The testcase "mmaptruncate" of ocfs2-test deadlocks occasionally.
In this testcase, we create a 2*CLUSTER_SIZE file and mmap() on it;
there are 2 process repeatedly performing the following operations
respectively: one is doing memset(mmaped_addr + 2*CLUSTER_SIZE - 1, 'a',
1), while the another is playing ftruncate(fd, 2*CLUSTER_SIZE) and then
ftruncate(fd, CLUSTER_SIZE) again and again.
This is the backtrace when the deadlock happens:
__wait_on_bit_lock+0x50/0xa0
__lock_page+0xb7/0xc0
ocfs2_write_begin_nolock+0x163f/0x1790 [ocfs2]
ocfs2_page_mkwrite+0x1c7/0x2a0 [ocfs2]
do_page_mkwrite+0x66/0xc0
handle_mm_fault+0x685/0x1350
__do_page_fault+0x1d8/0x4d0
trace_do_page_fault+0x37/0xf0
do_async_page_fault+0x19/0x70
async_page_fault+0x28/0x30
In ocfs2_write_begin_nolock(), we first grab the pages and then allocate
disk space for this write; ocfs2_try_to_free_truncate_log() will be
called if -ENOSPC is returned; if we're lucky to get enough clusters,
which is usually the case, we start over again.
But in ocfs2_free_write_ctxt() the target page isn't unlocked, so we
will deadlock when trying to grab the target page again.
Also, -ENOMEM might be returned in ocfs2_grab_pages_for_write().
Another deadlock will happen in __do_page_mkwrite() if
ocfs2_page_mkwrite() returns non-VM_FAULT_LOCKED, and along with a
locked target page.
These two errors fail on the same path, so fix them by unlocking the
target page manually before ocfs2_free_write_ctxt().
Jan Kara helps me clear out the JBD2 part, and suggest the hint for root
cause.
Changes since v1:
1. Also put ENOMEM error case into consideration.
Link: http://lkml.kernel.org/r/1474173902-32075-1-git-send-email-zren@suse.com
Signed-off-by: Eric Ren <zren@suse.com>
Reviewed-by: He Gang <ghe@suse.com>
Acked-by: Joseph Qi <joseph.qi@huawei.com>
Cc: Mark Fasheh <mfasheh@suse.de>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Antonio reports the following crash when using fuse under memory pressure:
kernel BUG at /build/linux-a2WvEb/linux-4.4.0/mm/workingset.c:346!
invalid opcode: 0000 [#1] SMP
Modules linked in: all of them
CPU: 2 PID: 63 Comm: kswapd0 Not tainted 4.4.0-36-generic #55-Ubuntu
Hardware name: System manufacturer System Product Name/P8H67-M PRO, BIOS 3904 04/27/2013
task: ffff88040cae6040 ti: ffff880407488000 task.ti: ffff880407488000
RIP: shadow_lru_isolate+0x181/0x190
Call Trace:
__list_lru_walk_one.isra.3+0x8f/0x130
list_lru_walk_one+0x23/0x30
scan_shadow_nodes+0x34/0x50
shrink_slab.part.40+0x1ed/0x3d0
shrink_zone+0x2ca/0x2e0
kswapd+0x51e/0x990
kthread+0xd8/0xf0
ret_from_fork+0x3f/0x70
which corresponds to the following sanity check in the shadow node
tracking:
BUG_ON(node->count & RADIX_TREE_COUNT_MASK);
The workingset code tracks radix tree nodes that exclusively contain
shadow entries of evicted pages in them, and this (somewhat obscure)
line checks whether there are real pages left that would interfere with
reclaim of the radix tree node under memory pressure.
While discussing ways how fuse might sneak pages into the radix tree
past the workingset code, Miklos pointed to replace_page_cache_page(),
and indeed there is a problem there: it properly accounts for the old
page being removed - __delete_from_page_cache() does that - but then
does a raw raw radix_tree_insert(), not accounting for the replacement
page. Eventually the page count bits in node->count underflow while
leaving the node incorrectly linked to the shadow node LRU.
To address this, make sure replace_page_cache_page() uses the tracked
page insertion code, page_cache_tree_insert(). This fixes the page
accounting and makes sure page-containing nodes are properly unlinked
from the shadow node LRU again.
Also, make the sanity checks a bit less obscure by using the helpers for
checking the number of pages and shadows in a radix tree node.
Fixes: 449dd6984d ("mm: keep page cache radix tree nodes in check")
Link: http://lkml.kernel.org/r/20160919155822.29498-1-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reported-by: Antonio SJ Musumeci <trapexit@spawn.link>
Debugged-by: Miklos Szeredi <miklos@szeredi.hu>
Cc: <stable@vger.kernel.org> [3.15+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit d9676fa152 ("ARCv2: Enable LOCKDEP"), changed
local_save_flags() to not return raw STATUS32 but encoded in the form
such that it could be fed directly to CLRI/SETI instructions.
However the STATUS32.E[] was not captured correctly as it corresponds to
bits [4:1] in the register and not [3:0]
Fixes: d9676fa152 ("ARCv2: Enable LOCKDEP")
Cc: Evgeny Voevodin <evgeny.voevodin@intel.com>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
Seem like values assigned as absolute number and not and
shift value, i.e. should be 0 for one node (2^0) and 1 for
couple of nodes (2^1)
Signed-off-by: Noam Camus <noamca@mellanox.com>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
In the end of "arc_init_IRQ" STATUS32.IE flag is going to be affected by
"flag" instruction but "flag" never touches IE flag on ARCv2. So "kflag"
instruction must be used instead of "flag".
Signed-off-by: Yuriy Kolerov <yuriy.kolerov@synopsys.com>
Cc: stable@vger.kernel.org #4.2+
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
We used to keep the .exit.* sections as linker would fail in final link
due to references from .debug_frame which itself could not be discardrd
due to the forced "write,alloc" attributes for it.
| LD init/built-in.o
| `.exit.text' referenced in section `.debug_frame' of arch/arc/built-in.o: defined in discarded section `.exit.text' of arch/arc/built-in.o
| Makefile:949: recipe for target 'vmlinux' failed
With .debug_frame now retired, this hack is no longer needed.
kernel binary is now a little bit smaller as well.
closes STAR 9000549913
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
This uses a new set of annoations viz. ENTRY_CFI/END_CFI to enabel cfi
ops generation.
Note that we didn't change the normal ENTRY/EXIT as we don't actually
want unwind info in the trap/exception/interrutp handlers which use
these, as unwinder then gets confused (it keeps recursing vs. stopping).
Semantically these are leaf routines and unwinding should stop when it
hits those routines.
Before
------
28.52% 1.19% 9929 hackbench libuClibc-1.0.17.so [.] __write_nocancel
|
---__write_nocancel
|--8.95%--EV_Trap
| --8.25%--sys_write
| |--3.93%--sock_write_iter
...
|--2.62%--memset <==== [LEAF entry as no unwind info]
^^^^^^
After
-----
29.46% 1.24% 13622 hackbench libuClibc-1.0.17.so [.] __write_nocancel
|
---__write_nocancel
|--9.31%--EV_Trap
| --8.62%--sys_write
| |--4.17%--sock_write_iter
...
|--6.19%--sys_write
| --6.19%--sock_write_iter
| unix_stream_sendmsg
| |--1.62%--sock_alloc_send_pskb
| |--0.89%--sock_def_readable
| |--0.88%--_raw_spin_unlock_irqrestore
| |--0.69%--memset
| | ^^^^^^ <==== [now in proper callframe]
| |
| --0.52%--skb_copy_datagram_from_iter
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
1. detect whether binutils supports the cfi pseudo ops
2. define conditional macros to generate the ops
3. define new ENTRY_CFI/END_CFI to annotate hand asm code.
- Needed because we don't want to emit dwarf info in general ENTRY/END
used by lowest level trap/exception/interrutp handlers as unwinder
gets confused trying to unwind out of them. We want unwinder to
instead stop when it hits onfo those routines
- These provide minimal start/end cfi ops assuming routine doesn't
touch stack memory/regs
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
This essentially removes ENTRY() assembler annotation for this symbol
since it didn't have a pairing END()
This in ahead of introducing cfi pseudo ops in ENTRY/END which expects
paired cfi_startproc/cfi_endproc
| ../arch/arc/kernel/entry.S: Assembler messages:
| ../arch/arc/kernel/entry.S:270: Error: previous CFI entry not closed (missing .cfi_endproc)
| ../scripts/Makefile.build:326: recipe for target 'arch/arc/kernel/entry-arcv2.o' failed
| make[4]: *** [arch/arc/kernel/entry-arcv2.o] Error 1
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
In .debug_frame based unwinding regime, we used to force -gdwarf-2 since
kernel unwinder only claimed to handle dwarf 2. This changed since commit
6d0d506012 ("ARC: dw2 unwind: Don't bail for CIE.version != 1")
which added some support beyond dwarf 2, atleast to handle CIE != 1
The ill-effect of -gdwarf-2 is that it forces generation of .debug_*
sections, which bloats loadable modules .ko files. For the curious, this
doesn't affect vmlinx binary since linker script discards .debug_* but
same discard is not yet implemented for modules.
So it seems we can drop the -gdwarf-2 toggle, which should not be needed
anyways given that we now use .eh_frame based unwinding.
I've verified using GNU 2016.09-engo10 that the actual unwind info is
not different with or w/o this toggle - but the debug_* sections are
gone for good.
before
-----
arc-linux-readelf -S q_proc.ko-unwinding-1-eh_frame-switch | grep debug
[15] .debug_info PROGBITS 00000000 000300 00d08d 00 0 0 1
[16] .rela.debug_info RELA 00000000 0162a0 008844 0c I 29 15 4
[17] .debug_abbrev PROGBITS 00000000 00d38d 0005f8 00 0 0 1
[18] .debug_loc PROGBITS 00000000 00d985 000070 00 0 0 1
[19] .rela.debug_loc RELA 00000000 01eae4 0000c0 0c I 29 18 4
[20] .debug_aranges PROGBITS 00000000 00d9f5 000040 00 0 0 1
[21] .rela.debug_arang RELA 00000000 01eba4 000030 0c I 29 20 4
[22] .debug_ranges PROGBITS 00000000 00da35 000018 00 0 0 1
[23] .rela.debug_range RELA 00000000 01ebd4 000030 0c I 29 22 4
[24] .debug_line PROGBITS 00000000 00da4d 000b5b 00 0 0 1
[25] .rela.debug_line RELA 00000000 01ec04 0000cc 0c I 29 24 4
[26] .debug_str PROGBITS 00000000 00e5a8 007831 01 MS 0 0 1
after
----
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
So finally after almost 8 years of dealing with .debug_frame, we are
finally switching to .eh_frame. The reason being stripped kernel
binaries had non-functional unwinder as .debug_frame was gone.
Also, in general .eh_frame seems more common way of doing unwinding.
This also folds a revert of f52e126cc7 ("ARC: unwind: ensure that
.debug_frame is generated (vs. .eh_frame)") to ensure that we start
getting .eh_frame
Reported-by: Daniel Mentz <danielmentz@google.com>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
We used to live with PERF_COUNT_HW_CACHE_REFERENCES and
PERF_COUNT_HW_CACHE_REFERENCES not specified on ARC.
Those events are actually aliases to 2 cache events that we do support
and so this change sets "cache-reference" and "cache-misses" events
in the same way as "L1-dcache-loads" and L1-dcache-load-misses.
And while at it adding debug info for cache events as well as doing a
subtle fix in HW events debug info - config value is much better
represented by hex so we may see not only event index but as well other
control bits set (if they exist).
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-snps-arc@lists.infradead.org
Cc: linux-kernel@vger.kernel.org
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
Build brekeage since last changes to generic atomic operations.
Added couple of missing macros which are now mandatory
Signed-off-by: Noam Camus <noamca@mellanox.com>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
ARCv2 ISA provides 64-bit exclusive load/stores so use them to implement
the 64-bit atomics and elide the spinlock based generic 64-bit atomics
boot tested with atomic64 self-test (and GOD bless the person who wrote
them, I realized my inline assmebly is sloppy as hell)
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will.deacon@arm.com>
Cc: linux-snps-arc@lists.infradead.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
HS release 3.0 provides for even more flexibility in specifying the
volatile address space for mapping peripherals.
With HS 2.1 @start was made flexible / programmable - with HS 3.0 even
@end can be setup (vs. fixed to 0xFFFF_FFFF before).
So add code to reflect that and while at it remove an unused struct
defintion
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
The cool thing is that same kernel image can run on
- nsim OSCI simulation platform
- SDPlite FPGA setups
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
As it was discussed quite some time ago (see
https://lkml.org/lkml/2015/11/5/862) it's a good practice to add
"model" property in .dts. Moreover as per ePAPR "model" property is
required and should look like "manufacturer,model" so we do here.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Vineet Gupta <vgupta@synopsys.com>
Cc: Jonas Gorski <jonas.gorski@gmail.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Rob Herring <robh@kernel.org>
Cc: Christian Ruppert <christian.ruppert@alitech.com>
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
Change my email address to my kernel.org account instead of the ARM one.
Signed-off-by: Javi Merino <javi.merino@arm.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This warning:
WARNING: CPU: 0 PID: 3331 at arch/x86/entry/common.c:45 enter_from_user_mode+0x32/0x50
CPU: 0 PID: 3331 Comm: ldt_gdt_64 Not tainted 4.8.0-rc7+ #13
Call Trace:
dump_stack+0x99/0xd0
__warn+0xd1/0xf0
warn_slowpath_null+0x1d/0x20
enter_from_user_mode+0x32/0x50
error_entry+0x6d/0xc0
? general_protection+0x12/0x30
? native_load_gs_index+0xd/0x20
? do_set_thread_area+0x19c/0x1f0
SyS_set_thread_area+0x24/0x30
do_int80_syscall_32+0x7c/0x220
entry_INT80_compat+0x38/0x50
... can be reproduced by running the GS testcase of the ldt_gdt test unit in
the x86 selftests.
do_int80_syscall_32() will call enter_form_user_mode() to convert context
tracking state from user state to kernel state. The load_gs_index() call
can fail with user gsbase, gsbase will be fixed up and proceed if this
happen.
However, enter_from_user_mode() will be called again in the fixed up path
though it is context tracking kernel state currently.
This patch fixes it by just fixing up gsbase and telling lockdep that IRQs
are off once load_gs_index() failed with user gsbase.
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Acked-by: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
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/1475197266-3440-1-git-send-email-wanpeng.li@hotmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Otherwise arch_task_struct_size == 0 and we die. While we're at it,
set X86_FEATURE_ALWAYS, too.
Reported-by: David Saggiorato <david@saggiorato.net>
Tested-by: David Saggiorato <david@saggiorato.net>
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Dave Hansen <dave@sr71.net>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Fixes: aaeb5c01c5b ("x86/fpu, sched: Introduce CONFIG_ARCH_WANTS_DYNAMIC_TASK_STRUCT and use it on x86")
Link: http://lkml.kernel.org/r/8de723afbf0811071185039f9088733188b606c9.1475103911.git.luto@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
We use __read_cr4() vs __read_cr4_safe() inconsistently. On
CR4-less CPUs, all CR4 bits are effectively clear, so we can make
the code simpler and more robust by making __read_cr4() always fix
up faults on 32-bit kernels.
This may fix some bugs on old 486-like CPUs, but I don't have any
easy way to test that.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: david@saggiorato.net
Link: http://lkml.kernel.org/r/ea647033d357d9ce2ad2bbde5a631045f5052fb6.1475178370.git.luto@kernel.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
We need to call GET_LE to read hdr->e_type.
Fixes: 57f90c3dfc ("x86/vdso: Error out if the vDSO isn't a valid DSO")
Reported-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Segher Boessenkool <segher@kernel.crashing.org>
Acked-by: Andy Lutomirski <luto@kernel.org>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: linux-next@vger.kernel.org
Link: http://lkml.kernel.org/r/20160929193442.GA16617@gate.crashing.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The condition for reading CR4 was wrong: there are some CPUs with
CPUID but not CR4. Rather than trying to make the condition exact,
use __read_cr4_safe().
Fixes: 18bc7bd523 ("x86/boot: Synchronize trampoline_cr4_features and mmu_cr4_features directly")
Reported-by: david@saggiorato.net
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Reviewed-by: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Link: http://lkml.kernel.org/r/8c453a61c4f44ab6ff43c29780ba04835234d2e5.1475178369.git.luto@kernel.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The code performing irqtime nsecs stats flushing to kcpustat is roughly
the same for hardirq and softirq. So lets consolidate that common code.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Wanpeng Li <wanpeng.li@hotmail.com>
Link: http://lkml.kernel.org/r/1474849761-12678-6-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The irqtime accounting currently implement its own ad hoc implementation
of u64_stats API. Lets rather consolidate it with the appropriate
library.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Wanpeng Li <wanpeng.li@hotmail.com>
Link: http://lkml.kernel.org/r/1474849761-12678-5-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Introduce light versions of u64_stats helpers for context where
either preempt or IRQs are disabled. This way we can make this library
usable by scheduler irqtime accounting which currenty implement its
ad-hoc version.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Wanpeng Li <wanpeng.li@hotmail.com>
Link: http://lkml.kernel.org/r/1474849761-12678-4-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The callers of the functions performing irqtime kcpustat updates have
IRQS disabled, no need to disable them again.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Wanpeng Li <wanpeng.li@hotmail.com>
Link: http://lkml.kernel.org/r/1474849761-12678-3-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
We can safely use the preempt-unsafe accessors for irqtime when we
flush its counters to kcpustat as IRQs are disabled at this time.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Wanpeng Li <wanpeng.li@hotmail.com>
Link: http://lkml.kernel.org/r/1474849761-12678-2-git-send-email-fweisbec@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
While going through enqueue/dequeue to review the movement of
set_curr_task() I noticed that the (2nd) update_min_vruntime() call in
dequeue_entity() is suspect.
It turns out, its actually wrong because it will consider
cfs_rq->curr, which could be the entry we just normalized. This mixes
different vruntime forms and leads to fail.
The purpose of the second update_min_vruntime() is to move
min_vruntime forward if the entity we just removed is the one that was
holding it back; _except_ for the DEQUEUE_SAVE case, because then we
know its a temporary removal and it will come back.
However, since we do put_prev_task() _after_ dequeue(), cfs_rq->curr
will still be set (and per the above, can be tranformed into a
different unit), so update_min_vruntime() should also consider
curr->on_rq. This also fixes another corner case where the enqueue
(which also does update_curr()->update_min_vruntime()) happens on the
rq->lock break in schedule(), between dequeue and put_prev_task.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Fixes: 1e87623178 ("sched: Fix ->min_vruntime calculation in dequeue_entity()")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Provide SCHED_WARN_ON as wrapper for WARN_ON_ONCE() to avoid
CONFIG_SCHED_DEBUG wrappery.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Almost all scheduler functions update state with the following
pattern:
if (queued)
dequeue_task(rq, p, DEQUEUE_SAVE);
if (running)
put_prev_task(rq, p);
/* update state */
if (queued)
enqueue_task(rq, p, ENQUEUE_RESTORE);
if (running)
set_curr_task(rq, p);
set_user_nice() however misses the running part, cure this.
This was found by asserting we never enqueue 'current'.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Now that the ia64 only set_curr_task() symbol is gone, provide a
helper just like put_prev_task().
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Rename the ia64 only set_curr_task() function to free up the name.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tony Luck <tony.luck@intel.com>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When a task switches to fair scheduling class, the period between now
and the last update of its utilization is accounted as running time
whatever happened during this period. This incorrect accounting applies
to the task and also to the task group branch.
When changing the property of a running task like its list of allowed
CPUs or its scheduling class, we follow the sequence:
- dequeue task
- put task
- change the property
- set task as current task
- enqueue task
The end of the sequence doesn't follow the normal sequence (as per
__schedule()) which is:
- enqueue a task
- then set the task as current task.
This incorrectordering is the root cause of incorrect utilization accounting.
Update the sequence to follow the right one:
- dequeue task
- put task
- change the property
- enqueue task
- set task as current task
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Morten.Rasmussen@arm.com
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: bsegall@google.com
Cc: dietmar.eggemann@arm.com
Cc: linaro-kernel@lists.linaro.org
Cc: pjt@google.com
Cc: yuyang.du@intel.com
Link: http://lkml.kernel.org/r/1473666472-13749-8-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
select_idle_siblings() is a known pain point for a number of
workloads; it either does too much or not enough and sometimes just
does plain wrong.
This rewrite attempts to address a number of issues (but sadly not
all).
The current code does an unconditional sched_domain iteration; with
the intent of finding an idle core (on SMT hardware). The problems
which this patch tries to address are:
- its pointless to look for idle cores if the machine is real busy;
at which point you're just wasting cycles.
- it's behaviour is inconsistent between SMT and !SMT hardware in
that !SMT hardware ends up doing a scan for any idle CPU in the LLC
domain, while SMT hardware does a scan for idle cores and if that
fails, falls back to a scan for idle threads on the 'target' core.
The new code replaces the sched_domain scan with 3 explicit scans:
1) search for an idle core in the LLC
2) search for an idle CPU in the LLC
3) search for an idle thread in the 'target' core
where 1 and 3 are conditional on SMT support and 1 and 2 have runtime
heuristics to skip the step.
Step 1) is conditional on sd_llc_shared->has_idle_cores; when a cpu
goes idle and sd_llc_shared->has_idle_cores is false, we scan all SMT
siblings of the CPU going idle. Similarly, we clear
sd_llc_shared->has_idle_cores when we fail to find an idle core.
Step 2) tracks the average cost of the scan and compares this to the
average idle time guestimate for the CPU doing the wakeup. There is a
significant fudge factor involved to deal with the variability of the
averages. Esp. hackbench was sensitive to this.
Step 3) is unconditional; we assume (also per step 1) that scanning
all SMT siblings in a core is 'cheap'.
With this; SMT systems gain step 2, which cures a few benchmarks --
notably one from Facebook.
One 'feature' of the sched_domain iteration, which we preserve in the
new code, is that it would start scanning from the 'target' CPU,
instead of scanning the cpumask in cpu id order. This avoids multiple
CPUs in the LLC scanning for idle to gang up and find the same CPU
quite as much. The down side is that tasks can end up hopping across
the LLC for no apparent reason.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
cmpxchg contained definitions for unused (x)add_* operations, dating back
to the original ticket spinlock implementation. Nowadays these are
unused so remove them.
Signed-off-by: Nikolay Borisov <n.borisov.lkml@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: hpa@zytor.com
Link: http://lkml.kernel.org/r/1474913478-17757-1-git-send-email-n.borisov.lkml@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Move the nr_busy_cpus thing from its hacky sd->parent->groups->sgc
location into the much more natural sched_domain_shared location.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Since struct sched_domain is strictly per cpu; introduce a structure
that is shared between all 'identical' sched_domains.
Limit to SD_SHARE_PKG_RESOURCES domains for now, as we'll only use it
for shared cache state; if another use comes up later we can easily
relax this.
While the sched_group's are normally shared between CPUs, these are
not natural to use when we need some shared state on a domain level --
since that would require the domain to have a parent, which is not a
given.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
There is no point in doing a call_rcu() for each domain, only do a
callback for the root sched domain and clean up the entire set in one
go.
Also make the entire call chain be called destroy_sched_domain*() to
remove confusion with the free_sched_domains() call, which does an
entirely different thing.
Both cpu_attach_domain() callers of destroy_sched_domain() can live
without the call_rcu() because at those points the sched_domain hasn't
been published yet.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Small cleanup; nothing uses the @cpu argument so make it go away.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The partial initialization of wait_queue_t in prepare_to_wait_event() looks
ugly. This was done to shrink .text, but we can simply add the new helper
which does the full initialization and shrink the compiled code a bit more.
And. This way prepare_to_wait_event() can have more users. In particular we
are ready to remove the signal_pending_state() checks from wait_bit_action_f
helpers and change __wait_on_bit_lock() to use prepare_to_wait_event().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Cc: Bart Van Assche <bvanassche@acm.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Neil Brown <neilb@suse.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20160906140055.GA6167@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
__wait_on_bit_lock() doesn't need abort_exclusive_wait() too. Right
now it can't use prepare_to_wait_event() (see the next change), but
it can do the additional finish_wait() if action() fails.
abort_exclusive_wait() no longer has callers, remove it.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Al Viro <viro@ZenIV.linux.org.uk>
Cc: Bart Van Assche <bvanassche@acm.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Neil Brown <neilb@suse.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20160906140053.GA6164@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
