ARC: gcc - Fix SIZE_TYPE to be "unsigned int" instead of long unsigned int"
PATCH 1/1 ARC: gcc - Fix SIZE_TYPE to be "unsigned int" instead of long unsigned int"
This makes size_t to be "unsigned" ssize_t which makes happy compiler on data type checks.
Fix is taken from current development branch of GCC for ARC and will be a part of the next release of ARC tools, so at that point patch should be dropped.
Fixes http://autobuild.buildroot.net/results/405/405da9a945511329929b18740b983c51b8dcc43e
--- package/gcc/arc-2014.08/200-size_type_unsigned_int.patch | 11 +++++++++++ 1 file changed, 11 insertions(+) create mode 100644 package/gcc/arc-2014.08/200-size_type_unsigned_int.patch
package/gcc/arc-2014.08/200-size_type_unsigned_int.patch new file +11 -0
@@ -0,0 +1,11 @@ +--- a/gcc/config/arc/arc.h ++++ b/gcc/config/arc/arc.h +@@ -487,7 +487,7 @@ if (GET_MODE_CLASS (MODE) == MODE_INT \ + /* Define this as 1 if `char' should by default be signed; else as 0. */ + #define DEFAULT_SIGNED_CHAR 0 + +-#define SIZE_TYPE "long unsigned int" ++#define SIZE_TYPE "unsigned int" + #define PTRDIFF_TYPE "long int" + #define WCHAR_TYPE "int" + #define WCHAR_TYPE_SIZE 32
Signature · 1 line
1.9.3
PATCH 1/1 ARC: gcc - Fix SIZE_TYPE to be "unsigned int" instead of long unsigned int"
Dear Alexey Brodkin,
Quoted 12 lines · Alexey Brodkin
This makes
size_tto be "unsigned"ssize_twhich makes happy compiler on data type checks.Fix is taken from current development branch of GCC for ARC and will be a part of the next release of ARC tools, so at that point patch should be dropped.
Fixes http://autobuild.buildroot.net/results/405/405da9a945511329929b18740b983c51b8dcc43e
Signed-off-by Alexey Brodkin <[email protected]>
Quoted 7 lines
Cc Anton Kolesov <[email protected]>Cc Thomas Petazzoni <[email protected]>Cc Peter Korsgaard <[email protected]>--- package/gcc/arc-2014.08/200-size_type_unsigned_int.patch | 11 +++++++++++ 1 file changed, 11 insertions(+) create mode 100644 package/gcc/arc-2014.08/200-size_type_unsigned_int.patch
Thanks, looks good, but...
package/gcc/arc-2014.08/200-size_type_unsigned_int.patch
Quoted 6 lines
package/gcc/arc-2014.08/200-size_type_unsigned_int.patch new file +0
-0
Patch lacks a description + SoB.
Thanks!
Thomas
Signature · 3 lines
Thomas Petazzoni, CTO, Free Electrons Embedded Linux, Kernel and Android engineering http://free-electrons.com
PATCH 1/1 ARC: gcc - Fix SIZE_TYPE to be "unsigned int" instead of long unsigned int"
Hi Thomas,
Quoted 8 lines · Thomas Petazzoni
Dear Alexey Brodkin,
package/gcc/arc-2014.08/200-size_type_unsigned_int.patch
Quoted 5 lines
package/gcc/arc-2014.08/200-size_type_unsigned_int.patch new file +0
-0Patch lacks a description + SoB.
Well first of all IMHO patch name is self-explanatory.
Than I didn't want to duplicate the same text in commit message and in patch itself.
If you think this is helpful I will happily add the same text and URL in the patch.
Let me know what do you think about it.
-Alexey
PATCH 1/1 ARC: gcc - Fix SIZE_TYPE to be "unsigned int" instead of long unsigned int"
Dear Alexey Brodkin,
Quoted 8 lines · Alexey Brodkin
Quoted 7 lines
package/gcc/arc-2014.08/200-size_type_unsigned_int.patch
Quoted 5 lines
package/gcc/arc-2014.08/200-size_type_unsigned_int.patch new file +0
-0Patch lacks a description + SoB.
Quoted 8 lines
Well first of all IMHO patch name is self-explanatory.
Than I didn't want to duplicate the same text in commit message and in patch itself.If you think this is helpful I will happily add the same text and URL in the patch.
Let me know what do you think about it.
We have a stupid rule, and stupid rules are meant to be complied with :-)
Joke aside, I think what you mentioned in the commit log is useful:
what the patch does, and the fact that it's already "upstream" and therefore should be dropped at the next version bump of the ARC gcc compiler.
Thanks!
Thomas
Signature · 3 lines
Thomas Petazzoni, CTO, Free Electrons Embedded Linux, Kernel and Android engineering http://free-electrons.com
PATCH 1/1 ARC: gcc - Fix SIZE_TYPE to be "unsigned int" instead of long unsigned int"
Hi Thomas,
Quoted 16 lines · Thomas Petazzoni
Dear Alexey Brodkin,
Quoted 12 lines · Alexey Brodkin
Quoted 7 lines
package/gcc/arc-2014.08/200-size_type_unsigned_int.patch
Quoted 5 lines
package/gcc/arc-2014.08/200-size_type_unsigned_int.patch new file +0
-0Patch lacks a description + SoB.
Well first of all IMHO patch name is self-explanatory.
Than I didn't want to duplicate the same text in commit message and in patch itself.
Quoted 7 lines
Quoted 4 lines
If you think this is helpful I will happily add the same text and URL in the patch.
Let me know what do you think about it.
We have a stupid rule, and stupid rules are meant to be complied with :-)
Understood - uniformity is good even if the rule itself is not that good :)
Quoted 4 lines
Joke aside, I think what you mentioned in the commit log is useful:
what the patch does, and the fact that it's already "upstream" and therefore should be dropped at the next version bump of the ARC gcc compiler.
Funny enough that initially I had the same comment in the patch itself, but while looking at output of "git format-patch" I found myself a bit frustrated - if I were a reviewer I would say that submitted "full" patch looks ridiculous.
Then I removed comment from commit message. That made me think that you guys won't apply a patch that won't show any useful info in git log.
So I decided to move a message to the Buildroot commit. This made sense for me because at least when I bump version of a package I look at history of this package and this way try to figure out what's still required and what's not.
Moreover looking at the next folder "package/gcc/4.9.1" I found that half of patches don't have any comment inside, some have URLs to bugs/commits in GCC, some are full-scale patches with verbose description.
Now you may understand my frustration.
Anyways, I'm about to send v2 with comments in both commit message and patch itself.
Thanks for your time.
-Alexey
PATCH 1/1 ARC: gcc - Fix SIZE_TYPE to be "unsigned int" instead of long unsigned int"
Dear Alexey Brodkin,
Quoted 5 lines · Alexey Brodkin
Quoted 2 lines
We have a stupid rule, and stupid rules are meant to be complied with :-)
Understood - uniformity is good even if the rule itself is not that good :)
Absolutely :)
Quoted 5 lines
Funny enough that initially I had the same comment in the patch itself, but while looking at output of "git format-patch" I found myself a bit frustrated - if I were a reviewer I would say that submitted "full" patch looks ridiculous.
Quoted 7 lines
Then I removed comment from commit message. That made me think that you guys won't apply a patch that won't show any useful info in git log.
So I decided to move a message to the Buildroot commit. This made sense for me because at least when I bump version of a package I look at history of this package and this way try to figure out what's still required and what's not.
Hehe, funny. No problem if there is actually some duplication between the git commit log, and the patch description itself.
Quoted 4 lines
Moreover looking at the next folder "package/gcc/4.9.1" I found that half of patches don't have any comment inside, some have URLs to bugs/commits in GCC, some are full-scale patches with verbose description.
Half of the patches are carried over since gcc 4.2 or so, in a pre-2008 era, at a point where the state of Buildroot was clearly not as good as it is today. I think you wouldn't want to look at a pre-2008 Buildroot tree :-)
That being said, is your comment an indication that you volunteer to do some research about these patches, write proper descriptions, or even better help getting them merged upstream? :-)
Joke aside again, it's by having less and less patches with no description that at some point all patches in Buildroot will have a description. This way nobody will feel your frustration because it will just appear to be the "normal thing".
Quoted 1 line
Now you may understand my frustration.
Yes, of course, I do understand.
Thomas
Signature · 3 lines
Thomas Petazzoni, CTO, Free Electrons Embedded Linux, Kernel and Android engineering http://free-electrons.com