I found a clever way to make the extra IOPL switching invisible to
non-paravirt compiles - since kernel_rpl is statically defined to be zero
there, and only non-zero rpl kernel have a problem restoring IOPL, as popf
does not restore IOPL flags unless run at CPL-0.
Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
The VMI ROM has a mode where hypercalls can be queued and batched. This turns
out to be a significant win during context switch, but must be done at a
specific point before side effects to CPU state are visible to subsequent
instructions. This is similar to the MMU batching hooks already provided.
The same hooks could be used by the Xen backend to implement a context switch
multicall.
To explain a bit more about lazy modes in the paravirt patches, basically, the
idea is that only one of lazy CPU or MMU mode can be active at any given time.
Lazy MMU mode is similar to this lazy CPU mode, and allows for batching of
multiple PTE updates (say, inside a remap loop), but to avoid keeping some
kind of state machine about when to flush cpu or mmu updates, we just allow
one or the other to be active. Although there is no real reason a more
comprehensive scheme could not be implemented, there is also no demonstrated
need for this extra complexity.
Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Cc: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Convert the PDA code to use %fs rather than %gs as the segment for
per-processor data. This is because some processors show a small but
measurable performance gain for reloading a NULL segment selector (as %fs
generally is in user-space) versus a non-NULL one (as %gs generally is).
On modern processors the difference is very small, perhaps undetectable.
Some old AMD "K6 3D+" processors are noticably slower when %fs is used
rather than %gs; I have no idea why this might be, but I think they're
sufficiently rare that it doesn't matter much.
This patch also fixes the math emulator, which had not been adjusted to
match the changed struct pt_regs.
[frederik.deweerdt@gmail.com: fixit with gdb]
[mingo@elte.hu: Fix KVM too]
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Ian Campbell <Ian.Campbell@XenSource.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Zachary Amsden <zach@vmware.com>
Cc: Eric Dumazet <dada1@cosmosbay.com>
Signed-off-by: Frederik Deweerdt <frederik.deweerdt@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Fernando Lopez-Lezcano reported frequent scheduling latencies and audio
xruns starting at the 2.6.18-rt kernel, and those problems persisted all
until current -rt kernels. The latencies were serious and unjustified by
system load, often in the milliseconds range.
After a patient and heroic multi-month effort of Fernando, where he
tested dozens of kernels, tried various configs, boot options,
test-patches of mine and provided latency traces of those incidents, the
following 'smoking gun' trace was captured by him:
_------=> CPU#
/ _-----=> irqs-off
| / _----=> need-resched
|| / _---=> hardirq/softirq
||| / _--=> preempt-depth
|||| /
||||| delay
cmd pid ||||| time | caller
\ / ||||| \ | /
IRQ_19-1479 1D..1 0us : __trace_start_sched_wakeup (try_to_wake_up)
IRQ_19-1479 1D..1 0us : __trace_start_sched_wakeup <<...>-5856> (37 0)
IRQ_19-1479 1D..1 0us : __trace_start_sched_wakeup (c01262ba 0 0)
IRQ_19-1479 1D..1 0us : resched_task (try_to_wake_up)
IRQ_19-1479 1D..1 0us : __spin_unlock_irqrestore (try_to_wake_up)
...
<idle>-0 1...1 11us!: default_idle (cpu_idle)
...
<idle>-0 0Dn.1 602us : smp_apic_timer_interrupt (c0103baf 1 0)
...
<...>-5856 0D..2 618us : __switch_to (__schedule)
<...>-5856 0D..2 618us : __schedule <<idle>-0> (20 162)
<...>-5856 0D..2 619us : __spin_unlock_irq (__schedule)
<...>-5856 0...1 619us : trace_stop_sched_switched (__schedule)
<...>-5856 0D..1 619us : trace_stop_sched_switched <<...>-5856> (37 0)
what is visible in this trace is that CPU#1 ran try_to_wake_up() for
PID:5856, it placed PID:5856 on CPU#0's runqueue and ran resched_task()
for CPU#0. But it decided to not send an IPI that no CPU - due to
TS_POLLING. But CPU#0 never woke up after its NEED_RESCHED bit was set,
and only rescheduled to PID:5856 upon the next lapic timer IRQ. The
result was a 600+ usecs latency and a missed wakeup!
the bug turned out to be an idle-wakeup bug introduced into the mainline
kernel this summer via an optimization in the x86_64 tree:
commit 495ab9c045
Author: Andi Kleen <ak@suse.de>
Date: Mon Jun 26 13:59:11 2006 +0200
[PATCH] i386/x86-64/ia64: Move polling flag into thread_info_status
During some profiling I noticed that default_idle causes a lot of
memory traffic. I think that is caused by the atomic operations
to clear/set the polling flag in thread_info. There is actually
no reason to make this atomic - only the idle thread does it
to itself, other CPUs only read it. So I moved it into ti->status.
the problem is this type of change:
if (!hlt_counter && boot_cpu_data.hlt_works_ok) {
- clear_thread_flag(TIF_POLLING_NRFLAG);
+ current_thread_info()->status &= ~TS_POLLING;
smp_mb__after_clear_bit();
while (!need_resched()) {
local_irq_disable();
this changes clear_thread_flag() to an explicit clearing of TS_POLLING.
clear_thread_flag() is defined as:
clear_bit(flag, &ti->flags);
and clear_bit() is a LOCK-ed atomic instruction on all x86 platforms:
static inline void clear_bit(int nr, volatile unsigned long * addr)
{
__asm__ __volatile__( LOCK_PREFIX
"btrl %1,%0"
hence smp_mb__after_clear_bit() is defined as a simple compile barrier:
#define smp_mb__after_clear_bit() barrier()
but the explicit TS_POLLING clearing introduced by the patch:
+ current_thread_info()->status &= ~TS_POLLING;
is not an atomic op! So the clearing of the TS_POLLING bit is freely
reorderable with the reading of the NEED_RESCHED bit - and both now
reside in different memory addresses.
CPU idle wakeup very much depends on ordered memory ops, the clearing of
the TS_POLLING flag must always be done before we test need_resched()
and hit the idle instruction(s). [Symmetrically, the wakeup code needs
to set NEED_RESCHED before it tests the TS_POLLING flag, so memory
ordering is paramount.]
Fernando's dual-core Athlon64 system has a sufficiently advanced memory
ordering model so that it triggered this scenario very often.
( And it also turned out that the reason why these latencies never
triggered on my testsystems is that i routinely use idle=poll, which
was the only idle variant not affected by this bug. )
The fix is to change the smp_mb__after_clear_bit() to an smp_mb(), to
act as an absolute barrier between the TS_POLLING write and the
NEED_RESCHED read. This affects almost all idling methods (default,
ACPI, APM), on all 3 x86 architectures: i386, x86_64, ia64.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Tested-by: Fernando Lopez-Lezcano <nando@ccrma.Stanford.EDU>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
IOPL is implicitly saved and restored on task switch,
so explicit check is no longer needed.
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Currently the idle loop has two nested loops -- one high level
in cpu_idle and in some low level idle functions another one.
Looping in the low level idle functions breaks the idle notifiers
because interrupts waking up sleep states need to execute
exit_idle() which is only in cpu_idle().
So don't do that, only loop in cpu_idle(). This only removes
code.
In some cases e.g. poll_idle the idle loop is a little longer
now because cpu_idle checks more things. I hope that isn't a problem
ACPI idle doesn't change behaviour because it never looped anyways.
Cc: len.brown@intel.com
Cc: eranian@hpl.hp.com
Signed-off-by: Andi Kleen <ak@suse.de>
Use the pcurrent field in the PDA to implement the "current" macro. This ends
up compiling down to a single instruction to get the current task.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Chuck Ebbert <76306.1226@compuserve.com>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Jan Beulich <jbeulich@novell.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
There are a few places where the change in struct pt_regs and the use of %gs
affect the userspace ABI. These are primarily debugging interfaces where
thread state can be inspected or extracted.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Chuck Ebbert <76306.1226@compuserve.com>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Jan Beulich <jbeulich@novell.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
This patch is the meat of the PDA change. This patch makes several related
changes:
1: Most significantly, %gs is now used in the kernel. This means that on
entry, the old value of %gs is saved away, and it is reloaded with
__KERNEL_PDA.
2: entry.S constructs the stack in the shape of struct pt_regs, and this
is passed around the kernel so that the process's saved register
state can be accessed.
Unfortunately struct pt_regs doesn't currently have space for %gs
(or %fs). This patch extends pt_regs to add space for gs (no space
is allocated for %fs, since it won't be used, and it would just
complicate the code in entry.S to work around the space).
3: Because %gs is now saved on the stack like %ds, %es and the integer
registers, there are a number of places where it no longer needs to
be handled specially; namely context switch, and saving/restoring the
register state in a signal context.
4: And since kernel threads run in kernel space and call normal kernel
code, they need to be created with their %gs == __KERNEL_PDA.
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Chuck Ebbert <76306.1226@compuserve.com>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Jan Beulich <jbeulich@novell.com>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
i386 port of the sLeAZY-fpu feature. Chuck reports that this gives him a +/-
0.4% improvement on his simple benchmark
x86_64 description follows:
Right now the kernel on x86-64 has a 100% lazy fpu behavior: after *every*
context switch a trap is taken for the first FPU use to restore the FPU
context lazily. This is of course great for applications that have very
sporadic or no FPU use (since then you avoid doing the expensive save/restore
all the time). However for very frequent FPU users... you take an extra trap
every context switch.
The patch below adds a simple heuristic to this code: After 5 consecutive
context switches of FPU use, the lazy behavior is disabled and the context
gets restored every context switch. If the app indeed uses the FPU, the trap
is avoided. (the chance of the 6th time slice using FPU after the previous 5
having done so are quite high obviously).
After 256 switches, this is reset and lazy behavior is returned (until there
are 5 consecutive ones again). The reason for this is to give apps that do
longer bursts of FPU use still the lazy behavior back after some time.
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
The scheduler on Andreas Friedrich's hyperthreading system stopped
working properly: the scheduler would never move tasks to another CPU!
The lask known working kernel was 2.6.8.
After a couple of attempts to corner the bug, the following smoking gun
was found:
BIOS reported wrong ACPI idfor the processor
CPU#1: set_cpus_allowed(), swapper:1, 3 -> 2
[<c0103bbe>] show_trace_log_lvl+0x34/0x4a
[<c0103ceb>] show_trace+0x2c/0x2e
[<c01045f8>] dump_stack+0x2b/0x2d
[<c0116a77>] set_cpus_allowed+0x52/0xec
[<c0101d86>] cpu_idle_wait+0x2e/0x100
[<c0259c57>] acpi_processor_power_exit+0x45/0x58
[<c0259752>] acpi_processor_remove+0x46/0xea
[<c025c6fb>] acpi_start_single_object+0x47/0x54
[<c025cee5>] acpi_bus_register_driver+0xa4/0xd3
[<c04ab2d7>] acpi_processor_init+0x57/0x77
[<c01004d7>] init+0x146/0x2fd
[<c0103a87>] kernel_thread_helper+0x7/0x10
a quick look at cpu_idle_wait() shows how broken that code is
on i386: it changes the init task's affinity map but never
restores it ...
and because all userspace tasks get forked by init, they all
inherited that single-CPU affinity mask. x86_64 cloned this
bug too.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Andreas Friedrich <andreas.friedrich@fujitsu-siemens.com>
Cc: Wolfgang Erig <Wolfgang.Erig@fujitsu-siemens.com>
Cc: Andrew Morton <akpm@osdl.org>
Cc: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Jan convinced me that it was unnecessary because the assembly stubs do
this already on the stack.
Cc: jbeulich@novell.com
Signed-off-by: Andi Kleen <ak@suse.de>
Intel processors starting with the Core Duo support
support processor native C-state using the MWAIT instruction.
Refer: Intel Architecture Software Developer's Manual
http://www.intel.com/design/Pentium4/manuals/253668.htm
Platform firmware exports the support for Native C-state to OS using
ACPI _PDC and _CST methods.
Refer: Intel Processor Vendor-Specific ACPI: Interface Specification
http://www.intel.com/technology/iapc/acpi/downloads/302223.htm
With Processor Native C-state, we use 'MWAIT' instruction on the processor
to enter different C-states (C1, C2, C3). We won't use the special IO
ports to enter C-state and no SMM mode etc required to enter C-state.
Overall this will mean better C-state support.
One major advantage of using MWAIT for all C-states is, with this and
"treat interrupt as break event" feature of MWAIT, we can now get accurate
timing for the time spent in C1, C2, .. states.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Len Brown <len.brown@intel.com>
Always make sure RIP/EIP is 0 in the registers stored on the top
of the stack of a kernel thread. This makes sure the unwinder code
won't try a fallback but knows the stack has ended.
AK: this patch is a bit mysterious. in theory they should be terminated
anyways, but it seems to fix at least one crash. Anyways double termination
probably doesn't hurt.
Signed-off-by: Andi Kleen <ak@suse.de>
In some places, particularly drivers and __init code, the init utsns is the
appropriate one to use. This patch replaces those with a the init_utsname
helper.
Changes: Removed several uses of init_utsname(). Hope I picked all the
right ones in net/ipv4/ipconfig.c. These are now changed to
utsname() (the per-process namespace utsname) in the previous
patch (2/7)
[akpm@osdl.org: CIFS fix]
Signed-off-by: Serge E. Hallyn <serue@us.ibm.com>
Cc: Kirill Korotaev <dev@openvz.org>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Herbert Poetzl <herbert@13thfloor.at>
Cc: Andrey Savochkin <saw@sw.ru>
Cc: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
We allow for the fact that the guest kernel may not run in ring 0. This
requires some abstraction in a few places when setting %cs or checking
privilege level (user vs kernel).
This is Chris' [RFC PATCH 15/33] move segment checks to subarch, except rather
than using #define USER_MODE_MASK which depends on a config option, we use
Zach's more flexible approach of assuming ring 3 == userspace. I also used
"get_kernel_rpl()" over "get_kernel_cs()" because I think it reads better in
the code...
1) Remove the hardcoded 3 and introduce #define SEGMENT_RPL_MASK 3 2) Add a
get_kernel_rpl() macro, and don't assume it's zero.
And:
Clean up of patch for letting kernel run other than ring 0:
a. Add some comments about the SEGMENT_IS_*_CODE() macros.
b. Add a USER_RPL macro. (Code was comparing a value to a mask
in some places and to the magic number 3 in other places.)
c. Add macros for table indicator field and use them.
d. Change the entry.S tests for LDT stack segment to use the macros
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Andi Kleen <ak@suse.de>
And add proper CFI annotation to it which was previously
impossible. This prevents "stuck" messages by the dwarf2 unwinder
when reaching the top of a kernel stack.
Includes feedback from Jan Beulich
Cc: jbeulich@novell.com
Signed-off-by: Andi Kleen <ak@suse.de>
Recent changes in i386 __switch_to() have a misplaced closing
parenthesis causing an unlikely() to terminate early.
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Use thread info flags to track use of debug registers and IO bitmaps.
- add TIF_DEBUG to track when debug registers are active
- add TIF_IO_BITMAP to track when I/O bitmap is used
- modify __switch_to() to use the new TIF flags
Performance tested on Pentium II, ten runs of LMbench context switch
benchmark (smaller is better:)
before after
avg 3.65 3.39
min 3.55 3.33
Signed-off-by: Stephane Eranian <eranian@hpl.hp.com>
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Acked-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
During some profiling I noticed that default_idle causes a lot of
memory traffic. I think that is caused by the atomic operations
to clear/set the polling flag in thread_info. There is actually
no reason to make this atomic - only the idle thread does it
to itself, other CPUs only read it. So I moved it into ti->status.
Converted i386/x86-64/ia64 for now because that was the easiest
way to fix ACPI which also manipulates these flags in its idle
function.
Cc: Nick Piggin <npiggin@novell.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Len Brown <len.brown@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
These are the i386-specific pieces to enable reliable stack traces. This is
going to be even more useful once CFI annotations get added to he assembly
code, namely to entry.S.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The only user of get_wchan is the proc fs - and proc can't be built modular.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When kretprobe probes the schedule() function, if the probed process exits
then schedule() will never return, so some kretprobe instances will never
be recycled.
In this patch the parent process will recycle retprobe instances of the
probed function and there will be no memory leak of kretprobe instances.
Signed-off-by: bibo mao <bibo.mao@intel.com>
Cc: Masami Hiramatsu <hiramatu@sdl.hitachi.co.jp>
Cc: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
>commit 76381fee7e
>Author: Vincent Hanquez <vincent.hanquez@cl.cam.ac.uk>
>Date: Thu Jun 23 00:08:46 2005 -0700
>
> [PATCH] xen: x86_64: use more usermode macro
>
> Make use of the user_mode macro where it's possible. This is useful for Xen
> because it will need only to redefine only the macro to a hypervisor call.
I am of the opinion that the above changeset is incomplete, i.e. it missed
converting some previous uses of user_mode to user_mode_vm. While most of
them could be considered just cosmetical, at least the one in die_nmi
doesn't appear to be.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Cc: Vincent Hanquez <vincent.hanquez@cl.cam.ac.uk>
Cc: Zachary Amsden <zach@vmware.com>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Show first field of kernel version in register dumps like x86_64 does.
Changes output from e.g.:
(2.6.16-rc1)
to:
(2.6.16-rc1 #12)
Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
)
From: Al Viro <viro@ftp.linux.org.uk>
task_pt_regs() needs the same offset-by-8 to match copy_thread()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This adds an option to remove vm86 support under CONFIG_EMBEDDED. Saves
about 5k.
This version eliminates most of the #ifdefs of the previous version and
instead uses function stubs in vm86.h. Also, release_vm86_irqs is moved
from asm-i386/irq.h to a more appropriate home in vm86.h so that the stubs
can live together.
$ size vmlinux-baseline vmlinux-novm86
text data bss dec hex filename
2920821 523232 190652 3634705 377611 vmlinux-baseline
2916268 523100 190492 3629860 376324 vmlinux-novm86
Signed-off-by: Matt Mackall <mpm@selenic.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Remove the "temporary debugging check" which has managed to live for quite
some time, and is clearly unneeded. The mm can never be live at this point,
so clearly checking the LDT in the mm->context is redundant as well.
Signed-off-by: Zachary Amsden <zach@vmware.com>
Cc: "Seth, Rohit" <rohit.seth@intel.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
So some 486 processors do have CR4 register. Allow them to present it in
register dumps by using the old fault technique rather than testing processor
family.
Thanks to Maciej for noticing this.
Signed-off-by: Zachary Amsden <zach@vmware.com>
Cc: "Seth, Rohit" <rohit.seth@intel.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix a bug in kprobes that can cause an Oops or even a crash when a return
probe is installed on one of the following functions: sys_execve,
do_execve, load_*_binary, flush_old_exec, or flush_thread. The fix is to
remove the call to kprobe_flush_task() in flush_thread(). This fix has
been tested on all architectures for which the return-probes feature has
been implemented (i386, x86_64, ppc64, ia64). Please apply.
BACKGROUND
Up to now, we have called kprobe_flush_task() under two situations: when a
task exits, and when it execs. Flushing kretprobe_instances on exit is
correct because (a) do_exit() doesn't return, and (b) one or more
return-probed functions may be active when a task calls do_exit(). Neither
is the case for sys_execve() and its callees.
Initially, the mistaken call to kprobe_flush_task() on exec was harmless
because we put the "real" return address of each active probed function
back in the stack, just to be safe, when we recycled its
kretprobe_instance. When support for ppc64 and ia64 was added, this safety
measure couldn't be employed, and was eventually dropped even for i386 and
x86_64. sys_execve() and its callees were informally blacklisted for
return probes until this fix was developed.
Acked-by: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Signed-off-by: Jim Keniston <jkenisto@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Make some changes to the NEED_RESCHED and POLLING_NRFLAG to reduce
confusion, and make their semantics rigid. Improves efficiency of
resched_task and some cpu_idle routines.
* In resched_task:
- TIF_NEED_RESCHED is only cleared with the task's runqueue lock held,
and as we hold it during resched_task, then there is no need for an
atomic test and set there. The only other time this should be set is
when the task's quantum expires, in the timer interrupt - this is
protected against because the rq lock is irq-safe.
- If TIF_NEED_RESCHED is set, then we don't need to do anything. It
won't get unset until the task get's schedule()d off.
- If we are running on the same CPU as the task we resched, then set
TIF_NEED_RESCHED and no further action is required.
- If we are running on another CPU, and TIF_POLLING_NRFLAG is *not* set
after TIF_NEED_RESCHED has been set, then we need to send an IPI.
Using these rules, we are able to remove the test and set operation in
resched_task, and make clear the previously vague semantics of
POLLING_NRFLAG.
* In idle routines:
- Enter cpu_idle with preempt disabled. When the need_resched() condition
becomes true, explicitly call schedule(). This makes things a bit clearer
(IMO), but haven't updated all architectures yet.
- Many do a test and clear of TIF_NEED_RESCHED for some reason. According
to the resched_task rules, this isn't needed (and actually breaks the
assumption that TIF_NEED_RESCHED is only cleared with the runqueue lock
held). So remove that. Generally one less locked memory op when switching
to the idle thread.
- Many idle routines clear TIF_POLLING_NRFLAG, and only set it in the inner
most polling idle loops. The above resched_task semantics allow it to be
set until before the last time need_resched() is checked before going into
a halt requiring interrupt wakeup.
Many idle routines simply never enter such a halt, and so POLLING_NRFLAG
can be always left set, completely eliminating resched IPIs when rescheduling
the idle task.
POLLING_NRFLAG width can be increased, to reduce the chance of resched IPIs.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Con Kolivas <kernel@kolivas.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Run idle threads with preempt disabled.
Also corrected a bugs in arm26's cpu_idle (make it actually call schedule()).
How did it ever work before?
Might fix the CPU hotplugging hang which Nigel Cunningham noted.
We think the bug hits if the idle thread is preempted after checking
need_resched() and before going to sleep, then the CPU offlined.
After calling stop_machine_run, the CPU eventually returns from preemption and
into the idle thread and goes to sleep. The CPU will continue executing
previous idle and have no chance to call play_dead.
By disabling preemption until we are ready to explicitly schedule, this bug is
fixed and the idle threads generally become more robust.
From: alexs <ashepard@u.washington.edu>
PPC build fix
From: Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
MIPS build fix
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Most of these guys are simply not needed (pulled by other stuff
via asm-i386/hardirq.h). One that is not entirely useless is hilarious -
arch/i386/oprofile/nmi_timer_int.c includes linux/irq.h... as a way to
get linux/errno.h
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The pushf/popf in switch_to are ONLY used to switch IOPL. Making this
explicit in C code is more clear. This pushf/popf pair was added as a
bugfix for leaking IOPL to unprivileged processes when using
sysenter/sysexit based system calls (sysexit does not restore flags).
When requesting an IOPL change in sys_iopl(), it is just as easy to change
the current flags and the flags in the stack image (in case an IRET is
required), but there is no reason to force an IRET if we came in from the
SYSENTER path.
This change is the minimal solution for supporting a paravirtualized Linux
kernel that allows user processes to run with I/O privilege. Other
solutions require radical rewrites of part of the low level fault / system
call handling code, or do not fully support sysenter based system calls.
Unfortunately, this added one field to the thread_struct. But as a bonus,
on P4, the fastest time measured for switch_to() went from 312 to 260
cycles, a win of about 17% in the fast case through this performance
critical path.
Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Some more assembler cleanups I noticed along the way.
Signed-off-by: Zachary Amsden <zach@vmware.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Subtle fix: load_TLS has been moved after saving %fs and %gs segments to avoid
creating non-reversible segments. This could conceivably cause a bug if the
kernel ever needed to save and restore fs/gs from the NMI handler. It
currently does not, but this is the safest approach to avoiding fs/gs
corruption. SMIs are safe, since SMI saves the descriptor hidden state.
Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
i386 Inline asm cleanup. Use cr/dr accessor functions.
Also, a potential bugfix. Also, some CR accessors really should be volatile.
Reads from CR0 (numeric state may change in an exception handler), writes to
CR4 (flipping CR4.TSD) and reads from CR2 (page fault) prevent instruction
re-ordering. I did not add memory clobber to CR3 / CR4 / CR0 updates, as it
was not there to begin with, and in no case should kernel memory be clobbered,
except when doing a TLB flush, which already has memory clobber.
I noticed that page invalidation does not have a memory clobber. I can't find
a bug as a result, but there is definitely a potential for a bug here:
#define __flush_tlb_single(addr) \
__asm__ __volatile__("invlpg %0": :"m" (*(char *) addr))
Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
sys_get_thread_area does not memset to 0 its struct user_desc info before
copying it to user space... since sizeof(struct user_desc) is 16 while the
actual datas which are filled are only 12 bytes + 9 bits (across the
bitfields), there is a (small) information leak.
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
These days %gs is normally the TLS segment, so it's no longer zero. As
a result, we shouldn't just assume that %fs/%gs tend to be zero
together, but test them independently instead.
Also, fix setting of debug registers to use the "next" pointer instead
of "current". It so happens that the scheduler will have set the new
current pointer before calling __switch_to(), but that's just an
implementation detail.
I believe at least for seccomp it's worth to turn off the tsc, not just for
HT but for the L2 cache too. So it's up to you, either you turn it off
completely (which isn't very nice IMHO) or I recommend to apply this below
patch.
This has been tested successfully on x86-64 against current cogito
repository (i686 compiles so I didn't bother testing ;). People selling
the cpu through cpushare may appreciate this bit for a peace of mind.
There's no way to get any timing info anymore with this applied
(gettimeofday is forbidden of course). The seccomp environment is
completely deterministic so it can't be allowed to get timing info, it has
to be deterministic so in the future I can enable a computing mode that
does a parallel computing for each task with server side transparent
checkpointing and verification that the output is the same from all the 2/3
seller computers for each task, without the buyer even noticing (for now
the verification is left to the buyer client side and there's no
checkpointing, since that would require more kernel changes to track the
dirty bits but it'll be easy to extend once the basic mode is finished).
Eliminating a cold-cache read of the cr4 global variable will save one
cacheline during the tlb flush while making the code per-cpu-safe at the
same time. Thanks to Mikael Pettersson for noticing the tlb flush wasn't
per-cpu-safe.
The global tlb flush can run from irq (IPI calling do_flush_tlb_all) but
it'll be transparent to the switch_to code since the IPI won't make any
change to the cr4 contents from the point of view of the interrupted code
and since it's now all per-cpu stuff, it will not race. So no need to
disable irqs in switch_to slow path.
Signed-off-by: Andrea Arcangeli <andrea@cpushare.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Clean CPU states in order to reuse smp boot code for CPU hotplug.
Signed-off-by: Li Shaohua<shaohua.li@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Trival patch for CPU hotplug. In CPU identify part, only did cleaup for intel
CPUs. Need do for other CPUs if they support S3 SMP.
Signed-off-by: Li Shaohua<shaohua.li@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
(The i386 CPU hotplug patch provides infrastructure for some work which Pavel
is doing as well as for ACPI S3 (suspend-to-RAM) work which Li Shaohua
<shaohua.li@intel.com> is doing)
The following provides i386 architecture support for safely unregistering and
registering processors during runtime, updated for the current -mm tree. In
order to avoid dumping cpu hotplug code into kernel/irq/* i dropped the
cpu_online check in do_IRQ() by modifying fixup_irqs(). The difference being
that on cpu offline, fixup_irqs() is called before we clear the cpu from
cpu_online_map and a long delay in order to ensure that we never have any
queued external interrupts on the APICs. There are additional changes to s390
and ppc64 to account for this change.
1) Add CONFIG_HOTPLUG_CPU
2) disable local APIC timer on dead cpus.
3) Disable preempt around irq balancing to prevent CPUs going down.
4) Print irq stats for all possible cpus.
5) Debugging check for interrupts on offline cpus.
6) Hacky fixup_irqs() to redirect irqs when cpus go off/online.
7) play_dead() for offline cpus to spin inside.
8) Handle offline cpus set in flush_tlb_others().
9) Grab lock earlier in smp_call_function() to prevent CPUs going down.
10) Implement __cpu_disable() and __cpu_die().
11) Enable local interrupts in cpu_enable() after fixup_irqs()
12) Don't fiddle with NMI on dead cpu, but leave intact on other cpus.
13) Program IRQ affinity whilst cpu is still in cpu_online_map on offline.
Signed-off-by: Zwane Mwaikambo <zwane@linuxpower.ca>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch adds function-return probes to kprobes for the i386
architecture. This enables you to establish a handler to be run when a
function returns.
1. API
Two new functions are added to kprobes:
int register_kretprobe(struct kretprobe *rp);
void unregister_kretprobe(struct kretprobe *rp);
2. Registration and unregistration
2.1 Register
To register a function-return probe, the user populates the following
fields in a kretprobe object and calls register_kretprobe() with the
kretprobe address as an argument:
kp.addr - the function's address
handler - this function is run after the ret instruction executes, but
before control returns to the return address in the caller.
maxactive - The maximum number of instances of the probed function that
can be active concurrently. For example, if the function is non-
recursive and is called with a spinlock or mutex held, maxactive = 1
should be enough. If the function is non-recursive and can never
relinquish the CPU (e.g., via a semaphore or preemption), NR_CPUS should
be enough. maxactive is used to determine how many kretprobe_instance
objects to allocate for this particular probed function. If maxactive <=
0, it is set to a default value (if CONFIG_PREEMPT maxactive=max(10, 2 *
NR_CPUS) else maxactive=NR_CPUS)
For example:
struct kretprobe rp;
rp.kp.addr = /* entrypoint address */
rp.handler = /*return probe handler */
rp.maxactive = /* e.g., 1 or NR_CPUS or 0, see the above explanation */
register_kretprobe(&rp);
The following field may also be of interest:
nmissed - Initialized to zero when the function-return probe is
registered, and incremented every time the probed function is entered but
there is no kretprobe_instance object available for establishing the
function-return probe (i.e., because maxactive was set too low).
2.2 Unregister
To unregiter a function-return probe, the user calls
unregister_kretprobe() with the same kretprobe object as registered
previously. If a probed function is running when the return probe is
unregistered, the function will return as expected, but the handler won't
be run.
3. Limitations
3.1 This patch supports only the i386 architecture, but patches for
x86_64 and ppc64 are anticipated soon.
3.2 Return probes operates by replacing the return address in the stack
(or in a known register, such as the lr register for ppc). This may
cause __builtin_return_address(0), when invoked from the return-probed
function, to return the address of the return-probes trampoline.
3.3 This implementation uses the "Multiprobes at an address" feature in
2.6.12-rc3-mm3.
3.4 Due to a limitation in multi-probes, you cannot currently establish
a return probe and a jprobe on the same function. A patch to remove
this limitation is being tested.
This feature is required by SystemTap (http://sourceware.org/systemtap),
and reflects ideas contributed by several SystemTap developers, including
Will Cohen and Ananth Mavinakayanahalli.
Signed-off-by: Hien Nguyen <hien@us.ibm.com>
Signed-off-by: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Signed-off-by: Frederik Deweerdt <frederik.deweerdt@laposte.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Zachary Amsden