There's an undesirable interaction with issuing MRA requests to
increase connection timeouts and the listen backlog.
When the rdma_cm receives a connection request, it queues an MRA with
the ib_cm. (The ib_cm will send an MRA if it receives a duplicate
REQ.) The rdma_cm will then create a new rdma_cm_id and give that to
the user, which in this case is the rdma_user_cm.
If the listen backlog maintained in the rdma_user_cm is full, it
destroys the rdma_cm_id, which in turns destroys the ib_cm_id. The
ib_cm_id generates a REJ because the state of the ib_cm_id has changed
to MRA sent, versus REQ received. When the backlog is full, we just
want to drop the REQ so that it is retried later.
Fix this by deferring queuing the MRA until after the user of the
rdma_cm has examined the connection request.
Signed-off-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Currently we hand off PAGE_SIZEd kmallocs to the page allocator in the
mistaken belief that the page allocator can handle these allocations
effectively. However, measurements indicate a minimum slowdown by the
factor of 8 (and that is only SMP, NUMA is much worse) vs the slub fastpath
which causes regressions in tbench.
Increase the number of kmalloc caches by one so that we again handle 4k
kmallocs directly from slub. 4k page buffering for the page allocator
will be performed by slub like done by slab.
At some point the page allocator fastpath should be fixed. A lot of the kernel
would benefit from a faster ability to allocate a single page. If that is
done then the 4k allocs may again be forwarded to the page allocator and this
patch could be reverted.
Reviewed-by: Pekka Enberg <penberg@cs.helsinki.fi>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Slub already has two ways of allocating an object. One is via its own
logic and the other is via the call to kmalloc_large to hand off object
allocation to the page allocator. kmalloc_large is typically used
for objects >= PAGE_SIZE.
We can use that handoff to avoid failing if a higher order kmalloc slab
allocation cannot be satisfied by the page allocator. If we reach the
out of memory path then simply try a kmalloc_large(). kfree() can
already handle the case of an object that was allocated via the page
allocator and so this will work just fine (apart from object
accounting...).
For any kmalloc slab that already requires higher order allocs (which
makes it impossible to use the page allocator fastpath!)
we just use PAGE_ALLOC_COSTLY_ORDER to get the largest number of
objects in one go from the page allocator slowpath.
On a 4k platform this patch will lead to the following use of higher
order pages for the following kmalloc slabs:
8 ... 1024 order 0
2048 .. 4096 order 3 (4k slab only after the next patch)
We may waste some space if fallback occurs on a 2k slab but we
are always able to fallback to an order 0 alloc.
Reviewed-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Currently we determine the gfp flags to pass to the page allocator
each time a slab is being allocated.
Determine the bits to be set at the time the slab is created. Store
in a new allocflags field and add the flags in allocate_slab().
Acked-by: Mel Gorman <mel@csn.ul.ie>
Reviewed-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
This adds a proper function for kmalloc page allocator pass-through. While it
simplifies any code that does slab tracing code a lot, I think it's a
worthwhile cleanup in itself.
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
- alloc_slabmgmt: initialize all slab fields in 1 place
- slab->nodeid was initialized twice: in alloc_slabmgmt
and immediately after it in cache_grow
Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
CC: Christoph Lameter <clameter@sgi.com>
Reviewed-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
The KVM configuration is no longer visible in the latest git tree. It looks
like it is selected by HAVE_SETUP_PER_CPU_AREA. I've moved HAVE_KVM to
under CONFIG_X86.
Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: Avi Kivity <avi@qumranet.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
the cpa API is page aligned - warn about any weird alignments.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
extern should not appear in C files. Also, the definitions
do not match the prototype currently, not sure what way you
want to go with this, I've switched the prototype to return
int, but I can see going to the void return as well.
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When the GART table is unmapped from the kernel direct mappings
during early bootup, make sure we have no leftover cachelines in it.
Note: the clflush done by set_memory_np() was not enough, because
clflush does not work on unmapped pages.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The EFI-runtime mapping code changed a larger memory area than it
should have, due to a pages/bytes parameter mixup.
noticed by Andi Kleen.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
dump_pagetable() can now become static.
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Jiri Kosina reported the following deadlock scenario with
show_unhandled_signals enabled:
[ 68.379022] gnome-settings-[2941] trap int3 ip:3d2c840f34
sp:7fff36f5d100 error:0<3>BUG: sleeping function called from invalid
context at kernel/rwsem.c:21
[ 68.379039] in_atomic():1, irqs_disabled():0
[ 68.379044] no locks held by gnome-settings-/2941.
[ 68.379050] Pid: 2941, comm: gnome-settings- Not tainted 2.6.25-rc1 #30
[ 68.379054]
[ 68.379056] Call Trace:
[ 68.379061] <#DB> [<ffffffff81064883>] ? __debug_show_held_locks+0x13/0x30
[ 68.379109] [<ffffffff81036765>] __might_sleep+0xe5/0x110
[ 68.379123] [<ffffffff812f2240>] down_read+0x20/0x70
[ 68.379137] [<ffffffff8109cdca>] print_vma_addr+0x3a/0x110
[ 68.379152] [<ffffffff8100f435>] do_trap+0xf5/0x170
[ 68.379168] [<ffffffff8100f52b>] do_int3+0x7b/0xe0
[ 68.379180] [<ffffffff812f4a6f>] int3+0x9f/0xd0
[ 68.379203] <<EOE>>
[ 68.379229] in libglib-2.0.so.0.1505.0[3d2c800000+dc000]
and tracked it down to:
commit 03252919b7
Author: Andi Kleen <ak@suse.de>
Date: Wed Jan 30 13:33:18 2008 +0100
x86: print which shared library/executable faulted in segfault etc. messages
the problem is that we call down_read() from an atomic context.
Solve this by returning from print_vma_addr() if the preempt count is
elevated. Update preempt_conditional_sti / preempt_conditional_cli to
unconditionally lift the preempt count even on !CONFIG_PREEMPT.
Reported-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
A CLOCK_REALTIME timer, which has an absolute expiry time less than
the clock realtime offset calls with a negative delta into the clock
events code and triggers the WARN_ON() there.
This is a false positive and needs to be prevented. Check the result
of timer->expires - timer->base->offset right away and return -ETIME
right away.
Thanks to Frans Pop, who reported the problem and tested the fixes.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Frans Pop <elendil@planet.nl>
Various user space callers ask for relative timeouts. While we fixed
that overflow issue in hrtimer_start(), the sites which convert
relative user space values to absolute timeouts themself were uncovered.
Instead of putting overflow checks into each place add a function
which does the sanity checking and convert all affected callers to use
it.
Thanks to Frans Pop, who reported the problem and tested the fixes.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Tested-by: Frans Pop <elendil@planet.nl>
mlx4_table_find (for FMR MPTs) requires that ICM memory already be
mapped. Before this fix, FMR allocation depended on ICM memory
already being mapped for the MPT entry. If all currently mapped
entries are taken, the find operation fails (even if the MPT ICM table
still had more entries, which were just not mapped yet).
This fix moves the mpt find operation to fmr_enable, to guarantee that
any required ICM memory mapping has already occurred.
Found by Oren Duer of Mellanox.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Currently mlx4_ib_fmr_alloc() calls mlx4_mr_enable() instead of
mlx4_fmr_enable(). The two functions are equivalent at the moment, but
this is not really correct (and the change is needed to fix a bug).
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
struct ipoib_cm_tx.ibwc is unused since commit 1b524963 ("IPoIB/cm:
Use common CQ for CM send completions"), so remove it.
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
In P_Key event handling, if the old P_Key is no longer available, the
driver must call ipoib_ib_dev_stop() -- just as it does when the P_Key
is still available (see procedure __ipoib_ib_dev_flush()).
When a P_Key becomes available, the driver will perform ipoib_open(),
which assumes that the QP is in RESET, the cm_id has been
destroyed/deleted, etc. If ipoib_ib_dev_stop() is not called as
described above, then these assumptions will be false, and the attempt
to bring the interface up will fail.
Found by Mellanox QA.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
CONFIG_DEVICE_TREE was the only user of CONFIG_WANT_DEVICE_TREE but
it was removed in commit id 2543133381
(bootwrapper: Build multiple cuImages).
This removes CONFIG_WANT_DEVICE_TREE from Kconfig and the defconfigs.
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: Paul Mackerras <paulus@samba.org>
arch/powerpc/platforms/cell/ras.c still has DEBUG #defined, which is no
longer necessary. Disable it - this disables two pr_debugs().
While we're there this file should have a copyright notice and license,
so add both.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This fixes the following section mismatch:
<-- snip -->
...
WARNING: vmlinux.o(.text+0x2fbca8): Section mismatch in reference from the function .hvc_rtas_init() to the function .devinit.text:.hvc_alloc()
...
<-- snip -->
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This fixes the following section mismatch:
<-- snip -->
...
WARNING: vmlinux.o(.text+0x55648): Section mismatch in reference from the function .free_node() to the function .init.text:.free_property()
...
<-- snip -->
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
This fixes the following section mismatches:
<-- snip -->
...
WARNING: vmlinux.o(.text+0xe49c): Section mismatch in reference from the function .vdso_do_func_patch64() to the function .init.text:.find_symbol64()
WARNING: vmlinux.o(.text+0xe4d0): Section mismatch in reference from the function .vdso_do_func_patch64() to the function .init.text:.find_symbol64()
WARNING: vmlinux.o(.text+0xe56c): Section mismatch in reference from the function .vdso_do_func_patch32() to the function .init.text:.find_symbol32()
WARNING: vmlinux.o(.text+0xe5a0): Section mismatch in reference from the function .vdso_do_func_patch32() to the function .init.text:.find_symbol32()
...
<-- snip -->
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Commit 2f569afd9c ("CONFIG_HIGHPTE vs.
sub-page page tables.") breaks compilation of arch/ppc since it
introduces the pgtable_t type which was not added to arch/ppc.
This adds the missing typedef.
Signed-off-by: Stefan Roese <sr@denx.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
PS3 firmware 1.94 added the source of power and reset events to the
payload of the system manager POWER_PRESSED and RESET_PRESSED events.
Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
General code cleanups for PS3 system-manager:
o Move all MODULE_ macros to bottom.
o Correct PS3_SM_WAKE_P_O_R value.
o Enhance comment on wakeup source values.
Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The PS3 Other OS boot flag is not checked when an LPAR reboot is done,
so the ps3-boot-game-os utility fails to reboot the system into the
Game OS. This fix changes the PS3 restart handler from requesting an
PS3_SM_NEXT_OP_LPAR_REBOOT to requesting an PS3_SM_NEXT_OP_SYS_REBOOT.
Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Fix a bug in the lv1_get_repository_node_value() routine of the PS3
bootwrapper. Changes in the PS3 system firmware 2.20 cause this bug
to hang the system when branching from the bootwrapper to the kernel
_start.
Since the video system has not yet been enabled at the time
the bug is hit, the system hangs with a blank screen. Earlier
firmwares don't cause such a catastrophic failure, and so this
bug went undetected.
Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
ps3_read_pm (pm_interval) should return an actual HW register value
because the pm_interval register is a counter register.
This removes the shadow pm_interval register.
Signed-off-by: Takashi Yamamoto <TakashiA.Yamamoto@jp.sony.com>
Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Fix the ps3_set_bookmark() routine of the PS3 logical performance
monitor driver.
To properly set a performance monitor bookmark the Cell processor
requires no instruction branches near the setting of the bookmark
SPR. Testing showed that the use of the db10cyc instruction did
not work correctly. This change replaces the db10cyc instruction
with 10 nop instructions.
Signed-off-by: Takashi Yamamoto <TakashiA.Yamamoto@jp.sony.com>
Signed-off-by: Geoff Levand <geoffrey.levand@am.sony.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
The powerpc show_regs prints CPU using smp_processor_id: change that to
raw_smp_processor_id, so that when it's showing a WARN_ON backtrace without
preemption disabled, DEBUG_PREEMPT doesn't mess up that warning with its own.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Acer Extensa 5220 -- OSI(Linux) is a NOP
Dell OptiPlex 755 -- OSI(Linux) turns GUSB into a NOP
Dell PowerEdge 1950 -- OSI(Linux) is a NOP
Dell Precision 690 -- OSI(Linux) touches USB (skips GUSB)
FSC ESPRIMO Mobile V5505 -- OSI(Linux) is a NOP
Lenovo LENOVO3000 V100 -- OSI(Linux) is a NOP
Lenovo X61x -- OSI(Linux) enables Linux specific AML
Sony Vaio VGN-NR11S_S - OSI(Linux) is a NOP
Signed-off-by: Len Brown <len.brown@intel.com>
sparse complains at drivers/pnp/pnpacpi/core.c:39 with the error:
Trying to use reserved word '__attribute__' as identifier
Expected ) in function declarator, got ".init.data"
and at drivers/pnp/pnpacpi/core.c:49:38 with the error:
undefined identifier 'excluded_id_list'
With the patch below these sparse complaints do not occur
Signed-off-by: Roel Kluin <12o3l@tiscali.nl>
Signed-off-by: Len Brown <len.brown@intel.com>
This acts as a reversion of 1c6b2ca5e0 in
the case of UP SH-4, where we still have the risk of a multiple hit
between the slow and fast paths. As seen on SH7780.
Signed-off-by: Hideo Saito <saito@densan.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
The _WAK global ACPI control method has to be called with the
argument representing the sleep state being exited. Make it happen.
Special thanks to Mirco Tischler <mt-ml@gmx.de> for reporting the
problem and debugging.
Reported-by: Mirco Tischler <mt-ml@gmx.de>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Len Brown <len.brown@intel.com>
Use generic_io_base to point out the pci io window, and make sure the
highest port address used is SH7751_PCI_IO_SIZE - 1.
This patch fixes pci io port access for the r2d boards - CONFIG_8139TOO_PIO
now works as expected. So does the alsa driver for CMI8738.
Signed-off-by: Magnus Damm <damm@igel.co.jp>
Acked-by: Katsuya MATSUBARA <matsu@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
This patch is a fix to make sure readsN/writesN are used over insN/outsN for
ioreadN_rep/iowriteN_rep.
The current state of the sh io code is that mmio operations like readN/writeN
and ioreadN/iowriteN are unaffected by the value of generic_io_base. This is
different fom port based io like inN/outN which gets adjusted using the value
in generic_io_base.
Without this patch ioreadN_rep/iowriteN_rep get their addresses adjusted.
The address for mmio access is adjusted using generic_io_base. This is wrong.
The ata core code currently crashes if generic_io_base is set.
This patch changes ioreadN_rep/iowriteN_rep to follow the same rules as the
rest of the mmio operations, ie don't adjust using generic_io_base.
Signed-off-by: Magnus Damm <damm@igel.co.jp>
Acked-by: Katsuya MATSUBARA <matsu@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
This patch makes sure ctrl_inN/outN are used instead of inN/outN for on chip
pci registers. Without this patch addresses may be adjusted using the value
in generic_io_base. This patch makes it possible to set generic_io_base and
have pci without reading and writing all over the place.
Signed-off-by: Magnus Damm <damm@igel.co.jp>
Acked-by: Katsuya MATSUBARA <matsu@igel.co.jp>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Sean Hefty