From: Jim Panetta
1) The value of rtems_bsdnet_ntpserver_count is equal to 0 when no
server is set, so the check for (rtems_bsdnet_ntpserver_count < 0)
in rtems_bsdnet_get_ntp() is wrong. The check should be "<= 0".
2) Binding the listening socket port to 0 does not work. Packets
appe
From: Chirayu Desai
* Move shared/console/bspreset.c to motorola_powerpc/startup/bspreset.c
---
.../libbsp/powerpc/motorola_powerpc/Makefile.am|4 +--
.../powerpc/motorola_powerpc/startup/bspreset.c| 27
c/src/lib/libbsp/powerpc/shared/console/bspreset.c | 27
t1/mmtest1.doc
> create mode 100644 testsuites/libtests/mmtest1/mmtests.scn
> create mode 100644 testsuites/libtests/mmtest1/system.h
>
> diff --git a/c/src/lib/libcpu/shared/include/mm.h
> b/c/src/lib/libcpu/shared/include/mm.h
> new file mode 100644
> index 000..02903f1
On Wed, Jul 24, 2013 at 12:43 PM, Rempel, Cynthia
wrote:
>>
>>From: ged...@gwmail.gwu.edu [ged...@gwmail.gwu.edu] on behalf of Gedare Bloom
>>[ged...@rtems.org]
>>Sent: Wednesday, July 24, 2013 9:19 AM
>>To: Rempel, Cynthia
Hi,
The approach we have taken so far with libmm is that it is "always on"
even if it is not supported by the BSP/CPU. In the cases that it is
not supported, the functions are nops that may have minimal effect on
runtime if users try to use the libmm. (Currently there is no API for
using libmm.)
I'm happy with the current state, which will execute but not do
anything. For the libmm usage I think this is OK. I see no compelling
reason to change the stub implementation to anything other than a nop.
Turning it into a fatal error makes testing it harder, since it
becomes a "fatal test" for BSP
tinue to
post to your blogs, I have been checking them each week to quietly
monitor progress, and some of you have been a little less active at
blogging.
Thanks,
Gedare
___
rtems-devel mailing list
rtems-devel@rtems.org
http://www.rtems.org/mailman/lis
Does dispatch_repressed == dispatch_disabled || isr_disabled ?
Is there a real difference between repressed and disabled?
On Fri, Jul 26, 2013 at 8:48 AM, Sebastian Huber
wrote:
> ---
> cpukit/score/Makefile.am |1 +
> cpukit/score/include/rtems/score/assert.h
Now that the free list is committed, can you recreate these patches for review?
On Sat, Jul 6, 2013 at 3:17 AM, Ashi wrote:
> Hi all, this patch includes a reimplement of POSIX Key manager and compared
> with first version of this patch, some improvement is added.
>
> *In short*:
> it enable unli
OK I think the word you chose is fine. I understood the intent if not
the technical difference from being disabled.
On Fri, Jul 26, 2013 at 9:09 AM, Sebastian Huber
wrote:
> On 2013-07-26 14:59, Gedare Bloom wrote:
>>
>> Does dispatch_repressed == dispatch_disabled || isr_disa
I can look at it and think through what it means hopefully early next week.
On Fri, Jul 26, 2013 at 10:48 AM, Sebastian Huber
wrote:
> Hello,
>
> please review this change carefully. It may introduce a subtle but
> devastating problem.
>
> --
> Sebastian Huber, embedded brains GmbH
>
> Address :
Chris (and everyone else), let us try to keep the first line of commit
messages brief so they fit on the first line. This makes reading the
"short log" and email subjects much easier.
-Gedare
On Sat, Jul 27, 2013 at 9:52 AM, Chris Johns wrote:
> Module:rtems
> Branch:
many pthreads in the
test?
-Gedare
On Jul 28, 2013 11:55 AM, "Ashi" wrote:
> I'm updating this patch this days, but get stuck with several problems:
>
> - when key deleting, should we just put the deleted node back to
> freechain or run workspace_free on these delete
It seems like it should work fine, as long as the ISR does not update
_Thread_Executing or reset the disable_level to 0 which I think would
both be bugs anyway...
On Fri, Jul 26, 2013 at 11:27 AM, Gedare Bloom wrote:
> I can look at it and think through what it means hopefully early next w
gdb connects to qemu.
you may need to specify in gdb some extra information about the
locations of files, for example:
file
/home/gedare/work/rtems/rtems/b-pc386/i386-rtems4.11/c/pc386/testsuites/samples/hello/hello.exe
if you are repeatedly debugging the same file connecting "remotely&quo
This seems like a good project. Ballista looks like it is "dead"
upstream. You'll want to scope your project to decide what / how much
of RTEMS you can reasonably instrument with fault injection. I also
wonder if it would be sensible to first start with a tool that can
support test input "fuzzing",
Is there a possibility an ISR happens during the
set_dispatch_level(1)? It can observe a 0-value dispatch level in that
case, and dispatch at the end of the isr?
On Mon, Jul 29, 2013 at 3:19 PM, Joel Sherrill
wrote:
> On 7/29/2013 9:17 AM, Gedare Bloom wrote:
>>
>> It seems like
As Chris said you may want to check how much free wkspace remains. Maybe
you can copy the object handler approach. Maybe later they can even share
code..
On Jul 29, 2013 9:38 PM, "Ashi" wrote:
>
>
> On Mon, Jul 29, 2013 at 11:18 AM, Gedare Bloom wrote:
>
>> I would
OK I agree
On Tue, Jul 30, 2013 at 3:38 AM, Sebastian Huber
wrote:
> On 2013-07-30 03:43, Gedare Bloom wrote:
>>
>> As Chris said you may want to check how much free wkspace remains. Maybe
>> you
>> can copy the object handler approach. Maybe later they can even share
On Tue, Jul 30, 2013 at 5:27 AM, Sebastian Huber
wrote:
> Hello R.,
>
>
> On 2013-07-30 11:19, R. Diez wrote:
>>
>> Hi all:
>>
>> I am still trying to port RTEMS to the Arduino Due. After having
>> implemented the start-up code and the basic serial port console, I am trying
>> to get the hello.exe
We should have a process to deprecate CONFIGURE options, although
usually we will just check if the user defines it and otherwise
provide a definition, which in this case will be the sum of the static
and dynamic configuration options(with dynamic set to 0 if undefined)
On Tue, Jul 30, 2013 at 11:
Mentors and students remember to fill out your evaluations on the GSoC
Melange site.
As always, if you have any private concerns or questions feel free to ask
myself, Joel, or Chris.
-Gedare
___
rtems-devel mailing list
rtems-devel@rtems.org
http
their respective
locations in the build tree. It is all a mess.
-Gedare
On Wed, Jul 31, 2013 at 10:56 AM, R. Diez wrote:
> Hi all:
>
> I am finding it difficult to choose a place in the source tree for my new .c
> and .h files. In a previous message in this mailing list, I read that
the organization and naming of the sptests is a mess to figure out
what tests are testing. i think there was some effort by the testing
framework student last year to work on this spaghetti of test cases..
I much prefer at the very least to give the test a meaningful name as
Sebastian suggested wit
Good work Zhongwei! This was a good effort and will be useful for many
in the future. I hope you can find other little RTEMS projects to work
on as your time and interest permits.
-Gedare
On Sun, Aug 4, 2013 at 8:23 PM, Ashi wrote:
> Here is the patch created by git format-patch.
>
&g
Is anything lost by going to POSIX threads in GCC?
On Fri, Aug 9, 2013 at 3:06 AM, Chris Johns wrote:
> Hi,
>
> The patch attached changes the RTEMS thread model in GCC from 'rtems' to
> 'posix'. The patch also fixes the include paths for newlib so the correct
> 'machine/_types.h' is picked up.
>
gt;
Can this "thread" parameter be used to "Update" the per-thread
scheduler metadata and obviate the call to _Scheduler_Update?
-Gedare
___
rtems-devel mailing list
rtems-devel@rtems.org
http://www.rtems.org/mailman/listinfo/rtems-devel
On Mon, Aug 12, 2013 at 9:12 AM, Sebastian Huber
wrote:
> On 2013-08-12 16:11, Gedare Bloom wrote:
>>
>> On Fri, Aug 9, 2013 at 2:35 AM, Sebastian Huber
>> wrote:
>>
>>>
>>> @@ -59,10 +60,12 @@ void _Scheduler_Handler_initialization( void );
>>
On Mon, Aug 12, 2013 at 12:15 PM, Ralf Corsepius
wrote:
> On 08/12/2013 09:57 AM, Sebastian Huber wrote:
>>
>> Hello,
>>
>> there was some lobbying against the deprecation of the i386 support of
>> GCC:
>>
>> http://gcc.gnu.org/ml/gcc/2012-12/msg00132.html
>>
>> Is this really justified? Why does
On Tue, Aug 13, 2013 at 9:42 AM, Sebastian Huber
wrote:
> Add Thread_Control::is_in_the_air field if configured for SMP. This
> helps to simplify the extract operation and avoids superfluous
> inter-processor interrupts. Move the processor allocation step into the
> enqueue operation.
>
> Add an
On Wed, Aug 14, 2013 at 11:21 AM, Rempel, Cynthia
wrote:
The RTEMS project has currently nothing to sign.
>>>
>>> Signed-off-by:
>>>
>>> is standard Linux kernel patch practice to declare that
>>> sender is the author or the patch or he proves other
>>> source (sender of patch) to be complian
On Wed, Aug 14, 2013 at 12:16 PM, Sebastian Huber
wrote:
> On 2013-08-14 17:41, Gedare Bloom wrote:
>>>
>>> + * XXX: Do not exchange parameters to do the negation of the order
>>> check.
>>> >+ */
>>
>> What does it mean here to use
ven a proper RTEMS name, e.g.
_CPU_Virtual_Disable_interrupts() or something.
* Add your copyright to the virtual/include/rtems/score/interrupts.h.
You substantially transformed this code and should lay claim to your
copyright.
-Gedare
On Wed, Jul 31, 2013 at 4:17 PM, Philipp Eppelt
wrote:
> Pat
is unused, do not include it.
-Gedare
On Wed, Jul 31, 2013 at 4:18 PM, Philipp Eppelt
wrote:
> Patch introducing libbsp/i386/virtPok BSP.
>
> ___
> rtems-devel mailing list
> rtems-devel@rtems.org
> http:/
Hi,
Would it be better to add an option to specify the maximum number of
parallel jobs to run? Also, is it 10 jobs or 20 jobs maximum right
now? I think there can be up to 10 controlled by n_confs, and up to 10
controlled by n_reconfs?
-Gedare
On Fri, Aug 2, 2013 at 2:49 AM, Rempel, Cynthia
This looks fine to commit. If no one does it within a week just ping
this message.
-Gedare
On Wed, Aug 14, 2013 at 4:12 AM, Nick Withers wrote:
> Hey all,
>
> Attached is a patch exposing rtems_verror().
>
> I'd personally like this as I'm writing up BSD err() / warn()
On Wed, Aug 14, 2013 at 2:52 PM, Chris Johns wrote:
> Gedare Bloom wrote:
>>
>> On Tue, Aug 13, 2013 at 9:42 AM, Sebastian Huber
>> wrote:
>>>
>>>
>>> /**
>>> + * @brief This field is true if the thread is in the air.
>>&
OK, I would say use a "-j" if that is not already in use, to be
similar to make's job parallelism option.
On Wed, Aug 14, 2013 at 3:25 PM, Rempel, Cynthia
wrote:
>>
>>From: ged...@gwmail.gwu.edu [ged...@gwmail.gwu.edu] on beh
You ought to test the patch that moves code from score to libcpu with
the tmtests to ensure that none of the timing changes. (Hm, does
pc386/qemu offer accurate tmtest timing results?)
-Gedare
On Wed, Aug 14, 2013 at 1:36 PM, Gedare Bloom wrote:
> Please merge the part of
CC:Vipul
Vipul, please have a look at the pc386 and determine if the changes
you provided have caused the compilation errors we are seeing.
-Gedare
On Thu, Aug 15, 2013 at 11:00 PM, Chris Johns wrote:
> Sree Harsha Konduri wrote:
>>
>> ../../../../../pc386/lib/librtemsbsp.
OK. If you find the bug in the modified clockdrv_shell.h think about
how it might be fixed.
On Fri, Aug 16, 2013 at 11:49 AM, Vipul Nayyar wrote:
> Hello,
>
> Will look more closely into the matter.
>
> FWIW, I compiled the pc386 bsp giving the same configure statement as Sree
> gave, in my perso
Send a separate email with this bug info, or file a bug report. This
is unrelated and was probably introduced by the code changes
contributed by Hengyi.
-Gedare
On Sat, Aug 17, 2013 at 2:38 AM, Sree Harsha Konduri
wrote:
> Hello,
>
> After applying the patch that Joel has sent out, t
bastian Huber
wrote:
> On 2013-08-14 17:41, Gedare Bloom wrote:
>>>
>>> diff --git a/cpukit/score/include/rtems/score/thread.h
>>> b/cpukit/score/include/rtems/score/thread.h
>>> >index d346ead..984bc0e 100644
>>> >--- a/cpukit/score/include
... ask if you don't figure it out easily. :)
-Gedare
On Mon, Aug 19, 2013 at 12:34 PM, Philipp Eppelt
wrote:
> Is there an automatic way to do all tmtests on qemu?
>
> The score-libcpu_native patch is ready. I just need to do the tests.
>
>
> On 08/15/2013 06:25 PM, Joel Sh
How much space overhead is the ISR lock? I prefer the solution (1) if
the overhead is not that much.
On Tue, Aug 20, 2013 at 8:59 AM, Sebastian Huber
wrote:
> Hello,
>
> on single processor systems the ISR disable/enable was the big hammer which
> ensured system-wide mutual exclusion. On SMP con
On Tue, Aug 20, 2013 at 4:24 AM, Sebastian Huber
wrote:
> On 2013-08-19 20:53, Sebastian Huber wrote:
>>
>> On 2013-08-19 20:42, Gedare Bloom wrote:
>>>
>>> That's good with me. I assumed you wanted to somehow encode both
>>> "!is_scheduled
printf( "...st_blocks %" PRIxblkcnt_t "\n", (long unsigned
int)statbuf.st_blocks );
Here the printf already should accommodate the size of the type of the
blksize_t and blkcnt_t. This is how I usually see this issue of
variably-sized types being printed. I don't know
On Fri, Aug 23, 2013 at 6:58 AM, Sebastian Huber
wrote:
> Return if the caller performed the extract operation. The caller may
> perform certain actions if it was the one who actually performed the
> extract operation.
> ---
> cpukit/score/include/rtems/score/threadqimpl.h | 12 +---
>
This seems OK. I have not looked deeply enough but it bugs me that
maybe there is a race condition on the_thread->Wait.return_code also?
-Gedare
On Fri, Aug 23, 2013 at 6:58 AM, Sebastian Huber
wrote:
> The _Thread_queue_Process_timeout() operation had several race
> conditions in the
Ah, Ok. Thanks for the clarification.
On Fri, Aug 23, 2013 at 2:28 PM, Sebastian Huber
wrote:
> On 2013-08-23 16:46, Gedare Bloom wrote:
>>
>> This seems OK. I have not looked deeply enough but it bugs me that
>> maybe there is a race condition on the_thread-&
On Fri, Aug 23, 2013 at 2:23 PM, Sebastian Huber
wrote:
> On 2013-08-23 16:38, Gedare Bloom wrote:
>>>
>>> >diff --git a/cpukit/score/include/rtems/score/threadqimpl.h
>>> > b/cpukit/score/include/rtems/score/threadqimpl.h
>>> >index c72982d..0762
On Sun, Aug 25, 2013 at 6:36 AM, Vipul Nayyar wrote:
> Hello,
>
> Covering for earlier incomplete work left..
> I've compiled a bit of information related to how the lpc24xx BSP is
> organized. The details can also be found here
> https://docs.google.com/document/d/1W9DMnpocXUKXiQMxHbbp0aG-ecuh94P
> +++ b/cpukit/score/include/rtems/score/mm.h
> @@ -0,0 +1,52 @@
> +/**
> + * @file
> + *
> + * @brief Manages use of MPU/MMU units to provide memory management.
> + */
> +
> +/*
> + * Copyright (c) 2013 Hesham Al-Matary.
> + * Copyright (c) 2013 Gedare Bloom.
> + *
On Mon, Aug 26, 2013 at 11:52 AM, Sebastian Huber
wrote:
> On 2013-08-20 14:59, Sebastian Huber wrote:
>>
>>
>> One big issue are the RTEMS chains:
>>
>>
>> http://www.rtems.org/onlinedocs/doxygen/cpukit/html/group__ClassicChains.html
>>
>> To make them SMP ready we have two options.
>>
>> 1. Inst
On Sun, Aug 25, 2013 at 8:14 PM, Hesham AL-Matary
wrote:
> ---
> c/src/lib/libcpu/arm/shared/include/arm-cp15.h | 17
> .../libcpu/arm/shared/include/arm_cp15_print_fsr.h | 84
> c/src/lib/libcpu/arm/shared/src/armv7-mm-mpu.c | 108
> +
> c/src
I don't have answers to your two questions specifically. However, I
did some unification work related to set vector over a year ago that
never got committed. See
http://git.rtems.org/gedare/rtems.git/log/?h=setvec
2012-04-16bsps: rename set_vector to bsp_set_vectorGedare Bloom
2012-04-16bsps
make/local.am
> diff --git a/testsuites/libtests/mmtest1/init.c
> b/testsuites/libtests/mmtest1/init.c
> new file mode 100644
> index 000..9e966ba
> --- /dev/null
> +++ b/testsuites/libtests/mmtest1/init.c
> @@ -0,0 +1,62 @@
> +/* Init
> + *
> + * This routi
On Mon, Aug 26, 2013 at 2:15 PM, Hesham Moustafa
wrote:
>
>
>
> On Mon, Aug 26, 2013 at 5:22 PM, Sebastian Huber
> wrote:
>>
>> On 2013-08-26 02:14, Hesham AL-Matary wrote:
>>>
>>> +/**
>>> + * @brief _Memory_management_Region Flags defs
>>> + */
>>> +#define RTEMS_MM_REGION_NO_PROTECTION
ome other clean up on the driver.
>
> With the attached patch, it compiles again and
> ticker runs.
>
> Please review and let me know if OK to commit.
>
>
> On 8/16/2013 11:03 AM, Gedare Bloom wrote:
>>
>> OK. If you find the bug in the modified clockdrv_shell.h
On Tue, Aug 27, 2013 at 4:15 AM, Sebastian Huber
wrote:
>
> For the ARM exception handling see:
>
> http://git.rtems.org/rtems/tree/cpukit/score/cpu/arm/armv4-exception-default.S
>
> They end up in:
>
> void _ARM_Exception_default( CPU_Exception_frame *frame )
> {
> rtems_fatal( RTEMS_FATAL_SOUR
tatic attribute. It does not make a lot of
sense to me for a variable to be static in a header file.
The tests in patch 0005 still need some fixes to deal with workspace
allocation, and eventually to make them flexible and not use
hard-coded addresses.
-Gedare
On Mon, Aug 26, 2013 at 8:23 PM, Hesham Mou
Yes I think the RB Heap would be an ideal solution. My intuition is
that the exception is caused exactly by the heap (workspace) metadata
being written, but Hesham should hook up a debugger to verify the
source of the exception.
-Gedare
On Wed, Aug 28, 2013 at 3:00 AM, Sebastian Huber
wrote
On Wed, Aug 28, 2013 at 9:19 AM, Sebastian Huber
wrote:
> Hello,
>
> as part of a GSoC project we made the atomic operations API compatible with
> the C11 operations and use them for the implementation. I added some tests
> today and now begin to think if this was really such a good idea:
>
> htt
Thanks Pavel. More below.
On Wed, Aug 28, 2013 at 10:31 AM, Pavel Pisa wrote:
> Hello All,
>
> I want to express my two cents even that I cannot find time
> to do more than chin-wag, sorry too many other project
> required for university an company stay in a operation.
>
> The containers basic AP
't know if that is the
right way we want to use locks, or if the lock should be encapsulated
inside of some kind of MM structure (even if that structure is only a
wrapper for the lock).
-Gedare
On Wed, Aug 28, 2013 at 12:11 PM, Hesham Moustafa
wrote:
> Hey all,
>
> Please review the
the naming rules and the leading underscore means it is
> private. So if the bsp or app init is to perform it, it needs a public style
> name
>
> Gedare Bloom wrote:
>
>
> Does the high level score interface need to include
> "_Memory_management_Initialize"? Is MMU/
etails of that routine but if it is called outside the
> implementation of POSIX, sapi, or classic, then we need to discuss it. From
> my perspective BSPs are just a special type of user code.
>
> Gedare Bloom wrote:
>
>
> I should have been a tad more clear. The
if it becomes part of the "bsp" files, then there should be a
libbsp/shared/mminit.c file with a default implementation, and then
each BSP can override with its particular implementation. So you could
put libbsp/arm/shared/mminit.c (or maybe arm-cp15-mminit.c).
-Gedare
On Fri, Aug 30,
cated into some "bsp_memory_management"
implementations.
On Sat, Aug 31, 2013 at 11:01 AM, Gedare Bloom wrote:
> if it becomes part of the "bsp" files, then there should be a
> libbsp/shared/mminit.c file with a default implementation, and then
> each BSP can overrid
/bsppredriverhook.c (it
omits "startup")
-Gedare
On Fri, Aug 30, 2013 at 6:49 PM, Vipul Nayyar wrote:
> Hello DrJoel,
>
> Haven't heard from you for a long time. I hope you're well. I also sincerely
> hope that you're following the discussion about the p
Also, tm27.h can be (usually is) overwritten by the BSP and located in
./include/tm27.h.
On Sat, Aug 31, 2013 at 11:08 AM, Gedare Bloom wrote:
> Vipul, would it make sense to also look for the specified files under
> the BSP in all its subdirectories and report that information in case
On Sat, Aug 31, 2013 at 11:11 AM, Gedare Bloom wrote:
> Also, tm27.h can be (usually is) overwritten by the BSP and located in
> ./include/tm27.h.
>
The same goes for coverhd.h
> On Sat, Aug 31, 2013 at 11:08 AM, Gedare Bloom wrote:
>> Vipul, would it make sense to also look
Welcome! The lm4f120 was newly added as a bsp. Trying out both with vanilla
rtems to "see" running code is a good place to start. Then testing libmm
would be good, especially to see what might be available on the TI board.
-Gedare
On Sep 2, 2013 1:33 AM, "Ritesh Harjani" w
ctions that start with an underscore, for example
_CPU_cache_flush_1_data_line(), _CPU_ISR_install_raw_handler(),
_ISR_Disable(), etc. Identify such possible violations.
-Gedare
On Wed, Sep 4, 2013 at 10:30 AM, Vipul Nayyar wrote:
>
> Hello,
>
> Eagerly waiting for your comments, suggestion
In case the length of cwd path plus the userScriptName exceeds
PATH_MAX (255), the strncat calls will overflow scriptFile. Also
check for getcwd failure.
---
cpukit/libmisc/shell/shell_script.c | 23 +++
1 files changed, 15 insertions(+), 8 deletions(-)
diff --git a/cpukit/l
On Thu, Sep 5, 2013 at 12:18 PM, Gedare Bloom wrote:
> In case the length of cwd path plus the userScriptName exceeds
> PATH_MAX (255), the strncat calls will overflow scriptFile. Also
> check for getcwd failure.
> ---
> cpukit/libmisc/shell/shell_script.c | 23 +
Change the type for storing the return from iconv to be signed.
---
cpukit/libfs/src/dosfs/msdos_conv_utf8.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/cpukit/libfs/src/dosfs/msdos_conv_utf8.c
b/cpukit/libfs/src/dosfs/msdos_conv_utf8.c
index a80db7e..18aebc6 100644
On Thu, Sep 5, 2013 at 1:17 PM, Gedare Bloom wrote:
> Change the type for storing the return from iconv to be signed.
> ---
> cpukit/libfs/src/dosfs/msdos_conv_utf8.c |2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/cpukit/libfs/src/dosfs/msdo
Well, I'm fixing without testing right now. A lot of these changes due
to Coverity probably are not introducing easily testable cases (other
than to ensure our existing tests pass). I'll try to give a good run
through yet today before I finish up. If nothing seems broken I'll
push
---
cpukit/libfs/src/imfs/imfs_load_tar.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/cpukit/libfs/src/imfs/imfs_load_tar.c
b/cpukit/libfs/src/imfs/imfs_load_tar.c
index 723699b..abb9f46 100644
--- a/cpukit/libfs/src/imfs/imfs_load_tar.c
+++ b/cpukit/libfs/src/imfs/i
In case obtain/release is unsuccessful, cause a fatal error.
---
cpukit/libblock/src/sparse-disk.c | 11 +--
1 files changed, 9 insertions(+), 2 deletions(-)
diff --git a/cpukit/libblock/src/sparse-disk.c
b/cpukit/libblock/src/sparse-disk.c
index 987ab6d..ab6a25d 100644
--- a/cpukit/li
The arguments to rtems_nvdisk_calc_crc16() are reversed when called from
rtems_nvdisk_page_checksum().
26053 Operands don't affect result
---
cpukit/libblock/src/nvdisk.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/cpukit/libblock/src/nvdisk.c b/cpukit/libblock/src/n
Raise a fatal error in case of failure.
---
cpukit/libcsupport/src/termios.c |9 +++--
1 files changed, 7 insertions(+), 2 deletions(-)
diff --git a/cpukit/libcsupport/src/termios.c b/cpukit/libcsupport/src/termios.c
index 976c376..c462ce6 100644
--- a/cpukit/libcsupport/src/termios.c
+++
The arguments to rtems_fdisk_calc_crc16() are reversed when called from
rtems_fdisk_page_checksum().
26052 Operands don't affect result
---
cpukit/libblock/src/flashdisk.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/cpukit/libblock/src/flashdisk.c b/cpukit/libblock/s
If open(fd) fails then avoid the error path that calls close(fd).
---
cpukit/libblock/src/bdpart-read.c |5 +++--
1 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/cpukit/libblock/src/bdpart-read.c
b/cpukit/libblock/src/bdpart-read.c
index fd8a038..cfb31d5 100644
--- a/cpukit/lib
Explicitly promote 32-bit integer to 64-bits for multiplication.
---
cpukit/libfs/src/dosfs/msdos_format.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/cpukit/libfs/src/dosfs/msdos_format.c
b/cpukit/libfs/src/dosfs/msdos_format.c
index a939df5..154282f 100644
--- a/cp
Fix the type of "cmpltd" to be ssize_t so that assigning it to -1
will terminate processing as intended.
---
cpukit/libfs/src/dosfs/msdos_dir.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/cpukit/libfs/src/dosfs/msdos_dir.c
b/cpukit/libfs/src/dosfs/msdos_dir.c
index 2
Free fs_info in case root_node is NULL.
---
cpukit/libfs/src/imfs/imfs_initsupp.c |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/cpukit/libfs/src/imfs/imfs_initsupp.c
b/cpukit/libfs/src/imfs/imfs_initsupp.c
index 1593076..7fe1e16 100644
--- a/cpukit/libfs/src/imfs/imfs_i
You'd have to ask Chris. I suspect it never did actually work exactly right.
-Gedare
On Thu, Sep 5, 2013 at 2:44 PM, Joel Sherrill wrote:
> How did this ever work?
>
>
> On 9/5/2013 1:29 PM, Gedare Bloom wrote:
>>
>> The arguments to rtems_nvdisk_calc_crc16()
On Thu, Sep 5, 2013 at 1:31 PM, Joel Sherrill wrote:
> Gedare.. are these from Coverity Scan? They seem
> remarkably picky otherwise. :)
>
Yes. I have some time to kill while I'm running some simulations.
> If they are, ask when you want another run submitted.
>
Will do. Th
If opts is NULL, then strchr() will dereference it and search for 's'.
1063863 Dereference after null check
---
cpukit/libfs/src/nfsclient/src/dirutils.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/cpukit/libfs/src/nfsclient/src/dirutils.c
b/cpukit/libfs/src/nfsclie
Upcast multiplication to 64-bit integer arithmetic.
---
cpukit/libmisc/uuid/gen_uuid.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/cpukit/libmisc/uuid/gen_uuid.c b/cpukit/libmisc/uuid/gen_uuid.c
index 0599637..4d10ff3 100644
--- a/cpukit/libmisc/uuid/gen_uuid.c
+++ b/
Check the return value from mkdir instead of failing silently.
1063857 Unchecked return value from library
---
cpukit/libmisc/untar/untar.c |6 +-
1 files changed, 5 insertions(+), 1 deletions(-)
diff --git a/cpukit/libmisc/untar/untar.c b/cpukit/libmisc/untar/untar.c
index 9ed00e9..5f16
Also eliminates possibly dead code in case the name_type can never
actually be MSDOS_NAME_INVALID.
1063860 Logically dead code.
---
cpukit/libfs/src/dosfs/msdos_misc.c |2 --
1 files changed, 0 insertions(+), 2 deletions(-)
diff --git a/cpukit/libfs/src/dosfs/msdos_misc.c
b/cpukit/libfs/src
Replace strcpy and strcat with counted variants.
---
cpukit/libfs/src/imfs/imfs_load_tar.c |8 +---
1 files changed, 5 insertions(+), 3 deletions(-)
diff --git a/cpukit/libfs/src/imfs/imfs_load_tar.c
b/cpukit/libfs/src/imfs/imfs_load_tar.c
index abb9f46..fd47ec2 100644
--- a/cpukit/libfs
OK. Can you send the patch against the current master head (rather
than revert)? I am not able to commit today, but maybe someone else
can.
-Gedare
On Fri, Sep 6, 2013 at 12:46 PM, Ralf Kirchner
wrote:
> Hi Gedare,
> Iconv() does not return any negative values. It returns "(size_t)-1
?
-Gedare
On Sep 8, 2013 5:59 AM, "Vipul Nayyar" wrote:
> Hello Gedare,
>
> I reviewed the current documentation for bsp development as you suggested. I
> didn't find any major discrepancies present already, but will look again
> closely. As far I think, the part
wrong.
>
> _BSP... Print exception frame has to be named incorrectly.
>
> If we ignore the cache and isr CPU dependent helpers, I would worry about
bsp reset and getworkarea. The latter almost certainly is a false positive.
The first could be some false positives.
>
> But as I mentioned
estions to this mailing list.
Cheers,
Gedare
___
rtems-devel mailing list
rtems-devel@rtems.org
http://www.rtems.org/mailman/listinfo/rtems-devel
Hi,
Is this condition even possible?
Also, would it make more sense to just assign task_load to 0, or maybe
not scale it (just set it to load[i])?
-Gedare
On Tue, Sep 10, 2013 at 1:36 AM, Rempel, Cynthia
wrote:
> Hi,
>
> Attached is a patch to check for dividing by zero in capt
1 - 100 of 602 matches
Mail list logo
