Miodrag Milanović
6ae143685a
Merge pull request #166 from ajeakins/master
...
Fix crash starting the GUI on macOS where we must request a core profile
2018-12-05 19:04:36 +01:00
David Shah
d298687dc2
ice40: Fix carry chain splitting
...
Signed-off-by: David Shah <dave@ds0.me>
2018-12-05 10:12:23 +00:00
Adrian Jeakins
92ddef9fc3
Fix crash starting the GUI on macOS where we must request a core profile.
...
See http://doc.qt.io/qt-5/qabstractopenglfunctions.html
2018-12-04 23:05:06 +00:00
Sylvain Munaut
99e1b6db47
build: Make use of the pipe option to avoid temporary files
...
This is really useful when building the ice40 with the gigantic .cc
files that generate multi gigabyte .s temporary files ... this way the
assembler just processed it in streaming way.
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
2018-12-04 20:20:50 +01:00
David Shah
51cda136b1
ice40: Don't split carry chain in simple feed-out cases
...
Signed-off-by: David Shah <dave@ds0.me>
2018-12-04 12:31:32 +00:00
David Shah
0c93b55650
ice40: Include I3 connectivity in chain
...
Thanks @smunaut
Signed-off-by: David Shah <dave@ds0.me>
2018-12-04 12:02:26 +00:00
David Shah
12aca1558f
Merge pull request #162 from whitequark/reset-fanout
...
ice40: add reset global promotion threshold
2018-12-04 10:14:12 +00:00
whitequark
7fad6058bd
ice40: add reset global promotion threshold.
2018-12-04 07:40:55 +00:00
Eddie Hung
20f0353f76
Add report to attosoc_tb.vhd
2018-12-02 15:41:30 -08:00
Eddie Hung
b1b8183967
Re-enable PLL in attosoc.v
2018-12-02 15:41:03 -08:00
Eddie Hung
8f0e888815
nextpnr now writes to log, netgen to overwrite
2018-12-02 15:40:20 -08:00
Eddie Hung
5aff7bbbc4
Fix INIT of pass-thru LUT to be "2" not "1"
2018-12-02 15:32:29 -08:00
David Shah
6e728c9a9b
Merge pull request #160 from dmsc/sb_ledda_ip
...
ice40: Add support for placing SB_LEDDA_IP block.
2018-12-02 08:48:58 +00:00
Daniel Serpell
d4b3c1d819
ice40: Add support for placing SB_LEDDA_IP block.
...
Signed-off-by: Daniel Serpell <daniel.serpell@gmail.com>
2018-12-01 22:27:04 -03:00
David Shah
dc549cd56b
Merge pull request #159 from YosysHQ/ecp5_pllplace
...
ecp5: Pre-place PLLs and use dedicated routes into globals
2018-12-01 09:14:34 +00:00
Eddie Hung
5ddfc32c75
Add attosoc.sh and attosoc_tb.vhd
2018-11-30 17:06:55 -08:00
Eddie Hung
4574a57efc
Add attosoc
2018-11-30 15:24:32 -08:00
Eddie Hung
fdca3d6d77
firmware.hex with delay loop
2018-11-30 15:23:09 -08:00
David Shah
5ddf99cf5d
ecp5: Pre-place PLLs and use dedicated routes into globals
...
Signed-off-by: David Shah <dave@ds0.me>
2018-11-30 16:09:56 +00:00
Eddie Hung
cac7ce2747
Cleanup
2018-11-29 17:20:51 -08:00
Eddie Hung
f4e7f4e690
Use wholesale attrs from ISE for MMCME2_ADV
2018-11-29 17:20:35 -08:00
Eddie Hung
2fdf937259
Assignment LUT inputs from fastest down
2018-11-29 17:08:05 -08:00
Eddie Hung
0327fa554a
Revert "Ahead of LUT input swapping, assign LUT<6 from A6 downwards"
...
This reverts commit ec96897c1d
.
2018-11-29 16:54:33 -08:00
Eddie Hung
c5165f7830
Duplicate arcs.clear()
2018-11-29 16:32:33 -08:00
Eddie Hung
d7dd945f55
Overwrite COMPENSATION attribute on MMCME2_ADV to "INTERNAL"
2018-11-29 16:32:08 -08:00
Eddie Hung
d8b6b231de
Move required attributes to pack
2018-11-29 15:38:28 -08:00
Eddie Hung
6985e80c01
Merge branch 'xc7' of gitlab.com:eddiehung/nextpnr into xc7
2018-11-29 13:32:32 -08:00
Miodrag Milanovic
535fc953d4
Use site x location to determine if it is one block or other
2018-11-29 12:44:02 -08:00
Eddie Hung
9f03d9eed3
Add PLL to bring 125MHz clock to 60MHz for picorv32
2018-11-29 12:25:39 -08:00
David Shah
58e9c6f32e
Merge pull request #158 from YosysHQ/improve_error
...
Error reporting improvements
2018-11-29 19:46:05 +00:00
David Shah
8af367ad0a
ice40: Add a warning for unconstrained IO
...
Signed-off-by: David Shah <dave@ds0.me>
2018-11-29 19:35:19 +00:00
David Shah
90138fc120
rulecheck: Improve message printed at start
...
Signed-off-by: David Shah <dave@ds0.me>
2018-11-29 19:28:15 +00:00
David Shah
4e05d09397
Improve reporting of unknown cell types
...
Signed-off-by: David Shah <dave@ds0.me>
2018-11-29 19:26:23 +00:00
David Shah
dbc14ea76d
json: Improve reporting of multiple drivers
...
Signed-off-by: David Shah <dave@ds0.me>
2018-11-29 19:20:51 +00:00
David Shah
fc08856537
Merge pull request #157 from whitequark/fanout-thresh
...
ice40: raise CE global promotion threshold
2018-11-29 09:12:47 +00:00
David Shah
3ff3b0f6b5
Merge pull request #156 from whitequark/fanout
...
ice40: print fanout of nets promoted to globals
2018-11-29 09:08:40 +00:00
Eddie Hung
4161856d49
Add support for MMCME2_ADV
2018-11-28 22:34:22 -08:00
whitequark
db96b88d79
ice40: raise CE global promotion threshold.
2018-11-29 00:12:48 +00:00
whitequark
a974124a7a
ice40: print fanout of nets promoted to globals.
2018-11-28 23:52:48 +00:00
Miodrag Milanovic
bfa2157ae6
compile fix for gui and proper size
2018-11-28 17:59:58 +01:00
David Shah
0872b63b0b
Merge pull request #155 from smunaut/issue_151
...
ice40: Update the way LVDS inputs are handled during bitstream generation
2018-11-28 16:20:36 +00:00
David Shah
48071e3650
Merge pull request #154 from smunaut/issue_141
...
ice40: Complete rework of the way PLLs are placed and validity checks
2018-11-28 16:20:26 +00:00
Miodrag Milanovic
f2fecc3c69
make gui run
2018-11-28 17:04:26 +01:00
Sylvain Munaut
ba958d1792
ice40: Try to be helpful and suggest using PAD PLL instead of CORE
...
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
2018-11-28 16:04:58 +01:00
Sylvain Munaut
a65b12e8d6
ice40: Revamp the whole PLL placement/validity check logic
...
We do a pre-pass on all the PLLs to place them before packing.
To place them:
- First pass with all the PADs PLLs since those can only fit at one
specific BEL depending on the input connection
- Second pass with all the dual outputs CORE PLLs. Those can go
anywhere where there is no conflicts with their A & B outputs and
used IO pins
- Third pass with the single output CORE PLLs. Those have the least
constrains.
During theses passes, we also check the validity of all their connections.
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
2018-11-28 16:04:43 +01:00
Sylvain Munaut
5f0f2b060b
ice40: Update the way LVDS inputs are handled during bitstream generation
...
* Instead of "patching" input_en, we completely separate config for
normal and LVDS pair.
- For normal pair, nothing changes
- For LVDS pairs, the IE/REN bits are always set as if the input buffer
are disabled. Then if input_en was set to 1 (i.e. the input is
actually for something), then we set the IoCtrl.LVDS bit.
- Also for LVDS, if input is used, pullups are forcibly disabled.
* When scanning for unused IOs, never process those part of a LVDS pair.
They will have been configured by the complement
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
2018-11-28 16:04:23 +01:00
David Shah
7a2ef27d6c
Merge pull request #153 from YosysHQ/global-options
...
ice40: Finer-grained control of global promotion
2018-11-28 07:43:00 +00:00
Eddie Hung
13e7798b34
Fix #endif placement
2018-11-27 18:11:19 -08:00
Eddie Hung
212b03999b
Gzip the torc_info data
2018-11-27 18:08:03 -08:00
Eddie Hung
440802bf9d
Add support for serialization of torc_info
2018-11-27 17:55:31 -08:00