Displaying 4 results from an estimated 4 matches for "field1trunc".
Did you mean:
field1trunced
2013 Aug 09
2
[LLVMdev] [RFC] Poor code generation for paired load
...generally better than loading each field on its own, this can lead to poor code generation when the operations extracting the fields are more expensive than a load or when “fancy” loads are available.
More generally, this may happen for smaller size too.
** Example **
1. %chunk64 = load i64
2. %field1trunced = trunc i64 %chunk64 to i32 // <— build field1 from chunk
3. %field1float = bitcast i32 field1trunced to float // <— build field1 from chunk
4. %field2shifted = lshr i64 %chunk64, 32 // <— build field2 from chunk
5. %field2trunced = trunc i64 %field2shifter...
2013 Aug 12
2
[LLVMdev] [RFC] Poor code generation for paired load
...eneration when the operations extracting the
>> fields are more expensive than a load or when “fancy” loads are available.
>>
>> More generally, this may happen for smaller size too.
>>
>>
>> ** Example **
>>
>> 1. %chunk64 = load i64
>> 2. %field1trunced = trunc i64 %chunk64 to i32 // <— build field1
>> from chunk
>> 3. %field1float = bitcast i32 field1trunced to float // <— build
>> field1 from chunk
>> 4. %field2shifted = lshr i64 %chunk64, 32 // <— build field2
>> from chunk...
2013 Aug 10
0
[LLVMdev] [RFC] Poor code generation for paired load
...s own,
> this can lead to poor code generation when the operations extracting the
> fields are more expensive than a load or when “fancy” loads are available.
>
> More generally, this may happen for smaller size too.
>
>
> ** Example **
>
> 1. %chunk64 = load i64
> 2. %field1trunced = trunc i64 %chunk64 to i32 // <— build field1
> from chunk
> 3. %field1float = bitcast i32 field1trunced to float // <— build
> field1 from chunk
> 4. %field2shifted = lshr i64 %chunk64, 32 // <— build field2
> from chunk
> 5. %field2trun...
2013 Aug 12
0
[LLVMdev] [RFC] Poor code generation for paired load
...s own,
> this can lead to poor code generation when the operations extracting the
> fields are more expensive than a load or when “fancy” loads are available.
>
> More generally, this may happen for smaller size too.
>
>
> ** Example **
>
> 1. %chunk64 = load i64
> 2. %field1trunced = trunc i64 %chunk64 to i32 // <— build field1
> from chunk
> 3. %field1float = bitcast i32 field1trunced to float // <— build
> field1 from chunk
> 4. %field2shifted = lshr i64 %chunk64, 32 // <— build field2
> from chunk
> 5. %field2trun...