PR1395/bsps
* Updated the changes from RTEMS-4.8.0, which were made since Oct. 2007.
* network/if_1GHz/if_wm.c: fixed some bugs in the 1GHz driver.
* pci/pci_interface.c:
+ Enabled PCI "Read", "Read Line", and "Read Multiple"
+ Agressive Prefetch to improve the performance of the PCI based
applications (e.g. 1GHz NIC).
* irq/BSP_irq.c : Replaced the irq/irq.c, and used GT_GPP_Value
register to monitor the cause of the level sensitive interrupts.
This unique solution solves various bugs in the 1GHz network drivers
Fixed bugs in compute_pic_masks_from_prio()
* pci/pci.c : Updated it to be consistent with the original pci.c
* written by Eric Valette. There is no change in its function.
* irq/irq_init.c : set defaultIrq->next_handler to be 0
* for BSP_SHARED_HANDLER_SUPPORT.
PR1395/bsps
* Updated the changes from RTEMS-4.8.0, which were made since Oct. 2007.
* network/if_1GHz/if_wm.c: fixed some bugs in the 1GHz driver.
* pci/pci_interface.c:
+ Enabled PCI "Read", "Read Line", and "Read Multiple"
+ Agressive Prefetch to improve the performance of the PCI based
applications (e.g. 1GHz NIC).
* irq/BSP_irq.c : Replaced the irq/irq.c, and used GT_GPP_Value
register to monitor the cause of the level sensitive interrupts.
This unique solution solves various bugs in the 1GHz network drivers
Fixed bugs in compute_pic_masks_from_prio()
* pci/pci.c : Updated it to be consistent with the original pci.c
* written by Eric Valette. There is no change in its function.
* irq/irq_init.c : set defaultIrq->next_handler to be 0
* for BSP_SHARED_HANDLER_SUPPORT.
PR 1385/cpukit
* mpc5xx/irq/irq_asm.S: When the type rtems_boolean was switched to the
C99 bool, the size changed from 4 bytes to 1 byte. The interrupt
dispatching code accesses two boolean variables for scheduling
purposes and the assembly implementations of this code did not get
updated.
PR 1385/cpukit
* shared/irq/irq_asm.S: When the type rtems_boolean was switched to the
C99 bool, the size changed from 4 bytes to 1 byte. The interrupt
dispatching code accesses two boolean variables for scheduling
purposes and the assembly implementations of this code did not get
updated.
PR 1385/cpukit
* irq/irq_asm.S: When the type rtems_boolean was switched to the C99
bool, the size changed from 4 bytes to 1 byte. The interrupt
dispatching code accesses two boolean variables for scheduling
purposes and the assembly implementations of this code did not get
updated.
PR 1385/cpukit
* startup/exception.S: When the type rtems_boolean was switched to the
C99 bool, the size changed from 4 bytes to 1 byte. The interrupt
dispatching code accesses two boolean variables for scheduling
purposes and the assembly implementations of this code did not get
updated.
* startup/bspstart.c: removed legacy code (inherited
from old mvme2307 BSP) -- for testing trapping into
PPCBug -- which is irrelevant on this BSP.
Removed warning about SPRG0 having been reassigned
-- this BSP is OK.
PR 1354/bsps
* network/network.c: In some cases it can occur that an empty mbuf is
put on the descriptor chain. (We found it especially then when
transmitting fragmented IP Packets.) Since the actual buffer
descriptor pointer will be incremented after every inserted mbuf
(txBd = sc->txBdBase + sc->txBdHead;) even if m->m_len of the current
mbuf was zero. This leads to the bug.
* psim-bottom, psim-gdb-bottom, psim-shared, psim-top.in: Add ability
to force psim configuration to include System V IPC devices even when
the test does not appear to be multiprocessing. This is used by some
RTEMS application developers for system simulation and could be used
to simulate a framebuffer if a UNIX process mirrored the buffer.
Michael South <msouth@msouth.org>
PR 1344/bsps
* cpuModel.h: Add use of TSC for nanoseconds granularity. i8254 is very
slow on some systems. TSC use is auto-detected by default.
Michael South <msouth@msouth.org>
PR 1344/bsps
* configure.ac, clock/ckinit.c, timer/timer.c: Add use of TSC for
nanoseconds granularity. i8254 is very slow on some systems. TSC use
is auto-detected by default.
