mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-01 08:23:21 +01:00
77f91408dd
Unlike most -- hopefully "all other", but I'm still checking -- memory
instructions we support, LOAD REVERSED and STORE REVERSED may access
the memory location several times. This means that they are not suitable
for volatile loads and stores.
This patch is a prerequisite for better atomic load and store support.
The same principle applies there: almost all memory instructions we
support are inherently atomic ("block concurrent"), but LOAD REVERSED
and STORE REVERSED are exceptions.
Other instructions continue to allow volatile operands. I will add
positive "allows volatile" tests at the same time as the "allows atomic
load or store" tests.
llvm-svn: 183002
100 lines
2.5 KiB
LLVM
100 lines
2.5 KiB
LLVM
; Test 32-bit byteswaps from memory to registers.
|
|
;
|
|
; RUN: llc < %s -mtriple=s390x-linux-gnu | FileCheck %s
|
|
|
|
declare i32 @llvm.bswap.i32(i32 %a)
|
|
|
|
; Check LRV with no displacement.
|
|
define i32 @f1(i32 *%src) {
|
|
; CHECK: f1:
|
|
; CHECK: lrv %r2, 0(%r2)
|
|
; CHECK: br %r14
|
|
%a = load i32 *%src
|
|
%swapped = call i32 @llvm.bswap.i32(i32 %a)
|
|
ret i32 %swapped
|
|
}
|
|
|
|
; Check the high end of the aligned LRV range.
|
|
define i32 @f2(i32 *%src) {
|
|
; CHECK: f2:
|
|
; CHECK: lrv %r2, 524284(%r2)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i32 *%src, i64 131071
|
|
%a = load i32 *%ptr
|
|
%swapped = call i32 @llvm.bswap.i32(i32 %a)
|
|
ret i32 %swapped
|
|
}
|
|
|
|
; Check the next word up, which needs separate address logic.
|
|
; Other sequences besides this one would be OK.
|
|
define i32 @f3(i32 *%src) {
|
|
; CHECK: f3:
|
|
; CHECK: agfi %r2, 524288
|
|
; CHECK: lrv %r2, 0(%r2)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i32 *%src, i64 131072
|
|
%a = load i32 *%ptr
|
|
%swapped = call i32 @llvm.bswap.i32(i32 %a)
|
|
ret i32 %swapped
|
|
}
|
|
|
|
; Check the high end of the negative aligned LRV range.
|
|
define i32 @f4(i32 *%src) {
|
|
; CHECK: f4:
|
|
; CHECK: lrv %r2, -4(%r2)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i32 *%src, i64 -1
|
|
%a = load i32 *%ptr
|
|
%swapped = call i32 @llvm.bswap.i32(i32 %a)
|
|
ret i32 %swapped
|
|
}
|
|
|
|
; Check the low end of the LRV range.
|
|
define i32 @f5(i32 *%src) {
|
|
; CHECK: f5:
|
|
; CHECK: lrv %r2, -524288(%r2)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i32 *%src, i64 -131072
|
|
%a = load i32 *%ptr
|
|
%swapped = call i32 @llvm.bswap.i32(i32 %a)
|
|
ret i32 %swapped
|
|
}
|
|
|
|
; Check the next word down, which needs separate address logic.
|
|
; Other sequences besides this one would be OK.
|
|
define i32 @f6(i32 *%src) {
|
|
; CHECK: f6:
|
|
; CHECK: agfi %r2, -524292
|
|
; CHECK: lrv %r2, 0(%r2)
|
|
; CHECK: br %r14
|
|
%ptr = getelementptr i32 *%src, i64 -131073
|
|
%a = load i32 *%ptr
|
|
%swapped = call i32 @llvm.bswap.i32(i32 %a)
|
|
ret i32 %swapped
|
|
}
|
|
|
|
; Check that LRV allows an index.
|
|
define i32 @f7(i64 %src, i64 %index) {
|
|
; CHECK: f7:
|
|
; CHECK: lrv %r2, 524287({{%r3,%r2|%r2,%r3}})
|
|
; CHECK: br %r14
|
|
%add1 = add i64 %src, %index
|
|
%add2 = add i64 %add1, 524287
|
|
%ptr = inttoptr i64 %add2 to i32 *
|
|
%a = load i32 *%ptr
|
|
%swapped = call i32 @llvm.bswap.i32(i32 %a)
|
|
ret i32 %swapped
|
|
}
|
|
|
|
; Check that volatile accesses do not use LRV, which might access the
|
|
; storage multple times.
|
|
define i32 @f8(i32 *%src) {
|
|
; CHECK: f8:
|
|
; CHECK: l [[REG:%r[0-5]]], 0(%r2)
|
|
; CHECK: lrvr %r2, [[REG]]
|
|
; CHECK: br %r14
|
|
%a = load volatile i32 *%src
|
|
%swapped = call i32 @llvm.bswap.i32(i32 %a)
|
|
ret i32 %swapped
|
|
}
|