dev/sinic.cc:
- Size the virtualRegs array based on the configured value
- Add debugging stuff for uniquely identifying vnic usage
- Only count totally unprocessed packets when notifying via RxDone
- Add initial virtual address support
- Fix some bugs in accessing packets out of order to make sure that
busy packets are processed first
- Add fifo watermark stuff
- Make number of vnics, zero/delay copy and watermarks parameters
dev/sinic.hh:
add rxUnique and txUnique to uniquely identify tx and rx VNICs
Create a separate list of Busy VNICs since more than one might
be busy and we want to service those first
Add more watermark stuff and new parameters
dev/sinicreg.hh:
Make the number of virtual nics a read-only parameter
add bits for ZeroCopy/DelayCopy
rename Virtual to Vaddr so it's not ambiguous
Add a flag for TxData/RxData to indicate a virtual address
Report rxfifo status in RxDone
python/m5/objects/Ethernet.py:
add more options for the fifo thresholds
add number of vnics as a parameter
add copy type as a parameter
add virtual addressing as a parameter
--HG--
extra : convert_revision : 850e2433b585d65469d4c5d85ad7ca820db10f4a
use Receive side scaling
dev/ns_gige.cc:
dev/ns_gige.hh:
dev/ns_gige_reg.h:
dev/sinic.cc:
dev/sinic.hh:
dev/sinicreg.hh:
add support for setting the RSS flag to notify the driver
to use RSS
--HG--
extra : convert_revision : 5f0c11668ae976634b3bf0caad669a9464a4c041
until the cache access occurs. The fundamental problem
is that a subsequent read that occurs functionally will
get a functionally incorrect result that can break
driver code.
dev/ns_gige.cc:
dev/ns_gige.hh:
dev/sinic.cc:
dev/sinic.hh:
get rid of pio_delay write and the associated code to move
the write to the cache access function
dev/sinicreg.hh:
no more write delays
python/m5/objects/Ethernet.py:
get rid of pio_delay write
--HG--
extra : convert_revision : 1dcb51b8f4514e717bc334a782dfdf06d29ae69d
arch/alpha/alpha_linux_process.cc:
Added using directive for AlphaISA namespace
arch/alpha/alpha_memory.hh:
arch/alpha/isa/branch.isa:
cpu/pc_event.hh:
Added typedefs for Addr
arch/alpha/alpha_tru64_process.cc:
arch/alpha/arguments.cc:
Added using directive for AlphaISA
arch/alpha/ev5.hh:
Added an include of arch/alpha/isa_traits.hh, and a using directive for the AlphaISA namespace.
arch/alpha/faults.hh:
Added a typedef for the Addr type, and changed the formatting of the faults slightly.
arch/alpha/isa/main.isa:
Untemplatized StaticInst, added a using for namespace AlphaISA to show up in decoder.cc and the exec.ccs, relocated makeNop to decoder.hh
arch/alpha/isa/mem.isa:
Untemplatized StaticInst and StaticInstPtr
arch/alpha/isa/pal.isa:
cpu/base_dyn_inst.cc:
Untemplatized StaticInstPtr
arch/alpha/isa_traits.hh:
Changed variables to be externs instead of static since they are part of a namespace and not a class.
arch/alpha/stacktrace.cc:
Untemplatized StaticInstPtr, and added a using directive for AlphaISA.
arch/alpha/stacktrace.hh:
Added some typedefs for Addr and MachInst, and untemplatized StaticInstPtr
arch/alpha/vtophys.cc:
Added a using directive for AlphaISA
arch/alpha/vtophys.hh:
Added the AlphaISA namespace specifier where needed
arch/isa_parser.py:
Changed the placement of the definition of the decodeInst function to be outside the namespaceInst namespace.
base/loader/object_file.hh:
cpu/o3/bpred_unit.hh:
Added a typedef for Addr
base/loader/symtab.hh:
Added a typedef for Addr, and added a TheISA to Addr in another typedef
base/remote_gdb.cc:
Added a using namespace TheISA, and untemplatized StaticInstPtr
base/remote_gdb.hh:
Added typedefs for Addr and MachInst
cpu/base.cc:
Added TheISA specifier to some variables exported from the isa.
cpu/base.hh:
Added a typedef for Addr, and TheISA to some variables from the ISA
cpu/base_dyn_inst.hh:
Untemplatized StaticInstPtr, and added TheISA specifier to some variables from the ISA.
cpu/exec_context.hh:
Added some typedefs for types from the isa, and added TheISA specifier to some variables from the isa
cpu/exetrace.hh:
Added typedefs for some types from the ISA, and untemplatized StaticInstPtr
cpu/memtest/memtest.cc:
cpu/o3/btb.cc:
dev/baddev.cc:
dev/ide_ctrl.cc:
dev/ide_disk.cc:
dev/isa_fake.cc:
dev/ns_gige.cc:
dev/pciconfigall.cc:
dev/platform.cc:
dev/sinic.cc:
dev/uart8250.cc:
kern/freebsd/freebsd_system.cc:
kern/linux/linux_system.cc:
kern/system_events.cc:
kern/tru64/dump_mbuf.cc:
kern/tru64/tru64_events.cc:
sim/process.cc:
sim/pseudo_inst.cc:
sim/system.cc:
Added using namespace TheISA
cpu/memtest/memtest.hh:
cpu/trace/opt_cpu.hh:
cpu/trace/reader/itx_reader.hh:
dev/ide_disk.hh:
dev/pcidev.hh:
dev/platform.hh:
dev/tsunami.hh:
sim/system.hh:
sim/vptr.hh:
Added typedef for Addr
cpu/o3/2bit_local_pred.hh:
Changed the include to use arch/isa_traits.hh instead of arch/alpha/isa_traits.hh. Added typedef for Addr
cpu/o3/alpha_cpu.hh:
Added typedefs for Addr and IntReg
cpu/o3/alpha_cpu_impl.hh:
Added this-> to setNextPC to fix a problem since it didn't depend on template parameters any more. Removed "typename" where it was no longer needed.
cpu/o3/alpha_dyn_inst.hh:
Cleaned up some typedefs, and untemplatized StaticInst
cpu/o3/alpha_dyn_inst_impl.hh:
untemplatized StaticInstPtr
cpu/o3/alpha_impl.hh:
Fixed up a typedef of MachInst
cpu/o3/bpred_unit_impl.hh:
Added a using TheISA::MachInst to a function
cpu/o3/btb.hh:
Changed an include from arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef for Addr
cpu/o3/commit.hh:
Removed a typedef of Impl::ISA as ISA, since TheISA takes care of this now.
cpu/o3/cpu.cc:
Cleaned up namespace issues
cpu/o3/cpu.hh:
Cleaned up namespace usage
cpu/o3/decode.hh:
Removed typedef of ISA, and changed it to TheISA
cpu/o3/fetch.hh:
Fized up typedefs, and changed ISA to TheISA
cpu/o3/free_list.hh:
Changed include of arch/alpha/isa_traits.hh to arch/isa_traits.hh
cpu/o3/iew.hh:
Removed typedef of ISA
cpu/o3/iew_impl.hh:
Added TheISA namespace specifier to MachInst
cpu/o3/ras.hh:
Changed include from arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef for Addr.
cpu/o3/regfile.hh:
Changed ISA to TheISA, and added some typedefs for Addr, IntReg, FloatReg, and MiscRegFile
cpu/o3/rename.hh:
Changed ISA to TheISA, and added a typedef for RegIndex
cpu/o3/rename_map.hh:
Added an include for arch/isa_traits.hh, and a typedef for RegIndex
cpu/o3/rob.hh:
Added a typedef for RegIndex
cpu/o3/store_set.hh:
cpu/o3/tournament_pred.hh:
Changed an include of arch/alpha/isa_traits.hh to arch/isa_traits.hh, and added a typedef of Addr
cpu/ozone/cpu.hh:
Changed ISA into TheISA, and untemplatized StaticInst
cpu/pc_event.cc:
Added namespace specifier TheISA to Addr types
cpu/profile.hh:
kern/kernel_stats.hh:
Added typedef for Addr, and untemplatized StaticInstPtr
cpu/simple/cpu.cc:
Changed using directive from LittleEndianGuest to AlphaISA, which will contain both namespaces. Added TheISA where needed, and untemplatized StaticInst
cpu/simple/cpu.hh:
Added a typedef for MachInst, and untemplatized StaticInst
cpu/static_inst.cc:
Untemplatized StaticInst
cpu/static_inst.hh:
Untemplatized StaticInst by using the TheISA namespace
dev/alpha_console.cc:
Added using namespace AlphaISA
dev/simple_disk.hh:
Added typedef for Addr and fixed up some formatting
dev/sinicreg.hh:
Added TheISA namespace specifier where needed
dev/tsunami.cc:
dev/tsunami_io.cc:
dev/tsunami_pchip.cc:
Added using namespace TheISA. It might be better for it to be AlphaISA
dev/tsunami_cchip.cc:
Added typedef for TheISA. It might be better for it to be AlphaISA
kern/linux/aligned.hh:
sim/pseudo_inst.hh:
Added TheISA namespace specifier to Addr
kern/linux/linux_threadinfo.hh:
Added typedef for Addr, and TheISA namespace specifier to StackPointerReg
kern/tru64/mbuf.hh:
Added TheISA to Addr type in structs
sim/process.hh:
Added typedefs of Addr, RegFile, and MachInst
sim/syscall_emul.cc:
Added using namespace TheISA, and a cast of VMPageSize to the int type
sim/syscall_emul.hh:
Added typecast for Addr, and TheISA namespace specifier for where needed
--HG--
extra : convert_revision : 91d4f6ca33a73b21c1f1771d74bfdea3b80eff45
separate the rx thread and tx thread and get rid of the dedicated flag.
dev/ns_gige.cc:
dev/ns_gige.hh:
dev/ns_gige_reg.h:
python/m5/objects/Ethernet.py:
dedicated flag goes away, we have new individual flags for
rx thread and tx thread
dev/sinic.cc:
Virtualize sinic
- The io registers are replicated many times in memory, allowing the NIC to
differentiate among several virtual interfaces.
- On the TX side, this allows multiple CPUs to initiate transmits at the same
time without locking in the software. If a partial packet is transmitted,
then the state machine blocks waiting for that virtual interface to complete
its packet. Then the state machine will move on to the next virtual
interface. The commands are kept in fifo order.
- On the RX side, multiple partial transmits can be simultaneously done.
Though a packet does not deallocate its fifo space until all preceeding
packets in the fifo are deallocated. To enable multiple receives, it
is necessary for each virtual nic to keep its own information about its
progress through the state machine.
dev/sinic.hh:
Virtualize sinic
Receive state must be virtualized since we allow the receipt of packets in
parallel.
dev/sinicreg.hh:
Virtualize sinic
separate rx thread and tx thread
create a soft interrupt and add a command to trigger it.
pad out the reserved bits in the RxDone and TxDone regs
--HG--
extra : convert_revision : c10bb23a46a89ffd1e08866c1f1621cb98069205
dev/ns_gige.cc:
add support for delaying pio writes until the cache access occurs
the only write we delay are for CR_TXE and CR_RXE
dev/sinic.cc:
dev/sinic.hh:
the txPioRequest and rxPioRequest things were more or less bogus
add support for delaying pio writes until the cache access occurs
dev/sinicreg.hh:
Add delay_read and delay_write to the register information struct
for now, we won't delay any reads, and we'll delay the writes that
initiate DMAs
python/m5/objects/Ethernet.py:
add a parameter to delay pio writes until the timing access
actually occurs.
--HG--
extra : convert_revision : 79b18ea2812c2935d7d5ea6eff1f55265114d05d
interrupts.
dev/sinic.cc:
- The prepareRead function sets all the variables in the register
file that depend on various state bits that change on the fly.
Includes RxDone, RxWait, TxDone, and TxWait
- Use the new register information accessor functions to grab
validity and size information for the read and write functions
- read all registers directly from the register space by offset
and size, not by actual name (less code)
- The side effect of reading the interrupt status (clearing it) now
happens outside the actual chunk of code where the value is loaded.
- Add an iprRead function for when we may want speculative access
to device registers through an ipr or special instruction.
- When RxData or TxData are written, their busy flag is set to
indicate that they have an outstanding transaction.
- The RxHigh and TxLow interrupts are special, they only interrupt
if the rxEmpty or txFull limits were hit
- Move reset to the command register
- Update more registers on reset, clear rxEmpty and txFull
- Data dumps only happen if EthernetData trace flag set
- When a DMA completes, kick the other engine if it was waiting
- implement all of the new interrupts
- serialize the new stuff
dev/sinic.hh:
- Put all registers with their proper size and alignment into
the regs struct so that we can copy multiple at a time.
- Provide accessor functions for accessing the registers with
different sizes.
- Flags to track when the rx fifo hit empty and the tx fifo became
full. These flags are used to determine what to do when below
the watermarks, and are reset when crossing the watermark.
- the txDmaEvent should actually trigger the txDmaDone function
- Add an iprRead function for when we may want speculative access
to device registers through an ipr or special instruction.
- The prepareRead function sets all the variables in the register
file that depend on various state bits that change on the fly.
- add rx_max_intr and dedicated (for dedicated thread) config params
dev/sinicreg.hh:
Add some new registers: Command, RxMaxIntr, RxFifoSize, TxFifoSize,
rename XxThreshold to XxFifoMark
Move Reset to the Command register
Add Thread to the Config register
New interrupts, better names
More info in RxDone and TxDone
Easier access to information on each register (size, read, write, name)
python/m5/objects/Ethernet.py:
Both sinic and nsgige have the dedicated thread
Add a parameter to configure the maximum number for receive
packets per interrupt
--HG--
extra : convert_revision : 407c5a993b6fb17326b4c623ee5d4b25fd69ac80
Nathan Binkert
Gabe Black
Steve Reinhardt