<html>
<head>
<base href="https://bugs.llvm.org/">
</head>
<body><table border="1" cellspacing="0" cellpadding="8">
<tr>
<th>Bug ID</th>
<td><a class="bz_bug_link
bz_status_NEW "
title="NEW - [SIMD] *dest = vec[lane] should generate store*_lane instruction"
href="https://bugs.llvm.org/show_bug.cgi?id=50800">50800</a>
</td>
</tr>
<tr>
<th>Summary</th>
<td>[SIMD] *dest = vec[lane] should generate store*_lane instruction
</td>
</tr>
<tr>
<th>Product</th>
<td>libraries
</td>
</tr>
<tr>
<th>Version</th>
<td>trunk
</td>
</tr>
<tr>
<th>Hardware</th>
<td>PC
</td>
</tr>
<tr>
<th>OS</th>
<td>Windows NT
</td>
</tr>
<tr>
<th>Status</th>
<td>NEW
</td>
</tr>
<tr>
<th>Severity</th>
<td>enhancement
</td>
</tr>
<tr>
<th>Priority</th>
<td>P
</td>
</tr>
<tr>
<th>Component</th>
<td>Backend: WebAssembly
</td>
</tr>
<tr>
<th>Assignee</th>
<td>unassignedbugs@nondot.org
</td>
</tr>
<tr>
<th>Reporter</th>
<td>clang@evan.coeusgroup.com
</td>
</tr>
<tr>
<th>CC</th>
<td>llvm-bugs@lists.llvm.org
</td>
</tr></table>
<p>
<div>
<pre>This is very similar to #50792 (which is the load*_lane version, this is for
store*_lane). Both are blocking the implementation of overloaded functions for
these operations in WAV.
Assigning an element from a vector to a pointer should use one of the
v128.store*_lane instructions, but right now uses an extract_lane and a scalar
store.
Example (Compiler Explorer: <a href="https://godbolt.org/z/a1cz8hexW">https://godbolt.org/z/a1cz8hexW</a>):
#include <stdint.h>
typedef int8_t i8x16 __attribute__((__vector_size__(16)));
typedef int16_t i16x8 __attribute__((__vector_size__(16)));
typedef int32_t i32x4 __attribute__((__vector_size__(16)));
typedef int64_t i64x2 __attribute__((__vector_size__(16)));
void
i8x16_store_lane(int8_t * dest, i8x16 src) {
*dest = src[1];
}
void
i16x8_store_lane(int16_t * dest, i16x8 src) {
*dest = src[1];
}
void
i32x4_store_lane(int32_t * dest, i32x4 src) {
*dest = src[1];
}
void
i64x2_store_lane(int64_t * dest, i64x2 src) {
*dest = src[1];
}
void
i8x16_load_lane_intrin(int8_t * dest, i8x16 src) {
__builtin_wasm_store8_lane(dest, src, 1);
}
void
i16x8_load_lane_intrin(int16_t * dest, i16x8 src) {
__builtin_wasm_store16_lane(dest, src, 1);
}
void
i32x4_load_lane_intrin(int32_t * dest, i32x4 src) {
__builtin_wasm_store32_lane(dest, src, 1);
}
void
i64x2_load_lane_intrin(int64_t * dest, i64x2 src) {
__builtin_wasm_store64_lane(dest, src, 1);
}</pre>
</div>
</p>
<hr>
<span>You are receiving this mail because:</span>
<ul>
<li>You are on the CC list for the bug.</li>
</ul>
</body>
</html>