WO1998045784A1 - Microprocessor-based device incorporating a cache for capturing software performance profiling data - Google Patents
Microprocessor-based device incorporating a cache for capturing software performance profiling data Download PDFInfo
- Publication number
- WO1998045784A1 WO1998045784A1 PCT/US1998/006838 US9806838W WO9845784A1 WO 1998045784 A1 WO1998045784 A1 WO 1998045784A1 US 9806838 W US9806838 W US 9806838W WO 9845784 A1 WO9845784 A1 WO 9845784A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- processor
- trace
- counter
- trace cache
- based device
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/362—Software debugging
- G06F11/3636—Software debugging by tracing the execution of the program
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/3604—Software analysis for verifying properties of programs
- G06F11/3612—Software analysis for verifying properties of programs by runtime analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/22—Detection or location of defective computer hardware by testing during standby operation or during idle time, e.g. start-up testing
- G06F11/26—Functional testing
- G06F11/261—Functional testing by simulating additional hardware, e.g. fault simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
- G06F11/3419—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment by assessing time
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3466—Performance evaluation by tracing or monitoring
- G06F11/348—Circuit details, i.e. tracer hardware
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/36—Preventing errors by testing or debugging software
- G06F11/362—Software debugging
- G06F11/3648—Software debugging using additional hardware
- G06F11/3656—Software debugging using additional hardware using a specific debug interface
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3466—Performance evaluation by tracing or monitoring
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2201/00—Indexing scheme relating to error detection, to error correction, and to monitoring
- G06F2201/86—Event-based monitoring
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2201/00—Indexing scheme relating to error detection, to error correction, and to monitoring
- G06F2201/88—Monitoring involving counting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2201/00—Indexing scheme relating to error detection, to error correction, and to monitoring
- G06F2201/885—Monitoring specific for caches
Definitions
- the invention relates to software performance profiling support in microprocessors, and more particularly to a microprocessor-based device incorporating an on-chip trace cache capable of capturing software performance profile data.
- Software performance profiling refers to examining the execution times, frequencies and calling patterns of different software procedures within a software program. Performance profiling can be a very useful tool to a software engineer attempting to optimize the execution times of software applications.
- Various techniques for performing software profiling are currently used, including many based on statistical analysis. When performing software profiling, execution times and subroutine call linkage are sometimes captured by external (off-chip) instrumentation that monitors the system buses of the computer system which is executing the software. Alternatively, software can be "instrumented” or modified to provide profiling information directly to the computer system on which the software is executed.
- In-circuit emulators provide certain advantages over other debug environments, offering complete control and visibility over memory and register contents, as well as overlay and trace memory in case system memory is insufficient.
- use of traditional in-circuit emulators which involves interfacing a custom emulator back-end with a processor socket to allow communication between emulation equipment and the target system, is becoming increasingly difficult and expensive in today's age of exotic packages and shrinking product life cycles.
- Instrumented code is often generated by a compiler configured to insert profiling information in ol der to analyze selected procedures. For example, on procedure call prologues and exit epilogues, the compiler may insert code used to activate counters that track execution times. As a specified program run call is executed, a lump to an inserted utine is performed to mark a counter/timer.
- the execution time of a parent procedure that calls other, ancillary procedures can be determined by subtracting the execution t ⁇ me(s) of the ancillary procedures from the total execution time of the parent procedure. By analyzing all of the procedures of a module, the total execution time of the module can be calculated. Of course, the execution time of a given procedure may vary depending on the state of variables within the procedure, requiring statistical sampling to be utilized.
- a processor-based device includes an on-chip trace cache and supporting circuitry for providing software performance profiling information.
- the trace cache gathers information concerning the execution time spent in selected procedures. Performance profiling information is thereby gathered without instrumenting code, negatively impacting program execution speeds, or using expensive off-chip support equipment
- a breakpoint or trigger control register is configured to initialize and trigger (start) a first on-chip counter upon entry into a selected procedure.
- a second breakpoint or trigger control register is used to stop the first counter when the procedure prologue of the selected procedure is entered
- Counter values reflecting the lapsed execution time of the selected procedure are then stored in the on- chip trace cache. Similar techniques can be used to measure other parameters such as interrupt handler execution times.
- a second counter is also provided.
- the second counter runs continually, but is reset to zero following a stop trigger event caused by the second trigger control register.
- the stop trigger event also causes the value of the second counter to be placed in the on-chip trace cache. This second counter value is useful for obtaining the frequency of occurrence of a procedure of interest, whereas the first counter provides information about the procedure's execution time.
- the ptofiie data can be analyzed by post-processing software resident in the computer system in which the selected pi ocedures are executed, by a host system utilizing a debug port, or via off-chip trace capture hardware Generally, only one procedure is profiled at a time. By examining the trace cache, the minimum. i average, and maximum tunes spent in a procedure, as well as other statistical data, can be determined.
- One beneficial aspect of the invention is that the procedure prologue and epilogue are not required to be modified. However, a compiler can still be utilized to add profiling information for use with the present invention.
- Both serial and parallel communication channels are provided for communicating the trace information to external devices.
- controllability and observability of the profile (or trace) cache are achieved through a software debug port that uses an IEEE-1 149.1-1990 compliant JTAG (Joint Test Action Group) interface or a similar standardized interface that is integrated into the processor-based device.
- JTAG Joint Test Action Group
- a processor-based device supplying a flexible, high-performance solution for furnishing software performance profiling information.
- the disclosed on-chip trace cache also alleviates various of the bandwidth and clock synchronization problems that arise in many existing solutions.
- f igure 1 is a block diagram of a software debug environment utilizing a software profiling and debug solution in accordance with the present invention
- Figure 2 is a block diagram providing details of an exemplary embedded processor product incorporating an on-chip trace cache according to the present invention
- Figure 3 is a simplified block diagram depicting the relationship between an exemplary trace cache and other components of an embedded processor product according to the present invention.
- Figure 4 is a flowchart illustrating software debug command passing according to one embodiment of the invention:
- Figure 5 is a flowchart illustrating enhanced command passing according to a second embodiment of the invention:
- Figure 6A illustrates performance profile counter sequences according to the present invention:
- Figure 6B illustrates the general format of a trace cache entry set for reporting software performance profiling information in accordance with the invention.
- Figures 7A - 7G illustrate the general format of a variety of optional trace cache entries for reporting instruction execution information.
- f igure 1 depicts an exemplary software debug environment illustrating a contemplated use of the present invention
- a target system T is shown containing an embedded processor device 102 accoiding to the present invention coupled to system memory 106.
- the embedded processor device 102 incorporates a processor core 104. a trace cache 200 ( Figure 2). and a debug port 100.
- the embedded processor device 102 may incorporate additional circuitry (not shown) for performing application specific functions, or may take the form of a stand-alone processor or digital signal processor.
- the debug port 100 uses an IEEE- 1 149.1 - 1990 compliant JTAG interface or other similar standardized serial port interface
- a host system H is used to execute debug control software 1 12 for transferring high-level commands and controlling the extraction and analysis of software performance profiling information generated by the target system T.
- the host system H and target system T of the disclosed embodiment of the invention communicate via a serial link 1 10.
- Most computers are equipped with a serial or parallel interface which can be inexpensively connected to the debug port 100 by means of a serial connector 108. allowing a variety of computers to function as a host system H.
- the serial connector 108 could be replaced with higher speed JTAG-to-network conversion equipment.
- the target system T can be configured to internally analyze software performance profile data.
- FIG. 2 depicts various elements of an enhanced embodiment of the debug port 100 capable of utilizing and controlling Trace cache 200. Many other configurations are possible, as will become apparent to those skilled in the art. and the various processor device 102 components described below arc shown for purposes of illustrating the benefits associated with providing the on- chip trace cache 200.
- the Trace control circuitry 218 and the trace cache 200 of the disclosed embodiment of the invention can also cooperate to capture software performance profiling information.
- the trace control circuitry 218 supports "tracing" to a trace pad interface port 220 or to the trace cache 200 and provides user control for selectively activating capture of software performance profiling data. Other features enabled by the trace control circuitry 218 include programmability of synchronization address generation and user specified trace records, as discussed in greater detail below.
- the JTAG pins essentially become a transportation mechanism, using existing pins, to enter profiling and other commands to be performed by the processor core 104. More specifically, the test clock signal TCK. the test mode select signal TMS. the test data input signal TDI and the test data output signal TDO provided to and driven by a JTAG Test Access Port (TAP) controller 204 are conventional JTAG support signals and known to those skilled in the ait. As discussed in more detail below, an "enhanced" embodiment of the debug port 100 adds the command acknowledge signal CMDACK.. the break request/trace capture signal BRTC. the stop transmit signal STOPTX.
- TCP JTAG Test Access Port
- T hese "sideband" signals offer extra functionality and improve communications speeds for the debug port 100
- T hese signals also aid in the operation of an optional parallel port 214 provided on special bond- out versions of the disclosed embedded processor device 102.
- the JTAG TAP controller 204 accepts standard JTAG serial data and control via the conventional JTAG signals.
- a serial debug shifter 212 is connected to the JTAG test data input signal TDI and test data output signal TDO. such that commands and data can then be loaded into and read from debug registers 210.
- the debug registers 210 include two debug registers for transmitting (TX DATA register) and receiving (RX DATA tegister) data, an instruction trace configuration register (ITCR).
- a control interface state machine 206 coordinates the loading/reading of data to/from the serial debug shifter 212 and the debug registers 210.
- a command decode and processing block 208 decodes commands/data and dispatches them to processor interface logic 202 and trace debug interface logic 216.
- the trace debug interface logic 216 and trace control logic 218 coordinate the communication of software performance profiling and other trace information from the trace cache 200 to the TAP controller 204.
- the processor interface logic 202 communicates directly with the processor core 104. as well as the trace control logic 218.
- parallel port logic 214 communicates with a control interface state machine 206 and the debug registers 210 to perform parallel data read/write operations in optional bond-out versions of the embedded processor device 102.
- the port 100 is enabled by writing the public JTAG instruction DEBUG into a JTAG instruction register contained within the TAP controller 204.
- the JTAG instruction register of the disclosed embodiment is a 38-bit register comprising a 32-bit data field (debug data[31 :0]), a four-bit command field to point to various internal registers and functions provided by the debug port 100. a command pending flag. and a command finished flag. It is possible for some commands to use bits from the debug data field as a sub-field to extend the number of available commands. 37 5 1 0
- This JTAG instruction register is selected by toggling the test mode select signal TMS.
- the test mode select signal TMS allows the JTAG path of clocking to be changed in the scan path, enabling multiple paths of varying lengths to be used.
- the JTAG instruction register is accessible via a short path.
- This register is configured to include a "soft" register for holding values to be loaded into or received from specified system registers.
- Figure 3 a simplified block diagram depicting the relationship between the exemplary trace cache 200 and other components of the embedded processor device 102 according to the present invention is shown.
- the trace cache 200 is a 128 entry first-in, first-out (FIFO) circular cache. Increasing the size of the trace cache 200 increases the amount of software performance profile and other instruction trace information that can be captured, although the amount of required silicon area may increase.
- the trace cache 200 of the disclosed embodiment of the invention stores a plurality of 20-b ⁇ t (or more) trace entries, such as software performance profiling and other trace information. Additional information, such as task identifiers and trace capture stop/start information, can also be placed in the trace cache 200.
- the contents of the trace cache 200 are provided to external hardware, such as the host system H, via either serial or parallel trace pins 230.
- the target system T can be configured to examine the contents of the trace cache 200 internally.
- Figure 4 provides a high-level flow chart of command passing when using a standard JTAG interface.
- the DEBUG instruction is written to the TAP controller 204 in step 402.
- step 404 the 38-b ⁇ t serial value is shifted in as a whole, with the command pending flag set and desired data (if applicable, otherwise zero) in the data field.
- Control proceeds to step 406 where the pending command is loaded/unloaded and the command finished flag checked.
- Completion of a command typically involves transferring a value between a data register and a processor register or memory/iO location.
- the processor 104 clears the command pending flag and sets the command finished flag, at the same time storing a value in the data field if applicable.
- the entire 38-b ⁇ t register is scanned to monitor the command finished and command pending flags. If the pending flag is reset to zero and the finished flag is set to one. the previous command has finished.
- the status of the flags is captured by the control interface state machine 206 A slave copy of the flags status is saved internally to determine if the next instruction should be loaded. The slave copy is maintained due to the possibility of a change in flag status between TAP controller 204 states. This allows the processor 104 to determine if the previous instruction has finished before loading the next instruction.
- step 410 If the finished flag is not set as determined in step 408, control proceeds to step 410 and the loading/unloading of the 38-b ⁇ t command is repeated. The command finished flag is also checked. Control then returns to step 408 If the finished flag is set as determined in step 408. control returns to step 406 for processing of the next command. DEBUG mode is exited via a typical JTAG process.
- the optional sideband signals are utilized in the enhanced debug port 100 to provide extra functionality.
- the optional sideband signals include a break request/trace capture signal BRTC that can function as a break request signal or a trace capture enable signal depending on the status of a bit set in the debug control/status register. If the break request trace capture signal BRTC is set to function as a break request signal, it is asserted to cause the processor 104 to enter debug mode (the processor 104 can also be stopped by scanning in a halt command via the convention J TAG signals). If set to function as a trace capture enable signal, asserting the break request/trace capture signal BRTC enables capturing of trace information. Deasserting the signal turns trace capture off. The signal takes effect on the next instruction boundary after it is detected and is synchronized with the internal processor clock.
- the break request/trace capture signal BRTC may be asserted at any tune.
- the trigger signal TRIG is configured to pulse whenever an internal processor breakpoint has been asserted.
- the trigger signal TRIG may be used to trigger an external capturing device such as a logic analyzer, and is synchronized with the trace record capture clock signal TRACECLK.
- an external capturing device such as a logic analyzer
- TRACECLK trace record capture clock signal
- the stop transmit signal STOPTX is asserted when the processor 104 has entered DEBUG mode and is ready tor register interrogation/modification, memory or I/O reads and writes through the debug port 100.
- the stop transmit signal STOPTX reflects the state of a bit in the debug control status register (DCSR).
- the stop transmit signal STOPTX is synchronous with the trace capture clock signal TRACECLK.
- the command acknowledge signal CMDACK is described in conjunction with Figure 5. which shows simplified command passing in the enhanced debug port 100 of Figure 2.
- a DEBUG instruction is written to the TAP controller 204 in step 502.
- Control proceeds to step 504 and the command acknowledge signal CMDACK is monitored by the host system H to determine command completion status. This signal is asserted high by the target system f simultaneously with the command finished flag and remains high until the next shift cycle begins.
- the command acknowledge signal CMDACK it is not necessary to shift out the JTAG instruction register to capture the command finished flag status.
- the command acknowledge signal CMDACK transitions high on the next rising edge of the test clock signal TCK after the command finished flag has changed from zero to one.
- a new shift sequence (step 506) is not started by the host system H until the command acknowledge signal CMDACK pin has been asserted high.
- the command acknowledge signal CMDACK is synchronous with the test clock signal TCK.
- the test clock signal TCK need not be clocked at all times, but is ideally clocked continuously when waiting for a command acknowledge signal CMDACK response.
- debug register block 210 Also included in debug register block 210 is an instruction trace configuration register (ITCR). This 32- bit register provides for the enabling/disabling and configuration of software performance profile and instruction trace debug functions. Numerous such functions are contemplated, including various levels of tracing, trace synchronization force counts, trace initialization, instruction tracing modes, clock divider ratio information, as well as additional functions shown in the following table.
- the ITCR is accessed through a JTAG instruction register write/read command as is the case with the other registers of the debug register block 210, or via a reserved instruction
- ICR Instruction Trace Configuration Register
- DCSR debug control/status register
- ICR Instruction Trace Configuration Register
- Another debug register the debug control/status register (DCSR). provides an indication of when the processor 104 has entered debug mode and allows the processor 104 to be forced into DEBUG mode through the enhanced JTAG interface.
- the DCSR also enables miscellaneous control features, such as. forcing a ready signal to the processor 104. controlling memory access space for accesses initiated through the debug port, disabling cache flush on entry to the DEBUG mode, the TX and RX bits, the parallel port 214 enable, forced breaks, forced global reset, and other functions.
- the ordering or presence of the various bits in either the ITCR oi DCSR is not considered critical to the operation of the invention.
- DCSR Debug Control/Status Register
- This data may consist, for example, of a character stream from a p ⁇ ntfO call or icgister information from a Task's Control Block (TCB).
- TBC Task's Control Block
- One contemplated method foi transferring the data is for the operating system to place the data in a known region, then via a trap instruction cause DEBUG mode to be entered.
- the host system H can then determine the reason that DEBUG mode was entered, and respond by retrieving the data from the reserved region. However, while the processor 104 is in DEBUG mode, normal processor execution is stopped. As noted above, this is undesirable for many real-time systems
- This situation is addressed according to the present invention by providing two debug registers in the debug port 100 for transmitting (TX DATA register) and receiving (RX DATA register) data. These registers can be accessed using the soft address and JTAG instruction register commands. As noted, after the host system H has written a debug instruction to the JTAG instruction register, the serial debug shifter 212 is coupled to the test data input signal TDI line and test data output signal TDO line.
- the piocessor 104 executes code causing it to transmit data, it first tests a TX bit in the ITCR. If the
- TX bit is set to zero then the processor 104 executes a processor instruction (either a memory or I/O write) to transfer the data to the TX DATA register.
- the debug port 100 sets the TX bit in the DCSR and ITCR. indicating to the host system H that it is ready to transmit data. Also, the STOPTX pin is set high. After the host system H completes reading the transmit data from the TX DATA register, the TX bit is set to zero.
- ITCR is then set to generate a signal to interrupt the processor 104.
- the interrupt is generated only when the TX bit in the ITCR transitions to zero.
- the processor 104 polls the ITCR to determine the status of the TX bit to further transmit data.
- the host system H desires to send data, it first tests a RX bit in the ITCR. If the RX bit is set to zero, the host system H writes the data to the RX DATA register and the RX bit is set to one in both the DCSR and
- a RXINT bit is then set in the ITCR to generate a signal to interrupt the processor 104. This interrupt is only generated when the RX in the ITCR transitions to one.
- the processor 104 polls the ITCR to verify the status of the RX bit. If the RX bit is set to one. the processor instruction is executed to lead data from the RX DATA register. After the data is read by the processor 104 from the RX DATA register the
- RX bit is set to zero.
- the host system H continuously reads the ITCR to determine the status of the RX bit to further send data.
- This technique enables an operating system or application to communicate with the host system H without stopping processor 104 execution. Communication is conveniently achieved via the debug port 100 with minimal impact to on-chip application resources. In some cases it is necessary to disable system interrupts. This requires that the RX and TX bits be examined by the processor 100. In this situation, the communication link is driven in a polled mode.
- PARALLEL INTERFACE TO DEBUG PORT 100 Some embedded systems require instruction trace to be examined while maintaining I/O and data processing operations. Without the use of a multi-tasking operating system, a bond-out version of the embedded processor device 102 is preferable to provide the trace data, as examining the trace cache 200 via the debug port 100 requires the processor 104 to be stopped. In the disclosed embodiment of the invention, a parallel port 214 is also provided in an optional bond-out version of the embedded processor device 102 to provide parallel command and data access to the debug port 100. This interface provides a 16-b ⁇ t data path that is multiplexed with the trace pad interface port 220.
- the parallel port 214 provides a 16-b ⁇ t wide bi-directional data bus (PDATA[ 15:0]), a 3-bit address bus (PADR[2.0j). a parallel debug port read/write select signal (PRW). a trace valid signal TV and an instruction trace record output clock TRACECLOCK (TC). Although not shared with the trace pad interface port 220. a parallel bus lequest grant signal pair PBREQ/PBGNT (not shown) are also provided.
- the parallel port 214 is enabled by setting a bit in the DCSR. Serial communications via the debug port 100 are not disabled when the parallel port 214 is enabled. 22 21 20 19 16 0
- the parallel port 214 is primarily intended for fast downloads/uploads to and from target system T memory However, the parallel port 214 may be used for all debug communications with the target system T whenever the processor 104 is stopped.
- the serial debug signals (standard or enhanced) are used for debug access to the target system T when the processor 104 is executing instructions.
- the parallel port 214 shares pins with the trace pad interface 220. requiring parallel commands to be initiated only while the processor 104 is stopped and the trace pad interface 220 is disconnected from the shared bus.
- the parallel bus request signal PBREQ and parallel bus grant signal PBGNT are provided to expedite multiplexing of the shared bus signals between the trace cache 200 and the parallel port 214
- the host interface to the parallel port 214 determines that the parallel bus request signal PBREQ is asserted, it begins driving the parallel port 214 signals and asserts the parallel bus grant signal PBGNT
- the parallel port 214 When entering or leaving DEBUG mode with the parallel port 214 enabled, the parallel port 214 is used for the processor state save and restore cycles.
- the parallel bus request signal PBREQ is asserted immediately before the beginning of a save state sequence penultimate to entry of DEBUG mode.
- the parallel bus lequest signal PBREQ is deasserted after latching the write data.
- the parallel port 214 host interface responds to parallel bus request signal PBREQ deassertion by t ⁇ -stating its parallel port drivers and deasserting the parallel bus grant signal PBGNT
- the parallel port 214 then enables the debug trace port pin drivers, completes the last restore state cycle, asserts the command acknowledge signal CMDACK. and returns control of the interface to trace control logic 218
- the address pins PADR[2.0] are used for selection of the field of the JTAG instruction register, which is mapped to the 16-bit data bus PDATA[ 15.0] as shown in the following table
- the command pending flag is automatically set when performing a write operation to the four-bit command register, and is cleared when the command finished flag is asserted.
- the host system H can monitor the command acknowledge signal CMDACK to determine when the finished flag has been asserted.
- Use of the parallel port 214 provides full visibility of execution history, without requiring throttling of the processor core 104.
- the trace cache 200 if needed, can be configured for use as a buffer to the parallel port 214 to alleviate any bandwidth matching issues.
- the operation of all debug supporting features can be controlled through the debug port 100 or via processor instructions. These processor instructions may be from a monitor program, target hosted debugger, or conventional pod-wear.
- the debug port 100 performs data moves which are initiated by serial data port commands rather than processor instructions.
- Operation of the processor core 104 from conventional pod-space is very similar to operating in DEBUG mode from a monitor program. All debug operations can be controlled via processor instructions. It makes no difference whether these instructions come from pod-space or regular memory. This enables an operating system to be extended to include additional debug capabilities. Of course, via privileged system calls such a ptrace(). operating systems have long supported debuggers
- the command LITCR loads an indexed record in the trace cache 200. as specified by a trace cache pointer
- ITREC.PTR with the contents of the EAX register of the processor core 104.
- the trace cache pointer ITREC.PTR is pre-incremented. such that the general operation of the command LITCR is as follows. ITREC.PTR ⁇ - ITREC.PTR + 1 ;
- the store instruction trace cache record command SITCR is used to retrieve and store (in the EAX register) an indexed record from the trace cache 200.
- the contents of the ECX register of the processor core 104 are used as an offset that is added to the trace cache pointer ITREC.PTR to create an index into the trace cache 200.
- the ECX register is post-incremented while the trace cache pointer ITREC.PTR is unaffected, such that:
- Extending an operating system to support on-chip trace has certain advantages within the communications industry. It enables the system I/O and communication activity to be maintained while a task is being traced. I iaditionally. the use of an in-circuit emulator has necessitated that the processor be stopped before the processor's state and trace can be examined (unlike ptrace( )). This disrupts continuous support of I/O data processing.
- the trace cache 200 is very useful when used with equipment in the field, if an unexpected system crash occurs, the trace cache 200 can be examined to observe the execution history leading up to the crash event. When used in portable systems or other environments in which power consumption is a concern, the trace cache 200 can be disabled as necessary via power management circuitry.
- an instruction trace record is 20 bits wide and consists of two fields, TCODE ( Trace Code) and TDATA (Trace Data), as well as a valid bit V.
- TCODE Trace Code
- TDATA Track Data
- the TCODE field is a code that identifies the type of data in the TDATA field.
- the TDATA field contains software performance profile and other trace information used for debug purposes.
- the embedded processor device 102 reports performance profiling data as well as data corresponding to ten other trace codes as set forth in the following table:
- the trace cache 200 is of limited storage capacity; thus a certain amount of "compression" in captured trace data is desirable.
- the following discussion assumes that an image of the program being traced is available to the host system H. If an address can be obtained from a program image (Ob
- trace information that can be captured includes: the target address of a trap or interrupt handler: the target address of a return instruction: a conditional branch instruction having a target address which is data register dependent (otherwise, all that is needed is a 1 -bit trace indicating if the bianch was taken or not); and, most frequently, addresses from procedure returns.
- Other information, such as task identifiers and trace capture stop/start information, can also be placed in the trace cache 200. The precise contents and nature of the trace records are not considered critical to the invention. Referring now to Figure 6A. exemplary performance profile counter sequences are illustrated.
- trigger control registers 219 are configured to start and stop a counter that measures lapsed time of execution for specified procedures.
- the precise implementation of the trigger control registers 219 is not considered critical to the invention, use of conventional breakpoint registers (such as any of the debug registers DR0-DR7 present in some prior microprocessor cores) to perform the triggering functions is preferred. Further, it should be noted that in the disclosed embodiment of the invention normal instruction execution is not interrupted while profiling information is gathered.
- a first on-chip trigger control register 219a is configured to trigger (start) a first counter upon entry into a specified procedure.
- a second trigger control register 219b is used to stop the counter upon entry into the prologue of the specified procedure.
- the first counter is started by the start t ⁇ ggei . it is initialized to zero.
- the stop trigger is generated as specified by the second trigger control register 219b.
- the trigger control registers 219 can also be used to pulse the trigger signal TRIG and to select program addresses where execution trace is to start and stop.
- a second counter is also used.
- the second counter runs continually, but is reset to zero following a stop trigger event.
- the stop trigger event also causes the value of the second counter to be placed in the trace cache 200.
- This second counter value is useful for obtaining the frequency of occurrence of a procedure of interest, whereas the first counter provides information about the procedure's execution time.
- the counter frequency (i.e.. resolution) of the first and second counters is programmable. Such programmabiiity may allow for better accuracy when profiling very low frequency or high frequency events. For example, the following table depicts two alternate and programmable accumulation frequencies for the counters: s
- Post-processing software in conjunction with optional off-chip trace capture hardware, can be utilized to analyze the profile data.
- the trace cache 200 is utilized to gather information concerning the execution time 5 spent in a selected procedure. Generally, only one procedure is profiled at a time. By examining the trace cache 200. the minimum average and maximum time spent in a procedure can be determined (within the limitations of the samples gathered). Code coverage profiling capabilities can also be added to show specific addresses executed and not executed during test runs.
- the trace cache 200 allows statistical analysis to be performed using as many samples as can be stored.
- Figure 7A illustrates an exemplary format for reporting conditional branch events. The outcome of up to
- the TDATA field is initially cleared except for the left-most bit. which is set to 1 As each new conditional branch is encountered, a new one bit entry is added on the left and any other entries are shifted to the right by one bit. 5
- Using a 128 entry trace cache 200 allows 320 bytes of information to be stored. Assuming a branch frequency of one branch every six instructions, the disclosed trace cache 200 therefore provides an effective trace record of 1.536 instructions. This estimate does not take into account the occurrence of call.
- FIG. 7C it may be desirable to start and stop trace gathering during certain sections of program execution; for example, when a task context switch occurs.
- trace capture When trace capture is stopped, no trace entries are entered into the trace cache 200. nor do any appear on the bond-out pins of the trace port 214.
- Different methods are contemplated for enabling and disabling trace capture. For example, an x86 command can be provided, or an existing x86 command can be utilized to toggle a bit in an I/O port location.
- on-chip trigger control registers 219 can be configured to indicate the addresses where trace capture should start/stop.
- a configuration option is also desirable to enable a current segment base address entry at the end of a trace prior to entering Debug mode. By contrast, it may not be desirable to provide segment base information when the base has not changed, such as when an interrupt has occurred.
- the trace synchronization entry contains the address of the last instruction retired before the interrupt handler commences.
- Figure 7E illustrates a trace entry used to report a change in segment parameters
- trace address values are combined with a segment base address to determine an instruction's linear address.
- the base address, as well as the default data operand size (32 or 16-b ⁇ t mode), are subject to change.
- the TCODE 01 1 1 entry also includes bits indicating the current segment size (32-b ⁇ t or 16-bit), the operating mode (real or protected), and a bit indicating whether paging is being utilized. Segment information generally relates to the previous segment, not a current (target) segment. Current segment information is obtained by stopping and examining the state of the processor core 104 USER SPECIFIED TRACE ENTRY
- an x86 instruction is preferably provided which enables a 16-bit data value to be placed in the trace stream at a desired execution position.
- the instruction can be implemented as a move to I/O space, with the operand being provided by memory or a register.
- the processor core 104 executes this instruction, the user specified trace entry is captured by the trace control logic 218 and placed in the trace cache 200.
- a synchronization register TSY C is provided in the disclosed embodiment to control the ⁇ n
- FIG. 7G depicts a trace synchronization entry.
- TODE 01
- Trace entry information can also be expanded to include data relating to code coverage or execution performance. This information is useful, for example, for code testing and performance tuning. Even without these enhancements, it is desirable to enable the processor core 104 to access the trace cache 200. In the case of a microcontroller device, this feature can be accomplished by mapping the trace cache 200 within a portion of I/O or memory space. A more general approach involves including an instruction which supports moving trace cache 200 data into system memory.
- the processor-based device provides a flexible, high-performance solution for furnishing software performance profiling information.
- the processor-based device incorporates an trace cache capable of capturing and providing the profiling information. Both serial and parallel communication channels are provided for communicating the profiling information to external devices.
- the disclosed on-chip trace cache alleviates various of the bandwidth and clock synchronization problems that arise in many existing solutions, and also allows less expensive external capture hardware to be utilized when such hardware is employed.
Abstract
Description
Claims
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR10-1999-7009274A KR100522193B1 (en) | 1997-04-08 | 1998-04-07 | Microprocessor-based device incorporating a cache for capturing software performance profiling data |
EP98918024A EP0974096B1 (en) | 1997-04-08 | 1998-04-07 | Microprocessor-based device incorporating a cache for capturing software performance profiling data |
JP54301998A JP4138021B2 (en) | 1997-04-08 | 1998-04-07 | Processor-based device, method for providing software performance profiling information, and software development system for generating and analyzing software performance profiling information |
DE69801156T DE69801156T2 (en) | 1997-04-08 | 1998-04-07 | MICROPROCESSOR OPERATED ARRANGEMENT WITH CACHE MEMORY FOR RECORDING SOFTWARE PERFORMANCE PROFILE DATA |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US4307097P | 1997-04-08 | 1997-04-08 | |
US60/043,070 | 1997-04-08 | ||
US08/992,610 US6154857A (en) | 1997-04-08 | 1997-12-17 | Microprocessor-based device incorporating a cache for capturing software performance profiling data |
US08/992,610 | 1997-12-17 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO1998045784A1 true WO1998045784A1 (en) | 1998-10-15 |
Family
ID=26720006
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US1998/006838 WO1998045784A1 (en) | 1997-04-08 | 1998-04-07 | Microprocessor-based device incorporating a cache for capturing software performance profiling data |
Country Status (6)
Country | Link |
---|---|
US (1) | US6154857A (en) |
EP (1) | EP0974096B1 (en) |
JP (1) | JP4138021B2 (en) |
KR (1) | KR100522193B1 (en) |
DE (1) | DE69801156T2 (en) |
WO (1) | WO1998045784A1 (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1139220A2 (en) * | 2000-03-02 | 2001-10-04 | Texas Instruments Incorporated | Obtaining and exporting on-chip data processor trace and timing information |
WO2001080012A2 (en) * | 2000-04-11 | 2001-10-25 | Analog Devices, Inc. | Non-intrusive application code profiling method and apparatus |
KR20020053021A (en) * | 2000-12-26 | 2002-07-04 | 마찌다 가쯔히꼬 | Microcomputer |
WO2003019375A1 (en) * | 2001-08-27 | 2003-03-06 | Telefonaktiebolaget L M Ericsson (Publ) | Dynamic tracing in a real-time system |
FR2871592A1 (en) * | 2004-06-15 | 2005-12-16 | Siemens Ag | Computer program predefined function execution time measuring method for engine controlling instrument, involves generating temporal information by time-recording function, and directly transferring information to computer system |
WO2008139162A2 (en) * | 2007-05-11 | 2008-11-20 | University Of Leicester | Debugging tool |
GB2461716A (en) * | 2008-07-09 | 2010-01-13 | Advanced Risc Mach Ltd | Monitoring circuitry for monitoring accesses to addressable locations in data processing apparatus that occur between the start and end events. |
GB2473850A (en) * | 2009-09-25 | 2011-03-30 | St Microelectronics | Cache configured to operate in cache or trace modes |
US8151250B2 (en) | 2007-01-22 | 2012-04-03 | Samsung Electronics Co., Ltd. | Program trace method using a relational database |
US8312253B2 (en) | 2008-02-22 | 2012-11-13 | Freescale Semiconductor, Inc. | Data processor device having trace capabilities and method |
WO2021228766A1 (en) | 2020-05-11 | 2021-11-18 | Politecnico Di Milano | A computing platform and method for synchronize the prototype execution and simulation of hardware device |
Families Citing this family (69)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3542463B2 (en) * | 1997-07-29 | 2004-07-14 | Necエレクトロニクス株式会社 | Integrated circuit device and operation control method thereof |
US6347368B1 (en) * | 1997-12-30 | 2002-02-12 | Jerry David Harthcock | Microcomputing device for exchanging data while executing an application |
JP3684831B2 (en) * | 1998-03-31 | 2005-08-17 | セイコーエプソン株式会社 | Microcomputer, electronic equipment and debugging system |
US7111290B1 (en) | 1999-01-28 | 2006-09-19 | Ati International Srl | Profiling program execution to identify frequently-executed portions and to assist binary translation |
US8121828B2 (en) | 1999-01-28 | 2012-02-21 | Ati Technologies Ulc | Detecting conditions for transfer of execution from one computer instruction stream to another and executing transfer on satisfaction of the conditions |
US8127121B2 (en) | 1999-01-28 | 2012-02-28 | Ati Technologies Ulc | Apparatus for executing programs for a first computer architechture on a computer of a second architechture |
US6954923B1 (en) | 1999-01-28 | 2005-10-11 | Ati International Srl | Recording classification of instructions executed by a computer |
US8074055B1 (en) | 1999-01-28 | 2011-12-06 | Ati Technologies Ulc | Altering data storage conventions of a processor when execution flows from first architecture code to second architecture code |
US6763452B1 (en) | 1999-01-28 | 2004-07-13 | Ati International Srl | Modifying program execution based on profiling |
US7941647B2 (en) | 1999-01-28 | 2011-05-10 | Ati Technologies Ulc | Computer for executing two instruction sets and adds a macroinstruction end marker for performing iterations after loop termination |
US7275246B1 (en) * | 1999-01-28 | 2007-09-25 | Ati International Srl | Executing programs for a first computer architecture on a computer of a second architecture |
US6370660B1 (en) * | 1999-04-21 | 2002-04-09 | Advanced Micro Devices, Inc. | Apparatus and method for providing a wait for status change capability for a host computer system |
US6374369B1 (en) * | 1999-05-21 | 2002-04-16 | Philips Electronics North America Corporation | Stochastic performance analysis method and apparatus therefor |
US6779107B1 (en) | 1999-05-28 | 2004-08-17 | Ati International Srl | Computer execution by opportunistic adaptation |
JP3767276B2 (en) * | 1999-09-30 | 2006-04-19 | 富士通株式会社 | System call information recording method and recording apparatus |
US6684348B1 (en) | 1999-10-01 | 2004-01-27 | Hitachi, Ltd. | Circuit for processing trace information |
US6732307B1 (en) | 1999-10-01 | 2004-05-04 | Hitachi, Ltd. | Apparatus and method for storing trace information |
US6615370B1 (en) * | 1999-10-01 | 2003-09-02 | Hitachi, Ltd. | Circuit for storing trace information |
US6918065B1 (en) | 1999-10-01 | 2005-07-12 | Hitachi, Ltd. | Method for compressing and decompressing trace information |
US6542940B1 (en) * | 1999-10-25 | 2003-04-01 | Motorola, Inc. | Method and apparatus for controlling task execution in a direct memory access controller |
JP2001184212A (en) * | 1999-12-24 | 2001-07-06 | Mitsubishi Electric Corp | Trace control circuit |
DE19963832A1 (en) * | 1999-12-30 | 2001-07-05 | Ericsson Telefon Ab L M | Program profiling |
US6687816B1 (en) * | 2000-04-04 | 2004-02-03 | Peoplesoft, Inc. | Configuration caching |
GB2368689B (en) * | 2000-06-28 | 2004-12-01 | Ibm | Performance profiling in a data processing system |
JP2002099447A (en) * | 2000-09-22 | 2002-04-05 | Fujitsu Ltd | Processor |
US20020073406A1 (en) * | 2000-12-12 | 2002-06-13 | Darryl Gove | Using performance counter profiling to drive compiler optimization |
JP3491617B2 (en) * | 2001-03-01 | 2004-01-26 | 日本電気株式会社 | Operation report creation method, operation report creation method, and operation report creation program |
US7047519B2 (en) * | 2001-09-26 | 2006-05-16 | International Business Machines Corporation | Dynamic setting of breakpoint count attributes |
US6961928B2 (en) * | 2001-10-01 | 2005-11-01 | International Business Machines Corporation | Co-ordinate internal timers with debugger stoppage |
US6931492B2 (en) * | 2001-11-02 | 2005-08-16 | International Business Machines Corporation | Method for using a portion of the system cache as a trace array |
US7073048B2 (en) * | 2002-02-04 | 2006-07-04 | Silicon Lease, L.L.C. | Cascaded microcomputer array and method |
US7107489B2 (en) * | 2002-07-25 | 2006-09-12 | Freescale Semiconductor, Inc. | Method and apparatus for debugging a data processing system |
US20040019828A1 (en) * | 2002-07-25 | 2004-01-29 | Gergen Joseph P. | Method and apparatus for debugging a data processing system |
US7013409B2 (en) * | 2002-07-25 | 2006-03-14 | Freescale Semiconductor, Inc. | Method and apparatus for debugging a data processing system |
DE10234469A1 (en) * | 2002-07-29 | 2004-02-12 | Siemens Ag | Program run-time measurement method for processor-controlled data processing unit, monitors clock signal during execution of program commands |
GB2393272A (en) * | 2002-09-19 | 2004-03-24 | Advanced Risc Mach Ltd | Controlling performance counters within a data processing system |
US20050125784A1 (en) * | 2003-11-13 | 2005-06-09 | Rhode Island Board Of Governors For Higher Education | Hardware environment for low-overhead profiling |
EP1531395A1 (en) * | 2003-11-17 | 2005-05-18 | Infineon Technologies AG | Method of determining information about the processes which run in a program-controlled unit during the execution of a program |
US7404178B2 (en) | 2004-02-18 | 2008-07-22 | Hewlett-Packard Development Company, L.P. | ROM-embedded debugging of computer |
WO2005109203A2 (en) * | 2004-05-12 | 2005-11-17 | Koninklijke Philips Electronics N.V. | Data processing system with trace co-processor |
CN101002169A (en) | 2004-05-19 | 2007-07-18 | Arc国际(英国)公司 | Microprocessor architecture |
JP4336251B2 (en) * | 2004-06-01 | 2009-09-30 | インターナショナル・ビジネス・マシーンズ・コーポレーション | Traceability system, trace information management method, trace information management program, and recording medium |
US7353429B2 (en) * | 2004-08-17 | 2008-04-01 | International Business Machines Corporation | System and method using hardware buffers for processing microcode trace data |
US7849364B2 (en) * | 2005-03-01 | 2010-12-07 | Microsoft Corporation | Kernel-mode in-flight recorder tracing mechanism |
US20060218204A1 (en) * | 2005-03-25 | 2006-09-28 | International Business Machines Corporation | Log stream validation in log shipping data replication systems |
US7475291B2 (en) * | 2005-03-31 | 2009-01-06 | International Business Machines Corporation | Apparatus and method to generate and save run time data |
US20060224925A1 (en) * | 2005-04-05 | 2006-10-05 | International Business Machines Corporation | Method and system for analyzing an application |
US7797686B2 (en) * | 2005-05-13 | 2010-09-14 | Texas Instruments Incorporated | Behavior of trace in non-emulatable code |
US20060255978A1 (en) * | 2005-05-16 | 2006-11-16 | Manisha Agarwala | Enabling Trace and Event Selection Procedures Independent of the Processor and Memory Variations |
US7590892B2 (en) * | 2005-05-16 | 2009-09-15 | Texas Instruments Incorporated | Method and system of profiling real-time streaming channels |
US7788645B2 (en) * | 2005-05-16 | 2010-08-31 | Texas Instruments Incorporated | Method for guaranteeing timing precision for randomly arriving asynchronous events |
US7607047B2 (en) * | 2005-05-16 | 2009-10-20 | Texas Instruments Incorporated | Method and system of identifying overlays |
US20060277435A1 (en) | 2005-06-07 | 2006-12-07 | Pedersen Frode M | Mechanism for storing and extracting trace information using internal memory in microcontrollers |
US20060294343A1 (en) * | 2005-06-27 | 2006-12-28 | Broadcom Corporation | Realtime compression of microprocessor execution history |
US7650539B2 (en) * | 2005-06-30 | 2010-01-19 | Microsoft Corporation | Observing debug counter values during system operation |
US7747088B2 (en) | 2005-09-28 | 2010-06-29 | Arc International (Uk) Limited | System and methods for performing deblocking in microprocessor-based video codec applications |
GB2447683B (en) * | 2007-03-21 | 2011-05-04 | Advanced Risc Mach Ltd | Techniques for generating a trace stream for a data processing apparatus |
JP2008293061A (en) * | 2007-05-22 | 2008-12-04 | Nec Electronics Corp | Semiconductor device, and debugging method of semiconductor device |
JPWO2009011028A1 (en) * | 2007-07-17 | 2010-09-09 | 株式会社アドバンテスト | Electronic device, host device, communication system, and program |
US20090037886A1 (en) * | 2007-07-30 | 2009-02-05 | Mips Technologies, Inc. | Apparatus and method for evaluating a free-running trace stream |
JP5029245B2 (en) * | 2007-09-20 | 2012-09-19 | 富士通セミコンダクター株式会社 | Profiling method and program |
US7802142B2 (en) * | 2007-12-06 | 2010-09-21 | Seagate Technology Llc | High speed serial trace protocol for device debug |
US8140911B2 (en) * | 2008-03-20 | 2012-03-20 | International Business Machines Corporation | Dynamic software tracing |
US10169187B2 (en) | 2010-08-18 | 2019-01-01 | International Business Machines Corporation | Processor core having a saturating event counter for making performance measurements |
JP5310819B2 (en) * | 2010-11-29 | 2013-10-09 | 株式会社デンソー | Microcomputer |
KR101301022B1 (en) * | 2011-12-23 | 2013-08-28 | 한국전자통신연구원 | Apparatus for protecting against external attack for processor based on arm core and method using the same |
JP5710543B2 (en) * | 2012-04-26 | 2015-04-30 | 京セラドキュメントソリューションズ株式会社 | Semiconductor integrated circuit |
US10289540B2 (en) * | 2016-10-06 | 2019-05-14 | International Business Machines Corporation | Performing entropy-based dataflow analysis |
US10761588B2 (en) * | 2018-08-09 | 2020-09-01 | Micron Technology, Inc. | Power configuration component including selectable configuration profiles corresponding to operating characteristics of the power configuration component |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5058114A (en) * | 1988-03-15 | 1991-10-15 | Hitachi, Ltd. | Program control apparatus incorporating a trace function |
US5371689A (en) * | 1990-10-08 | 1994-12-06 | Fujitsu Limited | Method of measuring cumulative processing time for modules required in process to be traced |
Family Cites Families (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3707725A (en) * | 1970-06-19 | 1972-12-26 | Ibm | Program execution tracing system improvements |
JPS59194245A (en) * | 1983-04-19 | 1984-11-05 | Nec Corp | Microprogram controller |
JPH01134541A (en) * | 1987-11-20 | 1989-05-26 | Toshiba Corp | Information processor |
DE69415600T2 (en) * | 1993-07-28 | 1999-07-15 | Koninkl Philips Electronics Nv | Microcontroller with hardware troubleshooting support based on the boundary scan method |
US5537541A (en) * | 1994-08-16 | 1996-07-16 | Digital Equipment Corporation | System independent interface for performance counters |
US5964893A (en) * | 1995-08-30 | 1999-10-12 | Motorola, Inc. | Data processing system for performing a trace function and method therefor |
US5774724A (en) * | 1995-11-20 | 1998-06-30 | International Business Machines Coporation | System and method for acquiring high granularity performance data in a computer system |
US5724505A (en) * | 1996-05-15 | 1998-03-03 | Lucent Technologies Inc. | Apparatus and method for real-time program monitoring via a serial interface |
US5898873A (en) * | 1996-11-12 | 1999-04-27 | International Business Machines Corporation | System and method for visualizing system operation trace chronologies |
GB9626367D0 (en) * | 1996-12-19 | 1997-02-05 | Sgs Thomson Microelectronics | Providing an instruction trace |
-
1997
- 1997-12-17 US US08/992,610 patent/US6154857A/en not_active Expired - Lifetime
-
1998
- 1998-04-07 JP JP54301998A patent/JP4138021B2/en not_active Expired - Lifetime
- 1998-04-07 WO PCT/US1998/006838 patent/WO1998045784A1/en active IP Right Grant
- 1998-04-07 EP EP98918024A patent/EP0974096B1/en not_active Expired - Lifetime
- 1998-04-07 DE DE69801156T patent/DE69801156T2/en not_active Expired - Lifetime
- 1998-04-07 KR KR10-1999-7009274A patent/KR100522193B1/en not_active IP Right Cessation
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5058114A (en) * | 1988-03-15 | 1991-10-15 | Hitachi, Ltd. | Program control apparatus incorporating a trace function |
US5371689A (en) * | 1990-10-08 | 1994-12-06 | Fujitsu Limited | Method of measuring cumulative processing time for modules required in process to be traced |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1139220A3 (en) * | 2000-03-02 | 2009-10-28 | Texas Instruments Incorporated | Obtaining and exporting on-chip data processor trace and timing information |
EP1139220A2 (en) * | 2000-03-02 | 2001-10-04 | Texas Instruments Incorporated | Obtaining and exporting on-chip data processor trace and timing information |
WO2001080012A2 (en) * | 2000-04-11 | 2001-10-25 | Analog Devices, Inc. | Non-intrusive application code profiling method and apparatus |
WO2001080012A3 (en) * | 2000-04-11 | 2002-05-30 | Analog Devices Inc | Non-intrusive application code profiling method and apparatus |
JP2003531436A (en) * | 2000-04-11 | 2003-10-21 | アナログ デバイセス インコーポレーテッド | Method and apparatus for non-intrusive application code profiling |
KR20020053021A (en) * | 2000-12-26 | 2002-07-04 | 마찌다 가쯔히꼬 | Microcomputer |
WO2003019375A1 (en) * | 2001-08-27 | 2003-03-06 | Telefonaktiebolaget L M Ericsson (Publ) | Dynamic tracing in a real-time system |
GB2395040A (en) * | 2001-08-27 | 2004-05-12 | Ericsson Telefon Ab L M | Dynamic tracing in a real-time system |
GB2395040B (en) * | 2001-08-27 | 2005-05-11 | Ericsson Telefon Ab L M | Dynamic tracing in a real-time system |
FR2871592A1 (en) * | 2004-06-15 | 2005-12-16 | Siemens Ag | Computer program predefined function execution time measuring method for engine controlling instrument, involves generating temporal information by time-recording function, and directly transferring information to computer system |
US8151250B2 (en) | 2007-01-22 | 2012-04-03 | Samsung Electronics Co., Ltd. | Program trace method using a relational database |
WO2008139162A3 (en) * | 2007-05-11 | 2009-02-26 | Univ Leicester | Debugging tool |
WO2008139162A2 (en) * | 2007-05-11 | 2008-11-20 | University Of Leicester | Debugging tool |
US8312253B2 (en) | 2008-02-22 | 2012-11-13 | Freescale Semiconductor, Inc. | Data processor device having trace capabilities and method |
GB2461716A (en) * | 2008-07-09 | 2010-01-13 | Advanced Risc Mach Ltd | Monitoring circuitry for monitoring accesses to addressable locations in data processing apparatus that occur between the start and end events. |
GB2461644A (en) * | 2008-07-09 | 2010-01-13 | Advanced Risc Mach Ltd | Circuitry for monitoring accesses to addressable locations in data processing apparatus that occur between start and end events |
GB2461644B (en) * | 2008-07-09 | 2012-11-14 | Advanced Risc Mach Ltd | Monitoring a data processing apparatus and summarising the monitoring data |
US9858169B2 (en) | 2008-07-09 | 2018-01-02 | Arm Limited | Monitoring a data processing apparatus and summarising the monitoring data |
GB2473850A (en) * | 2009-09-25 | 2011-03-30 | St Microelectronics | Cache configured to operate in cache or trace modes |
WO2021228766A1 (en) | 2020-05-11 | 2021-11-18 | Politecnico Di Milano | A computing platform and method for synchronize the prototype execution and simulation of hardware device |
Also Published As
Publication number | Publication date |
---|---|
KR20010006194A (en) | 2001-01-26 |
EP0974096B1 (en) | 2001-07-18 |
DE69801156D1 (en) | 2001-08-23 |
US6154857A (en) | 2000-11-28 |
JP2001519949A (en) | 2001-10-23 |
JP4138021B2 (en) | 2008-08-20 |
EP0974096A1 (en) | 2000-01-26 |
KR100522193B1 (en) | 2005-10-18 |
DE69801156T2 (en) | 2002-03-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6154857A (en) | Microprocessor-based device incorporating a cache for capturing software performance profiling data | |
EP0974094B1 (en) | Trace cache for a microprocessor-based device | |
US6009270A (en) | Trace synchronization in a processor | |
US6185732B1 (en) | Software debug port for a microprocessor | |
US6041406A (en) | Parallel and serial debug port on a processor | |
US6175914B1 (en) | Processor including a combined parallel debug and trace port and a serial port | |
EP0974093B1 (en) | Debug interface including a compact trace record storage | |
US5978902A (en) | Debug interface including operating system access of a serial/parallel debug port | |
US6314530B1 (en) | Processor having a trace access instruction to access on-chip trace memory | |
US6189140B1 (en) | Debug interface including logic generating handshake signals between a processor, an input/output port, and a trace logic | |
US6145123A (en) | Trace on/off with breakpoint register | |
EP0567722B1 (en) | System for analyzing and debugging embedded software through dynamic and interactive use of code markers | |
US6145100A (en) | Debug interface including timing synchronization logic | |
US6154856A (en) | Debug interface including state machines for timing synchronization and communication | |
US5724505A (en) | Apparatus and method for real-time program monitoring via a serial interface | |
US6142683A (en) | Debug interface including data steering between a processor, an input/output port, and a trace logic | |
US5754759A (en) | Testing and monitoring of programmed devices | |
US5265254A (en) | System of debugging software through use of code markers inserted into spaces in the source code during and after compilation | |
US6145122A (en) | Development interface for a data processor | |
US6854029B2 (en) | DSP bus monitoring apparatus and method | |
US5768152A (en) | Performance monitoring through JTAG 1149.1 interface | |
US5809293A (en) | System and method for program execution tracing within an integrated processor | |
US6513134B1 (en) | System and method for tracing program execution within a superscalar processor | |
EP0869434A2 (en) | Method for outputting trace information of a microprocessor | |
EP1184790B1 (en) | Trace cache for a microprocessor-based device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): JP KR |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE |
|
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 1998918024 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref country code: JP Ref document number: 1998 543019 Kind code of ref document: A Format of ref document f/p: F |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1019997009274 Country of ref document: KR |
|
WWP | Wipo information: published in national office |
Ref document number: 1998918024 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 1019997009274 Country of ref document: KR |
|
WWG | Wipo information: grant in national office |
Ref document number: 1998918024 Country of ref document: EP |
|
WWG | Wipo information: grant in national office |
Ref document number: 1019997009274 Country of ref document: KR |