public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug c/36735] New: incomplete tree dump with "-fdump-tree-original-raw"
@ 2008-07-05 0:39 grabner at icg dot tugraz dot at
2008-07-05 0:42 ` [Bug c/36735] " grabner at icg dot tugraz dot at
` (4 more replies)
0 siblings, 5 replies; 6+ messages in thread
From: grabner at icg dot tugraz dot at @ 2008-07-05 0:39 UTC (permalink / raw)
To: gcc-bugs
The tree dump created with the option "-fdump-tree-original-raw" is incomplete
for two reasons:
1.) at some places in the code, the "slim" option (TDF_SLIM) has been hardcoded
(this is easily demonstrated by running "gcc -fdump-tree-original-raw test.cpp"
and "gcc -fdump-tree-original-raw-slim test.cpp", which both produce the same
output)
2.) declaration expressions are not handled properly
This might be related to other bugs in the database which are not yet resolved
(e.g., #27193). The attached patch fixes both issues.
--
Summary: incomplete tree dump with "-fdump-tree-original-raw"
Product: gcc
Version: 4.3.1
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: grabner at icg dot tugraz dot at
GCC build triplet: (see above)
GCC host triplet: What's that? If this value is important to you, please
document
GCC target triplet: (see above)
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36735
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug c/36735] incomplete tree dump with "-fdump-tree-original-raw"
2008-07-05 0:39 [Bug c/36735] New: incomplete tree dump with "-fdump-tree-original-raw" grabner at icg dot tugraz dot at
@ 2008-07-05 0:42 ` grabner at icg dot tugraz dot at
2008-07-05 0:43 ` grabner at icg dot tugraz dot at
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: grabner at icg dot tugraz dot at @ 2008-07-05 0:42 UTC (permalink / raw)
To: gcc-bugs
------- Comment #1 from grabner at icg dot tugraz dot at 2008-07-05 00:41 -------
Created an attachment (id=15857)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15857&action=view)
short example
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36735
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug c/36735] incomplete tree dump with "-fdump-tree-original-raw"
2008-07-05 0:39 [Bug c/36735] New: incomplete tree dump with "-fdump-tree-original-raw" grabner at icg dot tugraz dot at
2008-07-05 0:42 ` [Bug c/36735] " grabner at icg dot tugraz dot at
@ 2008-07-05 0:43 ` grabner at icg dot tugraz dot at
2008-07-05 9:41 ` joseph at codesourcery dot com
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: grabner at icg dot tugraz dot at @ 2008-07-05 0:43 UTC (permalink / raw)
To: gcc-bugs
------- Comment #2 from grabner at icg dot tugraz dot at 2008-07-05 00:42 -------
Created an attachment (id=15858)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15858&action=view)
patch to fix the tree dump problem
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36735
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug c/36735] incomplete tree dump with "-fdump-tree-original-raw"
2008-07-05 0:39 [Bug c/36735] New: incomplete tree dump with "-fdump-tree-original-raw" grabner at icg dot tugraz dot at
2008-07-05 0:42 ` [Bug c/36735] " grabner at icg dot tugraz dot at
2008-07-05 0:43 ` grabner at icg dot tugraz dot at
@ 2008-07-05 9:41 ` joseph at codesourcery dot com
2008-07-05 19:23 ` [Bug middle-end/36735] " pinskia at gcc dot gnu dot org
2008-07-05 20:41 ` grabner at icg dot tugraz dot at
4 siblings, 0 replies; 6+ messages in thread
From: joseph at codesourcery dot com @ 2008-07-05 9:41 UTC (permalink / raw)
To: gcc-bugs
------- Comment #3 from joseph at codesourcery dot com 2008-07-05 09:40 -------
Subject: Re: incomplete tree dump with "-fdump-tree-original-raw"
On Sat, 5 Jul 2008, grabner at icg dot tugraz dot at wrote:
> Created an attachment (id=15858)
--> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15858&action=view)
> --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=15858&action=view)
> patch to fix the tree dump problem
Patches (against trunk) must go to gcc-patches after testing and with
ChangeLog entries and the other information required; see
<http://gcc.gnu.org/contribute.html> for details.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36735
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug middle-end/36735] incomplete tree dump with "-fdump-tree-original-raw"
2008-07-05 0:39 [Bug c/36735] New: incomplete tree dump with "-fdump-tree-original-raw" grabner at icg dot tugraz dot at
` (2 preceding siblings ...)
2008-07-05 9:41 ` joseph at codesourcery dot com
@ 2008-07-05 19:23 ` pinskia at gcc dot gnu dot org
2008-07-05 20:41 ` grabner at icg dot tugraz dot at
4 siblings, 0 replies; 6+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2008-07-05 19:23 UTC (permalink / raw)
To: gcc-bugs
------- Comment #4 from pinskia at gcc dot gnu dot org 2008-07-05 19:22 -------
Really these options are only to be used to debug GCC and not really supposed
to be used to outside of that.
> What's that? If this value is important to you, please document
It is documented via autoconf and really on
http://gcc.gnu.org/install/specific.html :).
--
pinskia at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
Component|c |middle-end
GCC build triplet|(see above) |
GCC host triplet|What's that? If this value |
|is important to you, please |
|document |
GCC target triplet|(see above) |
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36735
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug middle-end/36735] incomplete tree dump with "-fdump-tree-original-raw"
2008-07-05 0:39 [Bug c/36735] New: incomplete tree dump with "-fdump-tree-original-raw" grabner at icg dot tugraz dot at
` (3 preceding siblings ...)
2008-07-05 19:23 ` [Bug middle-end/36735] " pinskia at gcc dot gnu dot org
@ 2008-07-05 20:41 ` grabner at icg dot tugraz dot at
4 siblings, 0 replies; 6+ messages in thread
From: grabner at icg dot tugraz dot at @ 2008-07-05 20:41 UTC (permalink / raw)
To: gcc-bugs
------- Comment #5 from grabner at icg dot tugraz dot at 2008-07-05 20:40 -------
Ok, I inserted my best guess for host-target-build (since this particular
problem is not related to code generation, it shouldn't be too important here
anyway). A few proposals how to make it easier for first-time bug reporters to
create useful reports are here:
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36739
Regarding the "TDF_SLIM | ...", it still seems to me that these have been
inserted at some point in time for testing purposes and not yet been removed
since then. If a slim tree dump is desired, it can still be obtained by adding
the "-slim" switch. I am working on automatic differentiation of C code written
for gcc, and it seems natural to obtain the code tree directly from gcc. Since
the tree-dump options are documented in the man page, I took this approach, but
had to remove the TDF_SLIM override to get useful results.
--
grabner at icg dot tugraz dot at changed:
What |Removed |Added
----------------------------------------------------------------------------
GCC build triplet| |linux
GCC host triplet| |x86_64
GCC target triplet| |suse
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36735
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2008-07-05 20:41 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2008-07-05 0:39 [Bug c/36735] New: incomplete tree dump with "-fdump-tree-original-raw" grabner at icg dot tugraz dot at
2008-07-05 0:42 ` [Bug c/36735] " grabner at icg dot tugraz dot at
2008-07-05 0:43 ` grabner at icg dot tugraz dot at
2008-07-05 9:41 ` joseph at codesourcery dot com
2008-07-05 19:23 ` [Bug middle-end/36735] " pinskia at gcc dot gnu dot org
2008-07-05 20:41 ` grabner at icg dot tugraz dot at
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).