00:00
tpb has joined #symbiflow
02:47
Bertl_oO is now known as Bertl_zZ
04:28
citypw has joined #symbiflow
04:35
alainmarcel has joined #symbiflow
04:40
alainmarcel has left #symbiflow [#symbiflow]
05:23
citypw has quit [Remote host closed the connection]
06:15
freemint has quit [Ping timeout: 245 seconds]
06:40
freemint has joined #symbiflow
06:45
freemint has quit [Ping timeout: 250 seconds]
07:49
Bertl_zZ is now known as Bertl
08:24
freemint has joined #symbiflow
08:28
freemint has quit [Ping timeout: 245 seconds]
08:38
freemint has joined #symbiflow
08:42
freemint has quit [Ping timeout: 250 seconds]
09:43
freemint has joined #symbiflow
09:48
freemint has quit [Ping timeout: 264 seconds]
09:51
craigo has joined #symbiflow
09:53
freemint has joined #symbiflow
10:00
adjtm has quit [Ping timeout: 240 seconds]
10:19
freemint has quit [Remote host closed the connection]
10:19
freemint has joined #symbiflow
10:29
Vonter has joined #symbiflow
10:42
adjtm has joined #symbiflow
11:27
freemint has quit [Ping timeout: 245 seconds]
12:28
freemint has joined #symbiflow
12:55
citypw has joined #symbiflow
13:31
mankeli has joined #symbiflow
14:59
freemint has quit [Ping timeout: 250 seconds]
15:02
<
tpb >
Title: Pin libxml2 to version with fix. by litghost · Pull Request #1077 · SymbiFlow/symbiflow-arch-defs · GitHub (at github.com)
15:03
<
litghost >
How did you test it?
15:03
<
sf-slack >
<mkurc> Removed the build directory and started from scratch
15:03
<
tpb >
Title: conda-packages/libxml2 at master · SymbiFlow/conda-packages · GitHub (at github.com)
15:05
<
litghost >
That's the idea
15:06
<
litghost >
What error did you get?
15:06
freemint has joined #symbiflow
15:06
<
sf-slack >
<mkurc> `lxml.etree.SerialisationError: unknown error -296584060`
15:06
<
sf-slack >
<mkurc> during routing import
15:07
<
sf-slack >
<mkurc> After manually re-building and installing according to #1066 it worked correctly
15:08
<
sf-slack >
<mkurc> And this is the same error I encountered for libxml before #1077 when trying full xc7a50t graph the first time
15:08
<
litghost >
Did you run "all_conda" after cleaning the build dir? This could be a cmake dependency error
15:09
<
sf-slack >
<mkurc> No, I didn't. But when running single threaded make there are no issues
15:09
<
sf-slack >
<mkurc> And then I verified version installed
15:10
<
sf-slack >
<mkurc> and it was correct, as specified in the cmake file: 2.9.9 h14c3975_5
15:54
<
litghost >
mkurc: Can you run:
15:54
<
litghost >
find -name libxml2.so.2 | xargs sha256sum
15:54
<
litghost >
cd <conda dir>
15:55
<
litghost >
find -name etree.cpython*.so | xargs ldd
16:05
somlo has quit [Ping timeout: 250 seconds]
16:06
<
sf-slack >
<acomodi> @litghost BUFH passthrough seems to be working. I have successfully routed the `simple_ff_x1y0` without BUFH instantiations. Now I get an error during `print_route` for which an `
16:07
<
sf-slack >
<acomodi> ClusteredBlock is null for a specific location. I need to debug it. One thing I didn't do actually is remove the BUFH sites and tiles from the architecture file... could this influence VPR behavior?
16:11
<
sf-slack >
<acomodi> @litghost I have also force pushed on master+wip the new version with a working `wip/equivalent_sites` branch. How to create the name for the new tag? I see that now is sth like v8.0.0_rc1_XXXX_commit-hash: what should go in XXXX?
16:17
somlo has joined #symbiflow
16:20
citypw has quit [Ping timeout: 276 seconds]
16:34
freemint has quit [Ping timeout: 268 seconds]
16:36
<
litghost >
I have been tagging conda builds with the output of 'git describe'
16:37
freemint has joined #symbiflow
16:42
freemint has quit [Ping timeout: 268 seconds]
16:45
<
mithro >
litghost: Did the prjxray-db update look okay?
16:50
<
litghost >
Ya, they look okay
16:50
<
litghost >
I'll test the integrated symbiflow later, I'm trying to debug a VTR linked list bug :( :(
17:08
freemint has joined #symbiflow
17:26
adjtm has quit [Ping timeout: 240 seconds]
17:38
<
sf-slack >
<mkurc> @litghost: I've started again with freshly cloned repo, just to be sure.
17:38
<
sf-slack >
<mkurc> @litghost: c7283cbab4a165b891c0156ef618d9d151126b967502e6d0eca1ed5acc76b026 ./libxml2.so.2
17:39
<
litghost >
On my system there are two lxml instances, do yours have the same output?
17:39
freemint has quit [Ping timeout: 245 seconds]
17:40
<
litghost >
Also please run "sha256sum /home/mkurc/Repos/google-symbiflow-arch-defs-2/build/env/conda/lib/./python3.7/site-packages/lxml/../../../libxml2.so.2"
17:42
<
sf-slack >
<mkurc> c7283cbab4a165b891c0156ef618d9d151126b967502e6d0eca1ed5acc76b026
17:44
freemint has joined #symbiflow
17:44
<
sf-slack >
<mkurc> Well, actually only the one inside conda env should be relevant.
17:45
<
sf-slack >
<mkurc> When I run `conda list` it reports `libxml2 2.9.9 h14c3975_5 symbiflow`
17:45
<
litghost >
Can you check if you have an enviroment variable "PYTHON3" defined?
17:46
<
sf-slack >
<mkurc> outside conda env no.
17:46
<
litghost >
/home/mkurc/Repos/google-symbiflow-arch-defs-2/build/env/conda/bin/python3 repo_lxml.py
17:47
<
litghost >
Do you have PYTHONPATH defined?
17:49
<
litghost >
Also run:
17:49
<
litghost >
conda/bin/python3 -c "import lxml; print(lxml.__file__)"
17:49
<
litghost >
conda/bin/python3 -c "import lxml.etree; print(lxml.etree.__file__)"
17:51
<
sf-slack >
<mkurc> The two last commands output paths pointing to files inside the conda env:
17:51
<
sf-slack >
<mkurc> `build/env/conda/lib/python3.7/site-packages/lxml/__init__.py`
17:52
<
sf-slack >
<mkurc> and `/build/env/conda/lib/python3.7/site-packages/lxml/etree.cpython-37m-x86_64-linux-gnu.so`
17:52
<
sf-slack >
<mkurc> no PYTHONPATH outside conda env
17:52
<
litghost >
Try to see if "conda/bin/python3 repo_lxml.py" works or fails
17:52
<
litghost >
It will take a couple minutes (~10 on my machine)
17:52
<
sf-slack >
<mkurc> I'm already doing it
17:55
<
sf-slack >
<mkurc> @litghost: Ok, the repo_lxml.py test completed without errors.
17:55
<
litghost >
Ok, then the libxml2 in the tree has the fix
17:55
<
sf-slack >
<mkurc> I guess I must have had something wrong with the conda env
17:55
<
litghost >
Not neccarility
17:56
<
litghost >
Run "make VERBOSE=1 file_xc7_archs_artix7_channels.db"
17:56
<
litghost >
After removing build/xc7/archs/artix7/channels.db
17:56
<
litghost >
See what python executable is being invoked
17:58
adjtm has joined #symbiflow
18:01
<
tpb >
Title: Travis CI - Test and Deploy with Confidence (at travis-ci.com)
18:03
<
sf-slack >
<mkurc> @litghost Ok, this time routing import completed without any problems.
18:04
<
litghost >
Unclear what happened before
18:04
<
sf-slack >
<mkurc> Yep, I have no clue
18:14
<
mithro >
BTW If you are having trouble with getting kmurray to do a review, you can also ask Vaughn
18:15
<
sf-slack >
<acomodi> litghost: Yep, seen the failure now, I'll update it asap
18:32
<
mithro >
It also looks like Verilator is getting support for Python based models...
18:39
<
tpb >
Title: Python models by patstew · Pull Request #3 · verilator/verilator · GitHub (at github.com)
18:40
<
mithro >
Also other things I have discovered
18:40
<
tpb >
Title: GitHub - eirikpre/VSCode-SystemVerilog: SystemVerilog support in VS Code (at github.com)
18:40
<
tpb >
Title: GitHub - patstew/linter-verilator (at github.com)
18:40
<
tpb >
Title: GitHub - manucorporat/linter-verilog: Atom linter for Verilog, using icarus verilog. (at github.com)
20:54
<
hackerfoo >
418de8da
20:54
freemint has quit [Ping timeout: 245 seconds]
20:56
freemint has joined #symbiflow
20:57
Bertl is now known as Bertl_zZ
21:12
Vonter has quit [Ping timeout: 240 seconds]
21:27
Vonter has joined #symbiflow
23:59
tpb has quit [Remote host closed the connection]