Hi,
On 2018-04-07 14:23:38 -0400, Tom Lane wrote:
> I think I'd just drop those asserts altogether. The hardware is in charge
> of complaining about misaligned pointers.
I came around to that view. The problem with "natural" and atomic
alignment differing is only relevant for 64bit integers (4 b
Hi,
On 2018-04-07 14:23:38 -0400, Tom Lane wrote:
> I think I'd just drop those asserts altogether. The hardware is in charge
> of complaining about misaligned pointers.
Well, the problem is that some atomics operations on some platforms do
not fail for unaligned pointers, they just loose their
Andres Freund writes:
> On 2018-04-07 14:07:05 -0400, Tom Lane wrote:
>> TRAP: UnalignedPointer("(((uintptr_t) ((uintptr_t)(ptr)) + ((sizeof(*ptr)) -
>> 1)) & ~((uintptr_t) ((sizeof(*ptr)) - 1))) != (uintptr_t)(ptr)", File:
>> "../../../src/include/port/atomics.h", Line: 177)
> Yea, I just saw
On 2018-04-07 14:07:05 -0400, Tom Lane wrote:
> Andres Freund writes:
> > As Daniel pointed out in:
> > https://postgr.es/m/fb948276-7b32-4b77-83e6-d00167f8e...@yesql.se the
> > pg_atomic_flag fallback implementation is broken. That has gone
> > unnoticed because the fallback implementation wasn'
Andres Freund writes:
> As Daniel pointed out in:
> https://postgr.es/m/fb948276-7b32-4b77-83e6-d00167f8e...@yesql.se the
> pg_atomic_flag fallback implementation is broken. That has gone
> unnoticed because the fallback implementation wasn't testable until now:
> ...
> The attached fixes the bug
Hi,
On 2018-04-07 12:57:53 +, Magnus Hagander wrote:
> This is the same one you already committed right? Not a further one on top
> of that?
Yes, I pushed a few hours later.
> That said,+1 for not bothering to back patch it.
I did ;). I was more wondering about whether to backpatch the AB
This is the same one you already committed right? Not a further one on top
of that?
That said,+1 for not bothering to back patch it.
/Magnus
On Sat, Apr 7, 2018, 00:39 Andres Freund wrote:
> Hi,
>
> As Daniel pointed out in:
> https://postgr.es/m/fb948276-7b32-4b77-83e6-d00167f8e...@yesql.se
Hi,
As Daniel pointed out in:
https://postgr.es/m/fb948276-7b32-4b77-83e6-d00167f8e...@yesql.se the
pg_atomic_flag fallback implementation is broken. That has gone
unnoticed because the fallback implementation wasn't testable until now:
- /* ---
-* Can't run the test under the semaphore emu
