amdgcn: Add gfx1103 target
Add support for the gfx1103 RDNA3 APU integrated graphics devices. The ROCm documentation warns that these may not be supported, but it seems to work at least partially. gcc/ChangeLog: * config.gcc (amdgcn): Add gfx1103 entries. * config/gcn/gcn-hsa.h (NO_XNACK): Likewise. (gcn_local_sym_hash): Likewise. * config/gcn/gcn-opts.h (enum processor_type): Likewise. (TARGET_GFX1103): New macro. * config/gcn/gcn.cc (gcn_option_override): Handle gfx1103. (gcn_omp_device_kind_arch_isa): Likewise. (output_file_start): Likewise. (gcn_hsa_declare_function_name): Use TARGET_RDNA3, not just gfx1100. * config/gcn/gcn.h (TARGET_CPU_CPP_BUILTINS): Add __gfx1103__. * config/gcn/gcn.opt: Add gfx1103. * config/gcn/mkoffload.cc (EF_AMDGPU_MACH_AMDGCN_GFX1103): New. (main): Handle gfx1103. * config/gcn/t-omp-device: Add gfx1103 isa. * doc/install.texi (amdgcn): Add gfx1103. * doc/invoke.texi (-march): Likewise. libgomp/ChangeLog: * plugin/plugin-gcn.c (EF_AMDGPU_MACH): GFX1103. (gcn_gfx1103_s): New. (isa_hsa_name): Handle gfx1103. (isa_code): Likewise. (max_isa_vgprs): Likewise.
This commit is contained in:
parent
d1d8fd2884
commit
1bf18629c5
11 changed files with 50 additions and 16 deletions
|
@ -4560,7 +4560,7 @@ case "${target}" in
|
|||
for which in arch tune; do
|
||||
eval "val=\$with_$which"
|
||||
case ${val} in
|
||||
"" | fiji | gfx900 | gfx906 | gfx908 | gfx90a | gfx1030 | gfx1100)
|
||||
"" | fiji | gfx900 | gfx906 | gfx908 | gfx90a | gfx1030 | gfx1100 | gfx1103)
|
||||
# OK
|
||||
;;
|
||||
*)
|
||||
|
@ -4576,7 +4576,7 @@ case "${target}" in
|
|||
TM_MULTILIB_CONFIG=
|
||||
;;
|
||||
xdefault | xyes)
|
||||
TM_MULTILIB_CONFIG=`echo "gfx900,gfx906,gfx908,gfx90a,gfx1030,gfx1100" | sed "s/${with_arch},\?//;s/,$//"`
|
||||
TM_MULTILIB_CONFIG=`echo "gfx900,gfx906,gfx908,gfx90a,gfx1030,gfx1100,gfx1103" | sed "s/${with_arch},\?//;s/,$//"`
|
||||
;;
|
||||
*)
|
||||
TM_MULTILIB_CONFIG="${with_multilib_list}"
|
||||
|
|
|
@ -90,7 +90,7 @@ extern unsigned int gcn_local_sym_hash (const char *name);
|
|||
the ELF flags (e_flags) of that generated file must be identical to those
|
||||
generated by the compiler. */
|
||||
|
||||
#define NO_XNACK "march=fiji:;march=gfx1030:;march=gfx1100:;" \
|
||||
#define NO_XNACK "march=fiji:;march=gfx1030:;march=gfx1100:;march=gfx1103:;" \
|
||||
/* These match the defaults set in gcn.cc. */ \
|
||||
"!mxnack*|mxnack=default:%{march=gfx900|march=gfx906|march=gfx908:-mattr=-xnack};"
|
||||
#define NO_SRAM_ECC "!march=*:;march=fiji:;march=gfx900:;march=gfx906:;"
|
||||
|
@ -106,8 +106,8 @@ extern unsigned int gcn_local_sym_hash (const char *name);
|
|||
"%{" ABI_VERSION_SPEC "} " \
|
||||
"%{" NO_XNACK XNACKOPT "} " \
|
||||
"%{" NO_SRAM_ECC SRAMOPT "} " \
|
||||
"%{march=gfx1030|march=gfx1100:-mattr=+wavefrontsize64} " \
|
||||
"%{march=gfx1030|march=gfx1100:-mattr=+cumode} " \
|
||||
"%{march=gfx1030|march=gfx1100|march=gfx1103:-mattr=+wavefrontsize64} " \
|
||||
"%{march=gfx1030|march=gfx1100|march=gfx1103:-mattr=+cumode} " \
|
||||
"-filetype=obj"
|
||||
#define LINK_SPEC "--pie --export-dynamic"
|
||||
#define LIB_SPEC "-lc"
|
||||
|
|
|
@ -26,7 +26,8 @@ enum processor_type
|
|||
PROCESSOR_GFX908,
|
||||
PROCESSOR_GFX90a,
|
||||
PROCESSOR_GFX1030,
|
||||
PROCESSOR_GFX1100
|
||||
PROCESSOR_GFX1100,
|
||||
PROCESSOR_GFX1103
|
||||
};
|
||||
|
||||
#define TARGET_FIJI (gcn_arch == PROCESSOR_FIJI)
|
||||
|
@ -36,6 +37,7 @@ enum processor_type
|
|||
#define TARGET_GFX90a (gcn_arch == PROCESSOR_GFX90a)
|
||||
#define TARGET_GFX1030 (gcn_arch == PROCESSOR_GFX1030)
|
||||
#define TARGET_GFX1100 (gcn_arch == PROCESSOR_GFX1100)
|
||||
#define TARGET_GFX1103 (gcn_arch == PROCESSOR_GFX1103)
|
||||
|
||||
/* Set in gcn_option_override. */
|
||||
extern enum gcn_isa {
|
||||
|
|
|
@ -140,6 +140,7 @@ gcn_option_override (void)
|
|||
: gcn_arch == PROCESSOR_GFX90a ? ISA_CDNA2
|
||||
: gcn_arch == PROCESSOR_GFX1030 ? ISA_RDNA2
|
||||
: gcn_arch == PROCESSOR_GFX1100 ? ISA_RDNA3
|
||||
: gcn_arch == PROCESSOR_GFX1103 ? ISA_RDNA3
|
||||
: ISA_UNKNOWN);
|
||||
gcc_assert (gcn_isa != ISA_UNKNOWN);
|
||||
|
||||
|
@ -164,13 +165,15 @@ gcn_option_override (void)
|
|||
/* gfx803 "Fiji", gfx1030 and gfx1100 do not support XNACK. */
|
||||
if (gcn_arch == PROCESSOR_FIJI
|
||||
|| gcn_arch == PROCESSOR_GFX1030
|
||||
|| gcn_arch == PROCESSOR_GFX1100)
|
||||
|| gcn_arch == PROCESSOR_GFX1100
|
||||
|| gcn_arch == PROCESSOR_GFX1103)
|
||||
{
|
||||
if (flag_xnack == HSACO_ATTR_ON)
|
||||
error ("%<-mxnack=on%> is incompatible with %<-march=%s%>",
|
||||
(gcn_arch == PROCESSOR_FIJI ? "fiji"
|
||||
: gcn_arch == PROCESSOR_GFX1030 ? "gfx1030"
|
||||
: gcn_arch == PROCESSOR_GFX1100 ? "gfx1100"
|
||||
: gcn_arch == PROCESSOR_GFX1103 ? "gfx1103"
|
||||
: NULL));
|
||||
/* Allow HSACO_ATTR_ANY silently because that's the default. */
|
||||
flag_xnack = HSACO_ATTR_OFF;
|
||||
|
@ -3048,6 +3051,8 @@ gcn_omp_device_kind_arch_isa (enum omp_device_kind_arch_isa trait,
|
|||
return gcn_arch == PROCESSOR_GFX1030;
|
||||
if (strcmp (name, "gfx1100") == 0)
|
||||
return gcn_arch == PROCESSOR_GFX1100;
|
||||
if (strcmp (name, "gfx1103") == 0)
|
||||
return gcn_arch == PROCESSOR_GFX1103;
|
||||
return 0;
|
||||
default:
|
||||
gcc_unreachable ();
|
||||
|
@ -6558,6 +6563,11 @@ output_file_start (void)
|
|||
xnack = "";
|
||||
sram_ecc = "";
|
||||
break;
|
||||
case PROCESSOR_GFX1103:
|
||||
cpu = "gfx1103";
|
||||
xnack = "";
|
||||
sram_ecc = "";
|
||||
break;
|
||||
default: gcc_unreachable ();
|
||||
}
|
||||
|
||||
|
@ -6705,7 +6715,7 @@ gcn_hsa_declare_function_name (FILE *file, const char *name,
|
|||
xnack_enabled,
|
||||
LDS_SIZE);
|
||||
/* Not supported with 'architected flat scratch'. */
|
||||
if (gcn_arch != PROCESSOR_GFX1100)
|
||||
if (!TARGET_RDNA3)
|
||||
fprintf (file,
|
||||
"\t .amdhsa_reserve_flat_scratch\t0\n");
|
||||
if (gcn_arch == PROCESSOR_GFX90a)
|
||||
|
|
|
@ -51,6 +51,8 @@
|
|||
builtin_define ("__gfx1030"); \
|
||||
else if (TARGET_GFX1100) \
|
||||
builtin_define ("__gfx1100__"); \
|
||||
else if (TARGET_GFX1103) \
|
||||
builtin_define ("__gfx1103__"); \
|
||||
else \
|
||||
gcc_unreachable (); \
|
||||
} while (0)
|
||||
|
|
|
@ -46,6 +46,9 @@ Enum(gpu_type) String(gfx1030) Value(PROCESSOR_GFX1030)
|
|||
EnumValue
|
||||
Enum(gpu_type) String(gfx1100) Value(PROCESSOR_GFX1100)
|
||||
|
||||
EnumValue
|
||||
Enum(gpu_type) String(gfx1103) Value(PROCESSOR_GFX1103)
|
||||
|
||||
march=
|
||||
Target RejectNegative Negative(march=) Joined ToLower Enum(gpu_type) Var(gcn_arch) Init(PROCESSOR_FIJI)
|
||||
Specify the name of the target GPU.
|
||||
|
|
|
@ -61,6 +61,8 @@
|
|||
#define EF_AMDGPU_MACH_AMDGCN_GFX1030 0x36
|
||||
#undef EF_AMDGPU_MACH_AMDGCN_GFX1100
|
||||
#define EF_AMDGPU_MACH_AMDGCN_GFX1100 0x41
|
||||
#undef EF_AMDGPU_MACH_AMDGCN_GFX1103
|
||||
#define EF_AMDGPU_MACH_AMDGCN_GFX1103 0x44
|
||||
|
||||
#define EF_AMDGPU_FEATURE_XNACK_V4 0x300 /* Mask. */
|
||||
#define EF_AMDGPU_FEATURE_XNACK_UNSUPPORTED_V4 0x000
|
||||
|
@ -971,6 +973,8 @@ main (int argc, char **argv)
|
|||
elf_arch = EF_AMDGPU_MACH_AMDGCN_GFX1030;
|
||||
else if (strcmp (argv[i], "-march=gfx1100") == 0)
|
||||
elf_arch = EF_AMDGPU_MACH_AMDGCN_GFX1100;
|
||||
else if (strcmp (argv[i], "-march=gfx1103") == 0)
|
||||
elf_arch = EF_AMDGPU_MACH_AMDGCN_GFX1103;
|
||||
#define STR "-mstack-size="
|
||||
else if (startswith (argv[i], STR))
|
||||
gcn_stack_size = atoi (argv[i] + strlen (STR));
|
||||
|
@ -1013,6 +1017,7 @@ main (int argc, char **argv)
|
|||
case EF_AMDGPU_MACH_AMDGCN_GFX803:
|
||||
case EF_AMDGPU_MACH_AMDGCN_GFX1030:
|
||||
case EF_AMDGPU_MACH_AMDGCN_GFX1100:
|
||||
case EF_AMDGPU_MACH_AMDGCN_GFX1103:
|
||||
SET_XNACK_UNSET (elf_flags);
|
||||
SET_SRAM_ECC_UNSET (elf_flags);
|
||||
break;
|
||||
|
|
|
@ -1,4 +1,4 @@
|
|||
omp-device-properties-gcn: $(srcdir)/config/gcn/gcn.cc
|
||||
echo kind: gpu > $@
|
||||
echo arch: amdgcn gcn >> $@
|
||||
echo isa: fiji gfx803 gfx900 gfx906 gfx908 gfx90a gfx1030 gfx1100 >> $@
|
||||
echo isa: fiji gfx803 gfx900 gfx906 gfx908 gfx90a gfx1030 gfx1100 gfx1103 >> $@
|
||||
|
|
|
@ -1266,11 +1266,11 @@ default set of libraries is selected based on the value of
|
|||
@item amdgcn*-*-*
|
||||
@var{list} is a comma separated list of ISA names (allowed values: @code{fiji},
|
||||
@code{gfx900}, @code{gfx906}, @code{gfx908}, @code{gfx90a}, @code{gfx1030},
|
||||
@code{gfx1100}). It ought not include the name of the default ISA, specified
|
||||
via @option{--with-arch}. If @var{list} is empty, then there will be no
|
||||
multilibs and only the default run-time library will be built. If @var{list}
|
||||
is @code{default} or @option{--with-multilib-list=} is not specified, then
|
||||
the default set of libraries is selected.
|
||||
@code{gfx1100}, @code{gfx1103}). It ought not include the name of the default
|
||||
ISA, specified via @option{--with-arch}. If @var{list} is empty, then there
|
||||
will be no multilibs and only the default run-time library will be built. If
|
||||
@var{list} is @code{default} or @option{--with-multilib-list=} is not
|
||||
specified, then the default set of libraries is selected.
|
||||
|
||||
@item arm*-*-*
|
||||
@var{list} is a comma separated list of @code{aprofile} and
|
||||
|
@ -3940,7 +3940,8 @@ Instead of GNU Binutils, you will need to install LLVM 15, or later, and copy
|
|||
@file{bin/llvm-nm} to @file{amdgcn-amdhsa/bin/nm}, and
|
||||
@file{bin/llvm-ar} to both @file{bin/amdgcn-amdhsa-ar} and
|
||||
@file{bin/amdgcn-amdhsa-ranlib}. Note that LLVM 13.0.1 or LLVM 14 can be used
|
||||
by specifying a @code{--with-multilib-list=} that does not list @code{gfx1100}.
|
||||
by specifying a @code{--with-multilib-list=} that does not list @code{gfx1100}
|
||||
and @code{gfx1103}.
|
||||
|
||||
Use Newlib (4.3.0 or newer; 4.4.0 or later is recommended).
|
||||
|
||||
|
|
|
@ -21795,6 +21795,9 @@ Compile for RDNA2 gfx1030 devices (GFX10 series).
|
|||
@item gfx1100
|
||||
Compile for RDNA3 gfx1100 devices (GFX11 series).
|
||||
|
||||
@item gfx1103
|
||||
Compile for RDNA3 gfx1103 devices (GFX11 series).
|
||||
|
||||
@end table
|
||||
|
||||
@opindex msram-ecc
|
||||
|
|
|
@ -391,7 +391,8 @@ typedef enum {
|
|||
EF_AMDGPU_MACH_AMDGCN_GFX908 = 0x030,
|
||||
EF_AMDGPU_MACH_AMDGCN_GFX90a = 0x03f,
|
||||
EF_AMDGPU_MACH_AMDGCN_GFX1030 = 0x036,
|
||||
EF_AMDGPU_MACH_AMDGCN_GFX1100 = 0x041
|
||||
EF_AMDGPU_MACH_AMDGCN_GFX1100 = 0x041,
|
||||
EF_AMDGPU_MACH_AMDGCN_GFX1103 = 0x044
|
||||
} EF_AMDGPU_MACH;
|
||||
|
||||
const static int EF_AMDGPU_MACH_MASK = 0x000000ff;
|
||||
|
@ -1677,6 +1678,7 @@ const static char *gcn_gfx908_s = "gfx908";
|
|||
const static char *gcn_gfx90a_s = "gfx90a";
|
||||
const static char *gcn_gfx1030_s = "gfx1030";
|
||||
const static char *gcn_gfx1100_s = "gfx1100";
|
||||
const static char *gcn_gfx1103_s = "gfx1103";
|
||||
const static int gcn_isa_name_len = 7;
|
||||
|
||||
/* Returns the name that the HSA runtime uses for the ISA or NULL if we do not
|
||||
|
@ -1700,6 +1702,8 @@ isa_hsa_name (int isa) {
|
|||
return gcn_gfx1030_s;
|
||||
case EF_AMDGPU_MACH_AMDGCN_GFX1100:
|
||||
return gcn_gfx1100_s;
|
||||
case EF_AMDGPU_MACH_AMDGCN_GFX1103:
|
||||
return gcn_gfx1103_s;
|
||||
}
|
||||
return NULL;
|
||||
}
|
||||
|
@ -1745,6 +1749,9 @@ isa_code(const char *isa) {
|
|||
if (!strncmp (isa, gcn_gfx1100_s, gcn_isa_name_len))
|
||||
return EF_AMDGPU_MACH_AMDGCN_GFX1100;
|
||||
|
||||
if (!strncmp (isa, gcn_gfx1103_s, gcn_isa_name_len))
|
||||
return EF_AMDGPU_MACH_AMDGCN_GFX1103;
|
||||
|
||||
return EF_AMDGPU_MACH_UNSUPPORTED;
|
||||
}
|
||||
|
||||
|
@ -1765,6 +1772,7 @@ max_isa_vgprs (int isa)
|
|||
case EF_AMDGPU_MACH_AMDGCN_GFX1030:
|
||||
return 512; /* 512 SIMD32 = 256 wavefrontsize64. */
|
||||
case EF_AMDGPU_MACH_AMDGCN_GFX1100:
|
||||
case EF_AMDGPU_MACH_AMDGCN_GFX1103:
|
||||
return 1536; /* 1536 SIMD32 = 768 wavefrontsize64. */
|
||||
}
|
||||
GOMP_PLUGIN_fatal ("unhandled ISA in max_isa_vgprs");
|
||||
|
|
Loading…
Add table
Reference in a new issue