On 5/4/22 15:20, Andreas Schwab wrote:
> On Mai 04 2022, Martin Liška wrote:
>
>> diff --git a/gcc/gengtype-state.cc b/gcc/gengtype-state.cc
>> index ea566af3249..dfd9ea52785 100644
>> --- a/gcc/gengtype-state.cc
>> +++ b/gcc/gengtype-state.cc
>> @@ -473,43 +473,43 @@ read_a_state_token (void)
>> {
>> case 'a':
>> obstack_1grow (&bstring_obstack, '\a');
>> - c = getc (state_file);
>> + getc (state_file);
>> break;
>> case 'b':
>> obstack_1grow (&bstring_obstack, '\b');
>> - c = getc (state_file);
>> + getc (state_file);
>> break;
>> case 't':
>> obstack_1grow (&bstring_obstack, '\t');
>> - c = getc (state_file);
>> + getc (state_file);
>> break;
>> case 'n':
>> obstack_1grow (&bstring_obstack, '\n');
>> - c = getc (state_file);
>> + getc (state_file);
>> break;
>> case 'v':
>> obstack_1grow (&bstring_obstack, '\v');
>> - c = getc (state_file);
>> + getc (state_file);
>> break;
>> case 'f':
>> obstack_1grow (&bstring_obstack, '\f');
>> - c = getc (state_file);
>> + getc (state_file);
>> break;
>> case 'r':
>> obstack_1grow (&bstring_obstack, '\r');
>> - c = getc (state_file);
>> + getc (state_file);
>> break;
>> case '"':
>> obstack_1grow (&bstring_obstack, '\"');
>> - c = getc (state_file);
>> + getc (state_file);
>> break;
>> case '\\':
>> obstack_1grow (&bstring_obstack, '\\');
>> - c = getc (state_file);
>> + getc (state_file);
>> break;
>> case ' ':
>> obstack_1grow (&bstring_obstack, ' ');
>> - c = getc (state_file);
>> + getc (state_file);
>> break;
>
> This is surprising. Does that mean that an escape sequence must always
> be followed by a character that is thrown away?
Yeah, that's bogus. There's only one string that has an escape sequence.
If I apply:
diff --git a/gcc/basic-block.h b/gcc/basic-block.h
index e3fff1f6975..982fa724af0 100644
--- a/gcc/basic-block.h
+++ b/gcc/basic-block.h
@@ -160,7 +160,7 @@ struct GTY((chain_next ("%h.next_bb"), chain_prev
("%h.prev_bb"))) basic_block_d
is the better choice. */
typedef int __assert_gimple_bb_smaller_rtl_bb
[(int) sizeof (struct rtl_bb_info)
- - (int) sizeof (struct gimple_bb_info)];
+- (int) sizeof (struct gimple_bb_info)];
#define BB_FREQ_MAX 10000
Then I end up with:
CSTR=(int) sizeof (struct rtl_bb_info)
(int) sizeof (struct gimple_bb_info)
Where '-' is wrongly discarded.
> state_writer::write_state_a_string surely doesn't suggest that. I
> suspect that this part of the code has never been exercised.
>
So yes, I'm going to prepare a patch which would not skip next character.
Martin
