Commit Graph

49 Commits

Author SHA1 Message Date
gatecat 9b51c6e337 clangformat
Signed-off-by: gatecat <gatecat@ds0.me>
2024-09-30 14:51:33 +02:00
gatecat efb58711b0 ecp5: Split the SLICE bel into separate LUT/FF/RAMW bels 2022-04-07 18:02:36 +01:00
gatecat f36188f2e1 ecp5: LUT permutation support
Signed-off-by: gatecat <gatecat@ds0.me>
2021-12-13 20:22:06 +00:00
Matt Johnston 80dd442412 ecp5: Use a vector rather than dict
This improves router1 performance vs the default dict
Using it for wire2net, pip2net, wire_fanout
2021-12-12 22:09:11 +08:00
gatecat 2ffb081442 Fixing old emails and names in copyrights
Signed-off-by: gatecat <gatecat@ds0.me>
2021-06-12 13:22:38 +01:00
gatecat dcbb322447 Remove redundant code after hashlib move
Signed-off-by: gatecat <gatecat@ds0.me>
2021-06-02 15:05:20 +01:00
gatecat ff72454f83 Add hash() member functions
Signed-off-by: gatecat <gatecat@ds0.me>
2021-06-02 14:27:56 +01:00
gatecat c82df9e40d ecp5: Use new cluster API
Signed-off-by: gatecat <gatecat@ds0.me>
2021-05-06 11:59:58 +01:00
Keith Rothman fe4608386e Split nextpnr.h to allow for linear inclusion.
"nextpnr.h" is no longer the god header.  Important improvements:

 - Functions in log.h can be used without including
   BaseCtx/Arch/Context. This means that log_X functions can be called
   without included "nextpnr.h"

 - NPNR_ASSERT can be used without including "nextpnr.h" by including
   "nextpnr_assertions.h".  This allows NPNR_ASSERT to be used safely in
   any header file.

 - Types defined in "archdefs.h" are now available without including
   BaseCtx/Arch/Context.  This means that utility classes that will be
   used inside of BaseCtx/Arch/Context can be defined safely in a
   self-contained header.

Signed-off-by: Keith Rothman <537074+litghost@users.noreply.github.com>
2021-03-15 09:05:23 -07:00
gatecat 23413a4d12 Fix compiler warnings introduced by -Wextra
Signed-off-by: gatecat <gatecat@ds0.me>
2021-02-25 15:15:25 +00:00
gatecat 7922b3bfc4 Replace DelayInfo with DelayPair/DelayQuad
This replaces the arch-specific DelayInfo structure with new DelayPair
(min/max only) and DelayQuad (min/max for both rise and fall) structures
that form part of common code.

This further reduces the amount of arch-specific code; and also provides
useful data structures for timing analysis which will need to delay
with pairs/quads of delays as it is improved.

While there may be a small performance cost to arches that didn't
separate the rise/fall cases (arches that aren't currently separating
the min/max cases just need to be fixed...) in DelayInfo, my expectation
is that inlining will mean this doesn't make much difference.

Signed-off-by: gatecat <gatecat@ds0.me>
2021-02-19 11:31:33 +00:00
D. Shah 297cd026b9 Add default implementation of bel bucket functions
Signed-off-by: D. Shah <dave@ds0.me>
2021-02-05 19:19:17 +00:00
Keith Rothman da74a425d2 Run "make clangformat".
Signed-off-by: Keith Rothman <537074+litghost@users.noreply.github.com>
2021-02-02 07:44:49 -08:00
Keith Rothman 9fe546f279 Rename Partition -> BelBucket.
Signed-off-by: Keith Rothman <537074+litghost@users.noreply.github.com>
2021-02-02 07:34:56 -08:00
Keith Rothman d03d9d839b Working compile of ECP5.
Signed-off-by: Keith Rothman <537074+litghost@users.noreply.github.com>
2021-02-02 07:34:56 -08:00
Keith Rothman 71e210dd4b Refactor ECP5 to new Partition API.
Signed-off-by: Keith Rothman <537074+litghost@users.noreply.github.com>
2021-02-02 07:34:56 -08:00
Ross Schlaikjer 0043ae0807
Issue warning for mixed-mode inputs 2020-04-29 14:39:52 -04:00
Ross Schlaikjer 5e763b1afc
Alter MULT18X18D timing db based on register config
If the REG_INPUTA_CLK and REG_INPUTB_CLK values are set, then we should
use the faster setup/hold timings for the 18x8 multiplier.
Similarly, check the value of REG_OUTPUT_CLK for whether or not to use
faster timings for the output.

This is based on how I currently understand the registers to work - if
anyone knows the actual rules for when each timing applies please do
chime in to correct this implementation if necessary.

Along the same lines, this PR does not address the case when the
pipeline registers are enabled, since it is not clear to me how exactly
that affects the timing.
2020-04-28 20:01:29 -04:00
Ross Schlaikjer 3257bdc8a1
Actually just move all the logic to ArchInfo 2020-04-07 14:11:49 -04:00
Ross Schlaikjer 0bdf1e05f1
Extract regmode configuration to ArchInfo 2020-04-07 14:03:55 -04:00
Miodrag Milanovic 796d648995 Merge remote-tracking branch 'origin/master' into mmicko/ecp5_gui 2019-12-28 13:54:06 +01:00
David Shah b6e2159cec Work around Qt MOC issue with IdString enums
Signed-off-by: David Shah <dave@ds0.me>
2019-12-27 10:32:23 +00:00
David Shah 475fcd4425 ecp5: Add an error for out-of-sync constids and bba
Signed-off-by: David Shah <dave@ds0.me>
2019-10-26 20:38:28 +01:00
Miodrag Milanovic 847910d986 type needs to be part of hash for GroupId 2019-10-20 10:03:37 +02:00
Miodrag Milanovic bfbb6dbf69 Draw swbox, smaller slices, proper io 2019-10-20 09:41:30 +02:00
David Shah d04e5954a6 ecp5: Adding support for 36-bit wide PDP RAMs
Signed-off-by: David Shah <dave@ds0.me>
2019-10-01 12:01:33 +01:00
David Shah 998d055ea7 ecp5: Speed up timing analysis
Signed-off-by: David Shah <dave@ds0.me>
2019-02-25 11:49:25 +00:00
David Shah 458aa20161 ecp5: More optimal LUT6 placement
Signed-off-by: David Shah <dave@ds0.me>
2018-11-16 17:36:34 +00:00
David Shah 18813f2056 ecp5: Adding real timing data to database
Signed-off-by: David Shah <dave@ds0.me>
2018-11-16 13:26:28 +00:00
David Shah 7e1df82462 ecp5: Regression fix & format
Signed-off-by: David Shah <dave@ds0.me>
2018-11-15 11:54:28 +00:00
Clifford Wolf 06e0e1ffee Various router1 fixes, Add BelId/WireId/PipId::operator<()
Signed-off-by: Clifford Wolf <clifford@clifford.at>
2018-11-13 05:05:56 +01:00
David Shah 6a1b49c311 ecp5: Improve mixed no-FF/FF placement
Signed-off-by: David Shah <davey1576@gmail.com>
2018-09-30 18:39:53 +01:00
David Shah 0e0ad26f07 ecp5: Use ArchNetInfo to mark global nets to ignore
Signed-off-by: David Shah <davey1576@gmail.com>
2018-09-29 19:31:49 +01:00
David Shah b8206d71ca ecp5: Speedup placement using ArchCellInfo
Signed-off-by: David Shah <davey1576@gmail.com>
2018-08-18 18:14:18 +01:00
David Shah a3ae3f9791 ecp5: Update to use const IdStrings in place of PortPin/BelType
Signed-off-by: David Shah <davey1576@gmail.com>
2018-08-08 19:08:43 +02:00
Clifford Wolf 29dd98420b Remove getFrameDecal() API
Signed-off-by: Clifford Wolf <clifford@clifford.at>
2018-08-01 11:30:11 +02:00
David Shah f3127f7dfd ecp5: Add Bel graphics
Signed-off-by: David Shah <davey1576@gmail.com>
2018-07-31 14:39:37 +02:00
Clifford Wolf 78f40ca0af Change DelayInfo semantics to what we actually need
Signed-off-by: Clifford Wolf <clifford@clifford.at>
2018-07-21 13:52:59 +02:00
David Shah f138368e34 ecp5: Add simple constant packer
Signed-off-by: David Shah <davey1576@gmail.com>
2018-07-18 09:35:40 +02:00
Clifford Wolf ddfc535df7 Add ArchNetInfo and ArchCellInfo
Signed-off-by: Clifford Wolf <clifford@clifford.at>
2018-07-17 20:04:49 +02:00
David Shah e9668ed618 Fixing hash link problem
Signed-off-by: David Shah <davey1576@gmail.com>
2018-07-13 12:42:04 +02:00
Clifford Wolf b8a42ff53b Updates from clang-format
Signed-off-by: Clifford Wolf <clifford@clifford.at>
2018-07-12 22:04:13 +02:00
Clifford Wolf a436035424 Add Groups API
Signed-off-by: Clifford Wolf <clifford@clifford.at>
2018-07-12 17:22:29 +02:00
David Shah 35216298d5 ecp5: Update arch to use new graphics API
Signed-off-by: David Shah <davey1576@gmail.com>
2018-07-11 14:27:15 +02:00
David Shah 59cb1600d9 ecp5: Fixing arch bugs
Signed-off-by: David Shah <davey1576@gmail.com>
2018-07-11 10:42:09 +02:00
David Shah 83303bae5a ecp5: Implementing (at least stubs) most of arch.cc
Signed-off-by: David Shah <davey1576@gmail.com>
2018-07-11 10:41:36 +02:00
David Shah 7862d1b84b ecp5: Implementing core arch.h functions
Signed-off-by: David Shah <davey1576@gmail.com>
2018-07-11 10:41:36 +02:00
David Shah c4af52dd5b ecp5: Working on arch implementation
Signed-off-by: David Shah <davey1576@gmail.com>
2018-07-11 10:41:36 +02:00
David Shah eaae6b7dca ecp5: Begin planning data structures
Signed-off-by: David Shah <davey1576@gmail.com>
2018-07-11 10:40:16 +02:00