- 29 3月, 2012 2 次提交
-
-
由 Srivatsa S. Bhat 提交于
This was marked as obsolete for quite a while now.. Now it is time to remove it altogether. And while doing this, get rid of first_cpu() as well. Also, remove the redundant setting of cpu_online_mask in smp_prepare_cpus() because the generic code would have already set cpu 0 in cpu_online_mask. Reported-by: NTony Luck <tony.luck@intel.com> Signed-off-by: NSrivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: NAndrew Morton <akpm@linux-foundation.org> Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
-
由 Gilad Ben-Yossef 提交于
We have lots of infrastructure in place to partition multi-core systems such that we have a group of CPUs that are dedicated to specific task: cgroups, scheduler and interrupt affinity, and cpuisol= boot parameter. Still, kernel code will at times interrupt all CPUs in the system via IPIs for various needs. These IPIs are useful and cannot be avoided altogether, but in certain cases it is possible to interrupt only specific CPUs that have useful work to do and not the entire system. This patch set, inspired by discussions with Peter Zijlstra and Frederic Weisbecker when testing the nohz task patch set, is a first stab at trying to explore doing this by locating the places where such global IPI calls are being made and turning the global IPI into an IPI for a specific group of CPUs. The purpose of the patch set is to get feedback if this is the right way to go for dealing with this issue and indeed, if the issue is even worth dealing with at all. Based on the feedback from this patch set I plan to offer further patches that address similar issue in other code paths. This patch creates an on_each_cpu_mask() and on_each_cpu_cond() infrastructure API (the former derived from existing arch specific versions in Tile and Arm) and uses them to turn several global IPI invocation to per CPU group invocations. Core kernel: on_each_cpu_mask() calls a function on processors specified by cpumask, which may or may not include the local processor. You must not call this function with disabled interrupts or from a hardware interrupt handler or from a bottom half handler. arch/arm: Note that the generic version is a little different then the Arm one: 1. It has the mask as first parameter 2. It calls the function on the calling CPU with interrupts disabled, but this should be OK since the function is called on the other CPUs with interrupts disabled anyway. arch/tile: The API is the same as the tile private one, but the generic version also calls the function on the with interrupts disabled in UP case This is OK since the function is called on the other CPUs with interrupts disabled. Signed-off-by: NGilad Ben-Yossef <gilad@benyossef.com> Reviewed-by: NChristoph Lameter <cl@linux.com> Acked-by: NChris Metcalf <cmetcalf@tilera.com> Acked-by: NPeter Zijlstra <a.p.zijlstra@chello.nl> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Russell King <linux@arm.linux.org.uk> Cc: Pekka Enberg <penberg@kernel.org> Cc: Matt Mackall <mpm@selenic.com> Cc: Rik van Riel <riel@redhat.com> Cc: Andi Kleen <andi@firstfloor.org> Cc: Sasha Levin <levinsasha928@gmail.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Avi Kivity <avi@redhat.com> Acked-by: NMichal Nazarewicz <mina86@mina86.org> Cc: Kosaki Motohiro <kosaki.motohiro@gmail.com> Cc: Milton Miller <miltonm@bga.com> Cc: Russell King <linux@arm.linux.org.uk> Acked-by: NPeter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: NAndrew Morton <akpm@linux-foundation.org> Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>
-
- 26 3月, 2012 4 次提交
-
-
由 Russell King 提交于
Now that Neponset, UCB1x00 and SA1111 are all converted to use the IRQ allocation interfaces, we can enable sparse IRQ support for SA11x0 platforms.
-
由 Rob Herring 提交于
In preparation to convert SA1100 to sparse irq, set .nr_irqs for each machine and explicitly include mach/irqs.h as needed. Signed-off-by: NRob Herring <rob.herring@calxeda.com> Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
-
由 Russell King 提交于
Remove the new unused DMA controller definitions from mach/SA-1100.h. These are now private to the SA-11x0 DMA engine driver and contained within the driver. Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
-
由 Russell King 提交于
Now that all users are converted over to using the DMA engine API, we can get rid of the old platform dependent DMA driver. Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
-
- 25 3月, 2012 32 次提交
-
-
由 Richard Weinberger 提交于
Enable ext4, cgroups and devtmpfs. Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Richard Weinberger 提交于
x86_64 UML is unable to load modules if more than 504MiB of memory are used. This happens because on x86_64 the UML process has a quite high start address (typically around 0x6000000). If UML's memory is larger than 504MiB VMALLOC_START happens to be after 0x8000000. This is no problem unless one loads a module which was built with R_X86_64_32S relocations. Symbols with a location > 0x8000000 cannot be used with R_X86_64_32S To deal with this x86_64 UML has to be compiled with -mcmodel=large such that no R_X86_64_32S relocations are used. Signed-off-by: NRichard Weinberger <richard@nod.at> Reported-by: N전하늘 <allskyee@gmail.com>
-
由 Richard Weinberger 提交于
UML has no io memory but implements everything defined in generic-asm/io.h. Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Richard Weinberger 提交于
At some places io_remap_pfn_range() is needed. UML has to serve it like all other archs do. Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
nicked from patch by dwmw2 back in July Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> [richard@nod.at: Re-export SUBARCH in arch/um/Makefile] Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
... just strip NOSTDINC_FLAGS out of it for those Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
... the same one that controls whether elf_aux.o is included into the build, bringing the vsyscall_e... into it. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
just provide get_current_pid() to the userland side of things instead of get_current() + manual poking in its results Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
... and switch chan_interrupt() to directly calling close_one_chan(), so we can lose delay_free_irq argument of close_chan() as well. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
... since chan_interrupt() might schedule it if there's too much incoming data. Kill task argument of chan_interrupt(), while we are at it - it's always &line->task. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
looks like a half-arsed duplicate of line->enabled Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
put references to in and out chans associated with line into explicit struct chan * fields in it. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
now we can do that... Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
move config-independent parts of initialization into register_lines(), call setup_one_line() after it instead of abusing ->init_str. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Current code doesn't update the symlinks in /sys/dev/char when we add/remove tty lines. Fixing that allows to stop messing with ->valid before the driver registration, which is a Good Thing(tm) - we shouldn't have it set before we really have the things set up and ready for line_open(). We need tty_driver available to call tty_{un,}register_device(), so we just stash a reference to it into struct line_driver. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Pull parse_chan_pair() call into setup_one_line(), under the mutex. We really don't want open() to succeed before parse_chan_pair() had been done (or after it has failed, BTW). We also want "remove con<n>" to free irqs, etc., same as "config con<n>=none". Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
If two processes are opening the same line, the second to get into line_open() will decide that it doesn't need to do anything (correctly) or wait for anything. The latter, unfortunately, is incorrect - the first opener might not be through yet. We need to have exclusion covering the entire line_init(), including the blocking parts. Moreover, the next patch will need to widen the exclusion on mconsole side of things, also including the blocking bits, so let's just convert that sucker to mutex... Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
make line_setup() act on a separate array of conf strings + default conf, have lines array initialized explicitly by that data, bury LINE_INIT() macro from hell. Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Al Viro 提交于
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
由 Yong Zhang 提交于
Since commit [e58aa3d2: genirq: Run irq handlers with interrupts disabled], We run all interrupt handlers with interrupts disabled and we even check and yell when an interrupt handler returns with interrupts enabled (see commit [b738a50a: genirq: Warn when handler enables interrupts]). So now this flag is a NOOP and can be removed. Signed-off-by: NYong Zhang <yong.zhang0@gmail.com> Signed-off-by: NRichard Weinberger <richard@nod.at>
-
- 24 3月, 2012 2 次提交
-
-
由 Russell King 提交于
The sa1111 support will ioremap() the device; there is no need for platforms to setup a static mapping for this. Remove the static mapping for this device from badge4, jornada720 and neponset. Acked-by: NNicolas Pitre <nico@linaro.org> Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
-
由 Russell King 提交于
Use dev_err() to report device specific errors rather than printk(). Acked-by: NNicolas Pitre <nico@linaro.org> Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
-