* [libre-riscv-dev] [Bug 365] ROCM/Libre-SOC GPU Opcode interoperability
From: bugzilla-daemon @ 2020-06-07 21:23 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-365-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=365
--- Comment #11 from Yehowshua <yimmanuel3@gatech.edu> ---
(In reply to Jacob Lifshay from comment #7)
> From what I understand, ROCm is only for OpenCL/CUDA-style compute, it
> doesn't implement the Vulkan or OpenGL APIs:
> https://github.com/RadeonOpenCompute/ROCm/issues/706
> https://github.com/RadeonOpenCompute/ROCm/issues/131
Hi Jacob, you are correct - however - Mesa has a Vulkan implementation that can
sit on top of ROCM.
--
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 @ 2020-06-07 21:23 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 #82 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Cole Poirier from comment #79)
> (In reply to Luke Kenneth Casson Leighton from comment #75)
> > the normal solution for this is to "call in" something that *is* known,
> > and that would be the qemu emulator. except we kinda need some help from
> > michael to do that in a short timescale, or we need to investigate it
> > in a longer one. given that there's a critical task needed (investigating
> > the CR bit-ordering) i'd rather we not pull him off of that.
>
> So until we have the known qemu emulator I should just try writing
> appropriate unit tests for when we have the emulator working? Should I raise
> a bug report for investigating the qemu powerpc emulator on a longer
> timescale?
yes please.
> What should I say in the bug description?
that we need one example unit test of how to run some assembly code
"qemu vs simulator" rather than "qemu vs hardware"
or, even just "qemu with some asserts" but looking *exactly* like how
it's done in the simulator, right now.
and that ultimately it should be possible to have a runtime switch
(or a suite of tests) that allow us to verify *aaalll* the simulator
test_pipe_caller.py tests, in the following way:
* simulator vs qemu
* qemu vs hardware
* hardware vs simulator
later i would very much like to be able to add microwatt to that list.
--
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 @ 2020-06-07 21:18 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 #81 from Cole Poirier <colepoirier@gmail.com> ---
(In reply to Luke Kenneth Casson Leighton from comment #80)
> (In reply to Cole Poirier from comment #79)
> > This is our autogenerated from parsing csv files emulator?
>
> nono, that's actually the function that gets called *by* the emulator.
> it's decoder/isa/caller.py (which is the base class for the simulator)
>
> > Is this something I can help with
>
> actually, probably yes. it should be as brain-dead simple as
> putting things into the self.spr dict:
>
> self.spr['SRR0'] = self.pc.CIA
>
> and also modify the NIA. that sort of thing.
>
>
> > So is this TRAP function added in sort of ad hoc via caller.py, because it
> > isn't contained in fixedtrap.mdwn?
>
> it is... however look at the PDF SPEC (or see comment #1 which contains
> the pseudo-code. that's not a function, is it?
No indeed it seems more similar to a keyword. Ah, so just translating that
psuedo code into the TRAP method of ISACaller?
> so what i did was, identify the keyword "TRAP" and got it to output
> a *function* call "self.TRAP()". auto-generated decoder/isa/fixedtrap.py
> now contains this:
>
> class fixedtrap:
>
> @inject()
> def op_twi(self, RA):
> a = EXTS(RA[32:64])
> if lt(a, EXTS(SI)) & TO[0]:self.TRAP()
> if gt(a, EXTS(SI)) & TO[1]:self.TRAP()
>
>
> *now* it can call the function which was added in caller.py, *now* we
> can get that function to make the required modifications to the SPRs
> and to the PC (NIA - Next Instruction Address).
Aha, this is actually starting to make A LOT of sense :)
--
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 @ 2020-06-07 21:11 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 #80 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Cole Poirier from comment #79)
> (In reply to Luke Kenneth Casson Leighton from comment #75)
> > commit d2e0e6bea57795fb522e0805b16e6f2852472f98
> > Author: Luke Kenneth Casson Leighton <lkcl@lkcl.net>
> > Date: Sun Jun 7 07:03:58 2020 +0100
> >
> > add MSR to simulator context
> >
> > as comment says. i also updated the git submodule so that sprset now
> > contains the (new, untested) mtmsr/mfmsr. if you rerun pywriter.py
> > you'll see the code now gets created.
> >
> > and that in turn means that if you add a unit test that uses mfmsr
> > or mtmsr, the code in the auto-generated sprset.py for op_mfmsr/op_mtmsr
> > will get called.
> >
> > whether that (completely untested) code is *accurate* or not is another
> > matter.
> >
> > we now have a situation of two "unknowns"
> >
> > 1. unknown simulator
> > 2. unknown hardware code.
> >
> > the normal solution for this is to "call in" something that *is* known,
> > and that would be the qemu emulator. except we kinda need some help from
> > michael to do that in a short timescale, or we need to investigate it
> > in a longer one. given that there's a critical task needed (investigating
> > the CR bit-ordering) i'd rather we not pull him off of that.
>
> So until we have the known qemu emulator I should just try writing
> appropriate unit tests for when we have the emulator working? Should I raise
> a bug report for investigating the qemu powerpc emulator on a longer
> timescale? What should I say in the bug description?
>
> (In reply to Luke Kenneth Casson Leighton from comment #76)
> > i'm also attempting to add support for TRAP - as a function - in the
> > parser and simulator, *without* changing fixedtrap.mdwn, which is
> > taken directly from the 3.0B PDF. bit of a pain.
> >
> > now, if you create a test_tdi() function in trap test_pipe_caller.py,
> > the simulator *should* print out "TRAP TODO".
> >
> >
> > diff --git a/src/soc/decoder/isa/caller.py b/src/soc/decoder/isa/caller.py
> > index 61aa451..ae5d479 100644
> > --- a/src/soc/decoder/isa/caller.py
> > +++ b/src/soc/decoder/isa/caller.py
> > @@ -233,6 +233,11 @@ class ISACaller:
> > self.decoder = decoder2.dec
> > self.dec2 = decoder2
> >
> > + def TRAP(self, trap_addr=0x700):
> > + print ("TRAP: TODO")
> > + # store PC in SRR0, set PC to 0x700
> > + # store MSR in SRR1, set MSR to um errr something
> > +
> > def memassign(self, ea, sz, val):
> > self.mem.memassign(ea, sz, val)
>
> This is our autogenerated from parsing csv files emulator?
nono, that's actually the function that gets called *by* the emulator.
it's decoder/isa/caller.py (which is the base class for the simulator)
> Is this something I can help with
actually, probably yes. it should be as brain-dead simple as
putting things into the self.spr dict:
self.spr['SRR0'] = self.pc.CIA
and also modify the NIA. that sort of thing.
> So is this TRAP function added in sort of ad hoc via caller.py, because it
> isn't contained in fixedtrap.mdwn?
it is... however look at the PDF SPEC (or see comment #1 which contains
the pseudo-code. that's not a function, is it?
so what i did was, identify the keyword "TRAP" and got it to output
a *function* call "self.TRAP()". auto-generated decoder/isa/fixedtrap.py
now contains this:
class fixedtrap:
@inject()
def op_twi(self, RA):
a = EXTS(RA[32:64])
if lt(a, EXTS(SI)) & TO[0]:self.TRAP()
if gt(a, EXTS(SI)) & TO[1]:self.TRAP()
*now* it can call the function which was added in caller.py, *now* we
can get that function to make the required modifications to the SPRs
and to the PC (NIA - Next Instruction Address).
--
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 @ 2020-06-07 20:51 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 #79 from Cole Poirier <colepoirier@gmail.com> ---
(In reply to Luke Kenneth Casson Leighton from comment #75)
> commit d2e0e6bea57795fb522e0805b16e6f2852472f98
> Author: Luke Kenneth Casson Leighton <lkcl@lkcl.net>
> Date: Sun Jun 7 07:03:58 2020 +0100
>
> add MSR to simulator context
>
> as comment says. i also updated the git submodule so that sprset now
> contains the (new, untested) mtmsr/mfmsr. if you rerun pywriter.py
> you'll see the code now gets created.
>
> and that in turn means that if you add a unit test that uses mfmsr
> or mtmsr, the code in the auto-generated sprset.py for op_mfmsr/op_mtmsr
> will get called.
>
> whether that (completely untested) code is *accurate* or not is another
> matter.
>
> we now have a situation of two "unknowns"
>
> 1. unknown simulator
> 2. unknown hardware code.
>
> the normal solution for this is to "call in" something that *is* known,
> and that would be the qemu emulator. except we kinda need some help from
> michael to do that in a short timescale, or we need to investigate it
> in a longer one. given that there's a critical task needed (investigating
> the CR bit-ordering) i'd rather we not pull him off of that.
So until we have the known qemu emulator I should just try writing appropriate
unit tests for when we have the emulator working? Should I raise a bug report
for investigating the qemu powerpc emulator on a longer timescale? What should
I say in the bug description?
(In reply to Luke Kenneth Casson Leighton from comment #76)
> i'm also attempting to add support for TRAP - as a function - in the
> parser and simulator, *without* changing fixedtrap.mdwn, which is
> taken directly from the 3.0B PDF. bit of a pain.
>
> now, if you create a test_tdi() function in trap test_pipe_caller.py,
> the simulator *should* print out "TRAP TODO".
>
>
> diff --git a/src/soc/decoder/isa/caller.py b/src/soc/decoder/isa/caller.py
> index 61aa451..ae5d479 100644
> --- a/src/soc/decoder/isa/caller.py
> +++ b/src/soc/decoder/isa/caller.py
> @@ -233,6 +233,11 @@ class ISACaller:
> self.decoder = decoder2.dec
> self.dec2 = decoder2
>
> + def TRAP(self, trap_addr=0x700):
> + print ("TRAP: TODO")
> + # store PC in SRR0, set PC to 0x700
> + # store MSR in SRR1, set MSR to um errr something
> +
> def memassign(self, ea, sz, val):
> self.mem.memassign(ea, sz, val)
This is our autogenerated from parsing csv files emulator? Is this something I
can help with or are you handling it?
So is this TRAP function added in sort of ad hoc via caller.py, because it
isn't contained in fixedtrap.mdwn? Is it absent from fixedtrap.mdwn because
this is a different kind of trap?
--
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 @ 2020-06-07 20:45 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 #78 from Cole Poirier <colepoirier@gmail.com> ---
(In reply to Luke Kenneth Casson Leighton from comment #74)
> (In reply to Cole Poirier from comment #72)
>
> > I'm having a difficult time understanding how to write the first test for
> > class TrapTestCase looking both at cr/test/test_pipe_caller and
> > logical/test/test_pipe_caller... Do I have to translate all the code that
> > was set up for CR into setting up for TRAP? Trying to think through it I
> > can't figure it out, and I suspect it is related to my previous comment
> > about not understanding how the test_pipe_caller, XXXInput/OutputData,
> > CompALUOpSubset, and CompTrapOpSubset, and the decoder, relate to and
> > interact with each other... I have a little idea of each but I'm doing the
> > 'freeze cause I don't understand' right now.
>
> ok. suggestion. take alu test_pipe_caller.py and cut out absolutely every
> test_* except for one, make sure to pick one with only one assembly
> instruction.
>
> (or take copy of tpc.py, call it tpc2.py)
>
> then run it and load the generated alu_simulator.gtk in gtkwave.
>
> at the top level you will see an instruction.
>
> add that to the window.
>
> then just go through the hierarchy adding signals.
>
> look especially for these
>
> fn_unit
> dest1
> dest2
> src1_i
> src2_i
> issue_i
> busy_o
> rd_go_i
> rd_rel_o
> wr go req as well
>
> then go *back* to the source code (remember have SEVERAL files open at once)
>
> note that "hey anywhere i have m.submodules those show up in the hierarchy!"
>
> note that the Record contents also show up in the gtkwave output.
>
> look at the debug output and note the instruction assembly code 0xNnnnff1234
> etc got outputted and that it also matches that toplevel signal in gtkwave.
>
> the tree hierarchy in gtkwave will start to give you an idea of the
> interrelationships of the modules, in ways that staring at the static code
> simply cannot give you, and the "time" dimension will show you valuable
> insights too.
>
> once you have got used to this, add *one* more unit test back in, rerun
> tpc2.py, and hit ctrl-shift-r in gtkwave.
>
> you should be able to track the change in outputs, the result from the
> operation, with what the debug output shows.
Thank you, this is the process for MUCH greater understanding I mentioned
above. I plan on working on this today after doing a more polished draft of the
b-deck, and fixing the trap() convenience function. Probably will work on this
for a few days until I feel like I'm starting to get a handle on things. Will
also work on translating psuedo code to nmigen for SPR tomorrow.
--
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 @ 2020-06-07 20:42 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 #77 from Cole Poirier <colepoirier@gmail.com> ---
(In reply to Luke Kenneth Casson Leighton from comment #73)
> will explain tomorrow when it is not 4am.
You are very bad at taking your own advice :) Thank you for explaining... the
entire process I should follow to gain MUCH greater understanding of the SOC
and how it works... all at 4 am!!
> briefyly: every Record has a dictionary called fields. by walking this
> dictionary we can cheat and copy *only* a subset of fields... as long as the
> names all match up.
>
> see eq_from_execute1 and grep source code for where it is used.
>
>
> > One difference between OP_TRAP and OP_SC is that the end of the OP_TRAP
> > function is:
> >
> > ```
> > # take a copy of the current PC in SRR0
> > comb += srr0_o.data.eq(cia_i) # old PC
> > comb += srr0_o.ok.eq(1)
> > ```
> > While the end of the OP_SC function is:
> >
> > ```
> > # and store the (next-after-return) PC in SRR0
> > comb += srr0_o.data.eq(cia_i+4) # addr to begin from on return
> > comb += srr0_o.ok.eq(1)
> > ```
> >
> > So should the function be:
> >
> > ```
> > def trap(self, m, addr, nia_o, srr0_o, trap_addr):
> > comb = m.d.comb
> > comb += nia_o.data.eq(trap_addr)
> > comb += nia_o.ok.eq(1)
> >
> > comb += srr0_o.data.eq(addr) # addr to begin from on return
> > comb += srr0_o.ok.eq(1)
>
> yes! well done. ok i would cut out srr0_o from those args and just redo
> them inside the fn.
Thanks! Will do.
> so now also cookie cut srr0_0 and nia_o from lines 89 and 90 "convenience"
> variables. see how that works?
>
> so OP_SC should call self.trap(0xc00, cia_i+4) and OP_TRAP should be obvious.
>
>
> > ```
> >
> > The below is where my shakiness on the CompXOpSubset and the decoder starts
> > to confuse me. If you can briefly explain that to me here, I can use this
> > when doing the documentation after the tape-out deadline.
>
> tomorrow.
>
> late.
Hehe :)
--
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 @ 2020-06-07 20:35 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 #53 from Cesar Strauss <cestrauss@gmail.com> ---
commit 71a791cf5f05bd5858fdc60f4ccdca936edc1d75
Author: Cesar Strauss <cestrauss@gmail.com>
Date: Sun Jun 7 16:32:11 2020 -0300
Make the test ALU conform to the valid/ready protocol
Adjust the test case accordingly.
It was definitively not a one-line change. But it was fun.
Select any other operation than MUL, ADD, or SHR, and it will do a
combinatorial subtraction, with zero-delay and no registering.
The interesting thing about zero-delay is that, due to the absence of an
internal register, it will not release the input data until the result is
transferred out.
Basically, the handshake signals of the input and output ports are directly
connected between them, only the data is modified between the ports.
Nice change, working with RTL again. Will go back to the parallel unit tests,
now.
--
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 363] inconsistency between isel and mfcr and crand unit test on bit-ordering of CR
From: bugzilla-daemon @ 2020-06-07 20:31 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-363-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=363
--- Comment #10 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Michael Nolan from comment #9)
> GRR. I believe at some point I used the fields directly generated from
> power_fields.py (fields_x['BF'] or something like that), which needed the
> [0:-1]. But since it got changed to use the signals from the decoder, the
> [0:-1] chopped off the top bit.
ohh... :)
> I think I fixed another error in the decoder
> just like that recently,
i spotted that BFA didn't have [0:-1] on it, which was partly the clue
which led me to try the same on BF.
got there in the end. soc/simple/test/test_core.py now passes and
it runs *all* unit tests. still TODO, the XER checking though
--
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 365] ROCM/Libre-SOC GPU Opcode interoperability
From: bugzilla-daemon @ 2020-06-07 20:29 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-365-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=365
--- Comment #10 from Cole Poirier <colepoirier@gmail.com> ---
(In reply to Jacob Lifshay from comment #7)
> From what I understand, ROCm is only for OpenCL/CUDA-style compute, it
> doesn't implement the Vulkan or OpenGL APIs:
> https://github.com/RadeonOpenCompute/ROCm/issues/706
> https://github.com/RadeonOpenCompute/ROCm/issues/131
(In reply to Luke Kenneth Casson Leighton from comment #8)
> https://gpuopen.com/gpuperfapi/
>
> that's interesting. Version 3.5 (dec 2019):
> "Remove ROCm/HSA support."
Can you clarify the purpose of the present investigation for me?
Are we trying to find out if we can use the many man-years of development put
into the Radeon Open Compute Stack by making minimal modifications to our
external graphics api to make it compatible with HSA/ROCM?
Is HSA general across all radeon gpus? I remember reading that HSA was an
abstraction specific to only two of AMD's APUs
Are we looking not so much for OpenCL support, but instead the graphics stack
for OpenGL, Vulkan, DirectX, etc?
Or are we looking for both Compute and Graphics?
Would we take the route of making a fork of clang/llvm specific to Libre-SOC,
and trying to have these changes upstreamed to clang/llvm trunk?
All of these questions are asked in near total ignorance.
--
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 363] inconsistency between isel and mfcr and crand unit test on bit-ordering of CR
From: bugzilla-daemon @ 2020-06-07 20:25 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-363-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=363
--- Comment #9 from Michael Nolan <mtnolan2640@gmail.com> ---
GRR. I believe at some point I used the fields directly generated from
power_fields.py (fields_x['BF'] or something like that), which needed the
[0:-1]. But since it got changed to use the signals from the decoder, the
[0:-1] chopped off the top bit. I think I fixed another error in the decoder
just like that recently, I probably should have gone through and made sure
there weren't any like 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
* [libre-riscv-dev] [Bug 365] ROCM/Libre-SOC GPU Opcode interoperability
From: bugzilla-daemon @ 2020-06-07 20:21 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-365-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=365
--- Comment #9 from Cole Poirier <colepoirier@gmail.com> ---
(In reply to Luke Kenneth Casson Leighton from comment #6)
> sadly, luc (who is named explicitly in ARM NDAs that under no circumstances
> is he to be contacted for any reason by the signatories) has been under such
> constant attack for over a decade, for his work, that he has given up
> working on graphics drivers entirely.
That's devastatingly sad, what a loss for the libre community! Just so
incredibly upsetting.
> a friend of his found him some work to do and he now has enough money to pay
> for his family.
I'm thankful that he was able to survive, but it's still such a tragedy.
> some microsoft employees at one point gave serious consideration to engaging
> in a similar style of concerted attack against me, because of the reverse
> engineering that i did back in 1996-2000. they even called my employer,
> ISS, to arrange to have me fired or silenced.
>
> several senior employees inside microsoft, people who had been with the
> company since its beginning, had to explain to them in very blunt and clear
> terms that if they pissed me off, the knowledge and expertise that i had on
> the security vulnerabilities within the NT Operating System (of which those
> senior employees were keenly aware) could, if i focussed on revealing those
> vuulnerabilities day after day, week after week, could literally have
> brought their billion dollar company to its knees.
>
> they left me alone.
This in contrast to what happened to Luc is pretty darn cool. I'm very glad you
posed a great enough threat that you were left alone.
> luc verhaegen was not in a similar position because MALI and GPUs in general
> are not exactly critical components (unlike spectre, meltdown etc), and
> there is not a monopoly situation like there was with microsoft.
That's interesting, learning more about these market dynamics would probably
help me write better pitches. Not antagonistic, just knowing the state of the
markets we are trying to get into.
> now you know a little bit more about the background, why i started this
> project, and also why full transparency is so very important. it's because
> with full transparency there *is* no opportunity to exploit, blackmail or
> undermine software libre developers, and there *is* no need for people to
> frivolously have their time and expertise wasted on reverse engineering.
This is why I joined the project and am trying to take a more active role in
the company as well. You, Luke are the reason I joined the project, because of
your awesome capabilities gained through very smart "laziness" in your reverse
engineering and learning, but far more importantly your commitment to the
Titanian principles. I want to help you succeed in your mission to bring
genuine ethics to technology.
> it just so happens that this results in things being far easier for
> customers (like the fact that RTOSes such as the Amazon IOT one) get
> *direct* access to GPU capabilities, debugging is easier, extensibility is
> easier, development costs are dramatically reduced and so on.
I'm focusing heavily on this in the pitch. I really liked Cesar's comments from
a few weeks ago about the difference our open stack would make to his
productivity and capabilities when it comes to scientific data acquisition. I
think this definitely applies across the board from hobbyists right up through
mega corporations. With open standards and documentation, from the top of the
software stack, down through firmware and to the level of hardware schematic
diagrams, "debugging is easier, extensibility is easier, development costs are
dramatically reduced and so on."
--
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 363] inconsistency between isel and mfcr and crand unit test on bit-ordering of CR
From: bugzilla-daemon @ 2020-06-07 20:19 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-363-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=363
Luke Kenneth Casson Leighton <lkcl@lkcl.net> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|CONFIRMED |RESOLVED
Resolution|--- |FIXED
--- Comment #8 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
michael i've just been investigating by adding extra unit tests that
i believe to be reasonable:
def test_1_mcrf(self):
for i in range(20):
src = random.randint(0, 7)
dst = random.randint(0, 7)
lst = [f"mcrf {src}, {dst}"]
cr = random.randint(0, (1<<32)-1)
self.run_tst_program(Program(lst), initial_cr=cr)
def test_0_mcrf(self):
for i in range(8):
lst = [f"mcrf 5, {i}"]
cr = 0xfeff0001
self.run_tst_program(Program(lst), initial_cr=cr)
(i used numbering to get them to run first)
test_0_mcrf one fails straight away (even on the first loop):
self.assertEqual(expected_cr, real_cr, code)
AssertionError: 14 != 15 : mcrf 5, 0
a clue is here:
def op_mcrf(self, CR):
print ("op_mcrf", hex(CR.si.value), "BF", BF, "BFA", BFA)
CR.si[(4 * BF) + 32:(4 * BF) + 35 + 1] = CR.si[(4 * BFA) + 32:(4 * BFA)
+ 35 + 1]
print ("op_mcrf", hex(CR.si.value))
this prints out:
op_mcrf 0xfeff0001 BF 5 BFA 0
op_mcrf 0xfeff0f01
and... oink, the write_cr is invalid. it's set to 1.
cr_sel = yield dec2.e.write_cr.data
expected_cr = simulator.cr.get_range().value
print(f"CR whole: {expected_cr:x}, sel {cr_sel}")
CR whole: feff0f01, sel 1
ah HA!
--- a/src/soc/decoder/power_decoder2.py
+++ b/src/soc/decoder/power_decoder2.py
@@ -445,7 +445,7 @@ class DecodeCROut(Elaboratable):
comb += self.cr_bitfield.data.eq(0)
comb += self.cr_bitfield.ok.eq(self.rc_in) # only when RC=1
with m.Case(CROutSel.BF):
- comb += self.cr_bitfield.data.eq(self.dec.FormX.BF[0:-1])
+ comb += self.cr_bitfield.data.eq(self.dec.FormX.BF)
comb += self.cr_bitfield.ok.eq(1)
with m.Case(CROutSel.BT):
that's it! that's all it was. what a pain.
--
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 365] ROCM/Libre-SOC GPU Opcode interoperability
From: bugzilla-daemon @ 2020-06-07 17:52 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-365-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=365
--- Comment #8 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
https://gpuopen.com/gpuperfapi/
that's interesting. Version 3.5 (dec 2019):
"Remove ROCm/HSA support."
--
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 365] ROCM/Libre-SOC GPU Opcode interoperability
From: bugzilla-daemon @ 2020-06-07 17:41 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-365-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=365
Jacob Lifshay <programmerjake@gmail.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |programmerjake@gmail.com
--- Comment #7 from Jacob Lifshay <programmerjake@gmail.com> ---
From what I understand, ROCm is only for OpenCL/CUDA-style compute, it doesn't
implement the Vulkan or OpenGL APIs:
https://github.com/RadeonOpenCompute/ROCm/issues/706
https://github.com/RadeonOpenCompute/ROCm/issues/131
--
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 360] move RS to 1st or 2nd operand in CSV files
From: bugzilla-daemon @ 2020-06-07 15:08 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-360-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=360
--- Comment #2 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
MUL and DIV pipelines also need checking: however these actually have
3-operand RA RB RC so it *may* not be applicable.
--
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 07jun2020
From: Luke Kenneth Casson Leighton @ 2020-06-07 15:03 UTC (permalink / raw)
To: Libre-RISCV General Development
yesterday:
* added LDSTCompUnit unit test, got it partly working for LD. the
failures are down to L0CacheBuffer memory misalignments
* redesigned LDSTMultiCompUnit to look more like other derivatives
from MultiCompUnit
* added TestMemory to the compunit unit tests and added the ability to
initialise Simulator Memory
* went over TRAP with Cole
* discussed how to do interrupts and other exceptions (illegal
instruction) by changing the OP_xxx to OP_TRAP and executing that,
instead
* added a TRAP function (TODO) in the Simulator
* added MSR to the Simulator
* added mtmsr/mfmsr to the Simulator.
this morning/afternoon:
* tried to reduce the number of patches to auto-generated simulator
code from pseudocde (failed)
* added CA/CA32 to the parser-generator, to support sraw/srad creating
CA and CA32 in an unusual way
* updated the simulator to support the same
* updated rotator.py to match microwatt
* got the SHIFTROT unit tests working (in both fu/shift_rot
test_pipe_caller and test_shiftrot_compunit)
today:
* think about how to fix LDSTCompUnit (more specifically, the
prototype non-production L0CacheBuffer)
* continue debugging unit tests after adding CA/CA32.
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 339] create POWER9 ROTATE (SHIFTROT) pipeline
From: bugzilla-daemon @ 2020-06-07 14:32 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-339-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=339
--- Comment #8 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Luke Kenneth Casson Leighton from comment #7)
> sradi is different
>
> carry <- s & ((r& ¬m) != 0) CA <- carry CA32 <- carry
>
> best add CA and CA32 to the fields that can be returned.
sorted. it's a bit of a mess, had to special-case it, but it's
functional.
also noted that rotator.vhdl has changed (support for OP_EXTSWSLI
which i may add next)
--
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 @ 2020-06-07 6:26 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 #76 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
i'm also attempting to add support for TRAP - as a function - in the
parser and simulator, *without* changing fixedtrap.mdwn, which is
taken directly from the 3.0B PDF. bit of a pain.
now, if you create a test_tdi() function in trap test_pipe_caller.py,
the simulator *should* print out "TRAP TODO".
diff --git a/src/soc/decoder/isa/caller.py b/src/soc/decoder/isa/caller.py
index 61aa451..ae5d479 100644
--- a/src/soc/decoder/isa/caller.py
+++ b/src/soc/decoder/isa/caller.py
@@ -233,6 +233,11 @@ class ISACaller:
self.decoder = decoder2.dec
self.dec2 = decoder2
+ def TRAP(self, trap_addr=0x700):
+ print ("TRAP: TODO")
+ # store PC in SRR0, set PC to 0x700
+ # store MSR in SRR1, set MSR to um errr something
+
def memassign(self, ea, sz, val):
self.mem.memassign(ea, sz, val)
--
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 @ 2020-06-07 6:11 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 #75 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
commit d2e0e6bea57795fb522e0805b16e6f2852472f98
Author: Luke Kenneth Casson Leighton <lkcl@lkcl.net>
Date: Sun Jun 7 07:03:58 2020 +0100
add MSR to simulator context
as comment says. i also updated the git submodule so that sprset now
contains the (new, untested) mtmsr/mfmsr. if you rerun pywriter.py
you'll see the code now gets created.
and that in turn means that if you add a unit test that uses mfmsr
or mtmsr, the code in the auto-generated sprset.py for op_mfmsr/op_mtmsr
will get called.
whether that (completely untested) code is *accurate* or not is another matter.
we now have a situation of two "unknowns"
1. unknown simulator
2. unknown hardware code.
the normal solution for this is to "call in" something that *is* known,
and that would be the qemu emulator. except we kinda need some help from
michael to do that in a short timescale, or we need to investigate it
in a longer one. given that there's a critical task needed (investigating
the CR bit-ordering) i'd rather we not pull him off of that.
--
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 365] ROCM/Libre-SOC GPU Opcode interoperability
From: bugzilla-daemon @ 2020-06-07 4:30 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-365-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=365
--- Comment #6 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Cole Poirier from comment #5)
> In light of this, I think what would be vastly more valuable would be to
> solicit Luc Verhaegen’s input on this bug specifically. I believe I remember
> him commenting on list a few times, however I can’t recall with enough
> clarity to comment.
sadly, luc (who is named explicitly in ARM NDAs that under no circumstances is
he to be contacted for any reason by the signatories) has been under such
constant attack for over a decade, for his work, that he has given up working
on graphics drivers entirely.
a friend of his found him some work to do and he now has enough money to pay
for his family.
some microsoft employees at one point gave serious consideration to engaging in
a similar style of concerted attack against me, because of the reverse
engineering that i did back in 1996-2000. they even called my employer, ISS,
to arrange to have me fired or silenced.
several senior employees inside microsoft, people who had been with the company
since its beginning, had to explain to them in very blunt and clear terms that
if they pissed me off, the knowledge and expertise that i had on the security
vulnerabilities within the NT Operating System (of which those senior employees
were keenly aware) could, if i focussed on revealing those vuulnerabilities day
after day, week after week, could literally have brought their billion dollar
company to its knees.
they left me alone.
luc verhaegen was not in a similar position because MALI and GPUs in general
are not exactly critical components (unlike spectre, meltdown etc), and there
is not a monopoly situation like there was with microsoft.
now you know a little bit more about the background, why i started this
project, and also why full transparency is so very important. it's because
with full transparency there *is* no opportunity to exploit, blackmail or
undermine software libre developers, and there *is* no need for people to
frivolously have their time and expertise wasted on reverse engineering.
it just so happens that this results in things being far easier for customers
(like the fact that RTOSes such as the Amazon IOT one) get *direct* access to
GPU capabilities, debugging is easier, extensibility is easier, development
costs are dramatically reduced and so on.
--
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 365] ROCM/Libre-SOC GPU Opcode interoperability
From: bugzilla-daemon @ 2020-06-07 3:47 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-365-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=365
--- Comment #5 from Cole Poirier <colepoirier@gmail.com> ---
(In reply to Luke Kenneth Casson Leighton from comment #4)
> (In reply to Cole Poirier from comment #3)
> > Dave Airlie
>
> caused massive problems for luc verhaegen, setting him back years in his
> reverse engineering efforts of AMDGPU.
>
> airlied's disruptive interference and
> blatant disregard for luc's efforts caused AMD to terminate release of key
> strategic internal documentation that luc had spent several YEARS carefully
> negotiating to get public access to for the wider benefit of the entire
> libre community *and* ultimately of AMD.
YIKES!! I read about that a year ago from links you posted, I was very upset by
Airlies actions. I clearly forgot that the name of the villain in that story
was Airlie...
In light of this, I think what would be vastly more valuable would be to
solicit Luc Verhaegen’s input on this bug specifically. I believe I remember
him commenting on list a few times, however I can’t recall with enough clarity
to comment.
--
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 365] ROCM/Libre-SOC GPU Opcode interoperability
From: bugzilla-daemon @ 2020-06-07 3:31 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-365-13@https.bugs.libre-soc.org/>
https://bugs.libre-soc.org/show_bug.cgi?id=365
--- Comment #4 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Cole Poirier from comment #3)
> Dave Airlie
caused massive problems for luc verhaegen, setting him back years in his
reverse engineering efforts of AMDGPU.
airlied's disruptive interference and comp
blatant disregard for luc's efforts caused AMD to terminate release of key
strategic internal documentation that luc had spent several YEARS carefully
negotiating to get public access to for the wider benefit of the entire libre
community *and* ultimately of AMD.
--
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 @ 2020-06-07 3:23 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 #74 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Cole Poirier from comment #72)
> I'm having a difficult time understanding how to write the first test for
> class TrapTestCase looking both at cr/test/test_pipe_caller and
> logical/test/test_pipe_caller... Do I have to translate all the code that
> was set up for CR into setting up for TRAP? Trying to think through it I
> can't figure it out, and I suspect it is related to my previous comment
> about not understanding how the test_pipe_caller, XXXInput/OutputData,
> CompALUOpSubset, and CompTrapOpSubset, and the decoder, relate to and
> interact with each other... I have a little idea of each but I'm doing the
> 'freeze cause I don't understand' right now.
ok. suggestion. take alu test_pipe_caller.py and cut out absolutely every
test_* except for one, make sure to pick one with only one assembly
instruction.
(or take copy of tpc.py, call it tpc2.py)
then run it and load the generated alu_simulator.gtk in gtkwave.
at the top level you will see an instruction.
add that to the window.
then just go through the hierarchy adding signals.
look especially for these
fn_unit
dest1
dest2
src1_i
src2_i
issue_i
busy_o
rd_go_i
rd_rel_o
wr go req as well
then go *back* to the source code (remember have SEVERAL files open at once)
note that "hey anywhere i have m.submodules those show up in the hierarchy!"
note that the Record contents also show up in the gtkwave output.
look at the debug output and note the instruction assembly code 0xNnnnff1234
etc got outputted and that it also matches that toplevel signal in gtkwave.
the tree hierarchy in gtkwave will start to give you an idea of the
interrelationships of the modules, in ways that staring at the static code
simply cannot give you, and the "time" dimension will show you valuable
insights too.
once you have got used to this, add *one* more unit test back in, rerun
tpc2.py, and hit ctrl-shift-r in gtkwave.
you should be able to track the change in outputs, the result from the
operation, with what the debug output shows.
--
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 @ 2020-06-07 3:09 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 #73 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Cole Poirier from comment #71)
> (In reply to Luke Kenneth Casson Leighton from comment #68)
> > actually... cole, can you do the same thing as i just did with
> > msr_check_pr() create a function called trap(m, addr, nia)?
> >
> > except this time it is a member function of the class, so trap(self, m,
> > addr, nia)
> >
> > and just as you see i redid comb=m.d.comb in msr_check_pr, repeat/copy the
> > convenience variables that allow you to set self.o.nia etc etc
> >
> > the reason for making this function is because you can see it is used twice
> > already.
> >
> > also i am thinking it may be a good idea to add the address (0x700, etc)
> > actually to the trap_input_record and pass it in from the *decode* phase.
>
> Sorry, I'm very shaky on the dynamic input records and how they work with
> the decoder.
will explain tomorrow when it is not 4am.
briefyly: every Record has a dictionary called fields. by walking this
dictionary we can cheat and copy *only* a subset of fields... as long as the
names all match up.
see eq_from_execute1 and grep source code for where it is used.
> One difference between OP_TRAP and OP_SC is that the end of the OP_TRAP
> function is:
>
> ```
> # take a copy of the current PC in SRR0
> comb += srr0_o.data.eq(cia_i) # old PC
> comb += srr0_o.ok.eq(1)
> ```
> While the end of the OP_SC function is:
>
> ```
> # and store the (next-after-return) PC in SRR0
> comb += srr0_o.data.eq(cia_i+4) # addr to begin from on return
> comb += srr0_o.ok.eq(1)
> ```
>
> So should the function be:
>
> ```
> def trap(self, m, addr, nia_o, srr0_o, trap_addr):
> comb = m.d.comb
> comb += nia_o.data.eq(trap_addr)
> comb += nia_o.ok.eq(1)
>
> comb += srr0_o.data.eq(addr) # addr to begin from on return
> comb += srr0_o.ok.eq(1)
yes! well done. ok i would cut out srr0_o from those args and just redo them
inside the fn.
so now also cookie cut srr0_0 and nia_o from lines 89 and 90 "convenience"
variables. see how that works?
so OP_SC should call self.trap(0xc00, cia_i+4) and OP_TRAP should be obvious.
> ```
>
> The below is where my shakiness on the CompXOpSubset and the decoder starts
> to confuse me. If you can briefly explain that to me here, I can use this
> when doing the documentation after the tape-out deadline.
tomorrow.
late.
--
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