On 7/07/2019 6:48 pm, Erik Osterlund wrote:
Yeah that switch statement code and yet another plain non-volatile load/store
loop looks like complete nonsense unfortunately. It should at least use
Atomic::load/store.
Fortunately, on x86_64, I believe it will in practice yield word atomic copying
anyway by chance. But it should be fixed anyway. *sigh*
The requirement is for atomic word accesses, which properly aligned
addresses will always yield - we rely on that guarantee throughout to
avoid word-tearing. The issue is that the loop may be converted to a
'bulk' copying operation that may not provide atomic word accesses. So
unless the Atomic::load/store will prevent the loop from being converted
they are not in and off themselves required for correctness (else we
need to use them nearly everywhere).
The real danger is SPARC though and its BIS instructions. I don’t have the code
in front of me, but I really hope not to see that switch statement and
non-volatile loop in that pd_disjoint_words_atomic() function.
sparc uses the same loop.
Let's face it, almost no body expects the compiler to do these kinds of
transformations. :(
And I agree that the atomic copying API should be used when we need atomic
copying. And if it turns out the implementation of that API is not atomic, it
should be fixed in that atomic copying API.
I agree to some extent, but we assume atomic load/stores of words all
over the place - and rightly so. The issue here is that we need to hide
the loop inside an API that we can somehow prevent the C++ compiler from
screwing up. It's hardly intuitive or obvious when this is needed e.g if
I simply copy three adjacent words without a loop could the compiler
convert that to a block move that is non-atomic?
So I think this change looks good. But it looks like we are not done yet. :c
I agree that changing the current code to use the atomic copy API to
convey intent is fine.
Cheers,
David
-----
Thanks,
/Erik
On 7 Jul 2019, at 02:46, Daniel D. Daugherty <[email protected]>
wrote:
Added Erik O. to the "To:" list...
On 7/6/19 8:05 PM, Daniel D. Daugherty wrote:
On 7/6/19 6:06 PM, David Holmes wrote:
Hi Dan,
On 6/07/2019 11:53 pm, Daniel D. Daugherty wrote:
Greetings,
During the code review for the following fix:
JDK-8227117 normal interpreter table is not restored after single
stepping with TLH
https://bugs.openjdk.java.net/browse/JDK-8227117
Erik O. noticed a potential race with templateInterpreter.cpp: copy_table()
depending on C++ compiler optimizations. The following bug is being used
to fix this issue:
JDK-8227338 templateInterpreter.cpp: copy_table() needs to be safer
https://bugs.openjdk.java.net/browse/JDK-8227338
Here's the webrev URL:
http://cr.openjdk.java.net/~dcubed/8227338-webrev/0_for_jdk14/
So the original code uses a loop to copy, while the new code calls
Copy::disjoint_words_atomic, but the implementation of that on x64 is just a
loop same as the original AFAICS:
Yup. I figure Erik O. will jump in here with his reasoning... :-)
Thinking about it more... I think the answer is that we are switching to
calling code that specifies the type of behavior that we need:
Copy::disjoint_words_atomic()
is what we need when we're not at a safepoint. If, down the road, we find
that the compiler does something with pd_disjoint_words_atomic() that
breaks our expectation for pd_disjoint_words_atomic(), then we fix that
version of pd_disjoint_words_atomic() and all the callers will be good
again... Or something like that...
The version in src/hotspot/os_cpu/solaris_x86/copy_solaris_x86.inline.hpp
happens to be exactly our loop with no switch statement... which is
particularly funny given Erik's observations about what at least one
Solaris X64 compiler did to loops...
Still, I'm just guessing here on a Saturday night... hopefully Erik
will chime in here...
Dan
Dan
static void pd_disjoint_words_atomic(const HeapWord* from, HeapWord* to, size_t
count) {
#ifdef AMD64
switch (count) {
case 8: to[7] = from[7];
case 7: to[6] = from[6];
case 6: to[5] = from[5];
case 5: to[4] = from[4];
case 4: to[3] = from[3];
case 3: to[2] = from[2];
case 2: to[1] = from[1];
case 1: to[0] = from[0];
case 0: break;
default:
while (count-- > 0) {
*to++ = *from++;
}
break;
}
#else
David
-----
This fix has been tested via Mach5 Tier[1-3] on Oracle's usual platforms.
Mach5 tier[4-6] is running now. It has also been tested with the manual
jdb test from JDK-8227117 using 'release' and 'fastdebug' bits.
Thanks, in advance, for questions, comments or suggestions.
Dan
