Libre RISC-V Mailing list Archives (accessible via public-inbox)
 help / color / Atom feed
* Re: [libre-riscv-dev] go die/cancel signals in fsm
From: Luke Kenneth Casson Leighton @ 2020-07-22  9:51 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAC2bXD4+-OuvUk7m1s2F4pZZgkgaYL9RP_thL_ntL0QDdtkhfQ@mail.gmail.com>

On Wed, Jul 22, 2020 at 3:01 AM Jacob Lifshay <programmerjake@gmail.com> wrote:
>
> I looked through the vcd generated for soc.simple.test.test_core and
> was unable to locate the signals used to tell the pipeline stages that
> the instruction they are currently operating on is canceled.

that's because we're using SimpleHandshakeRedir rather than
MaskCancellable, and are not going to do cancellation in the 180nm
core unless we get more people, and i can focus in the remaining very
short amount of time on doing absolutely nothing but add in the
Dependency Matrices.

three months is already far too short to even consider doing that, so
it is extremely unlikely to be added.

> This is
> needed to not have to wait 63(!) more clock cycles for a canceled div
> instruction to finish computing before another div instruction can be
> started.

indeed.

ok so i added some code comments explaining what goes into CommonPipeSpec:
https://git.libre-soc.org/?p=soc.git;a=commitdiff;h=71af4fadd8a431079df13fc5812d745c133074be

i'll also update the README...

https://git.libre-soc.org/?p=soc.git;a=commitdiff;h=593c41881b455aeeb19fe98071647d7b4ffca15c

l.

_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] daily kan-ban update, 21jul2020
From: Luke Kenneth Casson Leighton @ 2020-07-22  9:08 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAEg7hu=rbqS=tjmW=OBGw5fYaixLr=LOcsvnJ09aO-uh4VVwLA@mail.gmail.com>

On Tue, Jul 21, 2020 at 10:33 PM Cole Poirier <colepoirier@gmail.com> wrote:
>
> Last week:
>
> * worked on coriolis2 chroot setup script (very slow going as it
> requires >5 GIGAbytes of downloads each test)

yeah don't put up with that - you should have mentioned it takes that
long, sooner.  use an apt proxy and squid.  "apt-get install
apt-cacher-ng squid".

>
> Today:
>
> * finish coriolis2 chroot setup script and HDL_Workflow/coriolis2 wiki
> documentation

excellent.

l.

_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] Fwd: [Mesa-dev] [XDC 2020] Virtual conference + Call for Proposals extended 2 weeks more
From: Luke Kenneth Casson Leighton @ 2020-07-22  9:06 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAEg7hunmmr1U+zMx9OAY2Mzu5=BQ46ypkLzDutmbDjF4ue1nEQ@mail.gmail.com>

On Tue, Jul 21, 2020 at 10:38 PM Cole Poirier <colepoirier@gmail.com> wrote:
>
> I'd like to collaborate with you two on this. I think the 'pitch' I
> sent you - Luke - for your corrections and augmentations a couple
> weeks ago (I assume the work on the core made working on this
> impossible) can be used as a basis/starting point for this
> presentation.

pretty much - i'm not really able to "switch tasks" without dropping
significant working knowledge.  if i'm going to write a presentation
it needs to be done 100% focus, all in one hit, nothing else but that
for 1-2 days.

l.

_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* [libre-riscv-dev] go die/cancel signals in fsm
From: Jacob Lifshay @ 2020-07-22  2:01 UTC (permalink / raw)
  To: Libre-RISCV General Development

I looked through the vcd generated for soc.simple.test.test_core and
was unable to locate the signals used to tell the pipeline stages that
the instruction they are currently operating on is canceled. This is
needed to not have to wait 63(!) more clock cycles for a canceled div
instruction to finish computing before another div instruction can be
started.

Help wanted,
Jacob

_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] Fwd: [Mesa-dev] [XDC 2020] Virtual conference + Call for Proposals extended 2 weeks more
From: Cole Poirier @ 2020-07-21 21:38 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAPweEDxb44ynto+KyVEdcd7Sy6sOH1hJkrpROB2PuP=a7ZGYJA@mail.gmail.com>

I'd like to collaborate with you two on this. I think the 'pitch' I
sent you - Luke - for your corrections and augmentations a couple
weeks ago (I assume the work on the core made working on this
impossible) can be used as a basis/starting point for this
presentation.

What do you think?

Cole

On Mon, Jul 20, 2020 at 4:30 AM Luke Kenneth Casson Leighton
<lkcl@lkcl.net> wrote:
>
> https://xdc2020.x.org/event/9/registrations/7/
>
> can i recommend registering, just to show some support for xdc?  there are
> only 33 people registered online (there is no charge)
>
> l.
>
>
>
> --
> ---
> crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
> _______________________________________________
> libre-riscv-dev mailing list
> libre-riscv-dev@lists.libre-riscv.org
> http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] daily kan-ban update, 21jul2020
From: Cole Poirier @ 2020-07-21 21:33 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAPweEDxiv-YfodPv6ZaU9o5D2Hm+ui7tbf4T3jhXbiZ82XRqFA@mail.gmail.com>

Last week:

* worked on coriolis2 chroot setup script (very slow going as it
requires >5 GIGAbytes of downloads each test)

Today:

* finish coriolis2 chroot setup script and HDL_Workflow/coriolis2 wiki
documentation
* work on finishing the setup CI testing on jacob's build server so we
know pretty quickly when there are dependency and version problems

Jacob, what is left to do in order to get the build server into a
shape that we will be happy with/will be useful?

I referred back to the original bug report "Bug 190 - Setup Gitlab CI
Runner for Kazan on a computer"
<https://bugs.libre-soc.org/show_bug.cgi?id=190> but the last comment
on that seems to be from before gitlab was even setup on the server.
Can you either point me to the current bug report for this, or leave a
comment on this old one with the steps I need to complete to finish
setting up our testing infrastructure?

Cole

On Tue, Jul 21, 2020 at 10:02 AM Luke Kenneth Casson Leighton
<lkcl@lkcl.net> wrote:
>
> tasks:
>
> https://bugs.libre-soc.org/showdependencytree.cgi?id=383&maxdepth=1&hide_resolved=1
>
> yesterday:
>
> * nmigen bi-weekly meeting, went well.
> * reviewed trap with help from samuel
> * set up core to be better started from litex
> * did a first cut litex core.py with help from florian
> * re-ran cxxsim, confirmed a bug with help from cesar.
>
> today:
>
> * helped JeanPaul get up-to-date
> * fixed missing exception MSR bit setting
> * decided to move both MSR and CIA into "state"
> * found a bug in CompALUMulti when it has no regs to read.
>
> this latter will need a special test case for, and some investigation.  so
> that is the rest of the day.
>
> l.
>
>
>
> --
> ---
> crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
> _______________________________________________
> libre-riscv-dev mailing list
> libre-riscv-dev@lists.libre-riscv.org
> http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* [libre-riscv-dev] daily kan-ban update, 21jul2020
From: Luke Kenneth Casson Leighton @ 2020-07-21 17:02 UTC (permalink / raw)
  To: Libre-RISCV General Development

tasks:

https://bugs.libre-soc.org/showdependencytree.cgi?id=383&maxdepth=1&hide_resolved=1

yesterday:

* nmigen bi-weekly meeting, went well.
* reviewed trap with help from samuel
* set up core to be better started from litex
* did a first cut litex core.py with help from florian
* re-ran cxxsim, confirmed a bug with help from cesar.

today:

* helped JeanPaul get up-to-date
* fixed missing exception MSR bit setting
* decided to move both MSR and CIA into "state"
* found a bug in CompALUMulti when it has no regs to read.

this latter will need a special test case for, and some investigation.  so
that is the rest of the day.

l.



-- 
---
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] nice timing diagram renderer
From: Luke Kenneth Casson Leighton @ 2020-07-21  3:48 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAPweEDzZQCsN3GArjUJMnJDzFSAZ5jqYBrk=1yB=zARy22f=1g@mail.gmail.com>

On Monday, July 20, 2020, Luke Kenneth Casson Leighton <lkcl@lkcl.net>
wrote:

>
>
> the infrastructure is in place.
>
> i have already set up docs.libre-soc.org and added a build / upload
> option to the top level Makefile:
>
> https://git.libre-soc.org/?p=soc.git;a=blob;f=Makefile;hb=HEAD
>


https://git.libre-soc.org/?p=soc.git;a=blob;f=index.rst;h=86921d65b56d96e48e0365238a9f37c33e7bfd6d;hb=8cf1f4f4919c4416c200371fb33d3a08a8f9bce7

the default is rst, i enabled napoleon plugin such that markdown is also
supported.

 hmmm the nginx config still needs sorting.

l.



-- 
---
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] nice timing diagram renderer
From: Luke Kenneth Casson Leighton @ 2020-07-20 22:12 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <12E4B1C8-0769-4496-8E6D-9052EB5FF376@gatech.edu>

On Monday, July 20, 2020, Yehowshua <yimmanuel3@gatech.edu> wrote:

> > I found a nice open source timing diagram renderer:
> > https://wavedrom.com/
>
>
> Yes. Wavedrom generates nice diagrams mainly for documentation.
> It does require a bit of fiddling to make it work.
>
> I don’t think it will help us until we start working on making
> good docs.


the infrastructure is in place.

i have already set up docs.libre-soc.org and added a build / upload option
to the top level Makefile:

https://git.libre-soc.org/?p=soc.git;a=blob;f=Makefile;hb=HEAD

there was a conversation on irc with mithro that prompted me to investigate
and set it up.

wavedrom itself made me hesitant about installing it / adding it as a
dependency because it is a node application (iirc).

by contrast the graphviz builder sphinx plugin was dead simple and worked
immediately, it uses yosys and graphviz which we already use.

l.



-- 
---
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] daily kan-ban update 20jul2020
From: Luke Kenneth Casson Leighton @ 2020-07-20 22:05 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <1e925e7e-a83f-cf97-75aa-2ab0baf1bf11@gmail.com>

On Monday, July 20, 2020, Cesar Strauss <cestrauss@gmail.com> wrote:

> Yesterday:
>
> * Investigate the possibility to generate useful, structured, commented,
> color-formatted gtkw templates in test cases, using the vcd.gtkw module
> from pyvcd (which is already a dependency of nMigen).


oo niice, because i constantly find i have to add colour to groups of
signals.

l.




-- 
---
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] nice timing diagram renderer
From: Yehowshua @ 2020-07-20 20:13 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAC2bXD4h0u4XsqRa8GpmS5RU1HMUgy_ndNgnod0A5zom-g7UZw@mail.gmail.com>

> I found a nice open source timing diagram renderer:
> https://wavedrom.com/


Yes. Wavedrom generates nice diagrams mainly for documentation.
It does require a bit of fiddling to make it work.

I don’t think it will help us until we start working on making
good docs.

Yehowshua
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* [libre-riscv-dev] nice timing diagram renderer
From: Jacob Lifshay @ 2020-07-20 20:10 UTC (permalink / raw)
  To: Libre-RISCV General Development

I found a nice open source timing diagram renderer:
https://wavedrom.com/

some pretty examples:
https://wavedrom.com/tutorial.html
https://wavedrom.com/tutorial2.html

I'm not 100% sure, but I think there is code to convert vcd files to
wavedrom's json format

Jacob
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* [libre-riscv-dev] daily kan-ban update 20jul2020
From: Cesar Strauss @ 2020-07-20 19:14 UTC (permalink / raw)
  To: libre-riscv-dev

Yesterday:

* Finished an example of a FSM-based Shifter:
src/soc/experiment/alu_fsm.py
https://bugs.libre-soc.org/show_bug.cgi?id=417

* Started investigating a difference in simulation results between pysim
and cxxsim, with regards to the Shifter example.
It was established that cxxsim probably has a bug.
https://github.com/nmigen/nmigen/issues/439

Today:

* Continue investigating the potential bug in cxxsim.
I think the test case can be simplified by taking out the SHIFT state
and keeping just the handshake states (IDLE and DONE). This would imply,
in turn, that a one bit register should be enough to trigger the problem.
Will add my findings to nMigen #439.

* Fix alu_fsm.py to work in pysim again.

Next:

* Shave an extra cycle from the Shifter, by short-circuiting IDLE and
going straight to SHIFT, when valid_i is already high at DONE. Most
likely, will do it in a new Shifter, to keep this one as simple as possible.

* Thinking about implementing an almost-free exponential speedup, by
shifting by 1,2,4,.. in sequence.

* Simplify my half-finished parallel test case in
src/soc/experiment/test/test_compalu_multi.py, which is too detailed.
Just test for functionality, leave correctness to formal proofs.

* Investigate the possibility to generate useful, structured, commented,
color-formatted gtkw templates in test cases, using the vcd.gtkw module
from pyvcd (which is already a dependency of nMigen).

Cesar

_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] Fwd: [Mesa-dev] [XDC 2020] Virtual conference + Call for Proposals extended 2 weeks more
From: Luke Kenneth Casson Leighton @ 2020-07-20 11:29 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAPweEDzRh4u33wTvJZaKy8DFmoX3MZA-pCuXEL14t4cntRzBQg@mail.gmail.com>

https://xdc2020.x.org/event/9/registrations/7/

can i recommend registering, just to show some support for xdc?  there are
only 33 people registered online (there is no charge)

l.



-- 
---
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] div pipe
From: Luke Kenneth Casson Leighton @ 2020-07-20 10:31 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAC2bXD5YA2_+xH_=zPUV-tcdTjAvF-QJa=cYQPCvYA75BvToEA@mail.gmail.com>

On Monday, July 20, 2020, Jacob Lifshay <programmerjake@gmail.com> wrote:

>
> > > if False and core_config.supported == {DP.UDivRem}:
> > >     self.
> >
> >
> >
> > > As far as I can tell, that should work
>
> Pushed a fix.


nice


>
> Fixed the comb tests since the generate process was 1us ahead of the
> check process, causing test failures.


whoops :)



-- 
---
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] Fwd: [Mesa-dev] [XDC 2020] Virtual conference + Call for Proposals extended 2 weeks more
From: Luke Kenneth Casson Leighton @ 2020-07-20 10:30 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAPweEDzwLgGsK=13kzK_L26xgbr2Nh0aUCYZq1fgJ0ZKwLoQtA@mail.gmail.com>

https://libre-soc.org/conferences/xdc2020/?updated

done.

jacob would you like to put in something about Kazan? a 2nd talk, i think
it would be really important and also emphasising that Kazan's roadmap
includes being an architecturally independent software renderer (like
SwiftShader) but also includes a Shader Compiler.

regarding actually doing the talk i think it will be video submissions and
we can write everything out and read the words as the slides progress.  we
can work together to do a bit of editing and review and come up with
something.  i used kino, recently, it does the job.

l.



-- 
---
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] Fwd: [Mesa-dev] [XDC 2020] Virtual conference + Call for Proposals extended 2 weeks more
From: Luke Kenneth Casson Leighton @ 2020-07-20  9:38 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAC2bXD7tWoPCnctYsBqc2zg=PLX36GXps73tYZub=4aLSPVrrQ@mail.gmail.com>

good reminder, we could really do with putting something in.


On Monday, July 20, 2020, Jacob Lifshay <programmerjake@gmail.com> wrote:

> ---------- Forwarded message ---------
>
>
> >
> > We would like to announce as well that the new CFP deadline is Sunday
> > July 19th 2020. Don't forget to submit your talk, demo and workshop
> > proposals!
> >
>
> As approved in last board's meeting [0], CfP is extended until two weeks
> before the conference, or until we fill all the slots (whichever happens
> first). Please submit your talk proposals early!


ok so they need talks.


Last two years we had lots of talks about new and fresh development and
> we prioritized those over other kind of talks, as they had more
> potential for discussions and hallway track. However, this year that
> doesn't make too much sense, so we encourage our community to submit any
> talk related to open-source graphics stack, including those that focus
> on project status updates.
>
> Sam
>
> [0] https://www.x.org/wiki/BoardOfDirectors/MeetingSummaries/2020/07-16/
>
> > Thanks,
> >
> > Sam
> >
> >
> > _______________________________________________
> > mesa-dev mailing list
> > mesa-dev@lists.freedesktop.org
> > https://lists.freedesktop.org/mailman/listinfo/mesa-dev
> >
>
> _______________________________________________
> mesa-dev mailing list
> mesa-dev@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev
>


-- 
---
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* [libre-riscv-dev] Fwd: [Mesa-dev] [XDC 2020] Virtual conference + Call for Proposals extended 2 weeks more
From: Jacob Lifshay @ 2020-07-20  8:16 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <b7a10661-2e4d-4462-98f6-de13226eeccc@igalia.com>

[-- Attachment #1: Type: text/plain, Size: 2682 bytes --]

---------- Forwarded message ---------
From: Samuel Iglesias Gonsálvez <siglesias@igalia.com>
Date: Mon, Jul 20, 2020, 00:41
Subject: Re: [Mesa-dev] [XDC 2020] Virtual conference + Call for Proposals
extended 2 weeks more
To: Szwichtenberg, Radoslaw <radoslaw.szwichtenberg@intel.com>,
events@lists.x.org <events@lists.x.org>, xorg-devel@lists.freedesktop.org <
xorg-devel@lists.freedesktop.org>, wayland-devel@lists.freedesktop.org <
wayland-devel@lists.freedesktop.org>, dri-devel@lists.freedesktop.org <
dri-devel@lists.freedesktop.org>, mesa-dev@lists.freedesktop.org <
mesa-dev@lists.freedesktop.org>, amd-gfx@lists.freedesktop.org <
amd-gfx@lists.freedesktop.org>, etnaviv@lists.freedesktop.org <
etnaviv@lists.freedesktop.org>, freedreno@lists.freedesktop.org <
freedreno@lists.freedesktop.org>, nouveau@lists.freedesktop.org <
nouveau@lists.freedesktop.org>, intel-gfx@lists.freedesktop.org <
intel-gfx@lists.freedesktop.org>
Cc: board@foundation.x.org <board@foundation.x.org>




On 7/3/20 4:41 PM, Samuel Iglesias Gonsálvez wrote:
> Hi,
>
> In the last meeting, X.Org Foundation board has decided that XDC 2020
> will be a virtual conference, given the uncertain COVID-19 situation in
> Europe by September, including the possibility of a second wave,
> outbreaks and travel restrictions, either in Poland or in other
> countries.
>
> XDC 2020 organization team agrees on this decision and it volunteered
> to organize our first virtual XDC!
>
> We would like to announce as well that the new CFP deadline is Sunday
> July 19th 2020. Don't forget to submit your talk, demo and workshop
> proposals!
>

As approved in last board's meeting [0], CfP is extended until two weeks
before the conference, or until we fill all the slots (whichever happens
first). Please submit your talk proposals early!

Last two years we had lots of talks about new and fresh development and
we prioritized those over other kind of talks, as they had more
potential for discussions and hallway track. However, this year that
doesn't make too much sense, so we encourage our community to submit any
talk related to open-source graphics stack, including those that focus
on project status updates.

Sam

[0] https://www.x.org/wiki/BoardOfDirectors/MeetingSummaries/2020/07-16/

> Thanks,
>
> Sam
>
>
> _______________________________________________
> mesa-dev mailing list
> mesa-dev@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev
>

_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev

[-- Attachment #2: signature.asc --]
[-- Type: text/plain, Size: 849 bytes --]

-----BEGIN PGP SIGNATURE-----
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=UDoL
-----END PGP SIGNATURE-----

[-- Attachment #3: Type: text/plain, Size: 177 bytes --]

_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] div pipe
From: Jacob Lifshay @ 2020-07-20  3:21 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAPweEDyK8ms7jXbby1=JejzkjcKevb8ueDrs6naMipeR_119_A@mail.gmail.com>

On Sun, Jul 19, 2020 at 7:14 PM Luke Kenneth Casson Leighton
<lkcl@lkcl.net> wrote:
>
> On Monday, July 20, 2020, Jacob Lifshay <programmerjake@gmail.com> wrote:
>
> > l
> > and replace every occurrence with:
> >
> > # TODO(programmerjake): re-enable once bit_width reduction is fixed
>
>
> yeh that makes sense.
>
>
> > if False and core_config.supported == {DP.UDivRem}:
> >     self.
>
>
>
> > As far as I can tell, that should work

Pushed a fix.

Fixed the comb tests since the generate process was 1us ahead of the
check process, causing test failures.

Jacob

_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] div pipe
From: Luke Kenneth Casson Leighton @ 2020-07-20  2:13 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAC2bXD5ODi8di2yqitLQ40MSychRA-2WGg0+zH7U-wKCe5A2WA@mail.gmail.com>

On Monday, July 20, 2020, Jacob Lifshay <programmerjake@gmail.com> wrote:

> l
> and replace every occurrence with:
>
> # TODO(programmerjake): re-enable once bit_width reduction is fixed


yeh that makes sense.


> if False and core_config.supported == {DP.UDivRem}:
>     self.



> As far as I can tell, that should work


yeh should do.

l.




-- 
---
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] div pipe
From: Jacob Lifshay @ 2020-07-20  2:06 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAPweEDzzY9OXscMQSh1HM26AtM+43PJpCa0RS44y+MmErFr8sw@mail.gmail.com>

On Sun, Jul 19, 2020, 18:39 Luke Kenneth Casson Leighton <lkcl@lkcl.net>
wrote:

> On Monday, July 20, 2020, Jacob Lifshay <programmerjake@gmail.com> wrote:
>
> >
> > effectively what's happening is the code is treating the compare_?hs
> > signals as if the binary value N means the fraction N/2^(3*fract_width)
> > (essentially fixed-point numbers), so changing the bit width of N without
> > adjusting the fract_width means there are too few integer bits and too
> many
> > fractional bits.
>
>
> ahh i remember.  that took a while to get right when connecting to fpdiv.
>
>
> > additionally, some of the shifts and bit sliced will have
> > to be adjusted to get the right bits.
> >
> > >
> > > no reversion should be needed.
> > >
> >
> > except that, last I checked DivPipeCore doesn't correctly do divisions
> when
> > configured to only support UDivRem, since that's when it enables the
> broken
> > code adjustments.
>
>
> yuk.  hmm a way to "cheat" there: tell it that it can handle sqrt as well,
> temporarily :)
>

I think a better way would be to find all the code like:

if core_config.supported == {DP.UDivRem}:
    self.compare_len = bw * 2
else:
    self.compare_len = bw * 3

and replace every occurrence with:

# TODO(programmerjake): re-enable once bit_width reduction is fixed
if False and core_config.supported == {DP.UDivRem}:
    self.compare_len = bw * 2
else:
    self.compare_len = bw * 3

As far as I can tell, that should work.

Jacob
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] div pipe
From: Luke Kenneth Casson Leighton @ 2020-07-20  1:38 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAC2bXD4ipWkR=1Z2hX+=HvMbDwtinhvC7CWgG6GFEsu57j1-Eg@mail.gmail.com>

On Monday, July 20, 2020, Jacob Lifshay <programmerjake@gmail.com> wrote:

>
> effectively what's happening is the code is treating the compare_?hs
> signals as if the binary value N means the fraction N/2^(3*fract_width)
> (essentially fixed-point numbers), so changing the bit width of N without
> adjusting the fract_width means there are too few integer bits and too many
> fractional bits.


ahh i remember.  that took a while to get right when connecting to fpdiv.


> additionally, some of the shifts and bit sliced will have
> to be adjusted to get the right bits.
>
> >
> > no reversion should be needed.
> >
>
> except that, last I checked DivPipeCore doesn't correctly do divisions when
> configured to only support UDivRem, since that's when it enables the broken
> code adjustments.


yuk.  hmm a way to "cheat" there: tell it that it can handle sqrt as well,
temporarily :)

that at least provides a functional codepath.  inefficient but functional.

l.



-- 
---
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] div pipe
From: Jacob Lifshay @ 2020-07-20  0:32 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAPweEDz79NiDvrAzFQ9kgMdBu1gX9EgvWY2yLXBR=FAMxU0=Uw@mail.gmail.com>

On Sun, Jul 19, 2020, 15:07 Luke Kenneth Casson Leighton <lkcl@lkcl.net>
wrote:

> On Sunday, July 19, 2020, Jacob Lifshay <programmerjake@gmail.com> wrote:
>
> > What we could do is just revert
> > the bitwidth changes if we want to have working code anyway.
>
>
> well, what i did was, if the configuration (initialisation) requests sqrt
> or rsqrt the bitwidth is set to the 3x
>

yeah, that part's correct as far as I can tell.
the part that's broken is where the width is set to 2x for div-only, since
effectively what's happening is the code is treating the compare_?hs
signals as if the binary value N means the fraction N/2^(3*fract_width)
(essentially fixed-point numbers), so changing the bit width of N without
adjusting the fract_width means there are too few integer bits and too many
fractional bits. additionally, some of the shifts and bit sliced will have
to be adjusted to get the right bits.

>
> no reversion should be needed.
>

except that, last I checked DivPipeCore doesn't correctly do divisions when
configured to only support UDivRem, since that's when it enables the broken
code adjustments.

see the test_bit_width_8_fract_width_4_radix_4[_comb]_div_only test
functions in TestDivPipeCore -- the code only breaks when div-only is
enabled: compare to test_bit_width_8_fract_width_4_radix_4[_comb]

Jacob
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] daily kan-ban update 19jul2020
From: Luke Kenneth Casson Leighton @ 2020-07-19 23:19 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <AF38DB3D-DCBC-42D7-A75A-1AC767AA7FB1@gatech.edu>

On Sunday, July 19, 2020, Yehowshua <yimmanuel3@gatech.edu> wrote:

> >
> > in our case that's 17,000 flattened signals.
>
> Verilator retains the hierarchy of whatever it consumes.
> SO you’ll get a hierarchical CPU and flat peripherals.


thank goodness for that.

many of the pipeline signals have not been given globally unique names,
meaning there's dozens of duplicate "ready_o" etc. and more.

l.



-- 
---
crowd-funded eco-conscious hardware: https://www.crowdsupply.com/eoma68
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

* Re: [libre-riscv-dev] daily kan-ban update 19jul2020
From: Yehowshua @ 2020-07-19 22:46 UTC (permalink / raw)
  To: Libre-RISCV General Development
In-Reply-To: <CAPweEDyY=RG6WXDiVQNFz7Q+o0zNDDp5LdSR0hKPUVH-FE9Seg@mail.gmail.com>

> 
> in our case that's 17,000 flattened signals.


Well, oMigen won’t flatten out verilog black boxes, but it
will flatten out the dram controller for example.

Which hopefully you don’t need to debug…

Verilator retains the hierarchy of whatever it consumes.
SO you’ll get a hierarchical CPU and flat peripherals.


Yehowshua
_______________________________________________
libre-riscv-dev mailing list
libre-riscv-dev@lists.libre-riscv.org
http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev

^ permalink raw reply

page:   latest

Libre RISC-V Mailing list Archives (accessible via public-inbox)

Archives are clonable: git clone --mirror https://inbox.libre-soc.org/libre-riscv-dev

Example config snippet for mirrors


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git