riscv-isa-sim.git
5 years agoc_slli c_srai
Luke Kenneth Casson Leighton [Thu, 18 Oct 2018 22:21:18 +0000 (23:21 +0100)]
c_slli c_srai

5 years agoc_lui
Luke Kenneth Casson Leighton [Thu, 18 Oct 2018 22:18:51 +0000 (23:18 +0100)]
c_lui

5 years agoc_jalr
Luke Kenneth Casson Leighton [Thu, 18 Oct 2018 22:17:49 +0000 (23:17 +0100)]
c_jalr

5 years agoc_jal
Luke Kenneth Casson Leighton [Thu, 18 Oct 2018 22:17:04 +0000 (23:17 +0100)]
c_jal

5 years agoc_beqz
Luke Kenneth Casson Leighton [Thu, 18 Oct 2018 22:16:24 +0000 (23:16 +0100)]
c_beqz

5 years agoaddi4spn
Luke Kenneth Casson Leighton [Thu, 18 Oct 2018 22:15:14 +0000 (23:15 +0100)]
addi4spn

5 years agoredirect sreg_t casts through function
Luke Kenneth Casson Leighton [Thu, 18 Oct 2018 22:14:36 +0000 (23:14 +0100)]
redirect sreg_t casts through function

5 years agotypedef on sv_reg_t to reg_t (and signed variant)
Luke Kenneth Casson Leighton [Thu, 18 Oct 2018 22:12:00 +0000 (23:12 +0100)]
typedef on sv_reg_t to reg_t (and signed variant)

still working on redirecting everything through a planned class
that can be polymorphic overloaded... eventually

5 years agouse unsigned long shift on sv csr setting
Luke Kenneth Casson Leighton [Thu, 18 Oct 2018 22:11:08 +0000 (23:11 +0100)]
use unsigned long shift on sv csr setting

5 years agoforgot to set clroffset
Luke Kenneth Casson Leighton [Thu, 18 Oct 2018 17:14:36 +0000 (18:14 +0100)]
forgot to set clroffset

5 years agoallow 4 CSR entries to be set at a time, on RV64
Luke Kenneth Casson Leighton [Wed, 17 Oct 2018 09:56:30 +0000 (10:56 +0100)]
allow 4 CSR entries to be set at a time, on RV64

5 years agominor alteration to CSRRWI SETVL / SETMVL to offset immediate by 1
Luke Kenneth Casson Leighton [Wed, 17 Oct 2018 00:58:15 +0000 (01:58 +0100)]
minor alteration to CSRRWI SETVL / SETMVL to offset immediate by 1

allows CSRRWI to make maximum use of only 5-bit immediate

5 years agoshuffle CSR offsets around, offset VL and MVL by one
Luke Kenneth Casson Leighton [Tue, 16 Oct 2018 22:40:26 +0000 (23:40 +0100)]
shuffle CSR offsets around, offset VL and MVL by one

VL and MVL now span from 1 to XLEN rather than 0 to XLEN-1

also making room for M-Mode and S-Mode CSRs

5 years agofix compiler warnings on printfs
Luke Kenneth Casson Leighton [Mon, 15 Oct 2018 11:56:24 +0000 (12:56 +0100)]
fix compiler warnings on printfs

5 years agofix annoying printf warning on fp compiles
Luke Kenneth Casson Leighton [Mon, 15 Oct 2018 11:55:38 +0000 (12:55 +0100)]
fix annoying printf warning on fp compiles

5 years agowhoops deref null pointer
Luke Kenneth Casson Leighton [Mon, 15 Oct 2018 09:26:14 +0000 (10:26 +0100)]
whoops deref null pointer

5 years agoc_beqz sv operational
Luke Kenneth Casson Leighton [Mon, 15 Oct 2018 08:58:22 +0000 (09:58 +0100)]
c_beqz sv operational

5 years agoput RVC_SP at back of cintpatterns list
Luke Kenneth Casson Leighton [Mon, 15 Oct 2018 07:56:51 +0000 (08:56 +0100)]
put RVC_SP at back of cintpatterns list

5 years agoadd rvc_sp redirection/offset overload
Luke Kenneth Casson Leighton [Mon, 15 Oct 2018 06:49:26 +0000 (07:49 +0100)]
add rvc_sp redirection/offset overload

also found weird bug where RVC_FRS1/2 were not being detected,
how it was not found earlier is a mystery, code should not
have compiled!

5 years agoneed to check whether SP (reg 2) is used, without redirection
Luke Kenneth Casson Leighton [Mon, 15 Oct 2018 05:52:02 +0000 (06:52 +0100)]
need to check whether SP (reg 2) is used, without redirection

5 years agoadd overload/redirection for WRITE_REG
Luke Kenneth Casson Leighton [Mon, 15 Oct 2018 05:43:24 +0000 (06:43 +0100)]
add overload/redirection for WRITE_REG

5 years agomove design to separate document
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 21:29:34 +0000 (22:29 +0100)]
move design to separate document

5 years agodrop all lui from restriction on parallelism
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 21:29:22 +0000 (22:29 +0100)]
drop all lui from restriction on parallelism

5 years agodisable jal in sv
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 12:30:48 +0000 (13:30 +0100)]
disable jal in sv

5 years agorv_xxx convert c_xxx
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 05:41:04 +0000 (06:41 +0100)]
rv_xxx convert c_xxx

5 years agorv_add in lh/sh
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 05:31:27 +0000 (06:31 +0100)]
rv_add in lh/sh

5 years agorv_add in store
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 05:28:16 +0000 (06:28 +0100)]
rv_add in store

5 years agoblt and use of rv_add
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 05:26:45 +0000 (06:26 +0100)]
blt and use of rv_add

5 years agoadd rv_ge
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 05:23:38 +0000 (06:23 +0100)]
add rv_ge

5 years agoadd rv_eq and rv_ne
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 05:21:42 +0000 (06:21 +0100)]
add rv_eq and rv_ne

5 years agoadd rv_eq and rv_ne
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 05:20:05 +0000 (06:20 +0100)]
add rv_eq and rv_ne

5 years agoadd rv_gt headers
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 05:16:54 +0000 (06:16 +0100)]
add rv_gt headers

5 years agoadd rv_sr
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 05:16:03 +0000 (06:16 +0100)]
add rv_sr

5 years agoadd shiftright
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 05:06:43 +0000 (06:06 +0100)]
add shiftright

5 years agoadd shiftleft and lessthan
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 05:02:59 +0000 (06:02 +0100)]
add shiftleft and lessthan

5 years agomissed a mul
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 04:50:24 +0000 (05:50 +0100)]
missed a mul

5 years agoreplace % operator with rv_rem
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 04:49:05 +0000 (05:49 +0100)]
replace % operator with rv_rem

5 years agoreplace ^ operator with rv_xor
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 04:46:37 +0000 (05:46 +0100)]
replace ^ operator with rv_xor

5 years agoreplace | operator with rv_or
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 04:45:40 +0000 (05:45 +0100)]
replace | operator with rv_or

5 years agoreplace & operator with rv_and
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 04:44:36 +0000 (05:44 +0100)]
replace & operator with rv_and

5 years agoreplace operator * with rv_mul
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 04:42:39 +0000 (05:42 +0100)]
replace operator * with rv_mul

5 years agoadd rv_div (signed and unsigned) to replace operator /
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 04:35:03 +0000 (05:35 +0100)]
add rv_div (signed and unsigned) to replace operator /

5 years agoredirect subtract through rv_sub
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 04:24:18 +0000 (05:24 +0100)]
redirect subtract through rv_sub

5 years agoredirect add to rv_add
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 04:22:43 +0000 (05:22 +0100)]
redirect add to rv_add

5 years agoredirect add to rv_add
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 04:22:18 +0000 (05:22 +0100)]
redirect add to rv_add

5 years agobit of a mess: attempted to create a complete arithmetic overload
Luke Kenneth Casson Leighton [Sun, 14 Oct 2018 04:06:46 +0000 (05:06 +0100)]
bit of a mess: attempted to create a complete arithmetic overload

had to back most of it out, and left in a change to the amo* functions
passing in a 2nd parameter to the higher-order-function

5 years agorename _zext_xlen
Luke Kenneth Casson Leighton [Sat, 13 Oct 2018 13:43:06 +0000 (14:43 +0100)]
rename _zext_xlen

5 years agoadd sv_reg_t
Luke Kenneth Casson Leighton [Sat, 13 Oct 2018 13:40:30 +0000 (14:40 +0100)]
add sv_reg_t

5 years agoredirect WRITE_FRD including different types (128/64/32)
Luke Kenneth Casson Leighton [Fri, 12 Oct 2018 17:22:43 +0000 (18:22 +0100)]
redirect WRITE_FRD including different types (128/64/32)

5 years agoadd WRITE_FRD macro redirect
Luke Kenneth Casson Leighton [Fri, 12 Oct 2018 17:05:04 +0000 (18:05 +0100)]
add WRITE_FRD macro redirect

5 years agochanged style, can revert changes to amomin/max
Luke Kenneth Casson Leighton [Fri, 12 Oct 2018 15:21:11 +0000 (16:21 +0100)]
changed style, can revert changes to amomin/max

5 years agoadd frs2 redirect
Luke Kenneth Casson Leighton [Fri, 12 Oct 2018 15:20:27 +0000 (16:20 +0100)]
add frs2 redirect

5 years agoadd RS3 replacement
Luke Kenneth Casson Leighton [Fri, 12 Oct 2018 15:15:32 +0000 (16:15 +0100)]
add RS3 replacement

5 years agosimplify sv_proc_t redirection of RS1-3 / FRS1 macrhos
Luke Kenneth Casson Leighton [Fri, 12 Oct 2018 15:12:31 +0000 (16:12 +0100)]
simplify sv_proc_t redirection of RS1-3 / FRS1 macrhos

5 years agoredirect RS2 to sv_proc_t class
Luke Kenneth Casson Leighton [Fri, 12 Oct 2018 14:16:07 +0000 (15:16 +0100)]
redirect RS2 to sv_proc_t class

5 years agoproof-of-concept, redirect RS1 to class sv_proc_t
Luke Kenneth Casson Leighton [Fri, 12 Oct 2018 12:27:11 +0000 (13:27 +0100)]
proof-of-concept, redirect RS1 to class sv_proc_t

5 years agocombination of redirection through a "property" class overloads WRITE_RD
Luke Kenneth Casson Leighton [Fri, 12 Oct 2018 02:38:33 +0000 (03:38 +0100)]
combination of redirection through a "property" class overloads WRITE_RD

WRITE_RD, formerly a macro, now replaced with a function.
also required a redirector function rd

5 years agoredirect instructions through a class called sv_proc_t
Luke Kenneth Casson Leighton [Thu, 11 Oct 2018 20:45:26 +0000 (21:45 +0100)]
redirect instructions through a class called sv_proc_t

preparing the groundwork for a total over-ride of macros such as
WRITE_RD, and so on, so that element width can be implemented

5 years agomore explicit testing, duplicating header file algorithms for div, rem and shift
Luke Kenneth Casson Leighton [Thu, 11 Oct 2018 12:17:11 +0000 (13:17 +0100)]
more explicit testing, duplicating header file algorithms for div, rem and shift

5 years agowhoops run from 0-255 not 0-254, and other test corrections
Luke Kenneth Casson Leighton [Thu, 11 Oct 2018 07:09:16 +0000 (08:09 +0100)]
whoops run from 0-255 not 0-254, and other test corrections

5 years agoadd some operator tests for int8_t being typecast to uint16_t
Luke Kenneth Casson Leighton [Thu, 11 Oct 2018 06:56:32 +0000 (07:56 +0100)]
add some operator tests for int8_t being typecast to uint16_t

5 years agoadd more experimenting on operators
Luke Kenneth Casson Leighton [Thu, 11 Oct 2018 06:34:13 +0000 (07:34 +0100)]
add more experimenting on operators

5 years agoadd operators test class
Luke Kenneth Casson Leighton [Wed, 10 Oct 2018 14:01:42 +0000 (15:01 +0100)]
add operators test class

5 years agoadd operators library to contain operator-overloads of +/-/*/div/>/>= etc
Luke Kenneth Casson Leighton [Wed, 10 Oct 2018 08:49:43 +0000 (09:49 +0100)]
add operators library to contain operator-overloads of +/-/*/div/>/>= etc

for all types int8, uint8, float16_t, float32_t etc.

all to be template-ified

5 years agoget predicated-vectorised branch working
Luke Kenneth Casson Leighton [Tue, 9 Oct 2018 18:33:53 +0000 (19:33 +0100)]
get predicated-vectorised branch working

5 years agosave branch address and predication merged result, and test after branch
Luke Kenneth Casson Leighton [Tue, 9 Oct 2018 16:01:12 +0000 (17:01 +0100)]
save branch address and predication merged result, and test after branch

if the predication was all good, go ahead with the branch

still not tested for predicated / vectorised branch yet however
at least scalar branches work

5 years agoadd explanatory comment
Luke Kenneth Casson Leighton [Tue, 9 Oct 2018 15:17:42 +0000 (16:17 +0100)]
add explanatory comment

5 years agoadd explanatory comment
Luke Kenneth Casson Leighton [Tue, 9 Oct 2018 15:02:00 +0000 (16:02 +0100)]
add explanatory comment

5 years agostart adding explicit twin-predicated branch identification (rs2)
Luke Kenneth Casson Leighton [Tue, 9 Oct 2018 14:51:12 +0000 (15:51 +0100)]
start adding explicit twin-predicated branch identification (rs2)

5 years agoextend sv register file from 64 to 128 after discussion.
Luke Kenneth Casson Leighton [Tue, 9 Oct 2018 10:39:49 +0000 (11:39 +0100)]
extend sv register file from 64 to 128 after discussion.

evaluation of even embedded GPUs shows that they have really enormous
register files.  a fp16 x 4 to express quads, times four, takes up eight
consecutive registers just on its own.

5 years agooverride setpc macro so that sv can redirect it in branch
Luke Kenneth Casson Leighton [Sun, 7 Oct 2018 14:06:39 +0000 (15:06 +0100)]
override setpc macro so that sv can redirect it in branch

this is slightly complicated.  setpc is a global macro, intended for
use in the templates.  however for SV it needs to be conditional,
so needs to redirect to a function in sv_insn_t.

that in turn needs quite a few extra parameters: the current loop
element offset, the argument to set_pc just in case actually it is
detected that this really is to be a branch not a predication scenario
and so on.

have not tried out vectorisation yet, at least straight non-vectorised
branch operations (unit tests) pass.

5 years agoswap #ifdef USING_NOREGS so that it is possible to redefine set_pc
Luke Kenneth Casson Leighton [Sun, 7 Oct 2018 08:09:21 +0000 (09:09 +0100)]
swap #ifdef USING_NOREGS so that it is possible to redefine set_pc

to avoid having to modify the branch instructions the set_pc macro will
be #undefd and redefined to modify the predication target

5 years agoadd rd bit-setting function
Luke Kenneth Casson Leighton [Sun, 7 Oct 2018 07:03:11 +0000 (08:03 +0100)]
add rd bit-setting function

5 years agoadd extra debug printing for c.lwsp
Luke Kenneth Casson Leighton [Sun, 7 Oct 2018 05:52:11 +0000 (06:52 +0100)]
add extra debug printing for c.lwsp

5 years agoadd rvc_swsp_imm sv overload, provides vector unit stride now
Luke Kenneth Casson Leighton [Sun, 7 Oct 2018 03:51:10 +0000 (04:51 +0100)]
add rvc_swsp_imm sv overload, provides vector unit stride now

5 years agoc.swsp and c.fswsp predication and offset enabling
Luke Kenneth Casson Leighton [Sat, 6 Oct 2018 19:18:32 +0000 (20:18 +0100)]
c.swsp and c.fswsp predication and offset enabling

5 years agoallow x2 (sp) to be redirected in C.LWSP
Luke Kenneth Casson Leighton [Sat, 6 Oct 2018 16:07:16 +0000 (17:07 +0100)]
allow x2 (sp) to be redirected in C.LWSP

5 years agotemporary hack disabling SV in anything other than user mode
Luke Kenneth Casson Leighton [Sat, 6 Oct 2018 15:49:20 +0000 (16:49 +0100)]
temporary hack disabling SV in anything other than user mode

5 years agowhoops inverted ldsp and lwsp immediates
Luke Kenneth Casson Leighton [Sat, 6 Oct 2018 15:48:55 +0000 (16:48 +0100)]
whoops inverted ldsp and lwsp immediates

5 years agocreate ldsp immediate offset overrides
Luke Kenneth Casson Leighton [Sat, 6 Oct 2018 09:24:25 +0000 (10:24 +0100)]
create ldsp immediate offset overrides

this allows C.LWSP/C.LDSP to do predicated load/stores from contiguous
blocks of memory

5 years agoadd in predication for immediate, for C.LWSP
Luke Kenneth Casson Leighton [Sat, 6 Oct 2018 05:51:46 +0000 (06:51 +0100)]
add in predication for immediate, for C.LWSP

5 years agoreorganise src and dest vector-element offsets
Luke Kenneth Casson Leighton [Fri, 5 Oct 2018 14:26:20 +0000 (15:26 +0100)]
reorganise src and dest vector-element offsets

pass in pointer to offsets from processor_t->state.srcoffs and destoffs
into sv_insn_t so that the element state information about how many
parallel elements have been executed is recorded.

in this way it will be possible to restart a loop at the right place
if there is an exception (such as a memory cache miss on a LOAD/STORE)

5 years agoadd srcoffs and destoffs sv state, alter CSRs
Luke Kenneth Casson Leighton [Fri, 5 Oct 2018 14:09:48 +0000 (15:09 +0100)]
add srcoffs and destoffs sv state, alter CSRs

remove SVREALVL, replace with SVSTATE

5 years agoreorganise twin-predication
Luke Kenneth Casson Leighton [Thu, 4 Oct 2018 13:11:02 +0000 (14:11 +0100)]
reorganise twin-predication

move the offset incrementing to outside of the sv_insn_t, and pass in
the src_offs and dest_offs by reference, mirroring and matching the
predication src and dest referencing

5 years agobig reorganisation to support twin-predication
Luke Kenneth Casson Leighton [Thu, 4 Oct 2018 09:16:10 +0000 (10:16 +0100)]
big reorganisation to support twin-predication

twin predication is on certain operations like LOAD, STORE, C.MV, FCVT
where the effect is similar to VSPLAT, VINSERT, and also bitmanip
scatter/gather.

5 years agoadd in twin-predication identification
Luke Kenneth Casson Leighton [Wed, 3 Oct 2018 09:42:23 +0000 (10:42 +0100)]
add in twin-predication identification

pass in second predicate for twin-predication operands

5 years agodecided not to change the behaviour of LOAD/STORE
Luke Kenneth Casson Leighton [Wed, 3 Oct 2018 06:16:34 +0000 (07:16 +0100)]
decided not to change the behaviour of LOAD/STORE

5 years agostart work on parallelsing LOAD, pass in parameter to reinterpret immed
Luke Kenneth Casson Leighton [Tue, 2 Oct 2018 15:01:22 +0000 (16:01 +0100)]
start work on parallelsing LOAD, pass in parameter to reinterpret immed

5 years agodebug print for floating-point regs
Luke Kenneth Casson Leighton [Tue, 2 Oct 2018 11:26:47 +0000 (12:26 +0100)]
debug print for floating-point regs

5 years agoadd comment explaining why invert isnt done in zeroing test
Luke Kenneth Casson Leighton [Mon, 1 Oct 2018 14:09:21 +0000 (15:09 +0100)]
add comment explaining why invert isnt done in zeroing test
(already inverted basically)

5 years agoadd comment explaining use of insn._rd() in zeroing
Luke Kenneth Casson Leighton [Mon, 1 Oct 2018 14:08:25 +0000 (15:08 +0100)]
add comment explaining use of insn._rd() in zeroing

5 years agowhoops vloop continuation logic the wrong way round
Luke Kenneth Casson Leighton [Mon, 1 Oct 2018 11:00:27 +0000 (12:00 +0100)]
whoops vloop continuation logic the wrong way round

the loop has to continue if there is one vectorised register left
rather than stop if there is *no* vectorised registers

5 years agoskip parallelisation of complex LR/SC operations
Luke Kenneth Casson Leighton [Mon, 1 Oct 2018 01:16:21 +0000 (02:16 +0100)]
skip parallelisation of complex LR/SC operations

5 years agoidentify type of instruction with additional #defines
Luke Kenneth Casson Leighton [Mon, 1 Oct 2018 01:04:57 +0000 (02:04 +0100)]
identify type of instruction with additional #defines

5 years agoadd a #define to id_regs.py which indicates name of the instruction
Luke Kenneth Casson Leighton [Sun, 30 Sep 2018 12:33:07 +0000 (13:33 +0100)]
add a #define to id_regs.py which indicates name of the instruction

the c preprocessor cannot cope with detecting what the name of the
instruction is (and when it does, if it is "and" that throws a
compile-error), so as a workaround have id_regs.py create a
#define INSN_ADD, #define INSN_ADD_I etc.

5 years agolist of instructions to avoid parallelising
Luke Kenneth Casson Leighton [Sun, 30 Sep 2018 09:37:05 +0000 (10:37 +0100)]
list of instructions to avoid parallelising

5 years agoupdate template comment
Luke Kenneth Casson Leighton [Sun, 30 Sep 2018 08:26:54 +0000 (09:26 +0100)]
update template comment

5 years agolots of debugging of predication, found other errors
Luke Kenneth Casson Leighton [Sun, 30 Sep 2018 08:15:17 +0000 (09:15 +0100)]
lots of debugging of predication, found other errors

lots of stuff here:
* changed sv reg and pred entry layout to make them a bit more
  human-readable: put the key and idx on byte-boundaries
* bug in SV REG setting (getting int table instead of fp one, twice)
* bug(s) in SV PRED table setting (zeroing the reg tables, using idx not
  key and more)
* stop lookup of predication if the *REGISTER* entry is inactive
  (but not if it is a scalar, because scalar predication is ok)
  this was the final piece of the puzzle that got predication working
* added a useful macro for creating SV REG and PRED CAM table entries

predication now working including zeroing

5 years agoadd sv support for zeroing predication in dest register
Luke Kenneth Casson Leighton [Sun, 30 Sep 2018 07:00:09 +0000 (08:00 +0100)]
add sv support for zeroing predication in dest register

bit of a major rework:

* access to the "unpredicated" (non-zero-hacked) register was needed
* therefore all rd/rs1-3/rvc_xxx functions had to have _ variants
* the underscored variants are not predicated
* this in turn meant that the offset for each register was wrong
  as it is incremented *after* being checked
* therefore a newoffs had to be added
* and the reset_cache function copies the newoffs values

bit of a mess but it works: this is a state machine after all...

5 years agoadd in predication to sv instruction execution
Luke Kenneth Casson Leighton [Sun, 30 Sep 2018 06:07:43 +0000 (07:07 +0100)]
add in predication to sv instruction execution

this relies on setting the value of the destination register
(and source registers) to zero.  a bad hack but it will do