www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - D Now in DWARF3 Standard!

reply John Demme <me teqdruid.com> writes:
According to
http://dwarf.freestandards.org/ShowIssue.php?issue=050408.1&type=closed

D has been added to the latest DWARF standard draft.  This means that D
has it's own language code.  According to
http://dwarf.freestandards.org/Dwarf3.doc (page 148)
D's been assigned 0x0013.

For those of you who don't know, DWARF is the standard for debugging
information for the ELF executable format... basically the debugging
standard on Linux.  Currently, Walter cheats and outputs the code for C.
This confuses the hell out of GDB.

Once Walter changes dmd to output D's new code (as he told me he would
once I got us assigned a code) I'll update my GDB patch to add this
code.  Currently the D symbol demangling in my GDB patch only works if
you don't compile the D source with -debug.  Hopefully while Walter is
mucking around with the DWARF debugging output, he'll fix the line
numbering to be somewhat close to working ;).  (Please, Walter?  Pretty
please?)

With this new code, and if Walter fixes DMD, I'll make an attempt at
getting my GDB patch into the main GDB source tree :)  Then D would
really get some credibility!  Things may be going slowly, but at least
they're moving.

-John Demme
Jul 24 2005
next sibling parent Factory <t t.com> writes:
In article <1122266146.3983.12.camel localhost>, 
me teqdruid.com says...
 With this new code, and if Walter fixes DMD, I'll make an attempt at
 getting my GDB patch into the main GDB source tree :)  Then D would
 really get some credibility!  Things may be going slowly, but at least
 they're moving.

Hmm I wouldn't mind a good debugger for D. - Factory
Jul 25 2005
prev sibling next sibling parent "Ben Hinkle" <ben.hinkle gmail.com> writes:
"John Demme" <me teqdruid.com> wrote in message 
news:1122266146.3983.12.camel localhost...
 According to
 http://dwarf.freestandards.org/ShowIssue.php?issue=050408.1&type=closed

 D has been added to the latest DWARF standard draft.  This means that D
 has it's own language code.  According to
 http://dwarf.freestandards.org/Dwarf3.doc (page 148)
 D's been assigned 0x0013.

 For those of you who don't know, DWARF is the standard for debugging
 information for the ELF executable format... basically the debugging
 standard on Linux.  Currently, Walter cheats and outputs the code for C.
 This confuses the hell out of GDB.

 Once Walter changes dmd to output D's new code (as he told me he would
 once I got us assigned a code) I'll update my GDB patch to add this
 code.  Currently the D symbol demangling in my GDB patch only works if
 you don't compile the D source with -debug.  Hopefully while Walter is
 mucking around with the DWARF debugging output, he'll fix the line
 numbering to be somewhat close to working ;).  (Please, Walter?  Pretty
 please?)

 With this new code, and if Walter fixes DMD, I'll make an attempt at
 getting my GDB patch into the main GDB source tree :)  Then D would
 really get some credibility!  Things may be going slowly, but at least
 they're moving.

Very cool! I think this would help a great deal.
Jul 25 2005
prev sibling next sibling parent "Charles" <noone nowhere.com> writes:
Thanks for doing this John , like you said D might not be bursting on to the
scene ( yet ) , but its slowly and consistently getting more and more
exposure :).

Charlie

"John Demme" <me teqdruid.com> wrote in message
news:1122266146.3983.12.camel localhost...
 According to
 http://dwarf.freestandards.org/ShowIssue.php?issue=050408.1&type=closed

 D has been added to the latest DWARF standard draft.  This means that D
 has it's own language code.  According to
 http://dwarf.freestandards.org/Dwarf3.doc (page 148)
 D's been assigned 0x0013.

 For those of you who don't know, DWARF is the standard for debugging
 information for the ELF executable format... basically the debugging
 standard on Linux.  Currently, Walter cheats and outputs the code for C.
 This confuses the hell out of GDB.

 Once Walter changes dmd to output D's new code (as he told me he would
 once I got us assigned a code) I'll update my GDB patch to add this
 code.  Currently the D symbol demangling in my GDB patch only works if
 you don't compile the D source with -debug.  Hopefully while Walter is
 mucking around with the DWARF debugging output, he'll fix the line
 numbering to be somewhat close to working ;).  (Please, Walter?  Pretty
 please?)

 With this new code, and if Walter fixes DMD, I'll make an attempt at
 getting my GDB patch into the main GDB source tree :)  Then D would
 really get some credibility!  Things may be going slowly, but at least
 they're moving.

 -John Demme

Jul 25 2005
prev sibling next sibling parent reply "Walter" <newshound digitalmars.com> writes:
"John Demme" <me teqdruid.com> wrote in message
news:1122266146.3983.12.camel localhost...
 According to
 http://dwarf.freestandards.org/ShowIssue.php?issue=050408.1&type=closed

 D has been added to the latest DWARF standard draft.  This means that D
 has it's own language code.  According to
 http://dwarf.freestandards.org/Dwarf3.doc (page 148)
 D's been assigned 0x0013.

Thanks, I'll make this change. Just to be sure, though, the first url shows it to be 0x001d, but you say it's 0x001d. I can't look at dwarf3.doc, my machine crashes trying to load it. Perhaps freestandards should use an open standard like html instead? <g>
 For those of you who don't know, DWARF is the standard for debugging
 information for the ELF executable format... basically the debugging
 standard on Linux.  Currently, Walter cheats and outputs the code for C.
 This confuses the hell out of GDB.

 Once Walter changes dmd to output D's new code (as he told me he would
 once I got us assigned a code) I'll update my GDB patch to add this
 code.  Currently the D symbol demangling in my GDB patch only works if
 you don't compile the D source with -debug.  Hopefully while Walter is
 mucking around with the DWARF debugging output, he'll fix the line
 numbering to be somewhat close to working ;).  (Please, Walter?  Pretty
 please?)

 With this new code, and if Walter fixes DMD, I'll make an attempt at
 getting my GDB patch into the main GDB source tree :)  Then D would
 really get some credibility!  Things may be going slowly, but at least
 they're moving.

 -John Demme

Jul 30 2005
next sibling parent reply pragma <pragma_member pathlink.com> writes:
In article <dch2bu$1nto$1 digitaldaemon.com>, Walter says...
"John Demme" <me teqdruid.com> wrote in message
news:1122266146.3983.12.camel localhost...
 According to
 http://dwarf.freestandards.org/ShowIssue.php?issue=050408.1&type=closed

 D has been added to the latest DWARF standard draft.  This means that D
 has it's own language code.  According to
 http://dwarf.freestandards.org/Dwarf3.doc (page 148)
 D's been assigned 0x0013.

Thanks, I'll make this change. Just to be sure, though, the first url shows it to be 0x001d, but you say it's 0x001d. I can't look at dwarf3.doc, my machine crashes trying to load it.

Walter, I could load the document just fine under Word/Office 2000. Here's a cut-and-paste of the language constants in the spec: DW_LANG_Pascal83 0x0009 DW_LANG_Modula2 0x000a DW_LANG_Java 0x000b DW_LANG_C99 0x000c DW_LANG_Ada95 0x000d DW_LANG_Fortran95DW_LANG_Fortran95DW_LANG_Fortran95 0x000e DW_LANG_PLI 0x000f DW_LANG_ObjC 0x0010 DW_LANG_ObjC_plus_plus 0x0011 DW_LANG_Upc 0x0012 DW_LANG_D 0x0013 DW_LANG_lo_user 0x8000 DW_LANG_hi_user 0xffff Looks like we're lucky number 13.
Perhaps freestandards should use an open
standard like html instead? <g>

Sad but true. Especially since the DWARF spec has such a strong affinity with Unix/Linux. - EricAnderton at yahoo
Jul 30 2005
parent Piotr Fusik <Piotr_member pathlink.com> writes:
 Looks like we're lucky number 13.

;-)
Aug 01 2005
prev sibling next sibling parent reply "Walter" <newshound digitalmars.com> writes:
Ok. I'll add it in, with a -gc switch for the old way, as it will take years
before all the gdb changes percolate through.

"John Demme" <me teqdruid.com> wrote in message
news:1122770369.12363.3.camel localhost...
 Right, so the first link is to my proposal.  I requested 0x001d, since
 0x000d was taken.  They decided to go in order, and gave us the next
 available one, 0x0013.

 I think they're using Word since they want to be able to track changes.
 This is only a draft, and once it's properly edited for grammer,
 spelling, ect, they'll release a copy in PDF- still not great, but
 better.

 John

 On Sat, 2005-07-30 at 16:26 -0700, Walter wrote:
 "John Demme" <me teqdruid.com> wrote in message
 news:1122266146.3983.12.camel localhost...
 According to



 D has been added to the latest DWARF standard draft.  This means that



 has it's own language code.  According to
 http://dwarf.freestandards.org/Dwarf3.doc (page 148)
 D's been assigned 0x0013.

Thanks, I'll make this change. Just to be sure, though, the first url


 it to be 0x001d, but you say it's 0x001d. I can't look at dwarf3.doc, my
 machine crashes trying to load it. Perhaps freestandards should use an


 standard like html instead? <g>

 For those of you who don't know, DWARF is the standard for debugging
 information for the ELF executable format... basically the debugging
 standard on Linux.  Currently, Walter cheats and outputs the code for



 This confuses the hell out of GDB.

 Once Walter changes dmd to output D's new code (as he told me he would
 once I got us assigned a code) I'll update my GDB patch to add this
 code.  Currently the D symbol demangling in my GDB patch only works if
 you don't compile the D source with -debug.  Hopefully while Walter is
 mucking around with the DWARF debugging output, he'll fix the line
 numbering to be somewhat close to working ;).  (Please, Walter?



 please?)

 With this new code, and if Walter fixes DMD, I'll make an attempt at
 getting my GDB patch into the main GDB source tree :)  Then D would
 really get some credibility!  Things may be going slowly, but at least
 they're moving.

 -John Demme



Jul 30 2005
parent reply "Walter" <newshound digitalmars.com> writes:
"John Demme" <me teqdruid.com> wrote in message
news:1122777734.12363.5.camel localhost...
 Not to keep harking on it to be an annoyance, but is there any chance
 you'll fix the DWARF line numbering bug while you're in there?  (It'd be
 real nice)

I'm not sure what this is, is there a message here you can point to which details it?
Jul 31 2005
parent "Walter" <newshound digitalmars.com> writes:
I'll go look at it again. It was working the last time I tried it.
Aug 01 2005
prev sibling parent reply J C Calvarese <technocrat7 gmail.com> writes:
In article <dch2bu$1nto$1 digitaldaemon.com>, Walter says...
"John Demme" <me teqdruid.com> wrote in message
news:1122266146.3983.12.camel localhost...
 According to
 http://dwarf.freestandards.org/ShowIssue.php?issue=050408.1&type=closed

 D has been added to the latest DWARF standard draft.  This means that D
 has it's own language code.  According to
 http://dwarf.freestandards.org/Dwarf3.doc (page 148)
 D's been assigned 0x0013.

Thanks, I'll make this change. Just to be sure, though, the first url shows it to be 0x001d, but you say it's 0x001d. I can't look at dwarf3.doc, my machine crashes trying to load it. Perhaps freestandards should use an open standard like html instead? <g>

FWIW, I was able to view the .doc with OpenOffice 1.1.4 (open source and free, http://www.openoffice.org/). But it wasn't on page 148 for me, it was on page 132. I found it by searching for "DW_LANG_D". jcc7
Jul 30 2005
parent "Walter" <newshound digitalmars.com> writes:
"John Demme" <me teqdruid.com> wrote in message
news:1122785162.12363.6.camel localhost...
 I viewed it in Abiword...  Damn I hate the doc format.  Give me anything
 but doc.

I just can't see using doc format: 1) it's proprietary 2) it changes substantially with updates to Word 3) I can't read the contents without installing Word (the various Word viewers, *including* the ones written by Microsoft, crash or fail on at least a third of the doc files I try them on) 4) doc files can contain hidden viruses 5) doc files can contain arbitrary bits of uninitialized memory in them from other programs, including things like passwords and personal data 6) doc files insert GUIDs specific to my computer 7) If you aren't careful, you'll wind up shipping a doc file with all your embarassing or compromising old versions of the text hidden and embedded within it. 8) Basically, I don't know and can't control what all is embedded in those files. 9) How am I going to read those files 20, 30, 40 years from now? I still have old Wordstar files I keep for legal reasons - fortunately, I have a Wordstar => text file converter. Frankly, any database file (such as documents, spreadsheets, accounts, etc.) should use a text based file format.
Jul 31 2005
prev sibling next sibling parent John Demme <me teqdruid.com> writes:
Right, so the first link is to my proposal.  I requested 0x001d, since
0x000d was taken.  They decided to go in order, and gave us the next
available one, 0x0013.

I think they're using Word since they want to be able to track changes.
This is only a draft, and once it's properly edited for grammer,
spelling, ect, they'll release a copy in PDF- still not great, but
better.

John

On Sat, 2005-07-30 at 16:26 -0700, Walter wrote:
 "John Demme" <me teqdruid.com> wrote in message
 news:1122266146.3983.12.camel localhost...
 According to
 http://dwarf.freestandards.org/ShowIssue.php?issue=050408.1&type=closed

 D has been added to the latest DWARF standard draft.  This means that D
 has it's own language code.  According to
 http://dwarf.freestandards.org/Dwarf3.doc (page 148)
 D's been assigned 0x0013.

Thanks, I'll make this change. Just to be sure, though, the first url shows it to be 0x001d, but you say it's 0x001d. I can't look at dwarf3.doc, my machine crashes trying to load it. Perhaps freestandards should use an open standard like html instead? <g>
 For those of you who don't know, DWARF is the standard for debugging
 information for the ELF executable format... basically the debugging
 standard on Linux.  Currently, Walter cheats and outputs the code for C.
 This confuses the hell out of GDB.

 Once Walter changes dmd to output D's new code (as he told me he would
 once I got us assigned a code) I'll update my GDB patch to add this
 code.  Currently the D symbol demangling in my GDB patch only works if
 you don't compile the D source with -debug.  Hopefully while Walter is
 mucking around with the DWARF debugging output, he'll fix the line
 numbering to be somewhat close to working ;).  (Please, Walter?  Pretty
 please?)

 With this new code, and if Walter fixes DMD, I'll make an attempt at
 getting my GDB patch into the main GDB source tree :)  Then D would
 really get some credibility!  Things may be going slowly, but at least
 they're moving.

 -John Demme


Jul 30 2005
prev sibling next sibling parent John Demme <me teqdruid.com> writes:
Not to keep harking on it to be an annoyance, but is there any chance
you'll fix the DWARF line numbering bug while you're in there?  (It'd be
real nice)

On Sat, 2005-07-30 at 18:31 -0700, Walter wrote:
 Ok. I'll add it in, with a -gc switch for the old way, as it will take years
 before all the gdb changes percolate through.
 
 "John Demme" <me teqdruid.com> wrote in message
 news:1122770369.12363.3.camel localhost...
 Right, so the first link is to my proposal.  I requested 0x001d, since
 0x000d was taken.  They decided to go in order, and gave us the next
 available one, 0x0013.

 I think they're using Word since they want to be able to track changes.
 This is only a draft, and once it's properly edited for grammer,
 spelling, ect, they'll release a copy in PDF- still not great, but
 better.

 John

 On Sat, 2005-07-30 at 16:26 -0700, Walter wrote:
 "John Demme" <me teqdruid.com> wrote in message
 news:1122266146.3983.12.camel localhost...
 According to



 D has been added to the latest DWARF standard draft.  This means that



 has it's own language code.  According to
 http://dwarf.freestandards.org/Dwarf3.doc (page 148)
 D's been assigned 0x0013.

Thanks, I'll make this change. Just to be sure, though, the first url


 it to be 0x001d, but you say it's 0x001d. I can't look at dwarf3.doc, my
 machine crashes trying to load it. Perhaps freestandards should use an


 standard like html instead? <g>

 For those of you who don't know, DWARF is the standard for debugging
 information for the ELF executable format... basically the debugging
 standard on Linux.  Currently, Walter cheats and outputs the code for



 This confuses the hell out of GDB.

 Once Walter changes dmd to output D's new code (as he told me he would
 once I got us assigned a code) I'll update my GDB patch to add this
 code.  Currently the D symbol demangling in my GDB patch only works if
 you don't compile the D source with -debug.  Hopefully while Walter is
 mucking around with the DWARF debugging output, he'll fix the line
 numbering to be somewhat close to working ;).  (Please, Walter?



 please?)

 With this new code, and if Walter fixes DMD, I'll make an attempt at
 getting my GDB patch into the main GDB source tree :)  Then D would
 really get some credibility!  Things may be going slowly, but at least
 they're moving.

 -John Demme




Jul 30 2005
prev sibling next sibling parent John Demme <me teqdruid.com> writes:
I viewed it in Abiword...  Damn I hate the doc format.  Give me anything
but doc.

On Sun, 2005-07-31 at 04:14 +0000, J C Calvarese wrote:
 In article <dch2bu$1nto$1 digitaldaemon.com>, Walter says...
"John Demme" <me teqdruid.com> wrote in message
news:1122266146.3983.12.camel localhost...
 According to
 http://dwarf.freestandards.org/ShowIssue.php?issue=050408.1&type=closed

 D has been added to the latest DWARF standard draft.  This means that D
 has it's own language code.  According to
 http://dwarf.freestandards.org/Dwarf3.doc (page 148)
 D's been assigned 0x0013.

Thanks, I'll make this change. Just to be sure, though, the first url shows it to be 0x001d, but you say it's 0x001d. I can't look at dwarf3.doc, my machine crashes trying to load it. Perhaps freestandards should use an open standard like html instead? <g>

FWIW, I was able to view the .doc with OpenOffice 1.1.4 (open source and free, http://www.openoffice.org/). But it wasn't on page 148 for me, it was on page 132. I found it by searching for "DW_LANG_D". jcc7

Jul 30 2005
prev sibling parent John Demme <me teqdruid.com> writes:
Uhhh... Well, the bug is that is doesn't work- at all!  I've been
yelling about it for a long, long time now.

digitalmars.D.bugs/3064

Is one of my posts about it.  I just tested again with DMD 0.127 and it
still doesn't work.  I basically haven't been able to -ever- step
through even a simple D program in GDB.  The above is not an isolated
case, just an example.  When I try to step through a program that's
linked with something like, say, Mango, it's even worse!

I thought you had told me in an email on 4/8/05 (if you save your
emails) that you were aware of the issue.

-John Demme

On Sun, 2005-07-31 at 11:12 -0700, Walter wrote:
 "John Demme" <me teqdruid.com> wrote in message
 news:1122777734.12363.5.camel localhost...
 Not to keep harking on it to be an annoyance, but is there any chance
 you'll fix the DWARF line numbering bug while you're in there?  (It'd be
 real nice)

I'm not sure what this is, is there a message here you can point to which details it?

Jul 31 2005