- 29 Oct, 2004 40 commits
-
-
Linus Torvalds authored
This avoids a bug where the compiler would overwrite the stackframe that the caller also considered to be a register save area. It also shrinks the code segment by a tiny amount by moving the failure case argument setup into the slow path. This not only makes the fast path smaller, but it makes it easier on gcc (gcc is not very good at generating code that uses fixed register names).
-
James Nelson authored
This file is no longer reqired - the MAKEDEV program makes the ida/ nodes automatically. Signed-off-by:
James Nelson <james4765@gmail.com> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
James Nelson authored
General cleanup of Documentation/cpqarray.txt. Removal of old and obsolete references, removed references to an external patch. Signed-off-by:
-
Marcel Holtmann authored
The MODULE_PARM is deprecated now and should be replaced by module_param() or module_param_named(). Signed-off-by:
Marcel Holtmann <marcel@holtmann.org> Signed-off-by:
-
Andrew Morton authored
Normal initcalls which call out to userspace may need to create pipes. Signed-off-by:
-
Robert Love authored
make dnotify configurable, via CONFIG_DNOTIFY. CONFIG_EMBEDDED is required for disabling dnotify. Signed-off-by:
-
Daniel Jacobowitz authored
When working on GDB support I found a typo. I assume the comment is correct. If you step to this particular instruction and backtrace, GDB gets lost. Signed-off-by:
-
Dmitry Torokhov authored
I8K: Switch to using module_param, allow switching 'power_status' through sysfs. Signed-off-by:
Dmitry Torokhov <dtor@mail.ru> Signed-off-by:
-
Antonino Daplas authored
Accepted Geert's offer to maintain the framebuffer layer. Signed-off-by:
Antonino Daplas <adaplas@pol.net> Signed-off-by:
-
Antonino Daplas authored
Check if cursor image has changed in intelfb Signed-off-by:
-
Antonino Daplas authored
Add __iomem annotations to vfb Signed-off-by:
-
Antonino Daplas authored
Add __iomem annotations to cirrusfb Signed-off-by:
-
Antonino Daplas authored
Add __iomem annotations to hgafb Signed-off-by:
-
Antonino Daplas authored
Add __iomem annotations to pm2fb Signed-off-by:
-
Antonino Daplas authored
Add __iomem annotations to cyber2000fb Signed-off-by:
-
Antonino Daplas authored
Do not memset the framebuffer memory in asiliantfb Signed-off-by:
-
Antonino Daplas authored
Add __iomem annotations to tdfxfb Signed-off-by:
-
Antonino Daplas authored
Add __iomem annotations to sstfb Signed-off-by:
-
Antonino Daplas authored
- Fix IO access in neofb - Use readl/writel - add __iomem annotations Signed-off-by:
-
Antonino Daplas authored
Convert MODULE_PARM to module_param in neofb Signed-off-by:
-
Antonino Daplas authored
fbdev: Convert MODULE_PARM to module_param in intelfb Signed-off-by:
-
Antonino Daplas authored
Remove module parameter 'disabled' from savagefb Signed-off-by:
-
Antonino Daplas authored
Convert MODULE_PARM to module_param in i810fb Signed-off-by:
-
Stephen C. Tweedie authored
The patch below adds online resize capability to ext3 based on Andreas patch for 2.4 and fixed up by Stephen. The patch also removes s_debts: s_debts is currently not used by ext3 (it is created, destroyed and checked but never set). Remove it for now. Resurrecting this will require adding it back in changed form. In existing form it's already unsafe wrt. byte-tearing as it performs unlocked byte increment/decrement on words which may be being accessed simultaneously on other CPUs. It is also the only in-memory dynamic table which needs to be extended by online-resize, so locking it will require care. Signed-off-by:
-
Andrew Morton authored
Enable the reservation code by default. So reservations default to "on". They can be disabled by mounting with the "noreservation" mount option. Signed-off-by:
-
Mingming Cao authored
rbtree implementation and other changes From: Stephen Tweedie <sct@redhat.com> contributions From: Badari Pulavarty <pbadari@us.ibm.com> and probably me. This is the ext3 block reservation patch. It improves the layout of ext3 files by establishing, for each inode, reserved areas of the disk in which only that file can allocate blocks. Those reserved areas are managed in an rbtree, via the in-core inode. It's a bit like ext2 preallocation only stronger in that it can span already-allocated blocks, including the per-blockgroup inode tables and bitmaps. The patch fixes ext3's worst performance problem: disastrous layout when multiple files are being concurrently grown. It increases the size of the inode by rather a lot. A todo item is to dynamically allocate the `struct reserve_window_node', so we don't need to carry this storage for inodes which aren't opened for writing. The feature is enabled by mounting with the "reservation" mount option. Reservations default to "off". Signed-off-by:
-
Mingming Cao authored
Cleans up the old ext3 preallocation code carried from ext2 but turned off. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
While changing the TTY layer, an API parameter was removed, so it was removed by almost all calls, changing their prototype. But one use of one such function was not updated, breaking UML compilation. This is the fix. Signed-off-by:
Paolo 'Blaisorblade' Giarrusso <blaisorblade_spam@yahoo.it> Signed-off-by:
-
Hirokazu Takata authored
This patch fixes a typo of arch/m32r/lib/delay.c. It is required to fix a compile error for non-CONFIG_ISA_DUAL_ISSUE configuration. NOTE: The m32r has a kind of ISA (instruction set architecture) like a 2-way VLIW. Originally the m32r has two types of 16-bit/32-bit instructions. Only 16-bit instruction pair can be executed in parallel on a m32r chip with dual issuing support (e.g. M32700). According to the insturction dual issuing of the m32r, a programmer/compiler can explicitly put two 16-bit instructions into an instruction word, which are to be executed in parallel. ex. Assembly coding style of two 16-bit instruncions; insn A and B: 1) insn A insn B ; sequential execution (implicit) 2) insn A -> insn B ; sequential execution (explicit) 3) insn A || insn B ; parallel execution Signed-off-by:Hirokazu Takata <takata@linux-m32r.org> Signed-off-by:
-
Andi Kleen authored
Straightforward patch to remove a warning in genapic about a unused function. Independently done by Chris Wright too. Signed-off-by:
Andi Kleen <ak@muc.de> Signed-off-by:
-
Andi Kleen authored
genapic broke early safe_smp_processor_id(), especially when you got a WARN_ON or oops early it would loops forever in show_trace. The reason was that the x86_cpu_to_apicid array wasn't correctly initialized. This patch fixes this by just testing for this case. Orginally from James Cleverdon Acked-by: Andi Kleen <ak@muc.de> Signed-off-by:
-
Benjamin Herrenschmidt authored
Now that pci_get_legacy_ide_irq() support has been merged, it's time to enable use of it by the Maple platform code. Signed-off-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Olof Johansson authored
Nathan Lynch pointed this out: The CPU sibling map is never initialized on iSeries. This makes the scheduler very unhappy if CONFIG_SCHED_SMT is enabled, causing an oops in find_busiest_group during boot. Below patch adds the expected init. Signed-off-by:
Olof Johansson <olof@austin.ibm.com> Signed-off-by:
-
Stephen Rothwell authored
The generic irq patches broke pci irqs on ppc64 iSeries. Signed-off-by:
Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by:
-
Benjamin Herrenschmidt authored
Tom's recent irq patch broke PowerMac (and possibly others). I think he forgot that PReP, CHRP and PowerMac are all built together in a single kernel image, thus all of those arch_initcall's will end up beeing called, even on the wrong machine... Signed-off-by:
-
Matt Porter authored
The patch fixes these warnings by doing two things: 1) Add the argument to the printk. 2) Rearrange the ifdef to eliminate the unused variable and function warnings. Signed-off-by:Scott Anderson <sanders@mvista.com> Signed-off-by:
Matt Porter <mporter@kernel.crashing.org> Signed-off-by:
-
Andrew Morton authored
Suppress page allocation warnings arising from radix_tree_node_alloc() allocations on behalf of swapper_space: these are expected under heavy load. Signed-off-by:
-
Andrew Morton authored
The data structures which are set up by key_init() are used by exec(). And we're using exec() super-early via the hotplug events from do_basic_setup():driver_init(). So call key_init() directly, prior to driver_init(). Cc: David S. Miller <davem@davemloft.net> Cc: David Howells <dhowells@redhat.com> Signed-off-by:
-
Pavel Machek authored
This adds typechecking to suspend types and powerdown types. This should solve at least part of suspend type confusion. There should be no code changes generated by this one. Acked-by:
Patrick Mochel <mochel@digitalimplant.org> Signed-off-by:
-
Toshihiro Iwamoto authored
It seems that O_DIRECT write sometimes leaks memory. Signed-off-by:
-
