[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [qemu-s390x] [PATCH v3 8/9] s390x/tcg: fix FP register pair checks
From: |
Thomas Huth |
Subject: |
Re: [qemu-s390x] [PATCH v3 8/9] s390x/tcg: fix FP register pair checks |
Date: |
Thu, 27 Sep 2018 13:29:26 +0200 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.9.1 |
On 2018-09-17 11:34, David Hildenbrand wrote:
> Valid register pairs are 0/2, 1/3, 4/6, 5/7, 8/10, 9/11, 12/14, 13/15.
>
> R1/R2 always selects the lower number, so the current checks are not
> correct as e.g. 2/4 could be selected as a pair.
>
> Reviewed-by: Richard Henderson <address@hidden>
> Signed-off-by: David Hildenbrand <address@hidden>
> ---
> target/s390x/translate.c | 22 +++++++++++-----------
> 1 file changed, 11 insertions(+), 11 deletions(-)
>
> diff --git a/target/s390x/translate.c b/target/s390x/translate.c
> index f9a78c4304..5cc65b0840 100644
> --- a/target/s390x/translate.c
> +++ b/target/s390x/translate.c
> @@ -1110,7 +1110,7 @@ typedef struct {
> #define IF_HFP3 0x0004 /* r3 points at fp reg for HFP instructions
> */
> #define IF_BFP 0x0008 /* binary floating point instruction */
> #define IF_DFP 0x0010 /* decimal floating point instruction */
> -#define IF_PRIV 0x0020 /* priviledged instruction */
> +#define IF_PRIV 0x0020 /* privileged instruction */
Ah, well, merge this with the previous patch, please.
> struct DisasInsn {
> unsigned opc:16;
> @@ -5985,6 +5985,12 @@ static bool is_afp_reg(int reg)
> return reg % 2 || reg > 6;
> }
>
> +static bool is_fp_pair(int reg)
> +{
> + /* 0,1,4,5,8,9,12,13: to exclude the others, check for single bit */
> + return !(reg & 0x2);
> +}
> +
> static DisasJumpType translate_one(CPUS390XState *env, DisasContext *s)
> {
> const DisasInsn *insn;
> @@ -6067,17 +6073,11 @@ static DisasJumpType translate_one(CPUS390XState
> *env, DisasContext *s)
> excp = PGM_SPECIFICATION;
> }
> }
> - if (spec & SPEC_r1_f128) {
> - r = get_field(&f, r1);
> - if (r > 13) {
> - excp = PGM_SPECIFICATION;
> - }
> + if (spec & SPEC_r1_f128 && !is_fp_pair(get_field(&f, r1))) {
> + excp = PGM_SPECIFICATION;
> }
> - if (spec & SPEC_r2_f128) {
> - r = get_field(&f, r2);
> - if (r > 13) {
> - excp = PGM_SPECIFICATION;
> - }
> + if (spec & SPEC_r2_f128 && !is_fp_pair(get_field(&f, r2))) {
> + excp = PGM_SPECIFICATION;
> }
> if (excp) {
> gen_program_exception(s, excp);
>
Without the "priviledged" hunk:
Reviewed-by: Thomas Huth <address@hidden>
- Re: [qemu-s390x] [PATCH v3 2/9] s390x/tcg: factor out and fix DATA exception injection, (continued)
- [qemu-s390x] [PATCH v3 3/9] s390x/tcg: store in the TB flags if AFP is enabled, David Hildenbrand, 2018/09/17
- [qemu-s390x] [PATCH v3 4/9] s390x/tcg: support flags for instructions, David Hildenbrand, 2018/09/17
- [qemu-s390x] [PATCH v3 5/9] s390x/tcg: add instruction flags for floating point instructions, David Hildenbrand, 2018/09/17
- [qemu-s390x] [PATCH v3 6/9] s390x/tcg: check for AFP-register, BFP and DFP data exceptions, David Hildenbrand, 2018/09/17
- [qemu-s390x] [PATCH v3 8/9] s390x/tcg: fix FP register pair checks, David Hildenbrand, 2018/09/17
- Re: [qemu-s390x] [PATCH v3 8/9] s390x/tcg: fix FP register pair checks,
Thomas Huth <=
- [qemu-s390x] [PATCH v3 9/9] s390x/tcg: refactor specification checking, David Hildenbrand, 2018/09/17
- [qemu-s390x] [PATCH v3 7/9] s390x/tcg: handle privileged instructions via flags, David Hildenbrand, 2018/09/17