Hi!

[from internal email]

On 2018-01-09T11:57:00+0100, Tom de Vries wrote:
> On 01/09/2018 12:17 AM, Julian Brown wrote:
>> +     (b) modifications to the copied data between the "spawning" point of
>> +     the asynchronous kernel and when it is executed will not be seen.
>> +     But, that is probably correct.  */
>
> Consider this example:
> [...]
> It passes without [certain code/changes],
> but fails with. So, this looks like a regression.

Confirmed (for GCN offloading).  Based on Tom's example, I've pushed
"Add 'libgomp.oacc-c-c++-common/async-data-1-{1,2}.c'" to master branch
in commit 88c40c36db8a52d2c630aa61ee54e33908e9daec, see attached.

GCN XFAIL 'libgomp.oacc-c-c++-common/async-data-1-1.c' to be removed by
'Fix OpenACC "ephemeral" asynchronous host-to-device copies' and
GCN XFAIL 'libgomp.oacc-c-c++-common/async-data-1-2.c' to be removed by
"Don't use libgomp 'cbuf' buffering with OpenACC 'async'".


Grüße
 Thomas
-----------------
Siemens Electronic Design Automation GmbH; Anschrift: Arnulfstraße 201, 80634 
München; Gesellschaft mit beschränkter Haftung; Geschäftsführer: Thomas 
Heurung, Frank Thürauf; Sitz der Gesellschaft: München; Registergericht 
München, HRB 106955

Reply via email to