* [libre-riscv-dev] [Bug 313] Create Branch Pipeline for POWER9
From: bugzilla-daemon @ 2020-06-02 20:30 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 #47 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Michael Nolan from comment #46)
> (In reply to Luke Kenneth Casson Leighton from comment #44)
>
> > bear in mind that DecodeA reads CTR, (which goes into read_fast1)
> > and DecodeB reads LR or TAR (which goes into read_fast2).
> >
> > so the above test needs to be split.
>
> So main_stage.py will need to check whether the instruction is a bcctr and
> use spr1, otherwise use spr2?
sigh yes. although what i was going to do was to just make BCREG use spr2
for CTR/LR/TAR and BC uses spr1=CTR.
i believe that may end up being simpler, because main_stage.py would not
need to decode XO[5/9]?
--
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 @ 2020-06-02 20:19 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 #46 from Michael Nolan <mtnolan2640@gmail.com> ---
(In reply to Luke Kenneth Casson Leighton from comment #44)
> bear in mind that DecodeA reads CTR, (which goes into read_fast1)
> and DecodeB reads LR or TAR (which goes into read_fast2).
>
> so the above test needs to be split.
So main_stage.py will need to check whether the instruction is a bcctr and use
spr1, otherwise use spr2?
--
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 @ 2020-06-02 20:11 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 #45 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
it's ok i got it, i appreciate you're dropping in on occasion at the moment.
--
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 @ 2020-06-02 19:09 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 #44 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Michael Nolan from comment #43)
> (In reply to Luke Kenneth Casson Leighton from comment #42)
> > michael i saw the change in fu/branch/test_pipe_caller.py to use spr2.
> > it now needs to be read_fast1.data and read_fast2.data,
> > not read_spr1.data and read_spr2.data.
>
> How do you think TAR should be handled? Should it be a fast SPR as well?
yes. it's accessed often enough.
if you can hack that into DecodeA and DecodeB, such that this is in
power_decoder2.py
rather than in the unit test, it would be great:
insn_type = yield dec2.e.insn_type
if insn_type == InternalOp.OP_BCREG.value:
xo9 = yield dec2.dec.FormXL.XO[9]
xo5 = yield dec2.dec.FormXL.XO[5]
if xo9 == 0:
yield branch.p.data_i.spr2.eq(sim.spr['LR'].value)
elif xo5 == 1:
yield branch.p.data_i.spr2.eq(sim.spr['TAR'].value)
else:
yield branch.p.data_i.spr2.eq(sim.spr['CTR'].value)
then, read_fast1 and read_fast2 can be used in both unit tests:
fu/compunit/test/test_branch_compunit.py
and
fu/branch/test/test_pipe_caller.py
if you can handle the modifications to power_decoder2.py that would be
great.
bear in mind that DecodeA reads CTR, (which goes into read_fast1)
and DecodeB reads LR or TAR (which goes into read_fast2).
so the above test needs to be split.
--
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 @ 2020-06-02 18:43 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 #43 from Michael Nolan <mtnolan2640@gmail.com> ---
(In reply to Luke Kenneth Casson Leighton from comment #42)
> michael i saw the change in fu/branch/test_pipe_caller.py to use spr2.
> it now needs to be read_fast1.data and read_fast2.data,
> not read_spr1.data and read_spr2.data.
How do you think TAR should be handled? Should it be a fast SPR as well?
--
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 @ 2020-06-02 18:27 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 #42 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
michael i saw the change in fu/branch/test_pipe_caller.py to use spr2.
it now needs to be read_fast1.data and read_fast2.data,
not read_spr1.data and read_spr2.data.
--
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 216] LOAD STORE buffer needed
From: bugzilla-daemon @ 2020-06-02 18:26 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-216-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=216
--- Comment #54 from Tobias Platen <libre-soc@platen-software.de> ---
Before running formal verification, I first had to upgrade nmigen, because my
code works using records. When using an old version of nmigen, I got the
following error.
> Object (rec <unnamed> data en) cannot be used as a key in value collections
The second one is a bug in nmigen when it invokes sby with an empty working
directory, which is the current directory in this case. I set it to "." with
the following lines added to the code.
+ if(len(spec_dir)==0):
+ spec_dir="."
--
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 325] create POWER9 TRAP pipeline
From: bugzilla-daemon @ 2020-06-02 17:54 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-325-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=325
--- Comment #17 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
Cole: example (A is to B, as C is to D). examine case OP_TRAP and case OP_SC.
A:
ctrl_tmp.irq_nia <= std_logic_vector(to_unsigned(16#C00#, 64));
ctrl_tmp.srr1 <= msr_copy(ctrl.msr);
B:
????
C:
ctrl_tmp.irq_nia <= std_logic_vector(to_unsigned(16#700#,
64));
ctrl_tmp.srr1 <= msr_copy(ctrl.msr);
D:
comb += self.o.nia.data.eq(0x700) # trap address
comb += self.o.nia.ok.eq(1)
it should be BLINDINGLY obvious that B should be:
comb += self.o.nia.data.eq(0xC00) # trap address
comb += self.o.nia.ok.eq(1)
therefore, in switch statement OP_SC, that's what goes into that function.
and that's it.
ta-daaa. OP_SC is completed.
wasn't that easy?
Q: do you understand or know what OP_SC is or does?
A: i don't... and don't care, and it is completely irrelevant to you, as well.
(it's *optional* to have understanding: you may *desire* understanding,
you may *like* to have understanding, but it is, fundamentally, *completely
irrelevant* to the actual task)
Q: do we *need* to understand or know what OP_SC is or does?
A: no we do not.
Q: will it work?
A: yes it will.
Q: will there be bugs?
A: most probably, and those can be found with unit tests.
Q: do absolutely all the required unit tests have to be written *right now*?
A: no they do not.
Q: do we need to freeze and lock up solid in total fear at our total and
complete lack of understanding just because those unit tests do not exist?
A: of course not.
so reading a 1300 page PDF is completely and utterly pointless. and making
reading that 1300 page PDF a hard, fixed, absolute critical dependency on
*completing this task* is a false assumption, isn't it?
--
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 02jun2020
From: Luke Kenneth Casson Leighton @ 2020-06-02 16:04 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <20200602174333.eb7a589872b96a9acdc787e3@platen-software.de>
On Tue, Jun 2, 2020 at 4:43 PM Tobias Platen
<libre-soc@platen-software.de> wrote:
> > > > Object (rec <unnamed> data en) cannot be used as a key in value collections
> > > This is now solved by upgrading nmigen.
> I also found another bug in nmigen, that I will report soon. Just a two line fix.
could you make sure to put links to both into #216, because otherwise
people will try to run the code, not realising that there's bugs, and
raise another bugreport?
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 02jun2020
From: Tobias Platen @ 2020-06-02 15:43 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <CAPweEDwiyUaaGex2fkw7wy8f7_nNm8yLaZNXW_DfasPxFDn2LA@mail.gmail.com>
On Tue, 2 Jun 2020 16:37:33 +0100
Luke Kenneth Casson Leighton <lkcl@lkcl.net> wrote:
> On Tue, Jun 2, 2020 at 4:03 PM Tobias Platen
> <libre-soc@platen-software.de> wrote:
> >
> > On Tue, 2 Jun 2020 16:44:45 +0200
> > Tobias Platen <libre-soc@platen-software.de> wrote:
> >
> > > today: more work on formal verification
> > >
> > > currently I am trying to find out how to solve the following thing, I think it is a bug in nmigen.
> > > Object (rec <unnamed> data en) cannot be used as a key in value collections
> > This is now solved by upgrading nmigen.
I also found another bug in nmigen, that I will report soon. Just a two line fix.
>
> ah excellent!
>
> _______________________________________________
> libre-riscv-dev mailing list
> libre-riscv-dev@lists.libre-riscv.org
> http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev
--
Tobias Platen <libre-soc[at]platen-software[dot]de>
_______________________________________________
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 02jun2020
From: Luke Kenneth Casson Leighton @ 2020-06-02 15:37 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <20200602170318.548577a2d143a59f9bd0c609@platen-software.de>
On Tue, Jun 2, 2020 at 4:03 PM Tobias Platen
<libre-soc@platen-software.de> wrote:
>
> On Tue, 2 Jun 2020 16:44:45 +0200
> Tobias Platen <libre-soc@platen-software.de> wrote:
>
> > today: more work on formal verification
> >
> > currently I am trying to find out how to solve the following thing, I think it is a bug in nmigen.
> > Object (rec <unnamed> data en) cannot be used as a key in value collections
> This is now solved by upgrading nmigen.
ah excellent!
_______________________________________________
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 02jun2020
From: Tobias Platen @ 2020-06-02 15:03 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <20200602164445.a04e701ea5d94cf1cb4e37f7@platen-software.de>
On Tue, 2 Jun 2020 16:44:45 +0200
Tobias Platen <libre-soc@platen-software.de> wrote:
> today: more work on formal verification
>
> currently I am trying to find out how to solve the following thing, I think it is a bug in nmigen.
> Object (rec <unnamed> data en) cannot be used as a key in value collections
This is now solved by upgrading nmigen.
>
> _______________________________________________
> libre-riscv-dev mailing list
> libre-riscv-dev@lists.libre-riscv.org
> http://lists.libre-riscv.org/mailman/listinfo/libre-riscv-dev
--
Tobias Platen <libre-soc[at]platen-software[dot]de>
_______________________________________________
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 02jun2020
From: Tobias Platen @ 2020-06-02 14:44 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <CAPweEDy-nKkUYj-2uikQpg54GGbZW+DOhy-wcL1jBdkt3XM+xw@mail.gmail.com>
today: more work on formal verification
currently I am trying to find out how to solve the following thing, I think it is a bug in nmigen.
Object (rec <unnamed> data en) cannot be used as a key in value collections
_______________________________________________
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 325] create POWER9 TRAP pipeline
From: bugzilla-daemon @ 2020-06-02 14:06 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-325-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=325
--- Comment #16 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
TODO: add OP_SC line but not exactly like it is done in microwatt:
it has to go in extra.csv so as to be able to detect (and exclude) scv.
this through bit 1 (actually bit 30 sigh) "if e_in.insn(1) = '1' then"
17 => (ALU, OP_SC, NONE, NONE, NONE, NONE,
'0', '0', '0', '0', ZERO, '0', NONE, '0', '0', '0', '0', '0', '0', NONE, '0',
'0'), -- sc
so this would be something like:
17 in binary ------ bit 30 here
010001------------------------1-,TRAP,OP_SC,....,sc,NONE
--
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 353] formal proof of soc.regfile classes RegFile and RegFileArray needed
From: bugzilla-daemon @ 2020-06-02 12:27 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-353-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=353
Luke Kenneth Casson Leighton <lkcl@lkcl.net> changed:
What |Removed |Added
----------------------------------------------------------------------------
NLnet milestone|--- |NLNet.2019.10.Formal
--
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] microwatt decoder tables: M-Form and X-Form switched RS and RB
From: Luke Kenneth Casson Leighton @ 2020-06-02 12:10 UTC (permalink / raw)
To: Libre-RISCV General Development; +Cc: openpower-hdl-cores
In-Reply-To: <CAPweEDxVDR0HgVB-doJm0gyHf0RwX3hLCutjzZZ84b5k06wAfA@mail.gmail.com>
ok. interesting. i start to see how POWER's design hangs together,
from a RISC / architectural perspective. RA, RB, RC, RT, RS - these
would have been divided into "lanes" - effectively register file and
inter-Function-Unit "broadcast" buses: 5 of them.
RA, RB, RC and RS were probably also connected to 4 read ports (on a
4R1W regfile), and i suspect that RA and RS were multiplexed onto the
same regfile write port (with stall signals or careful scheduling).
those broadcast bus lanes would have been shared between:
* read regfile ports
* write regfile ports
* Function Unit incoming and outgoing ports.
just as for LD/ST, RS/RB would have been "laned" so that the
part-result could be "broadcast" onto those buses (without needing
MUXes to do so), such that the Shift Function Unit can pick it up as
if the information had come from the register file ports, OP_EXTS was
also put onto "lane RC" because it can then be likewise used as a
micro-code block in exactly the same way.
likewise, the OR, AND and XOR operations, to allow direct inter-FU
communication without any kind of lane cross-over.
it's really elegant, because it basically encodes "operand forwarding"
into the actual architecture.
however the _normal_ focus of RISC is on the instruction encoding. in
the CDC 6600, there's 3 bits which are quite literally hard-wired to
enable a given Function Unit (of which there are, as might be obvious
from the 3-bits used: 8 types). a few extra gates - again directly
from the opcode - says whether the instruction is 1-operand or
2-operands.
i believe in the CDC 6600 it's something ridiculous like under 15
gates for the entire instruction decoder. now *that's* R(educed)
I(nstruction) S(et) C(oding)!
this therefore leaves me wondering if something similar was done for
POWER, where we've simply missed these patterns because they're just
not documented.
example: we know that ADD, NEG, SUB and even CMP can all use the exact
same underlying add hardware, with A-inversion, Out-inversion,
carry-in and carry-out detection engaged, optionally, to "augment" the
operation and turn what would otherwise be a plain add into a NEG, or
a SUB, and so on.
however... is this fact encoded *in the instructions*? i.e. is the
instruction set encoded such that:
* "bit 8 indicates invert A"
* "bit 9 indicates invert output"
* "pattern of bits in positions X thru Y indicate an *underlying*
generic ADD hardware is to be used"
Rc and OE definitely fall into this category, however my question is:
does this pattern go much deeper?
if this principle has been applied across the board, in the ISA
design, the POWER instruction decoder phase could be drastically
simplified. right now, the decoder is a massive 4 deep nested series
of Switch/Case statements, and consequently the number of gates
involved is enormous compared to other RISC architectures.
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 325] create POWER9 TRAP pipeline
From: bugzilla-daemon @ 2020-06-02 11:21 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-325-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=325
--- Comment #15 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
https://git.libre-soc.org/?p=soc.git;a=commitdiff;h=2955619bc39a83b09920f8452882c4c327d30fef
cole, this is important to understand: you *do not* need to "understand"
what is "going on". you do not need to read 1,300 page PDFs - in full -
memorised in their entirety - before being able to take action.
this can be extremely uncomfortable. it feels... "unnatural", like,
"i should know everything because if i don't i might make a mistake or
implement the wrong thing".
sound familiar?
please understand: "understanding" != "action".
the trick that i learned - a long time ago, from doing reverse-engineering -
is to find someone else's *working* implementation, and *without*
understanding,
perform as direct a one-to-one translation as is humanly and conveniently
possible.
forget optimisation.
forget "understanding".
forget even *bug* fixes.
just... get... it... translated.
through that process, it turns out that by walking over the code (which you
quotes don't understand quotes), this puts that code into your brain.
**THEN** when you read the spec - the 1300 page PDF - you go, "ohhhh, yeahhh, i
recognise that. that's those bits/flags/functions/insert-whatever that i saw
when i was reading and translating the code, line-by-line".
and *NOW* you have quotes understanding quotes.
without having taken that first step, i guarantee to you that you will get
nowhere, fast.
so just - and i mean this literally - translate the comments that i've put
from microwatt execute1.vhdl into nmigen. you can see how Michael and i
did that already for both OP_TRAP and OP_MFMSR.
i deliberately inserted the original microwatt code even for those, so that
you have a direct side-by-side comparison, and can follow the adage
"A is to B, as C is to D".
you'll also need to translate the bit-patterns MSR_* - this again should be
obvious and easy.
--
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] daily kan-ban update 02jun2020
From: Luke Kenneth Casson Leighton @ 2020-06-02 10:48 UTC (permalink / raw)
To: Libre-RISCV General Development
yesterday:
* got Branch CompUnit test on par with its equivalent test_pipe_caller.py
* swapped RS from position "3" to position "1" (read port 1) on all Logical ops,
OP_EXTS, and MTCRF
* started on ShiftRot CompUnit test. fixed bugs, discovered that sraw is
not setting CA/CA32, discovered that "rlwinm." is not implemented.
* fixed a bug in the CompUnit testing where the source operands were not
zero'd. ALUs picked up and used *old* operands (from a previous instruction)
* fixed two mask-related bugs in MultiCompUnit
* started on the linking of CompUnits to regfiles, through understanding of
the relationship between CompUnits, regspecs, and Decode2Execute1Type.
today:
* getting Decode2Execute1Type to "understand" regspecs: to provide the
"accessor" register information in the RegSpec API, so that it becomes
dead simple to connect Decoder, CompUnits, and Regfiles.
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 01jun2020
From: Luke Kenneth Casson Leighton @ 2020-06-02 10:18 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <CAC2bXD50ZtWsE_Yo8=KRTW_0qSEhZrca7uKqXHBUOStqb75Xew@mail.gmail.com>
On Tue, Jun 2, 2020 at 7:18 AM Jacob Lifshay <programmerjake@gmail.com> wrote:
>
> On Mon, Jun 1, 2020, 03:41 Luke Kenneth Casson Leighton <lkcl@lkcl.net>
> wrote:
>
> > * MUL and DIV pipelines (Jacob)
> >
>
> Working on that.
good. do bear in mind it's beginning to hold up development. i'm
working round that.
as i explained to Cole: there are two ways to do development:
1) do everything yourself. understand everything. do everything from
first principles. read every available piece of documentation.
2) find an existing implementation and translate it.
the first way takes months. we do not have months.
the second way takes days to weeks.
so if you can pick up the pace significantly by working from microwatt
source code, and translating it to nmigen, that would be good.
> Had some ISP-level internet issues yesterday, that's why
> my build server went offline for a while in case anyone was wondering.
hmm that's interesting. i cannot connect to gmail except by switching
off the HTTP proxy. it would seem that this is a global world-wide
issue.
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 01jun2020
From: Luke Kenneth Casson Leighton @ 2020-06-02 10:11 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <7E5444F2-118E-4C7F-AD1A-BAD020083784@getmailspring.com>
On Tue, Jun 2, 2020 at 1:23 AM Cole Poirier <colepoirier@gmail.com> wrote:
>
> Today (actual):
>
> Spent today reading the parts of OPENPOWER ISA v3.1 related to traps,
> not finished yet. I don't understand very much currently, but I think
> that this will progress as I read and investigate further, however,
> since I don't think I can read the spec well enough to understand what I
> have to implement for fu/trap/main_stage, if someone else has the time I
> think they should do it. Otherwise, I'll continue to work away at
> understanding enough of the spec to be able to implement the trap pipeline.
it is not in the least bit necessary to "understand" the spec. i do not.
what *is* necessary is to understand how to translate microwatt VHDL
source code into nmigen.
and to TRUST that microwatt is correct.
this is why i have put in code-snippets from microwatt into the bugreport.
this turns an exercise that would take months if not years by a factor
of approximately TEN.
quotes understanding quotes comes later.
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 01jun2020
From: Jacob Lifshay @ 2020-06-02 6:17 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <CAPweEDxaWfprCC=Aj61mpBdDS4QLyRRm+HzeHJaecJSaWDc=yQ@mail.gmail.com>
On Mon, Jun 1, 2020, 03:41 Luke Kenneth Casson Leighton <lkcl@lkcl.net>
wrote:
> * MUL and DIV pipelines (Jacob)
>
Working on that. Had some ISP-level internet issues yesterday, that's why
my build server went offline for a while in case anyone was wondering.
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 01jun2020
From: Cole Poirier @ 2020-06-02 0:22 UTC (permalink / raw)
To: Libre-RISCV General Development; +Cc: Libre-RISCV General Development
In-Reply-To: <CAPweEDy=11Ax6+O75QQTyRFyWffuese-_9xXmF2F+K6+Df7xGg@mail.gmail.com>
Today (actual):
Spent today reading the parts of OPENPOWER ISA v3.1 related to traps,
not finished yet. I don't understand very much currently, but I think
that this will progress as I read and investigate further, however,
since I don't think I can read the spec well enough to understand what I
have to implement for fu/trap/main_stage, if someone else has the time I
think they should do it. Otherwise, I'll continue to work away at
understanding enough of the spec to be able to implement the trap pipeline.
Cole
_______________________________________________
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] microwatt decoder tables: M-Form and X-Form switched RS and RB
From: Luke Kenneth Casson Leighton @ 2020-06-02 0:01 UTC (permalink / raw)
To: Libre-RISCV General Development; +Cc: openpower-hdl-cores
In-Reply-To: <20200601231311.GA428673@thinks.paulus.ozlabs.org>
On Tuesday, June 2, 2020, Paul Mackerras <paulus@ozlabs.org> wrote:
>
> I'm not quite following, sorry. When you talk about the second or
> third position or field, are you talking about where they are in the
> instruction word (and if so, are you counting left to right, or right
> to left),
there is the field names RA, RB, RC, RS, RT
there is the order (left to right) in which these occur in an X-Form,
M-Form etc.
then, in microwatt, there are 3 input regs, a/b/c.
these do not match up consistently, neither across Forms, nor in microwatt
input reg ordering when it comes to RA, RS and reg a and reg c.
* OP_EXTS for example places RS into microwatt reg c
> or are you talking about the input_reg_a/b/c entries in
> microwatt's decode tables?
>
> If you're talking about positions in the instruction word, then either
> I'm misunderstanding your paragraph about X-form above, or it is
> incorrect, because instructions such as "and" and "or" take their
> inputs from RS and RB, so op1 is not RA for them, though it is for the
> arithmetic instructions.
yet OP_EXTS, which is also arithmetic, take inputs ftom RS but place it
into the *third* microwatt register, c, despite microwatt positions 1 and 2
(reg a and reg b) being empty.
in the intervening time (good morning to you, btw) i worked out a couple of
things:
* i was wrong about fixedshift operations: the way that microwatt allocates
them to internal reg a/b/c is very sensible.
these cannot be reordered.
* anything from fixedlogical (and, or, xor), fixedarith (specifically exts)
and mtcrf, these all use RS which goes into microwatt reg c, *all* whilst
also having microwatt reg a NONE.
consequently these *can* be swapped. whereever in decode1.vhdl there is
"NONE, NONE, RS" this can be replaced "RS, NONE, NONE" and reg a used
instead of reg c.
whilst this has no significance at the moment, because everything is
primarily global in nature (in execute1.vhdl), if all fixedarith ALU
functions were moved to a separate module, that module would require
*three* 63 bit paths (reg a b c) where only 2 of those are ever active at
any one time.
by making the change above, the hypothetical fixedarith.vhdl module needs
only 2 input operands, not 3.
> (I asked some of the original Power architects why there was this
> nonuniformity in the Power ISA where rotate and logical instructions
> use RS and RB as their inputs, whereas arithmetic instructions use RA
> and RB as inputs. The answer was that in the Power1 implementation,
> they were very short of gates, so they made the logic for rotating and
> masking data for store instructions also serve as the logic for doing
> the rotate and mask instructions.
ooo efficient microcoding. niiice. and ah now i see why the masking
exists: it would have been used to perform the insertion of (or extraction
of) offset bytes into or out of a LD/ST operand.
i love it.
Since stores take their data from
> RS, that meant that the rotate and shift instructions also took the
> data to be rotated or shifted from RS.)
i like it. the paths would be straight: no crossover MUXes to get RS to 2
different ports.
i seriously considered making the LD/ST Function Unit also capable of ADD
operations. the add is there, you still need an immediate, so why not.
the reason why not: ADD in POWER9, by being broken into carry, negation
etc, would have added quite a lot of logic to LD/ST. and be able to
read/write carry, and CR0 and it gets too much.
i may however still get it to do "straight" (non carry, non invert) ADD and
ADDI.
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 305] Create Pipelined ALU similar to alu_hier.py
From: bugzilla-daemon @ 2020-06-01 23:24 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-305-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=305
Bug 305 depends on bug 358, which changed state.
Bug 358 Summary: new MCU-ALU test picked up RC / OE / CR handling issue
https://bugs.libre-soc.org/show_bug.cgi?id=358
What |Removed |Added
----------------------------------------------------------------------------
Status|CONFIRMED |RESOLVED
Resolution|--- |FIXED
--
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 358] new MCU-ALU test picked up RC / OE / CR handling issue
From: bugzilla-daemon @ 2020-06-01 23:24 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-358-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=358
Luke Kenneth Casson Leighton <lkcl@lkcl.net> changed:
What |Removed |Added
----------------------------------------------------------------------------
Resolution|--- |FIXED
Status|CONFIRMED |RESOLVED
--
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