USPatent applicationPatented

Apparatus and method for automatically extracting interface of embedded software

Granted 28 Aug 2012 · 1 office action

Life of the application

9 dated events
⤢ drag to zoom20082010201220142016201820202022202420262028ProsecutionOwnershipTerm & fees
ProsecutionOwnershipTerm & feeshover for detail · click to open

Abstract

Provided is an apparatus and method for automatically extracting an interface of embedded software. The method defines an interface of embedded software as a test item for test of the interface of the embedded software, and generates an interface symbol corresponding to the defined test item. The location of the interface symbol is determined, and a test case is generated based on the interface symbol and its location in the software. Embodiments of the invention thus facilitate the testing of software at interfaces between layers in the software.

Description

8 parts
›CROSS-REFERENCE TO RELATED APPLICATIONS

This U.S. non-provisional patent application claims priority under 35 U.S.C. §119 of Korean Patent Application No. 10-2007-0040127, filed on Apr. 25, 2007, the entire contents of which are hereby incorporated by reference.

›BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention disclosed herein relates to embedded software test, and more particularly but not by way of limitation, to a method for automatically generating a test case.

2. Description of Related Art

Typical embedded software is tested by monitoring and debugging an emulator. The monitoring may be synchronous or asynchronous. In synchronous monitoring, a source code debugger such as the remote Kernal GNU DeBugger KGDB is used to stop a program at a certain location in a source code and monitor a value of the corresponding location. In asynchronous monitoring, an independent monitoring daemon is used to monitor the entire embedded software, which may lead to inaccurate measurement of a source code that is to be monitored.

In both the synchronous monitoring scheme and the asynchronous monitoring technique, a code area, a data area, global variables, and a stack area may be monitored. Many such tests are performed on an ad-hoc basis rather than on an objective basis because determination of monitoring target symbols and analysis of the results may vary depending on the developers' experience/knowledge levels. Furthermore, simple monitoring techniques focus on test automation rather than identification of a test target (i.e., an input of a test case), generation of the test case, and analysis of the test results.

Embedded software is typically linked with hardware, a hardware abstraction layer (HAL), an OS kernel, device drivers, and upper-level application software. In an embedded system, the interfaces between such layers must be tested. This is, however, difficult to achieve using conventional test technology.

›SUMMARY OF THE INVENTION

An embodiment of the invention provides an apparatus and method for automatically extracting interface symbols to test an interface of embedded software.

An embodiment of the invention also provides an apparatus and method for automatically generating a test case to test an interface of embedded software.

An embodiment of the invention provides a method for testing software. The method includes: identifying an interface test feature associated with the software, the interface test feature being associated with an interface symbol, the interface symbol enabling measurement of the interface test feature; and extracting a location in the software associated with the interface symbol.

An embodiment of the invention provides an apparatus for testing software. The apparatus includes: a test target analyzer configured to receive the software, identify at least one interface between a plurality of layers in the software, and identify a location in the software for the at least one interface; and a test case generator coupled to the test target analyzer, the test case generator configured to generate a test case for the software based on the identified at least one interface and the location.

Embodiments of the invention thus facilitate the testing of software at interfaces between layers in the software.

›BRIEF DESCRIPTION OF THE FIGURES

The accompanying figures are included to provide a further understanding of the present invention, and are incorporated in and constitute a part of this specification. The drawings illustrate exemplary embodiments of the present invention and, together with the description, serve to explain principles of the present invention. In the figures:

FIG. 1 is a block diagram of an embedded software test system according to an embodiment of the present invention;

FIG. 2 is a block diagram of an embedded system according to an embodiment of the present invention;

FIG. 3 illustrates interfaces of the embedded system illustrated in FIG. 2 according to an embodiment of the present invention;

FIG. 4 illustrates standard file operations for an LCD device driver;

FIG. 5 illustrates an embodiment of the standard file operations illustrated in FIG. 4 ;

FIGS. 6 through 13 illustrate interface test items of an LCD device driver, which are shown in Table 2;

FIG. 14 illustrates a test script for test of collision with an OPI 3 — 6 kernel stack;

FIG. 15 is a block diagram illustrating generation of a test case for testing an interface of embedded software according to the present invention; and

FIG. 16 is a flowchart illustrating a method of generating a test case for testing an interface of embedded software according to an embodiment of the present invention.

›DETAILED DESCRIPTION OF EMBODIMENTS · 1 of 4

Preferred embodiments of the present invention will be described below in more detail with reference to the accompanying drawings. The present invention may, however, be embodied in different forms and should not be constructed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the present invention to those skilled in the art.

An embodiment of the invention provides an apparatus and method for automatically generating a test case for a LINUX-based LCD device driver. In this instance, an evaluation board can be emulated prior to actual product development. An embodiment of the invention exemplifies an LCD device driver as embedded software.

FIG. 1 is a block diagram of an embedded software test system according to an embodiment of the present invention.

Referring to FIG. 1 , a test system 100 for embedded software includes a host system 10 , an emulator 20 , an evaluation board 30 , and a LINUX server 40 . The LINUX server 40 includes embedded software 41 .

The host system 10 generates a test case for testing an interface of the embedded software 41 . In addition, the host system 10 controls the emulator 20 to execute the generated test case. The emulator 20 executes the test case generated by the host system 10 , and outputs the results. The evaluation board 30 is provided to test the embedded software 41 together with hardware. The LINUX server 40 loads the embedded software 41 into the evaluation board 30 under the control of the emulator 20 .

The host system 10 includes a test target analyzer 11 , a test case generator 12 , a test executing engine 13 , and a result analyzer 14 . The test target analyzer 11 receives a source (or object) code of the embedded software 41 from the LINUX server 40 . From an executable linker format (ELF), the test target analyzer 11 automatically identifies a location in the source (or object) code where an interface symbol corresponding to each interface is mapped. The test case generator 12 generates a test case for testing the interface of the embedded software 41 . The test case is generated in a script format so that it can be executed in the emulator 20 . The test execution engine 13 receives the test case from the test case generator 12 , provides the received test case to the emulator 20 , and controls the emulator 20 . The result analyzer 14 receives the execution results of the test case from the emulator 20 , and analyzes the received execution results.

Variations to the embodiment illustrated in FIG. 1 are possible. For example, a server other than a LINUX server may be used, and code formats other than those specified above may also be used, according to design choice. Likewise, the emulator 20 must not necessarily be a TRACE32 emulator as indicated in FIG. 1 and as described below.

The present invention tests the structure of embedded software (i.e., an LCD device driver) loaded into an embedded system, for example a mobile application (AP), and tests an interface in an LCD emulation test.

FIG. 2 is a block diagram of an embedded system according to an embodiment of the present invention. In the test environment, the embedded system is embodied in the evaluation board 30 . Hence, the embedded system of FIG. 2 is also referred to herein as the embedded system 30 .

Referring to FIG. 2 , the embedded system 30 includes a physical hardware layer 31 , a hardware abstraction layer (HAL) 32 , an operation system (OS) layer 35 , and an application layer 36 . The OS layer 35 includes embedded software 33 and kernel software 34 .

The embedded system 30 may include a variety of device drivers for supporting physical devices mounted on the evaluation board 30 , examples of which include data transfer-related device drivers (such as a USB driver and a serial bus driver) and audio-related device drivers such as Audio code 1997 (AC97) and Infrared Data Association (IrDA) links.

The embedded software 33 runs in such a way that software units in respective layers are tightly coupled with each other. The software units in the respective layers are classified into a kernel-dependent software unit (SU k ), a device-dependent software unit (SU d ), and a processor-dependent software unit (SU p ). An exemplary SU k is an OS application program interface for providing an OS service. An exemplary SU d is software of an LCD controller for supporting a physical device. An exemplary SU p is software of the HAL 32 , which is dependent on a target processor, such as for hardware device initialization and configuration.

FIG. 3 illustrates interfaces of the embedded system illustrated in FIG. 2 according to an embodiment of the present invention.

Referring to FIGS. 2 and 3 , the embedded software is tightly coupled with hardware units as well as with the software units mentioned above. The software units are directly connected to the hardware units when a desired value needs to be read and written through a random access memory (RAM) or a register. The hardware units are subdivided into a hardware unit_d (HU d ) and a hardware unit_i (HU i ), depending on their structures for coupling with the software units. Examples of the HU d include RAMs and registers that can be directly read/written from software to hardware. Examples of the HU i include universal asynchronous receiver/transmitters (UARTs), universal serial buses (USBs), and watchdog timers (WDTs) whose read/write operations are indirectly controlled through an address of the hardware unit_d HU d . Hardware-dependent Part Interfaces (HPI's) are divided into HPI 1 and HPI 2 . Using an HPI 1 , the SU d or the SU p accesses the HU d to read/write an expected value. Using an HPI 2 , the SU d or the SU p controls the HU i through the address of the HU d .

The LCD controller of the LCD device driver exemplified herein corresponds to the HU i .

Referring to FIGS. 2 and 3 , for analysis of an interface with the OS layer, a control flow of the OS varies depending on the types of OS services such as process management, inter-process communication and synchronization, exception handling, virtual memory management, physical memory management, time management, interrupt handling, IO management, networking, and file system management. That is, on the basis of a control flow for each OS service, OS-dependent Part Interfaces (OPI's) are divided into an OPI 1 , an OPI 2 , and an OPI 3 . Using the OPI 1 , the SU d or the SU p calls an SU k irrelevant to HU d /HU i control. In this case, the SU k corresponds to an application programming interface (API) for process management, inter-process communication, and exception handling. Using the OPI 2 , the SU d or the SU p calls a SU k relevant to HU d control. In this case, the SU k corresponds to an API for virtual memory management. Using the OPI 3 , the SU d or the SU p calls a SU k relevant to the HU d /HU i control. In this case, the SU k corresponds to an API for physical memory management, time management, interrupt handling, IO management, networking, or file system management.

›DETAILED DESCRIPTION OF EMBODIMENTS · 2 of 4

The five interfaces illustrated in FIG. 3 become the criteria of test coverage for testing an interface of the embedded software. Each interface is associated with a symbol for test case selection and pass/fail determination, and is further associated with a monitoring location for detecting the cause(s) of a fault in a debugging process. Interface test coverage requires testing of all applicable interfaces. The applicable interfaces are determined according to interface characteristics in the embedded software being tested.

An OS of an exemplary LCD device driver is based on LINUX. Therefore, the device driver follows the LINUX standard file operations, and upper-level application software can invoke (or enable) the device driver only through standard file operations such as a function open( ), a function close( ), a function read( ), a function write( ), a function mset( ), and a function ioctl( ).

FIG. 4 illustrates a function fb_open( ), a function fb_release( ), a function fb_read( ), a function fb_write( ), a function fb_mmap( ), and a function fb_ioctl( ) that are standard file operations for a frame buffer of the LCD device driver. FIG. 5 illustrates an embodiment of the standard file operations illustrated in FIG. 4 . Table 1 shows an interface for the function fb_write( ).

An interface of the LCD device driver is a unit where functions defined in LCD file operations are present between the software unit SU k /SU d /SU p and the hardware unit HU d /HU i in a call relationship.

Referring to FIGS. 4 and 5 , a function fb_write( ) invoked by a function write( ) uses an HPI 1 to write a desired value in a RAM through a function get_con_display( ), and uses an OPI 3 to write a user input value in the RAM through a function copy_from_user( ).

The LCD device driver includes a total of 74 interfaces for a function s3c2440fb_init( ), a function fb_open( ), a function fb_release( ), a function fb_read( ), a function fb_write( ), a function fb_mmap( ), and a function fb_ioctl( ). The 74 interfaces are classified into 32 HPI 1 's, 2 HPI 2 's, and 40 OPI 3 's, based on the interface pattern.

Table 2 illustrates the interface test characteristics of the LCD device driver and interface symbols that must be monitored for the respective characteristics.

From an executable linker format (ELF) file of a target source, the test target analyzer 11 identifies a location where an interface symbol of the interface test characteristics of the LCD device driver, which is shown in Table 2, is mapped in a test target source code. The ELF file is used to store all information on a source that is programmed by the developer for the use of a debugger. When a call stack is difficult to identify, for example, by backtrace, “objdump” is a command for identifying a source address and “readelf” is a command for indicating ELF file format information.

FIGS. 6 through 13 illustrate interface test items of an LCD device driver, which are shown in Table 2.

Referring to FIGS. 6 through 13 , in order to detect a location where the memory-related interface test characteristics of an HPI 1 is mapped to the source code, the size and address of a symbol are extracted from an ELF file generated by a command objdump −t and the type of the symbol is extracted from an ELF file generated by a command readelf −wi.

An HPI 1 — 1 static memory map boundary check is performed to test whether a static global variable ranges over a physical memory map boundary as illustrated in FIG. 6 . Thus, the address and size of the static global variable are an interface of the HPI 1 — 1 .

An HPI 1 — 2 kernel stack collision test is used to test whether a static global variable collides with a kernel stack as indicated by a dotted line in FIG. 7 . Thus, the address, size and type of the static global variable are an interface of the HPI 1 — 2 .

An HPI 1 — 3 user stack collision test is used to test whether a static global variable collides with a user stack as indicated by a dotted line in FIG. 8 . Thus, the address, size and type of the static global variable are an interface of the HPI 1 — 3 .

In order to detect a location where the timer-related interface test characteristics of an HPI 2 are mapped to a source code, a symbol related to a timer of a frame buffer is extracted from an ELF file generated by a command objdump −t.

An HPI 2 — 1 frame buffer response time test measures the time for a function fb_mmap( ) enabled by a function mset( ) of an application, to test a frame buffer response time. The time taken from the start of the function fb_mmap( ) to the return of the function fb_mmap( ) (i.e., completion of memory mapping) is measured. Thus, the called address of the function fb_mmap( ) is an interface of the HPI 2 — 1 .

In order to detect a location where the interrupt & memory management-related interface test characteristics of an OPI 3 are mapped to a source code, an address for an interface symbol of the OPI 3 is extracted from an ELF file generated by a command objdump −IS.

An OPI 3 — 1 interrupt context recovery test is used to test whether a context executed before the occurrence of an interrupt is correctly recovered after the execution of the interrupt. As illustrated in FIG. 9 , the called addresses of the register storage and recovery commands stmia and ldmia of do_IRQ after the call of _irq_svc are an interface of the OPI 3 — 1 .

The OPI 3 — 2 interrupt latency test is used to measure the time taken from the storage of a context to the recovery of the stored context after the occurrence of an interrupt. Thus, the called addresses of the register storage and recovery commands stmia and ldmia of do_IRQ are an interface of the OPI 3 — 2 .

An OPI 3 — 3 normal interrupt service routine test is used to test whether an interrupt service routine (ISR) for a generated interrupt is correctly executed. As illustrated in FIG. 10 , when do_IRQ is called due to the generation of an interrupt, an IRQ number is identified from an RO register and an ISR corresponding to the IRQ number is called by irq_desc[irq].action→handler. Thus, the called address of do_IRQ is an interface of the OPI 3 — 3 .

›DETAILED DESCRIPTION OF EMBODIMENTS · 3 of 4

An OPI 3 — 4 abnormal interrupt service routine test is used to test whether an ISR corresponding to an excessive interrupt is correctly executed as many times as the number of generated interrupts as illustrated in FIG. 10 . Thus, the called address of do_IRQ is an interface of the OPI 3 — 4 .

An OPI 3 — 5 dynamic memory map boundary check is performed to test whether a symbol dynamically allocated by a function kmalloc( ) ranges over a physical memory map boundary as illustrated in FIG. 6 . Thus, the called address of the function kmalloc( ) is an interface of the OPI 3 — 5 .

An OPI 3 — 6 kernel stack collision test is used to test whether there is a collision with a kernel stack in the case of access to a kernel space by a function _arch_copy_from_user( ), a function _arch_strncpy_from_user( ), and a function get_user( ), as indicated by a solid line in FIG. 7 . Thus, the called addresses of the function _arch_copy_from_user( ), the function _arch_strncpy_from_user( ), and the function get_user( ) are an interface of the OPI 3 — 6 .

An OPI 3 — 7 user stack collision test is used to test whether there is a collision with a user stack when data of a kernel space move into a user space by a function copy_to_user( ), a function put_user( ), and a function _arch_clear_user( ), as indicated by a solid line in FIG. 8 . Thus, the called addresses of the function copy_to_user( ), the function put_user( ), and the function _arch_clear_user( ) are an interface of the OPI 3 — 7 .

An OPI 3 — 8 heap collision test is used to test a collision between symbols in a kernel data segment when a function memset( ), a function _memzero( ), and a function memcpy( ) are called, as indicated by a solid line in FIG. 7 . Thus, the called addresses of the function memset( ), the function _memzero( ), and the function memcpy( ) are an interface of the OPI 3 — 8 .

An OPI 3 — 9 memory leakage test is used to test whether an address allocated by the function kmalloc( ) is released by a function kfree( ). Thus, the called address of the function kmalloc( ) is an interface of the OPI 3 — 9 .

In the case of an OPI 3 — 10 memory free latency test, because a function kfree( ) in a function kmem_cache_free_one( ) disconnects a link of an actual memory when the function kfree( ) is called as illustrated in FIG. 11 , the time taken from the call of the function kfree( ) to the call of the function kmem_cache_free_one( ) corresponds to a memory free latency. Thus, the called addresses of the function kfree( ) and the function kmem_cache_free_one( ) are an interface of the OPI 3 — 10 .

In the case of an OPI 3 — 11 mis-align check, because an alignment must be performed on a word basis to be able to prevent a page fault when a code is loaded into a memory, the virtual addresses of symbols in an init section, a code section, and a data section must end with 0, 4, 8, or C, as illustrated in FIG. 12 . Thus, the addresses for the symbols in the sections are an interface of the OPI 3 — 11 .

An OPI 3 — 12 critical region test is used to test whether a critical region is protected from a fast interrupt request (FIQ). As illustrated in FIG. 13 , the call addresses of commands mrs, orr and msr for disabling the FIQ simultaneously with critical region entry and the call addresses of commands mrs, bic and msr for enabling the FIQ simultaneously with critical region release are an interface of the OPI 3 — 12 .

The test case generator 12 generates a test case that covers a test target LCD device driver interface detected by the test target analyzer 11 . The test case is constructed to include an interface symbol corresponding to an input, which is to be monitored in each interface, data of the interface symbol, and an expected output.

A test case for each of the LCD device driver interface characteristics is determined according to each LCD device driver interface location and type. Table 3 shows expected output reference locations that can be used instead of expected output values. The reason for this is that an output value changes every time it is monitored through operation of the emulator 20 .

Test cases generated by the test case generator 12 are constructed in the format of a test script suitable for batch run by the emulator 20 . An interface symbol in Table 3 is a location for setting a break point (BP) of a test script, and an expected output reference location in Table 3 is a location that is to be monitored for determination of pass/fail at the location of the interface symbol.

FIG. 14 illustrates a test script for test of collision with an OPI 3 — 6 kernel stack.

Referring to FIG. 14 , BPs are set at the called addresses of functions _arch_copy_from_user( ), _arch_strncpy_from_user( ) and get_user( ). A value obtained by adding the size from an R2 register in the emulator must not collide with a stack pointer (SP) of a kernel stack at the set location. At this point, a stack point (SP) of the kernel stack is monitored by a debugger of the emulator 20 using the value of an R13_SVC register.

An exemplary embodiment of the present invention provides a method for automatically generating an interface test case for the interface test of a LINUX-based LCD device driver. The LINUX-based LCD device driver is exemplified as being developed on an S3C2440 board having a TRACE32-ICD emulator and an ARM processor.

FIG. 15 is a block diagram illustrating generation of a test case for testing an interface of embedded software according to the present invention.

Referring to FIG. 15 , in an automatic test case generation method for the interface test of an LCD device driver, a test target analyzer 11 automatically identifies interface test characteristics specified in a test model 41 and an interface symbol where each of the interface test characteristics is mapped in an object code. The test case generator 12 receives the identified interface symbol, and automatically generates a test case using pass/fail criteria based on the input interface symbol.

FIG. 16 is a flowchart illustrating a method of generating a test case for testing an interface of embedded software according to an embodiment of the present invention. In an embodiment of the present invention, an LCD device driver is exemplified as the embedded software 41 .

›DETAILED DESCRIPTION OF EMBODIMENTS · 4 of 4

Referring to FIGS. 1 and 16 , the LINUX server 40 provides an object code of the LCD device driver 41 to the test target analyzer 11 (step 210 ). The LINUX server 40 may also provide a source code of the LCD device driver 41 to the test target analyzer 11 . From the object code of the LCD device driver 41 , the test target analyzer 11 generates an ELF file using ARM processor-based commands objdump −t, objdump −I −S and readelf −wi.

The test target analyzer 11 identifies a layer of the LCD device driver 41 (step 220 ). That is, the test target analyzer 11 identifies an inter-layer interface of the LCD device driver 41 according to classified definitions. For example, it is determined whether the structure of sub-functions of the LCD device driver 41 corresponds to the interface defined in Table 1. The test target analyzer 11 identifies the call relationship of the LCD device driver (step 230 ). That is, only sub-functions called between interfaces are selected as a test target.

For the sub-functions called between interfaces, the test target analyzer 11 identifies an interface of the LCD device driver, which is defined in Table 2, and an interface symbol to be tested in the interface (step 240 ). In order to automatically identify a location where the interface symbol is mapped in the source code, the test target analyzer 11 maps the interface symbols of the LCD device driver corresponding to HPI 1 , HPI 2 , OPI 1 , OPI 2 and OPI 3 from the ELF file and the object code of the LCD device driver (step 250 ). The test target analyzer 11 extracts a location of the mapped symbol and the test case generator 12 sets each break point in the interface extracted by the test target analyzer 11 (step 260 ). The test case generator 12 sets each break point as an input of a test case (step 270 ). The test case generator 12 determines monitoring symbols that are to be monitored for determination of pass/fail for each interface (step 280 ). Based on this, an expected output of a corresponding interface is set (step 290 ). That is, a test case is generated using the input and the expected output. At this point, the test case is generated in a script format suitable for batch run in the emulator (step 300 ).

The method for generating the test case for the interface test of the LINUX-based LCD device driver according to the present invention combines the debugging/monitoring function of the emulator with the interface testing function, thereby supporting fault detection and fault cause estimation and automating the test for the LINUX-based LCD device driver.

Accordingly, the present invention automatically identifies the interface symbols for software of all layers that are organically linked with the LCD device driver, including the LINUX-based LCD device driver and the kernel related thereto.

In addition, the present invention automatically generates the test case for the interface test of the LINUX-based LCD device driver. The present invention automatically identifies the test case covering the interface of the identified LINUX-based LCD device driver, and adds a function for automatically determining the pass/fail of a test input symbol to the test case, thereby determining the test pass/fail automatically when the generated test case is executed.

As described above, the present invention makes it possible to automatically generate a test case for the interface test of a device driver.

In addition, the present invention can automatically identify the test case covering the interface of the device driver, and can add the function for automatically determining the pass/fail of the test input symbol to the test case, thereby making it possible to automatically determine the test pass/fail when the generated test case is executed.

The above-disclosed subject matter is to be considered illustrative, and not restrictive, and the appended claims are intended to cover all such modifications, enhancements, and other embodiments, which fall within the true spirit and scope of the present invention. Thus, to the maximum extent allowed by law, the scope of the present invention is to be determined by the broadest permissible interpretation of the following claims and their equivalents, and shall not be restricted or limited by the foregoing detailed description.

›Tables in the description — 3
TABLE 1
SU kSU dSU pHU dHU iPattern
——get_con_displayRAM—HPI 1
copy_from_user( )fb_write( )—RAM—OPI 3
TABLE 2
Interface Test CharacteristicsInterface Symbols
MemoryMemoryHPI 1 — 1 Static MemoryThe address and size of a
MapMap Boundary Check.symbol that is statically
BoundaryTest whether a virtualallocated in a LCD device
Checkmemory area of a staticdriver.
allocation variable is
allocated to a physical
memory map boundary.
MemoryHPI 1 — 2 Kernel StackThe type, address and
CollisionCollision Test.size of a symbol that is
Test whether a staticstatically allocated in a
allocation variableLCD device driver.
collides with a kernel
stack space.
HPI 1 — 3 User StackThe type, address and
Collision Test.size of a symbol that is
Test whether a staticstatically allocated in a
allocation variableLCD device driver.
collides with a user stack
space.
TimerFrameHPI 2 — 1 Frame BufferThe start and return times
BufferResponse Time Test.of fb_mmap( ) of a LCD
PerformanceTest frame bufferdevice driver.
performance.
InterruptInterruptOPI 3 — 1 Interrupt ContextThe called address of
HandlerRecovery Test._irq_svc, stmia and Idmia
Test whether the contextof do_irq( ).
before an interrupt
service is correctly
recovered.
OPI 3 — 2 Interrupt LatencyThe called address of
Test._irq_svc, stmia and Idmia
Test service performanceof do_irq( ).
measurements until
recovery after the
occurrence of an
interrupt.
InterruptOPI 3 — 3 Normal InterruptThe called address of
ServiceService Routine Test.do_irq( ).
RoutineTest whether a
corresponding service
routine is called for a
generated interrupt.
OPI 3 — 4 Abnormal InterruptThe called address of
Service Routine Test.do_irq( ).
Test whether a service
routine corresponding to
an excessive interrupt is
correctly called.
MemoryMemoryOPI 3 — 5 Dynamic MemoryThe address and size of a
ManagementMapMap Boundary Check.symbol that is dynamically
BoundaryTest whether a virtualallocated in a LCD device
memory area of adriver through kmalloc( ).
dynamic allocation
variable is allocated to a
physical memory map
boundary.
MemoryOPI 3 — 6 Kernel StackThe called addresses of
CollisionCollision Test.get_user( ),
Test whether a dynamic_arch_copy_from_user( ),
allocation variableand
collides with a kernel_arch_strncpy_from_user( )
stack space.in a LCD device driver.
OPI 3 — 7 User StackThe called addresses of
Collision Test.put_user( ),
Test whether a dynamic_arch_clear_user( ), and
allocation variable_arch_copy_to_user( ) in a
collides with a user stackLCD device driver.
space.
OPI 3 — 8 Heap CollisionThe called addresses of
Test.memset( ), _memzero( ),
Test whether a dynamicand memcpy( ) in a LCD
allocation variabledevice driver.
collides with a memory
area of a different
variable.
MemoryOPI 3 — 9 Memory leakageThe called address of
FreeTest.kfree( ) in a LCD device
Test whether adriver.
dynamically allocated
memory area is correctly
released.
OPI 3 — 10 Memory FreeThe called addresses of
Latency Test.kfree( ) and
Test of memory releasekmem_cache_free_one( )
performancein a LCD device driver.
measurements.
Page FaultOPI 3 — 11 Mis-Align Check.The virtual addresses of
Test address alignment.symbols in an init section,
a code section, and a data
section.
CriticalOPI 3 — 12 Critical RegionThe addresses of msr,
RegionTest.orr/bic, and mrs in a LCD
Test whether a criticaldevice driver.
region is protected from
FIQ.
TABLE 3
Expected OutputPass/Fail Determination
Interface SymbolsReference LocationsMethod
HPI 1 — 1Symbol address andMemory map containingCompare the sum of
symbol sizephysical memorysymbol address and
informationsymbol size with physical
memory address
HPI 1 — 2Symbol address andR13_SVC register forCompare the sum of
symbol sizemonitoring kernel stacksymbol address and
symbol size with kernel
stack
HPI 1 — 3Symbol address andR13_USR register forCompare the sum of
symbol sizemonitoring user stacksymbol address and
symbol size with user stack
HPI 2 — 1Start point of fb_mmap,Variable for measuringMeasure the time from start
end point of fb_mmapthe time from start ofto end of fb_mmap
fb_mmap to end of
fb_mmap
OPI 3 — 1Call address of stmia andR0~R12 registerCompare R0~R12 register
idmia of do_IRQ accordingcontaining contextvalue of stmia with R0~R12
to _irq_svcinformationregister value of Idmia
OPI 3 — 2Call address of stmia andVariable for measuringMeasure the time from
idmia of do_IRQ accordingthe time from stmia callstmia call to Idmia call
to _irq_svcto Idmia call
OPI 3 — 3Call address of do_IRQR0 register for obtainingCheck ISR execution
IRQ number after callsuitable for IRQ number
obtained from R0
OPI 3 — 4Call address of do_IRQR0 register for obtainingCheck ISR execution
IRQ number after callsuitable for IRQ number
obtained from R0
OPI 3 — 5Call address of kmalloc( )R0 register for obtainingObtain the size of kmalloc( )
size after call
Memory map containingCompare the sum of call
physical memoryaddress and size with
informationphysical memory address
OPI 3 — 6Call addresses ofR2 register for obtainingObtain the size of
_arch_copy_from_user( ),size after call_arch_copy_from_user( ),
_arch_strncpy_from_user( )_arch_strncpy_from_user( ),
and get_user( )get_user( )
R13_SVC register forCompare the sum of call
monitoring kernel stackaddress and size with
kernel stack
OPI 3 — 7Call addresses ofR2 register for obtainingObtain the size of
copy_to_user( ), put_user( )size after callcopy_to_user( ), put_user( ),
and _arch_clear_user( )_arch_clear_user( )
R13_USR register forCompare the sum of call
monitoring user stackaddress and size with user
stack
OPI 3 — 8Call addresses ofR2 register for obtainingObtain the size of
memset( ), -memzero( ) andsize after callmemset( ), _memzero( ),
memcpy( )memcpy( )
R1 register forCompare the sum of call
monitoring destinationaddress and size with
symbol address in heapdestination symbol address
space
OPI 3 — 9Call address of kfree( )R0 register for obtainingCheck whether RO value is
release address after0x0000000 and check
callwhether it is release for
allocated area
OPI 3 — 10Call addresses ofVariable for measuringMeasure the time from
kmem_cache_free_one( )the time from kfree( ) callkfree( ) to
and kfree( )tokmem_cache_free_one( )
kmem_cache_free_one( )
call
OPI 3 — 11Virtual addresses ofLast bit in each virtualCheck whether last bit of
symbols in init section,addresseach virtual address is 0, 4,
code section and data8, C
section
OPI 3 — 12Addresses of mrs, orr, msrFIQ bit of currentCheck whether CPSR FIQ
and mrs, bic, msrprogram status registerbit value is 1 or 0 at the call
of mrs, orr, msr and mrs,
bic, msr

Claims as granted

13 claims

Log in to read the claims of this application.

Log in to unlock

Classifications

5 codes
IPC · International Patent Classification
Section G — Physics
  • G06F9/44
USPC · US Patent Classification
717/124717/126717/135717/125

Claim changes

Soon
Coming soonHow the claims changed between publication and grant

See which claims were amended, added or cancelled during examination, with every added and removed word marked.

AmendedAddedCancelledUnchanged

The published claims of this application are not paired with the granted ones in what we hold.

File wrapper

⤢ drag to zoomJan 2008Jul 2008Jan 2009Jul 2009Jan 2010Jul 2010Jan 2011Jul 2011Jan 2012Jul 2012USPTOApplicantNon-final rejectionResponse after non-final
USPTOApplicanthover for detail · click to open
Pendency
4.6 y
1,679 days filing → grant
Office actions
1
non-final + final
Responses
1
no RCE
Interviews
1
examiner interview summaries
Examiner
Thuy Dao
art unit 2192 · TC 2100
Citations: 37 back · 1 forward

See the full prosecution history — every USPTO and applicant action on this file, in order.

Log in to unlock

Documents

Log in to open the documents of this file: the application as filed, every office action and response, the notice of allowance.

Log in to unlock

Chain of title

⤢ drag to zoom20082010201220142016201820202022202420262028Owner 1
Titlehover for detail · click to open

See the full assignment history — every owner this patent has passed through, with recordation dates and reel/frame numbers.

Log in to unlock