Typo in iph5527.c driver changes. ChangeSet@1.1243.28.11, 2003-06-03 02:03:11-07:00, davem@nuts.ninka.net [NET]: Use INIT_LIST_HEAD in arch/um/drivers/net_kern.c ChangeSet@1.1243.28.10, 2003-06-03 00:56:30-07:00, davem@nuts.ninka.net [IPV6]: In ipv6_add_dev dont call __ipv6_regen_rndid without initial reference held. Based upon a report from Andrew Morton. ChangeSet@1.1243.23.12, 2003-06-03 17:34:51+10:00, anton@samba.org ppc64: fix compile error introduced in threaded coredump patch ChangeSet@1.1243.23.11, 2003-06-03 17:08:06+10:00, anton@samba.org ppc64: FORCE_MAX_ZONEORDER is actually order + 1, from Dave Gibson ChangeSet@1.1243.28.9, 2003-06-03 00:01:29-07:00, davem@nuts.ninka.net [NET]: Eliminate init_etherdev usage from arch/um drivers. ChangeSet@1.1243.23.10, 2003-06-03 16:55:19+10:00, anton@samba.org ppc64: Increase maximum allocation size to 16MB for largepage support ChangeSet@1.1243.23.9, 2003-06-03 16:08:05+10:00, anton@samba.org ppc64: kcore support ChangeSet@1.1243.28.8, 2003-06-02 22:45:08-07:00, davem@nuts.ninka.net [NET]: Eliminate {init,register,unregister}_trdev(). ChangeSet@1.1243.23.8, 2003-06-03 15:10:38+10:00, anton@samba.org ppc64: copy_tofrom_user exception handling fix from Paul Mackerras ChangeSet@1.1243.28.7, 2003-06-02 21:35:01-07:00, davem@nuts.ninka.net [NET]: Fix typos in init_trdev changes to lanstreamer.c ChangeSet@1.1243.28.6, 2003-06-02 21:15:03-07:00, davem@nuts.ninka.net [NET]: Fix CONFIG_HIPPI build. ChangeSet@1.1243.27.2, 2003-06-03 13:54:34+10:00, anton@samba.org ppc64: ppc64 update ChangeSet@1.1243.28.5, 2003-06-02 20:50:46-07:00, viro@parcelfarce.linux.theplanet.co.uk [NET]: Eliminate init_fddidev. ChangeSet@1.1243.28.4, 2003-06-02 20:34:03-07:00, davem@nuts.ninka.net [NET]: Kill PTI fddi driver, never completed and no plans to finish. ChangeSet@1.1243.28.3, 2003-06-02 20:31:36-07:00, viro@parcelfarce.linux.theplanet.co.uk [NET]: Convert most tokenring drivers away from {init,register,unregister}_trdev, only ibmtr remains. ChangeSet@1.1243.28.2, 2003-06-02 20:21:06-07:00, viro@parcelfarce.linux.theplanet.co.uk [NET]: Eliminate init_hippi_dev and {un,}register_hipdev. ChangeSet@1.1243.28.1, 2003-06-02 20:17:43-07:00, viro@parcelfarce.linux.theplanet.co.uk [NET]: Eliminate {init,register,unregister}_fcdev. ChangeSet@1.1243.26.26, 2003-06-02 18:12:16-07:00, jim.houston@attbi.com [PATCH] preallocate signal queue resource - Posix timers This adds a new interface to kernel/signal.c which allows signals to be sent using preallocated sigqueue structures. It also modifies kernel/posix-timers.c to use this interface. The current timer code may fail to deliver a timer expiry signal if there are no sigqueue structures available at the time of the expiry. The Posix specification is clear that the signal queuing resource should be allocated at timer_create time. This allows the error to be returned to the application rather than silently losing the signal. This patch does not change the sigqueue structure allocation policy. I hope to revisit that in another patch. Here is the definition for the new interface: struct sigqueue *sigqueue_alloc(void) Preallocate a sigqueue structure for use with the functions described below. void sigqueue_free(struct sigqueue *q) Free a preallocated sigqueue structure. If the sigqueue structure being freed is still queued, it will be removed from the queue. I currently leave the signal pending. It may be delivered without the siginfo structure. int send_sigqueue(int sig, struct sigqueue *q, struct task_struct *p) This function is equivalent to send_sig_info(). It queues a signal to the specified thread using the supplied sigqueue structure. The caller is expected to fill in the siginfo_t which is part of the sigqueue structure. int send_group_sigqueue(int sig, struct sigqueue *q, struct task_struct *p) This function is equivalent to send_group_sig_info(). It queues the signal to a process allowing the system to select which thread will receive the signal in a multi-threaded process. Again, the sigqueue structure is used to queue the signal. Both send_sigqueue() and send_group_sigqueue() return 0 if the signal is queued. They return 1 if the signal was not queued because the process is ignoring the signal. Both versions include code to increment the si_overrun count if the sigqueue entry is for a Posix timer and they are called while the sigqueue entry is still queued. Yes, I know that the current code doesn't rearm the timer until the signal is delivered. Having this extra bit of code doesn't do any harm, and I plan to use it. These routines do not check if there already is a legacy (non-realtime) signal pending. They always queue the signal. This requires that collect_signal() always checks if there is another matching siginfo before clearing the signal bit. ChangeSet@1.1243.26.25, 2003-06-02 17:40:15-07:00, akpm@digeo.com [PATCH] remove 16-bit pid assumption from ipc/sem.c From: Manfred Spraul SysV sem operations that involve multiple semaphores can fail in the middle, and then sempid (pid of the last successful operation) must be restored. This happens with "sempid >>= 16" - broken due to the 32-bit pid values. The attached patch fixes that by reordering the updates of the semaphore fields. Additionally, the patch fixes the corruption of the sempid value that occurs if a wait-for-zero operation fails. The patch is more than two years old, and was in -dj and -ak kernels. ChangeSet@1.1243.26.24, 2003-06-02 17:40:07-07:00, akpm@digeo.com [PATCH] /proc/kcore fixes From: Tony Luck /proc/kcore has been broken on some architectures for a long time. Problems surround the fact that some architectures allocate memory for vmalloc() and thus modules at addresses below PAGE_OFFSET, which results in negative file offsets in the virtual core file image provided by /proc/kcore. There are also pending problems for discontig memory systems as /proc/kcore just pretends that there are no holes between "PAGE_OFFSET" and "high_memory", so an unwary user (ok super-user) can read non-existant memory which may do bad things. There may also be kernel objects that would be nice to view in /proc/kcore, but do not show up there. A pending change on ia64 to allow booting on machines that don't have physical memory in any convenient pre-determined place will make things even worse, because the kernel itself won't show up in the current implementation of /proc/kcore! The patch attached provides enough hooks that each architecture should be able to make /proc/kcore useful. The patch is INCOMPLETE in that *use* of those hooks is ONLY PROVIDED FOR IA64. Here's how it works. The default code in fs/proc/kcore.c doesn't set up any "elf_phdr" sections ... it is left to each architecture to make appropriate calls to "kclist_add()" to specify a base address and size for each piece of kernel virtual address space that needs to be made accessible through /proc/kcore. To get the old functionality, you'll need two calls that look something like: kclist_add(&kcore_mem, __va(0), max_low_pfn * PAGE_SIZE); kclist_add(&kcore_vmem, (void *)VMALLOC_START, VMALLOC_END-VMALLOC_START); The first makes all of memory visible (__i386__, __mc68000__ and __x86_64__ should use __va(PAGE_SIZE) to duplicate the original lack of access to page 0). The second provides a single map for all "vmalloc" space (the code still searches the vmlist to see what actually exists before accessing it). Other blocks of kernel virtual space can be added as needed, and removed again (with kclist_del()). E.g. discontiguous memory machines can add an entry for each block of memory. Architectures that allocate memory for modules someplace outside of vmalloc-land can add/remove entries on module insert and remove. The second piece of abstraction is the kc_vaddr_to_offset() and kc_offset_to_vaddr() macros. These provide mappings from kernel virtual addresses to offsets in the virtual file that /proc/kcore instantiates. I hope they are sufficient to avoid negative offset problems that plagued the old /proc/kcore. Default versions are provided for the old behaviour (mapping simply adds/subtracts PAGE_OFFSET). For ia64 I just need to use a different offset as all kernel virtual allocations are in the high 37.5% of the 64-bit virtual address space. x86_64 was the other architecture with this problem. I don't know enough (anything) about the kernel memory map on x86_64, so I hope these provide a big enough hook. I'm hoping that you have some low stuff, and some high stuff with a big hole in the middle ... in which case the macros might look something like: #define kc_vaddr_to_offset(v) ((v) < 0x1000000000000000 ? (v) : \ ((v) - 0xF000000000000000)) But if you have interesting stuff scattered across *every* part of the unsigned address range, then you won't be able to squeeze it all into a signed file offset. There are a couple of bug fixes too: 1) get_kcore_size() didn't account for the elf_prstatus, elf_prpsinfo and task_struct that are placed in the PT_NOTE section that is part of the header. We were saved on most configurations by the round-up to PAGE_SIZE ... but it's possible that some architectures or machines corrupted memory beyond the space allocated for the header. 2) The size of the PT_NOTES section was incorrectly set to the size of the last note, rather than the sum of the sizes of all the notes. ChangeSet@1.1243.26.23, 2003-06-02 17:39:59-07:00, akpm@digeo.com [PATCH] ext3: fix for blocksize < PAGE_CACHE_SIZE If a buffer_head is outside i_size, block_write_full_page() will leave it !buffer_mapped(). We shouldn't attach that buffer to the transaction for writeout! This bug has been in 2.5 for some time. ChangeSet@1.1243.26.22, 2003-06-02 17:39:51-07:00, akpm@digeo.com [PATCH] improved core support for time-interpolation From: David Mosberger Basically, what the patch does is provide two hooks such that platforms (and subplatforms) can provide time-interpolation in a way that guarantees that two causally related gettimeofday() calls will never see time going backwards (unless there is a settimeofday() call, of course). There is some evidence that the current scheme does work: we use it on ia64 both for cycle-counter-based interpolation and the SGI folks use it with a chipset-based high-performance counter. It seems like enough platforms do this sort of thing to provide _some_ support in the core, especially because it's rather tricky to guarantee that time never goes backwards (short of a settimeofday, of course). This patch is based on something Jes Sorensen wrote for the SGI Itanium 2 platform (which has a chipset-internal high-res clock). I adapted it so it can be used for cycle-counter interpolation also. The net effect is that "last_time_offset" can be removed completely from the kernel. The basic idea behind the patch is simply: every time you advance xtime by N nanoseconds, you call update_wall_time_hook(NSEC). Every time the time gets set (i.e., discontinuity is OK), reset_wall_time_hook() is called. ChangeSet@1.1243.26.21, 2003-06-02 17:39:42-07:00, akpm@digeo.com [PATCH] MTD build fix From: Adrian Bunk Fix this: drivers/mtd/maps/map_funcs.c: In function `simple_map_read64': drivers/mtd/maps/map_funcs.c:38: warning: implicit declaration of function `__raw_readll' ChangeSet@1.1243.26.20, 2003-06-02 17:39:34-07:00, akpm@digeo.com [PATCH] Some subarch warning fixes From: Andi Kleen Some recent subarch interface changes caused macro redefinition warnings for GET_APIC_ID and APIC_ID_MASK with the generic subarchitecture. Fixing it properly required some reorganization by giving the generic arch a mach_apicdef.h too. ChangeSet@1.1243.26.19, 2003-06-02 17:39:25-07:00, akpm@digeo.com [PATCH] Copy nanosecond stat values for i386 From: Andi Kleen A brown paper bag bug, noticed by Ralf Baechle. i386 needs to define STAT_HAVE_NSEC too, otherwise it won't copy the nanosecond values to user space. ChangeSet@1.1243.26.18, 2003-06-02 17:39:16-07:00, akpm@digeo.com [PATCH] dirty_writeback_centisecs fixes This /proc tunable sets the kupdate interval. It has a couple of problems: - No way to turn it off completely (userspace dirty memory management solutions require this). - If it has been set to one hour and then the user resets it to five seconds, that resetting will not take effect for up to an hour. Fix that up by providing a sysctl handler. Setting the tunable to zero now disables the kupdate function. ChangeSet@1.1243.26.17, 2003-06-02 17:39:07-07:00, akpm@digeo.com [PATCH] More irq balance fixes From: Andi Kleen John Stultz noticed that kirqd didn't start because of another logic error, which broke irq balancing. This was still a fallout from the generic subarchitecture changes. Actually it still refuses to balance anything on my test box, but perhaps I'm just not able to generate enough interrupts. Anyways, with this patch the thread is running again at least. ChangeSet@1.1243.26.16, 2003-06-02 17:38:59-07:00, akpm@digeo.com [PATCH] initialise vma->vm_next in do_mmap_pgoff() We end up inspecting this field in zap_page_range(), on the mmap error path. Best initialise it to something... ChangeSet@1.1243.26.15, 2003-06-02 17:38:50-07:00, akpm@digeo.com [PATCH] Additional fields in slabinfo From: Manfred Spraul We need to present more information in /proc/slabinfo now the magazine layer has been added. The slabinfo version has been updated to 2.0. ChangeSet@1.1243.26.14, 2003-06-02 17:38:42-07:00, akpm@digeo.com [PATCH] magazine layer for slab From: Manfred Spraul slab.c is not very efficient for passing objects between cpus. Usually this is a rare event, but with network routing and cpu-affine NICs it is possible that nearly all allocation operations will occur on one cpu, and nearly all free operations on another cpu. This causes slab memory to be returned to slab's free page list rather than being cached on behalf of the particular slab cache. The attached patch solves that by adding an array of objects that is shared by all cpus. Instead of multiple linked list operations per object, object pointers are now passed to/from the shared array (and thus between cpus) with memcopy operations. On uniprocessor, the default array size is 0, because the shared array and the per-cpu head array are redundant. Additionally, the patch exports more statistics in /proc/slabinfo and make the array sizes tunable by writing to /proc/slabinfo. Both changes break backward compatibility, user space scripts must look at the slabinfo version and act accordingly. The size of the new shared array may be altered at runtime, by writing to /proc/slabinfo. The new parameters for writing to /proc/slabinfo are: #echo "cache-name limit batchcount shared" > /proc/slabinfo For example "size-4096 120 60 8" improves the slab efficiency for network routing, because the default values (24 12 8) are too small for the large series generated due to irq mitigation. Note that only root has write permissions to /proc/slabinfo. These changes provided an overall 12% speedup in Robert Olson's gigE packet-formwarding testing on 2-way. ChangeSet@1.1243.26.13, 2003-06-02 17:35:28-07:00, viro@parcelfarce.linux.theplanet.co.uk [PATCH] ->minor_shift removal disk->minor_shift is not used anymore. Remove it. ChangeSet@1.1243.26.12, 2003-06-02 17:26:45-07:00, zippel@linux-m68k.org [PATCH] update kconfig documentation Fix various typos and and information about the new kconfig features. ChangeSet@1.1243.26.11, 2003-06-02 17:26:19-07:00, zippel@linux-m68k.org [PATCH] create configuration in the destination directory This creates the configuration in the destination directory instead of the current directory. ChangeSet@1.1243.26.10, 2003-06-02 17:26:00-07:00, zippel@linux-m68k.org [PATCH] front end updates conf: better choice interface don't ask for unchangable symbols mconf: mark unchangable symbols with '---' update exit text (from Sam Ravnborg ) qconf: update debug output ChangeSet@1.1243.26.9, 2003-06-02 17:25:40-07:00, zippel@linux-m68k.org [PATCH] add more warnings Add a number of warnings to avoid misuse of the previously added features (most important check for recursive dependencies). ChangeSet@1.1243.26.8, 2003-06-02 17:25:20-07:00, zippel@linux-m68k.org [PATCH] support for 'range' The 'range' keyword allows to define a lower and upper limit for integer and hex symbols. ChangeSet@1.1243.26.7, 2003-06-02 17:25:03-07:00, zippel@linux-m68k.org [PATCH] reverse dependency support The 'select' keyword defines a lower limit for symbols and allows to select other symbols when a symbol is selected, e.g.: config JOLIET bool "Microsoft Joliet CDROM extensions" select NLS This means when JOLIET is selected, NLS will be selected as well. ChangeSet@1.1243.26.6, 2003-06-02 17:24:48-07:00, zippel@linux-m68k.org [PATCH] expression support "default" accepts now not only a single symbol but also an expression which can be assigned to boolean and tristate symbols. ChangeSet@1.1243.26.5, 2003-06-02 17:24:37-07:00, zippel@linux-m68k.org [PATCH] add new keywords to parser Add the following new keywords: def_tristate, def_bool, def_boolean, select, enable and range. Add support for def_tristate and def_bool, which combines default and bool/tristate into a single statement and a allows simpler definition of derived symbols. ChangeSet@1.1243.26.4, 2003-06-02 17:24:27-07:00, zippel@linux-m68k.org [PATCH] Change P_ROOTMENU into a MENU_ROOT flag This changes P_ROOTMENU into a MENU_ROOT flag and also fixes some qconf usability problems. Some gconf fixes by Romain Lievin . ChangeSet@1.1243.26.3, 2003-06-02 17:24:18-07:00, zippel@linux-m68k.org [PATCH] Remove old code and macros Remove old code and debugging macros which were used by the cml1->kconfig converter. ChangeSet@1.1243.26.2, 2003-06-02 16:42:25-07:00, bcollins@debian.org [PATCH] Update IEEE1394 (r946) ETH1394: Use 64bit EUI as the hardware address. ETH1394: Support broadcast packets. SBP2 : Fix max_payload for > S400 CORE : Fix iso.c compilation by including linux/sched.h directly. ChangeSet@1.1243.26.1, 2003-06-02 13:52:47-07:00, davidel@xmailserver.org [PATCH] epoll race fix The was a race triggered by heavy MT usage that could have caused processes in D state. Bad Davide, bad ... Also, the semaphore is now per-epoll-fd and not global. Plus some comment adjustment. ChangeSet@1.1243.24.2, 2003-06-03 04:37:43+10:00, anton@samba.org ppc64: remove sys32.S ChangeSet@1.1243.18.12, 2003-06-02 10:48:25-07:00, greg@kroah.com [PATCH] USB: added .owner to kobil_sct driver ChangeSet@1.1243.18.11, 2003-06-02 10:46:42-07:00, wahrenbruch@kobil.de [PATCH] USB: kobil_sct.c added support for KAAN SIM Reader as promised - here is the patch for 2.5.70: Added support for KAAN SIM in kobil_sct. ChangeSet@1.1243.18.10, 2003-06-02 10:38:23-07:00, greg@kroah.com [PATCH] Root plug: remove USB bus walking functions, now use usb_find_device(). Also fixed compiler warnings about the dbg() function. ChangeSet@1.1243.18.9, 2003-06-02 10:38:03-07:00, greg@kroah.com [PATCH] USB: add usb_find_device() function to USB core. ChangeSet@1.1243.18.8, 2003-06-02 09:39:52-07:00, bunk@fs.tum.de [PATCH] SECURITY_ROOTPLUG must depend on USB The following patch lets SECURITY_ROOTPLUG depend on USB (otherwise there are link errors since Root Plug Support needs usb_bus_list{,_lock}): ChangeSet@1.1243.25.1, 2003-06-02 09:36:09-07:00, mhoffman@lightlink.com [PATCH] I2C: fix oops w83781d during rmmod This fixes the oops during w83781d module removal by putting the subclient registration back in. While I was in there, I split w83781d_detect in half in an attempt to reduce the goto madness. So, the /sys tree looks like this, where 48 & 49 are the subclients. There are no entries (besides name & power) for the subclients. /sys/bus/i2c/ |-- devices | |-- 0-002d -> ../../../devices/pci0/00:02.1/i2c-0/0-002d | |-- 0-0048 -> ../../../devices/pci0/00:02.1/i2c-0/0-0048 | `-- 0-0049 -> ../../../devices/pci0/00:02.1/i2c-0/0-0049 `-- drivers |-- i2c_adapter `-- w83781d |-- 0-002d -> ../../../../devices/pci0/00:02.1/i2c-0/0-002d |-- 0-0048 -> ../../../../devices/pci0/00:02.1/i2c-0/0-0048 `-- 0-0049 -> ../../../../devices/pci0/00:02.1/i2c-0/0-0049 Also, I fixed a bug where this driver would request and release an ISA region, then poke around in that region, then finally request it again. This patch against 2.5.70 works for me vs. an SMBus adapter. It needs re-testing against an ISA adapter since my particular chip is SMBus only. ChangeSet@1.1243.13.10, 2003-06-02 10:33:52-05:00, davidvh@cox.net [PATCH] Fix compilation errors in ppa and imm modules This patch is a resend from LKML since it has yet to be applied in either bk7. Patrick suggested I send it here. This patch fixes compilation failures in ppa.c and imm.c that were introduced in 2.5.70-bk1. * Removes the 'int hostno' parameter from imm_proc_info(). Parameter isn't used and breaks the pointer matching for Scsi_Host. * Added the prototype for imm_proc_info() in imm.h * Modified line 280 of ppa.c to match concept on line 263 of imm.c. ChangeSet@1.1243.23.4, 2003-06-02 02:57:55-07:00, davem@nuts.ninka.net [SPARC64]: Update defconfig. ChangeSet@1.1243.23.3, 2003-06-02 02:32:15-07:00, davem@nuts.ninka.net [SOUND]: Fix sparc cs4231 driver build due to pcm list changes. ChangeSet@1.1243.23.2, 2003-06-02 02:27:42-07:00, davem@nuts.ninka.net [SOUND]: vx/vx_core.c needs linux/init.h ChangeSet@1.1243.23.1, 2003-06-02 01:58:33-07:00, davem@nuts.ninka.net [SOUND]: Revert buggy ALSA merge ioctl32 changes. ChangeSet@1.1243.18.7, 2003-06-01 23:06:14-07:00, oliver@neukum.org [PATCH] USB: return errors when disabling a port this allows us to learn about a port that cannot be disabled. It's needed for a superrobust usb_reset_device(). ChangeSet@1.1243.18.6, 2003-06-01 23:00:22-07:00, mdharm-usb@one-eyed-alien.net [PATCH] USB: usb-storage: change result codes This patch changes to SAM_STAT_ result codes, which is (a) preferred, according to the code comments, and (b) removes some odd-looking bit-shifting. ChangeSet@1.1243.18.5, 2003-06-01 23:00:01-07:00, mdharm-usb@one-eyed-alien.net [PATCH] USB: usb-storage: usb_stor_control_msg() and stuff This patch replaces usb_control_msg() with usb_stor_control_msg() everywhere, which allows better abort/disconnect processing. Some comments are fixed-up. The GetMaxLUN function is moved later so URBs are initialized (now that it uses the new control_msg() ). There is also some locking cleanup during reset. ChangeSet@1.1243.18.4, 2003-06-01 22:59:36-07:00, mdharm-usb@one-eyed-alien.net [PATCH] USB: usb-storage: timeouts and aborts This patch adds timeouts to usb_stor_control_msg(). Now we will no longer have to use the usb_control_msg() routine in the usb core, so all our control messages can be interrupted by scsi aborts or disconnects. This also includes the new serial-number for auto-REQUEST-SENSE change. The new serial number has one bit toggled from the old, guaranteeing that it is unique. Following a suggestion of David Brownell, this makes the transport-reset function attempt to clear a halt condition on both bulk pipes even if one of them fails. ChangeSet@1.1243.18.3, 2003-06-01 22:59:08-07:00, mdharm-usb@one-eyed-alien.net [PATCH] USB: usb-storage: fix typo Typo fix. We need bitwise-OR here. ChangeSet@1.1243.18.2, 2003-06-01 22:38:42-07:00, greg@kroah.com Cset exclude: greg@kroah.com|ChangeSet|20030529215347|05329 ChangeSet@1.1243.1.16, 2003-06-01 19:11:48-07:00, viro@parcelfarce.linux.theplanet.co.uk [PATCH] Fix disk partitioning with multiple IDE disks Now that we use idedefault_driver.drives instead of ata_unused, the order of drives on the driver->drives becomes significant; note that when we added to ata_unused, we had done that to the end of list. So we should do the same with the "drives" list. ChangeSet@1.1243.1.14, 2003-06-01 20:49:32+02:00, perex@suse.cz ALSA update 0.9.4 - added drivers for Digigram VXPocket V2, VXPocket 440 (pcmcia) - added driver for Digigram VXP220 V2/Mic (PCI) - added driver Harmony chipset (parisc) - added OPL4 driver - code cleanups - removed 2.2 and 2.4 code - nm256 driver - added Dell Latitude LS workaround - ac97 driver - fixes in intialization - usb audio driver - strlcat() fixes ChangeSet@1.1243.20.7, 2003-06-01 01:11:53-07:00, davem@nuts.ninka.net [XFRM]: Handle use_time expiration properly. ChangeSet@1.1243.22.5, 2003-05-31 23:41:30-07:00, herbert@gondor.apana.org.au [XFRM]: Too many reference drops of delpol in xfrm_policy_insert. ChangeSet@1.1243.22.4, 2003-05-31 23:29:01-07:00, kaber@trash.net [PPP] fix memory leak in ioctl error path ChangeSet@1.1243.22.3, 2003-05-31 23:20:58-07:00, herbert@gondor.apana.org.au [XFRM_USER]: Rename confusing member of struct xfrm_usersa_id. ChangeSet@1.1243.22.2, 2003-05-31 21:20:59-07:00, herbert@gondor.apana.org.au [XFRM_USER]: Fix xfrm_state_lookup args in xfrm_add_sa. ChangeSet@1.1243.6.76, 2003-06-01 10:59:16+10:00, anton@samba.org ppc64: Add warning for unhandled irqs ChangeSet@1.1243.6.75, 2003-06-01 10:34:26+10:00, anton@samba.org ppc64: fix copy_from_user leak, from Milton Miller ChangeSet@1.1243.6.74, 2003-06-01 10:29:39+10:00, anton@samba.org ppc64: threaded coredump support ChangeSet@1.1243.20.5, 2003-05-31 18:44:38-03:00, acme@conectiva.com.br o net: reduce the data dependency of struct sock/struct tcp_tw_bucket With this the area that is shared among struct sock and struct tcp_tw_bucket is exact, i.e. there is no fields that are in struct sock and only reused for other reasons in tcp_tw_bucket, while keeping both struct free of holes. ChangeSet@1.1243.20.4, 2003-05-31 11:36:54-07:00, manfred@colorfullife.com [PATCH] Prepare for page unmapper Avoid touching the next page in the NMI handler stack validity check. That way, we can start unmapping pages that aren't in use, without triggering this case. ChangeSet@1.1243.20.3, 2003-05-31 11:04:12-07:00, elenstev@mesatop.com [PATCH] More ANSI C cleanup of zlib More zlib K&R to ANSI C function header conversions. ChangeSet@1.1243.20.2, 2003-05-31 10:41:15-07:00, vandrove@vc.cvut.cz [PATCH] matroxfb update to new API This updates the matroxfb driver to the new framebuffer API. I'm sorry that it is quite large, but due to completely changed underlying API there is no reasonable way how to split it into smaller pieces. - Removed support for text mode. No way for it with current API. - Removed support for hardware cursor. Generic cursor code has enough troubles as is, in software mode. - No reasonable fbset support... It is especially annoying on multihead system, as 'stty cols XXX rows YYY' does not change pixclock... - Removed fastfont support. No way for it with current API. (Mis)features inherited from generic fbdev API: - Cursor on other framebuffers than primary one does not blink. - Contents of visible, but not foreground, display is not updated. ChangeSet@1.1243.1.13, 2003-05-31 14:37:47+02:00, perex@suse.cz Merge suse.cz:/home/perex/bk/linux-sound/linux-2.5 into suse.cz:/home/perex/bk/linux-sound/linux-sound ChangeSet@1.1243.21.1, 2003-05-31 02:44:59-07:00, herbert@gondor.apana.org.au [XFRM]: u64 --> __u64 in linux/xfrm.h ChangeSet@1.1243.20.1, 2003-05-31 00:54:45-07:00, davem@nuts.ninka.net [SPARC64]: Fix sys_shmat handling for 64-bit binaries. ChangeSet@1.1243.19.1, 2003-05-30 23:55:21-03:00, acme@conectiva.com.br o n_hdlc: CodingStyle cleanups and removal of old stuff - remove include kerneld.h, long gone - some 80 column reformatting - removal of typedefs - c99 style documentation, using the existing function documentation - add MODULE_AUTHOR - remove GET_USER, COPY_TO_USER and friends - remove ssize_t typedef, already in the kernel headers - remove rw_count_t and rw_ret_t, use simple int and size_t as in the tty ops prototypes and avoiding casts for return - make strings in printk in init/exit module routines __initdata/__exitdata saving some bytes after module init and for the case where this driver is statically linked in the kernel (__exitdata is trown away) This makes this driver comply with parts of CodingStyle (more to be done) and makes it looks more like the rest of the kernel, making it easier to read/debug. Have been using this modifications with ADSL/syncppp for some time, working well. ChangeSet@1.1243.13.9, 2003-05-30 18:01:54-05:00, jejb@raven.il.steeleye.com Merge raven.il.steeleye.com:/home/jejb/BK/scsi-sg-2.5 into raven.il.steeleye.com:/home/jejb/BK/scsi-misc-2.5 ChangeSet@1.1243.17.1, 2003-05-30 17:41:42-05:00, dougg@torque.net [PATCH] sg driver for lk 2.5.70 This is a bug fix for a flawed patch I put in lk 2.4.19 (sg version 3.1.24) and has been carried through into the 2.5 series. Thanks to Andy Polyakov for pointing this out (more than once). The sg.h header didn't get updated last time I submitted an sg patch so this patch also addresses that. ChangeSet@1.1229.26.1, 2003-05-30 15:28:38-07:00, mochel@osdl.org Merge bk://ldm.bkbits.net/linux-2.5-core into osdl.org:/home/mochel/src/kernel/devel/linux-2.5-core ChangeSet@1.1243.16.1, 2003-05-30 12:58:33-07:00, bellucda@tiscali.it [PATCH] USB: replaced BKL in rio500.c ChangeSet@1.1243.10.15, 2003-05-30 11:56:54-07:00, paulkf@microgate.com [PATCH] tty_register_driver This patch reinstates the ability of tty devices to use dynamically allocated major numbers yet specify the minor numbers statically. The synclink drivers do this. ChangeSet@1.1243.6.71, 2003-05-30 11:54:31-07:00, henning@meier-geinitz.de [PATCH] USB: new vendor/product ids for scanner driver This patch adds some new vendor/product ids for the USB scanner driver. ChangeSet@1.1243.10.14, 2003-05-30 11:04:40-07:00, akpm@digeo.com [PATCH] fix generic_file_write() The recent writev() fix broke the invariant that ->commit_write _must_ be called after a successful ->prepare_write(). It leaves ext3 with a transaction stuck open and the filesystem locks up. ChangeSet@1.1243.14.7, 2003-05-30 11:01:54-07:00, torvalds@home.transmeta.com Heh. Jens clashes with himself. ChangeSet@1.1243.15.1, 2003-05-30 10:59:43-07:00, axboe@suse.de [PATCH] ide-cd/scsi/block fixups for SG_IO - Kill the bogus ret transformation in block/ioctl.c if we return -EINVAL, doesn't make any sense. - Don't allow sg_reserved_size to be set bigger than a request we can deal with... - timeout fixes. - Cleanup of user access. - Set SAM_STAT_CHECK_CONDITION, not CHECK_CONDITION which needs to be bit shifted 1 up. - Set sense_len correctly. - Account sense_len correctly, don't just increment by 1... - Use the correct pointer in post transform. - Fix oops in bio_map_user(), it must get the extra reference prior to calling bio_unmap_user() itself too. ChangeSet@1.1243.14.6, 2003-05-30 10:55:12-07:00, axboe@suse.de [PATCH] scsi_ioctl HZ fixes According to http://www.torque.net/sg/p/sg_v3_ho.html, SG HOWTO, SG_[GET|SET]_TIMEOUTs are measured in "jiffies," while timeout field of SG_IO structure - in milliseconds. Inconsistent? Yes. Yet it's no excuse to disregard the specification. "Jiffies" are USER_HZ, 10ms on IA-32 platforms and has to be scaled to kernel "jiffies," as suggested below. As for "(jiffies - start_time) * (1000 / HZ)" vs. "((jiffies - start_time) * 1000) / HZ." Just think that HZ is 1024 on some platforms... ChangeSet@1.1243.14.5, 2003-05-30 10:55:02-07:00, axboe@suse.de [PATCH] ide-cd buglets Assorted small ide-cd fixes, found and fixed by Andy Polyakov . - CHECK_CONDITION really wants to be SAM_STAT_CHECK_CONDITION, the damn bit shift by one bit again - Set sense_len correctly - Do post_transform() on the right buffer. ChangeSet@1.1243.14.4, 2003-05-30 10:54:52-07:00, axboe@suse.de [PATCH] copy the tag_map From: Milton Miller ChangeSet@1.1243.13.6, 2003-05-30 12:53:11-05:00, hch@lst.de [PATCH] kill remaining direct uses of scsi_register_host Together with a patch for the usb imaging drivers I just sent to Greg this will allow us to get rid of scsi_register_host as exported API, leaving it only for use of scsi_module.c. ChangeSet@1.1243.14.3, 2003-05-30 10:52:47-07:00, B.Zolnierkiewicz@elka.pw.edu.pl [PATCH] fix two IDE list_head problems Fix two problems related to list_head's (there are more, wip). Second bug was uncovered by wli's list_head debugging patch, thanks wli! - Remove ata_unused list and use &idedefault_driver->drives only, fixes list corruption (ata_unused will be later ressurected for hotplug). - Do not add same device twice to &idedefault_driver->drives, triggered by first calling ide_unregister_subdriver() and later ata_attach(). ChangeSet@1.1243.14.2, 2003-05-30 10:52:38-07:00, B.Zolnierkiewicz@elka.pw.edu.pl [PATCH] kill "hdX=noremap" Since Andries killed ide-geometry, remove "hdX=noremap" parameter as it is no longer needed. ChangeSet@1.1243.14.1, 2003-05-30 10:52:29-07:00, B.Zolnierkiewicz@elka.pw.edu.pl [PATCH] allow "hdX=scsi" for modular scsi/ide-scsi Allow a user to mark a device as for scsi emulation at boot even with modular scsi/ide-scsi. (from 2.4 patch by Matan Ziv-Av) ChangeSet@1.1243.13.5, 2003-05-30 12:48:39-05:00, hch@lst.de [PATCH] kill scsi_host_get_next The only reamining user was a horrible hack with a big XXX: in scsi_pc98.c. Remove it and add an even bigger XXX there, we don't want to keep this function (and the scsi hostlist) around just for this. ChangeSet@1.1243.13.4, 2003-05-30 12:47:41-05:00, hch@lst.de [PATCH] switch /proc/scsi/scsi to seq_file and proper device model ChangeSet@1.1243.13.3, 2003-05-30 12:46:55-05:00, hch@lst.de [PATCH] LDM-based scsi host lookup Replace scsi_host_hn_get with scsi_host_lookup that walks through the shost class and gets a proper reference to the host. This should fix the OOPS that Douglas saw last week. ChangeSet@1.1243.6.69, 2003-05-30 10:46:04-07:00, greg@kroah.com Merge kroah.com:/home/greg/linux/BK/bleed-2.5 into kroah.com:/home/greg/linux/BK/gregkh-2.5 ChangeSet@1.1243.13.2, 2003-05-30 12:45:58-05:00, jejb@raven.il.steeleye.com Fix __exit routine of NCR_D700 ChangeSet@1.1243.13.1, 2003-05-30 12:29:56-05:00, hch@lst.de [PATCH] driver model for scsi upper drivers, take 2 On Tue, May 27, 2003 at 09:32:18AM +0200, Christoph Hellwig wrote: > - removed the tape sysfs pseudodevice crap that caused hangs > - switched sg to a class_interface. This means sg can be used > on devices already claimed be an upper driver again. This > also means I had to remove the sg sysfs attributes temporarily > because the old mechanism is gone, but I'll restore them > differently in a followon patch. Yikes, this was the old patch again. Here's the right one: ChangeSet@1.1243.6.68, 2003-05-30 10:25:23-07:00, proski@gnu.org [PATCH] USB: name uninitialized in scanner.c Linux 2.5.69-bk18 prints something strange to the kernel log when the USB scanner is attached. It turns out drivers/usb/image/scanner.c uses uninitialized variable "name" in probe_scanner() in the printk() call. That means that random memory is read and output to the kernel log. ChangeSet@1.1243.6.67, 2003-05-30 10:25:11-07:00, stern@rowland.harvard.edu [PATCH] USB: fix address assignment after device reset Until my ambitious project gets going, this patch at least fixes the problem of assigning a device's new address following a device reset. The only change needed to David's original suggestion was to handle the pathway involved in registering root hubs. ChangeSet@1.1243.6.66, 2003-05-30 09:37:22-07:00, greg@kroah.com [PATCH] USB: remove some old references to /proc/bus/usb/drivers This is needed, as the file was deleted over a year ago... ChangeSet@1.1243.12.19, 2003-05-29 23:50:36-07:00, torvalds@home.transmeta.com Merge bk://kernel.bkbits.net/davem/net-2.5 into home.transmeta.com:/home/torvalds/v2.5/linux ChangeSet@1.1243.3.16, 2003-05-29 23:31:57-07:00, akpm@digeo.com [NET]: Convert rtnl_lock/register_netdevice/rtnl_unlock to register_netdev. ChangeSet@1.1243.12.18, 2003-05-29 23:23:26-07:00, akpm@digeo.com [PATCH] Fixes trivial error in From: Herbert Xu This patch adds a pair of missing quotes. ChangeSet@1.1243.12.17, 2003-05-29 23:23:15-07:00, akpm@digeo.com [PATCH] Fix writev when a segment generates EFAULT From: Martin Schwidefsky If writev() is passed a vector in which the second or later segment generates a fault it will currently return -EFAULT. It shouldn't. It should write what it can and return the number of bytes which were successfully copied. Fix that up by writing the partial result and then returning the right value. ChangeSet@1.1243.12.16, 2003-05-29 23:23:06-07:00, akpm@digeo.com [PATCH] svcsock use-after-free fix From: Neil Brown Extract ->stamp from skb *before* freeing it in svcsock.c As we sometime copy and free an skb, and sometime us it in-place, we must be careful to extract information from it *before* it might be freed, not after. Manfred's page-unmapping debug patch found this one. ChangeSet@1.1243.12.15, 2003-05-29 23:22:56-07:00, akpm@digeo.com [PATCH] support 64 bit pci_alloc_consistent From: Jes Sorensen This is patch which provides support for 64 bit address allocations from pci_alloc_consistent(), based on the address mask set through pci_set_consistent_dma_mask(). This is necessary on some platforms which are unable to provide physical memory in the lower 4GB block and do not provide IOMMU support for cards operating in certain bus modes, such as PCI-X on the SGI SN2. The default mask for pci_alloc_consistent() is still 32 bit as there are 64 bit capable hardware out there that doesn't support 64 bit addresses for descripters etc. Likewise, platforms which provide IOMMU support in all bus modes can ignore struct pci_dev->consistent_dma_mask and just return a 32 bit address as before. The patch also includes changes to tg3.c to make it use the new api as well as a documentation update. I have done my best on the documentation part, if anyone feel the can make my scribbles clearer, please do. Thanks to Dave Miller, Grant Grundler, James Bottomley, Colin Ngam, and Jeremy Higdon for input and code/documentation portions. ChangeSet@1.1243.12.14, 2003-05-29 23:22:46-07:00, akpm@digeo.com [PATCH] unregister_netdev cleanups Replace rtnl_lock(); register_netdevice(dev); rtnl_unlock(); with the equivalent register_netdev(); in numerous places. ChangeSet@1.1243.12.13, 2003-05-29 23:22:35-07:00, akpm@digeo.com [PATCH] Remove unneeded fcntl check The NR_OPEN check in F_DUPFD is unneeded. viro says: "We check the limits in locate_fd() (called by dupfd()). Check for NR_OPEN can (and should) be dropped - locate_fd() will never go beyond that (expand_fd() will check it and refuse to go). "IOW, simply lose the check. We _might_ want to check signedness, but that's it (IOW, check that arg will fit into 0..MAX_INT; second argument of dupfd() is an int). OTOH, we might actually make dupfd() et.al. take unsigned long and kill that crap completely." And indeed, the signedness is suspicious, so make various things in there unsigned too. ChangeSet@1.1243.12.12, 2003-05-29 23:22:26-07:00, akpm@digeo.com [PATCH] speed up the unlink speedup I'm not sure why I used igrab() in unlink(). igrab takes the oft-taken inode_lock. The caller has a ref, so a simple increment of i_count will suffice. ChangeSet@1.1243.12.11, 2003-05-29 23:22:16-07:00, akpm@digeo.com [PATCH] write_one_page() fixlets - set the number of pages to be written to "1". - Don't test PG_writeback twice. ChangeSet@1.1243.12.10, 2003-05-29 23:22:04-07:00, akpm@digeo.com [PATCH] i2o memleak comment From: Andy Whitcroft There's a spot in i2o where we deliberately leak some memory when the hardware plays up. The alternative is to let the hardware scribble on it at some unknown time in the future. Things like the Stanford checker keep alleging that this is a bug. So shut them up with a comment ChangeSet@1.1243.12.9, 2003-05-29 23:21:55-07:00, akpm@digeo.com [PATCH] kill lock_kernel() in inode_setattr() All we're doing in there is writing things into the inode. I see no need for the lock_kernel(). And holding lock_kernel() across mark_inode_dirty() hurts on big SMP. ChangeSet@1.1243.12.8, 2003-05-29 23:21:46-07:00, akpm@digeo.com [PATCH] irq balance logic fix From: Andi Kleen The logic is: the global variable is set to the magic value IRQBALANCE_CHECK_ARCH. It can be overwritten by a __setup function. If the magic value is still set when the irq balancer is started it asks the subarchitecture using the NO_BALANCE_IRQ macro. This is defined to a genapic field in the generic architecture, otherwise constant. Then the global variable is set and when it is true no balancing happens. Previously I had this wrong in that it always disabled it. This part should be correct, but it still doesn't seem to work. (I left the printk in there until the problem is debugged, could be removed of course) ChangeSet@1.1243.12.7, 2003-05-29 23:21:35-07:00, akpm@digeo.com [PATCH] zoran user-pointer fix From: Hollis Blanchard Fix a user pointer deref, found by the Stanford checker. ChangeSet@1.1243.12.6, 2003-05-29 23:21:26-07:00, akpm@digeo.com [PATCH] Fix CONFIG_PROCFS=n From: Manfred Spraul three build fixes for CONFIG_PROC_FS=n: include/linux/procfs.h: - add missing proc_pid_unhash, proc_pid_flush declarations. static inline functions that do nothing. - move semicolons around for kclist_{add,del}. gcc-3.2.2 doesn't like the current syntax. drivers/net/pppoe.c: - proc_net doesn't exist if CONFIG_PROC_FS=n. drivers/scsi/scsi_priv.h: - add missing brackets to macro definition. ChangeSet@1.1243.12.5, 2003-05-29 23:21:15-07:00, akpm@digeo.com [PATCH] [VISWS] irqreturn_t conversion From: Andrey Panin This small patch (against 2.5.70) updates visws_apic.c in accordance with linux irq handling changes. ChangeSet@1.1243.12.4, 2003-05-29 23:21:06-07:00, akpm@digeo.com [PATCH] export mmu_cr4_features to modules From: Jan Marek The DRM modules (i810) need this symbol. As this is a special-case for one particular in-kernel module I changed Jan's patch from EXPORT_SYMBOL to EXPORT_SYMBOL_GPL. ChangeSet@1.1243.12.3, 2003-05-29 23:20:55-07:00, akpm@digeo.com [PATCH] fix oops on resume from apm bios initiated suspend From: Milton Miller mm is NULL for kernel threads without their own context. active_mm is maintained the one we lazly switch from. Without this patch, apm bios initiated suspend events (eg panel close) cause an oops on resume in the LDT restore, killing kapmd, which causes further events to not be polled. ChangeSet@1.1243.12.2, 2003-05-29 23:20:46-07:00, akpm@digeo.com [PATCH] Fix suspend with pccardd running From: Pavel Machek This fixes suspend when pccards are used... ChangeSet@1.1243.12.1, 2003-05-29 23:20:35-07:00, akpm@digeo.com [PATCH] fix typo in coda We want an "|" in there, not "||". ChangeSet@1.1243.6.65, 2003-05-29 22:53:47-07:00, hwahl@hwahl.de [PATCH] USB: Patch for Samsung Digimax 410 *** a/drivers/usb/storage/unusual_devs.h 2003-05-22 20:54:26.000000000 +0200 ChangeSet@1.1243.6.64, 2003-05-29 22:44:25-07:00, david-b@pacbell.net [PATCH] USB: ethernet "gadget", rx overflows happen I tightened up rx overflow logic in the net2280 driver a while back, and it broke MTU size packets in this driver (host pads them out). This patch accomodates it by allocating a slightly larger buffer (almost 3*512). ChangeSet@1.1243.6.63, 2003-05-29 22:36:35-07:00, bcollins@debian.org [PATCH] USB Multi-input quirk ChangeSet@1.1243.10.9, 2003-05-29 22:28:31-07:00, gerg@snapgear.com [PATCH] m68knommu/pilot startup copy init segment to RAM The m68knommu/pilot startup code is not copying the init segment to RAM currently. Fix it to copy all of the data and init sections to RAM. ChangeSet@1.1243.10.8, 2003-05-29 22:28:21-07:00, gerg@snapgear.com [PATCH] support BOOTPARAM's on m68knommu/5206e targets Support pre-configured boot arguments on m68knommu/5206e targets. ChangeSet@1.1243.10.7, 2003-05-29 22:28:03-07:00, gerg@snapgear.com [PATCH] support BOOTPARAM's on m68knommu/5206 targets Support pre-configured boot arguments on m68knommu/5206 targets. ChangeSet@1.1243.10.6, 2003-05-29 22:27:54-07:00, gerg@snapgear.com [PATCH] conditional ROMfs copy for ARNEWSH/5206 setup Make the ROMfs copy in the startup code for ARNEWSH/5260 board conditional on actually using a ROMfs setup. ChangeSet@1.1243.6.62, 2003-05-29 21:51:44-07:00, greg@kroah.com Merge bk://kernel.bkbits.net/acme/usb-2.5 into kroah.com:/home/greg/linux/BK/gregkh-2.5 ChangeSet@1.1243.3.15, 2003-05-29 20:33:00-07:00, davem@nuts.ninka.net [ATM]: Fix driver Makefile clean-files. ChangeSet@1.1243.3.14, 2003-05-29 20:15:09-07:00, shemminger@osdl.org [NET]: Kill deprecated if_port_text and users. ChangeSet@1.1243.3.13, 2003-05-29 20:10:35-07:00, jmm@informatik.uni-bremen.de [CRYPTO]: Default CRYPTO and MD5 to y if IPV6_PRIVACY is enabled. ChangeSet@1.1243.3.12, 2003-05-29 20:04:32-07:00, shemminger@osdl.org [NET]: Sysfs netdev cleanup and bugfix. A couple of bugs in netdev_unregister_sysfs; still working on the harder refcount issues. - if driver sets get_stats after register then unregister will attempt to delete kobject that has not be initialized. - unregister should call kobject_unregister not kobject_del. cleanup's: - use strlcpy instead of snprintf - don't need to memset the stats kobject ChangeSet@1.1243.3.11, 2003-05-29 20:02:49-07:00, shemminger@osdl.org [BRIDGE]: Make delete bridge work with current unregister semantics. ChangeSet@1.1243.3.10, 2003-05-29 19:58:48-07:00, chas@cmf.nrl.navy.mil [ATM]: Move rategrid off stack in HE driver. ChangeSet@1.1243.3.9, 2003-05-29 19:57:25-07:00, chas@cmf.nrl.navy.mil [ATM]: HE driver misc irq handler cleanups. ChangeSet@1.1243.3.8, 2003-05-29 19:56:27-07:00, chas@cmf.nrl.navy.mil [ATM]: HE driver coding style conformance. ChangeSet@1.1243.10.5, 2003-05-29 21:07:23-04:00, scott.feldman@intel.com [netdrvr e100] move register_netdev below netdev struct init (i.e. the better fix) ChangeSet@1.1229.23.12, 2003-05-29 20:07:00-04:00, jgarzik@redhat.com Cset exclude: shemminger@osdl.org|ChangeSet|20030529205634|46794 The needed fix winds up breaking SG, checksumming, and other stuff in the process. ChangeSet@1.1243.10.3, 2003-05-29 16:02:10-07:00, elenstev@mesatop.com [PATCH] Yet more K&R to ANSI C conversions More K&R to ANSI C conversions for lib/zlib_deflate. ChangeSet@1.1243.10.2, 2003-05-29 16:02:01-07:00, elenstev@mesatop.com [PATCH] K&R to ANSI C conversions for zlib Here are some more K&R to ANSI C conversions. ChangeSet@1.1243.8.7, 2003-05-29 15:25:37-07:00, hch@lst.de [PATCH] fix scsi_register_host abuse in usb scanner drivers They should be using scsi_add_host directly. I had to rewrite half of the drivers, though to fix horrible braindamage like leaving dangling scsi structures around after ->disconnect. Gettig rid of the remaining scsi_register_host callers is required for the scsi stack to move forward so please try to forward this to Linus in a timely mannor, thanks! ChangeSet@1.1243.8.6, 2003-05-29 15:25:26-07:00, hch@lst.de [PATCH] use second arg to scsi_add_host in usb storage That way we don't need the addition scsi_set_device call. ChangeSet@1.1243.8.5, 2003-05-29 15:19:51-07:00, mdharm-usb@one-eyed-alien.net [PATCH] USB: storage: collapse one-use functions This patch collapses some one-use functions into their callers. It also clones some code for control transfers so we can implement abortable control transfers with timeout. This patch is from Alan Stern. Remove usb_stor_bulk_msg() and usb_stor_interrupt_msg(). Move their functionality into usb_stor_bulk_transfer_buf() and usb_stor_intr_transfer(). Move the functionality of usb_stor_control_msg() into usb_stor_ctrl_transfer(). Remove the unused act_len parameter from usb_stor_intr_transfer(). ChangeSet@1.1243.8.4, 2003-05-29 15:19:41-07:00, mdharm-usb@one-eyed-alien.net [PATCH] USB: storage: abort and disconnect handling. This patch re-organizes abort handling and enhances disconnect handling. Not only do we keep track of the state (ABORTING, IDLE, etc.), but during an abort we now introduce the idea of 'okay to send' or not. The idea is that we can now implement reset-after-abort properly. We also track if we're disconnecting, and use that data to determine if we can submit URBs or not. Which means we can now disconnect during an abort. This patch is from Alan Stern. ChangeSet@1.1243.8.3, 2003-05-29 15:17:04-07:00, oliver@neukum.org [PATCH] USB: allocate memory for reset earlier if we fail with -ENOMEM, we should do it before the device must be reparsed. ChangeSet@1.1243.8.2, 2003-05-29 15:03:07-07:00, greg@kroah.com USB: fix up unusual_devs.h merge mess Thanks to Per Winkvist for the info and patches to do this. ChangeSet@1.1243.7.2, 2003-05-30 08:01:28+10:00, anton@samba.org ppc64: remove ioperm ChangeSet@1.1243.6.61, 2003-05-29 19:00:55-03:00, acme@conectiva.com.br o drivers/usb/usb-skeleton: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.60, 2003-05-29 19:00:06-03:00, acme@conectiva.com.br o drivers/usb/storage/usb: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.59, 2003-05-29 18:57:15-03:00, acme@conectiva.com.br o drivers/usb/serial/whiteheat: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.58, 2003-05-29 18:56:36-03:00, acme@conectiva.com.br o drivers/usb/serial/visor: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.57, 2003-05-29 18:55:51-03:00, acme@conectiva.com.br o drivers/usb/serial/usb-serial: initialize struct usb_driver ->owner field ChangeSet@1.1243.8.1, 2003-05-29 14:53:47-07:00, greg@kroah.com USB: build gadget drivers if they are built into the kernel. Somehow this got merged away... ChangeSet@1.1243.6.56, 2003-05-29 18:53:34-03:00, acme@conectiva.com.br o drivers/usb/serial/safe_serial: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.55, 2003-05-29 18:46:05-03:00, acme@conectiva.com.br o drivers/usb/serial/pl2303: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.54, 2003-05-29 18:40:23-03:00, acme@conectiva.com.br o drivers/usb/serial/omninet: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.53, 2003-05-29 18:19:34-03:00, acme@conectiva.com.br o drivers/usb/serial/mct_u232: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.52, 2003-05-29 18:17:43-03:00, acme@conectiva.com.br o drivers/usb/serial/kl5kusb105: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.51, 2003-05-29 18:16:28-03:00, acme@conectiva.com.br o drivers/usb/serial/keyspan: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.50, 2003-05-29 18:13:31-03:00, acme@conectiva.com.br o drivers/usb/serial/ir-usb: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.49, 2003-05-29 18:12:51-03:00, acme@conectiva.com.br o drivers/usb/serial/ipaq: initialize struct usb_driver ->owner field ChangeSet@1.1243.4.2, 2003-05-30 07:12:22+10:00, anton@samba.org ppc64: Add some branch prediction ChangeSet@1.1243.6.48, 2003-05-29 18:11:57-03:00, acme@conectiva.com.br o drivers/usb/serial/io_ti: initialize struct usb_driver ->owner field ChangeSet@1.1229.23.11, 2003-05-29 16:56:34-04:00, shemminger@osdl.org [netdrvr e100] initialize callbacks before registering netdev Ouch. ChangeSet@1.1243.6.47, 2003-05-29 17:52:38-03:00, acme@conectiva.com.br o drivers/usb/serial/io_edgeport: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.46, 2003-05-29 17:51:43-03:00, acme@conectiva.com.br o drivers/usb/serial/ftdi_sio: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.45, 2003-05-29 17:50:41-03:00, acme@conectiva.com.br o drivers/usb/serial/empeg: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.44, 2003-05-29 17:48:26-03:00, acme@conectiva.com.br o drivers/usb/serial/digi_acceleport: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.43, 2003-05-29 17:46:57-03:00, acme@conectiva.com.br o drivers/usb/serial/cyberjack: initialize struct usb_driver ->owner field ChangeSet@1.1229.23.10, 2003-05-29 16:46:47-04:00, reeja.john@amd.com [netdrvr amd8111e] interrupt coalescing, libmii, bug fixes * Dynamic interrupt coalescing * mii lib support * dynamic IPG support (disabled by default) * jumbo frame fix * vlan fix * rx irq coalescing fix ChangeSet@1.1243.6.42, 2003-05-29 17:43:30-03:00, acme@conectiva.com.br o drivers/usb/serial/belkin_sa: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.41, 2003-05-29 17:42:31-03:00, acme@conectiva.com.br o drivers/usb/net/usbnet: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.40, 2003-05-29 17:41:41-03:00, acme@conectiva.com.br o drivers/usb/net/rtl8150: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.39, 2003-05-29 17:40:45-03:00, acme@conectiva.com.br o drivers/usb/net/pegasus: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.38, 2003-05-29 17:39:26-03:00, acme@conectiva.com.br o drivers/usb/net/cdc-ether: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.37, 2003-05-29 17:38:40-03:00, acme@conectiva.com.br o drivers/usb/net/catc: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.36, 2003-05-29 17:37:56-03:00, acme@conectiva.com.br o drivers/usb/misc/usblcd: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.35, 2003-05-29 17:37:11-03:00, acme@conectiva.com.br o drivers/usb/misc/rio500: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.34, 2003-05-29 17:36:06-03:00, acme@conectiva.com.br o drivers/usb/misc/emi26: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.33, 2003-05-29 17:35:16-03:00, acme@conectiva.com.br o drivers/usb/misc/auerswald: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.32, 2003-05-29 17:34:30-03:00, acme@conectiva.com.br o drivers/usb/media/vicam: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.31, 2003-05-29 17:32:16-03:00, acme@conectiva.com.br o drivers/usb/media/ultracam: remove MOD_{INC,DEC}_USE_COUNT ChangeSet@1.1243.6.30, 2003-05-29 17:30:36-03:00, acme@conectiva.com.br o drivers/usb/media/stv680: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.29, 2003-05-29 17:29:15-03:00, acme@conectiva.com.br o drivers/usb/media/se401: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.28, 2003-05-29 17:27:16-03:00, acme@conectiva.com.br o drivers/usb/media/pwc-if: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.27, 2003-05-29 17:24:16-03:00, acme@conectiva.com.br o drivers/usb/media/ov511: initialize struct usb_driver ->owner field Also remove MOD_{INC,DEC}_USE_COUNT ChangeSet@1.1243.6.26, 2003-05-29 17:23:09-03:00, acme@conectiva.com.br o drivers/usb/media/konicawc: remove MOD_{DEC,INC}_USE_COUNT ChangeSet@1.1243.6.25, 2003-05-29 17:21:34-03:00, acme@conectiva.com.br o drivers/usb/media/ibmcam: remove MOD_{INC,DEC}_USE_COUNT ChangeSet@1.1243.6.24, 2003-05-29 17:20:30-03:00, acme@conectiva.com.br o drivers/usb/media/dsbr100: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.23, 2003-05-29 17:19:34-03:00, acme@conectiva.com.br o drivers/usb/media/dabusb: initialize struct usb_driver ->owner field Also remove MOD_{INC,DEC}_USE_COUNT ChangeSet@1.1243.6.22, 2003-05-29 17:17:40-03:00, acme@conectiva.com.br o drivers/usb/input/xpad: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.21, 2003-05-29 17:16:45-03:00, acme@conectiva.com.br o drivers/usb/input/wacom: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.20, 2003-05-29 17:15:45-03:00, acme@conectiva.com.br o drivers/usb/input/usbmouse: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.19, 2003-05-29 17:14:51-03:00, acme@conectiva.com.br o drivers/usb/input/usbkbd: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.18, 2003-05-29 17:14:05-03:00, acme@conectiva.com.br o drivers/usb/input/powermate: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.17, 2003-05-29 17:13:18-03:00, acme@conectiva.com.br o drivers/usb/input/kbtab: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.16, 2003-05-29 17:11:54-03:00, acme@conectiva.com.br o drivers/usb/input/hiddev: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.15, 2003-05-29 17:10:51-03:00, acme@conectiva.com.br o drivers/usb/input/hid-core: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.14, 2003-05-29 17:09:40-03:00, acme@conectiva.com.br o drivers/usb/input/aiptek: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.13, 2003-05-29 17:08:24-03:00, acme@conectiva.com.br o drivers/usb/image/scanner: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.12, 2003-05-29 17:07:08-03:00, acme@conectiva.com.br o drivers/usb/image/microtek: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.11, 2003-05-29 17:06:19-03:00, acme@conectiva.com.br o drivers/usb/image/hpusbscsi: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.10, 2003-05-29 17:00:03-03:00, acme@conectiva.com.br o drivers/usb/core/hub: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.9, 2003-05-29 16:59:23-03:00, acme@conectiva.com.br o drivers/usb/core/devio: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.8, 2003-05-29 16:58:39-03:00, acme@conectiva.com.br o drivers/class/usb-midi: initialize struct usb_driver ->owner field And remove MOD_{INC,DEC}_USE_COUNT ChangeSet@1.1243.6.7, 2003-05-29 16:57:48-03:00, acme@conectiva.com.br o drivers/class/cdc-acm: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.6, 2003-05-29 16:56:04-03:00, acme@conectiva.com.br o drivers/class/bluetty: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.5, 2003-05-29 16:55:03-03:00, acme@conectiva.com.br o drivers/class/audio: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.4, 2003-05-29 16:52:59-03:00, acme@conectiva.com.br o drivers/net/irda/irda-usb: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.3, 2003-05-29 16:50:10-03:00, acme@conectiva.com.br o drivers/media/video/cpia_usb: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.2, 2003-05-29 16:48:53-03:00, acme@conectiva.com.br o drivers/isdn/hisax/st5481: initialize struct usb_driver ->owner field ChangeSet@1.1243.6.1, 2003-05-29 16:47:43-03:00, acme@conectiva.com.br o drivers/bluetooth/hci_usb: initialize struct usb_driver ->owner field ChangeSet@1.1229.23.9, 2003-05-29 12:35:03-04:00, alan@lxorguk.ukuu.org.uk [netdrvr tlan] fix 64-bit issues ChangeSet@1.1229.22.2, 2003-05-29 10:47:08-05:00, shaggy@shaggy.austin.ibm.com JFS: resize fixes Bmap control page was not always being updated. Superblock's s_size field was incorrectly set on big-endian hardware. ChangeSet@1.1243.1.11, 2003-05-29 08:34:26-07:00, gerg@snapgear.com [PATCH] security init call support in linker script Linker script updates for m68knommu architecture: - fix _ramend for DragonEngine2 board - add security init call support ChangeSet@1.1243.1.10, 2003-05-29 08:34:17-07:00, gerg@snapgear.com [PATCH] fix broken trace flag check in 68328 system call entry ChangeSet@1.1243.1.9, 2003-05-29 08:34:08-07:00, gerg@snapgear.com [PATCH] cleanup is_in_rom() checker Clean up the dodgy is_in_rom() code for m68knommu targets. Now that all the m68knommu sub-architectures (68x328, 68360 and ColdFire) have the same memory setup support we can make this code the same for all targets. ChangeSet@1.1243.1.8, 2003-05-29 08:33:59-07:00, gerg@snapgear.com [PATCH] remove obsolete BLKMEM driver reference ChangeSet@1.1243.1.7, 2003-05-29 08:33:49-07:00, gerg@snapgear.com [PATCH] fix calls to do_fork() Change the m68knommu specific calls to do_fork() to match its pid return vlue. ChangeSet@1.1243.3.7, 2003-05-29 04:08:04-07:00, davem@nuts.ninka.net [NET}: Fix typo in sock_set_flag changes. ChangeSet@1.1243.1.6, 2003-05-29 11:43:56+01:00, dwmw2@infradead.org Fix some accidental regressions which slipped in with the MTD merge. - Unrevert strncpy->strlcpy change in JEDEC chip driver - Fix partition handling in physmap map driver - Switch sa1100-flash map driver back to rmk's version. ChangeSet@1.1243.3.6, 2003-05-29 03:12:40-07:00, yoshfuji@linux-ipv6.org [IPV6]: Add ip6frag sysctls. ChangeSet@1.1243.3.5, 2003-05-29 03:10:47-07:00, jgarzik@redhat.com [ROSE]: Kill kfree of net_device->name. ChangeSet@1.1243.3.4, 2003-05-29 03:07:32-07:00, chas@cmf.nrl.navy.mil [ATM]: lane and mpoa module refcounting and locking cleanup. ChangeSet@1.1243.2.4, 2003-05-29 03:03:14-07:00, hch@lst.de [SPARC64]: Kill sys_aplib. ChangeSet@1.1243.3.3, 2003-05-29 02:57:26-07:00, hch@lst.de [NET]: Remove sdla from setup.c ChangeSet@1.1243.3.2, 2003-05-29 02:52:52-07:00, davem@nuts.ninka.net Merge bk://kernel.bkbits.net/acme/net-2.5 into nuts.ninka.net:/home/davem/src/BK/net-2.5 ChangeSet@1.1243.2.3, 2003-05-29 02:51:38-07:00, davem@nuts.ninka.net [SPARC64]: Do not export {un,}register_ioctl32_converstion twice. ChangeSet@1.1243.3.1, 2003-05-29 02:47:48-07:00, yoshfuji@linux-ipv6.org [IPV6]: Fix default router selection in some cases. ChangeSet@1.1229.20.2, 2003-05-29 16:34:00+10:00, anton@samba.org ppc64: Always pass non segment faults on the 0xc region up to do_page_fault ChangeSet@1.1243.2.2, 2003-05-28 23:01:34-07:00, davem@nuts.ninka.net [SPARC64]: Fix probe error handling in bbc_{envctrl,i2c}.c driver. ChangeSet@1.1243.2.1, 2003-05-28 22:53:32-07:00, davem@nuts.ninka.net [SPARC64]: Fix probe error handling in envctrl.c driver. ChangeSet@1.1229.23.8, 2003-05-29 01:07:33-04:00, jgarzik@redhat.com [netdrvr r8169] use alloc_etherdev, pci_disable_device ChangeSet@1.1243.1.4, 2003-05-28 21:21:29-07:00, anton@samba.org [PATCH] compat_wait4 fix sys_wait4 can return a pid and in this case we want to copy the struct rusage out to userspace. ChangeSet@1.1229.23.6, 2003-05-29 00:03:59-04:00, jgarzik@redhat.com [netdrvr 8139too] respond to "isn't this racy?" comment ChangeSet@1.1229.23.5, 2003-05-28 23:52:04-04:00, jgarzik@redhat.com [netdrvr] s/init_etherdev/alloc_etherdev/ in code comments, in 8139too and pci-skeleton drivers. ChangeSet@1.1243.1.3, 2003-05-28 20:16:14-07:00, david@gibson.dropbear.id.au [PATCH] Update orinoco driver to 0.13e This updates the orinoco driver, fixing many bugs and adding some minor features. It also adds a new module, orinoco_tmd for devices based on the TMD7168 PCI<->PCMCIA adaptor. ChangeSet@1.1229.23.4, 2003-05-28 23:08:58-04:00, jgarzik@redhat.com [netdrvr tlan] cleanup * use pci_{request,release}_regions for PCI devices * use alloc_etherdev (fixes race) * propagate error returns from pci_xxx function errors ChangeSet@1.1229.7.43, 2003-05-28 19:11:37-07:00, davem@nuts.ninka.net [TCP]: Do not access inet_sk() of a time-wait bucket. Bug discovered by Mandred Spraul. ChangeSet@1.1229.25.2, 2003-05-28 16:37:35-05:00, Matt_Domsch@dell.com dynids: free dynids on driver unload ChangeSet@1.1229.25.1, 2003-05-28 15:54:26-05:00, Matt_Domsch@dell.com dynids: use list_add_tail instead of list_add, such that later entries come later in the scanned list. ChangeSet@1.1229.23.3, 2003-05-28 09:48:17-07:00, torvalds@home.transmeta.com Merge master.kernel.org:/home/dwmw2/BK/mtd-forlinus-2.5 into home.transmeta.com:/home/torvalds/v2.5/linux ChangeSet@1.1229.23.2, 2003-05-28 09:41:03-07:00, torvalds@home.transmeta.com Make zlib_inflate look more like ANSI C code. Anybody who still thinks K&R makes sense should just be shot. ChangeSet@1.1229.24.2, 2003-05-28 16:56:12+01:00, dwmw2@infradead.org Final cleanups for MTD merge. ChangeSet@1.1229.24.1, 2003-05-28 16:03:33+01:00, dwmw2@infradead.org MTD and JFFS2 update. - JFFS2 bugfixes and performance improvements - Support for 64-bit flash arrangements - Optimise for linear mappings of flash, without out-of-line access functions - New map drivers - Updated NAND flash support, new board drivers - Support for DiskOnChip Millennium Plus and INFTL translation layer - Clean up all translation layers with a single blkdev helper library. - Fix races in MTD device registration/deregistration - Add support for new flash chips - Clean up partition parsing code More detailed comments in per-file changelogs. ChangeSet@1.1229.23.1, 2003-05-28 16:03:03+01:00, rmk@flint.arm.linux.org.uk Merge flint.arm.linux.org.uk:/usr/src/linux-bk-2.5/linux-2.5 into flint.arm.linux.org.uk:/usr/src/linux-bk-2.5/linux-2.5-pcmcia ChangeSet@1.1127.7.2, 2003-05-28 15:41:22+01:00, proski@org.rmk.(none) [PATCH] Fix crash when unloading yenta_socket in Linux 2.5.69 socket->base is unmapped in yenta_close(), which is called by cardbus_remove(). The value of socket->base is not changed to NULL, so it becomes invalid. Then cardbus_remove() calls class_device_unregister(), which calls pcmcia_unregister_socket(), which it turn tries to access memory space of the socket. ChangeSet@1.1229.1.145, 2003-05-28 16:36:26+02:00, perex@suse.cz Merge suse.cz:/home/perex/bk/linux-sound/linux-2.5 into suse.cz:/home/perex/bk/linux-sound/linux-sound ChangeSet@1.1229.21.3, 2003-05-28 14:37:32+01:00, rmk@flint.arm.linux.org.uk [ARM] Declare mmu_gathers using DEFINE_PER_CPU. ChangeSet@1.1229.21.2, 2003-05-28 14:27:30+01:00, rmk@flint.arm.linux.org.uk [ARM] Remove old 26-bit ARM keyboard drivers Also remove mouse_ps2.c which was never referenced from the Makefile. ChangeSet@1.1229.21.1, 2003-05-28 14:18:25+01:00, rmk@flint.arm.linux.org.uk [ARM] Fix GCC3.3 build error GCC 3.3 complains that r2 overlaps input operands when a u64 pointer is passed into __put_user(). Fix this by using ip as a temporary register instead. ChangeSet@1.1164.11.2, 2003-05-28 12:05:53+02:00, oliver@vermuden.neukum.org - major cleanup of the module code ChangeSet@1.1229.7.42, 2003-05-28 02:31:06-07:00, davem@nuts.ninka.net [IPV6]: Fix typo in defragmentation changes. ChangeSet@1.1229.7.41, 2003-05-28 01:57:49-07:00, yoshfuji@linux-ipv6.org [IPV6]: Fix possible dst leakage in ndisc_send_redirect. ChangeSet@1.1229.7.40, 2003-05-28 01:49:40-07:00, yoshfuji@linux-ipv6.org [IPV6]: Always remove fragment header. ChangeSet@1.1229.7.39, 2003-05-28 01:48:08-07:00, bdschuym@pandora.be [BRIDGE]: Remove unnecessary code in br_input. ChangeSet@1.1229.7.38, 2003-05-28 01:41:35-07:00, yoshfuji@linux-ipv6.org [IPV6]: Clean up ip6_dst_alloc() calls. ChangeSet@1.1229.17.24, 2003-05-28 05:09:54-03:00, acme@conectiva.com.br o net: abstract access to struct sock ->flags This makes: 1. simpler primitive to access struct sock flags, shorter 2. we check if the flag is valid by using enum sock_flags 3. we can change the implementation to an open coded bit operations if it proves to be faster than the more general bit manipulation routines now used, i.e. we only have to change sock.h, not the whole net tree like now ChangeSet@1.1229.7.37, 2003-05-28 00:45:59-07:00, davem@nuts.ninka.net [IPV4]: Add sysctl to control ipfrag_secret_interval. ChangeSet@1.1229.7.36, 2003-05-27 22:17:04-07:00, davem@nuts.ninka.net [IPV6]: Input full addresses into TCP_SYNQ hash function. ChangeSet@1.1229.7.35, 2003-05-27 22:16:23-07:00, davem@nuts.ninka.net [IPV4/IPV6]: Use Jenkins hash for fragment reassembly handling. ChangeSet@1.1229.7.34, 2003-05-27 22:07:08-07:00, herbert@gondor.apana.org.au [NET]: Missing refcount bump in flow cache. ChangeSet@1.1229.17.23, 2003-05-27 20:00:20-07:00, elenstev@mesatop.com [PATCH] Use '#ifdef' to test for CONFIG options ChangeSet@1.1229.17.22, 2003-05-27 19:58:25-07:00, ink@jurassic.park.msu.ru [PATCH] alpha: single-step breakpoints - updated fix Restore 2.4 behavior when setting the single step breakpoints. ChangeSet@1.1229.17.21, 2003-05-27 19:58:16-07:00, ink@jurassic.park.msu.ru [PATCH] alpha: typo in EISA bridge detection ChangeSet@1.1229.17.20, 2003-05-27 19:58:07-07:00, ink@jurassic.park.msu.ru [PATCH] alpha: fix panic on smp boot (fork_by_hand) ChangeSet@1.1229.17.19, 2003-05-27 19:57:15-07:00, ink@jurassic.park.msu.ru [PATCH] alpha: compile warning fix Make the "addr" arg to test_bit "const" to prevent flood of compile warnings in networking code. ChangeSet@1.1229.17.18, 2003-05-27 19:57:07-07:00, hirofumi@mail.parknet.co.jp [PATCH] FAT cluster chain cache per inode (5/5) This adds a cache of lastest accessed cluster to inode for sequential access. The following is 500M file of FAT-to-FAT copy test, this may be a most different case in usual operations, because maximum readahead window flush the all caches. 512 bytes blocksize, 4096 bytes cluster size. linux-2.5.69-bk12 root@devron (a)[1232]# time cp file file1 real 7m58.900s user 0m0.267s sys 6m44.258s linux-2.5.69-bk12+patch root@devron (a)[1576]# time cp file file1 real 2m44.309s user 0m0.270s sys 0m28.631s ChangeSet@1.1229.17.17, 2003-05-27 19:56:58-07:00, hirofumi@mail.parknet.co.jp [PATCH] FAT cluster chain cache per superblock (4/5) This shifts the data position caches from module to per-superblock, and cleanups. ChangeSet@1.1229.17.16, 2003-05-27 19:56:50-07:00, hirofumi@mail.parknet.co.jp [PATCH] Remove Documentation/filesystems/fat_cvf.txt (3/5) This removes the obsolete Documentation/filesystems/fat_cvf.txt. ChangeSet@1.1229.17.15, 2003-05-27 19:56:40-07:00, hirofumi@mail.parknet.co.jp [PATCH] Fix VFAT_IOCTL_READDIR_BOTH/_SHORT ioctl (2/5) This fixes the return value of ioctl() for enables using the same way as readdir(). put/get_user() return code check patch from John R R Leavitt ChangeSet@1.1229.17.14, 2003-05-27 19:56:32-07:00, hirofumi@mail.parknet.co.jp [PATCH] Adds the large partition (> 128GB) support to FAT (1/5) This adds large partition (> 128GB) support to FAT. ChangeSet@1.1229.17.13, 2003-05-27 19:32:00-07:00, miles@lsi.nec.co.jp [PATCH] const-qualify memory arg in v850's __test_bit This silences at least one compile-time warning... :-) ChangeSet@1.1229.17.12, 2003-05-27 19:31:21-07:00, miles@lsi.nec.co.jp [PATCH] Update irq.c on v850 to use irqreturn_t ChangeSet@1.1229.17.11, 2003-05-27 19:30:37-07:00, miles@lsi.nec.co.jp [PATCH] Add v850 support for hardware single-step (via ptrace) ChangeSet@1.1229.17.10, 2003-05-27 19:29:47-07:00, miles@lsi.nec.co.jp [PATCH] Whitespace and comment cleanups for v850 entry.S ChangeSet@1.1229.17.9, 2003-05-27 19:29:00-07:00, miles@lsi.nec.co.jp [PATCH] Add leading underline to new linker-script symbols on the v850 This is needed to match the output of the C compiler. ChangeSet@1.1229.17.8, 2003-05-27 19:28:32-07:00, miles@lsi.nec.co.jp [PATCH] Add __KERNEL__ guard to nb85e_cache.h on v850 This header ends up getting included by uClibc (though nothing in it is used), so this protection is necessary to avoid problems with kernel-only typedefs. ChangeSet@1.1229.17.7, 2003-05-27 19:28:16-07:00, miles@lsi.nec.co.jp [PATCH] Handle new do_fork return value on v850 ChangeSet@1.1229.17.6, 2003-05-27 19:28:03-07:00, miles@lsi.nec.co.jp [PATCH] Miscellaneous v850 whitespace and comment cleanups ChangeSet@1.1229.17.5, 2003-05-27 19:27:51-07:00, miles@lsi.nec.co.jp [PATCH] Include in arch/v850/kernel/rte_cb_leds.c This is to define `struct file'; apparently some include-file change removed a previous implicit include. ChangeSet@1.1229.17.4, 2003-05-27 19:27:40-07:00, miles@lsi.nec.co.jp [PATCH] Remove some unneeded register saving on the v850 These registers are now saved in a difference place, but the old code was inadvertently left in. ChangeSet@1.1229.17.3, 2003-05-27 18:13:21-07:00, viro@parcelfarce.linux.theplanet.co.uk [PATCH] procfs bug exposed by cdev changes fs/inode.c assumes that any ->delete_inode() will call clear_inode(). procfs instance doesn't. It had passed unpunished for a while; cdev changes combined with ALSA creating character devices in procfs made it fatal. Patch follows. It had fixed ALSA-triggered memory corruption here - what happens in vanilla 2.5.70 is that clear_inode() is not called when procfs character device inodes are freed. That leaves a freed inode on a cyclic list, with obvious unpleasantness following when we try to traverse it (e.g. when unregistering a device). ChangeSet@1.1229.19.2, 2003-05-28 09:03:10+10:00, anton@samba.org ppc64: merge conflicts ChangeSet@1.1229.19.1, 2003-05-28 09:00:16+10:00, anton@samba.org ppc64: merge some conflicts ChangeSet@1.1229.8.107, 2003-05-27 12:21:06-07:00, axboe@suse.de [PATCH] blk layer tag resize This allows drivers to resize their tag depth at run-time. ChangeSet@1.1229.8.106, 2003-05-27 12:16:20-07:00, viro@parcelfarce.linux.theplanet.co.uk [PATCH] Fix sound lockup - missing chardev init Argh. Missing initialization in char_dev.c - it's definitely responsible for crap on unload. Load side appears to be something else, though... ChangeSet@1.1229.18.2, 2003-05-27 13:31:10-05:00, stevef@smfhome1.austin.rr.com adjust for change of devname to const char (new mount format) ChangeSet@1.1229.18.1, 2003-05-27 07:51:12-07:00, cifs.adm@hostme.bitkeeper.com Merge bk://linux.bkbits.net/linux-2.5 into hostme.bitkeeper.com:/ua/repos/c/cifs/linux-2.5cifs ChangeSet@1.1229.16.1, 2003-05-27 07:39:14-07:00, axboe@suse.de [PATCH] remove buggy BUG_ON in ide-cd Alan (or someone else) added a buggy BUG_ON() in ide-cd. We can address > 32-bit just fine with 2kb block size. People are hitting this, just got one more report today... ChangeSet@1.1229.15.1, 2003-05-27 07:46:25-05:00, shaggy@shaggy.austin.ibm.com Merge jfs@jfs.bkbits.net:linux-2.5 into shaggy.austin.ibm.com:/shaggy/bk/jfs-2.5 ChangeSet@1.1164.11.1, 2003-05-27 13:06:46+02:00, oliver@vermuden.neukum.org 01-debug-cleanup.patch ChangeSet@1.1229.11.2, 2003-05-27 12:00:02+01:00, davej@codemonkey.org.uk [AGPGART] Yet another missed typedef compile fix. ChangeSet@1.1229.7.33, 2003-05-27 02:12:19-07:00, herbert@gondor.apana.org.au [IPSEC]: Order SPD using priority. ChangeSet@1.1229.7.32, 2003-05-27 00:51:13-07:00, yoshfuji@linux-ipv6.org [IPV6]: Fix possible oops in ndisc_send_na. ChangeSet@1.1229.7.31, 2003-05-27 00:50:12-07:00, davem@nuts.ninka.net [NET]: One too many IRQ_HANDLED added to sunqe.c driver. ChangeSet@1.1229.7.30, 2003-05-27 00:20:44-07:00, yoshfuji@linux-ipv6.org [IPV6]: Fix possible idev leakage in icmp.c ChangeSet@1.1229.7.29, 2003-05-27 00:10:59-07:00, yoshfuji@linux-ipv6.org [CRYPTO]: Fix compiler warnings in sha512.c ChangeSet@1.1229.7.27, 2003-05-26 23:59:47-07:00, davem@nuts.ninka.net Merge nuts.ninka.net:/home/davem/src/BK/network-2.5 into nuts.ninka.net:/home/davem/src/BK/net-2.5 ChangeSet@1.1229.8.103, 2003-05-26 23:07:28-07:00, torvalds@home.transmeta.com Remove a few zero-sized files, as noted by David Gibson. ChangeSet@1.1229.14.7, 2003-05-27 01:37:28-04:00, engebret@us.ibm.com [netdrvr pcnet32] bug fixes I would like to see a couple of the pcnet32 changes that I think we can agree on be put into the trees so a couple of the potential defects can be avoided. The following patch contains just these pieces. The only controversial one is an arbitrary change in the number of iterations in a while loop spinning on hardware state. No matter how this is done, I am not especially fond of this bit of code as it has no reasonable error recovery path -- however, as a half-way, incremental solution, increasing the polling time should help as the 100 value was certainly found to be insufficient. 1000 may not be sufficient either, but it is certainly no worse. Both of the other changes were hit in testing (and I belive the wmb() at a customer even), so it would help reduce some debug if these go in. Any feedback is appreciated - thanks. ChangeSet@1.1229.14.6, 2003-05-27 01:14:11-04:00, jgarzik@redhat.com [netdrvr eepro] update MODULE_AUTHOR per old-author request ChangeSet@1.1229.14.5, 2003-05-27 01:05:58-04:00, edward_peng@dlink.com.tw [netdrvr sundance] fix another flow control bug ChangeSet@1.1229.14.4, 2003-05-27 01:04:28-04:00, edward_peng@dlink.com.tw [netdrvr sundance] fix flow control bug ChangeSet@1.1229.14.3, 2003-05-27 01:01:53-04:00, bunk@fs.tum.de [wan lmc] remove 2.0.x-era code The patch below removes obsolete #if'd code for kernel 2.0 and 2.2 from drivers/net/wan/lmc/* (this includes the expansion of some #define's that were definded differently for different kernel versions). ChangeSet@1.1229.14.2, 2003-05-27 00:45:17-04:00, shmulik.hen@intel.com [netdrvr bonding] fix ABI version control problem This fix makes bonding not commit to a specific ABI version if the ioctl command is not supported by bonding. (It also removes the '\n' in the continuous printk reporting the link down event in bond_mii_monitor - it got in there by mistake in our previous patch set and caused log messages to appear funny in some situations). ChangeSet@1.1229.14.1, 2003-05-27 00:42:05-04:00, shmulik.hen@intel.com [netdrvr bonding] fix long failover in 802.3ad mode This patch fixes the bug reported by Jay on April 3rd regarding long failover time when releasing the last slave in the active aggregator. The fix, as suggested by Jay, is to follow the spec recommendation and send a LACPDU to the partner saying this port is no longer aggregatable and therefore trigger an immediate re-selection of a new aggregator instead of waiting the entire expiration timeout. ChangeSet@1.1229.8.101, 2003-05-26 21:13:50-07:00, gerg@snapgear.com [PATCH] compile m68knommu/ColdFire ints.c Add the m68knommu/Coldfire specific ints.c to build list. ChangeSet@1.1229.8.100, 2003-05-26 21:13:39-07:00, gerg@snapgear.com [PATCH] don't compile m68knommu/kernel ints.c Modify m68knommu/kernel Makefile to no longer compile removed common ints.c. ChangeSet@1.1229.8.99, 2003-05-26 21:13:29-07:00, gerg@snapgear.com [PATCH] remove common m68knommu ints.c Remove the m68knommu common ints.c. No longer needed with each sub-architecture now having its own. ChangeSet@1.1229.8.98, 2003-05-26 21:13:21-07:00, gerg@snapgear.com [PATCH] create m68knommu/coldfire specific ints.c Create a m68knommu/ColdFire specific ints.c. It is just simpler to have one for each sub-architecture (which means we currently need 3 for the 3 prominant m68knommu families). Each can handle the hardware setup differences, and there is a few at this level. ChangeSet@1.1229.8.97, 2003-05-26 21:06:15-07:00, neilb@cse.unsw.edu.au [PATCH] md: Replace bdev_partition_name with calls to bdevname ChangeSet@1.1229.8.96, 2003-05-26 21:06:04-07:00, neilb@cse.unsw.edu.au [PATCH] md: Remove dependance on MD_SB_DISKS in linear personality Linear uses one array sized by MD_SB_DISKS inside a structure. We move it to the end of the structure, declare it as size 0, and arrange for approprate extra space to be allocated on structure allocation. ChangeSet@1.1229.8.95, 2003-05-26 21:05:55-07:00, neilb@cse.unsw.edu.au [PATCH] md: Remove MD_SB_DISKS limits from raid1 raid1 uses MD_SB_DISKS to size two data structures, but the new version-1 superblock allows for more than this number of disks (and most actual arrays use many fewer). This patch sizes to two arrays dynamically. One becomes a separate kmalloced array. The other is moved to the end of the containing structure and appropriate extra space is allocated. Also, change r1buf_pool_alloc (which allocates buffers for a mempool for doing re-sync) to not get r1bio structures from the r1bio pool (which could exhaust the pool) but instead to allocate them separately. ChangeSet@1.1229.8.94, 2003-05-26 21:05:45-07:00, neilb@cse.unsw.edu.au [PATCH] md: Remove dependancy on MD_SB_DISKS from raid0 Arrays with type-1 superblock can have more than MD_SB_DISKS, so we remove the dependancy on that number from raid0, replacing several fixed sized arrays with one dynamically allocated array. ChangeSet@1.1229.8.93, 2003-05-26 21:05:35-07:00, neilb@cse.unsw.edu.au [PATCH] md: Remove dependancy on MD_SB_DISKS from raid5 One embeded array gets moved to end of structure and sized dynamically. ChangeSet@1.1229.8.92, 2003-05-26 21:05:25-07:00, neilb@cse.unsw.edu.au [PATCH] md: Remove dependancy on MD_SB_DISKS from multipath Multipath has a dependancy on MD_SB_DISKS which is no longer authoritative. We change it to use a separately allocated array. ChangeSet@1.1229.8.91, 2003-05-26 21:05:15-07:00, neilb@cse.unsw.edu.au [PATCH] md: Improve raid0 mapping code to simplify and reduce mem usage. To cope with a raid0 array with differing sized devices, raid0 divides an array into "strip zones". The first zone covers the start of all devices, upto an offset equal to the size of the smallest device. The second strip zone covers the remaining devices upto the size of the next smallest size, etc. In order to determing which strip zone a given address is in, the array is logically divided into slices the size of the smallest zone, and a 'hash' table is created listing the first and, if relevant, second zone in each slice. As the smallest slice can be very small (imagine an array with a 76G drive and a 75.5G drive) this hash table can be rather large. With this patch, we limit the size of the hash table to one page, at the possible cost of making several probes into the zone list before we find the correct zone. We also cope with the possibility that a zone could be larger than a 32bit sector address would allow. ChangeSet@1.1229.8.90, 2003-05-26 21:05:05-07:00, neilb@cse.unsw.edu.au [PATCH] md: Always allow a half-built md array to be stopped. When starting an array fails, we have to tear it down, but in some circumstances (particularly autostart_array) the reference count will be 3, so do_md_stop will fail. With this patch we only worry about the number of users is the array has been fully started. ChangeSet@1.1229.8.89, 2003-05-26 21:04:55-07:00, neilb@cse.unsw.edu.au [PATCH] md: Get rid of vmalloc/vfree from raid0 raid0 currently uses vmalloc instead of kmalloc. This patch changes to kmalloc. There is one allocation that can occasionally be very large - the hash_table. A subsequent patch will address this issue. ChangeSet@1.1229.8.88, 2003-05-26 21:04:45-07:00, neilb@cse.unsw.edu.au [PATCH] md: Fix simple off-by-one error in md driver. ChangeSet@1.1229.8.87, 2003-05-26 21:04:35-07:00, neilb@cse.unsw.edu.au [PATCH] md: Improve test for which raid1 device doesn't need to be written to. Instead of testing last_used (which could change in unusual circumstances) we test against the bdev that we read frmo, and don't write back to there. ChangeSet@1.1229.8.86, 2003-05-26 21:04:25-07:00, neilb@cse.unsw.edu.au [PATCH] md: Handle concurrent failure of two drives in raid5 If two drives both fail during a write request, raid5 doesn't cope properly and will eventually oops. With this patch, blocks that have already been 'written' are failed when double drive failure is noticed, as well as blocks that are about to be written. ChangeSet@1.1229.8.85, 2003-05-26 21:04:15-07:00, neilb@cse.unsw.edu.au [PATCH] md: Use new single page bio splitting for raid0 and linear Sometimes raid0 and linear are required to take a single page bio that spans two devices. We use bio_split to split such a bio into two. The the same time, bio.h is included by linux/raid/md.h so we don't included it elsewhere anymore. We also modify the mergeable_bvec functions to allow a bvec that doesn't fit if it is the first bvec to be added to the bio, and be careful never to return a negative length from a bvec_mergable funciton. ChangeSet@1.1229.8.84, 2003-05-26 21:04:04-07:00, neilb@cse.unsw.edu.au [PATCH] md: Export bio_split_pool for md to use. ChangeSet@1.1229.8.83, 2003-05-26 20:55:45-07:00, viro@www.linux.org.uk [PATCH] callout removal: callout is gone OK, that's it - no callout drivers left, so we drop the warning from tty_open() and kill the code that handles them in /proc/tty/drivers. ChangeSet@1.1229.8.82, 2003-05-26 20:55:36-07:00, viro@www.linux.org.uk [PATCH] callout removal: cleanup misc cleanup - switched to standard constant names for .subtype in the drivers that had private equivalents, removed unused ->callout_termios and friends from the last places where they were defined. ChangeSet@1.1229.8.81, 2003-05-26 20:55:24-07:00, viro@www.linux.org.uk [PATCH] callout removal: isdn callout removal: isdn ChangeSet@1.1229.8.80, 2003-05-26 20:55:15-07:00, viro@www.linux.org.uk [PATCH] callout removal: cris callout removal: cris ChangeSet@1.1229.8.79, 2003-05-26 20:55:05-07:00, viro@www.linux.org.uk [PATCH] callout removal: simserial callout removal: simserial ChangeSet@1.1229.8.78, 2003-05-26 20:54:55-07:00, viro@www.linux.org.uk [PATCH] callout removal: mips callout removal: mips ChangeSet@1.1229.8.77, 2003-05-26 20:54:45-07:00, viro@www.linux.org.uk [PATCH] callout removal: vacserial callout removal: vacserial ChangeSet@1.1229.8.76, 2003-05-26 20:54:36-07:00, viro@www.linux.org.uk [PATCH] callout removal: sicc callout removal: sicc ChangeSet@1.1229.8.75, 2003-05-26 20:54:26-07:00, viro@www.linux.org.uk [PATCH] callout removal: 8260_uart callout removal: 8260_uart ChangeSet@1.1229.8.74, 2003-05-26 20:54:15-07:00, viro@www.linux.org.uk [PATCH] callout removal: 8xx_uart callout removal: 8xx_uart ChangeSet@1.1229.8.73, 2003-05-26 20:54:05-07:00, viro@www.linux.org.uk [PATCH] callout removal: amiserial callout removal: amiserial ChangeSet@1.1229.8.72, 2003-05-26 20:53:55-07:00, viro@www.linux.org.uk [PATCH] callout removal: cyclades callout removal: cyclades ChangeSet@1.1229.8.71, 2003-05-26 20:53:45-07:00, viro@www.linux.org.uk [PATCH] callout removal: dz callout removal: dz ChangeSet@1.1229.8.70, 2003-05-26 20:53:34-07:00, viro@www.linux.org.uk [PATCH] callout removal: epca callout removal: epca ChangeSet@1.1229.8.69, 2003-05-26 20:53:24-07:00, viro@www.linux.org.uk [PATCH] callout removal: esp callout removal: esp ChangeSet@1.1229.8.68, 2003-05-26 20:53:15-07:00, viro@www.linux.org.uk [PATCH] callout removal: ip2 callout removal: ip2 ChangeSet@1.1229.8.67, 2003-05-26 20:53:05-07:00, viro@www.linux.org.uk [PATCH] callout removal: moxa callout removal: moxa ChangeSet@1.1229.8.66, 2003-05-26 20:52:54-07:00, viro@www.linux.org.uk [PATCH] callout removal: mxser callout removal: mxser ChangeSet@1.1229.8.65, 2003-05-26 20:52:44-07:00, viro@www.linux.org.uk [PATCH] callout removal: pcxx callout removal: pcxx ChangeSet@1.1229.8.64, 2003-05-26 20:52:34-07:00, viro@www.linux.org.uk [PATCH] callout removal: rocket callout removal: rocket ChangeSet@1.1229.8.63, 2003-05-26 20:52:24-07:00, viro@www.linux.org.uk [PATCH] callout removal: serial167 callout removal: serial167 ChangeSet@1.1229.8.62, 2003-05-26 20:52:15-07:00, viro@www.linux.org.uk [PATCH] callout removal: synclink callout removal: synclink ChangeSet@1.1229.8.61, 2003-05-26 20:52:05-07:00, viro@www.linux.org.uk [PATCH] callout removal: synclinkmp callout removal: synclinkmp ChangeSet@1.1229.8.60, 2003-05-26 20:51:55-07:00, viro@www.linux.org.uk [PATCH] callout removal: synclink_cs callout removal: synclink_cs ChangeSet@1.1229.8.59, 2003-05-26 20:51:45-07:00, viro@www.linux.org.uk [PATCH] callout removal: macserial callout removal: macserial ChangeSet@1.1229.8.58, 2003-05-26 20:51:35-07:00, viro@www.linux.org.uk [PATCH] callout removal: pc300 callout removal: pc300 ChangeSet@1.1229.8.57, 2003-05-26 20:51:25-07:00, viro@www.linux.org.uk [PATCH] callout removal: chdlc callout removal: chdlc ChangeSet@1.1229.8.56, 2003-05-26 20:51:14-07:00, viro@www.linux.org.uk [PATCH] callout removal: 68328 callout removal: 68328 ChangeSet@1.1229.8.55, 2003-05-26 20:51:05-07:00, viro@www.linux.org.uk [PATCH] callout removal: isicom callout removal: isicom ChangeSet@1.1229.8.54, 2003-05-26 20:50:55-07:00, viro@www.linux.org.uk [PATCH] callout removal: generic_serial callout removal: generic_serial ChangeSet@1.1229.8.53, 2003-05-26 20:50:45-07:00, viro@www.linux.org.uk [PATCH] callout removal: tx3912 callout removal: tx3912 ChangeSet@1.1229.8.52, 2003-05-26 20:50:34-07:00, viro@www.linux.org.uk [PATCH] callout removal: vme callout removal: vme ChangeSet@1.1229.8.51, 2003-05-26 20:50:24-07:00, viro@www.linux.org.uk [PATCH] callout removal: sci callout removal: sci ChangeSet@1.1229.8.50, 2003-05-26 20:50:15-07:00, viro@www.linux.org.uk [PATCH] callout removal: istallion callout removal: istallion ChangeSet@1.1229.8.49, 2003-05-26 20:50:05-07:00, viro@www.linux.org.uk [PATCH] callout removal: riscom8 callout removal: riscom8 ChangeSet@1.1229.8.48, 2003-05-26 20:49:55-07:00, viro@www.linux.org.uk [PATCH] callout removal: a2232 callout removal: a2232 ChangeSet@1.1229.8.47, 2003-05-26 20:49:45-07:00, viro@www.linux.org.uk [PATCH] callout removal: specialix callout removal: specialix ChangeSet@1.1229.8.46, 2003-05-26 20:49:34-07:00, viro@www.linux.org.uk [PATCH] callout removal: sx callout removal: sx ChangeSet@1.1229.8.45, 2003-05-26 20:49:25-07:00, viro@www.linux.org.uk [PATCH] callout removal: rio callout removal: rio ChangeSet@1.1229.8.44, 2003-05-26 20:49:15-07:00, viro@www.linux.org.uk [PATCH] callout removal: stallion callout removal: stallion ChangeSet@1.1229.8.43, 2003-05-26 20:49:05-07:00, viro@www.linux.org.uk [PATCH] callout removal: aurora callout removal: aurora ChangeSet@1.1229.8.42, 2003-05-26 20:48:56-07:00, viro@www.linux.org.uk [PATCH] callout removal: sgiserial callout removal: sgiserial ChangeSet@1.1229.8.41, 2003-05-26 20:48:45-07:00, viro@www.linux.org.uk [PATCH] callout removal: tc_zs callout removal: tc_zs ChangeSet@1.1229.8.40, 2003-05-26 20:48:35-07:00, viro@www.linux.org.uk [PATCH] callout removal: 68360 callout removal: 68360 ChangeSet@1.1229.8.39, 2003-05-26 20:48:25-07:00, viro@www.linux.org.uk [PATCH] callout removal: mcfserial callout removal: mcfserial ChangeSet@1.1229.8.38, 2003-05-26 20:48:16-07:00, viro@www.linux.org.uk [PATCH] callout removal: ircomm_tty callout removal: ircomm_tty ChangeSet@1.1229.8.37, 2003-05-26 19:49:21-07:00, bcollins@debian.org [PATCH] Update IEEE1394 (r939) - Adds fragementation support to eth1394 - Fix race conditition in packet completion task call - Fix lack of proper logic in tlabel allocation - Fix brokeness introduced by "stanford checker fixes for memset" in ohci1394 - Add trivial PM resume callback in ohci1394 to support sleep/resume. ChangeSet@1.1229.13.1, 2003-05-26 23:31:51-03:00, acme@conectiva.com.br o wanrouter: fix bug introduced by latest namespace fix Thanks to Adrian Bunk for reporting. ChangeSet@1.1229.12.1, 2003-05-26 19:10:22-07:00, akpm@digeo.com [PATCH] truncate and timestamps This patch will put us back to the 2.4 behaviour while preserving the truncation speedup. It's a bit dopey (why do the timestamp update in the fs at all?) but changing this stuff tends to cause subtle problems. ChangeSet@1.1229.8.35, 2003-05-26 21:02:24-05:00, jejb@raven.il.steeleye.com Merge raven.il.steeleye.com:/mnt1/jejb/BK/scsi-aic-2.5 into raven.il.steeleye.com:/home/jejb/BK/scsi-for-linus-2.5 ChangeSet@1.1229.8.34, 2003-05-26 20:55:10-05:00, jejb@raven.il.steeleye.com Merge raven.il.steeleye.com:/home/jejb/BK/scsi-misc-2.5 into raven.il.steeleye.com:/home/jejb/BK/scsi-for-linus-2.5 ChangeSet@1.1229.9.2, 2003-05-27 02:03:58+01:00, davej@codemonkey.org.uk [AGPGART] Remove useless early agp_init() from i810fb agp_init() just printk's a banner. This is unnecessary at this early stage. ChangeSet@1.1229.9.1, 2003-05-27 02:02:14+01:00, davej@codemonkey.org.uk [AGPGART] Compilation fix. Death of a typedef in an earlier cset broke i810fb ChangeSet@1.1229.8.33, 2003-05-26 17:57:56-07:00, torvalds@home.transmeta.com Linux v2.5.70 TAG: v2.5.70