Add kprobe-booster support on ia64.

Kprobe-booster improves the performance of kprobes by eliminating single-step,
where possible.  Currently, kprobe-booster is implemented on x86 and x86-64.
This is an ia64 port.

On ia64, kprobe-booster executes a copied bundle directly, instead of single
stepping.  Bundles which have B or X unit and which may cause an exception
(including break) are not executed directly.  And also, to prevent hitting
break exceptions on the copied bundle, only the hindmost kprobe is executed
directly if several kprobes share a bundle and are placed in different slots.
Note: set_brl_inst() is used for preparing an instruction buffer(it does not
modify any active code), so it does not need any atomic operation.
Signed-off-by: NMasami Hiramatsu <mhiramat@redhat.com>
Cc: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: bibo,mao <bibo.mao@intel.com>
Cc: Rusty Lynch <rusty.lynch@intel.com>
Cc: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NTony Luck <tony.luck@intel.com>

The sys_getpid() and sys_set_tid_address() behavior changed from

	return current->tgid

to

	struct pid *pid;
	pid = current->pids[PIDTYPE_PID].pid;
	return pid->numbers[pid->level].nr;

But the fast system calls on ia64 still operate the old way.  Patch them
appropriately to let ia64 work with pid namespaces.  Besides, this is one more
step in deprecating of pid and tgid on task_struct.

The fsys_getppid() is to be patched as well, but its logic is much
more complex now, so I will make it later.

One thing I'm not 100% sure is the trick with the IA64_UPID_SHIFT.  On order
to access the pid->level's element of an array I have to perform the following
calculations

	pid + sizeof(struct upid) * pid->level

The problem is that ia64 can only multiply float point registers, while all
the offsets I have in code are in rXX ones.  Fortunately, the sizeof(struct
upid) is 32 bytes on ia64 (and is very unlikely to ever change), so the
calculations get simpler:

	pid + pid->level << 5

So, I introduce the IA64_UPID_SHIFT and use the shl instruction.  I also
looked at how gcc compiles the similar place and found that it makes it with
shift as well.  Is this OK to do so?

Tested with ski emulator with 2.6.24 kernel, but fits 2.6.25-rc4 and
2.6.25-rc4-mm1 as well.
Signed-off-by: NPavel Emelyanov <xemul@openvz.org>
Cc: David Mosberger-Tang <davidm@hpl.hp.com>
Cc: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: Amy Griffis <amy.griffis@hp.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NTony Luck <tony.luck@intel.com>

In arch/ia64/sn/kernel/xpc_{main,partition}.c
Signed-off-by: NRobert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: NTony Luck <tony.luck@intel.com>

do_each_thread/while_each_thread is a double loop, so
should use 'goto' rather than 'break' to break out
the loop.
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

One should normally unlock in the reverse order of the lock calls,
and in this case there certainly is no reason not to.
Signed-off-by: NAlan D. Brunelle <alan.brunelle@hp.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Fix indenting of switch statement to follow Documentation/CodingStyle.
Signed-off-by: NHidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

While it is convenient that we can invoke kdump by asserting INIT
via button on chassis etc., there are some situations that invoking
kdump on fatal MCA is not welcomed rather than rebooting fast without
dump.

This patch adds a new flag 'kdump_on_fatal_mca' that is independent
from 'kdump_on_init' currently available.  Adding this flag enable
us to turning on/off of kdump depend on the event, INIT and/or fatal
MCA.  Default for this flag is to take the dump.
Signed-off-by: NHidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

ia64 named their handler kprobes_fault_handler while all other
arches used kprobe_fault_handler.  Change the function definition
and header declaration.
Signed-off-by: NHarvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

When EFI_DEBUG is defined to a non-zero value in arch/ia64/kernel/efi.c,
the efi memory regions are displayed. This patch enhances the
display code in a few ways:

1. Use TB, GB and MB as well as KB as units.
   Although this introduces rounding errors (KB doesn't as
   size is always a multiple of 4Kb), it does make
   things a lot more readable.

   Also as the range is also shown, it is possible to note the exact size
   if it is important. In my experience, the size field is mostly useful
   for getting a general idea of the size of a region.

   On the rx2620 that I use, there actually is an 8TB region (though not
   backed by physical memory, and 8TB really is a lot more readable than
   8589934592KB.

2. pad the size field with leading spaces to further improve readability

   ...
   ... (   8MB)
   ... ( 928MB)
   ... (   3MB)
   ...

   vs

   ...
   ... (8MB)
   ... (928MB)
   ... (3MB)
   ...

3. Pad the attr field out to 64bits using leading zeros,
   to further improve readability.

   ...
   mem05: type= 2, attr=0x0000000000000008, range=[0x0000000004000000-0x000000000481f000) (   8MB)
   mem06: type= 7, attr=0x0000000000000008, range=[0x000000000481f000-0x000000003e876000) ( 928MB)
   mem07: type= 5, attr=0x8000000000000008, range=[0x000000003e876000-0x000000003eb8e000) (   3MB)
   mem08: type= 4, attr=0x0000000000000008, range=[0x000000003eb8e000-0x000000003ee7a000) (   2MB)
   ...

   ...
   mem05: type= 2, attr=0x8, range=[0x0000000004000000-0x000000000481f000) (   8MB)
   mem06: type= 7, attr=0x8, range=[0x000000000481f000-0x000000003e876000) ( 928MB)
   mem07: type= 5, attr=0x8000000000000008, range=[0x000000003e876000-0x000000003eb8e000) (   3MB)
   mem08: type= 4, attr=0x8, range=[0x000000003eb8e000-0x000000003ee7a000) (   2MB)
   ...

4. Use %d instead of %u for the index field, as i is a signed int.

N.B: This code is not compiled unless EFI_DEBUG is non 0.
Signed-off-by: NSimon Horman <horms@verge.net.au>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Convert quirk printks to dev_printk().
Signed-off-by: NBjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

__FUNCTION__ is gcc-specific, use __func__

Long lines have been kept where they exist, some small spacing changes
have been done.
Signed-off-by: NHarvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Compilation of 2.6.25-rc2-mm1 on ia64 generates many warnings.

IA64 support 2 ELF format (IA64 binary and IA32 binary),
thus if 2 elf related header included, cause many warning or error.

about 2 week ago, J. Bruce Fields proposed this problem fixed patch.
(http://marc.info/?l=linux-ia64&m=120329313305695&w=2)
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: NKOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NTony Luck <tony.luck@intel.com>

When !CONFIG_SMP, cpu_physical_id() is ia64_get_lid(), which is
functionally identical to

    (ia64_getreg(_IA64_REG_CR_LID) >> 16) & 0xffff

so there's no need for two versions of this code.
Signed-off-by: NBjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Remove duplicate code, clean up goto's and indentation.
Signed-off-by: NPetr Tesarik <ptesarik@suse.cz>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Remove all code which does exactly the same thing as ptrace_request().
Signed-off-by: NPetr Tesarik <ptesarik@suse.cz>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Convert sys_ptrace() to arch_ptrace().
Signed-off-by: NPetr Tesarik <ptesarik@suse.cz>
Signed-off-by: NTony Luck <tony.luck@intel.com>

find_thread_for_addr() is no longer needed.  It was only used to find
the correct kernel RBS for a given memory address, but since the kernel
RBS is not needed any longer, this function can go away.
Signed-off-by: NPetr Tesarik <ptesarik@suse.cz>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Syncing is no longer needed, because user RBS is already
up-to-date.  Actually, if a debugger modified the contents
of the original RBS prior to changing PT_AR_BSP, the
modifications would get overwritten.
Signed-off-by: NPetr Tesarik <ptesarik@suse.cz>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Because the user RBS of a process is now completely stored in
user-mode when the process is ptrace-stopped, accesses to the
RBS should no longer augment any part of the kernel RBS.

This means we can get rid of most ia64_peek() and ia64_poke()
calls.
Signed-off-by: NPetr Tesarik <ptesarik@suse.cz>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Add CONFIG_HAVE_KRETPROBES to the arch/<arch>/Kconfig file for relevant
architectures with kprobes support.  This facilitates easy handling of
in-kernel modules (like samples/kprobes/kretprobe_example.c) that depend on
kretprobes being present in the kernel.

Thanks to Sam Ravnborg for helping make the patch more lean.

Per Mathieu's suggestion, added CONFIG_KRETPROBES and fixed up dependencies.
Signed-off-by: NAnanth N Mavinakayanahalli <ananth@in.ibm.com>
Acked-by: NMathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Acked-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This patch fixes the following compile error with a recent gcc:
  CC      kernel/kprobes.o
/home/bunk/linux/kernel-2.6/git/linux-2.6/kernel/kprobes.c:1066: error: __ksymtab_jprobe_return causes a section type conflict
Signed-off-by: NAdrian Bunk <bunk@kernel.org>
Signed-off-by: NTony Luck <tony.luck@intel.com>

This fixes regression introduced in 113134fc

Intel Tiger platforms hang when calling SAL_GET_PHYSICAL_ID_INFO
instead of properly returning -1 for unimplemented, so add a
version check.

SGI Altix platforms have an incorrect SAL version hard-coded into
their prom -- they encode 2.9, but actually implement 3.2 -- so
fix it up and allow ia64_sal_get_physical_id_info to keep
working.
Signed-off-by: NAlex Chiang <achiang@hp.com>
Acked-by: NRuss Anderson <rja@sgi.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

This patch moves the default ia64 defconfig to
arch/ia64/configs/generic_defconfig where it belongs and selects it as
the default defconfig through KBUILD_DEFCONFIG.
Signed-off-by: NAdrian Bunk <adrian.bunk@movial.fi>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Fix the problem that the following error message is sometimes displayed
at irq migration when vector domain is enabled.

    "Unexpected interrupt vector %d on CPU %d is not mapped to any IRQ!"

The cause of this problem is an interrupt is sent to the previous
target CPU after cleaning up vector to irq mapping table. To clean up
vector to irq map on the previous target CPU safty, change the irq
migration in multiple vector domain as follows. The original idea is
from x86 interrupt management code.

    - Delay vector to irq table cleanup until the interrupts are sent
      to new target CPUs. By this, it is ensured that target CPU is
      completely changed on the interrupt controller side.

    - Even after the interrupts are sent to new target CPUs, there can
      be pended interrupts remaining on the previous target CPU. So we
      need to delay clearning up vector to irq table until the pended
      interrupt is handled. For this, send IPI to the previous target
      CPU with lower priority vector and clean up vector to irq table
      in its handler.

This patch affects only to irq migration code with multiple vector
domain is enabled.
Signed-off-by: NKenji Kaneshige <kaneshige.kenji@jp.fujitsu.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

The similar check has been added to x86_32(i386) in commit
id 83bd0102.
So we add this check to ia64_ia32 and improve it a liitle bit in that
we need to check for stack overflow only when the signal is on stack.
Signed-off-by: NShi Weihua <shiwh@cn.fujitsu.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

The similar check has been added to x86_32(i386) in commit
id 83bd0102.
So we add this check to ia64 and improve it a liitle bit in that
we need to check for stack overflow only when the signal is on stack.
Signed-off-by: NShi Weihua <shiwh@cn.fujitsu.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Auto select CONFIG_NUMA and CONFIG_ACPI_NUMA when picking SN2, similar
to how they are selected automatically for CONFIG_IA64_GENERIC.
Signed-off-by: NJes Sorensen <jes@sgi.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Commit bdc80787 broke the build
for this config because the sim_defconfig selects CONFIG_HZ=250
but include/asm-ia64/param.h has an ifdef for the simulator to
force HZ to 32.  So we ended up with a kernel/timeconst.h set
for HZ=250 ... which then failed the check for the right HZ
value and died with:

Drop the #ifdef magic from param.h and make force CONFIG_HZ=32
directly for the simulator.
Signed-off-by: NTony Luck <tony.luck@intel.com>

We want to allow different implementations of pci_raw_ops for standard
and extended config space on x86.  Rather than clutter generic code with
knowledge of this, we make pci_raw_ops private to x86 and use it to
implement the new raw interface -- raw_pci_read() and raw_pci_write().
Signed-off-by: NMatthew Wilcox <willy@linux.intel.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

To allow flexible configuration of IDE introduce HAVE_IDE.
All archs except arm, um and s390 unconditionally select it.
For arm the actual configuration determine if IDE is supported.

This is a step towards introducing drivers/Kconfig for arm.
Signed-off-by: NSam Ravnborg <sam@ravnborg.org>
Acked-by: NRussell King - ARM Linux <linux@arm.linux.org.uk>
Acked-by: NBartlomiej Zolnierkiewicz <bzolnier@gmail.com>

The MCA code allocates bootmem memory for NR_CPUS, regardless
of how many cpus the system actually has.  This change allocates
memory only for cpus that actually exist.

On my test system with NR_CPUS = 1024, reserved memory was reduced by 130944k.

Before: Memory: 27886976k/28111168k available (8282k code, 242304k reserved, 5928k data, 1792k init)
After:  Memory: 28017920k/28111168k available (8282k code, 111360k reserved, 5928k data, 1792k init)
Signed-off-by: NRuss Anderson <rja@sgi.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

This is just Venki's patch[*] for x86 ported to ia64.

* http://marc.info/?l=linux-kernel&m=120249201318159&w=2Acked-by: NVenkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

When attaching to a stopped process, the RSE must be explicitly
synced to user-space, so the debugger can read the correct values.
Signed-off-by: NPetr Tesarik <ptesarik@suse.cz>
CC: Roland McGrath <roland@redhat.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

This is base kernel patch for ptrace RSE bug. It's basically a backport
from the utrace RSE patch I sent out several weeks ago. please review.

when a thread is stopped (ptraced), debugger might change thread's user
stack (change memory directly), and we must avoid the RSE stored in
kernel to override user stack (user space's RSE is newer than kernel's
in the case). To workaround the issue, we copy kernel RSE to user RSE
before the task is stopped, so user RSE has updated data.  we then copy
user RSE to kernel after the task is resummed from traced stop and
kernel will use the newer RSE to return to user.
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Signed-off-by: NPetr Tesarik <ptesarik@suse.cz>
CC: Roland McGrath <roland@redhat.com>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Since the RSE synchronization will need a TIF_ flag, but all

work-to-be-done bits are already used, so we have to multiplex
TIF_NOTIFY_RESUME again.
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Signed-off-by: NPetr Tesarik <ptesarik@suse.cz>
Signed-off-by: NTony Luck <tony.luck@intel.com>

Add ia64 hooks for the new syscalls that were added in
commit 4d672e7aSigned-off-by: NTony Luck <tony.luck@intel.com>

Fix typo in comments.

BTW: I have to fix coding style in arch/ia64/kernel/time.c also, otherwise
checkpatch.pl will be complaining.
Signed-off-by: NLi Zefan <lizf@cn.fujitsu.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This patch fixes the configuration dependencies in the vmcoreinfo data.

i386's "node_data" is defined in arch/x86/mm/discontig_32.c,
and x86_64's one is defined in arch/x86/mm/numa_64.c.
They depend on CONFIG_NUMA:
  arch/x86/mm/Makefile_32:7
    obj-$(CONFIG_NUMA) += discontig_32.o
  arch/x86/mm/Makefile_64:7
    obj-$(CONFIG_NUMA) += numa_64.o

ia64's "pgdat_list" is defined in arch/ia64/mm/discontig.c,
and it depends on CONFIG_DISCONTIGMEM and CONFIG_SPARSEMEM:
  arch/ia64/mm/Makefile:9-10
    obj-$(CONFIG_DISCONTIGMEM) += discontig.o
    obj-$(CONFIG_SPARSEMEM)    += discontig.o

ia64's "node_memblk" is defined in arch/ia64/mm/numa.c,
and it depends on CONFIG_NUMA:
  arch/ia64/mm/Makefile:8
    obj-$(CONFIG_NUMA)         += numa.o
Signed-off-by: NKen'ichi Ohmichi <oomichi@mxs.nes.nec.co.jp>
Acked-by: NSimon Horman <horms@verge.net.au>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This patchset is for the vmcoreinfo data.

The vmcoreinfo data has the minimum debugging information only for dump
filtering.  makedumpfile (dump filtering command) gets it to distinguish
unnecessary pages, and makedumpfile creates a small dumpfile.

This patch:

VMCOREINFO_SIZE() should be renamed VMCOREINFO_STRUCT_SIZE() since it's always
returning the size of the struct with a given name. This change would allow
VMCOREINFO_TYPEDEF_SIZE() to simply become VMCOREINFO_SIZE() since it need not
be used exclusively for typedefs.

This discussion is the following:
http://www.ussg.iu.edu/hypermail/linux/kernel/0709.3/0582.htmlSigned-off-by: NKen'ichi Ohmichi <oomichi@mxs.nes.nec.co.jp>
Acked-by: NDavid Rientjes <rientjes@google.com>
Acked-by: NSimon Horman <horms@verge.net.au>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This patchset adds a flags variable to reserve_bootmem() and uses the
BOOTMEM_EXCLUSIVE flag in crashkernel reservation code to detect collisions
between crashkernel area and already used memory.

This patch:

Change the reserve_bootmem() function to accept a new flag BOOTMEM_EXCLUSIVE.
If that flag is set, the function returns with -EBUSY if the memory already
has been reserved in the past.  This is to avoid conflicts.

Because that code runs before SMP initialisation, there's no race condition
inside reserve_bootmem_core().

[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: fix powerpc build]
Signed-off-by: NBernhard Walle <bwalle@suse.de>
Cc: <linux-arch@vger.kernel.org>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>