qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Blue Swirl <blauwirbel@gmail.com>
To: Stefan Weil <weil@mail.berlios.de>
Cc: qemu-devel <qemu-devel@nongnu.org>
Subject: Re: [Qemu-devel] [PATCH] tcg: Add consistency checks for op definitions
Date: Thu, 18 Feb 2010 21:30:01 +0200	[thread overview]
Message-ID: <f43fc5581002181130t5eead4cfw35f591aac07483d7@mail.gmail.com> (raw)
In-Reply-To: <1266250641-32585-1-git-send-email-weil@mail.berlios.de>

Thanks, applied.

On Mon, Feb 15, 2010 at 6:17 PM, Stefan Weil <weil@mail.berlios.de> wrote:
> When compiled with CONFIG_DEBUG_TCG, this code looks
> for missing, duplicate and wrong entries in the
> op definitions.
>
> Errors will raise an assertion at program start
> (all checks are done in the initial phase).
>
> The current code contains such errors, at least for
> i386 guest on i386 host.
>
> Signed-off-by: Stefan Weil <weil@mail.berlios.de>
> ---
>  tcg/tcg.c |   21 +++++++++++++++++++++
>  tcg/tcg.h |    3 +++
>  2 files changed, 24 insertions(+), 0 deletions(-)
>
> diff --git a/tcg/tcg.c b/tcg/tcg.c
> index 9949814..e6a1caf 100644
> --- a/tcg/tcg.c
> +++ b/tcg/tcg.c
> @@ -981,9 +981,16 @@ void tcg_add_target_add_op_defs(const TCGTargetOpDef *tdefs)
>         op = tdefs->op;
>         assert(op >= 0 && op < NB_OPS);
>         def = &tcg_op_defs[op];
> +#if defined(CONFIG_DEBUG_TCG)
> +        /* Duplicate entry in op definitions? */
> +        assert(!def->used);
> +        def->used = 1;
> +#endif
>         nb_args = def->nb_iargs + def->nb_oargs;
>         for(i = 0; i < nb_args; i++) {
>             ct_str = tdefs->args_ct_str[i];
> +            /* Incomplete TCGTargetOpDef entry? */
> +            assert(ct_str != NULL);
>             tcg_regset_clear(def->args_ct[i].u.regs);
>             def->args_ct[i].ct = 0;
>             if (ct_str[0] >= '0' && ct_str[0] <= '9') {
> @@ -1018,6 +1025,9 @@ void tcg_add_target_add_op_defs(const TCGTargetOpDef *tdefs)
>             }
>         }
>
> +        /* TCGTargetOpDef entry with too much information? */
> +        assert(i == TCG_MAX_OP_ARGS || tdefs->args_ct_str[i] == NULL);
> +
>         /* sort the constraints (XXX: this is just an heuristic) */
>         sort_constraints(def, 0, def->nb_oargs);
>         sort_constraints(def, def->nb_oargs, def->nb_iargs);
> @@ -1035,6 +1045,17 @@ void tcg_add_target_add_op_defs(const TCGTargetOpDef *tdefs)
>         tdefs++;
>     }
>
> +#if defined(CONFIG_DEBUG_TCG)
> +    for (op = 0; op < ARRAY_SIZE(tcg_op_defs); op++) {
> +        if (op < INDEX_op_call || op == INDEX_op_debug_insn_start) {
> +            /* Wrong entry in op definitions? */
> +            assert(!tcg_op_defs[op].used);
> +        } else {
> +            /* Missing entry in op definitions? */
> +            assert(tcg_op_defs[op].used);
> +        }
> +    }
> +#endif
>  }
>
>  #ifdef USE_LIVENESS_ANALYSIS
> diff --git a/tcg/tcg.h b/tcg/tcg.h
> index b218abe..aca9f27 100644
> --- a/tcg/tcg.h
> +++ b/tcg/tcg.h
> @@ -404,6 +404,9 @@ typedef struct TCGOpDef {
>     uint16_t copy_size;
>     TCGArgConstraint *args_ct;
>     int *sorted_args;
> +#if defined(CONFIG_DEBUG_TCG)
> +    int used;
> +#endif
>  } TCGOpDef;
>
>  typedef struct TCGTargetOpDef {
> --
> 1.6.6.1
>
>
>
>

      reply	other threads:[~2010-02-18 19:30 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-02-15 16:17 [Qemu-devel] [PATCH] tcg: Add consistency checks for op definitions Stefan Weil
2010-02-18 19:30 ` Blue Swirl [this message]

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=f43fc5581002181130t5eead4cfw35f591aac07483d7@mail.gmail.com \
    --to=blauwirbel@gmail.com \
    --cc=qemu-devel@nongnu.org \
    --cc=weil@mail.berlios.de \
    /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).