Merge branches 'doc.2015.07.15a' and 'torture.2015.07.15a' into HEAD

doc.2015.07.15a: Documentation updates.
torture.2015.07.15a: Torture-test updates.
This commit is contained in:
Paul E. McKenney 2015-08-04 08:42:02 -07:00
commit 3dbe43f6fb
5 changed files with 236 additions and 200 deletions

View File

@ -28,7 +28,7 @@ o You must use one of the rcu_dereference() family of primitives
o Avoid cancellation when using the "+" and "-" infix arithmetic
operators. For example, for a given variable "x", avoid
"(x-x)". There are similar arithmetic pitfalls from other
arithmetic operatiors, such as "(x*0)", "(x/(x+1))" or "(x%1)".
arithmetic operators, such as "(x*0)", "(x/(x+1))" or "(x%1)".
The compiler is within its rights to substitute zero for all of
these expressions, so that subsequent accesses no longer depend
on the rcu_dereference(), again possibly resulting in bugs due

View File

@ -3135,22 +3135,35 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
in a given burst of a callback-flood test.
rcutorture.fqs_duration= [KNL]
Set duration of force_quiescent_state bursts.
Set duration of force_quiescent_state bursts
in microseconds.
rcutorture.fqs_holdoff= [KNL]
Set holdoff time within force_quiescent_state bursts.
Set holdoff time within force_quiescent_state bursts
in microseconds.
rcutorture.fqs_stutter= [KNL]
Set wait time between force_quiescent_state bursts.
Set wait time between force_quiescent_state bursts
in seconds.
rcutorture.gp_cond= [KNL]
Use conditional/asynchronous update-side
primitives, if available.
rcutorture.gp_exp= [KNL]
Use expedited update-side primitives.
Use expedited update-side primitives, if available.
rcutorture.gp_normal= [KNL]
Use normal (non-expedited) update-side primitives.
If both gp_exp and gp_normal are set, do both.
If neither gp_exp nor gp_normal are set, still
do both.
Use normal (non-expedited) asynchronous
update-side primitives, if available.
rcutorture.gp_sync= [KNL]
Use normal (non-expedited) synchronous
update-side primitives, if available. If all
of rcutorture.gp_cond=, rcutorture.gp_exp=,
rcutorture.gp_normal=, and rcutorture.gp_sync=
are zero, rcutorture acts as if is interpreted
they are all non-zero.
rcutorture.n_barrier_cbs= [KNL]
Set callbacks/threads for rcu_barrier() testing.
@ -3177,9 +3190,6 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
Set time (s) between CPU-hotplug operations, or
zero to disable CPU-hotplug testing.
rcutorture.torture_runnable= [BOOT]
Start rcutorture running at boot time.
rcutorture.shuffle_interval= [KNL]
Set task-shuffle interval (s). Shuffling tasks
allows some CPUs to go into dyntick-idle mode
@ -3220,6 +3230,9 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
Test RCU's dyntick-idle handling. See also the
rcutorture.shuffle_interval parameter.
rcutorture.torture_runnable= [BOOT]
Start rcutorture running at boot time.
rcutorture.torture_type= [KNL]
Specify the RCU implementation to test.

View File

@ -194,22 +194,22 @@ There are some minimal guarantees that may be expected of a CPU:
(*) On any given CPU, dependent memory accesses will be issued in order, with
respect to itself. This means that for:
ACCESS_ONCE(Q) = P; smp_read_barrier_depends(); D = ACCESS_ONCE(*Q);
WRITE_ONCE(Q, P); smp_read_barrier_depends(); D = READ_ONCE(*Q);
the CPU will issue the following memory operations:
Q = LOAD P, D = LOAD *Q
and always in that order. On most systems, smp_read_barrier_depends()
does nothing, but it is required for DEC Alpha. The ACCESS_ONCE()
is required to prevent compiler mischief. Please note that you
should normally use something like rcu_dereference() instead of
open-coding smp_read_barrier_depends().
does nothing, but it is required for DEC Alpha. The READ_ONCE()
and WRITE_ONCE() are required to prevent compiler mischief. Please
note that you should normally use something like rcu_dereference()
instead of open-coding smp_read_barrier_depends().
(*) Overlapping loads and stores within a particular CPU will appear to be
ordered within that CPU. This means that for:
a = ACCESS_ONCE(*X); ACCESS_ONCE(*X) = b;
a = READ_ONCE(*X); WRITE_ONCE(*X, b);
the CPU will only issue the following sequence of memory operations:
@ -217,7 +217,7 @@ There are some minimal guarantees that may be expected of a CPU:
And for:
ACCESS_ONCE(*X) = c; d = ACCESS_ONCE(*X);
WRITE_ONCE(*X, c); d = READ_ONCE(*X);
the CPU will only issue:
@ -228,11 +228,11 @@ There are some minimal guarantees that may be expected of a CPU:
And there are a number of things that _must_ or _must_not_ be assumed:
(*) It _must_not_ be assumed that the compiler will do what you want with
memory references that are not protected by ACCESS_ONCE(). Without
ACCESS_ONCE(), the compiler is within its rights to do all sorts
of "creative" transformations, which are covered in the Compiler
Barrier section.
(*) It _must_not_ be assumed that the compiler will do what you want
with memory references that are not protected by READ_ONCE() and
WRITE_ONCE(). Without them, the compiler is within its rights to
do all sorts of "creative" transformations, which are covered in
the Compiler Barrier section.
(*) It _must_not_ be assumed that independent loads and stores will be issued
in the order given. This means that for:
@ -520,8 +520,8 @@ following sequence of events:
{ A == 1, B == 2, C = 3, P == &A, Q == &C }
B = 4;
<write barrier>
ACCESS_ONCE(P) = &B
Q = ACCESS_ONCE(P);
WRITE_ONCE(P, &B)
Q = READ_ONCE(P);
D = *Q;
There's a clear data dependency here, and it would seem that by the end of the
@ -547,8 +547,8 @@ between the address load and the data load:
{ A == 1, B == 2, C = 3, P == &A, Q == &C }
B = 4;
<write barrier>
ACCESS_ONCE(P) = &B
Q = ACCESS_ONCE(P);
WRITE_ONCE(P, &B);
Q = READ_ONCE(P);
<data dependency barrier>
D = *Q;
@ -574,8 +574,8 @@ access:
{ M[0] == 1, M[1] == 2, M[3] = 3, P == 0, Q == 3 }
M[1] = 4;
<write barrier>
ACCESS_ONCE(P) = 1
Q = ACCESS_ONCE(P);
WRITE_ONCE(P, 1);
Q = READ_ONCE(P);
<data dependency barrier>
D = M[Q];
@ -596,10 +596,10 @@ A load-load control dependency requires a full read memory barrier, not
simply a data dependency barrier to make it work correctly. Consider the
following bit of code:
q = ACCESS_ONCE(a);
q = READ_ONCE(a);
if (q) {
<data dependency barrier> /* BUG: No data dependency!!! */
p = ACCESS_ONCE(b);
p = READ_ONCE(b);
}
This will not have the desired effect because there is no actual data
@ -608,10 +608,10 @@ by attempting to predict the outcome in advance, so that other CPUs see
the load from b as having happened before the load from a. In such a
case what's actually required is:
q = ACCESS_ONCE(a);
q = READ_ONCE(a);
if (q) {
<read barrier>
p = ACCESS_ONCE(b);
p = READ_ONCE(b);
}
However, stores are not speculated. This means that ordering -is- provided
@ -619,7 +619,7 @@ for load-store control dependencies, as in the following example:
q = READ_ONCE_CTRL(a);
if (q) {
ACCESS_ONCE(b) = p;
WRITE_ONCE(b, p);
}
Control dependencies pair normally with other types of barriers. That
@ -647,11 +647,11 @@ branches of the "if" statement as follows:
q = READ_ONCE_CTRL(a);
if (q) {
barrier();
ACCESS_ONCE(b) = p;
WRITE_ONCE(b, p);
do_something();
} else {
barrier();
ACCESS_ONCE(b) = p;
WRITE_ONCE(b, p);
do_something_else();
}
@ -660,12 +660,12 @@ optimization levels:
q = READ_ONCE_CTRL(a);
barrier();
ACCESS_ONCE(b) = p; /* BUG: No ordering vs. load from a!!! */
WRITE_ONCE(b, p); /* BUG: No ordering vs. load from a!!! */
if (q) {
/* ACCESS_ONCE(b) = p; -- moved up, BUG!!! */
/* WRITE_ONCE(b, p); -- moved up, BUG!!! */
do_something();
} else {
/* ACCESS_ONCE(b) = p; -- moved up, BUG!!! */
/* WRITE_ONCE(b, p); -- moved up, BUG!!! */
do_something_else();
}
@ -676,7 +676,7 @@ assembly code even after all compiler optimizations have been applied.
Therefore, if you need ordering in this example, you need explicit
memory barriers, for example, smp_store_release():
q = ACCESS_ONCE(a);
q = READ_ONCE(a);
if (q) {
smp_store_release(&b, p);
do_something();
@ -690,10 +690,10 @@ ordering is guaranteed only when the stores differ, for example:
q = READ_ONCE_CTRL(a);
if (q) {
ACCESS_ONCE(b) = p;
WRITE_ONCE(b, p);
do_something();
} else {
ACCESS_ONCE(b) = r;
WRITE_ONCE(b, r);
do_something_else();
}
@ -706,10 +706,10 @@ the needed conditional. For example:
q = READ_ONCE_CTRL(a);
if (q % MAX) {
ACCESS_ONCE(b) = p;
WRITE_ONCE(b, p);
do_something();
} else {
ACCESS_ONCE(b) = r;
WRITE_ONCE(b, r);
do_something_else();
}
@ -718,7 +718,7 @@ equal to zero, in which case the compiler is within its rights to
transform the above code into the following:
q = READ_ONCE_CTRL(a);
ACCESS_ONCE(b) = p;
WRITE_ONCE(b, p);
do_something_else();
Given this transformation, the CPU is not required to respect the ordering
@ -731,10 +731,10 @@ one, perhaps as follows:
q = READ_ONCE_CTRL(a);
BUILD_BUG_ON(MAX <= 1); /* Order load from a with store to b. */
if (q % MAX) {
ACCESS_ONCE(b) = p;
WRITE_ONCE(b, p);
do_something();
} else {
ACCESS_ONCE(b) = r;
WRITE_ONCE(b, r);
do_something_else();
}
@ -746,18 +746,18 @@ You must also be careful not to rely too much on boolean short-circuit
evaluation. Consider this example:
q = READ_ONCE_CTRL(a);
if (a || 1 > 0)
ACCESS_ONCE(b) = 1;
if (q || 1 > 0)
WRITE_ONCE(b, 1);
Because the first condition cannot fault and the second condition is
always true, the compiler can transform this example as following,
defeating control dependency:
q = READ_ONCE_CTRL(a);
ACCESS_ONCE(b) = 1;
WRITE_ONCE(b, 1);
This example underscores the need to ensure that the compiler cannot
out-guess your code. More generally, although ACCESS_ONCE() does force
out-guess your code. More generally, although READ_ONCE() does force
the compiler to actually emit code for a given load, it does not force
the compiler to use the results.
@ -769,7 +769,7 @@ x and y both being zero:
======================= =======================
r1 = READ_ONCE_CTRL(x); r2 = READ_ONCE_CTRL(y);
if (r1 > 0) if (r2 > 0)
ACCESS_ONCE(y) = 1; ACCESS_ONCE(x) = 1;
WRITE_ONCE(y, 1); WRITE_ONCE(x, 1);
assert(!(r1 == 1 && r2 == 1));
@ -779,7 +779,7 @@ then adding the following CPU would guarantee a related assertion:
CPU 2
=====================
ACCESS_ONCE(x) = 2;
WRITE_ONCE(x, 2);
assert(!(r1 == 2 && r2 == 1 && x == 2)); /* FAILS!!! */
@ -798,8 +798,7 @@ In summary:
(*) Control dependencies must be headed by READ_ONCE_CTRL().
Or, as a much less preferable alternative, interpose
be headed by READ_ONCE() or an ACCESS_ONCE() read and must
have smp_read_barrier_depends() between this read and the
smp_read_barrier_depends() between a READ_ONCE() and the
control-dependent write.
(*) Control dependencies can order prior loads against later stores.
@ -815,15 +814,16 @@ In summary:
(*) Control dependencies require at least one run-time conditional
between the prior load and the subsequent store, and this
conditional must involve the prior load. If the compiler
is able to optimize the conditional away, it will have also
optimized away the ordering. Careful use of ACCESS_ONCE() can
help to preserve the needed conditional.
conditional must involve the prior load. If the compiler is able
to optimize the conditional away, it will have also optimized
away the ordering. Careful use of READ_ONCE_CTRL() READ_ONCE(),
and WRITE_ONCE() can help to preserve the needed conditional.
(*) Control dependencies require that the compiler avoid reordering the
dependency into nonexistence. Careful use of ACCESS_ONCE() or
barrier() can help to preserve your control dependency. Please
see the Compiler Barrier section for more information.
dependency into nonexistence. Careful use of READ_ONCE_CTRL()
or smp_read_barrier_depends() can help to preserve your control
dependency. Please see the Compiler Barrier section for more
information.
(*) Control dependencies pair normally with other types of barriers.
@ -848,11 +848,11 @@ barrier, an acquire barrier, a release barrier, or a general barrier:
CPU 1 CPU 2
=============== ===============
ACCESS_ONCE(a) = 1;
WRITE_ONCE(a, 1);
<write barrier>
ACCESS_ONCE(b) = 2; x = ACCESS_ONCE(b);
WRITE_ONCE(b, 2); x = READ_ONCE(b);
<read barrier>
y = ACCESS_ONCE(a);
y = READ_ONCE(a);
Or:
@ -860,7 +860,7 @@ Or:
=============== ===============================
a = 1;
<write barrier>
ACCESS_ONCE(b) = &a; x = ACCESS_ONCE(b);
WRITE_ONCE(b, &a); x = READ_ONCE(b);
<data dependency barrier>
y = *x;
@ -868,11 +868,11 @@ Or even:
CPU 1 CPU 2
=============== ===============================
r1 = ACCESS_ONCE(y);
r1 = READ_ONCE(y);
<general barrier>
ACCESS_ONCE(y) = 1; if (r2 = ACCESS_ONCE(x)) {
WRITE_ONCE(y, 1); if (r2 = READ_ONCE(x)) {
<implicit control dependency>
ACCESS_ONCE(y) = 1;
WRITE_ONCE(y, 1);
}
assert(r1 == 0 || r2 == 0);
@ -886,11 +886,11 @@ versa:
CPU 1 CPU 2
=================== ===================
ACCESS_ONCE(a) = 1; }---- --->{ v = ACCESS_ONCE(c);
ACCESS_ONCE(b) = 2; } \ / { w = ACCESS_ONCE(d);
WRITE_ONCE(a, 1); }---- --->{ v = READ_ONCE(c);
WRITE_ONCE(b, 2); } \ / { w = READ_ONCE(d);
<write barrier> \ <read barrier>
ACCESS_ONCE(c) = 3; } / \ { x = ACCESS_ONCE(a);
ACCESS_ONCE(d) = 4; }---- --->{ y = ACCESS_ONCE(b);
WRITE_ONCE(c, 3); } / \ { x = READ_ONCE(a);
WRITE_ONCE(d, 4); }---- --->{ y = READ_ONCE(b);
EXAMPLES OF MEMORY BARRIER SEQUENCES
@ -1340,10 +1340,10 @@ compiler from moving the memory accesses either side of it to the other side:
barrier();
This is a general barrier -- there are no read-read or write-write variants
of barrier(). However, ACCESS_ONCE() can be thought of as a weak form
for barrier() that affects only the specific accesses flagged by the
ACCESS_ONCE().
This is a general barrier -- there are no read-read or write-write
variants of barrier(). However, READ_ONCE() and WRITE_ONCE() can be
thought of as weak forms of barrier() that affect only the specific
accesses flagged by the READ_ONCE() or WRITE_ONCE().
The barrier() function has the following effects:
@ -1355,9 +1355,10 @@ The barrier() function has the following effects:
(*) Within a loop, forces the compiler to load the variables used
in that loop's conditional on each pass through that loop.
The ACCESS_ONCE() function can prevent any number of optimizations that,
while perfectly safe in single-threaded code, can be fatal in concurrent
code. Here are some examples of these sorts of optimizations:
The READ_ONCE() and WRITE_ONCE() functions can prevent any number of
optimizations that, while perfectly safe in single-threaded code, can
be fatal in concurrent code. Here are some examples of these sorts
of optimizations:
(*) The compiler is within its rights to reorder loads and stores
to the same variable, and in some cases, the CPU is within its
@ -1370,11 +1371,11 @@ code. Here are some examples of these sorts of optimizations:
Might result in an older value of x stored in a[1] than in a[0].
Prevent both the compiler and the CPU from doing this as follows:
a[0] = ACCESS_ONCE(x);
a[1] = ACCESS_ONCE(x);
a[0] = READ_ONCE(x);
a[1] = READ_ONCE(x);
In short, ACCESS_ONCE() provides cache coherence for accesses from
multiple CPUs to a single variable.
In short, READ_ONCE() and WRITE_ONCE() provide cache coherence for
accesses from multiple CPUs to a single variable.
(*) The compiler is within its rights to merge successive loads from
the same variable. Such merging can cause the compiler to "optimize"
@ -1391,9 +1392,9 @@ code. Here are some examples of these sorts of optimizations:
for (;;)
do_something_with(tmp);
Use ACCESS_ONCE() to prevent the compiler from doing this to you:
Use READ_ONCE() to prevent the compiler from doing this to you:
while (tmp = ACCESS_ONCE(a))
while (tmp = READ_ONCE(a))
do_something_with(tmp);
(*) The compiler is within its rights to reload a variable, for example,
@ -1415,9 +1416,9 @@ code. Here are some examples of these sorts of optimizations:
a was modified by some other CPU between the "while" statement and
the call to do_something_with().
Again, use ACCESS_ONCE() to prevent the compiler from doing this:
Again, use READ_ONCE() to prevent the compiler from doing this:
while (tmp = ACCESS_ONCE(a))
while (tmp = READ_ONCE(a))
do_something_with(tmp);
Note that if the compiler runs short of registers, it might save
@ -1437,21 +1438,21 @@ code. Here are some examples of these sorts of optimizations:
do { } while (0);
This transformation is a win for single-threaded code because it gets
rid of a load and a branch. The problem is that the compiler will
carry out its proof assuming that the current CPU is the only one
updating variable 'a'. If variable 'a' is shared, then the compiler's
proof will be erroneous. Use ACCESS_ONCE() to tell the compiler
that it doesn't know as much as it thinks it does:
This transformation is a win for single-threaded code because it
gets rid of a load and a branch. The problem is that the compiler
will carry out its proof assuming that the current CPU is the only
one updating variable 'a'. If variable 'a' is shared, then the
compiler's proof will be erroneous. Use READ_ONCE() to tell the
compiler that it doesn't know as much as it thinks it does:
while (tmp = ACCESS_ONCE(a))
while (tmp = READ_ONCE(a))
do_something_with(tmp);
But please note that the compiler is also closely watching what you
do with the value after the ACCESS_ONCE(). For example, suppose you
do with the value after the READ_ONCE(). For example, suppose you
do the following and MAX is a preprocessor macro with the value 1:
while ((tmp = ACCESS_ONCE(a)) % MAX)
while ((tmp = READ_ONCE(a)) % MAX)
do_something_with(tmp);
Then the compiler knows that the result of the "%" operator applied
@ -1475,12 +1476,12 @@ code. Here are some examples of these sorts of optimizations:
surprise if some other CPU might have stored to variable 'a' in the
meantime.
Use ACCESS_ONCE() to prevent the compiler from making this sort of
Use WRITE_ONCE() to prevent the compiler from making this sort of
wrong guess:
ACCESS_ONCE(a) = 0;
WRITE_ONCE(a, 0);
/* Code that does not store to variable a. */
ACCESS_ONCE(a) = 0;
WRITE_ONCE(a, 0);
(*) The compiler is within its rights to reorder memory accesses unless
you tell it not to. For example, consider the following interaction
@ -1509,40 +1510,43 @@ code. Here are some examples of these sorts of optimizations:
}
If the interrupt occurs between these two statement, then
interrupt_handler() might be passed a garbled msg. Use ACCESS_ONCE()
interrupt_handler() might be passed a garbled msg. Use WRITE_ONCE()
to prevent this as follows:
void process_level(void)
{
ACCESS_ONCE(msg) = get_message();
ACCESS_ONCE(flag) = true;
WRITE_ONCE(msg, get_message());
WRITE_ONCE(flag, true);
}
void interrupt_handler(void)
{
if (ACCESS_ONCE(flag))
process_message(ACCESS_ONCE(msg));
if (READ_ONCE(flag))
process_message(READ_ONCE(msg));
}
Note that the ACCESS_ONCE() wrappers in interrupt_handler()
are needed if this interrupt handler can itself be interrupted
by something that also accesses 'flag' and 'msg', for example,
a nested interrupt or an NMI. Otherwise, ACCESS_ONCE() is not
needed in interrupt_handler() other than for documentation purposes.
(Note also that nested interrupts do not typically occur in modern
Linux kernels, in fact, if an interrupt handler returns with
interrupts enabled, you will get a WARN_ONCE() splat.)
Note that the READ_ONCE() and WRITE_ONCE() wrappers in
interrupt_handler() are needed if this interrupt handler can itself
be interrupted by something that also accesses 'flag' and 'msg',
for example, a nested interrupt or an NMI. Otherwise, READ_ONCE()
and WRITE_ONCE() are not needed in interrupt_handler() other than
for documentation purposes. (Note also that nested interrupts
do not typically occur in modern Linux kernels, in fact, if an
interrupt handler returns with interrupts enabled, you will get a
WARN_ONCE() splat.)
You should assume that the compiler can move ACCESS_ONCE() past
code not containing ACCESS_ONCE(), barrier(), or similar primitives.
You should assume that the compiler can move READ_ONCE() and
WRITE_ONCE() past code not containing READ_ONCE(), WRITE_ONCE(),
barrier(), or similar primitives.
This effect could also be achieved using barrier(), but ACCESS_ONCE()
is more selective: With ACCESS_ONCE(), the compiler need only forget
the contents of the indicated memory locations, while with barrier()
the compiler must discard the value of all memory locations that
it has currented cached in any machine registers. Of course,
the compiler must also respect the order in which the ACCESS_ONCE()s
occur, though the CPU of course need not do so.
This effect could also be achieved using barrier(), but READ_ONCE()
and WRITE_ONCE() are more selective: With READ_ONCE() and
WRITE_ONCE(), the compiler need only forget the contents of the
indicated memory locations, while with barrier() the compiler must
discard the value of all memory locations that it has currented
cached in any machine registers. Of course, the compiler must also
respect the order in which the READ_ONCE()s and WRITE_ONCE()s occur,
though the CPU of course need not do so.
(*) The compiler is within its rights to invent stores to a variable,
as in the following example:
@ -1562,16 +1566,16 @@ code. Here are some examples of these sorts of optimizations:
a branch. Unfortunately, in concurrent code, this optimization
could cause some other CPU to see a spurious value of 42 -- even
if variable 'a' was never zero -- when loading variable 'b'.
Use ACCESS_ONCE() to prevent this as follows:
Use WRITE_ONCE() to prevent this as follows:
if (a)
ACCESS_ONCE(b) = a;
WRITE_ONCE(b, a);
else
ACCESS_ONCE(b) = 42;
WRITE_ONCE(b, 42);
The compiler can also invent loads. These are usually less
damaging, but they can result in cache-line bouncing and thus in
poor performance and scalability. Use ACCESS_ONCE() to prevent
poor performance and scalability. Use READ_ONCE() to prevent
invented loads.
(*) For aligned memory locations whose size allows them to be accessed
@ -1590,9 +1594,9 @@ code. Here are some examples of these sorts of optimizations:
This optimization can therefore be a win in single-threaded code.
In fact, a recent bug (since fixed) caused GCC to incorrectly use
this optimization in a volatile store. In the absence of such bugs,
use of ACCESS_ONCE() prevents store tearing in the following example:
use of WRITE_ONCE() prevents store tearing in the following example:
ACCESS_ONCE(p) = 0x00010002;
WRITE_ONCE(p, 0x00010002);
Use of packed structures can also result in load and store tearing,
as in this example:
@ -1609,22 +1613,23 @@ code. Here are some examples of these sorts of optimizations:
foo2.b = foo1.b;
foo2.c = foo1.c;
Because there are no ACCESS_ONCE() wrappers and no volatile markings,
the compiler would be well within its rights to implement these three
assignment statements as a pair of 32-bit loads followed by a pair
of 32-bit stores. This would result in load tearing on 'foo1.b'
and store tearing on 'foo2.b'. ACCESS_ONCE() again prevents tearing
in this example:
Because there are no READ_ONCE() or WRITE_ONCE() wrappers and no
volatile markings, the compiler would be well within its rights to
implement these three assignment statements as a pair of 32-bit
loads followed by a pair of 32-bit stores. This would result in
load tearing on 'foo1.b' and store tearing on 'foo2.b'. READ_ONCE()
and WRITE_ONCE() again prevent tearing in this example:
foo2.a = foo1.a;
ACCESS_ONCE(foo2.b) = ACCESS_ONCE(foo1.b);
WRITE_ONCE(foo2.b, READ_ONCE(foo1.b));
foo2.c = foo1.c;
All that aside, it is never necessary to use ACCESS_ONCE() on a variable
that has been marked volatile. For example, because 'jiffies' is marked
volatile, it is never necessary to say ACCESS_ONCE(jiffies). The reason
for this is that ACCESS_ONCE() is implemented as a volatile cast, which
has no effect when its argument is already marked volatile.
All that aside, it is never necessary to use READ_ONCE() and
WRITE_ONCE() on a variable that has been marked volatile. For example,
because 'jiffies' is marked volatile, it is never necessary to
say READ_ONCE(jiffies). The reason for this is that READ_ONCE() and
WRITE_ONCE() are implemented as volatile casts, which has no effect when
its argument is already marked volatile.
Please note that these compiler barriers have no direct effect on the CPU,
which may then reorder things however it wishes.
@ -1646,14 +1651,15 @@ The Linux kernel has eight basic CPU memory barriers:
All memory barriers except the data dependency barriers imply a compiler
barrier. Data dependencies do not impose any additional compiler ordering.
Aside: In the case of data dependencies, the compiler would be expected to
issue the loads in the correct order (eg. `a[b]` would have to load the value
of b before loading a[b]), however there is no guarantee in the C specification
that the compiler may not speculate the value of b (eg. is equal to 1) and load
a before b (eg. tmp = a[1]; if (b != 1) tmp = a[b]; ). There is also the
problem of a compiler reloading b after having loaded a[b], thus having a newer
copy of b than a[b]. A consensus has not yet been reached about these problems,
however the ACCESS_ONCE macro is a good place to start looking.
Aside: In the case of data dependencies, the compiler would be expected
to issue the loads in the correct order (eg. `a[b]` would have to load
the value of b before loading a[b]), however there is no guarantee in
the C specification that the compiler may not speculate the value of b
(eg. is equal to 1) and load a before b (eg. tmp = a[1]; if (b != 1)
tmp = a[b]; ). There is also the problem of a compiler reloading b after
having loaded a[b], thus having a newer copy of b than a[b]. A consensus
has not yet been reached about these problems, however the READ_ONCE()
macro is a good place to start looking.
SMP memory barriers are reduced to compiler barriers on uniprocessor compiled
systems because it is assumed that a CPU will appear to be self-consistent,
@ -1852,11 +1858,12 @@ Similarly, the reverse case of a RELEASE followed by an ACQUIRE does not
imply a full memory barrier. If it is necessary for a RELEASE-ACQUIRE
pair to produce a full barrier, the ACQUIRE can be followed by an
smp_mb__after_unlock_lock() invocation. This will produce a full barrier
if either (a) the RELEASE and the ACQUIRE are executed by the same
CPU or task, or (b) the RELEASE and ACQUIRE act on the same variable.
The smp_mb__after_unlock_lock() primitive is free on many architectures.
Without smp_mb__after_unlock_lock(), the CPU's execution of the critical
sections corresponding to the RELEASE and the ACQUIRE can cross, so that:
(including transitivity) if either (a) the RELEASE and the ACQUIRE are
executed by the same CPU or task, or (b) the RELEASE and ACQUIRE act on
the same variable. The smp_mb__after_unlock_lock() primitive is free
on many architectures. Without smp_mb__after_unlock_lock(), the CPU's
execution of the critical sections corresponding to the RELEASE and the
ACQUIRE can cross, so that:
*A = a;
RELEASE M
@ -2126,12 +2133,12 @@ three CPUs; then should the following sequence of events occur:
CPU 1 CPU 2
=============================== ===============================
ACCESS_ONCE(*A) = a; ACCESS_ONCE(*E) = e;
WRITE_ONCE(*A, a); WRITE_ONCE(*E, e);
ACQUIRE M ACQUIRE Q
ACCESS_ONCE(*B) = b; ACCESS_ONCE(*F) = f;
ACCESS_ONCE(*C) = c; ACCESS_ONCE(*G) = g;
WRITE_ONCE(*B, b); WRITE_ONCE(*F, f);
WRITE_ONCE(*C, c); WRITE_ONCE(*G, g);
RELEASE M RELEASE Q
ACCESS_ONCE(*D) = d; ACCESS_ONCE(*H) = h;
WRITE_ONCE(*D, d); WRITE_ONCE(*H, h);
Then there is no guarantee as to what order CPU 3 will see the accesses to *A
through *H occur in, other than the constraints imposed by the separate locks
@ -2151,18 +2158,18 @@ However, if the following occurs:
CPU 1 CPU 2
=============================== ===============================
ACCESS_ONCE(*A) = a;
WRITE_ONCE(*A, a);
ACQUIRE M [1]
ACCESS_ONCE(*B) = b;
ACCESS_ONCE(*C) = c;
WRITE_ONCE(*B, b);
WRITE_ONCE(*C, c);
RELEASE M [1]
ACCESS_ONCE(*D) = d; ACCESS_ONCE(*E) = e;
WRITE_ONCE(*D, d); WRITE_ONCE(*E, e);
ACQUIRE M [2]
smp_mb__after_unlock_lock();
ACCESS_ONCE(*F) = f;
ACCESS_ONCE(*G) = g;
WRITE_ONCE(*F, f);
WRITE_ONCE(*G, g);
RELEASE M [2]
ACCESS_ONCE(*H) = h;
WRITE_ONCE(*H, h);
CPU 3 might see:
@ -2881,11 +2888,11 @@ A programmer might take it for granted that the CPU will perform memory
operations in exactly the order specified, so that if the CPU is, for example,
given the following piece of code to execute:
a = ACCESS_ONCE(*A);
ACCESS_ONCE(*B) = b;
c = ACCESS_ONCE(*C);
d = ACCESS_ONCE(*D);
ACCESS_ONCE(*E) = e;
a = READ_ONCE(*A);
WRITE_ONCE(*B, b);
c = READ_ONCE(*C);
d = READ_ONCE(*D);
WRITE_ONCE(*E, e);
they would then expect that the CPU will complete the memory operation for each
instruction before moving on to the next one, leading to a definite sequence of
@ -2932,12 +2939,12 @@ However, it is guaranteed that a CPU will be self-consistent: it will see its
_own_ accesses appear to be correctly ordered, without the need for a memory
barrier. For instance with the following code:
U = ACCESS_ONCE(*A);
ACCESS_ONCE(*A) = V;
ACCESS_ONCE(*A) = W;
X = ACCESS_ONCE(*A);
ACCESS_ONCE(*A) = Y;
Z = ACCESS_ONCE(*A);
U = READ_ONCE(*A);
WRITE_ONCE(*A, V);
WRITE_ONCE(*A, W);
X = READ_ONCE(*A);
WRITE_ONCE(*A, Y);
Z = READ_ONCE(*A);
and assuming no intervention by an external influence, it can be assumed that
the final result will appear to be:
@ -2953,13 +2960,14 @@ accesses:
U=LOAD *A, STORE *A=V, STORE *A=W, X=LOAD *A, STORE *A=Y, Z=LOAD *A
in that order, but, without intervention, the sequence may have almost any
combination of elements combined or discarded, provided the program's view of
the world remains consistent. Note that ACCESS_ONCE() is -not- optional
in the above example, as there are architectures where a given CPU might
reorder successive loads to the same location. On such architectures,
ACCESS_ONCE() does whatever is necessary to prevent this, for example, on
Itanium the volatile casts used by ACCESS_ONCE() cause GCC to emit the
special ld.acq and st.rel instructions that prevent such reordering.
combination of elements combined or discarded, provided the program's view
of the world remains consistent. Note that READ_ONCE() and WRITE_ONCE()
are -not- optional in the above example, as there are architectures
where a given CPU might reorder successive loads to the same location.
On such architectures, READ_ONCE() and WRITE_ONCE() do whatever is
necessary to prevent this, for example, on Itanium the volatile casts
used by READ_ONCE() and WRITE_ONCE() cause GCC to emit the special ld.acq
and st.rel instructions (respectively) that prevent such reordering.
The compiler may also combine, discard or defer elements of the sequence before
the CPU even sees them.
@ -2973,13 +2981,14 @@ may be reduced to:
*A = W;
since, without either a write barrier or an ACCESS_ONCE(), it can be
since, without either a write barrier or an WRITE_ONCE(), it can be
assumed that the effect of the storage of V to *A is lost. Similarly:
*A = Y;
Z = *A;
may, without a memory barrier or an ACCESS_ONCE(), be reduced to:
may, without a memory barrier or an READ_ONCE() and WRITE_ONCE(), be
reduced to:
*A = Y;
Z = Y;

View File

@ -686,10 +686,20 @@ static struct rcu_torture_ops tasks_ops = {
#define RCUTORTURE_TASKS_OPS &tasks_ops,
static bool __maybe_unused torturing_tasks(void)
{
return cur_ops == &tasks_ops;
}
#else /* #ifdef CONFIG_TASKS_RCU */
#define RCUTORTURE_TASKS_OPS
static bool torturing_tasks(void)
{
return false;
}
#endif /* #else #ifdef CONFIG_TASKS_RCU */
/*
@ -825,9 +835,7 @@ rcu_torture_cbflood(void *arg)
}
if (err) {
VERBOSE_TOROUT_STRING("rcu_torture_cbflood disabled: Bad args or OOM");
while (!torture_must_stop())
schedule_timeout_interruptible(HZ);
return 0;
goto wait_for_stop;
}
VERBOSE_TOROUT_STRING("rcu_torture_cbflood task started");
do {
@ -846,6 +854,7 @@ rcu_torture_cbflood(void *arg)
stutter_wait("rcu_torture_cbflood");
} while (!torture_must_stop());
vfree(rhp);
wait_for_stop:
torture_kthread_stopping("rcu_torture_cbflood");
return 0;
}
@ -1090,7 +1099,8 @@ static void rcu_torture_timer(unsigned long unused)
p = rcu_dereference_check(rcu_torture_current,
rcu_read_lock_bh_held() ||
rcu_read_lock_sched_held() ||
srcu_read_lock_held(srcu_ctlp));
srcu_read_lock_held(srcu_ctlp) ||
torturing_tasks());
if (p == NULL) {
/* Leave because rcu_torture_writer is not yet underway */
cur_ops->readunlock(idx);
@ -1164,7 +1174,8 @@ rcu_torture_reader(void *arg)
p = rcu_dereference_check(rcu_torture_current,
rcu_read_lock_bh_held() ||
rcu_read_lock_sched_held() ||
srcu_read_lock_held(srcu_ctlp));
srcu_read_lock_held(srcu_ctlp) ||
torturing_tasks());
if (p == NULL) {
/* Wait for rcu_torture_writer to get underway */
cur_ops->readunlock(idx);
@ -1509,7 +1520,7 @@ static int rcu_torture_barrier_init(void)
int i;
int ret;
if (n_barrier_cbs == 0)
if (n_barrier_cbs <= 0)
return 0;
if (cur_ops->call == NULL || cur_ops->cb_barrier == NULL) {
pr_alert("%s" TORTURE_FLAG
@ -1788,12 +1799,15 @@ rcu_torture_init(void)
writer_task);
if (firsterr)
goto unwind;
fakewriter_tasks = kzalloc(nfakewriters * sizeof(fakewriter_tasks[0]),
GFP_KERNEL);
if (fakewriter_tasks == NULL) {
VERBOSE_TOROUT_ERRSTRING("out of memory");
firsterr = -ENOMEM;
goto unwind;
if (nfakewriters > 0) {
fakewriter_tasks = kzalloc(nfakewriters *
sizeof(fakewriter_tasks[0]),
GFP_KERNEL);
if (fakewriter_tasks == NULL) {
VERBOSE_TOROUT_ERRSTRING("out of memory");
firsterr = -ENOMEM;
goto unwind;
}
}
for (i = 0; i < nfakewriters; i++) {
firsterr = torture_create_kthread(rcu_torture_fakewriter,
@ -1820,7 +1834,7 @@ rcu_torture_init(void)
if (firsterr)
goto unwind;
}
if (test_no_idle_hz) {
if (test_no_idle_hz && shuffle_interval > 0) {
firsterr = torture_shuffle_init(shuffle_interval * HZ);
if (firsterr)
goto unwind;

View File

@ -5,6 +5,6 @@ CONFIG_PREEMPT_NONE=n
CONFIG_PREEMPT_VOLUNTARY=n
CONFIG_PREEMPT=y
CONFIG_DEBUG_LOCK_ALLOC=y
CONFIG_PROVE_LOCKING=n
#CHECK#CONFIG_PROVE_RCU=n
CONFIG_PROVE_LOCKING=y
#CHECK#CONFIG_PROVE_RCU=y
CONFIG_RCU_EXPERT=y