January 2003 - ixp1200 - lists.cs.princeton.edu

RE: [ixp1200] How to generate asm code from SDK
by Heaton, Eric D 04 Apr '03

04 Apr '03

Yan, This warning should have no effect on the production of an .asm file. Of course, if you had errors in your build, that is a different story. Check your build directory for a file named <c_file_with_main_in_it>.asm if you didn't specify a specific name with the -Fa switch. But note that you can look at the assembly code generated from a Microengine C file easily without using the -Fa switch. When using the Workbench (v2.01) in debug mode, choosing a thread to view will, by default, display the source code for the thread. However, there is a toggle switch on that window (it has a yellow and green page icon circularly pointing to each other) that will display the .list file containing the assembly for the source code line your cursor is currently on. Check it out! --eric h. -----Original Message----- From: Yan Luo [mailto:yluo@cs.ucr.edu] Sent: Sunday, January 12, 2003 12:02 PM To: ixp1200(a)CS.Princeton.EDU Subject: [ixp1200] How to generate asm code from SDK HI, I try to generate the assembly code from Micro C code using SDK 2.0, In the Build->setting, I check the "Produce assembly code file". When I compile , I got the following warning: (0): warning : IPO enabled; -Fa and -FA produce dummy .asm files It seems it did not produce asm file, and I cannot find asm file anywhere. How can I get the assembly code? Thanks, Yan

10 25

Re: [ixp1200] Textbook
by Douglas Comer 29 Jan '03

29 Jan '03

Wu, >> Your book looks great. I plan on using it for the IXP course I'll be >> offering in the Spring. It should be in the warehouse and ready to ship by Friday. >> I was wondering if you planned on setting up an >> e-mail list for questions about the book and its labs or if you expect >> all questions to be directed to the Princeton list. I have indeed set up a mailing list for the book: npbook.cs.purdue.edu However, I hope to use the book list just for conceptual questions, and forward all the detailed IXP questions to the Princeton list. Is that what you expected? Cheers, Doug

6 5

Textbook
by Douglas Comer 28 Jan '03

28 Jan '03

Folks, By now, everone should have received information from the publisher about my latest textbook: Network Systems Design Using Network Processors IXP1200 version Prentice Hall, 2003, ISBN 0-13-141792-4 For those of you using the book this semester: it will ship from the warehouse next Tuesday (the 21st). In the mean time, if you need copies of early chapters, please let me know. I have received a couple of questions about the content, so I put the Table Of Contents on: http://www.cs.purdue.edu/homes/comer/netbooks.html#nsd Cheers, Doug

4 3

Unable to pump data stream into L2L3NAT_8_1Gig_hw example project
by Eng Kean 26 Jan '03

26 Jan '03

Hi, I have done some modification on L2L3NAT_8_1Gig_hw project so that it can be run under simulation mode. Following are changes I have made: 1. Change to Simulation mode 2. Define SIMULATION,BUILDFREELIST for Microengine 5. 3. Uncomment the following statement in EgressInit[]: immed[temp2, 0x00d7] immed[temp1, 0x00db] And comment out the following statements: immed[temp2, 0x00b7] immed[temp1, 0x00bb] 4. Create a script file to initialize SRAM, DRAM, and Scratchpad 5. Set data stream and assign I/O ports to pump in packets. But, I found that data stream could not be pumped into RFIFO. Thus the content of RFIFO are zeros and TFIFO are uninitialized. Anybody have encountered this problem? How to solve it? Thanks. Regards, Eng Kean __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com

1 0

Help with sdram_read_ind function
by Deepa Srinivasan 22 Jan '03

22 Jan '03

Hello, I have an array of pointers in SDRAM (arrayPtrs) that point to other SDRAM locations which contain the actual data. I am trying to read this data using the sdram_read_ind function as follows: //Declared globally __declspec(sdram) data_t* arrayPtrs[DIMS]; .... ..... read_data() { .... sdram_read_write_ind_t ind1; __declspec(sdram_read_reg) data_t *temp; ind1.ctx = 0; ind1.ov_ctx = 0; ind1.reserved = 0; ind1.byte_mask = 0; ind1.ov_byte_mask = 0; ind1.ref_count = 1; ind1.ov_ref_count = 0; ind1.xadd = 0; ind1.abs = 0; ind1.ov_xadd = 0; ind1.ueng_addr = 0; ind1.ov_ueng_addr = 0; sdram_read_ind( temp, arrayPtrs[0], 1, ind1, queue_default, sig_done); ctx_wait(sdram); .... .... } With this code, I get the following compiler error: (for the line of code that contains: "ind1,") error : intrinsic data buffer must be a local variable passed by direct reference error : intrinsic data buffer must be a non-shared variable passed by direct reference What am I missing here? Thanks for any help. ~Deepa

1 0

How to generate asm code from SDK
by Yan Luo 12 Jan '03

12 Jan '03

HI, I try to generate the assembly code from Micro C code using SDK 2.0, In the Build->setting, I check the "Produce assembly code file". When I compile , I got the following warning: (0): warning : IPO enabled; -Fa and -FA produce dummy .asm files It seems it did not produce asm file, and I cannot find asm file anywhere. How can I get the assembly code? Thanks, Yan

1 0

RE: [ixp1200] problems when assembling micro code
by Harikishan V 11 Jan '03

11 Jan '03

I think you have to use the '.segment 1' statement in the microcode and keep the code that exceeds 1024 microwords. You can get the details on how to use this in section 2.5.17 of Microcode Programmers Reference Mannual. regards Hari -----Original Message----- From: Gregory Prier [mailto:greg@Lever.cs.ucla.edu] Sent: Saturday, January 11, 2003 4:52 AM To: ixp mailing list Subject: [ixp1200] problems when assembling micro code I have a relatively large micro code program that I am trying to assemble. I get these error messages: (17009): error : The numeric address value 1035 in field "branch_field" is too large to be represented in field. F:\src\dward\interface_ace\include\SlowIngress_h.uc(509): error : The previous error occurred while expanding macro. This repeats 4 times, and then: (0): error : Page, common_code, exceeds the 1024 address limit. (0): error : Unexpected error occurred during encoding phase. We are using an ixp1200 with a 2k instruction store. As far as we can tell our program should fit into this space. Has anyone else seen this same problem and how did they fix it? Thank you for the help. GKP ---------------------------------------------------------------------- Greg Prier Previously: A Mudder 3564 Boelter Hall Currently: A Bruin working UCLA CA 90095 toward a PhD (310) 825-8899 After that: No idea "Hell, there are no rules here. We're trying to accomplish something." Thomas Edison

1 0

problems when assembling micro code
by Gregory Prier 10 Jan '03

10 Jan '03

I have a relatively large micro code program that I am trying to assemble. I get these error messages: (17009): error : The numeric address value 1035 in field "branch_field" is too large to be represented in field. F:\src\dward\interface_ace\include\SlowIngress_h.uc(509): error : The previous error occurred while expanding macro. This repeats 4 times, and then: (0): error : Page, common_code, exceeds the 1024 address limit. (0): error : Unexpected error occurred during encoding phase. We are using an ixp1200 with a 2k instruction store. As far as we can tell our program should fit into this space. Has anyone else seen this same problem and how did they fix it? Thank you for the help. GKP ---------------------------------------------------------------------- Greg Prier Previously: A Mudder 3564 Boelter Hall Currently: A Bruin working UCLA CA 90095 toward a PhD (310) 825-8899 After that: No idea "Hell, there are no rules here. We're trying to accomplish something." Thomas Edison

1 0

Intel bug in ue debug code?
by Austen McDonald 06 Jan '03

06 Jan '03

Someone from Intel please answer this: Im trying to write a wrapper for the ioctl commands to mimic the kernel level API for the uDbg library. The function Im working w/ is int uDbg_SetNumUword(uDbg_Handle_T dbgH, unsigned int uEngMask, unsigned int uWordAddr, unsigned short numWords, unsigned int *uWord) Inside SA1_CoreLibs/ue/main.c where the ioctl is actually handled, we find this code: case UENG_SETNUMUWORD: { uEng_setNumUword_T d; if (copy_from_user(&d, (void *)arg, sizeof(d))) return -EFAULT; size = d.numWords * sizeof(unsigned int); buf = kmalloc(size, GFP_KERNEL); if (buf == NULL) return -ENOMEM; if (copy_from_user(buf, (void *)d.uWord, size)) { kfree(buf); return -EFAULT; } if (uEng_setNumUword(d.uEng, d.uAddr, d.numWords, &d.uWord)) { printk("ue: UENG_SETNUMUWORD failed\n"); kfree(buf); return -EFAULT; } kfree(buf); break; } and SA1_CoreLibs/debug_1200/debug_1200.hxx defines Debug_SetNumUword_T to be typedef struct Debug_SetNumUword_S{ unsigned int uEngMask; unsigned int uWordAddr; unsigned short numWords; unsigned int uWord; } Debug_SetNumUword_T; But this cant be right---isnt the purpose of buf to copy the data from the POINTER uWord (whereas now its not a pointer, its just and int) AND shouldn't buf be passed to uEng_setNumUword instead of &d.uWord inside the case statement? Let me know if Im an idiot or if this has been fixed in a newer version... Austen McDonald austen(a)cc.gatech.edu God does not care about our mathematical difficulties. He integrates empirically. -- Albert Einstein

1 0

L2L3NAT_8_1Gig_hw memory initialization location
by Eng Kean 03 Jan '03

03 Jan '03

anybody know which SRAM, SDRAM, and Scratchpad memory location should be initialized before running simulation? so that a script to initialize that specific memory location. thanks. __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com

4 3