On 28/05/2024 23:33, Tobias Burnus wrote:
While most of the nvptx systems I have access to don't have the support
for CU_DEVICE_ATTRIBUTE_PAGEABLE_MEMORY_ACCESS_USES_HOST_PAGE_TABLES,
one has:
Tesla V100-SXM2-16GB (as installed, e.g., on ORNL's Summit) does support
this feature. And with that feature, unified-shared memory support does
work, presumably by handling automatic page migration when a page fault
occurs.
Hence: Enable USM support for those. When doing so, all 'requires
unified_shared_memory' tests of sollve_vv pass :-)
I am not quite sure whether there are unintended side effects, hence, I
have not enabled support for it in general. In particular, 'declare
target enter(global_var)' seems to be mishandled (I think it should be
link + pointer updated to point to the host; cf. description for
'self_maps'). Thus, it is not enabled by default but only when USM has
been requested.
OK for mainline?
Comments? Remarks? Suggestions?
Tobias
PS: I guess some more USM tests should be added…
+ /* If USM has been requested and is supported by all devices
+ of this type, set the capability accordingly. */
+ if (omp_requires_mask & GOMP_REQUIRES_UNIFIED_SHARED_MEMORY)
+ current_device.capabilities |= GOMP_OFFLOAD_CAP_SHARED_MEM;
+
This breaks my USM patches that add the omp_alloc support (because it
now short-circuits all of those code-paths), and it's just not true for
devices where all host memory isn't magically addressable on the device.
Is there another way to detect truly shared memory?
Andrew
