- 19 8月, 2020 1 次提交
-
-
由 linjiawei 提交于
-
- 15 8月, 2020 1 次提交
-
-
由 Yinan Xu 提交于
-
- 14 8月, 2020 1 次提交
-
-
由 Yinan Xu 提交于
-
- 12 8月, 2020 1 次提交
-
-
由 William Wang 提交于
-
- 11 8月, 2020 1 次提交
-
-
由 William Wang 提交于
-
- 04 8月, 2020 2 次提交
- 02 8月, 2020 1 次提交
-
-
由 LinJiawei 提交于
The goal of this commit is to remove 'implict val p: XSConfig' in our code becasue it's inconvenient
-
- 21 7月, 2020 1 次提交
-
-
由 Yinan Xu 提交于
-
- 20 7月, 2020 2 次提交
- 19 7月, 2020 2 次提交
- 18 7月, 2020 1 次提交
-
-
由 zoujr 提交于
-
- 16 7月, 2020 1 次提交
-
-
由 Yinan Xu 提交于
-
- 14 7月, 2020 1 次提交
-
-
由 LinJiawei 提交于
-
- 13 7月, 2020 1 次提交
-
-
由 LinJiawei 提交于
-
- 12 7月, 2020 1 次提交
-
-
由 LinJiawei 提交于
-
- 08 5月, 2020 1 次提交
-
-
由 LinJiawei 提交于
FPUv0.1 can pass all rvf/rvd tests in both riscv-tests and berkeley-softfloat Signed-off-by: NLinJiawei <linjiav@outlook.com>
-
- 11 12月, 2019 2 次提交
-
-
由 William Wang 提交于
-
由 William Wang 提交于
* According to The RISC-V Instruction Set Manual Volume II: Privileged Architecture, for instruction-fetch access or page-fault exceptions on systems with variable-length instructions, m/stval will contain the virtual address of the portion of the instruction that caused the fault while m/sepc will point to the beginning of the instruction.
-
- 23 11月, 2019 6 次提交
-
-
由 William Wang 提交于
-
由 William Wang 提交于
-
由 zhangzifei 提交于
-
由 zhangzifei 提交于
-
由 zhangzifei 提交于
change inner pc/target/npc... to VAddrBits(39) && epc/val... keep XLEN, sign-ext-write/cut-off-read && signExt(pc) for difftest && auipc/jal/jalr/ecall use the signExt(pc) && lr/sc don't change && pass busybox
-
由 zhangzifei 提交于
WIP: add AddrBits(64)/VAddrBits(39)/PAddrBits(32) && change btb/cache tagBits && change tlb ppn2Len. Next: add SimpleBusBundle addr bits param
-
- 21 11月, 2019 1 次提交
-
-
由 Zihao Yu 提交于
-
- 20 11月, 2019 2 次提交
-
-
由 zhangzifei 提交于
cancel fu.tlb, move sfence_vma decode to fu.mou && cancel TLBEXUIO, turn to BoringUtils.addSink/addSource
-
由 William Wang 提交于
When SPF/LPF was raised, LSU used to wait until its FSM go back to s_idle then commit. However, CSR will modify mode/status immediately when it gets PF signal. Here comes the problem: when a PF is raised, redirect may not happen because mode/status i not right.
-
- 19 11月, 2019 2 次提交
-
-
由 zhangzifei 提交于
-
由 William Wang 提交于
-
- 18 11月, 2019 3 次提交
-
-
由 zhangzifei 提交于
add PipelineConnect for request && add register to keep metas/datas && cputest/microbenc-test pass, but ipc is low
-
由 William Wang 提交于
-
由 William Wang 提交于
-
- 17 11月, 2019 4 次提交
-
-
由 Zihao Yu 提交于
* hasInstrPageFault should only be valid when io.in.valid === true.B * the badaddr of instruction PF is in io.cfIn.pc
-
由 William Wang 提交于
-
由 William Wang 提交于
This modification is used to solve the following case: ``` ffffffe000000094: 8d4d or a0,a0,a1 ffffffe000000096: 12000073 sfence.vma ffffffe00000009a: 18051073 csrw satp,a0 ffffffe00000009e: 00000517 auipc a0,0x0 ``` In that case, when executing `ffffffe00000009e`, noop get paddr from the new page table
-
由 William Wang 提交于
-
- 16 11月, 2019 1 次提交
-
-
由 William Wang 提交于
Remaining work: * Fix PF exceptions according to nemu commit: * e4d03123: riscv64,mmu: amo load should trigger AMO/Store exceptions * 06873d26: riscv64,mmu: mstatus.sum do not take effect when fetching instructions * AMO is not working correctly when PF exceptions raised * IDU gets wrong inst in xv6 sh
-