Original patch from Mikulas Patocka <mpatocka@redhat.com>
Mike Anderson was doing an OLTP benchmark on a computer with 48 physical
disks mapped to one logical device via device mapper.
He found that there was a slowdown on request_queue->lock in function
generic_unplug_device. The slowdown is caused by the fact that when some
code calls unplug on the device mapper, device mapper calls unplug on all
physical disks. These unplug calls take the lock, find that the queue is
already unplugged, release the lock and exit.
With the below patch, performance of the benchmark was increased by 18%
(the whole OLTP application, not just block layer microbenchmarks).
So I'm submitting this patch for upstream. I think the patch is correct,
because when more threads call simultaneously plug and unplug, it is
unspecified, if the queue is or isn't plugged (so the patch can't make
this worse). And the caller that plugged the queue should unplug it
anyway. (if it doesn't, there's 3ms timeout).
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
generic_file_splice_write() duplicates remove_suid() just because it
doesn't hold i_mutex. But it grabs i_mutex inside splice_from_pipe()
anyway, so this is rather pointless.
Move locking to generic_file_splice_write() and call remove_suid() and
__splice_from_pipe() instead.
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
put_io_context() drops the RCU read lock before calling into cfq_dtor(),
however we need to hold off freeing there before grabbing and
dereferencing the first object on the list.
So extend the rcu_read_lock() scope to cover the calling of cfq_dtor(),
and optimize cfq_free_io_context() to use a new variant for
call_for_each_cic() that assumes the RCU read lock is already held.
Hit in the wild by Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
For most initialization purposes, calling blk_queue_init_tags() without
the queue lock held is OK. Only if called for resizing an existing map
must the lock be held. Ditto for tag cleanup, the maps are reference
counted.
So switch the general queue flag setting to the unlocked variant, but
retain the locked variant for resizing.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Use the existing arch_alloc_page/arch_free_page callbacks to do
the guest page state transitions between stable and unused.
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
The wrong function got coverted ;)
CC drivers/s390/char/tty3270.o
drivers/s390/char/tty3270.c:1747:
warning: initialization from incompatible pointer type
Acked-by: Alan Cox <alan@redhat.com>
Cc: Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
This removes redundant arch code for generic ptrace requests
already handled by ptrace_request and compat_ptrace_request.
It simplifies things to just have the standard entry points,
and use the generic compat_sys_ptrace.
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Fix the following compile warning:
drivers/s390/s390mach.c: In function 's390_collect_crw_info':
drivers/s390/s390mach.c:77: warning: ignoring return value of 'down_interruptibl
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
New format cssid.ssid.devno is now parsed correctly.
Signed-off-by: Michael Ernst <mernst@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
The only sporadically used CIO_DEBUG messages are replaced by ordinary
CIO_MSG_EVENT messages. The CIO_MSG_EVENT messages debug levels are
consolidated.
Signed-off-by: Michael Ernst <mernst@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
From: Martin Schwidefsky <schwidefsky@de.ibm.com>
This patch fixes a bug with cpu bound guest on kvm-s390. Sometimes it
was impossible to deliver a signal to a spinning guest. We used
preemption as a circumvention. The preemption notifiers called
vcpu_load, which checked for pending signals and triggered a host
intercept. But even with preemption, a sigkill was not delivered
immediately.
This patch changes the low level host interrupt handler to check for the
SIE instruction, if TIF_WORK is set. In that case we change the
instruction pointer of the return PSW to rerun the vcpu_run loop. The kvm
code sees an intercept reason 0 if that happens. This patch adds accounting
for these types of intercept as well.
The advantages:
- works with and without preemption
- signals are delivered immediately
- much better host latencies without preemption
Acked-by: Carsten Otte <cotte@de.ibm.com>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
On return from syscall or interrupt, we have to check if we return to
userspace (likely) and if there is work todo (less likely) to decide
if we handle the work. We can optimize this check: we first check for
the less likely work case and then check for userspace.
This patch is also a preparation for an additional patch, that fixes a bug
in KVM dealing with cpu bound guests.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Concurrency isn't a big deal here since we have requests in flight
at this point, but do the locked variant to set a better example.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Remove module will not free L1 memory used which caused by
memory access after free. This patch fixes it.
Signed-off-by: Meihui Fan <mhfan@hhcn.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
When transferring to IRQ5 from an exception, save SYSCFG in memory across the
transfer and clear the trace bit.
When we get a single step exception, check whether we can safely clear the
trace bit in SYSCFG. We can (and should) clear it after the first instruction
of the interrupt handler; the first insn saves SYSCFG to the stack in all
handlers.
Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
When delivering a signal, disable single stepping but call
ptrace_notify if it was enabled before. The idea was taken
from the x86 port.
Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
Fix some really ancient code that was correct only for the m68k port.
Delete unused (i.e. copied from m68k) entries in asm-offsets.c.
Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
In the double fault handler, set up the PT_RETI slot so that
we print out the correct return address in the dumping code.
Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
The following cleanup patch:
add __user markings to a few userspace system functions
mysteriously added a "&" operator that doesn't belong in there, breaking the
atomic sections code.
Signed-off-by: Bernd Schmidt <bernds_cb1@t-online.de>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
This reverts commit 22eecde2f9. Uli
reports that it breaks UML on x86-64 with the Fedora 8 gcc (gcc 4.1.2),
causing a crash on startup. See
http://marc.info/?l=linux-kernel&m=121011722806093&w=2
for a trace.
Reported-by: Ulrich Drepper <drepper@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently, iw_cxgb3 is severely limited on the amount of userspace
memory that can be registered in in a single memory region, which
causes big problems for applications that expect to be able to
register 100s of MB.
The problem is that the driver uses a single kmalloc()ed buffer to
hold the physical buffer list (PBL) for the entire memory region
during registration, which means that 8 bytes of contiguous memory are
required for each page of memory being registered. For example, a 64
MB registration will require 128 KB of contiguous memory with 4 KB
pages, and it unlikely that such an allocation will succeed on a busy
system.
This is purely a driver problem: the temporary page list buffer is not
needed by the hardware, so we can fix this by writing the PBL to the
hardware in page-sized chunks rather than all at once. We do this by
splitting the memory registration operation up into several steps:
- Allocate PBL space in adapter memory for the full registration
- Copy PBL to adapter memory in chunks
- Allocate STag and enable memory region
This also allows several other cleanups to the __cxio_tpt_op()
interface and related parts of the driver.
This change leaves the reregister memory region and memory window
operations broken, but they already didn't work due to other
longstanding bugs, so fixing them will be left to a later patch.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Was a holdover from the old kernel_thread based cifsd
code. We needed to know that the thread had set the task variable
before proceeding. Now that kthread_run returns the new task, this
doesn't appear to be needed anymore.
As best I can tell, this sleep was intended to try to prevent
cifs_umount from freeing the cifsSesInfo struct before cifsd had
exited. Now that cifsd is using the kthread API, we know that
when kthread_stop returns that cifsd has exited, so I don't
think this is needed any longer.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Acked-by: Christop Hellwig <hch@infradead.org>
Signed-off-by: Steve French <sfrench@us.ibm.com>
We die because we forget to convert initrd_start and
initrd_end to virtual addresses.
Reported by Mikael Pettersson
Signed-off-by: David S. Miller <davem@davemloft.net>
Sparc doesn't have some of the OF interfaces this driver
wants to use.
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Current iw_cxgb3 code adds PBL memory to the driver's gen_pool in 2 MB
chunks. This limits the largest single allocation that can be done to
the same size, which means that with 4 KB pages, each of which takes 8
bytes of PBL memory, the largest memory region that can be allocated
is 1 GB (256K PBL entries * 4 KB/entry).
Remove this limit by adding all the PBL memory in a single gen_pool
chunk, if possible. Add code that falls back to smaller chunks if
gen_pool_add() fails, which can happen if there is not sufficient
contiguous lowmem for the internal gen_pool bitmap.
Signed-off-by: Roland Dreier <rolandd@cisco.com>
Commit 33dcdac2df ("kill ->put_inode")
removed the final use of i_op->put_inode, but left the now totally
unused "op" variable in iput().
Get rid of it.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix warning from pmd_bad() at bootup on a HIGHMEM64G HIGHPTE x86_32.
That came from 9fc34113f6 x86: debug pmd_bad();
but we understand now that the typecasting was wrong for PAE in the previous
version: pagetable pages above 4GB looked bad and stopped Arjan from booting.
And revert that cded932b75 x86: fix pmd_bad
and pud_bad to support huge pages. It was the wrong way round: we shouldn't
weaken every pmd_bad and pud_bad check to let huge pages slip through - in
part they check that we _don't_ have a huge page where it's not expected.
Put the x86 pmd_bad() and pud_bad() definitions back to what they have long
been: they can be improved (x86_32 should use PTE_MASK, to stop PAE thinking
junk in the upper word is good; and x86_64 should follow x86_32's stricter
comparison, to stop thinking any subset of required bits is good); but that
should be a later patch.
Fix Hans' good observation that follow_page() will never find pmd_huge()
because that would have already failed the pmd_bad test: test pmd_huge in
between the pmd_none and pmd_bad tests. Tighten x86's pmd_huge() check?
No, once it's a hugepage entry, it can get quite far from a good pmd: for
example, PROT_NONE leaves it with only ACCESSED of the KERN_PGTABLE bits.
However... though follow_page() contains this and another test for huge
pages, so it's nice to keep it working on them, where does it actually get
called on a huge page? get_user_pages() checks is_vm_hugetlb_page(vma) to
to call alternative hugetlb processing, as does unmap_vmas() and others.
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Earlier-version-tested-by: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Jeff Chua <jeff.chua.linux@gmail.com>
Cc: Hans Rosenfeld <hans.rosenfeld@amd.com>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
fcntl_setlk()/close() race prevention has a subtle hole - we need to
make sure that if we *do* have an fcntl/close race on SMP box, the
access to descriptor table and inode->i_flock won't get reordered.
As it is, we get STORE inode->i_flock, LOAD descriptor table entry vs.
STORE descriptor table entry, LOAD inode->i_flock with not a single
lock in common on both sides. We do have BKL around the first STORE,
but check in locks_remove_posix() is outside of BKL and for a good
reason - we don't want BKL on common path of close(2).
Solution is to hold ->file_lock around fcheck() in there; that orders
us wrt removal from descriptor table that preceded locks_remove_posix()
on close path and we either come first (in which case eviction will be
handled by the close side) or we'll see the effect of close and do
eviction ourselves. Note that even though it's read-only access,
we do need ->file_lock here - rcu_read_lock() won't be enough to
order the things.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
And with that last patch to affs killing the last put_inode instance we
can finally, after many years of transition kill this racy and awkward
interface.
(It's kinda funny that even the description in
Documentation/filesystems/vfs.txt was entirely wrong..)
Also remove a very misleading comment above the defintion of
struct super_operations.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Jens Axboe