* rtems/score/cpu_asm.h: Enhanced to include register offsets.
* Makefile.am: Install rtems/score/cpu_asm.h.
* cpu.c: Significantly enhanced including the implementation of
_CPU_ISR_Get_level.
* cpu_asm.S: Improved behavior of context switch and interrupt
dispatching.
* rtems/score/arm.h: Improved the CPU model name determination.
* rtems/score/cpu.h: Improved interrupt disable/enable functions.
* New driver for the Intel EtherExpressPro (82559ER) chip.
* network/if_fxp.c, network/if_fxpreg.h, network/if_fxpvar.h,
network/pci.h: New file.
* network/Makefile.am: Modified to reflect above.
* console/console.c: Others on the rtems-users list have expressed
concern about this run-time selection, so I've enclosed the changes
in a #ifdef RTEMS_RUNTIME_CONSOLE_SELECT conditional.
* comm/uart.c: I am using a PC-104 card with no video output. I
found that things would lock up if a printk was followed closely by
a printf when BSPConsolePort = BSP_UART_COM2 and
BSPPrintkPort = BSP_UART_COM1. With this change in place,
printf/printk calls can be intermingled with no apparent problems.
* macros/rtems/score/object.inl: Corrected implementation of
_Objects_Open, _Objects_Close, and _Objects_Namespace_remove
to be consistent with the inline implementation.
* src/dpmemident.c, src/msgqident.c, src/partident.c,
src/ratemonident.c, src/regionident.c, src/semident.c,
src/timerident.c: Corrected use of name parameter to
be consistent with fixes implemented to string name processing
which were committed with the POSIX Message Queue update.
* include/rtems/posix/cancel.h, src/cancel.c, src/cancelrun.c,
src/mqueue.c, src/pthread.c, src/semaphore.c, src/setcancelstate.c,
src/setcanceltype.c, src/testcancel.c: Per PR164, corrected the
behavior of thread cancellation and did some cleanup as a side-effect.
* psxcancel/Makefile.am, psxcancel/init.c, psxcancel/psxcancel.scn:
Updated as part of PR164 which reported problems with the RTEMS
implementation of pthread_cancel.
* Jump table for single entry point removed.
* include/rtems/directives.h, src/entrytable.c: Removed.
* include/Makefile.am, src/Makefile.am, src/exinit.c: Modified to
reflect above.
* Mega patch merge to change the format of the object IDs to
loosen the dependency between the SCORE and the various APIs.
There was considerable work to simplify the object name management
and it appears that the name_table field is no longer needed.
This patch also includes the addition of the internal mutex
which is currently only used to protect some types of allocation
and deallocation. This significantly can reduce context
switch latency under certain circumstances. In particular,
some heap/region operations were O(n) and had dispatching
disabled. This should help enormously. With this merge,
the patch is not as clean as it should be. In particular,
the documentation has not been modified to reflect the new object
ID layout, the IDs in the test screens are not updated, and
_Objects_Get_information needs to be a real routine not inlined.
As part of this patch a lot of MP code for thread/proxy blocking
was made conditional and cleaned up.
* src/bdbuf.c: Modified as part of above.
* Mega patch merge to change the format of the object IDs to
loosen the dependency between the SCORE and the various APIs.
There was considerable work to simplify the object name management
and it appears that the name_table field is no longer needed.
This patch also includes the addition of the internal mutex
which is currently only used to protect some types of allocation
and deallocation. This significantly can reduce context
switch latency under certain circumstances. In particular,
some heap/region operations were O(n) and had dispatching
disabled. This should help enormously. With this merge,
the patch is not as clean as it should be. In particular,
the documentation has not been modified to reflect the new object
ID layout, the IDs in the test screens are not updated, and
_Objects_Get_information needs to be a real routine not inlined.
As part of this patch a lot of MP code for thread/proxy blocking
was made conditional and cleaned up.
* src/cre_mbf.c, src/cre_sem.c, src/eventflags.c, src/fmempool.c,
src/itronsem.c, src/mbox.c, src/msgbuffer.c, src/port.c, src/task.c,
src/vmempool.c: Modified as part of above.
* Mega patch merge to change the format of the object IDs to
loosen the dependency between the SCORE and the various APIs.
There was considerable work to simplify the object name management
and it appears that the name_table field is no longer needed.
This patch also includes the addition of the internal mutex
which is currently only used to protect some types of allocation
and deallocation. This significantly can reduce context
switch latency under certain circumstances. In particular,
some heap/region operations were O(n) and had dispatching
disabled. This should help enormously. With this merge,
the patch is not as clean as it should be. In particular,
the documentation has not been modified to reflect the new object
ID layout, the IDs in the test screens are not updated, and
_Objects_Get_information needs to be a real routine not inlined.
As part of this patch a lot of MP code for thread/proxy blocking
was made conditional and cleaned up.
* include/rtems/posix/key.h, src/cond.c, src/condinit.c, src/intr.c,
src/key.c, src/keycreate.c, src/keydelete.c, src/killinfo.c,
src/mqueue.c, src/mqueuecreatesupp.c, src/mutex.c, src/mutexinit.c,
src/psignal.c, src/pthread.c, src/semaphore.c,
src/semaphorecreatesupp.c: Modified as part of above.
* Mega patch merge to change the format of the object IDs to
loosen the dependency between the SCORE and the various APIs.
There was considerable work to simplify the object name management
and it appears that the name_table field is no longer needed.
This patch also includes the addition of the internal mutex
which is currently only used to protect some types of allocation
and deallocation. This significantly can reduce context
switch latency under certain circumstances. In particular,
some heap/region operations were O(n) and had dispatching
disabled. This should help enormously. With this merge,
the patch is not as clean as it should be. In particular,
the documentation has not been modified to reflect the new object
ID layout, the IDs in the test screens are not updated, and
_Objects_Get_information needs to be a real routine not inlined.
As part of this patch a lot of MP code for thread/proxy blocking
was made conditional and cleaned up.
* include/rtems.h, inline/rtems/rtems/region.inl,
macros/rtems/rtems/region.inl, src/dpmem.c, src/dpmemcreate.c,
src/msg.c, src/msgqcreate.c, src/part.c, src/partcreate.c,
src/ratemon.c, src/ratemoncreate.c, src/region.c, src/regioncreate.c,
src/regiondelete.c, src/regionextend.c, src/regiongetinfo.c,
src/regiongetsegment.c, src/regiongetsegmentsize.c,
src/regionident.c, src/regionreturnsegment.c, src/rtemstimer.c,
src/sem.c, src/semcreate.c, src/taskcreate.c, src/taskident.c,
src/tasks.c, src/timercreate.c: Modified as part of above.
* Mega patch merge to change the format of the object IDs to
loosen the dependency between the SCORE and the various APIs.
There was considerable work to simplify the object name management
and it appears that the name_table field is no longer needed.
This patch also includes the addition of the internal mutex
which is currently only used to protect some types of allocation
and deallocation. This significantly can reduce context
switch latency under certain circumstances. In particular,
some heap/region operations were O(n) and had dispatching
disabled. This should help enormously. With this merge,
the patch is not as clean as it should be. In particular,
the documentation has not been modified to reflect the new object
ID layout, the IDs in the test screens are not updated, and
_Objects_Get_information needs to be a real routine not inlined.
As part of this patch a lot of MP code for thread/proxy blocking
was made conditional and cleaned up.
* include/confdefs.h, src/exinit.c, src/extension.c, src/itronapi.c,
src/posixapi.c, src/rtemsapi.c: Modified as part of above.
* Mega patch merge to change the format of the object IDs to
loosen the dependency between the SCORE and the various APIs.
There was considerable work to simplify the object name management
and it appears that the name_table field is no longer needed.
This patch also includes the addition of the internal mutex
which is currently only used to protect some types of allocation
and deallocation. This significantly can reduce context
switch latency under certain circumstances. In particular,
some heap/region operations were O(n) and had dispatching
disabled. This should help enormously. With this merge,
the patch is not as clean as it should be. In particular,
the documentation has not been modified to reflect the new object
ID layout, the IDs in the test screens are not updated, and
_Objects_Get_information needs to be a real routine not inlined.
As part of this patch a lot of MP code for thread/proxy blocking
was made conditional and cleaned up.
* include/Makefile.am, include/rtems/score/coremsg.h,
include/rtems/score/coremutex.h, include/rtems/score/coresem.h,
include/rtems/score/object.h, include/rtems/score/threadq.h,
inline/rtems/score/object.inl, inline/rtems/score/thread.inl,
macros/rtems/score/object.inl, src/Makefile.am, src/coremsg.c,
src/coremutex.c, src/coresem.c, src/mpci.c,
src/objectcomparenameraw.c, src/objectextendinformation.c,
src/objectinitializeinformation.c, src/objectnametoid.c,
src/thread.c, src/threadclose.c, src/threadget.c, src/threadq.c,
src/threadqextractwithproxy.c: Modified as part of above.
* include/rtems/score/apimutex.h, src/objectgetnoprotection.c: New
files.
* rtems/score/cpu.h: Fixed comments and renamed
CPU_SYSTEM_INITIALIZATION_THREAD_EXTRA_STACK to
CPU_MPCI_RECEIVE_SERVER_EXTRA_STACK to be consistent with other code.
* capture/capture-cli.c, cpuuse/cpuuse.c, monitor/mon-monitor.c,
monitor/mon-object.c, monitor/monitor.h: Corrected use of
_Objects_Information_table now that it is a two dimensional
array based upon API and class. In addition, in the monitor,
corrected an error which occured when a target has 64 bit pointers.
