Hwacha source code

[Xinfei Guo]

Is the Hwacha source code in public now? or is it going to be? I didn't find it in the http://github.com/ucb-bar repository. Thanks!

Xinfei

[Ben Keller]

The Hwacha source is not currently public, but my understanding is that it will be open-sourced soon.  Stay tuned!

-Ben

On Jun 1, 2016, at 14:19, Xinfei Guo <xg...@virginia.edu> wrote:

Is the Hwacha source code in public now? or is it going to be? I didn't find it in the http://github.com/ucb-bar repository. Thanks!

Xinfei

--
You received this message because you are subscribed to the Google Groups "RISC-V HW Dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hw-dev+un...@groups.riscv.org.
To post to this group, send email to hw-...@groups.riscv.org.
Visit this group at https://groups.google.com/a/groups.riscv.org/group/hw-dev/.
To view this discussion on the web visit https://groups.google.com/a/groups.riscv.org/d/msgid/hw-dev/CACpn_Nkw-%2Br6AeFkdxzKRei5cZRBHTnHw%2BKaKMs-VNiUTWiWcw%40mail.gmail.com.

[Parviz Palangpour]

Does anyone know if Hwacha is still going to be open sourced? I understand Hwacha predates the RISC-V vector extension but I thought there might be a tangential announcement during the "V Vector Extension Proposal" talk from the workshop today -- I'm not able to attend but didn't see anything referring to it in the lowRISC live blog: http://www.lowrisc.org/blog/2016/11/fifth-risc-v-workshop-day-two/

[Alex Bradbury]

On 30 November 2016 at 18:28, Parviz Palangpour <par...@gmail.com> wrote:
> Does anyone know if Hwacha is still going to be open sourced? I understand
> Hwacha predates the RISC-V vector extension but I thought there might be a
> tangential announcement during the "V Vector Extension Proposal" talk from
> the workshop today -- I'm not able to attend but didn't see anything
> referring to it in the lowRISC live blog:
> http://www.lowrisc.org/blog/2016/11/fifth-risc-v-workshop-day-two/

I didn't note it, but Krste did comment that Hwacha should be opened 'soon'.

Best,

Alex

[Colin Schmidt]

It was not released today but will still be released. I want it to be as usable out of the box as possible to avoid being inundated with support requests.

To unsubscribe from this group and stop receiving emails from it, send an email to hw-dev+unsubscribe@groups.riscv.org.

To post to this group, send email to hw-...@groups.riscv.org.
Visit this group at https://groups.google.com/a/groups.riscv.org/group/hw-dev/.

To view this discussion on the web visit https://groups.google.com/a/groups.riscv.org/d/msgid/hw-dev/b022f93b-401e-47f2-a096-791a8fbda18e%40groups.riscv.org.

[promach]

May I know where the source of Hwacha is ?

https://github.com/ucb-bar/riscv-benchmarks/tree/master/hwacha is only the c code, not the chisel code.

On Thursday, December 1, 2016 at 2:35:33 AM UTC+8, Colin Schmidt wrote:

It was not released today but will still be released. I want it to be as usable out of the box as possible to avoid being inundated with support requests.

On Wed, Nov 30, 2016 at 10:28 AM, Parviz Palangpour <par...@gmail.com> wrote:

Does anyone know if Hwacha is still going to be open sourced? I understand Hwacha predates the RISC-V vector extension but I thought there might be a tangential announcement during the "V Vector Extension Proposal" talk from the workshop today -- I'm not able to attend but didn't see anything referring to it in the lowRISC live blog: http://www.lowrisc.org/blog/2016/11/fifth-risc-v-workshop-day-two/

On Thursday, June 2, 2016 at 9:51:36 AM UTC-5, bkeller wrote:

The Hwacha source is not currently public, but my understanding is that it will be open-sourced soon.  Stay tuned!

-Ben

On Jun 1, 2016, at 14:19, Xinfei Guo <xg...@virginia.edu> wrote:

Is the Hwacha source code in public now? or is it going to be? I didn't find it in the http://github.com/ucb-bar repository. Thanks!

Xinfei

--
You received this message because you are subscribed to the Google Groups "RISC-V HW Dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hw-dev+un...@groups.riscv.org.
To post to this group, send email to hw-...@groups.riscv.org.
Visit this group at https://groups.google.com/a/groups.riscv.org/group/hw-dev/.
To view this discussion on the web visit https://groups.google.com/a/groups.riscv.org/d/msgid/hw-dev/CACpn_Nkw-%2Br6AeFkdxzKRei5cZRBHTnHw%2BKaKMs-VNiUTWiWcw%40mail.gmail.com.

--
You received this message because you are subscribed to the Google Groups "RISC-V HW Dev" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hw-dev+un...@groups.riscv.org.
To post to this group, send email to hw-...@groups.riscv.org.
Visit this group at https://groups.google.com/a/groups.riscv.org/group/hw-dev/.

To view this discussion on the web visit https://groups.google.com/a/groups.riscv.org/d/msgid/hw-dev/b022f93b-401e-47f2-a096-791a8fbda18e%40groups.riscv.org.

[Luke Kenneth Casson Leighton]

On Wed, Jun 27, 2018 at 4:36 AM, promach <feiph...@gmail.com> wrote:

> May I know where the source of Hwacha is ?

many people have asked and have not received a response.

> https://github.com/ucb-bar/riscv-benchmarks/tree/master/hwacha is only the c
> code, not the chisel code.

that looks like it is the source code of the benchmark code for
hwacha, not the source code of hwacha.

i was able to determine through this:
http://opensource.rock-chips.com/images/4/4e/Rockchip_RK3288_Datasheet_V2.2-20170301.pdf

that there also exists a version of spike that has RVV fully-implemented.

where may the source code for that be obtained?

l.

[promach]

I just came across https://github.com/babydream/lab3/tree/master/hwacha

Anyone have used this before in real FPGA hardware ?

[Luke Kenneth Casson Leighton]

---

On Wed, Jun 27, 2018 at 9:35 AM, promach <feiph...@gmail.com> wrote:

> I just came across https://github.com/babydream/lab3/tree/master/hwacha

FINALLY! all right! people have been looking for and asking that
for MONTHS if not YEARS. dang. well done.

l.

[fei phung]

Have anyone used RISCV Hwacha Vector Extension on Zedboard before ? Could anyone give some personal advice on this ?

https://github.com/ucb-bar/fpga-zynq

https://github.com/ucb-bar/riscv-benchmarks/tree/master/hwacha

http://msyksphinz.hatenablog.com/entry/advent20161215

https://github.com/freechipsproject/chisel3/issues/143

https://github.com/babydream/lab3/tree/master/hwacha

https://www2.eecs.berkeley.edu/Pubs/TechRpts/2015/EECS-2015-262.pdf

https://riscv.org/wp-content/uploads/2016/12/Wed0930-RISC-V-Vectors-Asanovic-UC-Berkeley-SiFive.pdf

[Colin Schmidt]

The source code of Hwacha linked in this thread is for an older version with a different(unsupported) ISA then the 2015 tech reports. It was uploaded by a student in one of the UC Berkeley classes despite them being told not to upload the labs(for various reasons).

It is also not the standard vector extension discussed in the linked talked or being worked on by the Vector extension working group.

We are preparing for an open source release of a current version of Hwacha soon. I will make an announcement on the hw-dev list serve when that happens since I know that many in the community are eagerly awaiting it.

Thanks for all of your interest,

Colin

[Luke Kenneth Casson Leighton]

On Sun, Jul 1, 2018 at 9:03 PM, Colin Schmidt <col...@eecs.berkeley.edu> wrote:

> We are preparing for an open source release of a current version of Hwacha
> soon. I will make an announcement on the hw-dev list serve when that happens
> since I know that many in the community are eagerly awaiting it.

colin i appreciate the update. this conversation i am, in advance,
informing you that i am following the advice listed here, so that you
understand clearly in advance that i am wriing with "good intent":
http://www.crnhq.org/CR-Kit.aspx?rw=c#assertiveness

whilst i cannot speak for other people, i understand from what you
say that you believe i am "awaiting eagerly". i am not. i am seeking
*active participation* and am prevented and prohibited from doing so.
this leads to a feeling of total disempowerment and a *lot* of
frustration. what i would like instead is for there to be no barriers
to discussion and participation, and the RISC-V Membership Agreement
is a barrier (due to conflicts between Sections 1.9 and 5, for anyone
in the Software Libre world).

there are some groups that are happy for the Working Groups to do all
of the work. that's great. if they are students they will learn from
what the WGs do. if they are commercial companies they will save
money by doing so.

neither of these groups needs to participate. they just need to
"eagerly await".

but that leaves a group of specialist people who know what "truly
open" means who are looking at the working practices of the RISC-V
Foundation and see a repeat of Google's failed "Project Ara" policies.
"it's open... but only as long as you join our clique and agree to our
privileged rules".

i would be interested to hear your thoughts... and i await a time in
the immediate future where access to the source code, documentation,
specification, mailing list, forum, spike simulation, gcc and binutils
tools are opened up *without restriction* - not "when it's ready" or
"when it's finished" - because that's what "open" really is.

bottom line, you don't need to "prepare" for a release, you just do
it. right now.

l.