diff mbox series

[3/6,gdb/symtab] Do more zero-initialization of type::fields

Message ID 20230830191336.15885-3-tdevries@suse.de
State Committed
Headers
Series [1/6,gdb/symtab] Fix uninitialized memory in buildsym_compunit::finish_block_internal |

Checks

Context Check Description
linaro-tcwg-bot/tcwg_gdb_check--master-aarch64 fail Patch failed to apply
linaro-tcwg-bot/tcwg_gdb_build--master-arm fail Patch failed to apply
linaro-tcwg-bot/tcwg_gdb_build--master-aarch64 fail Patch failed to apply
linaro-tcwg-bot/tcwg_gdb_check--master-arm fail Patch failed to apply

Commit Message

Tom de Vries Aug. 30, 2023, 7:13 p.m. UTC 
  Now that we've introduced type::{alloc_fields,copy_fields}, the places where
no zero-initialization of allocated fields is done are easy to spot:
...
$ find gdb* -type f | grep -v ChangeLog | xargs grep alloc_fields | grep false
gdb/coffread.c:  type->alloc_fields (nfields, false);
gdb/coffread.c:  type->alloc_fields (nsyms, false);
gdb/stabsread.c:	  ftype->alloc_fields (nsemi, false);
gdb/gdbtypes.c:  resolved_type->alloc_fields (nfields, false);
gdb/gdbtypes.c:  alloc_fields (nfields, false);
gdb/gdbtypes.c:  alloc_fields (nfields, false);
gdb/mdebugread.c:	t->alloc_fields (nfields, false);
gdb/mdebugread.c:		  ftype->alloc_fields (nparams, false);
...

All hits in gdbtypes.c are ok.  There are two hits in the two variants of
copy_fields, and there's already a comment for the third.

AFAICT, the other ones are not, so fix those by dropping the "false" argument.

Tested on x86_64-linux.
---
 gdb/coffread.c   | 4 ++--
 gdb/mdebugread.c | 4 ++--
 gdb/stabsread.c  | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

Comments

Tom Tromey Aug. 30, 2023, 8:19 p.m. UTC | #1 
>>>>> "Tom" == Tom de Vries via Gdb-patches <gdb-patches@sourceware.org> writes:

Tom> Now that we've introduced type::{alloc_fields,copy_fields}, the places where
Tom> no zero-initialization of allocated fields is done are easy to spot:
Tom> ...
Tom> $ find gdb* -type f | grep -v ChangeLog | xargs grep alloc_fields | grep false
Tom> gdb/coffread.c:  type->alloc_fields (nfields, false);
Tom> gdb/coffread.c:  type->alloc_fields (nsyms, false);
Tom> gdb/stabsread.c:	  ftype->alloc_fields (nsemi, false);
Tom> gdb/gdbtypes.c:  resolved_type->alloc_fields (nfields, false);
Tom> gdb/gdbtypes.c:  alloc_fields (nfields, false);
Tom> gdb/gdbtypes.c:  alloc_fields (nfields, false);
Tom> gdb/mdebugread.c:	t->alloc_fields (nfields, false);
Tom> gdb/mdebugread.c:		  ftype->alloc_fields (nparams, false);
Tom> ...

Tom> All hits in gdbtypes.c are ok.  There are two hits in the two variants of
Tom> copy_fields, and there's already a comment for the third.

Tom> AFAICT, the other ones are not, so fix those by dropping the "false" argument.

Looks good.
Approved-By: Tom Tromey <tom@tromey.com>

Honestly it probably wouldn't hurt to just always zero-init these.
If type allocation is ever a performance problem, then I suspect
something's gone pretty wrong.

Tom
diff mbox series

Patch

diff --git a/gdb/coffread.c b/gdb/coffread.c
index e132dd1a1f7..e251f110ff8 100644
--- a/gdb/coffread.c
+++ b/gdb/coffread.c
@@ -2032,7 +2032,7 @@  coff_read_struct_type (int index, int length, int lastsym,
     }
   /* Now create the vector of fields, and record how big it is.  */
 
-  type->alloc_fields (nfields, false);
+  type->alloc_fields (nfields);
 
   /* Copy the saved-up fields into the field vector.  */
 
@@ -2110,7 +2110,7 @@  coff_read_enum_type (int index, int length, int lastsym,
   else /* Assume ints.  */
     type->set_length (gdbarch_int_bit (gdbarch) / TARGET_CHAR_BIT);
   type->set_code (TYPE_CODE_ENUM);
-  type->alloc_fields (nsyms, false);
+  type->alloc_fields (nsyms);
 
   /* Find the symbols for the values and put them into the type.
      The symbols can be found in the symlist that we put them on
diff --git a/gdb/mdebugread.c b/gdb/mdebugread.c
index 688501e157a..ad9967b75fa 100644
--- a/gdb/mdebugread.c
+++ b/gdb/mdebugread.c
@@ -1034,7 +1034,7 @@  parse_symbol (SYMR *sh, union aux_ext *ax, char *ext_sh, int bigend,
 
 	t->set_code (type_code);
 	t->set_length (sh->value);
-	t->alloc_fields (nfields, false);
+	t->alloc_fields (nfields);
 
 	if (type_code == TYPE_CODE_ENUM)
 	  {
@@ -1195,7 +1195,7 @@  parse_symbol (SYMR *sh, union aux_ext *ax, char *ext_sh, int bigend,
 
 	      if (nparams > 0)
 		{
-		  ftype->alloc_fields (nparams, false);
+		  ftype->alloc_fields (nparams);
 
 		  iparams = 0;
 		  for (struct symbol *sym : block_iterator_range (cblock))
diff --git a/gdb/stabsread.c b/gdb/stabsread.c
index c29d1c0d397..ad9258a9f20 100644
--- a/gdb/stabsread.c
+++ b/gdb/stabsread.c
@@ -982,7 +982,7 @@  define_symbol (CORE_ADDR valu, const char *string, int desc, int type,
 	    }
 
 	  /* Allocate parameter information fields and fill them in.  */
-	  ftype->alloc_fields (nsemi, false);
+	  ftype->alloc_fields (nsemi);
 	  while (*p++ == ';')
 	    {
 	      struct type *ptype;