qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: Warner Losh <imp@bsdimp.com>, qemu-devel@nongnu.org
Cc: qemu-trivial@nongnu.org, Klye Evans <kevans@FreeBSD.org>,
	Michael Tokarev <mjt@tls.msk.ru>, Stacey Son <sson@FreeBSD.org>,
	Laurent Vivier <laurent@vivier.eu>
Subject: Re: [PATCH 22/24] bsd-user/arm/target_arch_signal.h: arm set_mcontext
Date: Thu, 28 Oct 2021 10:53:39 -0700	[thread overview]
Message-ID: <344ac222-daa9-c752-8c6a-21773efb8656@linaro.org> (raw)
In-Reply-To: <20211019164447.16359-23-imp@bsdimp.com>

On 10/19/21 9:44 AM, Warner Losh wrote:
> +    regs->regs[15] = tswap32(gr[TARGET_REG_PC]);
> +    cpsr = tswap32(gr[TARGET_REG_CPSR]);
> +    cpsr_write(regs, cpsr, CPSR_USER | CPSR_EXEC, CPSRWriteByInstr);

Hmm.  What's the expected behaviour if the saved CPSR state does not match the PC state 
wrt thumb?

I'm ok if this should fail in spectacular ways, I just wanna know.

I *think* what will happen at the moment is that qemu will go into a whacky state in which 
the translator will read and interpret unaligned data.

I have a pending patch set for arm to raise unaligned exceptions for mis-aligned pc.  For 
arm32 mode, this is fine, and we'll raise the exception.  But for thumb mode, this is 
architecturally impossible, and the translator will assert.

The assert is going to be a problem.  There are a couple of options:

(1) TARGET_REG_PC wins: unset PC[0] and adjust CPSR[T] to match.

(2) CPSR_T wins: unset pc[0] if CPSR[T] is set, unchanged otherwise.  (In the Arm ARM 
psueodcode, pc[0] is hardwired to 0 in thumb mode.)

(3) Don't worry about matching PC[0] and CPSR[T], but do prevent an impossible situation 
and unset PC[0] always.  If PC[1] is set, and CPSR[T] is unset, then we'll raise unaligned 
when the cpu restarts.

And, finally, you're missing the mc_vfp_* handling.


r~


  parent reply	other threads:[~2021-10-28 17:55 UTC|newest]

Thread overview: 95+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-19 16:44 [PATCH 00/24] bsd-user: arm (32-bit) support Warner Losh
2021-10-19 16:44 ` [PATCH 01/24] bsd-user/arm/target_arch_sysarch.h: Use consistent include guards Warner Losh
2021-10-23  7:29   ` Kyle Evans
2021-10-28 15:08   ` Richard Henderson
2021-10-19 16:44 ` [PATCH 02/24] bsd-user/arm/target_syscall.h: Add copyright and update name Warner Losh
2021-10-23  7:30   ` Kyle Evans
2021-10-28 15:08   ` Richard Henderson
2021-10-19 16:44 ` [PATCH 03/24] bsd-user/arm/target_arch_cpu.c: Target specific TLS routines Warner Losh
2021-10-23  7:30   ` Kyle Evans
2021-10-28 15:08   ` Richard Henderson
2021-10-19 16:44 ` [PATCH 04/24] bsd-user/arm/target_arch_cpu.h: CPU Loop definitions Warner Losh
2021-10-23  7:31   ` Kyle Evans
2021-10-28 15:14   ` Richard Henderson
2021-10-28 17:36     ` Warner Losh
2021-10-19 16:44 ` [PATCH 05/24] bsd-user/arm/target_arch_cpu.h: Implement target_cpu_clone_regs Warner Losh
2021-10-23  7:31   ` Kyle Evans
2021-10-28 15:15   ` Richard Henderson
2021-10-19 16:44 ` [PATCH 06/24] bsd-user/arm/target_arch_cpu.h: Dummy target_cpu_loop implementation Warner Losh
2021-10-23  7:32   ` Kyle Evans
2021-10-28 15:15   ` Richard Henderson
2021-10-19 16:44 ` [PATCH 07/24] bsd-user/arm/target_arch_cpu.h: Implment trivial EXCP exceptions Warner Losh
2021-10-26  5:52   ` Kyle Evans
2021-10-28 15:19   ` Richard Henderson
2021-10-19 16:44 ` [PATCH 08/24] bsd-user/arm/target_arch_cpu.h: Implement data abort exceptions Warner Losh
2021-10-26  5:47   ` Kyle Evans
2021-10-28 15:29   ` Richard Henderson
2021-10-28 16:56     ` Warner Losh
2021-10-19 16:44 ` [PATCH 09/24] bsd-user/arm/target_arch_cpu.h: Implement system call dispatch Warner Losh
2021-10-23  7:33   ` Kyle Evans
2021-10-23 15:17     ` Warner Losh
2021-10-28 15:35   ` Richard Henderson
2021-10-28 17:56     ` Warner Losh
2021-10-19 16:44 ` [PATCH 10/24] bsd-user/arm/target_arch_reg.h: Implement core dump register copying Warner Losh
2021-10-26  5:48   ` Kyle Evans
2021-10-28 15:36   ` Richard Henderson
2021-10-19 16:44 ` [PATCH 11/24] bsd-user/arm/target_arch_vmparam.h: Parameters for arm address space Warner Losh
2021-10-26  5:52   ` Kyle Evans
2021-10-28 15:37   ` Richard Henderson
2021-10-19 16:44 ` [PATCH 12/24] bsd-user/arm/target_arch_sigtramp.h: Signal Trampoline for arm Warner Losh
2021-10-26  5:51   ` Kyle Evans
2021-10-28 15:42   ` Richard Henderson
2021-10-28 19:35     ` Warner Losh
2021-10-19 16:44 ` [PATCH 13/24] bsd-user/arm/target_arch_thread.h: Routines to create and switch to a thread Warner Losh
2021-10-26  6:01   ` Kyle Evans
2021-10-26  6:11     ` Kyle Evans
2021-10-27 15:35       ` Warner Losh
2021-10-27 15:40         ` Kyle Evans
2021-10-28 15:57   ` Richard Henderson
2021-10-28 19:45     ` Warner Losh
2021-10-29 16:06       ` Richard Henderson
2021-10-19 16:44 ` [PATCH 14/24] bsd-user/arm/target_arch_elf.h: arm defines for ELF Warner Losh
2021-10-26  6:07   ` Kyle Evans
2021-10-28 16:02   ` Richard Henderson
2021-10-19 16:44 ` [PATCH 15/24] bsd-user/arm/target_arch_elf.h: arm get hwcap Warner Losh
2021-10-26  6:02   ` Kyle Evans
2021-10-28 16:06   ` Richard Henderson
2021-10-19 16:44 ` [PATCH 16/24] bsd-user/arm/target_arch_elf.h: arm get_hwcap2 impl Warner Losh
2021-10-26  6:02   ` Kyle Evans
2021-10-28 16:08   ` Richard Henderson
2021-10-19 16:44 ` [PATCH 17/24] bsd-user/arm/target_arch_signal.h: arm specific signal registers and stack Warner Losh
2021-10-26  6:03   ` Kyle Evans
2021-10-28 16:18   ` Richard Henderson
2021-10-28 16:48     ` Warner Losh
2021-10-19 16:44 ` [PATCH 18/24] bsd-user/arm/target_arch_signal.h: arm machine context for signals Warner Losh
2021-10-26  6:03   ` Kyle Evans
2021-10-28 17:04   ` Richard Henderson
2021-10-28 17:18   ` Richard Henderson
2021-10-28 20:16     ` Warner Losh
2021-10-19 16:44 ` [PATCH 19/24] bsd-user/arm/target_arch_signal.h: arm user context and trapframe " Warner Losh
2021-10-26  6:07   ` Kyle Evans
2021-10-27 15:48     ` Warner Losh
2021-10-28 17:22   ` Richard Henderson
2021-10-30  2:44     ` Warner Losh
2021-10-19 16:44 ` [PATCH 20/24] bsd-user/arm/target_arch_signal.h: arm set_sigtramp_args Warner Losh
2021-10-26  6:10   ` Kyle Evans
2021-10-28 17:25   ` Richard Henderson
2021-10-28 17:35     ` Kyle Evans
2021-10-28 22:22       ` Warner Losh
2021-10-28 22:41     ` Warner Losh
2021-10-30  2:47       ` Warner Losh
2021-10-19 16:44 ` [PATCH 21/24] bsd-user/arm/target_arch_signal.h: arm get_mcontext Warner Losh
2021-10-26  6:08   ` Kyle Evans
2021-10-28 17:27   ` Richard Henderson
2021-10-19 16:44 ` [PATCH 22/24] bsd-user/arm/target_arch_signal.h: arm set_mcontext Warner Losh
2021-10-26  6:12   ` Kyle Evans
2021-10-28 17:53   ` Richard Henderson [this message]
2021-10-29  0:07     ` Warner Losh
2021-10-29  4:34       ` Warner Losh
2021-10-28 17:57   ` Richard Henderson
2021-10-19 16:44 ` [PATCH 23/24] bsd-user/arm/target_arch_signal.h: arm get_ucontext_sigreturn Warner Losh
2021-10-26  6:12   ` Kyle Evans
2021-10-28 17:59   ` Richard Henderson
2021-10-19 16:44 ` [PATCH 24/24] bsd-user: add arm target build Warner Losh
2021-10-26  6:21   ` Kyle Evans
2021-10-28 18:02     ` Richard Henderson

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=344ac222-daa9-c752-8c6a-21773efb8656@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=imp@bsdimp.com \
    --cc=kevans@FreeBSD.org \
    --cc=laurent@vivier.eu \
    --cc=mjt@tls.msk.ru \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-trivial@nongnu.org \
    --cc=sson@FreeBSD.org \
    --subject='Re: [PATCH 22/24] bsd-user/arm/target_arch_signal.h: arm set_mcontext' \
    /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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).