Re: [RFA] Process record and replay, 8/10

[Eli Zaretskii <eliz@gnu.org>]

> Date: Thu, 6 Nov 2008 15:50:44 +0800
> From: teawater <teawater@gmail.com>
> 
> This patch add code to make I386 architecture support process record and replay.

Thanks.

> +      printf_unfiltered (_("Process record: read memeory 0x%s error.\n"),
                                                    ^^^^^^^
A typo.  (There are several more like it.)

Also, I suggest to say "error 0x%s", not "0x%s error".  The latter is
confusing for the ears of an English speaker, I think.

> +      /* XXX: index == 4 is always invalid */

Why the XXX in this comment?

> +      /* arith & logic */
> +    case 0x00 ... 0x05:
> +    case 0x08 ... 0x0d:
> +    case 0x10 ... 0x15:
> +    case 0x18 ... 0x1d:
> +    case 0x20 ... 0x25:
> +    case 0x28 ... 0x2d:
> +    case 0x30 ... 0x35:
> +    case 0x38 ... 0x3d:

Is this valid ISO C?

> +	    if (record_debug)
> +	      printf_unfiltered (_
> +				 ("Process record ignores the memory change of instruction in address 0x%s because it can't get the value of the segment register.\n"),
                                                                                           ^^^^^^^^^^
"at address".

By the way, do we need debug messages to be translatable?  Other
similar places in the patches don't have them in _().

> +    case 0x9b:
> +      printf_unfiltered (_
> +			 ("Process record don't support instruction fwait.\n"));
                                          ^^^^^^^^^^^^^
"doesn't support"

By the way, what happens if the code stream includes one of these
``unsupported'' instructions?  What will the user see at replay time?