Alexei Starovoitov writes:
> On Wed, Mar 27, 2019 at 05:18:35PM +, Jiong Wang wrote:
>>
>> > On 27 Mar 2019, at 17:17, Alexei Starovoitov
>> > wrote:
>> >
>> > On Wed, Mar 27, 2019 at 05:06:01PM +, Jiong Wang wrote:
>> >>
>> >>> On 27 Mar 2019, at 17:00, Alexei Starovoitov
>> >>>
On Wed, Mar 27, 2019 at 05:18:35PM +, Jiong Wang wrote:
>
> > On 27 Mar 2019, at 17:17, Alexei Starovoitov
> > wrote:
> >
> > On Wed, Mar 27, 2019 at 05:06:01PM +, Jiong Wang wrote:
> >>
> >>> On 27 Mar 2019, at 17:00, Alexei Starovoitov
> >>> wrote:
> >>>
> >>> On Tue, Mar 26, 2019
> On 27 Mar 2019, at 17:17, Alexei Starovoitov
> wrote:
>
> On Wed, Mar 27, 2019 at 05:06:01PM +, Jiong Wang wrote:
>>
>>> On 27 Mar 2019, at 17:00, Alexei Starovoitov
>>> wrote:
>>>
>>> On Tue, Mar 26, 2019 at 06:05:29PM +, Jiong Wang wrote:
After previous patches, verifier h
On Wed, Mar 27, 2019 at 05:06:01PM +, Jiong Wang wrote:
>
> > On 27 Mar 2019, at 17:00, Alexei Starovoitov
> > wrote:
> >
> > On Tue, Mar 26, 2019 at 06:05:29PM +, Jiong Wang wrote:
> >> After previous patches, verifier has marked those instructions that really
> >> need zero extension
> On 27 Mar 2019, at 17:00, Alexei Starovoitov
> wrote:
>
> On Tue, Mar 26, 2019 at 06:05:29PM +, Jiong Wang wrote:
>> After previous patches, verifier has marked those instructions that really
>> need zero extension on dst_reg.
>>
>> It is then for all back-ends to decide how to use such
On Tue, Mar 26, 2019 at 06:05:29PM +, Jiong Wang wrote:
> After previous patches, verifier has marked those instructions that really
> need zero extension on dst_reg.
>
> It is then for all back-ends to decide how to use such information to
> eliminate unnecessary zero extension codegen during
After previous patches, verifier has marked those instructions that really
need zero extension on dst_reg.
It is then for all back-ends to decide how to use such information to
eliminate unnecessary zero extension codegen during JIT compilation.
One approach is:
1. Verifier insert explicit zero
