3 Jun
2016
3 Jun
'16
2:23 a.m.
uClibc-ng tries to be compatible with GNU libc and defines
__GLIBC__ and pretend to be version 2.2.
We once changed it to 2.10, but then some hard to fix problems
in different software packages (gcc) occured.
It would be better if we disable the special GNU libc checks
for uClibc-ng here. uClibc-ng implements the required scanf
functionality.
Signed-off-by: Waldemar Brodkorb <wbx(a)uclibc-ng.org>
---
configure.ac | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index f36b18c..4661c0d 100644
--- a/configure.ac
+++ b/configure.ac
@@ -581,7 +581,7 @@ AC_CACHE_VAL([scanf_cv_alloc_modifier],
#include <stdio.h>
#include <unistd.h>
- #ifdef __GLIBC__
+ #if defined(__GLIBC__) && !defined(__UCLIBC__)
#if !(__GLIBC_PREREQ(2, 7))
#error %m is not available
--
2.1.4
3 Jun
3 Jun
8:54 a.m.
New subject: [LEDE-DEV] [PATCH] fix uClibc-ng scanf check
Hi Waldemar
On 03/06/2016 04:23, Waldemar Brodkorb wrote:
uClibc-ng tries to be compatible with GNU libc and
defines
__GLIBC__ and pretend to be version 2.2.
We once changed it to 2.10, but then some hard to fix problems
in different software packages (gcc) occured.
It would be better if we disable the special GNU libc checks
for uClibc-ng here. uClibc-ng implements the required scanf
functionality.
Signed-off-by: Waldemar Brodkorb <wbx(a)uclibc-ng.org>
---
configure.ac | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index f36b18c..4661c0d 100644
--- a/configure.ac
+++ b/configure.ac
@@ -581,7 +581,7 @@ AC_CACHE_VAL([scanf_cv_alloc_modifier],
#include <stdio.h>
#include <unistd.h>
- #ifdef __GLIBC__
+ #if defined(__GLIBC__) && !defined(__UCLIBC__)
#if !(__GLIBC_PREREQ(2, 7))
#error %m is not available
this is the raw patch. please send a patch that adds this patch at the
right place inside the tree.
John
9:58 a.m.
Hi Waldemar,
On Fri, 2016-06-03 at 04:23 +0200, Waldemar Brodkorb wrote:
uClibc-ng tries to be compatible with GNU libc and
defines
__GLIBC__ and pretend to be version 2.2.
We once changed it to 2.10, but then some hard to fix problems
in different software packages (gcc) occured.
It would be better if we disable the special GNU libc checks
for uClibc-ng here. uClibc-ng implements the required scanf
functionality.
Signed-off-by: Waldemar Brodkorb <wbx(a)uclibc-ng.org>
---
configure.ac | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index f36b18c..4661c0d 100644
--- a/configure.ac
+++ b/configure.ac
@@ -581,7 +581,7 @@ AC_CACHE_VAL([scanf_cv_alloc_modifier],
#include <stdio.h>
#include <unistd.h>
- #ifdef __GLIBC__
+ #if defined(__GLIBC__) && !defined(__UCLIBC__)
#if !(__GLIBC_PREREQ(2, 7))
#error %m is not available
Even though this is a very minor and clean change I don't like this
approach. We're again implicitly assume something.
IMHO much better approach would be to include a compile test for
small source that uses scanf() with "%as"/"%ms".
That way we may remove all dependencies on either GLIBC/UCLIBC/MUSL etc.
-Alexey
1:43 p.m.
Hi Alexey,
On Fri, Jun 3, 2016 at 12:58 PM, Alexey Brodkin
<Alexey.Brodkin(a)synopsys.com> wrote:
IMHO much better approach would be to include a
compile test for
small source that uses scanf() with "%as"/"%ms".
I doubt a direct compile test is possible here, one need to run scanf
code to find out whether %m is actually supported.
--
Thanks.
-- Max
1:51 p.m.
Hi Max,
On Fri, 2016-06-03 at 16:43 +0300, Max Filippov wrote:
Hi Alexey,
On Fri, Jun 3, 2016 at 12:58 PM, Alexey Brodkin
<Alexey.Brodkin(a)synopsys.com> wrote:
Right, I see now.
So then indeed Waldemar's patch looks good.
Sorry for the noise.
-Alexey
IMHO much better approach would be to include a compile test for
small source that uses scanf() with "%as"/"%ms".
I doubt a
direct compile test is possible here, one need to run scanf
code to find out whether %m is actually supported.
3:14 p.m.
New subject: [LEDE-DEV] [PATCH] fix uClibc-ng scanf check
Hi John,
John Crispin wrote,
Hi Waldemar
On 03/06/2016 04:23, Waldemar Brodkorb wrote:
It isn't for LEDE, it is a patch for util-linux git tree.
That is the reason I changed To: field to util-linux maintainer.
I still preparing a patch for LEDE, I have some issues regarding
librt dependency.
best regards
Waldemar
uClibc-ng tries to be compatible with GNU libc
and defines
__GLIBC__ and pretend to be version 2.2.
We once changed it to 2.10, but then some hard to fix problems
in different software packages (gcc) occured.
It would be better if we disable the special GNU libc checks
for uClibc-ng here. uClibc-ng implements the required scanf
functionality.
Signed-off-by: Waldemar Brodkorb <wbx(a)uclibc-ng.org>
---
configure.ac | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index f36b18c..4661c0d 100644
--- a/configure.ac
+++ b/configure.ac
@@ -581,7 +581,7 @@ AC_CACHE_VAL([scanf_cv_alloc_modifier],
#include <stdio.h>
#include <unistd.h>
- #ifdef __GLIBC__
+ #if defined(__GLIBC__) && !defined(__UCLIBC__)
#if !(__GLIBC_PREREQ(2, 7))
#error %m is not available
this is the raw patch. please send a patch that adds this patch at the
right place inside the tree.
3:16 p.m.
Hi Alexey,
Alexey Brodkin wrote,
Hi Waldemar,
On Fri, 2016-06-03 at 04:23 +0200, Waldemar Brodkorb wrote:
Just for completeness, I saw your other mail already.
Musl has no problem with this code, because it is util-linux
fallback code for old GNU libc. Unfortunately it matches the
pretended version from uClibc-ng. So I think it is the right
solution, if the fallback code will not be removed.
best regards
Waldemar
uClibc-ng tries to be compatible with GNU libc
and defines
__GLIBC__ and pretend to be version 2.2.
We once changed it to 2.10, but then some hard to fix problems
in different software packages (gcc) occured.
It would be better if we disable the special GNU libc checks
for uClibc-ng here. uClibc-ng implements the required scanf
functionality.
Signed-off-by: Waldemar Brodkorb <wbx(a)uclibc-ng.org>
---
configure.ac | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index f36b18c..4661c0d 100644
--- a/configure.ac
+++ b/configure.ac
@@ -581,7 +581,7 @@ AC_CACHE_VAL([scanf_cv_alloc_modifier],
#include <stdio.h>
#include <unistd.h>
- #ifdef __GLIBC__
+ #if defined(__GLIBC__) && !defined(__UCLIBC__)
#if !(__GLIBC_PREREQ(2, 7))
#error %m is not available
Even though this is a very minor and clean change I don't like this
approach. We're again implicitly assume something.
IMHO much better approach would be to include a compile test for
small source that uses scanf() with "%as"/"%ms".
That way we may remove all dependencies on either GLIBC/UCLIBC/MUSL etc.
3:36 p.m.
Hi Waldemar,
On Fri, 2016-06-03 at 17:16 +0200, Waldemar Brodkorb wrote:
Hi Alexey,
Alexey Brodkin wrote,
Thanks for explanation.
-Alexey
Hi Waldemar,
On Fri, 2016-06-03 at 04:23 +0200, Waldemar Brodkorb wrote:
Just for
completeness, I saw your other mail already.
Musl has no problem with this code, because it is util-linux
fallback code for old GNU libc. Unfortunately it matches the
pretended version from uClibc-ng. So I think it is the right
solution, if the fallback code will not be removed.
uClibc-ng tries to be compatible with GNU libc and defines
__GLIBC__ and pretend to be version 2.2.
We once changed it to 2.10, but then some hard to fix problems
in different software packages (gcc) occured.
It would be better if we disable the special GNU libc checks
for uClibc-ng here. uClibc-ng implements the required scanf
functionality.
Signed-off-by: Waldemar Brodkorb <wbx(a)uclibc-ng.org>
---
configure.ac | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index f36b18c..4661c0d 100644
--- a/configure.ac
+++ b/configure.ac
@@ -581,7 +581,7 @@ AC_CACHE_VAL([scanf_cv_alloc_modifier],
#include <stdio.h>
#include <unistd.h>
- #ifdef __GLIBC__
+ #if defined(__GLIBC__) && !defined(__UCLIBC__)
#if !(__GLIBC_PREREQ(2, 7))
#error %m is not available
Even though this is a very minor and clean change
I don't like this
approach. We're again implicitly assume something.
IMHO much better approach would be to include a compile test for
small source that uses scanf() with "%as"/"%ms".
That way we may remove all dependencies on either GLIBC/UCLIBC/MUSL etc. 
6 Jun
6 Jun
8:05 a.m.
New subject: [LEDE-DEV] [PATCH] fix uClibc-ng scanf check
On Fri, Jun 03, 2016 at 10:54:57AM +0200, John Crispin wrote:
> configure.ac | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
Applied, thanks.
Karel
--
Karel Zak <kzak(a)redhat.com>
http://karelzak.blogspot.com
2906
days inactive
2909
days old
8 comments
5 participants
participants (5)
-
Alexey Brodkin -
John Crispin -
Karel Zak -
Max Filippov -
Waldemar Brodkorb