Issue with illegal_reg_rv32e when RV32E

[esdatmisarian]

Instruction sw x20, -1982(x3) is not seen as illegal with RV32E
Waveform:

if (RV32E) begin : gen_rv32e_reg_check_active
assign illegal_reg_rv32e = ((rf_raddr_a_o[4] & (alu_op_a_mux_sel_o == OP_A_REG_A)) |
(rf_raddr_b_o[4] & (alu_op_b_mux_sel_o == OP_B_REG_B)) |
(rf_waddr_o[4] & rf_we));

version:
5e4798f

[kliuMsft]

Very good finding, thanks!

@marnovandermaas, this might be an issue for upstream ibex as well.

[marnovandermaas]

I think this is related to: lowRISC/ibex#2178

[esdatmisarian]

Yes totally agree. Rgds From: Marno van der Maas ***@***.***> Sent: Wednesday, July 17, 2024 11:03 AM To: microsoft/cheriot-ibex ***@***.***> Cc: Sedat SAYAR ***@***.***>; Author ***@***.***> Subject: Re: [microsoft/cheriot-ibex] Issue with illegal_reg_rv32e when RV32E (Issue #40) I think this is related to: lowRISC/ibex#2178<lowRISC/ibex#2178> — Reply to this email directly, view it on GitHub<#40 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANX5NG2WLUIGFVYOHGYN5DLZMYXNJAVCNFSM6AAAAABK6VFRVGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEMZSHAYDIMZRHE>. You are receiving this because you authored the thread.Message ID: ***@***.******@***.***>>

[kliuMsft]

New commit a57ec46 should fix the issue, please help to verify.

[esdatmisarian]

Sorry for latency in feedback, it is well corrected. * Verification is OK Rgds From: Kunyan Liu ***@***.***> Sent: Thursday, July 18, 2024 3:01 AM To: microsoft/cheriot-ibex ***@***.***> Cc: Sedat SAYAR ***@***.***>; Author ***@***.***> Subject: Re: [microsoft/cheriot-ibex] Issue with illegal_reg_rv32e when RV32E (Issue #40) New commit a57ec46<a57ec46> should fix the issue, please help to verify. — Reply to this email directly, view it on GitHub<#40 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANX5NGZLEAA4VD644PQA5BLZM4HVNAVCNFSM6AAAAABK6VFRVGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEMZUHE3TKMBRHA>. You are receiving this because you authored the thread.Message ID: ***@***.******@***.***>>

[kliuMsft]

Thanks for the confirmation. I am closing the issue then.