#
ea7e6d59 |
| 23-Dec-2024 |
xiaofeibao <[email protected]> |
timing(vecExcpMod): add pipe from rab and rat's signals
|
#
5540bdc7 |
| 05-Dec-2024 |
xiaofeibao <[email protected]> |
fix(rob): fix bug of canAcceptForDispatch
|
#
ddb49062 |
| 19-Oct-2024 |
Xuan Hu <[email protected]> |
fix(VecExcp): commit vls exception after Rab commiting all reg pairs
|
#
8ab9d9d0 |
| 19-Oct-2024 |
Xuan Hu <[email protected]> |
fix(Rab): no need to limit diff commit valid only assert in commit state (#3760)
|
#
ea2894c8 |
| 06-Oct-2024 |
Xuan Hu <[email protected]> |
fix(ROB): vlsNeedCommit only assert one cycle to avoid dup message to RAB (#3702)
|
#
e43bb916 |
| 20-Sep-2024 |
Xuan Hu <[email protected]> |
feat(VecLoad): add VecLoadExcp module to handle merging old/new data
* When NF not 0, the register indices are arranged group by group. But in exception handle progress, all registers needed to merg
feat(VecLoad): add VecLoadExcp module to handle merging old/new data
* When NF not 0, the register indices are arranged group by group. But in exception handle progress, all registers needed to merge will be handled first, and then the registers needed to move will be handled later. * The need merge vdIdx can be until 8, so 4 bits reg is needed. * If the instruction is indexed, the eew of vd is sew from vtype. Otherwise, the eew of vd is encoded in instruction. * Use ivemulNoLessThanM1 and dvemulNoLessThanM1 to produce vemul_i_d to avoid either demul or iemul is less than M1. * For whole register load, need handle NF(nf + 1) dest regs. * Use data EMUL to calculate number of dest reg. * GetE8OffsetInVreg will return the n-th 8bit which idx mapped to. * Since xs will flush pipe, when vstart is not 0 and execute vector mem inst, the value of vstart in CSR is the first element of this vector instruction. When exception occurs, the vstart in writeback bundle is the new one, So writebacked vstart should never be used as the beginning of vector mem operation. * Non-seg indexed load use non-sequential vd. * When "index emul" / "data emul" equals 2, the old vd is located in vuopidx 0, 2, 4, 6, the new vd is located in vuopidx 1, 3, 5, 7. * Make rename's input not ready until VecExcpMod not busy. * Delay trap passed to difftest until VecExcpMod not busy. * Rab commit to VecExcpMod as it commit to Rat, and select real load reg maps in VecExcpMod. * Use isDstMask to distinguish vlm and other vle. * When isWhole, vd regs are sequential.
show more ...
|
#
7a5f6e11 |
| 20-Sep-2024 |
lewislzh <[email protected]> |
fix(rab): parameterization of commit/walk num logic (#3618)
|
#
63d67ef3 |
| 14-Sep-2024 |
Tang Haojin <[email protected]> |
build: enable always-basic-diff for make verilog (#3574)
This commit turns on basic difftest features again, no matter it's for
simulation or physical design. This commit aims at allowing designs f
build: enable always-basic-diff for make verilog (#3574)
This commit turns on basic difftest features again, no matter it's for
simulation or physical design. This commit aims at allowing designs for
physical design to be verified.
show more ...
|
#
65d838c0 |
| 10-Sep-2024 |
xiaofeibao-xjtu <[email protected]> |
timing(Rab): fix timing of state reg (#3517)
|
#
618b89e6 |
| 12-Jun-2024 |
lewislzh <[email protected]> |
Backend fixtiming: fix rab/exuwb/wbtorob timing (#3032)
rab: fix commit/walk/special walk Count from popcount to priority mux exuwb: fix exuwb Nto1 logic: add int/fp/vec 3 wbpath to wbarbite
Backend fixtiming: fix rab/exuwb/wbtorob timing (#3032)
rab: fix commit/walk/special walk Count from popcount to priority mux exuwb: fix exuwb Nto1 logic: add int/fp/vec 3 wbpath to wbarbiter wbtorob: fix writebacknum count: delete extra count for exu which cannot be compressed
show more ...
|
#
d3a32fa0 |
| 30-Apr-2024 |
xiaofeibao <[email protected]> |
Rab: use Reg for debug
|
#
780712aa |
| 19-Mar-2024 |
xiaofeibao-xjtu <[email protected]> |
backend: new rob 8 banks read and 8 commit width
|
#
81535d7b |
| 15-Mar-2024 |
sinsanction <[email protected]> |
Backend: remove unused extra RF read ports, connect real commit vtype to VTypeGen
|
#
5f8b6c9e |
| 07-Mar-2024 |
sinceforYy <[email protected]> |
Backend: add clock gating to valid singal
|
#
6b102a39 |
| 22-Nov-2023 |
Haojin Tang <[email protected]> |
Rab: shrink rab entry width
|
#
cda1c534 |
| 24-Oct-2023 |
xiaofeibao-xjtu <[email protected]> |
Rob: optimize timing, remove vconfig debugIO
|
#
7d086385 |
| 26-Dec-2023 |
Xuan Hu <[email protected]> |
Backend: disallow snapshot when there are tail uops at the same cycle
* When creating snapshot at the middle of split uops, RAT record the state before this snapshot, but ROB record the enqPtr conta
Backend: disallow snapshot when there are tail uops at the same cycle
* When creating snapshot at the middle of split uops, RAT record the state before this snapshot, but ROB record the enqPtr containing some states of last cycle.
show more ...
|
#
ffc4f3c2 |
| 27-Oct-2023 |
Haojin Tang <[email protected]> |
Rab: use `diffPtr` instead of `diffPtrOH` for better build speed
|
#
9b9e991b |
| 25-Oct-2023 |
Haojin Tang <[email protected]> |
Rab: use snapshot enqueue condition of rob
|
#
c2887b4f |
| 25-Oct-2023 |
Haojin Tang <[email protected]> |
Revert "Snapshot: fix enq condition"
This reverts commit a3ea5c1d41581cbff4cfc5175c0a4699c3786b0a.
|
#
a3ea5c1d |
| 25-Oct-2023 |
Xuan Hu <[email protected]> |
Snapshot: fix enq condition
|
#
c4b56310 |
| 20-Oct-2023 |
Haojin Tang <[email protected]> |
snapshot: flush conditionally when redirect comes
|
#
82640bc3 |
| 18-Oct-2023 |
Haojin Tang <[email protected]> |
Rab: set io.canEnq to true immediately when state becomes idle
|
#
83ba63b3 |
| 11-Oct-2023 |
Xuan Hu <[email protected]> |
fix merge error
|
#
f1ba628b |
| 26-Sep-2023 |
Haojin Tang <[email protected]> |
Rob: fix FP CSR issue when rob compressing
|