Skip to content

Commit

Permalink
Improve API of literal save and the snapshot (command line) tool (#2507)
Browse files Browse the repository at this point in the history 
Removed 'jerry_parse_and_save_literals' and introduced
'jerry_get_literals_from_snapshot' instead which works
on snapshot buffers rather than source code. Added literal
saving feature to snapshot merge in the snapshot command
line tool. Also added missing 'jerry_cleanup()' calls to the
snapshot tool. Improved the console messages of the snapshot
tool.

Based on previous work of Tamas Zakor <ztamas@inf.u-szeged.hu>

JerryScript-DCO-1.0-Signed-off-by: László Langó llango.u-szeged@partner.samsung.com
  • Loading branch information
@LaszloLango @akosthekiss
LaszloLango authored and akosthekiss committed Sep 6, 2018
1 parent 054717f commit 99c9a22
Show file tree
Hide file tree
Showing 5 changed files with 354 additions and 180 deletions.
59 changes: 34 additions & 25 deletions docs/02.API-REFERENCE.md
View file
Expand Up @@ -678,7 +678,7 @@ main (void)

- [jerry_init](#jerry_init)
- [jerry_cleanup](#jerry_cleanup)
- [jerry_parse_and_save_literals](#jerry_parse_and_save_literals)
- [jerry_get_literals_from_snapshot](#jerry_get_literals_from_snapshot)


## jerry_get_memory_stats
Expand Down Expand Up @@ -5349,30 +5349,28 @@ main (void)
- [jerry_parse_and_save_function_snapshot](#jerry_parse_and_save_function_snapshot)


## jerry_parse_and_save_literals
## jerry_get_literals_from_snapshot

**Summary**

Collect the used literals from the given source code and save them into a specific file in a list or C format.
These literals are generated by the parser, they are valid identifiers and none of them are magic string.
Collect the used literals from the given snapshot and save them into a buffer in list or C format.
None of these literals are magic strings. In C format only valid identifiers are collected.

**Prototype**

```c
size_t
jerry_parse_and_save_literals (const jerry_char_t *source_p,
size_t source_size,
bool is_strict,
uint32_t *buffer_p,
size_t buffer_size,
bool is_c_format);
jerry_get_literals_from_snapshot (const uint32_t *snapshot_p,
size_t snapshot_size,
jerry_char_t *lit_buf_p,
size_t lit_buf_size,
bool is_c_format);
```

- `source_p` - script source, it must be a valid utf8 string.
- `source_size` - script source size, in bytes.
- `is_strict` - strict mode.
- `buffer_p` - buffer to save literals to.
- `buffer_size` - the buffer's size.
- `snapshot_p` - input snapshot buffer.
- `snapshot_size` - snapshot size, in bytes.
- `lit_buf_p` - buffer to save literals to.
- `lit_buf_size` - the buffer's size.
- `is_c_format` - the output format would be C-style (true) or a simple list (false).
- return value
- the size of the literal-list, if it was generated succesfully (i.e. the list of literals isn't empty,
Expand All @@ -5393,20 +5391,31 @@ main (void)
{
jerry_init (JERRY_INIT_EMPTY);

static uint32_t save_literal_buffer[256];
static jerry_char_t literal_buffer[256];
static uint32_t snapshot_buffer[256];
const jerry_char_t *code_for_literal_save_p = (const jerry_char_t *) "var obj = { a:'aa', bb:'Bb' }";
size_t code_for_literal_save_size = strlen ((const char *) code_for_literal_save_p);

jerry_value_t generate_result = jerry_generate_snapshot (NULL,
0,
code_for_literal_save_p,
code_for_literal_save_size,
0,
snapshot_buffer,
256);
size_t snapshot_size = (size_t) jerry_get_number_value (generate_result);
jerry_release_value (generate_result);

size_t literal_sizes = jerry_parse_and_save_literals (code_for_literal_save_p,
strlen ((const char *) code_for_literal_save_p),
false,
save_literal_buffer,
sizeof (save_literal_buffer) / sizeof (uint32_t),
true);
const size_t literal_size = jerry_get_literals_from_snapshot (snapshot_buffer,
snapshot_size,
literal_buffer,
256,
true);

if (literal_sizes != 0)
if (literal_size != 0)
{
FILE *literal_file_p = fopen ("literals.txt", "w");
fwrite (save_literal_buffer, sizeof (uint8_t), literal_sizes, literal_file_p);
FILE *literal_file_p = fopen ("literals.h", "wb");
fwrite (literal_buffer, sizeof (uint8_t), literal_size, literal_file_p);
fclose (literal_file_p);
}

Expand Down
138 changes: 66 additions & 72 deletions jerry-core/api/jerry-snapshot.c
View file
Expand Up @@ -1259,7 +1259,7 @@ jerry_merge_snapshots (const uint32_t **inp_buffers_p, /**< array of (pointers t
functions_size += header_p->lit_table_offset - start_offset;

scan_snapshot_functions (data_p + start_offset,
data_p + header_p->lit_table_offset,
literal_base_p,
lit_pool_p,
literal_base_p);
}
Expand Down Expand Up @@ -1582,48 +1582,44 @@ ecma_string_is_valid_identifier (const ecma_string_t *string_p)
#endif /* JERRY_ENABLE_SNAPSHOT_SAVE */

/**
* Copy certain string literals into the given buffer in a specified format,
* which are valid identifiers and none of them are magic string.
* Get the literals from a snapshot. Copies certain string literals into the given
* buffer in a specified format.
*
* Note:
* Only valid identifiers are saved in C format.
*
* @return size of the literal-list in bytes, at most equal to the buffer size,
* if the source parsed successfully and the list of the literals isn't empty,
* if the list of the literals isn't empty,
* 0 - otherwise.
*/
size_t
jerry_parse_and_save_literals (const jerry_char_t *source_p, /**< script source */
size_t source_size, /**< script source size */
bool is_strict, /**< strict mode */
uint32_t *buffer_p, /**< [out] buffer to save literals to */
size_t buffer_size, /**< the buffer's size */
bool is_c_format) /**< format-flag */
jerry_get_literals_from_snapshot (const uint32_t *snapshot_p, /**< input snapshot buffer */
size_t snapshot_size, /**< size of the input snapshot buffer */
jerry_char_t *lit_buf_p, /**< [out] buffer to save literals to */
size_t lit_buf_size, /**< the buffer's size */
bool is_c_format) /**< format-flag */
{
#ifdef JERRY_ENABLE_SNAPSHOT_SAVE
ecma_value_t parse_status;
ecma_compiled_code_t *bytecode_data_p;

#ifdef JERRY_ENABLE_LINE_INFO
JERRY_CONTEXT (resource_name) = ECMA_VALUE_UNDEFINED;
#endif /* JERRY_ENABLE_LINE_INFO */

parse_status = parser_parse_script (NULL,
0,
source_p,
source_size,
is_strict,
&bytecode_data_p);
const uint8_t *snapshot_data_p = (uint8_t *) snapshot_p;
const jerry_snapshot_header_t *header_p = (const jerry_snapshot_header_t *) snapshot_data_p;

if (ECMA_IS_VALUE_ERROR (parse_status))
if (snapshot_size <= sizeof (jerry_snapshot_header_t)
|| header_p->magic != JERRY_SNAPSHOT_MAGIC
|| header_p->version != JERRY_SNAPSHOT_VERSION
|| !snapshot_check_global_flags (header_p->global_flags))
{
ecma_free_value (JERRY_CONTEXT (error_value));
/* Invalid snapshot format */
return 0;
}

ecma_free_value (parse_status);
JERRY_ASSERT ((header_p->lit_table_offset % sizeof (uint32_t)) == 0);
const uint8_t *literal_base_p = snapshot_data_p + header_p->lit_table_offset;

ecma_collection_header_t *lit_pool_p = ecma_new_values_collection ();
ecma_save_literals_add_compiled_code (bytecode_data_p, lit_pool_p);

ecma_bytecode_deref (bytecode_data_p);
scan_snapshot_functions (snapshot_data_p + header_p->func_offsets[0],
literal_base_p,
lit_pool_p,
literal_base_p);

lit_utf8_size_t literal_count = 0;
ecma_value_t *iterator_p = ecma_collection_iterator_init (lit_pool_p);
Expand Down Expand Up @@ -1657,10 +1653,8 @@ jerry_parse_and_save_literals (const jerry_char_t *source_p, /**< script source
return 0;
}

uint8_t *destination_p = (uint8_t *) buffer_p;

uint8_t *const buffer_start_p = destination_p;
uint8_t *const buffer_end_p = destination_p + buffer_size;
jerry_char_t *const buffer_start_p = lit_buf_p;
jerry_char_t *const buffer_end_p = lit_buf_p + lit_buf_size;

JMEM_DEFINE_LOCAL_ARRAY (literal_array, literal_count, ecma_string_t *);
lit_utf8_size_t literal_idx = 0;
Expand Down Expand Up @@ -1697,43 +1691,43 @@ jerry_parse_and_save_literals (const jerry_char_t *source_p, /**< script source
if (is_c_format)
{
/* Save literal count. */
destination_p = jerry_append_chars_to_buffer (destination_p,
buffer_end_p,
"jerry_length_t literal_count = ",
0);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p,
buffer_end_p,
"jerry_length_t literal_count = ",
0);

destination_p = jerry_append_number_to_buffer (destination_p, buffer_end_p, literal_count);
lit_buf_p = jerry_append_number_to_buffer (lit_buf_p, buffer_end_p, literal_count);

/* Save the array of literals. */
destination_p = jerry_append_chars_to_buffer (destination_p,
buffer_end_p,
";\n\njerry_char_t *literals[",
0);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p,
buffer_end_p,
";\n\njerry_char_t *literals[",
0);

destination_p = jerry_append_number_to_buffer (destination_p, buffer_end_p, literal_count);
destination_p = jerry_append_chars_to_buffer (destination_p, buffer_end_p, "] =\n{\n", 0);
lit_buf_p = jerry_append_number_to_buffer (lit_buf_p, buffer_end_p, literal_count);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p, buffer_end_p, "] =\n{\n", 0);

for (lit_utf8_size_t i = 0; i < literal_count; i++)
{
destination_p = jerry_append_chars_to_buffer (destination_p, buffer_end_p, " \"", 0);
destination_p = jerry_append_ecma_string_to_buffer (destination_p, buffer_end_p, literal_array[i]);
destination_p = jerry_append_chars_to_buffer (destination_p, buffer_end_p, "\"", 0);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p, buffer_end_p, " \"", 0);
lit_buf_p = jerry_append_ecma_string_to_buffer (lit_buf_p, buffer_end_p, literal_array[i]);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p, buffer_end_p, "\"", 0);

if (i < literal_count - 1)
{
destination_p = jerry_append_chars_to_buffer (destination_p, buffer_end_p, ",", 0);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p, buffer_end_p, ",", 0);
}

destination_p = jerry_append_chars_to_buffer (destination_p, buffer_end_p, "\n", 0);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p, buffer_end_p, "\n", 0);
}

destination_p = jerry_append_chars_to_buffer (destination_p,
buffer_end_p,
"};\n\njerry_length_t literal_sizes[",
0);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p,
buffer_end_p,
"};\n\njerry_length_t literal_sizes[",
0);

destination_p = jerry_append_number_to_buffer (destination_p, buffer_end_p, literal_count);
destination_p = jerry_append_chars_to_buffer (destination_p, buffer_end_p, "] =\n{\n", 0);
lit_buf_p = jerry_append_number_to_buffer (lit_buf_p, buffer_end_p, literal_count);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p, buffer_end_p, "] =\n{\n", 0);
}

/* Save the literal sizes respectively. */
Expand All @@ -1743,51 +1737,51 @@ jerry_parse_and_save_literals (const jerry_char_t *source_p, /**< script source

if (is_c_format)
{
destination_p = jerry_append_chars_to_buffer (destination_p, buffer_end_p, " ", 0);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p, buffer_end_p, " ", 0);
}

destination_p = jerry_append_number_to_buffer (destination_p, buffer_end_p, str_size);
destination_p = jerry_append_chars_to_buffer (destination_p, buffer_end_p, " ", 0);
lit_buf_p = jerry_append_number_to_buffer (lit_buf_p, buffer_end_p, str_size);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p, buffer_end_p, " ", 0);

if (is_c_format)
{
/* Show the given string as a comment. */
destination_p = jerry_append_chars_to_buffer (destination_p, buffer_end_p, "/* ", 0);
destination_p = jerry_append_ecma_string_to_buffer (destination_p, buffer_end_p, literal_array[i]);
destination_p = jerry_append_chars_to_buffer (destination_p, buffer_end_p, " */", 0);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p, buffer_end_p, "/* ", 0);
lit_buf_p = jerry_append_ecma_string_to_buffer (lit_buf_p, buffer_end_p, literal_array[i]);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p, buffer_end_p, " */", 0);

if (i < literal_count - 1)
{
destination_p = jerry_append_chars_to_buffer (destination_p, buffer_end_p, ",", 0);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p, buffer_end_p, ",", 0);
}
}
else
{
destination_p = jerry_append_ecma_string_to_buffer (destination_p, buffer_end_p, literal_array[i]);
lit_buf_p = jerry_append_ecma_string_to_buffer (lit_buf_p, buffer_end_p, literal_array[i]);
}

destination_p = jerry_append_chars_to_buffer (destination_p, buffer_end_p, "\n", 0);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p, buffer_end_p, "\n", 0);
}

if (is_c_format)
{
destination_p = jerry_append_chars_to_buffer (destination_p, buffer_end_p, "};\n", 0);
lit_buf_p = jerry_append_chars_to_buffer (lit_buf_p, buffer_end_p, "};\n", 0);
}

JMEM_FINALIZE_LOCAL_ARRAY (literal_array);

return destination_p <= buffer_end_p ? (size_t) (destination_p - buffer_start_p) : 0;
return lit_buf_p <= buffer_end_p ? (size_t) (lit_buf_p - buffer_start_p) : 0;
#else /* !JERRY_ENABLE_SNAPSHOT_SAVE */
JERRY_UNUSED (source_p);
JERRY_UNUSED (source_size);
JERRY_UNUSED (is_strict);
JERRY_UNUSED (buffer_p);
JERRY_UNUSED (buffer_size);
JERRY_UNUSED (snapshot_p);
JERRY_UNUSED (snapshot_size);
JERRY_UNUSED (lit_buf_p);
JERRY_UNUSED (lit_buf_size);
JERRY_UNUSED (is_c_format);

return 0;
#endif /* JERRY_ENABLE_SNAPSHOT_SAVE */
} /* jerry_parse_and_save_literals */
} /* jerry_get_literals_from_snapshot */


/**
* Generate snapshot function from specified source and arguments
Expand Down
4 changes: 2 additions & 2 deletions jerry-core/include/jerryscript-snapshot.h
View file
Expand Up @@ -65,8 +65,8 @@ jerry_value_t jerry_load_function_snapshot (const uint32_t *function_snapshot_p,

size_t jerry_merge_snapshots (const uint32_t **inp_buffers_p, size_t *inp_buffer_sizes_p, size_t number_of_snapshots,
uint32_t *out_buffer_p, size_t out_buffer_size, const char **error_p);
size_t jerry_parse_and_save_literals (const jerry_char_t *source_p, size_t source_size, bool is_strict,
uint32_t *buffer_p, size_t buffer_size, bool is_c_format);
size_t jerry_get_literals_from_snapshot (const uint32_t *snapshot_p, size_t snapshot_size,
jerry_char_t *lit_buf_p, size_t lit_buf_size, bool is_c_format);
/**
* @}
*/
Expand Down

0 comments on commit 99c9a22

Please sign in to comment.