From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 23481 invoked by alias); 5 Mar 2004 03:31:23 -0000 Mailing-List: contact gdb-patches-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sources.redhat.com Received: (qmail 23469 invoked from network); 5 Mar 2004 03:31:20 -0000 Received: from unknown (HELO nevyn.them.org) (66.93.172.17) by sources.redhat.com with SMTP; 5 Mar 2004 03:31:20 -0000 Received: from drow by nevyn.them.org with local (Exim 4.30 #1 (Debian)) id 1Az63U-0003Ip-5b; Thu, 04 Mar 2004 22:31:20 -0500 Date: Fri, 05 Mar 2004 03:31:00 -0000 From: Daniel Jacobowitz To: Elena Zannoni Cc: gdb-patches@sources.redhat.com Subject: Re: RFA: Don't use obsavestring in dwarf2read Message-ID: <20040305033120.GD5320@nevyn.them.org> Mail-Followup-To: Elena Zannoni , gdb-patches@sources.redhat.com References: <20040112015726.GA7151@nevyn.them.org> <20040202182218.GA3405@nevyn.them.org> <16418.39156.566837.685666@localhost.redhat.com> <20040205194821.GA30363@nevyn.them.org> <16418.43222.486879.784465@localhost.redhat.com> <20040205204756.GA2465@nevyn.them.org> <16418.52967.149289.557083@localhost.redhat.com> <20040208044115.GD13033@nevyn.them.org> <16432.56112.64946.492536@localhost.redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <16432.56112.64946.492536@localhost.redhat.com> User-Agent: Mutt/1.5.1i X-SW-Source: 2004-03.o/txt/msg00082.txt Message-ID: <20040305033100.-JATj2mcValeNt1pfezo7u3oiirTOAZtBk7MSH_D4lE@z> Sorry 'bout the delay getting back to this... I got distracted by working on intercu and thought I'd already done it. On Mon, Feb 16, 2004 at 10:01:04AM -0500, Elena Zannoni wrote: > There is a similar comment right before the dwarf2_pinfo structure, > can you somehow either unify the two or 'link' them together? Both peices > of information should be available together. Actually having this > comment just before an ifdeffed out section of code, makes me wonder > if it won't be overlooked. I've moved it up above the first non-#include (which is a #define) so that should be a little clearer. I also moved the relevant bits of the dwarf2_pinfo comment up to join it. > Can you please add a clear comment above the assignment about filename being on the obstack? Sure. This is what I've committed to HEAD only - just comment changes from the version you OK'd. Let me know if the new comments are still lacking. Still no testsuite result changes on i386-linux, as expected. -- Daniel Jacobowitz MontaVista Software Debian GNU/Linux Developer 2004-03-04 Daniel Jacobowitz * dwarf2read.c: Add comment describing memory lifetimes. (struct dwarf2_pinfo): Update comment. (dwarf2_add_field, dwarf2_add_member_fn, read_structure_scope) (read_enumeration, new_symbol): Don't use obsavestring. Index: dwarf2read.c =================================================================== RCS file: /cvs/src/src/gdb/dwarf2read.c,v retrieving revision 1.135 diff -u -p -r1.135 dwarf2read.c --- dwarf2read.c 21 Feb 2004 02:13:35 -0000 1.135 +++ dwarf2read.c 5 Mar 2004 03:17:40 -0000 @@ -50,6 +50,20 @@ #include "gdb_assert.h" #include +/* A note on memory usage for this file. + + At the present time, this code reads the debug info sections into + the objfile's objfile_obstack. A definite improvement for startup + time, on platforms which do not emit relocations for debug + sections, would be to use mmap instead. The object's complete + debug information is loaded into memory, partly to simplify + absolute DIE references. + + Whether using obstacks or mmap, the sections should remain loaded + until the objfile is released, and pointers into the section data + can be used for any other data associated to the objfile (symbol + names, type names, location expressions to name a few). */ + #ifndef DWARF2_REG_TO_REGNUM #define DWARF2_REG_TO_REGNUM(REG) (REG) #endif @@ -444,12 +458,11 @@ static int isreg; /* Object lives in re /* We put a pointer to this structure in the read_symtab_private field of the psymtab. - The complete dwarf information for an objfile is kept in the - objfile_obstack, so that absolute die references can be handled. + Most of the information in this structure is related to an entire - object file and could be passed via the sym_private field of the objfile. - It is however conceivable that dwarf2 might not be the only type - of symbols read from an object file. */ + object file and could be passed via the sym_private field of the + objfile. It is possible to have both dwarf2 and some other form + of debug symbols in one object file. */ struct dwarf2_pinfo { @@ -2664,8 +2677,10 @@ dwarf2_add_field (struct field_info *fip attr = dwarf2_attr (die, DW_AT_name, cu); if (attr && DW_STRING (attr)) fieldname = DW_STRING (attr); - fp->name = obsavestring (fieldname, strlen (fieldname), - &objfile->objfile_obstack); + + /* The name is already allocated along with this objfile, so we don't + need to duplicate it for the type. */ + fp->name = fieldname; /* Change accessibility for artificial fields (e.g. virtual table pointer or virtual base class pointer) to private. */ @@ -2696,11 +2711,11 @@ dwarf2_add_field (struct field_info *fip /* Get physical name. */ physname = dwarf2_linkage_name (die, cu); - SET_FIELD_PHYSNAME (*fp, obsavestring (physname, strlen (physname), - &objfile->objfile_obstack)); + /* The name is already allocated along with this objfile, so we don't + need to duplicate it for the type. */ + SET_FIELD_PHYSNAME (*fp, physname ? physname : ""); FIELD_TYPE (*fp) = die_type (die, cu); - FIELD_NAME (*fp) = obsavestring (fieldname, strlen (fieldname), - &objfile->objfile_obstack); + FIELD_NAME (*fp) = fieldname; } else if (die->tag == DW_TAG_inheritance) { @@ -2868,8 +2883,9 @@ dwarf2_add_member_fn (struct field_info /* Fill in the member function field info. */ fnp = &new_fnfield->fnfield; - fnp->physname = obsavestring (physname, strlen (physname), - &objfile->objfile_obstack); + /* The name is already allocated along with this objfile, so we don't + need to duplicate it for the type. */ + fnp->physname = physname ? physname : ""; fnp->type = alloc_type (objfile); if (die->type && TYPE_CODE (die->type) == TYPE_CODE_FUNC) { @@ -3000,7 +3016,7 @@ read_structure_scope (struct die_info *d struct objfile *objfile = cu->objfile; struct type *type; struct attribute *attr; - const char *name = NULL; + char *name = NULL; const char *previous_prefix = processing_current_prefix; struct cleanup *back_to = NULL; /* This says whether or not we want to try to update the structure's @@ -3045,8 +3061,9 @@ read_structure_scope (struct die_info *d } else { - TYPE_TAG_NAME (type) = obsavestring (name, strlen (name), - &objfile->objfile_obstack); + /* The name is already allocated along with this objfile, so + we don't need to duplicate it for the type. */ + TYPE_TAG_NAME (type) = name; need_to_update_name = (cu->language == language_cplus); } } @@ -3251,7 +3268,7 @@ read_enumeration (struct die_info *die, attr = dwarf2_attr (die, DW_AT_name, cu); if (attr && DW_STRING (attr)) { - const char *name = DW_STRING (attr); + char *name = DW_STRING (attr); if (processing_has_namespace_info) { @@ -3263,8 +3280,9 @@ read_enumeration (struct die_info *die, } else { - TYPE_TAG_NAME (type) = obsavestring (name, strlen (name), - &objfile->objfile_obstack); + /* The name is already allocated along with this objfile, so + we don't need to duplicate it for the type. */ + TYPE_TAG_NAME (type) = name; } } @@ -5677,11 +5695,11 @@ new_symbol (struct die_info *die, struct { /* FIXME: carlton/2003-11-10: Should this use SYMBOL_SET_NAMES instead? (The same problem also - arises a further down in the function.) */ - SYMBOL_LINKAGE_NAME (sym) - = obsavestring (TYPE_TAG_NAME (type), - strlen (TYPE_TAG_NAME (type)), - &objfile->objfile_obstack); + arises further down in this function.) */ + /* The type's name is already allocated along with + this objfile, so we don't need to duplicate it + for the symbol. */ + SYMBOL_LINKAGE_NAME (sym) = TYPE_TAG_NAME (type); } } @@ -5712,11 +5730,11 @@ new_symbol (struct die_info *die, struct sizeof (struct symbol)); *typedef_sym = *sym; SYMBOL_DOMAIN (typedef_sym) = VAR_DOMAIN; + /* The symbol's name is already allocated along with + this objfile, so we don't need to duplicate it for + the type. */ if (TYPE_NAME (SYMBOL_TYPE (sym)) == 0) - TYPE_NAME (SYMBOL_TYPE (sym)) = - obsavestring (SYMBOL_NATURAL_NAME (sym), - strlen (SYMBOL_NATURAL_NAME (sym)), - &objfile->objfile_obstack); + TYPE_NAME (SYMBOL_TYPE (sym)) = SYMBOL_NATURAL_NAME (sym); add_symbol_to_list (typedef_sym, list_to_add); } }