Fast firmware loading method and circuit for photoelectric DSP chip and optical module

Abstract

The invention provides a firmware rapid loading method and circuit of a photoelectric DSP chip and an optical module.The method is applied to a DSP unit, the DSP unit is used for being installed in the optical module and executing a starting program, the DSP unit is electrically connected with an MCU of the optical module and obtains loading information of the MCU, and the step of obtaining the loading information of the MCU comprises the substeps that first information is obtained, acquiring second information according to the first information; the file size of the first information is smaller than that of the second information, and the first information comprises communication parameter information used for obtaining the second information; the second information comprises target firmware of the DSP unit. On the basis of saving the hardware cost, the loading speed of the DSP chip firmware is improved, the development work of the control program of the MCU is reduced, and the method has good compatibility.

Description

Technical Field

[0001] The present invention relates to the technical field of optical communications, and in particular to a method, circuit and optical module for fast loading firmware of an optoelectronic DSP chip. Background Art

[0002] To reduce the manufacturing cost of optical modules, most optical module manufacturers typically do not configure independent non-volatile memory (such as Flash) for digital signal processing (DSP) chips. Instead, they integrate the firmware required by the DSP chip into the same Flash memory used by the optical module's microcontroller unit (MCU). Consequently, the DSP chip's boot process relies on the MCU reading the DSP firmware data from its shared Flash memory and transmitting it to the DSP chip via a specific communication interface. While this solution consolidates storage resources and controls costs, it also introduces technical challenges regarding firmware transmission speed and program compatibility. Therefore, optimizing firmware transmission speed and program compatibility has become a key issue that DSP chip manufacturers continue to focus on and urgently need to address.

[0003] One existing solution relies on the DSP chip's boot read-only memory (BootROM) to download firmware via low-speed protocols such as I2C / UART. However, this solution suffers from slow speed. Optical modules typically have limited I2C interfaces, so GPIO pins are often used to simulate I2C communication. To ensure communication accuracy, the maximum communication rate is 400 kHz. Downloading a 1MB file takes 20 seconds, which is too long to meet the DSP chip's fast startup requirements.

[0004] Another existing solution relies on the DSP chip's BootROM to download firmware via high-speed SPI / USB. Although this solution achieves faster firmware transmission speeds based on SPI / USB, different DSP chips have different download formats. When using different DSP chips, the optical module's MCU program must be developed for the proprietary transmission protocol of the DSP chip's BootROM. This requires a significant or even complete rewrite of the existing MCU's communication program, resulting in poor compatibility. Summary of the Invention

[0005] The first purpose of the present invention is to provide a method for fast loading firmware of an optoelectronic DSP chip, so as to solve the problems of slow loading speed and poor compatibility of the existing optoelectronic DSP chip.

[0006] The second object of the present invention is to provide a circuit for implementing the method for fast loading firmware of the above-mentioned optoelectronic DSP chip.

[0007] A third object of the present invention is to provide an optical module comprising a firmware fast loading circuit of the above-mentioned optoelectronic DSP chip.

[0008] In order to achieve the above-mentioned first purpose, the present invention provides a method for quickly loading firmware of an optoelectronic DSP chip, which is applied to a DSP unit. The DSP unit is used to be installed in an optical module, execute a startup program, and the DSP unit is electrically connected to the MCU unit of the optical module, and obtains loading information of the MCU unit, wherein: obtaining the loading information of the MCU unit includes: obtaining first information and obtaining second information based on the first information; the file size of the first information is smaller than the file size of the second information, and the first information includes communication parameter information for obtaining the second information; the second information includes the target firmware of the DSP unit.

[0009] As can be seen from the above scheme, the DSP unit of the present invention first downloads a smaller first message from the MCU unit, and then downloads a larger second message from the MCU unit based on the first message, thereby achieving rapid loading of the required target firmware. Because the present invention encapsulates the secondary development of the DSP unit's proprietary download protocol within the first message, the MCU unit does not need to worry about the specific implementation of the DSP unit's download format. When the DSP unit needs to be replaced, the control program within the MCU unit only needs to be fine-tuned based on the first message corresponding to the replaced DSP unit (for example, changing the clock rate, pin configuration, firmware storage address, etc.), rather than making major adjustments to the MCU control program (for example, adjusting the file format, transmission sequence, control commands, etc. based on the proprietary download protocol for the replaced DSP unit). This significantly reduces the development work of the MCU control program and improves the compatibility of the MCU control program. Furthermore, because the MCU unit can directly transmit the second message containing the entire target firmware to the DSP unit at once without worrying about the specific content of the target, the MCU unit's transmission interface can be simplified.

[0010] A further solution is that the first information includes an information part and an execution part; the second information includes firmware parsing information; the startup program also includes: after obtaining the first information from the MCU unit, loading the first information; determining the communication parameter information based on the information part, and jumping to the execution part; the execution part includes: controlling communication with the MCU unit based on the communication parameter information, and obtaining the second information from the MCU unit; determining the firmware parsing information from the second information, loading the target firmware in the firmware storage area based on the firmware parsing information, and jumping to the firmware storage area.

[0011] It can be seen from this that the firmware parsing information used to parse the target firmware and the target firmware can be encapsulated in the second information and transmitted to the DSP unit at one time, so that the DSP unit will store the received second information in a temporary cache area (for example, a preset address in the SRAM of the DSP unit), and then obtain the firmware parsing information from the second information, and then load the target firmware in the temporary cache area at the loading address indicated by the firmware parsing information according to the firmware information.

[0012] A further solution is that the first information includes an information part and an execution part; the second information includes firmware parsing information; the startup program also includes: after obtaining the first information from the MCU unit, loading the first information; determining the communication parameter information based on the information part, and jumping to the execution part; the execution part includes: controlling communication with the MCU unit based on the communication parameter information, and obtaining the second information from the MCU unit; determining the firmware parsing information from the second information, loading the target firmware in the firmware storage area based on the firmware parsing information, and jumping to the firmware storage area.

[0013] It can be seen that the present invention encapsulates the firmware parsing information used to parse the target firmware in the first information, so that the startup program can parse the firmware parsing information from the information part before obtaining the second information. Therefore, when the DSP unit obtains the target firmware of the second information, it can directly load the target firmware according to the firmware parsing information. Since there is no need to load it in the temporary cache area, the loading speed of the target firmware can be further improved.

[0014] A further solution is that the startup program includes determining verification information based on the information part; after obtaining the target firmware and before jumping to the firmware storage area, it includes: verifying the target firmware based on the verification information, jumping to the firmware storage area after the verification passes, and resetting if the verification fails.

[0015] It can be seen that verification can be used to verify the sender of the target firmware and protect the target firmware.

[0016] A further solution is that the startup program includes determining decompression information according to the information part; after obtaining the target firmware and before jumping to the firmware storage area, it includes: decompressing the target firmware based on the decompression information.

[0017] This shows that the storage space of the MCU unit can be saved, and when the decompression speed is greater than the transmission speed, the speed of firmware loading can be further improved.

[0018] A further solution is to perform an integrity check on the first information before obtaining the second information based on the first information.

[0019] This shows that it is possible to ensure that the complete first information is received.

[0020] A further solution is that when obtaining the first information from the MCU unit, the first information is obtained from the MCU unit through a first preset communication method; when obtaining the target firmware from the MCU unit, the target firmware is obtained through a second preset communication method based on the communication parameter information; the signal transmission rate when applying the second preset communication method is greater than the signal transmission rate when applying the first preset communication method.

[0021] It can be seen from this that through the existing different communication methods, the smaller first information can be downloaded with a low-speed clock, and then the larger DSP firmware can be downloaded with a high-speed clock. Since the first information is small, it can be conveniently transmitted quickly using multiple existing communication methods, reducing the requirements for the transmission rate of the communication method, facilitating implementation and reducing development work.

[0022] In order to achieve the above-mentioned second purpose, the present invention provides a firmware fast loading circuit for an optoelectronic DSP chip, which includes: a DSP unit and an MCU unit, the DSP unit being connected to the MCU unit; the DSP unit including a boot read-only memory and a random access memory, the boot read-only memory storing a boot program; the MCU unit including a non-volatile memory, the non-volatile memory storing first information and second information; the DSP unit implements the above-mentioned firmware fast loading method, wherein the first information and second information obtained by the DSP unit are loaded into the random access memory.

[0023] It can be seen from the above scheme that the present invention uses the MCU unit to store the target firmware required for starting the DSP unit, and uses the first information to download the second information containing the target firmware to the DSP unit. While saving hardware costs, it can also ensure the transmission speed of the second information and the compatibility of the control program of the MCU unit.

[0024] A further solution is that the DSP unit further includes a one-time programmable memory, in which a root key for decrypting the target firmware is stored.

[0025] This shows that the received target firmware can be easily decrypted.

[0026] In order to achieve the third objective mentioned above, the present invention provides an optical module, comprising a housing, wherein the housing includes a firmware fast loading circuit for the optoelectronic DSP chip mentioned above.

[0027] It can be seen from the above scheme that the optical module provided by the present invention enables the user to develop the control program of the MCU unit without having to develop it according to the private download protocol of the DSP unit, but instead develop it according to the first information of the provided DSP unit. Since there is no need to pay attention to the specific implementation of the private download protocol, the user's MCU unit control program development work can be reduced, and it is convenient for the user to replace different types of DSP units. It has good compatibility and improves product competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a circuit framework diagram of an embodiment of a firmware fast loading circuit for an optoelectronic DSP chip of the present invention.

[0029] Figure 2The figure is a flow chart of an embodiment of a method for fast loading firmware of an optoelectronic DSP chip according to the present invention.

[0030] Figure 3 yes Figure 2 Specific flow chart of step S6 in FIG.

[0031] The present invention will be further described below with reference to the accompanying drawings and embodiments. DETAILED DESCRIPTION

[0032] After the DSP unit of the present invention is powered on, it first obtains the first information from the MCU unit, and then obtains the second information containing the target firmware from the MCU unit through the first information, loads the target firmware into the firmware storage area, and jumps to the firmware storage area for execution to complete the fast startup. In this process, the MCU unit does not need to care about the specific implementation of the DSP chip download format.

[0033] Implementation example of a fast firmware loading circuit for an optoelectronic DSP chip: See also Figure 1 This embodiment includes a DSP unit 1 and an MCU unit 2. The DSP unit 1 includes a boot read-only memory 11, a random access memory 12, and a one-time programmable memory 13. The MCU unit 2 includes a non-volatile memory 21. The DSP unit 1 is electrically connected to the MCU unit 2.

[0034] The startup ROM 11 stores a startup program. The startup program is a program that is executed after the DSP unit 1 is powered on or reset. When the DSP unit 1 leaves the factory, the startup program is fixed in the startup ROM 11 and cannot be changed. In this embodiment, the startup ROM 11 is specifically a BootROM.

[0035] The startup program is used to obtain first information from the MCU unit 2, load the first information into the random access memory 12, perform integrity check and analysis on the obtained first information, and jump to the execution part of the first information after passing the integrity check and analysis.

[0036] The random access memory 12 is used to store temporary data of the DSP unit 1, including storing the first information and the second information obtained by the DSP unit 1 from the MCU unit 2. In this embodiment, the random access memory 12 is specifically an SRAM.

[0037] The one-time programmable memory 13 is used to store a key for decrypting the target firmware and a key for verifying the signature of the target firmware. In this embodiment, the one-time programmable memory 13 is specifically an eFuse.

[0038] The first information and the second information are stored in the non-volatile memory 21. In this embodiment, the non-volatile memory 21 is specifically Flash.

[0039] When the first information is executed as a downloader program in the DSP unit 1, the second information is downloaded from the MCU unit 2. The second information includes the target firmware, so that the target firmware is loaded and executed in the non-volatile memory 21. The target firmware is the firmware required to be loaded when the DSP unit 1 starts.

[0040] The first information includes an information portion and an execution portion. In this embodiment, the information portion and the execution portion are arranged in chronological order. The information portion is the file header, and the execution portion is the file body. The file header includes metadata describing the file body. The file body includes code to implement DSP firmware download. By parsing the file header, basic attribute information, firmware parsing information, communication parameter information, verification information, and decompression information can be obtained. The parsed firmware parsing information, communication parameter information, verification information, and decompression information are loaded into random access memory 12 for use when executing the file body.

[0041] The basic attribute information describes the version information and checksum, and the checksum is used to implement integrity verification of the file header and file body.

[0042] The firmware parsing information describes information about the DSP firmware to be received, including the size of the DSP firmware and the address at which the DSP firmware is loaded into the random access memory 12. The DSP firmware may include multiple sub-firmwares, the size of the DSP firmware includes the size of each sub-firmware, and the address at which the DSP firmware is loaded includes the address at which each sub-firmware is loaded. Using the firmware parsing information, the DSP unit 1 can parse the received target firmware and load the target firmware into the random access memory 12.

[0043] The communication parameter information describes the communication parameters that need to be configured when the DSP unit 1 downloads the target firmware from the MCU unit 2. In this embodiment, the communication parameter information includes that the communication mode is SPI communication and that the DSP unit 1, as the host, provides the MCU unit 2 with the required SPI clock information.

[0044] The verification information describes whether verification is required for the received target firmware and the attribute information corresponding to the required verification items when verification is required. In this embodiment, the verification items include signature verification items and decryption items.

[0045] The signature verification item attributes include signature verification enablement information, signature verification algorithm information, and key information required for signature verification. The signature verification enablement information describes whether signature verification is required for the received target firmware. The signature verification algorithm information describes the signature verification algorithm used when verifying the received target firmware. The key information required for signature verification specifies the address of the signature verification key corresponding to the currently used signature verification algorithm in the one-time programmable memory 13.

[0046] The decryption item's attribute information includes decryption enable information, decryption algorithm information, and decryption key information. The decryption enable information describes whether the received target firmware needs to be decrypted. The decryption algorithm information describes the decryption algorithm used to decrypt the received target firmware. The decryption key information describes the address of the decryption root key corresponding to the currently used decryption algorithm in the one-time programmable memory 13.

[0047] The decompression information describes whether the received target firmware needs to be decompressed, and if so, the decompression algorithm to use. If the target firmware is large, it can be pre-compressed in MCU unit 2. DSP unit 1 then retrieves the compressed target firmware and determines the appropriate decompression algorithm based on the decompression information for local decompression. This saves storage space in MCU unit 2 and further speeds up the loading of the target firmware in DSP unit 1.

[0048] When jumping to the file body for execution, the target firmware is obtained from the MCU unit 2 according to the communication parameter information, and possible verification operations and decompression operations are performed on the target firmware based on the specific verification information and decompression information. Then, the target firmware is loaded into the random access memory 12 based on the firmware parsing information. The area where the target firmware is stored in the random access memory 12 is the firmware storage area. Finally, the program jumps to the firmware storage area to load the target firmware and complete the rapid loading of the firmware of the DSP chip.

[0049] In different embodiments, the DSP unit 1 may be integrated on one chip, and the MCU unit 2 may be integrated on another chip.

[0050] In different embodiments, the above-mentioned DSP unit 1 includes an independent DSP chip, which is responsible for the execution of the specific startup program, the first information and the target firmware. The read-only memory 11, the random access memory 12, and the one-time programmable memory 13 are electrically connected to the DSP chip as discrete components; the above-mentioned DSP unit 1 includes an independent MCU chip, which is electrically connected to the DSP chip to realize data transmission, and the non-volatile memory 21 is electrically connected to the MCU chip as a discrete component.

[0051] Embodiment of the method for fast loading firmware of optoelectronic DSP chip: This embodiment is implemented based on the firmware fast loading circuit of the above-mentioned optoelectronic DSP chip.

[0052] See also Figure 2 This embodiment is implemented by executing a computer program, and specifically includes the following steps: S1: Obtain first information from the MCU unit through a first preset communication method and load the information into a random access memory.

[0053] The first preset communication mode of this embodiment is the I2C communication mode, the file size of the first information is 3 KB, and it is transmitted at a rate of 100 Kbps.

[0054] It should be noted that, since the file size of the first information is relatively small (generally less than 4KB), which is generally much smaller than the file size of the second information (about 500KB), the I2C communication method can ensure fast and accurate transmission of the first information.

[0055] In other embodiments, the first preset communication mode may also be UART or JTAG.

[0056] S2: Determine whether the first information passes the integrity check.

[0057] In this embodiment, the integrity of the file header and file body is checked using a CRC32 checksum, combined with the checksum provided by the file header of the first information. If the integrity check fails, a problem has occurred in the transmission of the first information, and step S9 is executed to reset the process, returning to step S1. If the integrity check passes, the first information has been transmitted intact from the MCU unit to the DSP unit, and step S3 is then continued.

[0058] S3: Parse the file header of the first information.

[0059] The firmware parsing information, communication parameter information, verification information, and decompression information obtained from the file header are loaded into a random access memory for subsequent steps.

[0060] S4: Jump to the file body of the first information.

[0061] S5: Obtain second information from the MCU unit through a second preset communication method.

[0062] The signal transmission rate of the second preset communication mode is greater than that of the first preset communication mode. In this embodiment, the second preset communication mode is an SPI communication mode, and SPI communication with the MCU unit is implemented based on the communication parameter information. The DSP unit acts as a host and the MCU unit acts as a slave.

[0063] In other embodiments, according to different communication parameter information, the second preset communication mode may also be MDIO or USB.

[0064] S6: Determine whether the target firmware can be executed normally.

[0065] By determining whether the target firmware in the second information is executable, the execution of the wrong target firmware can be avoided. If it is determined that it is not executable, it indicates that there is a problem with the target firmware itself or the transmission process, and step S9 needs to be executed to perform a reset operation and return to step S1. If it is determined that it is executable, step S7 is continued.

[0066] S7: Load the target firmware into the firmware storage area of ​​the random access memory.

[0067] The target firmware is loaded into a firmware storage area of ​​a random access memory according to the firmware parsing information.

[0068] S8: Jump to the firmware storage area.

[0069] Among them, according to the firmware parsing information, jump to the first address of the firmware storage area for execution.

[0070] Continue to see Figure 3 , Figure 3 The specific process of the above step S6 is shown.

[0071] First, step S61 is executed to determine whether signature verification is required. In particular, whether signature verification is required for the received target firmware is determined based on the signature verification enabling information. If yes, step S62 is continued; if not, step S63 is jumped to.

[0072] When it is determined that signature verification is required according to the signature verification enabling information, step S62 is executed to determine whether the target firmware has passed the ECC verification. After it is determined that signature verification is required according to the signature verification enabling information, the signature verification algorithm to be used is determined according to the signature verification algorithm information, and the signature verification key to be used is determined according to the key information required for signature verification. This embodiment uses the ECC algorithm for signature verification, and obtains the corresponding signature verification key ECC_PUBLIC_KEY in the one-time programmable memory. The signature verification operation can ensure that the target firmware has not been tampered with by the middleman. When the target firmware passes the ECC verification, step S63 is continued; if the target firmware does not pass the ECC verification, jump to step S9.

[0073] If the target firmware passes ECC verification, step S63 is executed to determine whether decryption is required. The decryption enable information is used to determine whether the received target firmware needs to be decrypted. If decryption is required, step S64 is executed to obtain the key combination corresponding to the target firmware based on the EFUSE decryption root key and decrypt the target firmware. If not, the process jumps to step S65.

[0074] When decryption is required, the decryption algorithm to be used is determined based on the decryption algorithm information. The decryption root key corresponding to the decryption algorithm in the one-time programmable memory is determined based on the key information required for decryption. This embodiment uses the SM4 decryption algorithm. The decryption algorithm information includes a key derivation hierarchy (specifically, three levels) and three levels of CONTENT input (C, D, and E). SM4 decryption is performed on C, D, and E in sequence using the decryption root key SM4_KEY corresponding to the one-time programmable memory to obtain the final firmware decryption key. The target firmware is then decrypted using SM4 using the final firmware decryption key. This encryption and decryption of the target firmware protects the target firmware from being leaked.

[0075] Then, step S65 is continued to determine whether decompression is required. Based on the decompression information, it is determined whether the received target firmware needs to be decompressed. If decompression is required, step S66 is continued to decompress the target firmware, wherein the target firmware is decompressed according to the decompression algorithm indicated by the decompression information. If not, the process jumps to step S7.

[0076] Optical module example: The optical module of this embodiment includes a housing, and the housing includes a firmware fast loading circuit of the optoelectronic DSP chip of the above embodiment.

[0077] In summary, the present invention uses a two-stage startup scheme to shorten firmware startup time. First, a smaller first message is loaded via a low-speed bus, and then a larger second message containing the DSP firmware is downloaded via a high-speed bus. The total startup time can be compressed to less than 0.5 seconds. The present invention eliminates the need for adding non-volatile memory to store firmware in the DSP chip, reducing hardware costs. Furthermore, the MCU program does not need to worry about the specific implementation of the DSP chip download format, eliminating compatibility issues. The MCU program can transfer the entire DSP firmware to the DSP chip at once, without having to worry about the specific content of each submodule at each level of the DSP firmware, simplifying its transmission interface.

[0078] Finally, it should be emphasized that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various changes and modifications. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for fast firmware loading for an optoelectronic DSP chip, applied to a DSP unit installed in an optical module, executing a startup program, electrically connected to an MCU unit of the optical module, and obtaining loading information from the MCU unit, characterized in that: Obtaining loading information of the MCU unit includes: obtaining first information and obtaining second information based on the first information; the file size of the first information is smaller than the file size of the second information, and the first information includes communication parameter information for obtaining the second information; the second information includes the target firmware of the DSP unit.

2. The method for fast firmware loading of an optoelectronic DSP chip according to claim 1, wherein: The first information includes an information part and an execution part; the second information includes firmware parsing information; The startup procedure includes: After acquiring first information from the MCU unit, loading the first information; Determine the communication parameter information according to the information part, and jump to the execution part; The execution part includes: controlling communication with the MCU unit based on the communication parameter information, and obtaining the second information from the MCU unit; The firmware parsing information is determined from the second information, the target firmware is loaded into a firmware storage area based on the firmware parsing information, and the target firmware is jumped to the firmware storage area.

3. The method for fast firmware loading of an optoelectronic DSP chip according to claim 1, wherein: The first information includes an information part and an execution part; The startup procedure includes: After acquiring first information from the MCU unit, loading the first information; Determine the firmware parsing information and the communication parameter information according to the information part, and jump to the execution part; The execution part includes: controlling communication with the MCU unit based on the communication parameter information, and obtaining the second information from the MCU unit; The target firmware is loaded into a firmware storage area based on the firmware parsing information, and the target firmware is jumped to the firmware storage area.

4. The method for fast firmware loading of an optoelectronic DSP chip according to claim 3, wherein: The initiation procedure includes determining verification information based on the information portion; After the target firmware is acquired and before the target firmware is jumped to the firmware storage area, the method includes: verifying the target firmware based on the verification information, jumping to the firmware storage area after the verification passes, and resetting the target firmware if the verification fails.

5. The method for fast loading firmware of an optoelectronic DSP chip according to claim 3, wherein: The initiation procedure includes determining decompression information based on the information portion; After the target firmware is acquired and before jumping to the firmware storage area, the method includes: performing a decompression operation on the target firmware based on the decompression information.

6. The method for fast firmware loading of an optoelectronic DSP chip according to claim 1, wherein: Before obtaining the second information based on the first information, perform an integrity check on the first information.

7. The method for fast firmware loading of an optoelectronic DSP chip according to any one of claims 1 to 5, characterized in that: When acquiring the first information from the MCU unit, acquiring the first information from the MCU unit through a first preset communication method; When acquiring the second information from the MCU unit, acquiring the second information from the MCU unit through a second preset communication method based on the communication parameter information; The signal transmission rate when the second preset communication mode is applied is greater than the signal transmission rate when the first preset communication mode is applied.

8. A fast firmware loading circuit for an optoelectronic DSP chip, characterized in that: include: A DSP unit and an MCU unit, wherein the DSP unit is connected to the MCU unit; The DSP unit includes a startup read-only memory and a random access memory, wherein the startup read-only memory stores a startup program; The MCU unit includes a non-volatile memory, wherein the non-volatile memory stores first information and second information; The DSP unit implements the method for fast loading firmware of the optoelectronic DSP chip according to any one of claims 1 to 7, wherein the first information and the second information acquired by the DSP unit are loaded into the random access memory.

9. The fast firmware loading circuit for an optoelectronic DSP chip according to claim 8, wherein: The DSP unit further includes a one-time programmable memory storing a root key for decrypting the second information.

10. An optical module, comprising a housing, characterized in that: The housing includes a fast firmware loading circuit for the optoelectronic DSP chip according to claim 8 or 9.

Images