Found using the Clang Static Analyzer.
* libdiskfs/io-write.c (diskfs_S_io_write): Remove dead assignment.
---
libdiskfs/io-write.c |1 -
1 file changed, 1 deletion(-)
diff --git a/libdiskfs/io-write.c b/libdiskfs/io-write.c
index 26e0be4..2967c4c 100644
--- a/libdiskfs/io-write.c
+++ b/libd
Found using the Clang Static Analyzer.
* libdiskfs/dir-renamed.c (diskfs_rename_dir): Fix error handling.
---
libdiskfs/dir-renamed.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libdiskfs/dir-renamed.c b/libdiskfs/dir-renamed.c
index d73dc28..9b7ec3a 100644
--- a/libdis
Found using the Clang Static Analyzer.
* ext2fs/dir.c (diskfs_lookup_hard): Fix error handling.
---
ext2fs/dir.c |2 ++
1 file changed, 2 insertions(+)
diff --git a/ext2fs/dir.c b/ext2fs/dir.c
index c075246..a7eeaaa 100644
--- a/ext2fs/dir.c
+++ b/ext2fs/dir.c
@@ -195,6 +195,8 @@ diskfs_look
Hi folks :)
I've been using the Clang Static Analyzer to analyze the Hurd and GNU
Mach source. The results are here (until a more permanent home is
found and an automated build system is in place):
http://darnassus.sceen.net/~teythoon/qa/
I've seen Marin has joined in fixing bugs in Mach based o
Previously the increment was outside the loop body, turning it into a
dead increment.
Move the increment into the loop body as it is done in the true branch
above. This fixes the offsets recorded in the default_pager_page_t
objects.
Found using the Clang Static Analyzer.
* mach-defpager/default_
Bother, said Pooh.
* Makefile: Do not remove -Wall from CFLAGS.
---
mach-defpager/Makefile |3 ---
1 file changed, 3 deletions(-)
diff --git a/mach-defpager/Makefile b/mach-defpager/Makefile
index 5a98d69..e38a0be 100644
--- a/mach-defpager/Makefile
+++ b/mach-defpager/Makefile
@@ -35,6 +35,
Found using the Clang Static Analyzer.
* libshouldbeinlibc/timefmt.c (fmt_past_time): Fix dead assignment,
normalize adjacent white space.
---
libshouldbeinlibc/timefmt.c |4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libshouldbeinlibc/timefmt.c b/libshouldbeinlibc/t
Justus Winter, le Sat 16 Nov 2013 10:58:13 +0100, a écrit :
> The code in libftpconn and libstore/do-bunzip2.c is very noisy.
It's probably better to make libstore just use libbz2 than fixing
unmaintained code :)
Samuel
Marin Ramesa, le Fri 15 Nov 2013 19:39:20 +0100, a écrit :
> MACH_INLINE is already defined where needed. Inline functions from seg.h and
> tss.h get compiled when included from other files, therefore seg.c serves no
> real purpose. I think it's safe to remove.
It does serve purpose: the inline is
Justus Winter, le Fri 15 Nov 2013 20:47:37 +0100, a écrit :
> * libihash/ihash.c: Clean up the included header files.
> * libshouldbeinlibc/cacheq.c: Likewise.
> * libshouldbeinlibc/canon-host.c: Likewise.
> * libshouldbeinlibc/fsysops.c: Likewise.
> * libshouldbeinlibc/idvec-auth.c: Likewise.
> *
Justus Winter, le Fri 15 Nov 2013 20:52:23 +0100, a écrit :
> @@ -224,30 +216,22 @@ ports_manage_port_operations_multithread (struct
> port_bucket *bucket,
>
>if (master)
> {
> - pthread_spin_lock (&lock);
> - if (totalthreads != 1)
> - {
> - pthread_s
Justus Winter, le Sat 16 Nov 2013 10:58:14 +0100, a écrit :
> Found using the Clang Static Analyzer.
>
> * libdiskfs/io-write.c (diskfs_S_io_write): Remove dead assignment.
Ack
> ---
> libdiskfs/io-write.c |1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/libdiskfs/io-write.c b/libdis
Justus Winter, le Sat 16 Nov 2013 10:58:15 +0100, a écrit :
> Found using the Clang Static Analyzer.
>
> * libdiskfs/dir-renamed.c (diskfs_rename_dir): Fix error handling.
Ack!
It was really bad, AIUI it meant creating 32000 directories in the same
parent leads to leaked locked mutexes.
> ---
>
Justus Winter, le Sat 16 Nov 2013 10:58:16 +0100, a écrit :
> Found using the Clang Static Analyzer.
>
> * ext2fs/dir.c (diskfs_lookup_hard): Fix error handling.
Ack.
> ---
> ext2fs/dir.c |2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/ext2fs/dir.c b/ext2fs/dir.c
> index c075246.
Justus Winter, le Sat 16 Nov 2013 10:58:17 +0100, a écrit :
> Found using the Clang Static Analyzer.
>
> * libshouldbeinlibc/timefmt.c (fmt_past_time): Fix dead assignment,
> normalize adjacent white space.
Ack.
> ---
> libshouldbeinlibc/timefmt.c |4 ++--
> 1 file changed, 2 insertions(+
Justus Winter, le Sat 16 Nov 2013 10:58:18 +0100, a écrit :
> Previously the increment was outside the loop body, turning it into a
> dead increment.
>
> Move the increment into the loop body as it is done in the true branch
> above. This fixes the offsets recorded in the default_pager_page_t
> ob
Justus Winter, le Sat 16 Nov 2013 10:58:19 +0100, a écrit :
> Bother, said Pooh.
>
> * Makefile: Do not remove -Wall from CFLAGS.
Ack.
> ---
> mach-defpager/Makefile |3 ---
> 1 file changed, 3 deletions(-)
>
> diff --git a/mach-defpager/Makefile b/mach-defpager/Makefile
> index 5a98d69..e
Justus Winter, le Sat 16 Nov 2013 10:58:13 +0100, a écrit :
> Please pay close attention to this one. The Hurd seems to work fine
> with this patch.
>
> [PATCH 6/6] mach-defpager: do not remove -Wall from CFLAGS
Well, -Wall shouldn't have any impact on the generated code. Does that
really produc
Quoting Samuel Thibault (2013-11-16 11:16:02)
> > - pthread_spin_lock (&lock);
> > - if (nreqthreads == 1)
> > + __atomic_sub_fetch (&totalthreads, 1, __ATOMIC_RELAXED);
> > + if (__atomic_sub_fetch (&nreqthreads, 1, __ATOMIC_RELAXED) == 0)
> > {
> > /*
On Sat, Nov 16, 2013 at 11:41:03AM +0100, Justus Winter wrote:
> I just realized that I tested these changes *with* the
> libports_stability.patch, so the threads do not actually time out and
> so this particular code is never reached. Maybe we should defer this
> patch until Richards patch destruc
On 16.11.2013 10:58:13, Justus Winter wrote:
> I've seen Marin has joined in fixing bugs in Mach based on these kind
> of analysis, very cool!
Yeah, I've been following your reports. It seems this kind of analysis
does emit some false alarms, as seen with my recent patch on null
pointer derefer
Hello,
Attached is a gzipped series of 26 patches which do a small cleanup of
the kernel debugger source. Cleanups include:
1. Removal of register qualifiers.
2. Addition of ifndefs in header files.
3. Addition of comments after endifs.
ddb_cleanup.tar.gz
Description: application/compressed-ta
Hi!
On Tue, 2013-11-12 at 15:47:00 +0100, Marin Ramesa wrote:
> Attached is a gzipped series of 32 patches which remove register
> qualifiers from kern source. I did not want to spam the list with 32
> patches, so I'm sending them as an attachment.
Personally I find tarred patch series annoying
On 16.11.2013 16:15:45, Guillem Jover wrote:
> Hi!
>
> On Tue, 2013-11-12 at 15:47:00 +0100, Marin Ramesa wrote:
> > Attached is a gzipped series of 32 patches which remove register
> > qualifiers from kern source. I did not want to spam the list with
> > 32 patches, so I'm sending them as an at
On Sat, Nov 16, 2013 at 04:39:01PM +0100, Marin Ramesa wrote:
> I assumed it's inconvinient for people to suddenly find 32 messages in
We think of it (and our viewers make us see it) as one thread, not 32
messages.
--
Richard Braun
On 16.11.2013 17:12:37, Richard Braun wrote:
> On Sat, Nov 16, 2013 at 04:39:01PM +0100, Marin Ramesa wrote:
> > I assumed it's inconvinient for people to suddenly find 32 messages
> > in
>
> We think of it (and our viewers make us see it) as one thread, not 32
> messages.
I have threading turne
Marin Ramesa, le Sat 16 Nov 2013 16:39:01 +0100, a écrit :
> Plus, it's a repetitive patch series, so I assumed the better solution
> is to compress it and send it as an attachment.
More than being repetitive, register dropping is trivial to review and
obviously correct, so there is essentially no
These two are preliminary patches generated from dereference of null
pointer reports. Let's see how this goes. If I got it right, I'll
continue to work on this. If I got it wrong, I'll let someone more
experienced deal with these kind of warnings.
Below are the links to reports.
[PATCH 1/2] i386/
When ldt equals zero, and default branch is taken from the switch
statement, and sel is not equal to zero, comparison results in
a dereference of a null pointer. Avoid this.
* i386/i386/user_ldt.c (ldt): Check if it equals zero.
---
i386/i386/user_ldt.c | 7 ---
1 file changed, 4 insertions(
When fp_thread is not NULL and is not the current thread, and fp_save() does
not alter the machine state, check if ifps is NULL before setting fp_valid
to avoid dereference of null pointer.
* i386/i386/fpu.c (ifps): Check if it's NULL.
---
i386/i386/fpu.c | 3 ++-
1 file changed, 2 insertions(+
30 matches
Mail list logo
