mirror of
https://github.com/Gericom/teak-llvm.git
synced 2025-06-20 12:05:48 -04:00
57098f5ac3
This patch has the following changes
A new flag "-mhvx-length={64B|128B}" is introduced to specify the length of the vector.
Previously we have used "-mhvx-double" for 128 Bytes. This adds the target-feature "+hvx-length{64|128}b"
The "-mhvx" flag must be provided on command line to enable HVX for Hexagon. If no -mhvx-length flag
is specified, a default length is picked from the arch mentioned in this priority order from either -mhvx=vxx
or -mcpu. For v60 and v62 the default length is 64 Byte. For unknown versions, the length is 128 Byte. The
-mhvx flag adds the target-feature "+hvxv{hvx_version}"
The 64 Byte mode is soon going to be deprecated. A warning is emitted if 64 Byte is enabled. A warning is
still emitted for the default 64 Byte as well. This warning can be suppressed with a -Wno flag.
The "-mhvx-double" and "-mno-hvx-double" flags are deprecated. A warning is emitted if the driver sees
them on commandline. "-mhvx-double" is an alias to "-mhvx-length=128B"
The compilation will error out if -mhvx-length is specified with out an -mhvx/-mhvx= flag
The macro HVX_LENGTH is defined and is set to the length of the vector.
Eg: #define HVX_LENGTH 64
The macro HVX_ARCH is defined and is set to the version of the HVX.
Eg: #define HVX_ARCH 62
Differential Revision: https://reviews.llvm.org/D38852
llvm-svn: 316102
24 lines
796 B
C
24 lines
796 B
C
// RUN: %clang_cc1 -triple hexagon-unknown-elf -target-feature +hvx -target-feature +hvx-length64b -emit-llvm -o - %s | FileCheck %s
|
|
|
|
typedef int v64 __attribute__((__vector_size__(64)))
|
|
__attribute__((aligned(64)));
|
|
|
|
int g;
|
|
|
|
void foo(v64 v0, v64 v1, v64 *p) {
|
|
int r;
|
|
v64 q0;
|
|
asm ("%0 = vgtw(%1.w,%2.w)" : "=q"(q0) : "v"(v0), "v"(v1));
|
|
// CHECK: call <16 x i32> asm "$0 = vgtw($1.w,$2.w)", "=q,v,v"(<16 x i32>{{.*}}, <16 x i32>{{.*}})
|
|
*p = q0;
|
|
|
|
asm ("%0 = memw(##%1)" : "=r"(r) : "s"(&g));
|
|
// CHECK: call i32 asm "$0 = memw(##$1)", "=r,s"(i32* @g)
|
|
}
|
|
|
|
void fred(unsigned *p, unsigned m, unsigned v) {
|
|
asm ("memw(%0++%1) = %2" : : "r"(p),"a"(m),"r"(v) : "memory");
|
|
// CHECK: call void asm sideeffect "memw($0++$1) = $2", "r,a,r,~{memory}"(i32* %{{.*}}, i32 %{{.*}}, i32 %{{.*}})
|
|
}
|
|
|