--- Begin Message ---
|
Subject: |
struct charset.code_space[15] contains garbage |
|
Date: |
Sun, 01 May 2011 09:59:25 -0700 |
|
User-agent: |
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.2.14) Gecko/20110223 Thunderbird/3.1.8 |
While testing the 32+64 port I noticed that a too-wide value
is stuffed into charset.code_space[15], which means that
slot has a garbage value (at least, it's garbage on typical
machines with 32-bit int). As far as I can see, the garbage
value is never used, so it's a bit cleaner to never compute
or store it.
I plan to install the following patch to do that.
This patch is relevant to ordinary 32- and 64-bit hosts, too,
so I'm separating it out.
* charset.h (struct charset.code_space): Now has 15 elements, not 16.
* charset.c (Fdefine_charset_internal): Don't initialize
charset.code_space[15]. The value was garbage, on hosts with
32-bit int.
=== modified file 'src/charset.c'
--- src/charset.c 2011-04-26 06:17:52 +0000
+++ src/charset.c 2011-05-01 06:28:23 +0000
@@ -869,7 +869,7 @@
ASET (attrs, charset_name, args[charset_arg_name]);
val = args[charset_arg_code_space];
- for (i = 0, dimension = 0, nchars = 1; i < 4; i++)
+ for (i = 0, dimension = 0, nchars = 1; ; i++)
{
int min_byte, max_byte;
@@ -880,10 +880,12 @@
charset.code_space[i * 4] = min_byte;
charset.code_space[i * 4 + 1] = max_byte;
charset.code_space[i * 4 + 2] = max_byte - min_byte + 1;
+ if (max_byte > 0)
+ dimension = i + 1;
+ if (i == 3)
+ break;
nchars *= charset.code_space[i * 4 + 2];
charset.code_space[i * 4 + 3] = nchars;
- if (max_byte > 0)
- dimension = i + 1;
}
val = args[charset_arg_dimension];
=== modified file 'src/charset.h'
--- src/charset.h 2011-04-11 06:48:18 +0000
+++ src/charset.h 2011-05-01 16:22:33 +0000
@@ -155,10 +155,11 @@
byte code of the (N+1)th dimension, <code_space>[4N+1] is a
maximum byte code of the (N+1)th dimension, <code_space>[4N+2] is
(<code_space>[4N+1] - <code_space>[4N] + 1), <code_space>[4N+3]
- is a number of characters containd in the first to (N+1)th
- dismesions. We get `char-index' of a `code-point' from this
+ is the number of characters contained in the first through (N+1)th
+ dimensions, except that there is no <code_space>[15].
+ We get `char-index' of a `code-point' from this
information. */
- int code_space[16];
+ int code_space[15];
/* If B is a byte of Nth dimension of a code-point, the (N-1)th bit
of code_space_mask[B] is set. This array is used to quickly
--- End Message ---
--- Begin Message ---
|
Subject: |
Merged fixes for 8600, 8601, 8602, and (partially) for 8545 |
|
Date: |
Fri, 06 May 2011 00:29:56 -0700 |
|
User-agent: |
Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.2.17) Gecko/20110424 Thunderbird/3.1.10 |
I committed to the Emacs trunk a merge (bzr 104134) that has fixes for
the following bugs:
* Bug#8600 - The fix removes the garbage element of code_space.
* Bug#8601 - Here I assumed that the "* 2" is a typo.
* Bug#8602 - This fixes some large-int-to-float screwups in
the Lisp reader.
* Bug#8545 - This fixes the bug where the code should have called
va_copy, but didn't. Also, I changed a limit so that
the MOST_POSITIVE_FIXNUM limit for strings applies to
their length, i.e., does not include the null termination
byte. Stefan hasn't had time to chime in, but if this
last change turns out to be incorrect I will back it out.
This merge doesn't entirely fix Bug#8545, so I'll leave that bug open;
the others I'll close.
--- End Message ---