[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Guile-commits] GNU Guile branch, boehm-demers-weiser-gc, updated. relea
From: |
Ludovic Courtès |
Subject: |
[Guile-commits] GNU Guile branch, boehm-demers-weiser-gc, updated. release_1-9-2-280-gba54a20 |
Date: |
Tue, 01 Sep 2009 00:07:09 +0000 |
This is an automated email from the git hooks/post-receive script. It was
generated because a ref change was pushed to the repository containing
the project "GNU Guile".
http://git.savannah.gnu.org/cgit/guile.git/commit/?id=ba54a2026beaadb4e7566d4b9e2c9e4c7cd793e6
The branch, boehm-demers-weiser-gc has been updated
via ba54a2026beaadb4e7566d4b9e2c9e4c7cd793e6 (commit)
via 13a9455669c2a8d1e4ed59cb8736bf23e91eaa55 (commit)
from 4812ce85ddf1d04c49436ada34152ac7751a8b50 (commit)
Those revisions listed above that are new to this repository have
not appeared on any other notification email; so we list those
revisions in full, below.
- Log -----------------------------------------------------------------
commit ba54a2026beaadb4e7566d4b9e2c9e4c7cd793e6
Author: Ludovic Courtès <address@hidden>
Date: Tue Sep 1 02:02:43 2009 +0200
Remove the distinction between inline/outline storage for stringbufs.
* libguile/strings.c (STRINGBUF_HEADER_SIZE, STRINGBUF_HEADER_BYTES):
New macros.
(STRINGBUF_F_INLINE, STRINGBUF_INLINE, STRINGBUF_OUTLINE_CHARS,
STRINGBUF_OUTLINE_LENGTH, STRINGBUF_INLINE_CHARS,
STRINGBUF_INLINE_LENGTH, STRINGBUF_MAX_INLINE_LEN): Remove.
(STRINGBUF_CHARS, STRINGBUF_WIDE_CHARS): Adjust to return a fixed
location.
(STRINGBUF_LENGTH): Get the length from word 1.
(make_stringbuf, make_wide_stringbuf): Adjust to use a contiguous
memory region.
(wide_stringbuf): Renamed from `widen_stringbuf'. Adjust similarly.
Return the new stringbuf. Callers updated.
(narrow_stringbuf): Likewise.
(scm_sys_string_dump, scm_sys_symbol_dump): Remove `stringbuf-inline'
pair.
* test-suite/tests/strings.test ("string internals")["null strings are
inlined", "short Latin-1 encoded strings are inlined", "long Latin-1
encoded strings are not inlined", "short UCS-4 encoded strings are not
inlined", "long UCS-4 encoded strings are not inlined"]: Remove.
* test-suite/tests/symbols.test ("symbol internals")["null symbols are
inlined", "short Latin-1 encoded symbols are inlined", "long Latin-1
encoded symbols are not inlined", "short UCS-4 encoded symbols are not
inlined", "long UCS-4 encoded symbols are not inlined"]: Remove.
commit 13a9455669c2a8d1e4ed59cb8736bf23e91eaa55
Author: Ludovic Courtès <address@hidden>
Date: Tue Sep 1 00:38:40 2009 +0200
Fix leaky handling of `scm_take_locale_{symbol,string} ()'.
* libguile/strings.c (scm_i_take_stringbufn, scm_i_c_take_symbol):
Remove.
(scm_take_locale_stringn): Rewrite in terms of `scm_from_locale_stringn
()'.
* libguile/strings.h (scm_i_c_take_symbol, scm_i_take_stringbufn):
Remove declarations.
-----------------------------------------------------------------------
Summary of changes:
libguile/strings.c | 258 +++++++++++++++--------------------------
libguile/strings.h | 4 -
test-suite/tests/strings.test | 20 ---
test-suite/tests/symbols.test | 19 ---
4 files changed, 94 insertions(+), 207 deletions(-)
diff --git a/libguile/strings.c b/libguile/strings.c
index dfa0690..5199981 100644
--- a/libguile/strings.c
+++ b/libguile/strings.c
@@ -61,49 +61,32 @@
* cow-strings, but it failed randomly with more than 10 threads, say.
* I couldn't figure out what went wrong, so I used the conservative
* approach implemented below.
- *
- * A stringbuf needs to know its length, but only so that it can be
- * reported when the stringbuf is freed.
- *
- * There are 3 storage strategies for stringbufs: inline, outline, and
- * wide.
- *
- * Inline strings are small 8-bit strings stored within the double
- * cell itself. Outline strings are larger 8-bit strings with GC
- * allocated storage. Wide strings are 32-bit strings with allocated
- * storage.
*
- * There was little value in making wide string inlineable, since
- * there is only room for three inlined 32-bit characters. Thus wide
- * stringbufs are never inlined.
+ * There are 2 storage strategies for stringbufs: 8-bit and wide. 8-bit
+ * strings are ISO-8859-1-encoded strings; wide strings are 32-bit (UCS-4)
+ * strings.
*/
+/* The size in words of the stringbuf header (type tag + size). */
+#define STRINGBUF_HEADER_SIZE 2U
+
+#define STRINGBUF_HEADER_BYTES (STRINGBUF_HEADER_SIZE * sizeof (SCM))
+
#define STRINGBUF_F_SHARED 0x100
-#define STRINGBUF_F_INLINE 0x200
#define STRINGBUF_F_WIDE 0x400 /* If true, strings have UCS-4
encoding. Otherwise, strings
are Latin-1. */
#define STRINGBUF_TAG scm_tc7_stringbuf
#define STRINGBUF_SHARED(buf) (SCM_CELL_WORD_0(buf) & STRINGBUF_F_SHARED)
-#define STRINGBUF_INLINE(buf) (SCM_CELL_WORD_0(buf) & STRINGBUF_F_INLINE)
#define STRINGBUF_WIDE(buf) (SCM_CELL_WORD_0(buf) & STRINGBUF_F_WIDE)
-#define STRINGBUF_OUTLINE_CHARS(buf) ((unsigned char *) SCM_CELL_WORD_1(buf))
-#define STRINGBUF_OUTLINE_LENGTH(buf) (SCM_CELL_WORD_2(buf))
-#define STRINGBUF_INLINE_CHARS(buf) ((unsigned char *)
SCM_CELL_OBJECT_LOC(buf,1))
-#define STRINGBUF_INLINE_LENGTH(buf) (((size_t)SCM_CELL_WORD_0(buf))>>16)
+#define STRINGBUF_CHARS(buf) ((unsigned char *) \
+ SCM_CELL_OBJECT_LOC (buf, \
+ STRINGBUF_HEADER_SIZE))
+#define STRINGBUF_LENGTH(buf) (SCM_CELL_WORD_1 (buf))
-#define STRINGBUF_CHARS(buf) (STRINGBUF_INLINE (buf) \
- ? STRINGBUF_INLINE_CHARS (buf) \
- : STRINGBUF_OUTLINE_CHARS (buf))
-
-#define STRINGBUF_WIDE_CHARS(buf) ((scm_t_wchar *) SCM_CELL_WORD_1(buf))
-#define STRINGBUF_LENGTH(buf) (STRINGBUF_INLINE (buf) \
- ? STRINGBUF_INLINE_LENGTH (buf) \
- : STRINGBUF_OUTLINE_LENGTH (buf))
-
-#define STRINGBUF_MAX_INLINE_LEN (3*sizeof(scm_t_bits))
+#define STRINGBUF_WIDE_CHARS(buf) ((scm_t_wchar *) STRINGBUF_CHARS (buf))
#define SET_STRINGBUF_SHARED(buf) \
(SCM_SET_CELL_WORD_0 ((buf), SCM_CELL_WORD_0 (buf) | STRINGBUF_F_SHARED))
@@ -124,6 +107,8 @@ make_stringbuf (size_t len)
can be dropped.
*/
+ SCM buf;
+
#if SCM_STRING_LENGTH_HISTOGRAM
if (len < 1000)
lenhist[len]++;
@@ -131,18 +116,15 @@ make_stringbuf (size_t len)
lenhist[1000]++;
#endif
- if (len <= STRINGBUF_MAX_INLINE_LEN-1)
- {
- return scm_double_cell (STRINGBUF_TAG | STRINGBUF_F_INLINE | (len << 16),
- 0, 0, 0);
- }
- else
- {
- char *mem = scm_gc_malloc_pointerless (len + 1, "string");
- mem[len] = '\0';
- return scm_double_cell (STRINGBUF_TAG, (scm_t_bits) mem,
- (scm_t_bits) len, (scm_t_bits) 0);
- }
+ buf = PTR2SCM (scm_gc_malloc_pointerless (STRINGBUF_HEADER_BYTES + len + 1,
+ "string"));
+
+ SCM_SET_CELL_TYPE (buf, STRINGBUF_TAG);
+ SCM_SET_CELL_WORD_1 (buf, (scm_t_bits) len);
+
+ STRINGBUF_CHARS (buf)[len] = 0;
+
+ return buf;
}
/* Make a stringbuf with space for LEN 32-bit UCS-4-encoded
@@ -150,7 +132,9 @@ make_stringbuf (size_t len)
static SCM
make_wide_stringbuf (size_t len)
{
- scm_t_wchar *mem;
+ SCM buf;
+ size_t raw_len;
+
#if SCM_STRING_LENGTH_HISTOGRAM
if (len < 1000)
lenhist[len]++;
@@ -158,99 +142,82 @@ make_wide_stringbuf (size_t len)
lenhist[1000]++;
#endif
- mem = scm_gc_malloc (sizeof (scm_t_wchar) * (len + 1), "string");
- mem[len] = 0;
- return scm_double_cell (STRINGBUF_TAG | STRINGBUF_F_WIDE, (scm_t_bits) mem,
- (scm_t_bits) len, (scm_t_bits) 0);
-}
+ raw_len = (len + 1) * sizeof (scm_t_wchar);
+ buf = PTR2SCM (scm_gc_malloc_pointerless (STRINGBUF_HEADER_BYTES + raw_len,
+ "string"));
-/* Return a new stringbuf whose underlying storage consists of the LEN+1
- octets pointed to by STR (the last octet is zero). */
-SCM
-scm_i_take_stringbufn (char *str, size_t len)
-{
- scm_gc_register_collectable_memory (str, len + 1, "stringbuf");
+ SCM_SET_CELL_TYPE (buf, STRINGBUF_TAG | STRINGBUF_F_WIDE);
+ SCM_SET_CELL_WORD_1 (buf, (scm_t_bits) len);
- return scm_double_cell (STRINGBUF_TAG, (scm_t_bits) str,
- (scm_t_bits) len, (scm_t_bits) 0);
+ STRINGBUF_WIDE_CHARS (buf)[len] = 0;
+
+ return buf;
}
-/* Convert a stringbuf containing 8-bit Latin-1-encoded characters to
- one containing 32-bit UCS-4-encoded characters. */
-static void
-widen_stringbuf (SCM buf)
+/* Return a UCS-4-encoded stringbuf containing the (possibly Latin-1-encoded)
+ characters from BUF. */
+static SCM
+wide_stringbuf (SCM buf)
{
- size_t i, len;
- scm_t_wchar *mem;
+ SCM new_buf;
if (STRINGBUF_WIDE (buf))
- return;
-
- if (STRINGBUF_INLINE (buf))
+ new_buf = buf;
+ else
{
- len = STRINGBUF_INLINE_LENGTH (buf);
+ size_t i, len;
+ scm_t_wchar *mem;
- mem = scm_gc_malloc (sizeof (scm_t_wchar) * (len + 1), "string");
- for (i = 0; i < len; i++)
- mem[i] =
- (scm_t_wchar) STRINGBUF_INLINE_CHARS (buf)[i];
- mem[len] = 0;
+ len = STRINGBUF_LENGTH (buf);
- SCM_SET_CELL_WORD_0 (buf, SCM_CELL_WORD_0 (buf) ^ STRINGBUF_F_INLINE);
- SCM_SET_CELL_WORD_0 (buf, SCM_CELL_WORD_0 (buf) | STRINGBUF_F_WIDE);
- SCM_SET_CELL_WORD_1 (buf, mem);
- SCM_SET_CELL_WORD_2 (buf, len);
- }
- else
- {
- len = STRINGBUF_OUTLINE_LENGTH (buf);
+ new_buf = make_wide_stringbuf (len);
- mem = scm_gc_malloc (sizeof (scm_t_wchar) * (len + 1), "string");
+ mem = STRINGBUF_WIDE_CHARS (new_buf);
for (i = 0; i < len; i++)
- mem[i] =
- (scm_t_wchar) STRINGBUF_OUTLINE_CHARS (buf)[i];
+ mem[i] = (scm_t_wchar) STRINGBUF_CHARS (buf)[i];
mem[len] = 0;
-
- scm_gc_free (STRINGBUF_OUTLINE_CHARS (buf), len + 1, "string");
-
- SCM_SET_CELL_WORD_0 (buf, SCM_CELL_WORD_0 (buf) | STRINGBUF_F_WIDE);
- SCM_SET_CELL_WORD_1 (buf, mem);
- SCM_SET_CELL_WORD_2 (buf, len);
}
+
+ return new_buf;
}
-/* Convert a stringbuf of 32-bit UCS-4-encoded characters to one
- containing 8-bit Latin-1-encoded characters, if possible. */
-static void
+/* Return a Latin-1-encoded stringbuf containing the (possibly UCS-4-encoded)
+ characters from BUF, if possible. */
+static SCM
narrow_stringbuf (SCM buf)
{
- size_t i, len;
- scm_t_wchar *wmem;
- char *mem;
+ SCM new_buf;
if (!STRINGBUF_WIDE (buf))
- return;
+ new_buf = buf;
+ else
+ {
+ size_t i, len;
+ scm_t_wchar *wmem;
+ unsigned char *mem;
- len = STRINGBUF_OUTLINE_LENGTH (buf);
- i = 0;
- wmem = STRINGBUF_WIDE_CHARS (buf);
- while (i < len)
- if (wmem[i++] > 0xFF)
- return;
+ len = STRINGBUF_LENGTH (buf);
+ wmem = STRINGBUF_WIDE_CHARS (buf);
+
+ for (i = 0; i < len; i++)
+ if (wmem[i] > 0xFF)
+ /* BUF cannot be narrowed. */
+ return buf;
- mem = scm_gc_malloc (sizeof (char) * (len + 1), "string");
- for (i = 0; i < len; i++)
- mem[i] = (unsigned char) wmem[i];
+ new_buf = make_stringbuf (len);
- scm_gc_free (wmem, sizeof (scm_t_wchar) * (len + 1), "string");
+ mem = STRINGBUF_CHARS (new_buf);
+ for (i = 0; i < len; i++)
+ mem[i] = (unsigned char) wmem[i];
+ mem[len] = 0;
+ }
- SCM_SET_CELL_WORD_0 (buf, SCM_CELL_WORD_0 (buf) ^ STRINGBUF_F_WIDE);
- SCM_SET_CELL_WORD_1 (buf, mem);
- SCM_SET_CELL_WORD_2 (buf, len);
+ return new_buf;
}
scm_i_pthread_mutex_t stringbuf_write_mutex = SCM_I_PTHREAD_MUTEX_INITIALIZER;
+
/* Copy-on-write strings.
*/
@@ -463,7 +430,7 @@ scm_i_is_narrow_string (SCM str)
int
scm_i_try_narrow_string (SCM str)
{
- narrow_stringbuf (STRING_STRINGBUF (str));
+ SET_STRING_STRINGBUF (str, narrow_stringbuf (STRING_STRINGBUF (str)));
return scm_i_is_narrow_string (str);
}
@@ -665,7 +632,7 @@ void
scm_i_string_set_x (SCM str, size_t p, scm_t_wchar chr)
{
if (chr > 0xFF && scm_i_is_narrow_string (str))
- widen_stringbuf (STRING_STRINGBUF (str));
+ SET_STRING_STRINGBUF (str, wide_stringbuf (STRING_STRINGBUF (str)));
if (scm_i_is_narrow_string (str))
{
@@ -679,6 +646,7 @@ scm_i_string_set_x (SCM str, size_t p, scm_t_wchar chr)
}
}
+
/* Symbols.
Basic symbol creation and accessing is done here, the rest is in
@@ -744,18 +712,6 @@ scm_i_c_make_symbol (const char *name, size_t len,
(scm_t_bits) hash, SCM_UNPACK (props));
}
-/* Return a new symbol that uses the LEN bytes pointed to by NAME as its
- underlying storage. */
-SCM
-scm_i_c_take_symbol (char *name, size_t len,
- scm_t_bits flags, unsigned long hash, SCM props)
-{
- SCM buf = scm_i_take_stringbufn (name, len);
-
- return scm_double_cell (scm_tc7_symbol | flags, SCM_UNPACK (buf),
- (scm_t_bits) hash, SCM_UNPACK (props));
-}
-
/* Returns the number of characters in SYM. This may be different
from the memory size of SYM. */
size_t
@@ -860,9 +816,6 @@ SCM_DEFINE (scm_sys_string_dump, "%string-dump", 1, 0, 0,
(SCM str),
"The number of characters in this stringbuf\n"
"@item stringbuf-shared\n"
"@code{#t} if this stringbuf is shared\n"
- "@item stringbuf-inline\n"
- "@code{#t} if this stringbuf's characters are stored in the\n"
- "cell itself, or @code{#f} if they were allocated in memory\n"
"@item stringbuf-wide\n"
"@code{#t} if this stringbuf's characters are stored in a\n"
"32-bit buffer, or @code{#f} if they are stored in an 8-bit\n"
@@ -870,7 +823,7 @@ SCM_DEFINE (scm_sys_string_dump, "%string-dump", 1, 0, 0,
(SCM str),
"@end table")
#define FUNC_NAME s_scm_sys_string_dump
{
- SCM e1, e2, e3, e4, e5, e6, e7, e8, e9, e10;
+ SCM e1, e2, e3, e4, e5, e6, e7, e8, e9;
SCM buf;
SCM_VALIDATE_STRING (1, str);
@@ -930,20 +883,14 @@ SCM_DEFINE (scm_sys_string_dump, "%string-dump", 1, 0, 0,
(SCM str),
else
e8 = scm_cons (scm_from_locale_symbol ("stringbuf-shared"),
SCM_BOOL_F);
- if (STRINGBUF_INLINE (buf))
- e9 = scm_cons (scm_from_locale_symbol ("stringbuf-inline"),
- SCM_BOOL_T);
- else
- e9 = scm_cons (scm_from_locale_symbol ("stringbuf-inline"),
- SCM_BOOL_F);
if (STRINGBUF_WIDE (buf))
- e10 = scm_cons (scm_from_locale_symbol ("stringbuf-wide"),
- SCM_BOOL_T);
+ e9 = scm_cons (scm_from_locale_symbol ("stringbuf-wide"),
+ SCM_BOOL_T);
else
- e10 = scm_cons (scm_from_locale_symbol ("stringbuf-wide"),
- SCM_BOOL_F);
+ e9 = scm_cons (scm_from_locale_symbol ("stringbuf-wide"),
+ SCM_BOOL_F);
- return scm_list_n (e1, e2, e3, e4, e5, e6, e7, e8, e9, e10, SCM_UNDEFINED);
+ return scm_list_n (e1, e2, e3, e4, e5, e6, e7, e8, e9, SCM_UNDEFINED);
}
#undef FUNC_NAME
@@ -963,9 +910,6 @@ SCM_DEFINE (scm_sys_symbol_dump, "%symbol-dump", 1, 0, 0,
(SCM sym),
"The number of characters in this stringbuf\n"
"@item stringbuf-shared\n"
"@code{#t} if this stringbuf is shared\n"
- "@item stringbuf-inline\n"
- "@code{#t} if this stringbuf's characters are stored in the\n"
- "cell itself, or @code{#f} if they were allocated in memory\n"
"@item stringbuf-wide\n"
"@code{#t} if this stringbuf's characters are stored in a\n"
"32-bit buffer, or @code{#f} if they are stored in an 8-bit\n"
@@ -973,7 +917,7 @@ SCM_DEFINE (scm_sys_symbol_dump, "%symbol-dump", 1, 0, 0,
(SCM sym),
"@end table")
#define FUNC_NAME s_scm_sys_symbol_dump
{
- SCM e1, e2, e3, e4, e5, e6, e7, e8;
+ SCM e1, e2, e3, e4, e5, e6, e7;
SCM buf;
SCM_VALIDATE_SYMBOL (1, sym);
e1 = scm_cons (scm_from_locale_symbol ("symbol"),
@@ -1012,19 +956,13 @@ SCM_DEFINE (scm_sys_symbol_dump, "%symbol-dump", 1, 0,
0, (SCM sym),
else
e6 = scm_cons (scm_from_locale_symbol ("stringbuf-shared"),
SCM_BOOL_F);
- if (STRINGBUF_INLINE (buf))
- e7 = scm_cons (scm_from_locale_symbol ("stringbuf-inline"),
- SCM_BOOL_T);
- else
- e7 = scm_cons (scm_from_locale_symbol ("stringbuf-inline"),
- SCM_BOOL_F);
if (STRINGBUF_WIDE (buf))
- e8 = scm_cons (scm_from_locale_symbol ("stringbuf-wide"),
+ e7 = scm_cons (scm_from_locale_symbol ("stringbuf-wide"),
SCM_BOOL_T);
else
- e8 = scm_cons (scm_from_locale_symbol ("stringbuf-wide"),
+ e7 = scm_cons (scm_from_locale_symbol ("stringbuf-wide"),
SCM_BOOL_F);
- return scm_list_n (e1, e2, e3, e4, e5, e6, e7, e8, SCM_UNDEFINED);
+ return scm_list_n (e1, e2, e3, e4, e5, e6, e7, SCM_UNDEFINED);
}
#undef FUNC_NAME
@@ -1556,25 +1494,17 @@ scm_i_from_utf8_string (const scm_t_uint8 *str)
/* Create a new scheme string from the C string STR. The memory of
STR may be used directly as storage for the new string. */
+/* FIXME: GC-wise, the only way to use the memory area pointed to by STR
+ would be to register a finalizer to eventually free(3) STR, which isn't
+ worth it. Should we just deprecate the `scm_take_' functions? */
SCM
scm_take_locale_stringn (char *str, size_t len)
{
- SCM buf, res;
+ SCM res;
- if (len == (size_t) -1)
- len = strlen (str);
- else
- {
- /* Ensure STR is null terminated. A realloc for 1 extra byte should
- often be satisfied from the alignment padding after the block, with
- no actual data movement. */
- str = scm_realloc (str, len + 1);
- str[len] = '\0';
- }
+ res = scm_from_locale_stringn (str, len);
+ free (str);
- buf = scm_i_take_stringbufn (str, len);
- res = scm_double_cell (STRING_TAG,
- SCM_UNPACK (buf), (scm_t_bits) 0, (scm_t_bits) len);
return res;
}
diff --git a/libguile/strings.h b/libguile/strings.h
index 95dc7ac..fff7c97 100644
--- a/libguile/strings.h
+++ b/libguile/strings.h
@@ -164,9 +164,6 @@ SCM_INTERNAL SCM scm_i_make_symbol (SCM name, scm_t_bits
flags,
SCM_INTERNAL SCM
scm_i_c_make_symbol (const char *name, size_t len,
scm_t_bits flags, unsigned long hash, SCM props);
-SCM_INTERNAL SCM
-scm_i_c_take_symbol (char *name, size_t len,
- scm_t_bits flags, unsigned long hash, SCM props);
SCM_INTERNAL const char *scm_i_symbol_chars (SCM sym);
SCM_INTERNAL const scm_t_wchar *scm_i_symbol_wide_chars (SCM sym);
SCM_INTERNAL size_t scm_i_symbol_length (SCM sym);
@@ -181,7 +178,6 @@ SCM_INTERNAL char **scm_i_allocate_string_pointers (SCM
list);
SCM_INTERNAL void scm_i_get_substring_spec (size_t len,
SCM start, size_t *cstart,
SCM end, size_t *cend);
-SCM_INTERNAL SCM scm_i_take_stringbufn (char *str, size_t len);
/* Debugging functions */
diff --git a/test-suite/tests/strings.test b/test-suite/tests/strings.test
index 3f24537..c78fe55 100644
--- a/test-suite/tests/strings.test
+++ b/test-suite/tests/strings.test
@@ -63,26 +63,6 @@
(let ((s (substring/read-only "zyx" 0)))
(assq-ref (%string-dump s) 'read-only)))
- (pass-if "null strings are inlined"
- (let ((s ""))
- (assq-ref (%string-dump s) 'stringbuf-inline)))
-
- (pass-if "short Latin-1 encoded strings are inlined"
- (let ((s "m"))
- (assq-ref (%string-dump s) 'stringbuf-inline)))
-
- (pass-if "long Latin-1 encoded strings are not inlined"
- (let ((s "0123456789012345678901234567890123456789"))
- (not (assq-ref (%string-dump s) 'stringbuf-inline))))
-
- (pass-if "short UCS-4 encoded strings are not inlined"
- (let ((s "\u0100"))
- (not (assq-ref (%string-dump s) 'stringbuf-inline))))
-
- (pass-if "long UCS-4 encoded strings are not inlined"
- (let ((s "\u010012345678901234567890123456789"))
- (not (assq-ref (%string-dump s) 'stringbuf-inline))))
-
(pass-if "new Latin-1 encoded strings are not shared"
(let ((s "abc"))
(not (assq-ref (%string-dump s) 'stringbuf-shared))))
diff --git a/test-suite/tests/symbols.test b/test-suite/tests/symbols.test
index b6dbb9d..c87aa21 100644
--- a/test-suite/tests/symbols.test
+++ b/test-suite/tests/symbols.test
@@ -49,25 +49,6 @@
(string=? (symbol->string s)
(assq-ref (%symbol-dump s) 'stringbuf-chars))))
- (pass-if "the null symbol is inlined"
- (let ((s '#{}#))
- (assq-ref (%symbol-dump s) 'stringbuf-inline)))
-
- (pass-if "short Latin-1-encoded symbols are inlined"
- (let ((s 'm))
- (assq-ref (%symbol-dump s) 'stringbuf-inline)))
-
- (pass-if "long Latin-1-encoded symbols are not inlined"
- (let ((s 'x0123456789012345678901234567890123456789))
- (not (assq-ref (%symbol-dump s) 'stringbuf-inline))))
-
- (pass-if "short UCS-4-encoded symbols are not inlined"
- (let ((s (string->symbol "\u0100")))
- (not (assq-ref (%symbol-dump s) 'stringbuf-inline))))
-
- (pass-if "long UCS-4-encoded symbols are not inlined"
- (let ((s (string->symbol "\u010012345678901234567890123456789")))
- (not (assq-ref (%symbol-dump s) 'stringbuf-inline))))
(with-test-prefix "hashes"
hooks/post-receive
--
GNU Guile
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- [Guile-commits] GNU Guile branch, boehm-demers-weiser-gc, updated. release_1-9-2-280-gba54a20,
Ludovic Courtès <=