From: <ltaylorsimpson@gmail.com>
To: "'Brian Cain'" <brian.cain@oss.qualcomm.com>, <qemu-devel@nongnu.org>
Cc: <richard.henderson@linaro.org>, <philmd@linaro.org>,
<quic_mathbern@quicinc.com>, <ale@rev.ng>, <anjo@rev.ng>,
<quic_mliebel@quicinc.com>, <alex.bennee@linaro.org>,
<quic_mburton@quicinc.com>, <sidneym@quicinc.com>,
"'Brian Cain'" <bcain@quicinc.com>
Subject: RE: [PATCH 21/38] target/hexagon: Add system reg insns
Date: Fri, 7 Mar 2025 19:32:53 -0600 [thread overview]
Message-ID: <030601db8fca$04123730$0c36a590$@gmail.com> (raw)
In-Reply-To: <20250301052628.1011210-22-brian.cain@oss.qualcomm.com>
> -----Original Message-----
> From: Brian Cain <brian.cain@oss.qualcomm.com>
> Sent: Friday, February 28, 2025 11:26 PM
> To: qemu-devel@nongnu.org
> Cc: brian.cain@oss.qualcomm.com; richard.henderson@linaro.org;
> philmd@linaro.org; quic_mathbern@quicinc.com; ale@rev.ng; anjo@rev.ng;
> quic_mliebel@quicinc.com; ltaylorsimpson@gmail.com;
> alex.bennee@linaro.org; quic_mburton@quicinc.com;
> sidneym@quicinc.com; Brian Cain <bcain@quicinc.com>
> Subject: [PATCH 21/38] target/hexagon: Add system reg insns
>
> From: Brian Cain <bcain@quicinc.com>
>
> Signed-off-by: Brian Cain <brian.cain@oss.qualcomm.com>
> ---
> target/hexagon/macros.h | 2 +
> target/hexagon/hex_common.py | 15 +-
> target/hexagon/imported/encode_pp.def | 213 +++++++++++++++------
> target/hexagon/imported/system.idef | 262
> +++++++++++++++++++++++---
> 4 files changed, 410 insertions(+), 82 deletions(-)
>
> diff --git a/target/hexagon/macros.h b/target/hexagon/macros.h index
> 6e4a3a1697..b0e9610d98 100644
> --- a/target/hexagon/macros.h
> +++ b/target/hexagon/macros.h
> @@ -675,3 +675,5 @@ static inline TCGv gen_read_ireg(TCGv result, TCGv
> val, int shift) #endif
>
> #endif
> +
> +#define fPREDUSE_TIMING()
Why wasn't this in the previous patch with all the changes to macros.h?
> diff --git a/target/hexagon/hex_common.py
> b/target/hexagon/hex_common.py index 7fb11a0819..9147701333 100755
> --- a/target/hexagon/hex_common.py
> +++ b/target/hexagon/hex_common.py
> @@ -1235,11 +1235,18 @@ def init_registers():
> for reg in new_regs:
> new_registers[f"{reg.regtype}{reg.regid}"] = reg
>
> +def is_new_reg(tag, regid):
> + if regid[0] in "NO":
> + return True
> + return regid[0] == "P" and \
> + f"{regid}N" in semdict[tag] and \
> + f"{regid}V" not in semdict[tag]
> +
> def get_register(tag, regtype, regid):
> - if f"{regtype}{regid}V" in semdict[tag]:
> - return registers[f"{regtype}{regid}"]
> - else:
> - return new_registers[f"{regtype}{regid}"]
> + regid = f"{regtype}{regid}"
> + is_new = is_new_reg(tag, regid)
> + reg = new_registers[regid] if is_new else registers[regid]
> + return reg
This looks OK but is out of place in this patch.
>
> def helper_ret_type(tag, regs):
> ## If there is a scalar result, it is the return type diff --git
> a/target/hexagon/imported/encode_pp.def
> b/target/hexagon/imported/encode_pp.def
> index 0cd30a5e85..37faf62b1b 100644
> --- a/target/hexagon/imported/encode_pp.def
> +++ b/target/hexagon/imported/encode_pp.def
> @@ -1,5 +1,5 @@
> /*
> - * Copyright(c) 2019-2023 Qualcomm Innovation Center, Inc. All Rights
> Reserved.
> + * Copyright(c) 2019-2020 Qualcomm Innovation Center, Inc. All Rights
> Reserved.
Why are you changing the copyright date badkward?
> *
> * This program is free software; you can redistribute it and/or modify
> * it under the terms of the GNU General Public License as published by @@
> -16,6 +16,7 @@
> */
>
> /*
> + * encode32.def
> * Encodings for 32 bit instructions
> *
> */
> @@ -341,6 +342,8 @@ DEF_ENC32(L4_pload##TAG##tnew_abs,ICLASS_LD"
> 1 11 "OPC" iiiii PP110tti 1--ddd
> DEF_ENC32(L4_pload##TAG##fnew_abs,ICLASS_LD" 1 11 "OPC" iiiii PP111tti
> 1--ddddd")
>
>
> +
> +
> /* 0 000 misc: dealloc,loadw_locked,dcfetch */
> STD_LD_ENC(bzw4,"0 101")
> STD_LD_ENC(bzw2,"0 011")
> @@ -375,6 +378,7 @@ DEF_ANTICLASS32(ICLASS_LD" 1110 000----- PP------ --
> ------",LD_ADDR_POST_REG)
>
> DEF_ENC32(L2_deallocframe, ICLASS_LD" 000 0 000 sssss PP0----- ---
> ddddd")
> DEF_ENC32(L4_return, ICLASS_LD" 011 0 000 sssss PP0000-- ---ddddd")
> +
> DEF_ENC32(L4_return_t, ICLASS_LD" 011 0 000 sssss PP0100vv ---ddddd")
> DEF_ENC32(L4_return_f, ICLASS_LD" 011 0 000 sssss PP1100vv ---ddddd")
> DEF_ENC32(L4_return_tnew_pt, ICLASS_LD" 011 0 000 sssss PP0110vv ---
> ddddd") @@ -382,15 +386,18 @@ DEF_ENC32(L4_return_fnew_pt,
> ICLASS_LD" 011 0 000 sssss PP1110vv ---ddddd")
> DEF_ENC32(L4_return_tnew_pnt, ICLASS_LD" 011 0 000 sssss PP0010vv ---
> ddddd") DEF_ENC32(L4_return_fnew_pnt, ICLASS_LD" 011 0 000 sssss
> PP1010vv ---ddddd")
>
> -DEF_ENC32(L2_loadw_locked,ICLASS_LD" 001 0 000 sssss PP000---
> 000ddddd")
> -
> +/** Load Acquire Store Release Encoding **/
>
> +DEF_ENC32(L2_loadw_locked, ICLASS_LD" 001 0 000 sssss PP000---
> 000ddddd")
> +DEF_ENC32(L4_loadd_locked, ICLASS_LD" 001 0 000 sssss PP010---
> 000ddddd")
There are lots of changes here that look like formatting/whitespace. It would be easier to review if you could minimize these.
> diff --git a/target/hexagon/imported/system.idef
> b/target/hexagon/imported/system.idef
> index 7c6568e75e..fd7ef18b3e 100644
> --- a/target/hexagon/imported/system.idef
> +++ b/target/hexagon/imported/system.idef
> @@ -25,44 +25,262 @@
> /* User->OS interface */
> /********************************************/
>
> -Q6INSN(J2_trap0,"trap0(#u8)",ATTRIBS(A_COF),
> +Q6INSN(J2_trap0,"trap0(#u8)",ATTRIBS(A_COF,A_NOTE_NOPACKET,A_RES
> TRICT_N
> +OPACKET),
> "Trap to Operating System",
> - fTRAP(0,uiV);
> + fTRAP(0,uiV);
The formatting/whitespace changes are not as rampant in this file, but this is one example.
> )
next prev parent reply other threads:[~2025-03-08 1:33 UTC|newest]
Thread overview: 120+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-03-01 5:25 [PATCH 00/38] hexagon system emu, part 1/3 Brian Cain
2025-03-01 5:25 ` [PATCH 01/38] docs: Add hexagon sysemu docs Brian Cain
2025-03-05 19:29 ` ltaylorsimpson
2025-03-01 5:25 ` [PATCH 02/38] docs/system: Add hexagon CPU emulation Brian Cain
2025-03-05 19:36 ` ltaylorsimpson
2025-03-05 20:12 ` Brian Cain
2025-03-05 21:21 ` ltaylorsimpson
2025-03-05 21:28 ` Brian Cain
2025-03-01 5:25 ` [PATCH 03/38] target/hexagon: Add System/Guest register definitions Brian Cain
2025-03-06 20:54 ` ltaylorsimpson
2025-04-16 17:54 ` ltaylorsimpson
2025-04-16 19:43 ` Brian Cain
2025-04-16 22:02 ` ltaylorsimpson
2025-09-02 0:17 ` Brian Cain
2025-03-01 5:25 ` [PATCH 04/38] target/hexagon: Make gen_exception_end_tb non-static Brian Cain
2025-03-06 20:55 ` ltaylorsimpson
2025-03-01 5:25 ` [PATCH 05/38] target/hexagon: Switch to tag_ignore(), generate via get_{user, sys}_tags() Brian Cain via
2025-03-06 21:07 ` ltaylorsimpson
2025-03-01 5:25 ` [PATCH 06/38] target/hexagon: Add privilege check, use tag_ignore() Brian Cain
2025-03-06 21:11 ` ltaylorsimpson
2025-03-06 22:01 ` Richard Henderson
2025-09-02 0:24 ` Brian Cain
2025-03-01 5:25 ` [PATCH 07/38] target/hexagon: Add a placeholder fp exception Brian Cain
2025-03-06 21:22 ` ltaylorsimpson
2025-03-01 5:25 ` [PATCH 08/38] target/hexagon: Add guest, system reg number defs Brian Cain
2025-03-06 21:30 ` ltaylorsimpson
2025-03-08 0:35 ` Sid Manning
2025-09-02 0:25 ` Brian Cain
2025-03-01 5:25 ` [PATCH 09/38] target/hexagon: Add guest, system reg number state Brian Cain
2025-03-06 21:32 ` ltaylorsimpson
2025-03-12 19:15 ` Philippe Mathieu-Daudé
2025-09-02 0:27 ` Brian Cain
2025-03-01 5:26 ` [PATCH 10/38] target/hexagon: Add TCG values for sreg, greg Brian Cain
2025-03-06 21:38 ` ltaylorsimpson
2025-09-02 0:28 ` Brian Cain
2025-03-01 5:26 ` [PATCH 11/38] target/hexagon: Add guest/sys reg writes to DisasContext Brian Cain
2025-03-06 21:40 ` ltaylorsimpson
2025-03-01 5:26 ` [PATCH 12/38] target/hexagon: Add imported macro, attr defs for sysemu Brian Cain
2025-03-07 19:01 ` ltaylorsimpson
2025-09-02 0:36 ` Brian Cain
2025-03-01 5:26 ` [PATCH 13/38] target/hexagon: Define DCache states Brian Cain
2025-03-07 19:03 ` ltaylorsimpson
2025-03-01 5:26 ` [PATCH 14/38] target/hexagon: Add new macro definitions for sysemu Brian Cain
2025-03-07 19:35 ` ltaylorsimpson
2025-09-02 0:38 ` Brian Cain
2025-03-01 5:26 ` [PATCH 15/38] target/hexagon: Add handlers for guest/sysreg r/w Brian Cain
2025-03-07 19:46 ` ltaylorsimpson
2025-09-02 0:40 ` Brian Cain
2025-03-01 5:26 ` [PATCH 16/38] target/hexagon: Add placeholder greg/sreg r/w helpers Brian Cain
2025-03-07 20:45 ` ltaylorsimpson
2025-03-01 5:26 ` [PATCH 17/38] target/hexagon: Add vmstate representation Brian Cain
2025-03-07 21:19 ` ltaylorsimpson
2025-03-01 5:26 ` [PATCH 18/38] target/hexagon: Make A_PRIV, "J2_trap*" insts need_env() Brian Cain
2025-03-07 21:20 ` ltaylorsimpson
2025-03-01 5:26 ` [PATCH 19/38] target/hexagon: Define register fields for system regs Brian Cain
2025-03-07 21:21 ` ltaylorsimpson
2025-03-01 5:26 ` [PATCH 20/38] target/hexagon: Implement do_raise_exception() Brian Cain
2025-03-07 21:28 ` ltaylorsimpson
2025-09-02 0:41 ` Brian Cain
2025-03-01 5:26 ` [PATCH 21/38] target/hexagon: Add system reg insns Brian Cain
2025-03-08 1:32 ` ltaylorsimpson [this message]
2025-09-02 0:44 ` Brian Cain
2025-03-01 5:26 ` [PATCH 22/38] target/hexagon: Add sysemu TCG overrides Brian Cain
2025-03-08 1:43 ` ltaylorsimpson
2025-09-02 0:46 ` Brian Cain
2025-03-01 5:26 ` [PATCH 23/38] target/hexagon: Add implicit attributes to sysemu macros Brian Cain
2025-03-11 22:30 ` ltaylorsimpson
2025-09-02 0:47 ` Brian Cain
2025-03-01 5:26 ` [PATCH 24/38] target/hexagon: Add TCG overrides for int handler insts Brian Cain
2025-03-08 1:46 ` ltaylorsimpson
2025-03-01 5:26 ` [PATCH 25/38] target/hexagon: Add TCG overrides for thread ctl Brian Cain
2025-03-08 1:47 ` ltaylorsimpson
2025-03-01 5:26 ` [PATCH 26/38] target/hexagon: Add TCG overrides for rte, nmi Brian Cain
2025-03-11 22:33 ` ltaylorsimpson
2025-03-01 5:26 ` [PATCH 27/38] target/hexagon: Add sreg_{read,write} helpers Brian Cain
2025-03-11 23:22 ` ltaylorsimpson
2025-09-02 0:53 ` Brian Cain
2025-03-01 5:26 ` [PATCH 28/38] target/hexagon: Initialize htid, modectl regs Brian Cain
2025-03-11 23:26 ` ltaylorsimpson
2025-03-12 14:02 ` Sid Manning
2025-03-12 19:19 ` Philippe Mathieu-Daudé
2025-03-12 23:10 ` Brian Cain
2025-03-12 23:40 ` Philippe Mathieu-Daudé
2025-03-13 18:47 ` ltaylorsimpson
2025-03-13 19:06 ` Richard Henderson
2025-03-19 16:08 ` Sid Manning
2025-03-20 15:34 ` Richard Henderson
2025-03-20 17:38 ` Sid Manning
2025-09-02 0:56 ` Brian Cain
2025-03-01 5:26 ` [PATCH 29/38] target/hexagon: Add locks, id, next_PC to state Brian Cain
2025-03-11 23:33 ` ltaylorsimpson
2025-03-01 5:26 ` [PATCH 30/38] target/hexagon: Add a TLB count property Brian Cain
2025-03-11 23:41 ` ltaylorsimpson
2025-03-12 14:01 ` Sid Manning
2025-03-01 5:26 ` [PATCH 31/38] target/hexagon: Add {TLB, k0}lock, cause code, wait_next_pc Brian Cain via
2025-03-11 23:44 ` ltaylorsimpson
2025-03-12 16:58 ` [PATCH 31/38] target/hexagon: Add {TLB,k0}lock, " Sid Manning
2025-03-01 5:26 ` [PATCH 32/38] target/hexagon: Add stubs for modify_ssr/get_exe_mode Brian Cain
2025-03-11 23:43 ` ltaylorsimpson
2025-03-01 5:26 ` [PATCH 33/38] target/hexagon: Add gdb support for sys regs Brian Cain
2025-03-12 16:27 ` ltaylorsimpson
2025-03-12 19:10 ` Sid Manning
2025-03-12 19:27 ` Sid Manning
2025-03-12 19:46 ` Matheus Tavares Bernardino
2025-09-02 1:15 ` Brian Cain
2025-03-01 5:26 ` [PATCH 34/38] target/hexagon: Add initial MMU model Brian Cain
2025-03-12 17:04 ` ltaylorsimpson
2025-09-02 1:20 ` Brian Cain
2025-03-12 19:20 ` Philippe Mathieu-Daudé
2025-03-12 21:15 ` Sid Manning
2025-03-12 23:32 ` Philippe Mathieu-Daudé
2025-03-01 5:26 ` [PATCH 35/38] target/hexagon: Add IRQ events Brian Cain
2025-03-12 17:06 ` ltaylorsimpson
2025-03-01 5:26 ` [PATCH 36/38] target/hexagon: Add clear_wait_mode() definition Brian Cain
2025-03-12 17:08 ` ltaylorsimpson
2025-03-01 5:26 ` [PATCH 37/38] target/hexagon: Define f{S,G}ET_FIELD macros Brian Cain
2025-03-12 17:11 ` ltaylorsimpson
2025-03-01 5:26 ` [PATCH 38/38] target/hexagon: Add hex_interrupts support Brian Cain
2025-03-12 17:32 ` ltaylorsimpson
2025-09-02 1:22 ` Brian Cain
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='030601db8fca$04123730$0c36a590$@gmail.com' \
--to=ltaylorsimpson@gmail.com \
--cc=ale@rev.ng \
--cc=alex.bennee@linaro.org \
--cc=anjo@rev.ng \
--cc=bcain@quicinc.com \
--cc=brian.cain@oss.qualcomm.com \
--cc=philmd@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=quic_mathbern@quicinc.com \
--cc=quic_mburton@quicinc.com \
--cc=quic_mliebel@quicinc.com \
--cc=richard.henderson@linaro.org \
--cc=sidneym@quicinc.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.