Age | Commit message (Collapse) | Author | Files | Lines |
|
We encountered workloads that have very long wake up list on large
systems. A waker takes a long time to traverse the entire wake list and
execute all the wake functions.
We saw page wait list that are up to 3700+ entries long in tests of
large 4 and 8 socket systems. It took 0.8 sec to traverse such list
during wake up. Any other CPU that contends for the list spin lock will
spin for a long time. It is a result of the numa balancing migration of
hot pages that are shared by many threads.
Multiple CPUs waking are queued up behind the lock, and the last one
queued has to wait until all CPUs did all the wakeups.
The page wait list is traversed with interrupt disabled, which caused
various problems. This was the original cause that triggered the NMI
watch dog timer in: https://patchwork.kernel.org/patch/9800303/ . Only
extending the NMI watch dog timer there helped.
This patch bookmarks the waker's scan position in wake list and break
the wake up walk, to allow access to the list before the waker resume
its walk down the rest of the wait list. It lowers the interrupt and
rescheduling latency.
This patch also provides a performance boost when combined with the next
patch to break up page wakeup list walk. We saw 22% improvement in the
will-it-scale file pread2 test on a Xeon Phi system running 256 threads.
[ v2: Merged in Linus' changes to remove the bookmark_wake_function, and
simply access to flags. ]
Reported-by: Kan Liang <kan.liang@intel.com>
Tested-by: Kan Liang <kan.liang@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Another merge window, another MAINTAINERS file disaster.
People have serious problems with the alphabet and sorting, and poor
Jérôme Glisse and Radim Krčmář get their names mangled by locale issues,
turning them into some mangled mess (probably others do too, but those
two stood out when sorting things again).
And we now have two copies of the same 'AS3645A LED FLASH CONTROLLER
DRIVER' in the tree and in the MAINTAINERS file, but that's a separate
issue - the duplication is real, and I left them as two entries for the
same name.
This does not try to sort the actual section pattern entries, although I
may end up doing that later.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Commits:
7dcf90e9e032 ("PCI: hv: Use vPCI protocol version 1.2")
628f54cc6451 ("x86/hyper-v: Support extended CPU ranges for TLB flush hypercalls")
added the same definition and they came in through different trees.
Fix the duplication.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Cc: K. Y. Srinivasan <kys@microsoft.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephen Hemminger <sthemmin@microsoft.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: devel@linuxdriverproject.org
Link: http://lkml.kernel.org/r/20170911150620.3998-1-vkuznets@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
Allocate the hypervisor callback IDT entry early in the boot sequence.
The previous code would allocate the entry as part of registering the handler
when the vmbus driver loaded, and this caused a problem for the IDT cleanup
that Thomas is working on for v4.15.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: apw@canonical.com
Cc: devel@linuxdriverproject.org
Cc: gregkh@linuxfoundation.org
Cc: jasowang@redhat.com
Cc: olaf@aepfle.de
Link: http://lkml.kernel.org/r/20170908231557.2419-1-kys@exchange.microsoft.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
Jeremy Fitzhardinge is stepping down as a paravirt maintainer. I'll
replace him.
While at it, update the file list to the actual pattern.
Signed-off-by: Juergen Gross <jgross@suse.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: akataria@vmware.com
Cc: chrisw@sous-sol.org
Cc: jeremy@goop.org
Cc: rusty@rustcorp.com.au
Cc: virtualization@lists.linux-foundation.org
Link: http://lkml.kernel.org/r/20170905143407.9227-1-jgross@suse.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
With removal of lguest some of the paravirt functions are no longer
needed:
->read_cr4()
->store_idt()
->set_pmd_at()
->set_pud_at()
->pte_update()
Remove them.
Signed-off-by: Juergen Gross <jgross@suse.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: akataria@vmware.com
Cc: boris.ostrovsky@oracle.com
Cc: chrisw@sous-sol.org
Cc: jeremy@goop.org
Cc: rusty@rustcorp.com.au
Cc: virtualization@lists.linux-foundation.org
Cc: xen-devel@lists.xenproject.org
Link: http://lkml.kernel.org/r/20170904102527.25409-1-jgross@suse.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
cpu_init() is weird: it's called rather late (after early
identification and after most MMU state is initialized) on the boot
CPU but is called extremely early (before identification) on secondary
CPUs. It's called just late enough on the boot CPU that its CR4 value
isn't propagated to mmu_cr4_features.
Even if we put CR4.PCIDE into mmu_cr4_features, we'd hit two
problems. First, we'd crash in the trampoline code. That's
fixable, and I tried that. It turns out that mmu_cr4_features is
totally ignored by secondary_start_64(), though, so even with the
trampoline code fixed, it wouldn't help.
This means that we don't currently have CR4.PCIDE reliably initialized
before we start playing with cpu_tlbstate. This is very fragile and
tends to cause boot failures if I make even small changes to the TLB
handling code.
Make it more robust: initialize CR4.PCIDE earlier on the boot CPU
and propagate it to secondary CPUs in start_secondary().
( Yes, this is ugly. I think we should have improved mmu_cr4_features
to actually control CR4 during secondary bootup, but that would be
fairly intrusive at this stage. )
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Reported-by: Sai Praneeth Prakhya <sai.praneeth.prakhya@intel.com>
Tested-by: Sai Praneeth Prakhya <sai.praneeth.prakhya@intel.com>
Cc: Borislav Petkov <bpetkov@suse.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Fixes: 660da7c9228f ("x86/mm: Enable CR4.PCIDE on supported systems")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
Jiri reported a resume-from-hibernation failure triggered by PCID.
The root cause appears to be rather odd. The hibernation asm
restores a CR3 value that comes from the image header. If the image
kernel has PCID on, it's entirely reasonable for this CR3 value to
have one of the low 12 bits set. The restore code restores it with
CR4.PCIDE=0, which means that those low 12 bits are accepted by the
CPU but are either ignored or interpreted as a caching mode. This
is odd, but still works. We blow up later when the image kernel
restores CR4, though, since changing CR4.PCIDE with CR3[11:0] != 0
is illegal. Boom!
FWIW, it's entirely unclear to me what's supposed to happen if a PAE
kernel restores a non-PAE image or vice versa. Ditto for LA57.
Reported-by: Jiri Kosina <jikos@kernel.org>
Tested-by: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bpetkov@suse.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 660da7c9228f ("x86/mm: Enable CR4.PCIDE on supported systems")
Link: http://lkml.kernel.org/r/18ca57090651a6341e97083883f9e814c4f14684.1504847163.git.luto@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
If we hit the VM_BUG_ON(), we're detecting a genuinely bad situation,
but we're very unlikely to get a useful call trace.
Make it a warning instead.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bpetkov@suse.de>
Cc: Jiri Kosina <jikos@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/3b4e06bbb382ca54a93218407c93925ff5871546.1504847163.git.luto@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
If using a kernel with CONFIG_XFS_RT=y and we set the RHINHERIT flag on
a directory in a filesystem that does not have a realtime device and
create a new file in that directory, it gets marked as a real time file.
When data is written and a fsync is issued, the filesystem attempts to
flush a non-existent rt device during the fsync process.
This results in a crash dereferencing a null buftarg pointer in
xfs_blkdev_issue_flush():
BUG: unable to handle kernel NULL pointer dereference at 0000000000000008
IP: xfs_blkdev_issue_flush+0xd/0x20
.....
Call Trace:
xfs_file_fsync+0x188/0x1c0
vfs_fsync_range+0x3b/0xa0
do_fsync+0x3d/0x70
SyS_fsync+0x10/0x20
do_syscall_64+0x4d/0xb0
entry_SYSCALL64_slow_path+0x25/0x25
Setting RT inode flags does not require special privileges so any
unprivileged user can cause this oops to occur. To reproduce, confirm
kernel is compiled with CONFIG_XFS_RT=y and run:
# mkfs.xfs -f /dev/pmem0
# mount /dev/pmem0 /mnt/test
# mkdir /mnt/test/foo
# xfs_io -c 'chattr +t' /mnt/test/foo
# xfs_io -f -c 'pwrite 0 5m' -c fsync /mnt/test/foo/bar
Or just run xfstests with MKFS_OPTIONS="-d rtinherit=1" and wait.
Kernels built with CONFIG_XFS_RT=n are not exposed to this bug.
Fixes: f538d4da8d52 ("[XFS] write barrier support")
Cc: <stable@vger.kernel.org>
Signed-off-by: Richard Wareing <rwareing@fb.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Once we encounter I/O interruption during issuing discards, we will delay
long time before next round, but if system status is I/O idle during the
time, it may loses opportunity to issue discards. So this patch changes
to hurry up to issue discard after io interruption.
Besides, this patch also fixes to issue discards accurately with assigned
rate.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
Fix below incorrect display when reading discard_granularity sysfs node.
$ cat /sys/fs/f2fs/<device>/discard_granularity
$ 16
$ echo 32 > /sys/fs/f2fs/<device>/discard_granularity
$ cat /sys/fs/f2fs/<device>/discard_granularity
$ 16
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
Add a bugon in f2fs_evict_inode to detect inconsistent status between
inode cache and related node page cache.
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
When packaging the perf userland application into an AppImage, the
wait() call in perf stat returned too early. It turned out that some
other child process exited, but not the one perf stat launched:
$ sudo strace -e fork,execve,clone,wait4 -f ./perf-x86_64.AppImage stat sleep 1
execve("./perf-git.3a73b7f9-x86_64.AppImage", ["./perf-git.3a73b7f9-x86_64.AppIm"..., "stat", "sleep", "1"], 0x7ffec1bbf050 /* 18 vars */) = 0
clone(child_stack=NULL, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0x7f6a6e7efe50) = 3912
strace: Process 3912 attached
[pid 3912] clone(child_stack=NULL, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0x7f6a6e7efe50) = 3914
strace: Process 3914 attached
[pid 3912] +++ exited with 0 +++
[pid 3911] --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_EXITED, si_pid=3912, si_uid=0, si_status=0, si_utime=0, si_stime=0} ---
[pid 3914] clone(strace: Process 3915 attached
child_stack=0x7f6a6d9fefb0, flags=CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHAND|CLONE_THREAD|CLONE_SYSVSEM|CLONE_SETTLS|CLONE_PARENT_SETTID|CLONE_CHILD_CLEARTID, parent_tidptr=0x7f6a6d9ff9d0, tls=0x7f6a6d9ff700, child_tidptr=0x7f6a6d9ff9d0) = 3915
[pid 3911] execve("/tmp/.mount_perf-g6VYMpl/AppRun", ["./perf-git.3a73b7f9-x86_64.AppIm"..., "stat", "sleep", "1"], 0x14aab70 /* 21 vars */) = 0
[pid 3911] clone(child_stack=NULL, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0x7f4ae113c4d0) = 3916
strace: Process 3916 attached
[pid 3911] wait4(-1, [{WIFEXITED(s) && WEXITSTATUS(s) == 0}], 0, NULL) = 3912
[pid 3916] execve("/usr/libexec/perf-core/sleep", ["sleep", "1"], 0x27d3650 /* 22 vars */) = -1 ENOENT (No such file or directory)
[pid 3916] execve("/tmp/./sleep", ["sleep", "1"], 0x27d3650 /* 22 vars */) = -1 ENOENT (No such file or directory)
[pid 3916] execve("/home/milian/.bin/sleep", ["sleep", "1"], 0x27d3650 /* 22 vars */) = -1 ENOENT (No such file or directory)
[pid 3916] execve("/usr/lib/icecream/libexec/icecc/bin/sleep", ["sleep", "1"], 0x27d3650 /* 22 vars */) = -1 ENOENT (No such file or directory)
[pid 3916] execve("/ssd2/milian/projects/compiled/other/bin/sleep", ["sleep", "1"], 0x27d3650 /* 22 vars */) = -1 ENOENT (No such file or directory)
[pid 3916] execve("/home/milian/.bin/kf5/sleep", ["sleep", "1"], 0x27d3650 /* 22 vars */) = -1 ENOENT (No such file or directory)
[pid 3916] execve("/ssd2/milian/projects/compiled/kf5/bin/sleep", ["sleep", "1"], 0x27d3650 /* 22 vars */) = -1 ENOENT (No such file or directory)
[pid 3916] execve("/home/milian/projects/compiled/other/bin/sleep", ["sleep", "1"], 0x27d3650 /* 22 vars */) = -1 ENOENT (No such file or directory)
[pid 3916] execve("/home/milian/projects/compiled/kf5/bin/sleep", ["sleep", "1"], 0x27d3650 /* 22 vars */) = -1 ENOENT (No such file or directory)
[pid 3916] execve("/usr/local/sbin/sleep", ["sleep", "1"], 0x27d3650 /* 22 vars */) = -1 ENOENT (No such file or directory)
[pid 3916] execve("/usr/local/bin/sleep", ["sleep", "1"], 0x27d3650 /* 22 vars */) = -1 ENOENT (No such file or directory)
[pid 3916] execve("/usr/bin/sleep", ["sleep", "1"], 0x27d3650 /* 22 vars */
Performance counter stats for 'sleep 1':
<not counted> task-clock
<not counted> context-switches
<not counted> cpu-migrations
<not counted> page-faults
<not counted> cycles
<not counted> instructions
<not counted> branches
<not counted> branch-misses
0.000047194 seconds time elapsed
[pid 3916] --- SIGTERM {si_signo=SIGTERM, si_code=SI_USER, si_pid=3911, si_uid=0} ---
[pid 3916] +++ killed by SIGTERM +++
[pid 3911] --- SIGCHLD {si_signo=SIGCHLD, si_code=CLD_KILLED, si_pid=3916, si_uid=0, si_status=SIGTERM, si_utime=0, si_stime=0} ---
[pid 3915] --- SIGPIPE {si_signo=SIGPIPE, si_code=SI_USER, si_pid=3914, si_uid=0} ---
[pid 3911] +++ exited with 0 +++
[pid 3915] --- SIGHUP {si_signo=SIGHUP, si_code=SI_USER, si_pid=3914, si_uid=0} ---
[pid 3915] +++ exited with 0 +++
+++ exited with 0 +++
This patch uses waitpid instead to ensure the call waits for the
debuggee application launched by 'perf stat'. This fixes 'perf stat'
when launched from an AppImage:
$ ./perf-x86_64.AppImage stat sleep 1
Performance counter stats for 'sleep 1':
0.357235 task-clock (msec) # 0.000 CPUs utilized
1 context-switches # 0.003 M/sec
0 cpu-migrations # 0.000 K/sec
50 page-faults # 0.140 M/sec
1269602 cycles # 3.554 GHz
654278 instructions # 0.52 insn per cycle
129963 branches # 363.803 M/sec
7082 branch-misses # 5.45% of all branches
1.000633420 seconds time elapsed
Signed-off-by: Milian Wolff <milian.wolff@kdab.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20170912152523.4497-1-milian.wolff@kdab.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
Previously the part behind "perf-" was interpreted as an internal perf
command. If the suffix could not be handled, the execution was stopped.
This makes it impossible to launch perf binaries that got renamed to
have the `perf-` prefix. This is e.g. the case for appimages (e.g.
"perf-x86_64.AppImage"), but would also apply to all other scenarios
where users symlink or rename perf themselves:
Status quo with the broken behavior:
$ ln -s ./perf ./perf-custom-suffix
$ ./perf-custom-suffix list
cannot handle custom-suffix internally$
Also note the missing newline at the end of the error message.
With this patch applied, the above works properly:
$ ./perf-custom-suffix list
List of pre-defined events (to be used in -e):
...
Signed-off-by: Milian Wolff <milian.wolff@kdab.com>
Acked-by: David Ahern <dsahern@gmail.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Yao Jin <yao.jin@linux.intel.com>
Link: http://lkml.kernel.org/r/20170911111422.31903-1-milian.wolff@kdab.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
I'm forever late for editing my kernel cmdline, add a runtime knob to
disable the "sched_debug" thing.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20170907150614.142924283@infradead.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
Migrating tasks to offline CPUs is a pretty big fail, warn about it.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20170907150614.094206976@infradead.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
The load balancer applies cpu_active_mask to whatever sched_domains it
finds, however in the case of active_balance there is a hole between
setting rq->{active_balance,push_cpu} and running the stop_machine
work doing the actual migration.
The @push_cpu can go offline in this window, which would result in us
moving a task onto a dead cpu, which is a fairly bad thing.
Double check the active mask before the stop work does the migration.
CPU0 CPU1
<SoftIRQ>
stop_machine(takedown_cpu)
load_balance() cpu_stopper_thread()
... work = multi_cpu_stop
stop_one_cpu_nowait( /* wait for CPU0 */
.func = active_load_balance_cpu_stop
);
</SoftIRQ>
cpu_stopper_thread()
work = multi_cpu_stop
/* sync with CPU1 */
take_cpu_down()
<idle>
play_dead();
work = active_load_balance_cpu_stop
set_task_cpu(p, CPU1); /* oops!! */
Reported-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20170907150614.044460912@infradead.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
On CPU hot unplug, when parking the last kthread we'll try and
schedule into idle to kill the CPU. This last schedule can (and does)
trigger newidle balance because at this point the sched domains are
still up because of commit:
77d1dfda0e79 ("sched/topology, cpuset: Avoid spurious/wrong domain rebuilds")
Obviously pulling tasks to an already offline CPU is a bad idea, and
all balancing operations _should_ be subject to cpu_active_mask, make
it so.
Reported-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Fixes: 77d1dfda0e79 ("sched/topology, cpuset: Avoid spurious/wrong domain rebuilds")
Link: http://lkml.kernel.org/r/20170907150613.994135806@infradead.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
Currently section->from_system_config is being checked multiple times.
item->from_system_config should be checked instead, when iterating thru
the items in a section. Fix it.
Signed-off-by: Taeung Song <treeze.taeung@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/1504754325-9724-1-git-send-email-treeze.taeung@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
We currently update the 'next' variable only with a single step value.
But it's possible the 'adv' update is bigger than single 'step' value.
This would leave 'next' value under counted and force unnecessary
ui_progress__ops->update calls.
Calculate the amount of steps we need for 'adv' update and increase the
'next' with that amounts of steps.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20170908120510.22515-3-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
Unlikely, but we could have ui_progress__init being called with total <
16, which would set the next and step variables to 0. That would force
unnecessary ui_progress__ops->update calls because 'next' would never
raise.
Forcing the next and step values to be always > 0.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20170908120510.22515-2-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
Do not carry the perf.data file descriptor into the workload process and
close it when perf executes the workload.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20170908084621.31595-2-jolsa@kernel.org
[ Add definitions for O_CLOEXEC for older systems ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
Do not use -D_FORTIFY_SOURCE=2 for DEBUG build as it seems to mess up
with debuginfo, which results in bad gdb experience.
We already do that for tools/perf/.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20170908084621.31595-1-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
When cross compiling perf and I want to link against a self-compiled
libunwind, I usually make the custom path where the libunwind headers
exist visible by adding the libunwind prefix to the include path when
compiling perf, i.e.:
~~~~~
$ ls $HOME/projects/compiled/other/include/
libunwind-coredump.h libunwind.h libunwind-x86_64.h
libunwind-common.h libunwind-dynamic.h libunwind-ptrace.h
unwind.h
$ make EXTRA_CFLAGS="-I$HOME/projects/compiled/other/include/
~~~~~~
Note the `unwind.h` header from libunwind which leads to compile
errors when compiling tests/dwarf-unwind.c, since it shadows perf's
util/unwind.h:
~~~~~
tests/dwarf-unwind.c:41:32: error: ‘struct unwind_entry’ declared inside parameter list will not be visible outside of this definition or declaration [-Werror]
static int unwind_entry(struct unwind_entry *entry, void *arg)
^~~~~~~~~~~~
tests/dwarf-unwind.c: In function ‘unwind_entry’:
tests/dwarf-unwind.c:44:22: error: dereferencing pointer to incomplete type ‘struct unwind_entry’
char *symbol = entry->sym ? entry->sym->name : NULL;
^~
tests/dwarf-unwind.c: In function ‘unwind_thread’:
tests/dwarf-unwind.c:92:8: error: implicit declaration of function ‘unwind__get_entries’; did you mean ‘unwind_entry’? [-Werror=implicit-function-declaration]
err = unwind__get_entries(unwind_entry, &cnt, thread,
^~~~~~~~~~~~~~~~~~~
unwind_entry
tests/dwarf-unwind.c:92:8: error: nested extern declaration of ‘unwind__get_entries’ [-Werror=nested-externs]
~~~~~~
Fix this compile error by specificing an explicit include of perf's
unwind.h in the util folder.
Signed-off-by: Milian Wolff <milian.wolff@kdab.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Yao Jin <yao.jin@linux.intel.com>
Link: http://lkml.kernel.org/r/20170906150209.12579-1-milian.wolff@kdab.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
When cross building to android r15c (and older versions) on Fedora 26
we notice these:
/opt/android-ndk-r15c/platforms/android-24/arch-arm/usr/include/sys/cdefs.h:332:0: note: this is the location of the previous definition
For __aligned, __packed and __noreturn, so guard those with ifdefs to
avoid drowning useful warnings in these.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/n/tip-d7w3fa9c22dtmrwbedos6ie1@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
|
Commit b9ac5c274b8c ("ovl: hash overlay non-dir inodes by copy up origin")
verifies that the origin lower inode stored in the overlayfs inode matched
the inode of a copy up origin dentry found by lookup.
There is a false positive result in that check when lower fs does not
support file handles and copy up origin cannot be followed by file handle
at lookup time.
The false negative happens when finding an overlay inode in cache on a
copied up overlay dentry lookup. The overlay inode still 'remembers' the
copy up origin inode, but the copy up origin dentry is not available for
verification.
Relax the check in case copy up origin dentry is not available.
Fixes: b9ac5c274b8c ("ovl: hash overlay non-dir inodes by copy up...")
Cc: <stable@vger.kernel.org> # v4.13
Reported-by: Jordi Pujol <jordipujolp@gmail.com>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
The refreshed argument isn't used by any caller, get rid of it.
Use a helper for just updating the inode (no need to fill in a kstat).
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
If the IOCB_DSYNC flag is set a sync is not being performed by
fuse_file_write_iter.
Honor IOCB_DSYNC/IOCB_SYNC by setting O_DYSNC/O_SYNC respectively in the
flags filed of the write request.
We don't need to sync data or metadata, since fuse_perform_write() does
write-through and the filesystem is responsible for updating file times.
Original patch by Vitaly Zolotusky.
Reported-by: Nate Clark <nate@neworld.us>
Cc: Vitaly Zolotusky <vitaly@unitc.com>.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Commit 0b6e9ea041e6 ("fuse: Add support for pid namespaces") broke
Sandstorm.io development tools, which have been sending FUSE file
descriptors across PID namespace boundaries since early 2014.
The above patch added a check that prevented I/O on the fuse device file
descriptor if the pid namespace of the reader/writer was different from the
pid namespace of the mounter. With this change passing the device file
descriptor to a different pid namespace simply doesn't work. The check was
added because pids are transferred to/from the fuse userspace server in the
namespace registered at mount time.
To fix this regression, remove the checks and do the following:
1) the pid in the request header (the pid of the task that initiated the
filesystem operation) is translated to the reader's pid namespace. If a
mapping doesn't exist for this pid, then a zero pid is used. Note: even if
a mapping would exist between the initiator task's pid namespace and the
reader's pid namespace the pid will be zero if either mapping from
initator's to mounter's namespace or mapping from mounter's to reader's
namespace doesn't exist.
2) The lk.pid value in setlk/setlkw requests and getlk reply is left alone.
Userspace should not interpret this value anyway. Also allow the
setlk/setlkw operations if the pid of the task cannot be represented in the
mounter's namespace (pid being zero in that case).
Reported-by: Kenton Varda <kenton@sandstorm.io>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Fixes: 0b6e9ea041e6 ("fuse: Add support for pid namespaces")
Cc: <stable@vger.kernel.org> # v4.12+
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Seth Forshee <seth.forshee@canonical.com>
|
|
ALSA sequencer core has a mechanism to load the enumerated devices
automatically, and it's performed in an off-load work. This seems
causing some race when a sequencer is removed while the pending
autoload work is running. As syzkaller spotted, it may lead to some
use-after-free:
BUG: KASAN: use-after-free in snd_rawmidi_dev_seq_free+0x69/0x70
sound/core/rawmidi.c:1617
Write of size 8 at addr ffff88006c611d90 by task kworker/2:1/567
CPU: 2 PID: 567 Comm: kworker/2:1 Not tainted 4.13.0+ #29
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011
Workqueue: events autoload_drivers
Call Trace:
__dump_stack lib/dump_stack.c:16 [inline]
dump_stack+0x192/0x22c lib/dump_stack.c:52
print_address_description+0x78/0x280 mm/kasan/report.c:252
kasan_report_error mm/kasan/report.c:351 [inline]
kasan_report+0x230/0x340 mm/kasan/report.c:409
__asan_report_store8_noabort+0x1c/0x20 mm/kasan/report.c:435
snd_rawmidi_dev_seq_free+0x69/0x70 sound/core/rawmidi.c:1617
snd_seq_dev_release+0x4f/0x70 sound/core/seq_device.c:192
device_release+0x13f/0x210 drivers/base/core.c:814
kobject_cleanup lib/kobject.c:648 [inline]
kobject_release lib/kobject.c:677 [inline]
kref_put include/linux/kref.h:70 [inline]
kobject_put+0x145/0x240 lib/kobject.c:694
put_device+0x25/0x30 drivers/base/core.c:1799
klist_devices_put+0x36/0x40 drivers/base/bus.c:827
klist_next+0x264/0x4a0 lib/klist.c:403
next_device drivers/base/bus.c:270 [inline]
bus_for_each_dev+0x17e/0x210 drivers/base/bus.c:312
autoload_drivers+0x3b/0x50 sound/core/seq_device.c:117
process_one_work+0x9fb/0x1570 kernel/workqueue.c:2097
worker_thread+0x1e4/0x1350 kernel/workqueue.c:2231
kthread+0x324/0x3f0 kernel/kthread.c:231
ret_from_fork+0x25/0x30 arch/x86/entry/entry_64.S:425
The fix is simply to assure canceling the autoload work at removing
the device.
Reported-by: Andrey Konovalov <andreyknvl@google.com>
Tested-by: Andrey Konovalov <andreyknvl@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
|
|
Add a jump target so that a bit of exception handling can be better reused
at the end of this function.
This issue was detected by using the Coccinelle software.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Takashi Iwai <tiwai@suse.de>
|
|
On a senario like writing out the first dirty page of the inode
as the inline data, we only cleared dirty flags of the pages, but
didn't clear the dirty tags of those pages in the radix tree.
If we don't clear the dirty tags of the pages in the radix tree, the
inodes which contain the pages will be marked with I_DIRTY_PAGES again
and again, and writepages() for the inodes will be invoked in every
writeback period. As a result, nothing will be done in every
writepages() for the inodes and it will just consume CPU time
meaninglessly.
Signed-off-by: Daeho Jeong <daeho.jeong@samsung.com>
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
This patch activates SSR in gc_urgent mode.
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
In android, we'd better wait for fstrim completion instead of issuing the
discard commands asynchronous.
Reviewed-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
|
|
A NULL pointer crash was reported for the case of having the BFQ IO
scheduler attached to the underlying blk-mq paths of a DM multipath
device. The crash occured in blk_mq_sched_insert_request()'s call to
e->type->ops.mq.insert_requests().
Paolo Valente correctly summarized why the crash occured with:
"the call chain (dm_mq_queue_rq -> map_request -> setup_clone ->
blk_rq_prep_clone) creates a cloned request without invoking
e->type->ops.mq.prepare_request for the target elevator e. The cloned
request is therefore not initialized for the scheduler, but it is
however inserted into the scheduler by blk_mq_sched_insert_request."
All said, a request-based DM multipath device's IO scheduler should be
the only one used -- when the original requests are issued to the
underlying paths as cloned requests they are inserted directly in the
underlying dispatch queue(s) rather than through an additional elevator.
But commit bd166ef18 ("blk-mq-sched: add framework for MQ capable IO
schedulers") switched blk_insert_cloned_request() from using
blk_mq_insert_request() to blk_mq_sched_insert_request(). Which
incorrectly added elevator machinery into a call chain that isn't
supposed to have any.
To fix this introduce a blk-mq private blk_mq_request_bypass_insert()
that blk_insert_cloned_request() calls to insert the request without
involving any elevator that may be attached to the cloned request's
request_queue.
Fixes: bd166ef183c2 ("blk-mq-sched: add framework for MQ capable IO schedulers")
Cc: stable@vger.kernel.org
Reported-by: Bart Van Assche <Bart.VanAssche@wdc.com>
Tested-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
If the 'kmalloc' fails, we must go through the existing error handling
path.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Fixes: 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks")
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
The way I'd implemented the new helper memcpy_and_pad with
__FORTIFY_INLINE caused compiler warnings for certain kernel
configurations.
This helper is only used in a single place at this time, and thus
doesn't benefit much from fortification. So simplify the code
by dropping fortification support for now.
Fixes: 01f33c336e2d "string.h: add memcpy_and_pad()"
Signed-off-by: Martin Wilck <mwilck@suse.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
Adds support for the new Host Memory Buffer Minimum Descriptor Entry Size
and Host Memory Maximum Descriptors Entries field that were added in
TP 4002 HMB Enhancements. These allow the controller to advertise
limits for the usual number of segments in the host memory buffer, as
well as a minimum usable per-segment size.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <keith.busch@intel.com>
|
|
We want to catch command execution errors when resetting the device, so
propagate errors from the Set Features when setting up the host memory
buffer. We keep ignoring memory allocation failures, as the spec
clearly says that the controller must work without a host memory buffer.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Cc: stable@vger.kernel.org
|
|
The initial chunk size for host memory buffer allocation is currently
PAGE_SIZE << MAX_ORDER. MAX_ORDER order allocation is usually failed
without CONFIG_DMA_CMA. So the HMB allocation is retried with chunk size
PAGE_SIZE << (MAX_ORDER - 1) in general, but there is no problem if the
retry allocation works correctly.
Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
[hch: rebased]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Cc: stable@vger.kernel.org
|
|
nvme_alloc_host_mem currently contains two loops that are interwinded,
and the outer retry loop turns out to be broken. Fix this by untangling
the two.
Based on a report an initial patch from Akinobu Mita.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reported-by: Akinobu Mita <akinobu.mita@gmail.com>
Tested-by: Akinobu Mita <akinobu.mita@gmail.com>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Cc: stable@vger.kernel.org
|
|
nvme_nvm_ns_supported assumes every device is a pci_dev, which leads to
reading an incorrect field, or possible even a dereference of unallocated
memory for fabrics controllers.
Fix this by introducing a quirk for lighnvm capable devices instead.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Matias Bjørling <mb@lightnvm.io>
Reviewed-by: Keith Busch <keith.busch@intel.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
|
|
Fix possible integer overflow in __blkdev_sectors_to_bio_pages if
sector_t is 32-bit.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Fixes: 615d22a51c04 ("block: Fix __blkdev_issue_zeroout loop")
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Users who are booting off their Opal enabled drives are having
issues when they have a shadow MBR set up after s3/resume cycle.
When the Drive has a shadow MBR setup the MBRDone flag is set to
false upon power loss (S3/S4/S5). When the MBRDone flag is false
I/O to LBA 0 -> LBA_END_MBR are remapped to the shadow mbr
of the drive. If the drive contains useful data in the 0 -> end_mbr
range upon s3 resume the user can never get to that data as the
drive will keep remapping it to the MBR. To fix this when we unlock
on S3 resume, we need to tell the drive that we're done with the
shadow mbr (even though we didnt use it) by setting true to MBRDone.
This way the drive will stop the remapping and the user can access
their data.
Acked-by Jon Derrick: <jonathan.derrick@intel.com>
Signed-off-by: Scott Bauer <scott.bauer@intel.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
__get_request() can call trace_block_getrq() with bio=NULL which causes
block_get_rq::TP_fast_assign() to deref a NULL pointer and panic.
Syzkaller fuzzer panics with
linux-next (1d53d908b79d7870d89063062584eead4cf83448):
kasan: GPF could be caused by NULL-ptr deref or user memory access
general protection fault: 0000 [#1] SMP KASAN
Modules linked in:
CPU: 0 PID: 2983 Comm: syzkaller401111 Not tainted 4.13.0-rc7-next-20170901+ #13
task: ffff8801cf1da000 task.stack: ffff8801ce440000
RIP: 0010:perf_trace_block_get_rq+0x697/0x970 include/trace/events/block.h:384
RSP: 0018:ffff8801ce4473f0 EFLAGS: 00010246
RAX: ffff8801cf1da000 RBX: 1ffff10039c88e84 RCX: 1ffffd1ffff84d27
RDX: dffffc0000000001 RSI: 1ffff1003b643e7a RDI: ffffe8ffffc26938
RBP: ffff8801ce447530 R08: 1ffff1003b643e6c R09: ffffe8ffffc26964
R10: 0000000000000002 R11: fffff91ffff84d2d R12: ffffe8ffffc1f890
R13: ffffe8ffffc26930 R14: ffffffff85cad9e0 R15: 0000000000000000
FS: 0000000002641880(0000) GS:ffff8801db200000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 000000000043e670 CR3: 00000001d1d7a000 CR4: 00000000001406f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
trace_block_getrq include/trace/events/block.h:423 [inline]
__get_request block/blk-core.c:1283 [inline]
get_request+0x1518/0x23b0 block/blk-core.c:1355
blk_old_get_request block/blk-core.c:1402 [inline]
blk_get_request+0x1d8/0x3c0 block/blk-core.c:1427
sg_scsi_ioctl+0x117/0x750 block/scsi_ioctl.c:451
sg_ioctl+0x192d/0x2ed0 drivers/scsi/sg.c:1070
vfs_ioctl fs/ioctl.c:45 [inline]
do_vfs_ioctl+0x1b1/0x1530 fs/ioctl.c:685
SYSC_ioctl fs/ioctl.c:700 [inline]
SyS_ioctl+0x8f/0xc0 fs/ioctl.c:691
entry_SYSCALL_64_fastpath+0x1f/0xbe
block_get_rq::TP_fast_assign() has multiple redundant ->dev assignments.
Only one of them is NULL tolerant. Favor the NULL tolerant one.
Fixes: 74d46992e0d9 ("block: replace bi_bdev with a gendisk pointer and partitions index")
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Greg Thelen <gthelen@google.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
The following revert:
2b85b3d22920 ("x86/acpi: Restore the order of CPU IDs")
... got rid of __generic_processor_info(), but forgot to remove its
declaration in mpspec.h.
Remove the declaration and update the comments as well.
Signed-off-by: Dou Liyang <douly.fnst@cn.fujitsu.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: lenb@kernel.org
Link: http://lkml.kernel.org/r/1505101403-29100-1-git-send-email-douly.fnst@cn.fujitsu.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
Work around kernel-doc warning ('*' in Sphinx doc means "emphasis"):
../kernel/sched/fair.c:7584: WARNING: Inline emphasis start-string without end-string.
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/f18b30f9-6251-6d86-9d44-16501e386891@infradead.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
This reverts commit 81f95076281fdd3bc382e004ba1bce8e82fccbce.
It causes random failures of firmware loading at resume time (well,
random for me, it seems to be more reliable for others) because the
firmware disabling is not actually synchronous with any particular
resume event, and at least the btusb driver that uses a workqueue to
load the firmware at resume seems to occasionally hit the "firmware
loading is disabled" logic because the firmware loader hasn't gotten the
resume event yet.
Some kind of sanity check for not trying to load firmware when it's not
possible might be a good thing, but this commit was not it.
Greg seems to have silently suffered the same issue, and pointed to the
likely culprit, and Gabriel C verified the revert fixed it for him too.
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Pointed-at-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Tested-by: Gabriel C <nix.or.die@gmail.com>
Cc: Luis R. Rodriguez <mcgrof@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
After removing linux/vmalloc.h from asm-generic/io.h, the following
warning occurs on openrisc:
In file included from arch/openrisc/include/asm/io.h:33:0,
from include/linux/io.h:25,
from drivers/tty/serial/earlycon.c:19:
arch/openrisc/include/asm/pgtable.h:424:2: warning: 'struct vm_area_struct' declared inside parameter list
unsigned long address, pte_t *pte)
^
arch/openrisc/include/asm/pgtable.h:424:2: warning: its scope is only this definition or declaration, which is probably not what you want
Reported-by: kbuild test robot <lkp@intel.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Stafford Horne <shorne@gmail.com>
|