* [libre-riscv-dev] [Bug 340] formal proof of POWER9 SHIFTROT pipeline needed
From: bugzilla-daemon @ 2020-05-22 20:31 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-340-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=340
Luke Kenneth Casson Leighton <lkcl@lkcl.net> changed:
What |Removed |Added
----------------------------------------------------------------------------
total budget (EUR)|0 |250
for completion of| |
task and all| |
subtasks| |
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
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] [Bug 195] Formal correctness framework is needed for Power ISA
From: bugzilla-daemon @ 2020-05-22 20:31 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-195-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=195
Luke Kenneth Casson Leighton <lkcl@lkcl.net> changed:
What |Removed |Added
----------------------------------------------------------------------------
budget (EUR) for|10700 |10450
completion of task| |
(excludes budget| |
allocated to| |
subtasks)| |
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
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] [Bug 324] create POWER9 DIV pipeline
From: bugzilla-daemon @ 2020-05-22 21:03 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-324-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=324
--- Comment #3 from Jacob Lifshay <programmerjake@gmail.com> ---
after looking through the spec some, it turns out that div and rem don't
actually trap on division by zero, so that makes them much easier to implement.
The numerical results are undefined on division-by-zero/overflow, so I will
need to run some tests on a Power processor to see which results we should
produce, hopefully they are the same as for RISC-V.
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
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] dns entries for talos server & server maintenance
From: Jacob Lifshay @ 2020-05-22 21:21 UTC (permalink / raw)
To: Libre-RISCV General Development
to make it easier to connect to without having to remember the IP
address, I think we should add a dns entry for the talos server, I'd
suggest talos1.libre-soc.org. As a reminder, the IP address is
23.155.224.75
Also, someone forgot to enable automatic updates, so I updated the
software manually. Sorry if that broke something.
The server needs to be rebooted to use the new kernel that got
installed, so, Luke, I'll let you do that since you have stuff running
on it.
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] dns entries for talos server & server maintenance
From: Luke Kenneth Casson Leighton @ 2020-05-22 22:54 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <CAC2bXD6+fOdpdaNJBL9O_TkMVMyQ8SrnqzJWbUMTMx+Ct-O=ew@mail.gmail.com>
On Fri, May 22, 2020 at 10:21 PM Jacob Lifshay <programmerjake@gmail.com> wrote:
>
> to make it easier to connect to without having to remember the IP
> address, I think we should add a dns entry for the talos server, I'd
> suggest talos1.libre-soc.org. As a reminder, the IP address is
> 23.155.224.75
good idea.
> Also, someone forgot to enable automatic updates,
no, it was not "forgotten" - the absolute last thing we need is
"automatically introduced problems", particularly if - especially as -
the setup is using debian/testing.
debian/stable and "production" servers with carefully pre-planned
setups, or "toys" where nobody else is relying on it, or "your own
personal machine" is absolutely fine to enable potentially-damaging
"automatic updates".
> so I updated the
> software manually. Sorry if that broke something.
yes please don't do that without consulting everyone, first.
to give you some idea: the notification schedule that servers used to
be on at Imperial College was 90 days advance warning of softtware
updates, and one year for a reboot.
> The server needs to be rebooted to use the new kernel that got
> installed,
which is precisely why you don't "upgrade" a server, especially
without notice or discussion. we are now critically dependent on that
machine having 100% uninterrupted power.
if there is a power-outage in the data centre that causes an
unscheduled reboot, resulting in that machine going down, and we are
critically reliant on it for testing at the time, someone has to drop
everything they're doing, scramble to look up the procedure for
logging in to the virtual console, and then work out how to recover
the machine.
> so, Luke, I'll let you do that since you have stuff running
> on it.
no, i'm not going to reboot it, because if it doesn't come back up
someone has to spend the time recovering it.
lesson learned: never "upgrade" servers without consulting people, ok?
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] [Bug 324] create POWER9 DIV pipeline
From: bugzilla-daemon @ 2020-05-22 22:58 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-324-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=324
--- Comment #4 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Jacob Lifshay from comment #3)
> after looking through the spec some, it turns out that div and rem don't
> actually trap on division by zero, so that makes them much easier to
> implement. The numerical results are undefined on division-by-zero/overflow,
> so I will need to run some tests on a Power processor to see which results
> we should produce, hopefully they are the same as for RISC-V.
good idea.
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
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] dns entries for talos server & server maintenance
From: Luke Kenneth Casson Leighton @ 2020-05-22 23:01 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <CAPweEDxxSdP8c7gDdWUidkwvH8diP1jYhqX==R5O7kG74N7CNw@mail.gmail.com>
just logged in: yes it's definitely running debian/testing: *really* -
don't ever do an "apt-get dist-upgrade" or any form of automated
upgrade on a debian/testing system, ever.
the only thing that should ever be done on a debian/testing system is
to see if an "apt-get install {package}" will try to replace, upgrade
or remove any packages, to TERMINATE that and CAREFULLY review the
list of package alterations.
given that this is a shared server that is also a critical resource,
that process involves consulting everyone who may be affected, ok?
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] dns entries for talos server & server maintenance
From: Luke Kenneth Casson Leighton @ 2020-05-22 23:06 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <CAPweEDzugy2fQmcyAwuQjj2meDoXktHwebp-q1gg1F6q+jkRjg@mail.gmail.com>
the good news is, you prompted me to log in and add michael's ssh key
to the machine. he also needs to run some tests.
it's ssh -p922 mtnolan@talos1.libre-soc.org, michael.
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] dns entries for talos server & server maintenance
From: Jacob Lifshay @ 2020-05-23 1:08 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <CAPweEDzugy2fQmcyAwuQjj2meDoXktHwebp-q1gg1F6q+jkRjg@mail.gmail.com>
On Fri, May 22, 2020 at 4:02 PM Luke Kenneth Casson Leighton
<lkcl@lkcl.net> wrote:
>
> just logged in: yes it's definitely running debian/testing: *really* -
> don't ever do an "apt-get dist-upgrade" or any form of automated
> upgrade on a debian/testing system, ever.
Ok, sorry. I saw stable in the apt lists and assumed it was on debian
stable. Also, the debian security source isn't working for some reason
(missing file? got a 404), so that source didn't work.
So, the list of changes is in /var/log/apt/term.log I think.
>
> the only thing that should ever be done on a debian/testing system is
> to see if an "apt-get install {package}" will try to replace, upgrade
> or remove any packages, to TERMINATE that and CAREFULLY review the
> list of package alterations.
>
> given that this is a shared server that is also a critical resource,
> that process involves consulting everyone who may be affected, ok?
Ok.
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] [Bug 314] Create POWER9 Condition Register pipeline
From: bugzilla-daemon @ 2020-05-23 1:17 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-314-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=314
Luke Kenneth Casson Leighton <lkcl@lkcl.net> changed:
What |Removed |Added
----------------------------------------------------------------------------
Summary|Create Condition Register |Create POWER9 Condition
|pipeline |Register pipeline
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
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] [Bug 314] Create POWER9 Condition Register pipeline
From: bugzilla-daemon @ 2020-05-23 1:21 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-314-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=314
--- Comment #16 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
hmmmm there are not many good choices for putting this instruction in. it
requires RA RC INT regs, and full CR's BI field as input.
for output it needs INT RS
CR is the closest least disruptive candidate, if a 2nd INT reg is added: RB.
when OP_ISEL =>
crbit := to_integer(unsigned(insn_bc(e_in.insn)));
if e_in.cr(31-crbit) = '1' then
result := a_in;
else
result := b_in;
end if;
result_en := '1';
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
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] [Bug 314] Create POWER9 Condition Register pipeline
From: bugzilla-daemon @ 2020-05-23 1:25 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-314-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=314
--- Comment #17 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
https://git.libre-soc.org/?p=soc.git;a=blob;f=src/soc/fu/cr/pipe_data.py;h=2894144fd814444a7bdb75ae6c695af8519ae084;hb=HEAD
can add b to CRInputData.
https://libre-soc.org/openpower/isa/fixedtrap/
what on earth isel is doing in the trap instruction group is beyond me. could
be i just put the pseudocode in the wrong file
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
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] [Bug 313] Create Branch Pipeline for POWER9
From: bugzilla-daemon @ 2020-05-23 1:29 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-313-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=313
--- Comment #39 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
https://git.libre-soc.org/?p=soc.git;a=blob;f=src/soc/fu/branch/pipe_data.py;h=28099eaa84f065065132208abaa9d8fa4a6f003f;hb=HEAD
whoops the regspec for cr needs to be 0:3 range. 4 means "select bit 4 only".
ie only a 1 bit wide CR.
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
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 22may2020
From: Cesar Strauss @ 2020-05-23 1:56 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <3C8C9F70-B1BB-4A9C-8EEB-14EA91B6E0C4@getmailspring.com>
Hi Cole and all,
On 05/22/2020 16:10, Cole Poirier wrote:
> Would you mind giving a one line description here of what Cesar does?
> I'd be interested to know as well.
I work in a research institute, on data acquisition and control of
scientific instruments for astronomy.
Let me expand a bit on that.
Some of the projects I'm involved, past and present:
1) Microwave receivers for a 14 m radio telescope.
2) Control of a 26-element radio interferometer array
3) Image acquisition for an infrared camera
4) Data acquisition for a X-ray detector in a stratospheric balloon.
5) Miniaturized version of the above, to fly on a nano-satellite (2U
cubesat)
6) Spherical-mass gravitational wave detector
Some phenomena that the scientists observe with the above:
1) Molecular clouds and star formation
2) Variability in the active nuclei of galaxies (quasars)
3) Accretion disks around black holes
4) Gamma ray burst
5) Solar flares and coronal mass ejections
6) Space weather
Some technologies I use:
1) Data acquisition and remote control of voltmeters and spectrum
analyzers via RS-232, GPIB, Ethernet and USB
2) Multi-channel analog data acquisition boards
3) Microcontrollers like 8051, PIC, ARM Cortex, Arduinos
4) FPGA for glue logic, data acquisition and signal processing
5) Single-board computer
6) Cryogenics for low noise temperature radio frequency receivers
Software tools:
1) GCC, C/C++ compiler for the PC and some microcontrollers
2) SDCC, C compiler for smaller microcontrollers
3) Icarus Verilog for FPGA simulation
4) gEDA/gaf, Pcb and Gerbv for PCB design
5) Python, for quick prototyping, data processing, plotting.
6) R, Octave, for data processing
My motivation to participate in Libre-SOC:
A libre, supported, fully documented SoC, free of proprietary drivers,
firmware and bootloaders, that I can trust to use in my workstation,
server and as building blocks of scientific instruments.
Regards,
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] daily kan-ban update 22may2020
From: Luke Kenneth Casson Leighton @ 2020-05-23 2:41 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <a1e0dff9-477f-59a0-d5da-6b0b97d75e16@gmail.com>
On Saturday, May 23, 2020, Cesar Strauss <cestrauss@gmail.com> wrote:
>
> My motivation to participate in Libre-SOC:
>
> A libre, supported, fully documented SoC, free of proprietary drivers,
> firmware and bootloaders, that I can trust to use in my workstation,
> server and as building blocks of scientific instruments.
very cool.
and even in FPGA, one that had good number crunching capability might be
useful, i assume.
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] dns entries for talos server & server maintenance
From: Luke Kenneth Casson Leighton @ 2020-05-23 2:46 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <CAC2bXD6KYBJeMEX+R+B76zH=Tp1VyOdUv7d4hPZR7Mon=c9eKg@mail.gmail.com>
On Saturday, May 23, 2020, Jacob Lifshay <programmerjake@gmail.com> wrote:
> On Fri, May 22, 2020 at 4:02 PM Luke Kenneth Casson Leighton
> <lkcl@lkcl.net> wrote:
> >
> > just logged in: yes it's definitely running debian/testing: *really* -
> > don't ever do an "apt-get dist-upgrade" or any form of automated
> > upgrade on a debian/testing system, ever.
>
> Ok, sorry. I saw stable in the apt lists and assumed it was on debian
> stable.
the list is amalgamated, including from /etc/apt/sources.list.d so it is
necessary to really really pay attention and read every line.
the way debian works is that unless apt pinning is active, the highest
lexicographically named version of a package is picked.
so with stable *and* testing in the list, and no apt pin preferences file,
testing gets picked... unless a testing package has not been built.
>
> Also, the debian security source isn't working for some reason
> (missing file? got a 404), so that source didn't work.
either it doesn't actually exist or it is spelled wrong.
>
> So, the list of changes is in /var/log/apt/term.log I think.
> >
> > the only thing that should ever be done on a debian/testing system is
> > to see if an "apt-get install {package}" will try to replace, upgrade
> > or remove any packages, to TERMINATE that and CAREFULLY review the
> > list of package alterations.
> >
> > given that this is a shared server that is also a critical resource,
> > that process involves consulting everyone who may be affected, ok?
>
> Ok.
>
> Jacob
>
> _______________________________________________
> libre-riscv-dev mailing list
> libre-riscv-dev@lists.libre-riscv.org
> http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-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] [Bug 313] Create Branch Pipeline for POWER9
From: bugzilla-daemon @ 2020-05-23 3:06 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-313-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=313
--- Comment #40 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
looking at putting CR_ISEL into CR pipe, it also uses BC so i checked
DecodeCRin and it puts all 5 bits of BC into the reg bitfield.
that seems to be an error, i think it should be BC[2:5] like the others
i grepped BC in the csv files and at present only BR uses BC in the CRin field.
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
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 22may2020
From: Cesar Strauss @ 2020-05-23 10:24 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <CAPweEDx_YjWZRzNb4qN7m4TundvyWAKTWhzPW1Eff4ph4kYOhw@mail.gmail.com>
On 05/22/2020 23:41, Luke Kenneth Casson Leighton wrote:
> On Saturday, May 23, 2020, Cesar Strauss <cestrauss@gmail.com> wrote:
>
>>
>> My motivation to participate in Libre-SOC:
>>
>> A libre, supported, fully documented SoC, free of proprietary drivers,
>> firmware and bootloaders, that I can trust to use in my workstation,
>> server and as building blocks of scientific instruments.
>
>
> very cool.
>
> and even in FPGA, one that had good number crunching capability might be
> useful, i assume.
>
If you mean a libre FPGA, I am all for it.
A libre, supported, fully documented FPGA, with open architecture and
bitstream format.
If some people (hint!) were to start a libre-FPGA project, it would be
absolutely wonderful.
Independent of having a number crunching capability or not.
For synthesis, placing and routing, we do have libre tools. But, to
actually program the Configurable Logic Blocks (CLB), the routing switch
matrices, etc., inside the FPGA, we are forced to either use the
proprietary tools, or use libre tools based on a reverse engineered
bitstream format. It's a complete lock-in.
The community did produce a completely libre flow, from HDL to
bitstream, for Lattice chips. But the bitstream format is still
proprietary, and had to be reverse engineered. It is not assured that we
always will be able to do so, as new families are introduced. It is a
waste of effort, and features will be lacking.
I think there are experimental FPGAs developed in academia, but not
reaching commercial availability.
Cheers,
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
* [libre-riscv-dev] [Bug 314] Create POWER9 Condition Register pipeline
From: bugzilla-daemon @ 2020-05-23 10:36 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-314-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=314
--- Comment #18 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
i'm so excited, i'm so happy, i added an instruction (isel).
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
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] [Bug 332] Formal correctness proof needed for CR pipeline
From: bugzilla-daemon @ 2020-05-23 10:59 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-332-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=332
--- Comment #4 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
michael could you check this when you have a moment?
commit 27440addbb83c3c13949246b495b8c474b1c2c90 (HEAD -> master, origin/master)
Author: Luke Kenneth Casson Leighton <lkcl@lkcl.net>
Date: Sat May 23 11:58:43 2020 +0100
add CR_ISEL formal proof to CR pipeline
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
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 22may2020
From: Luke Kenneth Casson Leighton @ 2020-05-23 11:03 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <5c96dbc3-3893-717a-6bea-5f2a9f8c3c5c@gmail.com>
On Sat, May 23, 2020 at 11:24 AM Cesar Strauss <cestrauss@gmail.com> wrote:
>
> On 05/22/2020 23:41, Luke Kenneth Casson Leighton wrote:
> > and even in FPGA, one that had good number crunching capability might be
> > useful, i assume.
> >
>
> If you mean a libre FPGA, I am all for it.
:)
> I think there are experimental FPGAs developed in academia, but not
> reaching commercial availability.
yes we found one a few months ago, written in chisel (!) and
considered putting it into silicon. however it is a big project in
its own right, you'd have to go through the full business case
justification and properly commit to it.
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 22may2020
From: Cesar Strauss @ 2020-05-23 11:53 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <CAPweEDx_YjWZRzNb4qN7m4TundvyWAKTWhzPW1Eff4ph4kYOhw@mail.gmail.com>
Em 22/05/2020 23:41, Luke Kenneth Casson Leighton escreveu:
> On Saturday, May 23, 2020, Cesar Strauss <cestrauss@gmail.com> wrote:
>
>>
>> My motivation to participate in Libre-SOC:
>>
>> A libre, supported, fully documented SoC, free of proprietary drivers,
>> firmware and bootloaders, that I can trust to use in my workstation,
>> server and as building blocks of scientific instruments.
>
>
> very cool.
>
> and even in FPGA, one that had good number crunching capability might be
> useful, i assume.
>
If you mean, programming an FPGA board with a libre-SOC core, it would
have to be seriously optimized for size, not speed. These big FPGAs
boards are very expensive, compared to an equivalent Single Board Computer.
There is no way a relatively simple DSP algorithm, implemented on the
raw resources of an FPGA, could be beaten by a CPU/GPU core executing
this same algorithm, in the same FPGA.
Yes, people put CPU cores in FPGAs, but mainly to interface to the
outside world, providing a convenient interface to the FPGA. The main
data processing is still done in the free resources which remains after
placing the CPU core.
Of course, I could use parts of libre-SOC, like the floating points
pipelines. But, fixed-point algorithms are more common than
floating-point in FPGA applications.
Regards,
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] daily kan-ban update 22may2020
From: Luke Kenneth Casson Leighton @ 2020-05-23 12:08 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <6a1e761a-6e41-9431-647d-5e374b4d4134@gmail.com>
On Saturday, May 23, 2020, Cesar Strauss <cestrauss@gmail.com> wrote:
>
> Of course, I could use parts of libre-SOC, like the floating points
> pipelines. But, fixed-point algorithms are more common than
> floating-point in FPGA applications.
one of the big advantages of an OoO design is the opportunity for parallel
processing interwoven cleanly behind a sequential "API" (the instruction
set).
so let us suppose that you needed for example a significant number of fixed
point correlators, done in a specialist hardware-optimised mathematical way
that is massively inconvenient even for a DSP or a standard CPU.
however a full FPGA implementation is also inconvenient because the level
of parallelism required, when interacting closely with a sequential
program, makes the FPGA correlator a huge pain to program.
what you need would therefore be a CPU with sequential behaviour that could
just call a sequential instruction in a loop, to activate the special
custom correlator, but because that CPU is an out of order parallel
processor multiple such custom instructions (custom correlators) get issued
in parallel and the dependencies automatically resolved for you.
here you would get the best of both worlds.
another thought, for when we have time: if you can clearly express the
primitives needed, it would be a fascinating use-case evaluation to see
whether the ASIC would be any good, for the kind of DSP work that you do,
and if not, what operations would actually be needed.
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
* [libre-riscv-dev] [Bug 336] ALU CompUnit needs to recognise that RA (src1) can be zero
From: bugzilla-daemon @ 2020-05-23 12:51 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-336-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=336
--- Comment #5 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
Cesar, hi,
just to let you know, this should *not* interfere with adding is_zero support.
it however will be necessary to use this:
if hasattr(self.oper_i, "zero_a"):
# ok now we know that this operation requires zero_a support, we can
# mux in zero_a. otherwise we must put register a straight through
i am doing a minor code-morph on MultiCompUnit which makes it suitable for
general use across multiple pipelines, as specified by "regspecs". an example
is below.
note that this example - which uses CompCROpSubset **NOT** CompALUOpSubset -
is why we will need the "detection" above.
please do not let this stop you from adding zero_a detection support, let us
keep in touch and coordinate here.
i will have to add "detect if there is an operand B immediate" in the same way,
so you can see how it is done.
class CompCROpSubset(Record):
def __init__(self, name=None):
layout = (('insn_type', InternalOp),
('fn_unit', Function),
('insn', 32),
('read_cr_whole', 1),
('write_cr_whole', 1),
)
class CRInputData(IntegerData):
regspec = [('INT', 'a', '0:63'), # 64 bit range
('INT', 'b', '0:63'), # 6B bit range
('CR', 'full_cr', '0:31'), # 32 bit range
('CR', 'cr_a', '0:3'), # 4 bit range
('CR', 'cr_b', '0:3'), # 4 bit range
('CR', 'cr_c', '0:3')] # 4 bit range
class CROutputData(IntegerData):
regspec = [('INT', 'o', '0:63'), # 64 bit range
('CR', 'full_cr', '0:31'), # 32 bit range
('CR', 'cr_o', '0:3')] # 4 bit range
...
class CRPipeSpec(CommonPipeSpec):
regspec = (CRInputData.regspec, CROutputData.regspec)
opsubsetkls = CompCROpSubset
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
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] [Bug 336] ALU CompUnit needs to recognise that RA (src1) can be zero
From: bugzilla-daemon @ 2020-05-23 13:07 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-336-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=336
--- Comment #6 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
ok done for imm_data, here is the location, it is virtually a cookie-cut of
what
is done for imm_data, bear in mind it will need to be src_l.q[0] and because
src_l.q[0] covers RA, where src_l.q[1] covers RB. likewise the src1_or_imm
needs to go into sl[0] not sl[1]
```
# if the operand subset has "zero_a" we implicitly assume that means
# src_i[0] is an INT register type where zero can be multiplexed in,
instead.
# see https://bugs.libre-soc.org/show_bug.cgi?id=336
#if hasattr(oper_r, "zero_a"):
# select zero immediate if opcode says so. however also change the
latch
# to trigger *from* the opcode latch instead.
# ...
# ...
# if the operand subset has "imm_data" we implicitly assume that means
# "this is an INT ALU/Logical FU jobbie, RB is multiplexed with the
immediate"
if hasattr(oper_r, "imm_data"):
# select immediate if opcode says so. however also change the
latch
# to trigger *from* the opcode latch instead.
op_is_imm = oper_r.imm_data.imm_ok
```
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________
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: next prev 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