* [libre-riscv-dev] [Bug 265] new server from raptorcs needs set up with build/test rnvironment
From: bugzilla-daemon @ 2020-03-22 23:36 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-265-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=265
--- Comment #1 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
veera would you be happy to log in and manage this server? if so please send me
an ssh pub key.
we have to agree not to do anything illegal with this server, we are guests on
raptor engineering's network.
it will need fail2ban configuring, i have it installed already. whitelisting
of the VPN 10.6.0.0/16, changing ssh to recognise port 922, enabling ssh
fail2ban and also enanling recidive.
also i think installing ssmtpd so that it can at least send email out, use
10.6.0.1 as the SMTP relay.
the VPN also has access to libre-riscv.org, to the build server that jacob set
up, and to my laptop. so we need to be careful.
--
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 265] New: new server from raptorcs needs set up with build/test rnvironment
From: bugzilla-daemon @ 2020-03-22 23:28 UTC (permalink / raw)
To: libre-riscv-dev
http://bugs.libre-riscv.org/show_bug.cgi?id=265
Bug ID: 265
Summary: new server from raptorcs needs set up with build/test
rnvironment
Product: Libre-RISC-V Website
Version: unspecified
Hardware: Other
OS: Linux
Status: CONFIRMED
Severity: enhancement
Priority: ---
Component: website
Assignee: lkcl@lkcl.net
Reporter: lkcl@lkcl.net
CC: libre-riscv-dev@lists.libre-riscv.org
NLnet milestone: ---
we have been donated access to a monster Talos II workstation with 128GB RAM
and a 24 core 4x threaded processor.
i have debian 10 ppc64 installed on it, and we need to get nmigen, and a stack
of test software compiled and running.
this invludes coriolis2, nmigen, jacob's algorithmic FP library, the works.
it is a critical resource for example in being able to verify correctness of
the IEEE754 FP library, by having a compliant system against which testing can
be done particularly for rounding modes.
also it can act as another build server.
the NVMe is only 128GB so we cannot go mad with 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 264] New: ISA switch needs to be a privileged operation
From: bugzilla-daemon @ 2020-03-22 23:19 UTC (permalink / raw)
To: libre-riscv-dev
http://bugs.libre-riscv.org/show_bug.cgi?id=264
Bug ID: 264
Summary: ISA switch needs to be a privileged operation
Product: Libre-SOC's first SoC
Version: unspecified
Hardware: Other
OS: Linux
Status: CONFIRMED
Severity: enhancement
Priority: ---
Component: Specification
Assignee: lkcl@lkcl.net
Reporter: lkcl@lkcl.net
CC: libre-riscv-dev@lists.libre-riscv.org
NLnet milestone: ---
we've had a request from the OpenPOWER foundation that ISA switching be a
privileged operation.
this to prevent endusers from mix and matching arbitrary assembler from
multiple compilers and then expecting "vanilla" systems to support the
resultant excruciatingly nauseous binaries.
a privileged operation will allow say a thread to be established by the kernel
with the same memory space as the userspace application but with a different
executable (the 3D shader compiled application)
in particular we have a bit of a problem when it comes to inlining POWER
assembler with SVPrefix: POWER opcodes can be either LE or BE, and
unfortunately it is the TOP 6 bits which specify the opcode.
if they are LE, it is not ok, exactly, to read 32 bits and have the decoder
then work out if the opcode is actually a 16 bit Compressed instruction.
that just doesn't work. it's "ok" for 48 and 64 bit SVPrefix (just a bit
weird) however for C it most definitely is not.
however for BE it just so happens that those top 6 bits of a 32 bit opcode will
be byte reversed and so will be in the very first byte of an instruction
stream.
this kind of mess is also why it should be a privileged op to do the ISA
switch, really.
needs thinking through, properly.
--
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 257] Implement demo Load/Store queueing algorithm
From: bugzilla-daemon @ 2020-03-22 11:37 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-257-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=257
--- Comment #14 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
https://git.libre-riscv.org/?p=soc.git;a=blob;f=src/soc/scoreboard/addr_split.py;h=40f4569c38defa7fbf31b68f186f064d2df3e732;hb=f58721bc872b0e0b487a14dfa6a816774b433892
putting together a basic LD/ST address-splitter. however it includes
signalling which will connect directly to CompLDST.
basically, the address+len expands to a bit-mask, that could be mis-aligned,
so the "normal" half is sent truncated on a cache-line-boundary and the "top"
half is split into a *second* LD/ST.
what that means is that the LDSTCompUnit now needs to wait for *TWO* LD/STs
to complete.
therefore, we need valid/data signalling for *two* LD/STs and, on successful
completion of *both*, the data needs "re-assembly" based on the *two* masks
--
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 184] new mailing lists proposal for libre-soc.org
From: bugzilla-daemon @ 2020-03-21 23:24 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-184-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=184
vklr@vkten.in <vklr@vkten.in> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |vklr@vkten.in
--- Comment #6 from vklr@vkten.in <vklr@vkten.in> ---
(In reply to Jacob Lifshay from comment #4)
> (In reply to Luke Kenneth Casson Leighton from comment #0)
> > These should be *new* lists on libre-soc.org (retire libre-riscv-dev)
> >
> > 1) Libre-SOC-Dev, for hardware and software development (productive)
> > discussion
> > 2) Libre-SOC-Org, for discussion about related to the organization
> > itself, such as funding, investing, finances, domain change, website
> > design, members, ethics, business, administration, you get my point
> > 3) Libre-SOC-Misc, for everything else (the non serious stuff that I
> > can safely ignore unlike the first two which should have important
> > stuff only)
> > 4) Bugtracker separate list, readonly announce only, and replyto set to
> > bugzilla ready for when/if incoming bugzilla is ever set up
>
> I think the above proposal is a good idea. When we finally shut-down
> libre-riscv-dev, I think we should send out a single notification email to
> everyone who's subscribed even if they have mail delivery turned off. We
> should also announce our name & domain change on crowdsupply.
>
> Is everyone ok with the above plan?
I am also fine with 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 186] Create decoder for SOC: Power ISA and RISC-V
From: bugzilla-daemon @ 2020-03-21 20:10 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-186-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=186
--- Comment #95 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Michael Nolan from comment #94)
> Yeah, seems reasonable. This wouldn't be cycle accurate or anything right,
well, cycle accurate just means that everything on each clock is exactly as it
would be after one clock
it's actually harder to do a nonaccurate simulator as it involves JIT
translation etc.
cycle accurate means doing TLBs in the simulator, too, and that means we have
something to test the hardware against, there, too.
> just a simulator for our eventual backend?
yes. however adding a power frontend on it as we found ain't so bad. can use
the same code.
if you look here you can see how basic the test simulator is
https://git.libre-riscv.org/?p=soc.git;a=blob;f=src/soc/experiment/score6600.py;h=babce2439233f84c262c08795b873a6167197ec9;hb=HEAD#l859
adjusting that so it uses InternalOp is... er... not hard? :)
--
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 186] Create decoder for SOC: Power ISA and RISC-V
From: bugzilla-daemon @ 2020-03-21 18:59 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-186-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=186
--- Comment #94 from Michael Nolan <mtnolan2640@gmail.com> ---
(In reply to Luke Kenneth Casson Leighton from comment #93)
> y'know what, michael... you see in soc experiments comp6600 there is the
> beginnings of a Simulator?
>
> i have an idea, it might be worthwhile to run with that to its logical
> obvious implication, namely to actually have a simulator, in soc, in python,
> which interprets the InternalOps.
>
> we then have a reference implementation which we can document properly etc.
> with links to the PDF.
>
> also it means we have to write the code and so understsnd it.
>
> speed is not crucial as the nmigen simulator will be screamingly slow. we
> can use cocotb later.
>
> also at some point we can drop-in power-gem5.
>
> what does everyone think?
Yeah, seems reasonable. This wouldn't be cycle accurate or anything right, just
a simulator for our eventual backend?
--
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] Please go through the new public-inbox elaborate setup revison
From: Luke Kenneth Casson Leighton @ 2020-03-21 15:52 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <20200320210511.GA31305@lily.local>
On Friday, March 20, 2020, Veera <vklr@vkten.in> wrote:
> On Fri, Mar 20, 2020 at 06:30:00PM +0000, Luke Kenneth Casson Leighton
> wrote:
> > On Fri, Mar 20, 2020 at 1:00 PM Veera <vklr@vkten.in> wrote:
> > >
> > > Hi,
> > >
> > > I finally worked out what was required from public-inbox + exim4 +
> mailman2.
> >
> > hurrah!
> >
> > > Please see the Bugzilla page http://bugs.libre-riscv.org/sh
> ow_bug.cgi?id=181
> > > for the attached doc which has elaborate setup instructions.
> >
> > excellent. we now need to decide which lists we want (names). i
> > suggest "retiring" the libre-riscv-dev list. also, it doesn't
>
> You said we need archived http urls of libre-riscv-dev lists to save google
> search engine hits we have already. What about it?
well... if we retire libre-riscv-dev that doesnt matter. just leave it
as-is. however bugzilla etc it will matter.
> > entirely make sense to have two archives, one pipermail, one
> > public-inbox. i'm advocating we just have public-inbox web archives
> > as it's clear that they're clonable.
> >
>
> Hey, but pipermail archives are nice to browse.
i know.
except mirroring them and replying is a pain.
i wonder if pba is easy to improve the look, certainly not black on white
>
> In that case we should set mailman settings:
> DEFAULT_ARCHIVE_PRIVATE = 1
> # So no public viewable archive dir is made
> ARCHIVE_TO_MBOX = 1
> # do not use builtin mailman html archiving, archive to mbox only
>
> Mbox archives are backable, while maildir is difficult to backup and
> restore.
i'm not worrying about mbox / maildir diffs, people can always run
converters between the two
>
> > i *really* need offline mail, i will be using a 3G/4G/WIFI router for
> > the next... err... N months.
> >
>
> The public-inbox Xapian search is good.
yes i installed it.
>
> And raw power of linux command line search tools with plains files
> available
> in git will be great.
the git repos for pba seem to store in base64, unless that is V1.
what is nice is all of pba can be stored and installed *locally*. the
entire lot.
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 184] new mailing lists proposal for libre-soc.org
From: bugzilla-daemon @ 2020-03-21 15:47 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-184-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=184
--- Comment #5 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Jacob Lifshay from comment #4)
> (In reply to Luke Kenneth Casson Leighton from comment #0)
> > These should be *new* lists on libre-soc.org (retire libre-riscv-dev)
> >
> > 1) Libre-SOC-Dev, for hardware and software development (productive)
> > discussion
> > 2) Libre-SOC-Org, for discussion about related to the organization
> > itself, such as funding, investing, finances, domain change, website
> > design, members, ethics, business, administration, you get my point
> > 3) Libre-SOC-Misc, for everything else (the non serious stuff that I
> > can safely ignore unlike the first two which should have important
> > stuff only)
> > 4) Bugtracker separate list, readonly announce only, and replyto set to
> > bugzilla ready for when/if incoming bugzilla is ever set up
>
> I think the above proposal is a good idea. When we finally shut-down
> libre-riscv-dev, I think we should send out a single notification email to
> everyone who's subscribed even if they have mail delivery turned off.
i can do that.
i'd also like to just "migrate" everyone currently on the dev list to... the
dev list and let them decide from there
> We
> should also announce our name & domain change on crowdsupply.
yes.
>
> Is everyone ok with the above plan?
fine with me.
--
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 186] Create decoder for SOC: Power ISA and RISC-V
From: bugzilla-daemon @ 2020-03-21 15:44 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-186-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=186
--- Comment #93 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
y'know what, michael... you see in soc experiments comp6600 there is the
beginnings of a Simulator?
i have an idea, it might be worthwhile to run with that to its logical obvious
implication, namely to actually have a simulator, in soc, in python, which
interprets the InternalOps.
we then have a reference implementation which we can document properly etc.
with links to the PDF.
also it means we have to write the code and so understsnd it.
speed is not crucial as the nmigen simulator will be screamingly slow. we can
use cocotb later.
also at some point we can drop-in power-gem5.
what does everyone think?
--
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 257] Implement demo Load/Store queueing algorithm
From: bugzilla-daemon @ 2020-03-21 15:36 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-257-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=257
--- Comment #13 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Luke Kenneth Casson Leighton from comment #12)
> hmmm connected to LDSTCompAlu we need something that, after splitting the
> address and len into two, can "reconstruct" the LD/ST
this basically adds 16 to the real address and masks off the top bytes that
overrun a 16 byte cache line.
there needs to be two latches which are in "passthrough" mode (in case the twin
data is available immediately) and if one is not available, wait until it is
before responding "ok".
also needs to pass through exceptions back to the LDSTCompAlu
the interesting thing is, the interface is almost certainly identical to the
minerva LoadStoreInterface class.
--
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 263] New: LD/ST batching needed
From: bugzilla-daemon @ 2020-03-21 15:30 UTC (permalink / raw)
To: libre-riscv-dev
http://bugs.libre-riscv.org/show_bug.cgi?id=263
Bug ID: 263
Summary: LD/ST batching needed
Product: Libre-SOC's first SoC
Version: unspecified
Hardware: Other
OS: Linux
Status: CONFIRMED
Severity: enhancement
Priority: ---
Component: Source Code
Assignee: lkcl@lkcl.net
Reporter: lkcl@lkcl.net
CC: libre-riscv-dev@lists.libre-riscv.org
NLnet milestone: ---
http://bugs.libre-riscv.org/show_bug.cgi?id=257#c12
during VL decode phase "batching" is needed. care to be taken to ensure that
failed LD/ST at the end does not wipe out all LD/STs in the same batch.
--
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 257] Implement demo Load/Store queueing algorithm
From: bugzilla-daemon @ 2020-03-21 15:28 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-257-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=257
--- Comment #12 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
hmmm connected to LDSTCompAlu we need something that, after splitting the
address and len into two, can "reconstruct" the LD/ST
i'm also going to suggest that we have a preanalysis phase on LD/STs for VL and
make it SIMD-like, up to 64 bit.
so if a VL=4 LD comes in on a halfword operation it is turned into a 64 bit
request.
the only thing to watch out for there is mem faults. the last 16 bits failing
should *not* cause the other 3 16 bit LDs in that "batch" to fail.
we can incrementally add this as an optimisation, later
--
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 212] test power isa decoder against gnu/llvm assemblers/disassemblers
From: bugzilla-daemon @ 2020-03-21 11:49 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-212-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=212
Luke Kenneth Casson Leighton <lkcl@lkcl.net> changed:
What |Removed |Added
----------------------------------------------------------------------------
See Also| |http://bugs.libre-riscv.org
| |/show_bug.cgi?id=186
--- Comment #1 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(continuing from
--
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] powerpc endian modes
From: Luke Kenneth Casson Leighton @ 2020-03-21 11:24 UTC (permalink / raw)
To: Libre-RISCV General Development
https://en.wikipedia.org/wiki/PowerPC#Endian_modes
moo??
_______________________________________________
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 181] test and install public-inbox
From: bugzilla-daemon @ 2020-03-21 7:01 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-181-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=181
--- Comment #58 from vklr@vkten.in <vklr@vkten.in> ---
(In reply to Luke Kenneth Casson Leighton from comment #57)
> (In reply to vklr@vkten.in from comment #55)
> > (In reply to Luke Kenneth Casson Leighton from comment #49)
> > > (In reply to vklr@vkten.in from comment #48)
> >
> > location /publicinbox/ {
> > access_log /var/log/nginx/public-inbox-httpd_access.log;
> > error_log /var/log/nginx/public-inbox-httpd_error.log;
> > proxy_pass https://127.0.0.1:8080/;
> > proxy_buffering off;
> > }
>
> this fails not in nginx but in public-inbox-httpd. the URL is passed in
> full including the word "publicinbox" and it cannot cope, it starts looking
> for a list named "publicinbox".
>
> a rewrite rule or double proxypass is needed.
>
> this is why i went with the domain name
>
> > Also please install libplack-middleware-reverseproxy-perl
> > It's needed with nginx.
>
> i have that.
I tested locally it is working. Please be careful with location "/" endings.
Anyway if this feature is not needed then?
--
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 181] test and install public-inbox
From: bugzilla-daemon @ 2020-03-21 4:18 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-181-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=181
--- Comment #57 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to vklr@vkten.in from comment #55)
> (In reply to Luke Kenneth Casson Leighton from comment #49)
> > (In reply to vklr@vkten.in from comment #48)
> >
> > > > ok just for a laugh i'm going to run public-inbox-httpd, which is
> > > > now set up on http://inbox.libre-riscv.org/libre-riscv-dev
> > > So you are using subdomains by adding A records in DNS.
> >
> > yes. i found that proxy redirect would send to the wrong
> > path. i don't know why.
> >
> > > You can also use nginx reverse proxy so that
> > > http://libre-riscv.org:8080/pi/libre-riscv-dev becomes some path as
> > > http://libre-riscv.org:80/pi/libre-riscv-dev
> >
> > i had to use this:
> >
> > location /{
> > access_log /var/log/nginx/public-inbox-httpd_access.log;
> > error_log /var/log/nginx/public-inbox-httpd_error.log;
> > proxy_set_header HOST $host;
> > proxy_set_header X-Real-IP $remote_addr;
> > proxy_set_header X-Forwarded-Proto $scheme;
> > proxy_pass http://127.0.0.1:8080$request_uri;
> > proxy_buffering off;
> > }
> >
> > efforts to change it to work as location /publicinbox met with...
> > resistance :)
>
> please try this:
>
> location /publicinbox/ {
> access_log /var/log/nginx/public-inbox-httpd_access.log;
> error_log /var/log/nginx/public-inbox-httpd_error.log;
> proxy_pass https://127.0.0.1:8080/;
> proxy_buffering off;
> }
this fails not in nginx but in public-inbox-httpd. the URL is passed in full
including the word "publicinbox" and it cannot cope, it starts looking for a
list named "publicinbox".
a rewrite rule or double proxypass is needed.
this is why i went with the domain name
> Also please install libplack-middleware-reverseproxy-perl
> It's needed with nginx.
i have 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
* Re: [libre-riscv-dev] New formal verification checking tool
From: Luke Kenneth Casson Leighton @ 2020-03-21 4:13 UTC (permalink / raw)
To: Libre-RISCV General Development
In-Reply-To: <CAEz=somJ3zNUJkCPhjGpsu7KFnGUcSUgzKr4bvNuVch7KmMd4A@mail.gmail.com>
On Saturday, March 21, 2020, Samuel Falvo II <sam.falvo@gmail.com> wrote:
> I saw a tweet on my timeline which announced a new test coverage analysis
> tool which uses Yosys' support for formal methods to calculate your test
> coverage.
interesting. appreciated
--
---
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 257] Implement demo Load/Store queueing algorithm
From: bugzilla-daemon @ 2020-03-21 4:12 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-257-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=257
--- Comment #11 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
(In reply to Jacob Lifshay from comment #10)
> I've been thinking that we would have a cache with a cache line size of 64
> bytes (or 32 if we really have to).
eek. ok. right. 64 / 4 = 16 that gives 4 stripes to 1R1W Memory inside the
cache. that way we do not need multi ported SRAM in the cache.
if we have byte enable lines the 16 bit mask can be passed to each byte.
> we want the cache line size to be the same for all our caches since it
> greatly simplifies cache to cache transfers.
ok
> So, for a 32kB L1 cache, it would have 512 lines.
ye gods.
> If we had it be 8-way set
> associative, then there would be 64 sets and each set would have 8 lines.
4 way makes more sense (a little easier) as the byte enable mask is 16 bits.
however 8 way does as well because each 16 byte enable mask is split in two.
> Having a L1 cache much smaller than 16kB would be quite detrimental due to
> the excessive cache latency and lack of bandwidth to the L2.
yes.
> I'm going to be unavailable till at least Sat night, then will work on
> publishing v0.2 of algebraics (updating transitive dependencies for
> simple-soft-float due to PyO3 v0.9 being released) then work on this.
ok.
the earlier idea of a tiny pre L1 cache may on reflection be unnecessary.
i drew out a diagram a couple days ago, it basically involved:
* a PriorityPicker to find the first entry in the AddressMatrix that is "live"
* that then says which entry is to be used to broadcast the high address bits
(14 to 48) to all other live LD/STs
* bear in mind all "live" addresses have had to calculate a comparison on bits
4 thru 13 of the address, we also include *these* (ANDed) in the comparison
* anything that succeeds on matching bits 4 to 13 from the AddressMatcher
matrix *and* gives a hit on the Priority Picked bits 14 to 48, its 16bit "mask"
is permitted to go through an ORer to the L1 cache byte-enable on the Memory.
a refinement of that is to allow up to *four* of those to be picked, one to be
routed to each "way".
that is one of the things that i wrote the MultiPriorityPicker for.
--
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] New formal verification checking tool
From: Samuel Falvo II @ 2020-03-21 3:20 UTC (permalink / raw)
To: Libre-RISCV General Development
I saw a tweet on my timeline which announced a new test coverage analysis
tool which uses Yosys' support for formal methods to calculate your test
coverage.
I don't know how applicable this would be to this project, but thought I'd
pass it along in the hopes that it'd be useful.
https://www.youtube.com/watch?v=NKzqRum1ksg&feature=emb_title
--
Samuel A. Falvo II
_______________________________________________
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 257] Implement demo Load/Store queueing algorithm
From: bugzilla-daemon @ 2020-03-21 1:56 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-257-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=257
--- Comment #10 from Jacob Lifshay <programmerjake@gmail.com> ---
I've been thinking that we would have a cache with a cache line size of 64
bytes (or 32 if we really have to).
we want the cache line size to be the same for all our caches since it greatly
simplifies cache to cache transfers.
So, for a 32kB L1 cache, it would have 512 lines. If we had it be 8-way set
associative, then there would be 64 sets and each set would have 8 lines.
Having a L1 cache much smaller than 16kB would be quite detrimental due to the
excessive cache latency and lack of bandwidth to the L2.
I'm going to be unavailable till at least Sat night, then will work on
publishing v0.2 of algebraics (updating transitive dependencies for
simple-soft-float due to PyO3 v0.9 being released) then work on this.
--
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 181] test and install public-inbox
From: bugzilla-daemon @ 2020-03-21 1:14 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-181-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=181
--- Comment #56 from vklr@vkten.in <vklr@vkten.in> ---
> please try this:
>
> location /publicinbox/ {
> access_log /var/log/nginx/public-inbox-httpd_access.log;
> error_log /var/log/nginx/public-inbox-httpd_error.log;
> proxy_pass https://127.0.0.1:8080/;
> proxy_buffering off;
> }
>
> Also please install libplack-middleware-reverseproxy-perl
> It's needed with nginx.
this location directive within server directive for http://libre-riscv.org/
--
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 181] test and install public-inbox
From: bugzilla-daemon @ 2020-03-21 1:13 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-181-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=181
--- Comment #55 from vklr@vkten.in <vklr@vkten.in> ---
(In reply to Luke Kenneth Casson Leighton from comment #49)
> (In reply to vklr@vkten.in from comment #48)
>
> > > ok just for a laugh i'm going to run public-inbox-httpd, which is
> > > now set up on http://inbox.libre-riscv.org/libre-riscv-dev
> > So you are using subdomains by adding A records in DNS.
>
> yes. i found that proxy redirect would send to the wrong
> path. i don't know why.
>
> > You can also use nginx reverse proxy so that
> > http://libre-riscv.org:8080/pi/libre-riscv-dev becomes some path as
> > http://libre-riscv.org:80/pi/libre-riscv-dev
>
> i had to use this:
>
> location /{
> access_log /var/log/nginx/public-inbox-httpd_access.log;
> error_log /var/log/nginx/public-inbox-httpd_error.log;
> proxy_set_header HOST $host;
> proxy_set_header X-Real-IP $remote_addr;
> proxy_set_header X-Forwarded-Proto $scheme;
> proxy_pass http://127.0.0.1:8080$request_uri;
> proxy_buffering off;
> }
>
> efforts to change it to work as location /publicinbox met with...
> resistance :)
please try this:
location /publicinbox/ {
access_log /var/log/nginx/public-inbox-httpd_access.log;
error_log /var/log/nginx/public-inbox-httpd_error.log;
proxy_pass https://127.0.0.1:8080/;
proxy_buffering off;
}
Also please install libplack-middleware-reverseproxy-perl
It's needed with nginx.
--
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 257] Implement demo Load/Store queueing algorithm
From: bugzilla-daemon @ 2020-03-21 0:48 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-257-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=257
--- Comment #9 from Luke Kenneth Casson Leighton <lkcl@lkcl.net> ---
hang on a minute...
if we use a very small but wide L1 cache, with the following characteristics,
does it do the job?
* 8 sets
* 16 way set associative
* byte per way
the 16 bit mask *is* the 16 way selector
then i think, if you make a N-in/out multiplexor to write to each byte, the
job's done.
errr...
--
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 181] test and install public-inbox
From: bugzilla-daemon @ 2020-03-21 0:28 UTC (permalink / raw)
To: libre-riscv-dev
In-Reply-To: <bug-181-13@http.bugs.libre-riscv.org/>
http://bugs.libre-riscv.org/show_bug.cgi?id=181
--- Comment #54 from vklr@vkten.in <vklr@vkten.in> ---
Forgot to add.
Please edit: test1.git/description
So that nice description is shown by public-inbox-httpd
--
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