
Re: VFRECIP/VFRSQRT instructions
I had been hoping that the reference C code would scratch that itch, but you're probably right. I've added a tiny example and a huge example for each.
My mistake. I was thinking of the fact that
I had been hoping that the reference C code would scratch that itch, but you're probably right. I've added a tiny example and a huge example for each.
My mistake. I was thinking of the fact that

By
Andrew Waterman
·
#345
·


Re: VFRECIP/VFRSQRT instructions
On 8/12/20 7:05 PM, Andrew Waterman wrote:
Probably so.
That's probably OK. It's much less confusing. I wonder if two examples for each (recip and rsqrt) would help. One with a denormal input and
On 8/12/20 7:05 PM, Andrew Waterman wrote:
Probably so.
That's probably OK. It's much less confusing. I wonder if two examples for each (recip and rsqrt) would help. One with a denormal input and

By
Bill Huffman
·
#344
·


Re: VFRECIP/VFRSQRT instructions
Yeah, but big is OK, I think.
I didn't change the rsqrt table at all. Since the subnormal cases are mostly uninteresting, I think the NOTE that positive subnormal and normal inputs always produce
Yeah, but big is OK, I think.
I didn't change the rsqrt table at all. Since the subnormal cases are mostly uninteresting, I think the NOTE that positive subnormal and normal inputs always produce

By
Andrew Waterman
·
#343
·


Re: VFRECIP/VFRSQRT instructions
On 8/12/20 4:21 PM, Andrew Waterman wrote:
It is pretty big... I'm just looking at the recip at this point. I have a couple of thoughts:
In the "Output" column for the 5 new positive and negative
On 8/12/20 4:21 PM, Andrew Waterman wrote:
It is pretty big... I'm just looking at the recip at this point. I have a couple of thoughts:
In the "Output" column for the 5 new positive and negative

By
Bill Huffman
·
#342
·


Re: VFRECIP/VFRSQRT instructions
How about this... it's a beast, but I think it works. https://github.com/riscv/riscvvspec/blob/vfrecip/vspec.adoc#149vectorfloatingpointreciprocalestimateinstruction
How about this... it's a beast, but I think it works. https://github.com/riscv/riscvvspec/blob/vfrecip/vspec.adoc#149vectorfloatingpointreciprocalestimateinstruction

By
Andrew Waterman
·
#341
·


Re: VFRECIP/VFRSQRT instructions
On 8/12/20 3:32 PM, Andrew Waterman wrote:
I almost suggested expanding the current table. That makes it quite a bit larger. But then, it also means there's no need to clarify the relationship
On 8/12/20 3:32 PM, Andrew Waterman wrote:
I almost suggested expanding the current table. That makes it quite a bit larger. But then, it also means there's no need to clarify the relationship

By
Bill Huffman
·
#340
·


Re: VFRECIP/VFRSQRT instructions
Yeah, let me play around with the presentation a bit. I'm not sure whether breaking it into two tables or expanding the current table will be clearer, but your suggestion holds either way. Thanks
Yeah, let me play around with the presentation a bit. I'm not sure whether breaking it into two tables or expanding the current table will be clearer, but your suggestion holds either way. Thanks

By
Andrew Waterman
·
#339
·


Re: VFRECIP/VFRSQRT instructions
On 8/11/20 4:11 PM, Andrew Waterman wrote:
With that reorientation to what the instruction means, it looks correct. I have a couple of comments:
Just above the table you use the concept of the
On 8/11/20 4:11 PM, Andrew Waterman wrote:
With that reorientation to what the instruction means, it looks correct. I have a couple of comments:
Just above the table you use the concept of the

By
Bill Huffman
·
#338
·


Re: VFRECIP/VFRSQRT instructions
Ah, that clarifies your earlier question. Yeah, LMK what you think.
Ah, that clarifies your earlier question. Yeah, LMK what you think.

By
Andrew Waterman
·
#337
·


Re: VFRECIP/VFRSQRT instructions
On 8/11/20 3:00 PM, Andrew Waterman wrote:
Ah, so you're counting the 7bit (plus hidden bit) result as the absolutely correct answer. There's no relationship here to the infinite precision
On 8/11/20 3:00 PM, Andrew Waterman wrote:
Ah, so you're counting the 7bit (plus hidden bit) result as the absolutely correct answer. There's no relationship here to the infinite precision

By
Bill Huffman
·
#336
·


Re: Integer Overflow/Saturation Operations
As Cohen responded. The examples provided only work, as you rightly noted, if you have implied guard bits in the system.
This is fine, with the caveat that the ISA definition doesn't provision for the
As Cohen responded. The examples provided only work, as you rightly noted, if you have implied guard bits in the system.
This is fine, with the caveat that the ISA definition doesn't provision for the

By
Mikael <mikael.mortensen@...>
·
#335
·


Re: Fixed Point (Chapter 13): Clarification Request
I like the new definition of fixed point. Its quite crisp.
Building on the definition we can now argue that any integer number representation is: integer value/2^N
With N being a positive
I like the new definition of fixed point. Its quite crisp.
Building on the definition we can now argue that any integer number representation is: integer value/2^N
With N being a positive

By
Mikael <mikael.mortensen@...>
·
#334
·


Re: VFRECIP/VFRSQRT instructions
There are no cases where UF should be raised because there are no cases where denormalization causes loss of precision. When the result is subnormal, it is only subnormal by either one or two
There are no cases where UF should be raised because there are no cases where denormalization causes loss of precision. When the result is subnormal, it is only subnormal by either one or two

By
Andrew Waterman
·
#333
·


Re: VFRECIP/VFRSQRT instructions
Hi Andrew,
I'm looking at the cases where the reciprocal is near the boundary between finite and infinite or between normal and denormal. Are you trying to get the boundaries approximately right?
Hi Andrew,
I'm looking at the cases where the reciprocal is near the boundary between finite and infinite or between normal and denormal. Are you trying to get the boundaries approximately right?

By
Bill Huffman
·
#332
·


Re: VFRECIP/VFRSQRT instructions
I've PRed a full definition of these instructions. Please sanitycheck my work:
I've PRed a full definition of these instructions. Please sanitycheck my work:

By
Andrew Waterman
·
#331
·


Re: Fixed Point (Chapter 13): Clarification Request
I always understood fixpoint to have an implicit denominator that was restricted to a positive (integer) power of 2 (which could be fixed, or could be configured).
Arbitrary denominators mean that
I always understood fixpoint to have an implicit denominator that was restricted to a positive (integer) power of 2 (which could be fixed, or could be configured).
Arbitrary denominators mean that

By
Allen Baum
·
#330
·


Re: Fixed Point (Chapter 13): Clarification Request
Nick,
Thank you for your response and proposed clarification.
This proposal for how to use the numbers fundamentally realigns our interpretation of how this would be used. This is much closer to the
Nick,
Thank you for your response and proposed clarification.
This proposal for how to use the numbers fundamentally realigns our interpretation of how this would be used. This is much closer to the

By
CDS <cohen.steed@...>
·
#329
·


Re: Fixed Point (Chapter 13): Clarification Request
Hi Coheen,
Thanks for the discussion on the fixedpoint vector instructions. Most of Chapter 13 predates my involvement with the Task Group, but I think I am able to address one of your
Hi Coheen,
Thanks for the discussion on the fixedpoint vector instructions. Most of Chapter 13 predates my involvement with the Task Group, but I think I am able to address one of your

By
Nick Knight
·
#328
·


Re: Fixed Point (Chapter 13): Clarification Request
perhaps i am not upto date on this topic. but addresses are fixed point. (integers). and you need vector support for vector loads using the vector accumulators.(indexs). the math, other than
perhaps i am not upto date on this topic. but addresses are fixed point. (integers). and you need vector support for vector loads using the vector accumulators.(indexs). the math, other than

By
swallach
·
#327
·


Re: Integer Overflow/Saturation Operations
Andy,
Thank you for your response.
The concern I'm raising is less about "How do I avoid overflow?" and more about "Why are we avoiding the specification of saturating instructions, or an overflow
Andy,
Thank you for your response.
The concern I'm raising is less about "How do I avoid overflow?" and more about "Why are we avoiding the specification of saturating instructions, or an overflow

By
CDS <cohen.steed@...>
·
#326
·
