From: Paolo Bonzini <pbonzini@redhat.com>
To: Richard Henderson <rth@twiddle.net>, qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] [PATCH 2/3] target-i386: Fix SSE status flag corruption
Date: Tue, 25 Feb 2014 09:22:08 +0100 [thread overview]
Message-ID: <530C52B0.6070602@redhat.com> (raw)
In-Reply-To: <1393313432-15327-3-git-send-email-rth@twiddle.net>
Il 25/02/2014 08:30, Richard Henderson ha scritto:
> When we restore the mxcsr register with FXRSTOR, or set it with gdb,
> we need to update the various SSE status flags in CPUX86State
>
> Reported-by: Richard Purdie <richard.purdie@linuxfoundation.org>
> Signed-off-by: Richard Henderson <rth@twiddle.net>
> ---
> Differs from Purdie's patch primarily in fixing gdb too. And that
> required exporting update_sse_status. Which suggested that the name
> and interface be changed to match the norm.
>
>
> r~
> ---
> target-i386/cpu.h | 3 +++
> target-i386/fpu_helper.c | 15 ++++++++-------
> target-i386/gdbstub.c | 2 +-
> 3 files changed, 12 insertions(+), 8 deletions(-)
>
> diff --git a/target-i386/cpu.h b/target-i386/cpu.h
> index 1b94f0f..5d3f143 100644
> --- a/target-i386/cpu.h
> +++ b/target-i386/cpu.h
> @@ -1259,6 +1259,9 @@ static inline void cpu_load_efer(CPUX86State *env, uint64_t val)
> }
> }
>
> +/* fpu_helper.c */
> +void cpu_set_mxcsr(CPUX86State *env, uint32_t val);
> +
> /* svm_helper.c */
> void cpu_svm_check_intercept_param(CPUX86State *env1, uint32_t type,
> uint64_t param);
> diff --git a/target-i386/fpu_helper.c b/target-i386/fpu_helper.c
> index c0427fe..de7ba76 100644
> --- a/target-i386/fpu_helper.c
> +++ b/target-i386/fpu_helper.c
> @@ -1179,7 +1179,7 @@ void helper_fxrstor(CPUX86State *env, target_ulong ptr, int data64)
>
> if (env->cr[4] & CR4_OSFXSR_MASK) {
> /* XXX: finish it */
> - env->mxcsr = cpu_ldl_data(env, ptr + 0x18);
> + cpu_set_mxcsr(env, cpu_ldl_data(env, ptr + 0x18));
> /* cpu_ldl_data(env, ptr + 0x1c); */
> if (env->hflags & HF_CS64_MASK) {
> nb_xmm_regs = 16;
> @@ -1229,12 +1229,14 @@ floatx80 cpu_set_fp80(uint64_t mant, uint16_t upper)
> #define SSE_RC_CHOP 0x6000
> #define SSE_FZ 0x8000
>
> -static void update_sse_status(CPUX86State *env)
> +void cpu_set_mxcsr(CPUX86State *env, uint32_t mxcsr)
> {
> int rnd_type;
>
> + env->mxcsr = mxcsr;
> +
> /* set rounding mode */
> - switch (env->mxcsr & SSE_RC_MASK) {
> + switch (mxcsr & SSE_RC_MASK) {
> default:
> case SSE_RC_NEAR:
> rnd_type = float_round_nearest_even;
> @@ -1252,16 +1254,15 @@ static void update_sse_status(CPUX86State *env)
> set_float_rounding_mode(rnd_type, &env->sse_status);
>
> /* set denormals are zero */
> - set_flush_inputs_to_zero((env->mxcsr & SSE_DAZ) ? 1 : 0, &env->sse_status);
> + set_flush_inputs_to_zero((mxcsr & SSE_DAZ) ? 1 : 0, &env->sse_status);
>
> /* set flush to zero */
> - set_flush_to_zero((env->mxcsr & SSE_FZ) ? 1 : 0, &env->fp_status);
> + set_flush_to_zero((mxcsr & SSE_FZ) ? 1 : 0, &env->fp_status);
> }
>
> void helper_ldmxcsr(CPUX86State *env, uint32_t val)
> {
> - env->mxcsr = val;
> - update_sse_status(env);
> + cpu_set_mxcsr(env, val);
> }
>
> void helper_enter_mmx(CPUX86State *env)
> diff --git a/target-i386/gdbstub.c b/target-i386/gdbstub.c
> index 15bebef..d34e535 100644
> --- a/target-i386/gdbstub.c
> +++ b/target-i386/gdbstub.c
> @@ -222,7 +222,7 @@ int x86_cpu_gdb_write_register(CPUState *cs, uint8_t *mem_buf, int n)
> return 4;
>
> case IDX_MXCSR_REG:
> - env->mxcsr = ldl_p(mem_buf);
> + cpu_set_mxcsr(env, ldl_p(mem_buf));
> return 4;
> }
> }
>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
next prev parent reply other threads:[~2014-02-25 8:22 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-02-25 7:30 [Qemu-devel] [PATCH 0/3] target-i386 updates Richard Henderson
2014-02-25 7:30 ` [Qemu-devel] [PATCH 1/3] target-i386: Fix CC_OP_CLR vs PF Richard Henderson
2014-02-27 15:53 ` [Qemu-devel] [Qemu-stable] " Michael Roth
2014-02-25 7:30 ` [Qemu-devel] [PATCH 2/3] target-i386: Fix SSE status flag corruption Richard Henderson
2014-02-25 8:22 ` Paolo Bonzini [this message]
2014-02-25 7:30 ` [Qemu-devel] [PATCH 3/3] target-i386: Fix ucomis and comis memory access 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=530C52B0.6070602@redhat.com \
--to=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=rth@twiddle.net \
/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 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).