mirrors/gcc
1
0
Fork 0
mirror of https://gcc.gnu.org/git/gcc.git synced 2024-11-23 02:44:18 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

libgomp: Enable USM for AMD APUs and MI200 devices

Browse Source 
If HSA_AMD_SYSTEM_INFO_SVM_ACCESSIBLE_BY_DEFAULT is true,
all GPUs on the system support unified shared memory. That's
the case for APUs and MI200 devices when XNACK is enabled.

XNACK can be enabled by setting HSA_XNACK=1 as env var for
supported devices; otherwise, if disable, USM code will
use host fallback.

gcc/ChangeLog:

	* config/gcn/gcn-hsa.h (gcn_local_sym_hash): Fix typo.

include/ChangeLog:

	* hsa.h (HSA_AMD_SYSTEM_INFO_SVM_ACCESSIBLE_BY_DEFAULT): Add
	enum value.

libgomp/ChangeLog:

	* libgomp.texi (gcn): Update USM handling
	* plugin/plugin-gcn.c (GOMP_OFFLOAD_get_num_devices): Handle
	USM if HSA_AMD_SYSTEM_INFO_SVM_ACCESSIBLE_BY_DEFAULT is true.
This commit is contained in:
Tobias Burnus 2024-05-29 15:29:06 +02:00
parent 4ccb3366ad
commit 18f477980c
4 changed files with 28 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
gcc/config/gcn/gcn-hsa.h
View File

@ -80,7 +80,7 @@ extern unsigned int gcn_local_sym_hash (const char *name);
writes a new AMD GPU object file and the ABI version needs to be the
same. - LLVM <= 17 defaults to 4 while LLVM >= 18 defaults to 5.
GCC supports LLVM >= 13.0.1 and only LLVM >= 14 supports version 5.
Note that Fiji is only suppored with LLVM <= 17 as version 3 is no longer
Note that Fiji is only supported with LLVM <= 17 as version 3 is no longer
supported in LLVM >= 18. */
#define ABI_VERSION_SPEC "march=fiji:--amdhsa-code-object-version=3;" \
"!march=*|march=*:--amdhsa-code-object-version=4"

4
include/hsa.h
View File

@ -466,7 +466,9 @@ typedef enum {
/**
* String containing the ROCr build identifier.
*/
HSA_AMD_SYSTEM_INFO_BUILD_VERSION = 0x200
HSA_AMD_SYSTEM_INFO_BUILD_VERSION = 0x200,
HSA_AMD_SYSTEM_INFO_SVM_ACCESSIBLE_BY_DEFAULT = 0x202
} hsa_system_info_t;
/**

9
libgomp/libgomp.texi
View File

@ -6360,8 +6360,13 @@ The implementation remark:
such that the next reverse offload region is only executed after the previous
one returned.
@item OpenMP code that has a @code{requires} directive with
@code{unified_shared_memory} will remove any GCN device from the list of
available devices (``host fallback'').
@code{unified_shared_memory} is only supported if all AMD GPUs have the
@code{HSA_AMD_SYSTEM_INFO_SVM_ACCESSIBLE_BY_DEFAULT} property; for
discrete GPUs, this may require setting the @code{HSA_XNACK} environment
variable to @samp{1}; for systems with both an APU and a discrete GPU that
does not support XNACK, consider using @code{ROCR_VISIBLE_DEVICES} to
enable only the APU. If not supported, all AMD GPU devices are removed
from the list of available devices (``host fallback'').
@item The available stack size can be changed using the @code{GCN_STACK_SIZE}
environment variable; the default is 32 kiB per thread.
@item Low-latency memory (@code{omp_low_lat_mem_space}) is supported when the

17
libgomp/plugin/plugin-gcn.c
View File

@ -3355,8 +3355,25 @@ GOMP_OFFLOAD_get_num_devices (unsigned int omp_requires_mask)
if (hsa_context.agent_count > 0
&& ((omp_requires_mask
& ~(GOMP_REQUIRES_UNIFIED_ADDRESS
| GOMP_REQUIRES_UNIFIED_SHARED_MEMORY
| GOMP_REQUIRES_REVERSE_OFFLOAD)) != 0))
return -1;
/* Check whether host page access is supported; this is per system level
(all GPUs supported by HSA). While intrinsically true for APUs, it
requires XNACK support for discrete GPUs. */
if (hsa_context.agent_count > 0
&& (omp_requires_mask & GOMP_REQUIRES_UNIFIED_SHARED_MEMORY))
{
bool b;
hsa_system_info_t type = HSA_AMD_SYSTEM_INFO_SVM_ACCESSIBLE_BY_DEFAULT;
hsa_status_t status = hsa_fns.hsa_system_get_info_fn (type, &b);
if (status != HSA_STATUS_SUCCESS)
GOMP_PLUGIN_error ("HSA_AMD_SYSTEM_INFO_SVM_ACCESSIBLE_BY_DEFAULT "
"failed");
if (!b)
return -1;
}
return hsa_context.agent_count;
}