For x86 hybrid CPUs like Alder Lake, the order of CPU selection should
be based strictly on CPU priority. Don't include cluster topology for
hybrid CPUs to avoid interference with such CPU selection order.
On Alder Lake, the Atom CPU cluster has more capacity (4 Atom CPUs) vs
Big core cluster (2 hyperthread CPUs). This could potentially bias CPU
selection towards Atom over Big Core, when Big core CPU has higher
priority.
Fixes: 66558b730f ("sched: Add cluster scheduler level for x86")
Suggested-by: Tim Chen <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Acked-by: Tim Chen <[email protected]>
Tested-by: Ricardo Neri <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Boris reported that in one of his randconfig builds, objtool got
infinitely stuck. Turns out there's trivial list corruption in the
pv_ops tracking when a function is both in a static table and in a code
assignment.
Avoid re-adding function to the pv_ops[] lists when they're already on
it.
Fixes: db2b0c5d7b ("objtool: Support pv_opsindirect calls for noinstr")
Reported-by: Borislav Petkov <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Tested-by: Borislav Petkov <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Add a few signature bytes after the static call trampoline and verify
those bytes match before patching the trampoline. This avoids patching
random other JMPs (such as CFI jump-table entries) instead.
These bytes decode as:
d: 53 push %rbx
e: 43 54 rex.XB push %r12
And happen to spell "SCT".
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
The vmalloc guard pages are added on top of each allocation, thereby
isolating any two allocations from one another. The top guard of the
lower allocation is the bottom guard guard of the higher allocation etc.
Therefore VM_NO_GUARD is dangerous; it breaks the basic premise of
isolating separate allocations.
There are only two in-tree users of this flag, neither of which use it
through the exported interface. Ensure it stays this way.
Link: https://lkml.kernel.org/r/YUMfdA36fuyZ+/[email protected]
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Reviewed-by: David Hildenbrand <[email protected]>
Acked-by: Will Deacon <[email protected]>
Acked-by: Kees Cook <[email protected]>
Cc: Andrey Konovalov <[email protected]>
Cc: Mel Gorman <[email protected]>
Cc: Uladzislau Rezki <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
Current BPF codegen doesn't respect X86_FEATURE_RETPOLINE* flags and
unconditionally emits a thunk call, this is sub-optimal and doesn't
match the regular, compiler generated, code.
Update the i386 JIT to emit code equal to what the compiler emits for
the regular kernel text (IOW. a plain THUNK call).
Update the x86_64 JIT to emit code similar to the result of compiler
and kernel rewrites as according to X86_FEATURE_RETPOLINE* flags.
Inlining RETPOLINE_AMD (lfence; jmp *%reg) and !RETPOLINE (jmp *%reg),
while doing a THUNK call for RETPOLINE.
This removes the hard-coded retpoline thunks and shrinks the generated
code. Leaving a single retpoline thunk definition in the kernel.
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Borislav Petkov <[email protected]>
Acked-by: Alexei Starovoitov <[email protected]>
Acked-by: Josh Poimboeuf <[email protected]>
Tested-by: Alexei Starovoitov <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Take an idea from the 32bit JIT, which uses the multi-pass nature of
the JIT to compute the instruction offsets on a prior pass in order to
compute the relative jump offsets on a later pass.
Application to the x86_64 JIT is slightly more involved because the
offsets depend on program variables (such as callee_regs_used and
stack_depth) and hence the computed offsets need to be kept in the
context of the JIT.
This removes, IMO quite fragile, code that hard-codes the offsets and
tries to compute the length of variable parts of it.
Convert both emit_bpf_tail_call_*() functions which have an out: label
at the end. Additionally emit_bpt_tail_call_direct() also has a poke
table entry, for which it computes the offset from the end (and thus
already relies on the previous pass to have computed addrs[i]), also
convert this to be a forward based offset.
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Borislav Petkov <[email protected]>
Acked-by: Alexei Starovoitov <[email protected]>
Acked-by: Josh Poimboeuf <[email protected]>
Tested-by: Alexei Starovoitov <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Try and replace retpoline thunk calls with:
LFENCE
CALL *%\reg
for spectre_v2=retpoline,amd.
Specifically, the sequence above is 5 bytes for the low 8 registers,
but 6 bytes for the high 8 registers. This means that unless the
compilers prefix stuff the call with higher registers this replacement
will fail.
Luckily GCC strongly favours RAX for the indirect calls and most (95%+
for defconfig-x86_64) will be converted. OTOH clang strongly favours
R11 and almost nothing gets converted.
Note: it will also generate a correct replacement for the Jcc.d32
case, except unless the compilers start to prefix stuff that, it'll
never fit. Specifically:
Jncc.d8 1f
LFENCE
JMP *%\reg
1:
is 7-8 bytes long, where the original instruction in unpadded form is
only 6 bytes.
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Borislav Petkov <[email protected]>
Acked-by: Josh Poimboeuf <[email protected]>
Tested-by: Alexei Starovoitov <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Handle the rare cases where the compiler (clang) does an indirect
conditional tail-call using:
Jcc __x86_indirect_thunk_\reg
For the !RETPOLINE case this can be rewritten to fit the original (6
byte) instruction like:
Jncc.d8 1f
JMP *%\reg
NOP
1:
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Borislav Petkov <[email protected]>
Acked-by: Josh Poimboeuf <[email protected]>
Tested-by: Alexei Starovoitov <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Rewrite retpoline thunk call sites to be indirect calls for
spectre_v2=off. This ensures spectre_v2=off is as near to a
RETPOLINE=n build as possible.
This is the replacement for objtool writing alternative entries to
ensure the same and achieves feature-parity with the previous
approach.
One noteworthy feature is that it relies on the thunks to be in
machine order to compute the register index.
Specifically, this does not yet address the Jcc __x86_indirect_thunk_*
calls generated by clang, a future patch will add this.
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Borislav Petkov <[email protected]>
Acked-by: Josh Poimboeuf <[email protected]>
Tested-by: Alexei Starovoitov <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Stick all the retpolines in a single symbol and have the individual
thunks as inner labels, this should guarantee thunk order and layout.
Previously there were 16 (or rather 15 without rsp) separate symbols and
a toolchain might reasonably expect it could displace them however it
liked, with disregard for their relative position.
However, now they're part of a larger symbol. Any change to their
relative position would disrupt this larger _array symbol and thus not
be sound.
This is the same reasoning used for data symbols. On their own there
is no guarantee about their relative position wrt to one aonther, but
we're still able to do arrays because an array as a whole is a single
larger symbol.
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Borislav Petkov <[email protected]>
Acked-by: Josh Poimboeuf <[email protected]>
Tested-by: Alexei Starovoitov <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Any one instruction can only ever call a single function, therefore
insn->mcount_loc_node is superfluous and can use insn->call_node.
This shrinks struct instruction, which is by far the most numerous
structure objtool creates.
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Borislav Petkov <[email protected]>
Acked-by: Josh Poimboeuf <[email protected]>
Tested-by: Alexei Starovoitov <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Currently AMD/Hygon do not populate l2c_id, this means that for SMT
enabled systems they report an L2 per thread. This is ofcourse not
true but was harmless so far.
However, since commit: 66558b730f ("sched: Add cluster scheduler
level for x86") the scheduler topology setup requires:
SMT <= L2 <= LLC
Which leads to noisy warnings and possibly weird behaviour on affected
chips.
Therefore change the topology generation such that if l2c_id is not
populated it follows the SMT topology, thereby satisfying the
constraint.
Fixes: 66558b730f ("sched: Add cluster scheduler level for x86")
Reported-by: Tom Lendacky <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Tested-by: Tom Lendacky <[email protected]>
As reported by syzbot and experienced by Pavel, using cpus_read_lock()
in wake_up_all_idle_cpus() generates lock inversion (against mmap_sem
and possibly others).
Instead, shrink the preempt disable region by iterating all CPUs and
checking the online status for each individual CPU while having
preemption disabled.
Fixes: 8850cb663b ("sched: Simplify wake_up_*idle*()")
Reported-by: [email protected]
Reported-by: Pavel Machek <[email protected]>
Reported-by: Qian Cai <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Tested-by: Qian Cai <[email protected]>
Mike reported that rcuwait went walk-about and is causing failures on
the PREEMPT_RT builds, restore it.
Reported-by: Mike Galbraith <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
The compilers can't deal with obvious DCE vs that warning, resulting
in code like:
if (0) {
sched sched_statistics *stats;
stats = __schedstats_from_se(se);
...
}
triggering the warning. Kill the warning to make the robots stop
reporting this.
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Nathan Chancellor <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
vmlinux.o: warning: objtool: rcu_nmi_enter()+0x36: call to __kasan_check_read() leaves .noinstr.text section
noinstr cannot have atomic_*() functions in because they're explicitly
annotated, use arch_atomic_*().
Fixes: 2be57f7328 ("rcu: Weaken ->dynticks accesses and updates")
Reported-by: Stephen Rothwell <[email protected]>
Reviewed-by: Thomas Gleixner <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Signed-off-by: Paul E. McKenney <[email protected]>
RCU managed to grow a few noinstr violations:
vmlinux.o: warning: objtool: rcu_dynticks_eqs_enter()+0x0: call to rcu_dynticks_task_trace_enter() leaves .noinstr.text section
vmlinux.o: warning: objtool: rcu_dynticks_eqs_exit()+0xe: call to rcu_dynticks_task_trace_exit() leaves .noinstr.text section
Fix them by adding __always_inline to the relevant trivial functions.
Also replace the noinstr with __always_inline for the existing
rcu_dynticks_task_*() functions since noinstr would force noinline
them, even when empty, which seems silly.
Fixes: 7d0c9c50c5 ("rcu-tasks: Avoid IPIing userspace/idle tasks if kernel is so built")
Reported-by: Stephen Rothwell <[email protected]>
Reviewed-by: Thomas Gleixner <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Signed-off-by: Paul E. McKenney <[email protected]>
Simplify and make wake_up_if_idle() more robust, also don't iterate
the whole machine with preempt_disable() in it's caller:
wake_up_all_idle_cpus().
This prepares for another wake_up_if_idle() user that needs a full
do_idle() cycle.
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Acked-by: Vasily Gorbik <[email protected]>
Tested-by: Vasily Gorbik <[email protected]> # on s390
Link: https://lkml.kernel.org/r/[email protected]
Clarify and tighten try_invoke_on_locked_down_task().
Basically the function calls @func under task_rq_lock(), except it
avoids taking rq->lock when possible.
This makes calling @func unconditional (the function will get renamed
in a later patch to remove the try).
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Acked-by: Vasily Gorbik <[email protected]>
Tested-by: Vasily Gorbik <[email protected]> # on s390
Link: https://lkml.kernel.org/r/[email protected]
We need to make sure that all requeue operations take the hash bucket
locks in the same order to avoid deadlock. Simplify the current
double_lock_hb implementation by making sure hb1 is always the
"smallest" bucket to avoid extra checks.
[André: Add commit description]
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Signed-off-by: André Almeida <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: André Almeida <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
When !SCHEDSTATS schedstat_enabled() is an unconditional 0 and the
whole block doesn't exist, however GCC figures the scoped variable
'stats' is unused and complains about it.
Upgrade the warning from -Wunused-variable to -Wunused-but-set-variable
by writing it in two statements. This fixes the build because the new
warning is in W=1.
Given that whole if(0) {} thing, I don't feel motivated to change
things overly much and quite strongly feel this is the compiler being
daft.
Fixes: cb3e971c435d ("sched: Make struct sched_statistics independent of fair sched class")
Reported-by: Stephen Rothwell <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Tao suggested a two-pass task selection to avoid the retry loop.
Not only does it avoid the retry loop, it results in *much* simpler
code.
This also fixes an issue spotted by Josh Don where, for SMT3+, we can
forget to update max on the first pass and get to do an extra round.
Suggested-by: Tao Zhou <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Josh Don <[email protected]>
Reviewed-by: Vineeth Pillai (Microsoft) <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Occasionally objtool encounters symbol (as opposed to section)
relocations in .altinstructions. Typically they are the alternatives
written by elf_add_alternative() as encountered on a noinstr
validation run on vmlinux after having already ran objtool on the
individual .o files.
Basically this is the counterpart of commit 44f6a7c075 ("objtool:
Fix seg fault with Clang non-section symbols"), because when these new
assemblers (binutils now also does this) strip the section symbols,
elf_add_reloc_to_insn() is forced to emit symbol based relocations.
As such, teach get_alt_entry() about different relocation types.
Fixes: 9bc0bb5072 ("objtool/x86: Rewrite retpoline thunk calls")
Reported-by: Stephen Rothwell <[email protected]>
Reported-by: Borislav Petkov <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Acked-by: Josh Poimboeuf <[email protected]>
Tested-by: Nathan Chancellor <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
It turns out that a single page of stack is trivial to overflow with
all the tracing gunk enabled. Raise the exception stacks to 2 pages,
which is still half the interrupt stacks, which are at 4 pages.
Reported-by: Michael Wang <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Current code has an explicit check for hitting the task stack guard;
but overflowing any of the other stacks will get you a non-descript
general #DF warning.
Improve matters by using get_stack_info_noinstr() to detetrmine if and
which stack guard page got hit, enabling a better stack warning.
In specific, Michael Wang reported what turned out to be an NMI
exception stack overflow, which is now clearly reported as such:
[] BUG: NMI stack guard page was hit at 0000000085fd977b (stack is 000000003a55b09e..00000000d8cce1a5)
Reported-by: Michael Wang <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Tested-by: Michael Wang <[email protected]>
Link: https://lkml.kernel.org/r/YUTE/[email protected]
Since commit c8137ace56 ("x86/iopl: Restrict iopl() permission
scope") it's possible to emulate iopl(3) using ioperm(), except for
the CLI/STI usage.
Userspace CLI/STI usage is very dubious (read broken), since any
exception taken during that window can lead to rescheduling anyway (or
worse). The IOPL(2) manpage even states that usage of CLI/STI is highly
discouraged and might even crash the system.
Of course, that won't stop people and HP has the dubious honour of
being the first vendor to be found using this in their hp-health
package.
In order to enable this 'software' to still 'work', have the #GP treat
the CLI/STI instructions as NOPs when iopl(3). Warn the user that
their program is doing dubious things.
Fixes: a24ca99768 ("x86/iopl: Remove legacy IOPL option")
Reported-by: Ondrej Zary <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Thomas Gleixner <[email protected]>
Cc: [email protected] # v5.5+
Link: https://lkml.kernel.org/r/[email protected]
Normally objtool will now follow indirect calls; there is no need.
However, this becomes a problem with noinstr validation; if there's an
indirect call from noinstr code, we very much need to know it is to
another noinstr function. Luckily there aren't many indirect calls in
entry code with the obvious exception of paravirt. As such, noinstr
validation didn't work with paravirt kernels.
In order to track pv_ops[] call targets, objtool reads the static
pv_ops[] tables as well as direct assignments to the pv_ops[] array,
provided the compiler makes them a single instruction like:
bf87: 48 c7 05 00 00 00 00 00 00 00 00 movq $0x0,0x0(%rip)
bf92 <xen_init_spinlocks+0x5f>
bf8a: R_X86_64_PC32 pv_ops+0x268
There are, as of yet, no warnings for when this goes wrong :/
Using the functions found with the above means, all pv_ops[] calls are
now subject to noinstr validation.
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
In order to allow objtool to make sense of all the various paravirt
functions, it needs to either parse whole pv_ops[] tables, or observe
individual assignments in the form:
bf87: 48 c7 05 00 00 00 00 00 00 00 00 movq $0x0,0x0(%rip)
bf92 <xen_init_spinlocks+0x5f>
bf8a: R_X86_64_PC32 pv_ops+0x268
As is, xen_cpu_ops[] is at offset +0 in pv_ops[] and could thus be
parsed as a 'normal' pv_ops[] table, however xen_irq_ops[] and
xen_mmu_ops[] are not.
Worse, both the latter two are compiled into the individual assignment
for by current GCC, but that's not something one can rely on.
Therefore, convert all three into full pv_ops[] tables. This has the
benefit of not needing to teach objtool about the offsets and
resulting in more conservative code-gen.
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Juergen Gross <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
The code in rwbase_write_lock() is a little non-obvious vs the
read+set 'trylock', extract the sequence into a helper function to
clarify the code.
This also provides a single site to fix fast-path ordering.
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Thomas Gleixner <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Andi reported that objtool on vmlinux.o consumes more memory than his
system has, leading to horrific performance.
This is in part because we keep a struct instruction for every
instruction in the file in-memory. Shrink struct instruction by
removing the CFI state (which includes full register state) from it
and demand allocating it.
Given most instructions don't actually change CFI state, there's lots
of repetition there, so add a hash table to find previous CFI
instances.
Reduces memory consumption (and runtime) for processing an
x86_64-allyesconfig:
pre: 4:40.84 real, 143.99 user, 44.18 sys, 30624988 mem
post: 2:14.61 real, 108.58 user, 25.04 sys, 16396184 mem
Suggested-by: Andi Kleen <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
The asm_cpu_bringup_and_idle() function is required to push the return
value on the stack in order to make ORC happy, but the only reason
objtool doesn't complain is because of a happy accident.
The thing is that asm_cpu_bringup_and_idle() doesn't return, so
validate_branch() never terminates and falls through to the next
function, which in the normal case is the hypercall_page. And that, as
it happens, is 4095 NOPs and a RET.
Make asm_cpu_bringup_and_idle() terminate on it's own, by making the
function it calls as a dead-end. This way we no longer rely on what
code happens to come after.
Fixes: c3881eb58d ("x86/xen: Make the secondary CPU idle tasks reliable")
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Juergen Gross <[email protected]>
Reviewed-by: Miroslav Benes <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Dan reported that rt_mutex_adjust_prio_chain() can be called with
.orig_waiter == NULL however commit a055fcc132 ("locking/rtmutex: Return
success on deadlock for ww_mutex waiters") unconditionally dereferences it.
Since both call-sites that have .orig_waiter == NULL don't care for the
return value, simply disable the deadlock squash by adding the NULL check.
Notably, both callers use the deadlock condition as a termination condition
for the iteration; once detected, it is sure that (de)boosting is done.
Arguably step [3] would be a more natural termination point, but it's
dubious whether adding a third deadlock detection state would improve the
code.
Fixes: a055fcc132 ("locking/rtmutex: Return success on deadlock for ww_mutex waiters")
Reported-by: Dan Carpenter <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Acked-by: Sebastian Andrzej Siewior <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
ww_mutexes can legitimately cause a deadlock situation in the lock graph
which is resolved afterwards by the wait/wound mechanics. The rtmutex chain
walk can detect such a deadlock and returns EDEADLK which in turn skips the
wait/wound mechanism and returns EDEADLK to the caller. That's wrong
because both lock chains might get EDEADLK or the wrong waiter would back
out.
Detect that situation and return 'success' in case that the waiter which
initiated the chain walk is a ww_mutex with context. This allows the
wait/wound mechanics to resolve the situation according to the rules.
[ tglx: Split it apart and added changelog ]
Reported-by: Sebastian Siewior <[email protected]>
Fixes: add461325e ("locking/rtmutex: Extend the rtmutex core to support ww_mutex")
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
rtmutex based ww_mutexes can legitimately create a cycle in the lock graph
which can be observed by a blocker which didn't cause the problem:
P1: A, ww_A, ww_B
P2: ww_B, ww_A
P3: A
P3 might therefore be trapped in the ww_mutex induced cycle and run into
the lock depth limitation of rt_mutex_adjust_prio_chain() which returns
-EDEADLK to the caller.
Disable the deadlock detection walk when the chain walk observes a
ww_mutex to prevent this looping.
[ tglx: Split it apart and added changelog ]
Reported-by: Sebastian Siewior <[email protected]>
Fixes: add461325e ("locking/rtmutex: Extend the rtmutex core to support ww_mutex")
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Eugene tripped over the case where rq_lock(), as called in a
for_each_possible_cpu() loop came apart because rq->core hadn't been
setup yet.
This is a somewhat unusual, but valid case.
Rework things such that rq->core is initialized to point at itself. IOW
initialize each CPU as a single threaded Core. CPU online will then join
the new CPU (thread) to an existing Core where needed.
For completeness sake, have CPU offline fully undo the state so as to
not presume the topology will match the next time it comes online.
Fixes: 9edeaea1bc ("sched: Core-wide rq->lock")
Reported-by: Eugene Syromiatnikov <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Reviewed-by: Josh Don <[email protected]>
Tested-by: Eugene Syromiatnikov <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Add a ww acquire context pointer to the waiter and various functions and
add the ww_mutex related invocations to the proper spots in the locking
code, similar to the mutex based variant.
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Some ww_mutex helper functions use pointers for the underlying mutex and
mutex_waiter. The upcoming rtmutex based implementation needs to share
these functions. Add and use defines for the types and replace the direct
types in the affected functions.
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Signed-off-by: Thomas Gleixner <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Peter Zijlstra