QEMU-Devel Archive on lore.kernel.org
 help / color / Atom feed
From: Jonathan Behrens <fintelia@gmail.com>
To: Richard Henderson <richard.henderson@linaro.org>
Cc: Palmer Dabbelt <palmer@sifive.com>,
	"open list:RISC-V" <qemu-riscv@nongnu.org>,
	Anup Patel <Anup.Patel@wdc.com>,
	"qemu-devel@nongnu.org Developers" <qemu-devel@nongnu.org>,
	Atish Patra <Atish.Patra@wdc.com>,
	Alistair Francis <Alistair.Francis@wdc.com>,
	Alistair Francis <alistair23@gmail.com>
Subject: Re: [Qemu-devel] [Qemu-riscv] [PATCH v1 10/28] target/riscv: Convert mie and mstatus to pointers
Date: Thu, 19 Sep 2019 12:58:40 -0400
Message-ID: <CANnJOVEsF57FBbNxJ_WvL3O7iJ_mZijYikC_aNhxKcP856qTzw@mail.gmail.com> (raw)
In-Reply-To: <18b87754-d5b2-e4f1-bdc5-92ad26b97379@linaro.org>

On Thu, Sep 19, 2019 at 10:50 AM Richard Henderson
<richard.henderson@linaro.org> wrote:
>
> On 9/18/19 4:47 PM, Alistair Francis wrote:
> > I'm not a fan of the pointer method that I'm using, but to me it seems
> > the least worst in terms of handling future code, keeping everythign
> > consistnent and avoiding complex access rules.
>
> FWIW, I prefer the "banked" register method used by ARM.
>
> enum {
>     M_REG_NS = 0,    /* non-secure mode */
>     M_REG_S = 1,     /* secure mode */
>     M_REG_NUM_BANKS = 2,
> };
>
> ...
>
>         uint32_t vecbase[M_REG_NUM_BANKS];
>         uint32_t basepri[M_REG_NUM_BANKS];
>         uint32_t control[M_REG_NUM_BANKS];
>
> The major difference that I see is that a pointer can only represent a single
> state at a single time.  With an index, different parts of the code can ask
> different questions that may have different states.  E.g. "are we currently in
> secure mode" vs "will the exception return to secure mode".

This makes a lot of sense to me. It means that any individual control register
has an unambiguous name that doesn't change based on context. They aren't quite
the same names as used in the architecture specification (mie & vsie
vs. mie[NOVIRT] & mie[VIRT]), but they are reasonably close. It also means other
parts of the code can't ignore that there are two different versions of the
registers in play. Perhaps the biggest benefit though is that you can sidestep
swapping on mode changes *and* avoid needing any super fancy logic in the access
functions:

int read_mstatus(...) {
    target_ulong novirt_mask = ...;
    *val = env->mstatus[NOVIRT] & novirt_mask | env->mstatus[virt_mode()];
}

int read_vsstatus(...) {
    *val = env->mstatus[VIRT];
}

int write_mstatus(...) {
    ...
    target_ulong novirt_mask = ...;
    env->mstatus[NOVIRT] = (env->mstatus[NOVIRT] & ~novirt_mask) |
                           (newval & novirt_mask);
    env->mstatus[virt_mode()] = (env->mstatus[virt_mode()] & novirt_mask) |
                                (newval & ~novirt_mask);
}


  reply index

Thread overview: 75+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-23 23:37 [Qemu-devel] [PATCH v1 00/28] Add RISC-V Hypervisor Extension v0.4 Alistair Francis
2019-08-23 23:37 ` [Qemu-devel] [PATCH v1 01/28] target/riscv: Add the Hypervisor extension Alistair Francis
2019-08-27 15:26   ` Chih-Min Chao
2019-09-10 13:43   ` Palmer Dabbelt
2019-08-23 23:37 ` [Qemu-devel] [PATCH v1 02/28] target/riscv: Add the virtulisation mode Alistair Francis
2019-08-27 15:44   ` [Qemu-devel] [Qemu-riscv] " Chih-Min Chao
2019-08-28  0:08     ` Alistair Francis
2019-09-10 13:44   ` [Qemu-devel] " Palmer Dabbelt
2019-09-16 15:57     ` Alistair Francis
2019-08-23 23:37 ` [Qemu-devel] [PATCH v1 03/28] target/riscv: Add the force HS exception mode Alistair Francis
2019-08-27 15:46   ` Chih-Min Chao
2019-09-10 14:48   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 04/28] target/riscv: Fix CSR perm checking for HS mode Alistair Francis
2019-09-10 14:48   ` Palmer Dabbelt
2019-10-16 20:56     ` Alistair Francis
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 05/28] target/riscv: Add the Hypervisor CSRs to CPUState Alistair Francis
2019-08-27 15:50   ` [Qemu-devel] [Qemu-riscv] " Chih-Min Chao
2019-09-10 14:48   ` [Qemu-devel] " Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 06/28] target/riscv: Print priv and virt in disas log Alistair Francis
2019-09-10 14:48   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 07/28] target/riscv: Dump Hypervisor registers if enabled Alistair Francis
2019-09-10 14:48   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 08/28] target/riscv: Add Hypervisor CSR access functions Alistair Francis
2019-09-10 14:48   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 09/28] target/riscv: Add Hypervisor virtual CSRs accesses Alistair Francis
2019-09-10 14:48   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 10/28] target/riscv: Convert mie and mstatus to pointers Alistair Francis
2019-09-11  8:24   ` Palmer Dabbelt
2019-09-11 14:54     ` [Qemu-devel] [Qemu-riscv] " Jonathan Behrens
2019-09-17 23:33       ` Alistair Francis
2019-09-18  1:59         ` Jonathan Behrens
2019-09-18 23:47           ` Alistair Francis
2019-09-19 14:50             ` Richard Henderson
2019-09-19 16:58               ` Jonathan Behrens [this message]
2019-10-25 20:28                 ` Alistair Francis
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 11/28] target/riscv: Add background register swapping function Alistair Francis
2019-09-11 14:17   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 12/28] target/riscv: Add support for virtual interrupt setting Alistair Francis
2019-09-14 20:30   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 13/28] target/ricsv: Flush the TLB on virtulisation mode changes Alistair Francis
2019-09-14 20:30   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 14/28] target/riscv: Generate illegal instruction on WFI when V=1 Alistair Francis
2019-09-14 20:30   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 15/28] riscv: plic: Always set sip.SEIP bit for HS Alistair Francis
2019-09-14 20:32   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 16/28] target/riscv: Add hypvervisor trap support Alistair Francis
2019-09-20 14:01   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 17/28] target/riscv: Add Hypervisor trap return support Alistair Francis
2019-10-01 18:33   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 18/28] target/riscv: Add hfence instructions Alistair Francis
2019-10-01 18:34   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 19/28] target/riscv: Disable guest FP support based on virtual status Alistair Francis
2019-10-01 18:34   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 20/28] target/riscv: Mark both sstatus and vsstatus as dirty Alistair Francis
2019-10-01 18:34   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 21/28] target/riscv: Respect MPRV and SPRV for floating point ops Alistair Francis
2019-10-02 23:52   ` Palmer Dabbelt
2019-10-16 21:01     ` Alistair Francis
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 22/28] target/riscv: Allow specifying MMU stage Alistair Francis
2019-10-03 15:53   ` Palmer Dabbelt
2019-10-07 18:05     ` Alistair Francis
2019-10-16 19:02       ` Palmer Dabbelt
2019-10-16 21:25         ` Alistair Francis
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 23/28] target/riscv: Allow specifying number of MMU stages Alistair Francis
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 24/28] target/riscv: Implement second stage MMU Alistair Francis
2019-10-07 16:15   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 25/28] target/riscv: Call the second stage MMU in virtualisation mode Alistair Francis
2019-10-08 17:54   ` Palmer Dabbelt
2019-08-23 23:38 ` [Qemu-devel] [PATCH v1 26/28] target/riscv: Add support for the 32-bit MSTATUSH CSR Alistair Francis
2019-10-08 18:36   ` Palmer Dabbelt
2019-08-23 23:39 ` [Qemu-devel] [PATCH v1 27/28] target/riscv: Add the MSTATUS_MPV_ISSET helper macro Alistair Francis
2019-10-08 18:36   ` Palmer Dabbelt
2019-10-16 21:14     ` Alistair Francis
2019-08-23 23:39 ` [Qemu-devel] [PATCH v1 28/28] target/riscv: Allow enabling the Hypervisor extension Alistair Francis
2019-10-08 18:53   ` Palmer Dabbelt

Reply instructions:

You may reply publically 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=CANnJOVEsF57FBbNxJ_WvL3O7iJ_mZijYikC_aNhxKcP856qTzw@mail.gmail.com \
    --to=fintelia@gmail.com \
    --cc="qemu-devel@nongnu.org \
    --cc=Alistair.Francis@wdc.com \
    --cc=Anup.Patel@wdc.com \
    --cc=Atish.Patra@wdc.com \
    --cc=alistair23@gmail.com \
    --cc=palmer@sifive.com \
    --cc=qemu-riscv@nongnu.org \
    --cc=richard.henderson@linaro.org \
    /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

QEMU-Devel Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/qemu-devel/0 qemu-devel/git/0.git
	git clone --mirror https://lore.kernel.org/qemu-devel/1 qemu-devel/git/1.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 qemu-devel qemu-devel/ https://lore.kernel.org/qemu-devel \
		qemu-devel@nongnu.org
	public-inbox-index qemu-devel

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.nongnu.qemu-devel


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git