<div>Hi there,</div>
<div> </div>
<div> I met some problems when I tried to debug the 2.5.16 kernel with BDI2000 for a customerized MPC8xx. </div>
<div>I listed my two trials, can any one tell me what is wrong with my approachs,</div>
<div> </div>
<div><font size="2">
<p>[Try 1] =============================================></p>
<p>(1) Build the kernel image, the map file -- System.map file</p>
<p>.....</p>
<p>c0000000 T _start</p>
<p>c0000000 T __start</p>
<p>c0000000 T _stext</p>
<p>c0000018 t turn_on_mmu</p>
<p>c0000100 t Reset</p>
<p>..... </p>
<p>(2) Copy the vmlinux into /tftpboot for load the iamge</p>
<p>(3) I mapped the logic address 0xc0000000 into physical address 0x00100000 in target </p>
<p>by loading the file through BDI telnet console, the telnet screen shows</p>
<p>BDI>load 0x40100000 vmlinux elf</p>
<p>Loading vmlinux , please wait ....</p>
<p>- File offset 0x00010000 to address 0x00100000 size 1728646</p>
<p>Loading program file passed</p>
<p> </p>
<p>(4) Start the GDB command, and single step from address 0x00100000. </p>
<p>I tried two version of ppc gdb, both have the same result. The location of one of </p>
<p>the gdb (through DENX) is,</p>
<p>$ which ppc_8xx-gdb</p>
<p>/opt/eldk/usr/bin/ppc_8xx-gdb</p>
<p> </p>
<p>(5) And start the gdb through the following command,</p>
<p>$ ppc_8xx-gdb vmlinux (under kernel top directory)</p>
<p>(gdb) target remote <a href="http://10.199.5.234:2001">10.199.5.234:2001</a></p>
<p>Remote debugging using <a href="http://10.199.5.234:2001">10.199.5.234:2001</a></p>
<p></p>
<p>(gdb) si</p>
<p>0x0010000c in ?? ()</p>
<p>(gdb)</p>
<p>0x00100010 in ?? ()</p>
<p>(gdb)</p>
<p>0x00100014 in ?? ()</p>
<p>....</p>
<p></p>
<p>Things go well till I reach the </p>
<p></p>
<p>(gdb)</p>
<p>0x00100030 in ?? ()</p>
<p>(gdb)</p>
<p>Program received signal SIGSTOP, Stopped (signal).</p>
<p>0x00100030 in ?? ()</p>
<p>(gdb)</p>
<p>Program received signal SIGSTOP, Stopped (signal).</p>
<p>0x00100030 in ?? ()</p>
<p>(gdb)</p>
<p> </p>
<p>If I check the code carefully in /arch/ppc/kernel_8xx.S, 0x00100030 address </p>
<p>matches the rfi comamnd in </p>
<p>turn_on_mmu:</p>
<p>mfmsr r0</p>
<p>ori r0,r0,MSR_DR|MSR_IR</p>
<p>mtspr SPRN_SRR1,r0</p>
<p>lis r0,start_here@h</p>
<p>ori r0,r0,start_here@l</p>
<p>mtspr SPRN_SRR0,r0</p>
<p>SYNC</p>
<p>rfi /* enables MMU */</p>
<p> </p>
<p>(6) I check the BDI configuration (8xx.cfg) file, seems I had enabled the MMI </p>
<p> </p>
<p>[TARGET]</p>
<p>CPUCLOCK 32000000 ;the CPU clock rate after processing the init list</p>
<p>BDIMODE AGENT ;the BDI working mode (LOADONLY | AGENT)</p>
<p>BREAKMODE SOFT ;SOFT or HARD, HARD uses PPC hardware breakpoints</p>
<p>;STEPMODE HWBP ;TRACE or HWBP, HWPB uses one or two hardware breakpoints</p>
<p>;STARTUP STOP 5000 ;let the monitor initialize the system</p>
<p>;DCACHE FLUSH ;flush data cache, needs a workspace of 32 bytes</p>
<p>;WORKSPACE 0x00000040 ;workspace in target RAM for data cache flush</p>
<p>MMU XLAT ;translate effective to physical address</p>
<p>;PTBASE 0x00000f0 ;????</p>
<p>;REGLIST SPR ;select register to transfer to GDB</p>
<p>REGLIST ALL ;select register to transfer to GDB</p>
<p>I assume I do not need set PTBASE, since I am working on the very new version of kernel</p>
<p>(2.6.15).</p>
<p> </p>
<p>Question, seems to me the MMU in BDI does still map the logic address into physical </p>
<p>address correctly, what did I miss here?</p>
<p>[Try 2 ] ===================================================></p>
<p>Exactly follow the BDI user manual,</p>
<p>(1) load image, BDI telnet screen shows</p>
<p>BDI>load 0x40100000 vmlinux elf</p>
<p>Loading vmlinux , please wait ....</p>
<p>- File offset 0x00010000 to address 0x00100000 size 1728646</p>
<p>Loading program file passed</p>
<p> </p>
<p>(2) set hardware break point </p>
<p>BDI>bi 0xc0000000 0xc00fffff</p>
<p>Breakpoint identification is 0</p>
<p></p>
<p>(3) issue go comamnd from telnet screen</p>
<p>BDI>go</p>
<p>- TARGET: stopped</p>
<p></p>
<p>(4) Disable hardware break point</p>
<p>BDI> ci</p>
<p></p>
<p>(5) Start gdb from another terminal,</p>
<p>$ ppc_8xx-gdb vmlinux</p>
<p></p>
<p>(6) Attached target, </p>
<p>(gdb) target remote <a href="http://10.199.5.234:2001">10.199.5.234:2001</a></p>
<p>Remote debugging using <a href="http://10.199.5.234:2001">10.199.5.234:2001</a></p>
<p>0x00001100 in ?? ()</p>
<p></p>
<p>(7) Set software break point,</p>
<p>(gdb) b start_kernel</p>
<p></p>
<p>(8) let go</p>
<p>(gdb) c</p>
<p>Continuing.</p>
<p>Program received signal SIGTRAP, Trace/breakpoint trap.</p>
<p>0x00001100 in ?? ()</p>
<p></p>
<p>Seems to me the MMU still not work with BDI.</p>
<p>Can any one one point out what is the problem? </p>
<p>Thanks!</p>
<p> </p></font></div>