digitalmars.D.announce - DMD 0.157 release
- Walter Bright <newshound digitalmars.com> May 10 2006
- "Ameer Armaly" <ameer_armaly hotmail.com> May 10 2006
- pragma <pragma_member pathlink.com> May 10 2006
- Frank Benoit <keinfarbton nospam.xyz> May 10 2006
- Walter Bright <newshound digitalmars.com> May 10 2006
- Thomas Kuehne <thomas-dloop kuehne.cn> May 10 2006
- Walter Bright <newshound digitalmars.com> May 11 2006
- Thomas Kuehne <thomas-dloop kuehne.cn> May 11 2006
- Dejan Lekic <dejan nu6.org> May 11 2006
- Stewart Gordon <smjg_1998 yahoo.com> May 11 2006
- John Demme <me teqdruid.com> May 11 2006
- Walter Bright <newshound digitalmars.com> May 11 2006
Adds partial symbolic debug info support for Linux. (Doesn't look like much, but it's a lot of code to support this!) Doesn't yet do the internals of structs or classes. http://www.digitalmars.com/d/changelog.html
May 10 2006
"Ameer Armaly" <ameer_armaly hotmail.com> "Walter Bright" <newshound digitalmars.com> wrote in message news:e3tefc$1n5d$1 digitaldaemon.com...Adds partial symbolic debug info support for Linux. (Doesn't look like much, but it's a lot of code to support this!) Doesn't yet do the internals of structs or classes.
completed.http://www.digitalmars.com/d/changelog.html
May 10 2006
In article <e3tefc$1n5d$1 digitaldaemon.com>, Walter Bright says...Adds partial symbolic debug info support for Linux. (Doesn't look like much, but it's a lot of code to support this!) Doesn't yet do the internals of structs or classes. http://www.digitalmars.com/d/changelog.html
Nice! This will make my development trips into LinuxLand a lot easier. Thanks! - EricAnderton at yahoo
May 10 2006
Walter Bright schrieb:Adds partial symbolic debug info support for Linux. (Doesn't look like much, but it's a lot of code to support this!) Doesn't yet do the internals of structs or classes. http://www.digitalmars.com/d/changelog.html
Great. Does this allready work with gdb? Or is additional work necessary for the gdb-patches?
May 10 2006
Frank Benoit wrote:Great. Does this allready work with gdb? Or is additional work necessary for the gdb-patches?
Yes, it works already with gdb, though it works better with the D patched one.
May 10 2006
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Walter Bright schrieb am 2006-05-10:Adds partial symbolic debug info support for Linux.
While you are at it: How about adding line number debug info for inline assembler code? Thomas -----BEGIN PGP SIGNATURE----- iD8DBQFEYuca3w+/yD4P9tIRAuGfAJ9gP6tb5s7gGDvjgboBvjITM3cx6QCdG+K4 9GwB+8k5h3Ffnhy7+fmOHjw= =dXEA -----END PGP SIGNATURE-----
May 10 2006
Thomas Kuehne wrote:Walter Bright schrieb am 2006-05-10:Adds partial symbolic debug info support for Linux.
While you are at it: How about adding line number debug info for inline assembler code?
I'll add it to the queue!
May 11 2006
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Walter Bright schrieb am 2006-05-11:Thomas Kuehne wrote:Walter Bright schrieb am 2006-05-10:Adds partial symbolic debug info support for Linux.
While you are at it: How about adding line number debug info for inline assembler code?
I'll add it to the queue!
Thanks -----BEGIN PGP SIGNATURE----- iD8DBQFEY5E43w+/yD4P9tIRArpHAKCdwtuGrw1s4DQZFCzhehSjCnPB+ACfafyS bWwPd2TKmVnwaW7HtPvwLrc= =a+jh -----END PGP SIGNATURE-----
May 11 2006
What happened to 0.156? Stewart. -- -----BEGIN GEEK CODE BLOCK----- Version: 3.1 GCS/M d- s:- C++ a->--- UB P+ L E W++ N+++ o K- w++ O? M V? PS- PE- Y? PGP- t- 5? X? R b DI? D G e++++ h-- r-- !y ------END GEEK CODE BLOCK------ My e-mail is valid but not my primary mailbox. Please keep replies on the 'group where everyone may benefit.
May 11 2006
Stewart Gordon wrote:What happened to 0.156?
http://www.digitalmars.com/d/changelog.html#new0156 It fixed some printf's, that W had missed to comment. --anders
May 11 2006
Walter Bright wrote:Adds partial symbolic debug info support for Linux. (Doesn't look like much, but it's a lot of code to support this!) Doesn't yet do the internals of structs or classes. http://www.digitalmars.com/d/changelog.html
Thank you thank you thank you thank you. Supporting struct and class internals would be great, but I guess we'll take this some step at a time... How about strings? GDB seems to detect them as "unsigned long long"... Once you get struct internals working, you should probably make dynamic arrays structs, but until then, why not make strings appear as char*? They should then work with GDB, since DMD puts a null character in them. If there's anything that GDB doesn't quite support, but DWARF seems to, then let me know... No promises, but I'll poke around GDB and (if sucessful) add stuff to the patch. It's very possible that there's stuff for dealing with classes and such in DWARF and GDB, but won't currently work with GDB since GDB treats D code like C code. I can probably fix at least some of that. ~John Demme
May 11 2006
John Demme wrote:Thank you thank you thank you thank you.
You're welcome!Supporting struct and class internals would be great, but I guess we'll take this some step at a time...
Yes.How about strings? GDB seems to detect them as "unsigned long long"... Once you get struct internals working, you should probably make dynamic arrays structs, but until then, why not make strings appear as char*? They should then work with GDB, since DMD puts a null character in them.
The trouble is that strings are arrays. Arrays are a 64 bit type. If dmd puts it out as a struct, then gdb wrongly assumes that it's returned like a struct, instead of in the EDX:EAX register pair. (windbg has the same problem)If there's anything that GDB doesn't quite support, but DWARF seems to, then let me know... No promises, but I'll poke around GDB and (if sucessful) add stuff to the patch. It's very possible that there's stuff for dealing with classes and such in DWARF and GDB, but won't currently work with GDB since GDB treats D code like C code. I can probably fix at least some of that.
The number one thing would be to support the D array type. A DW_TAG_dynamic_array_type would work. The number two thing would be to support register variables in register pairs. Although you can specify that to codeview, there seems no way to do it in dwarf.
May 11 2006