Santifort, Conor

Amber ARM-compatible core

Hi,

I followed the Amber Open Source Project User Guide to run a simulation (page 7).  However:

$ ./run.sh hello-world
Test hello-world, type 4
make -s -C ../mini-libc MIN_SIZE=1
../tools/check_mem_size.sh boot-loader-serial.mem "@000020"
make -s -C ../mini-libc MIN_SIZE=1
./run.sh: line 239: fuse: command not found

I have installed fuse-2.9.2-11.el7.x86_64. 

Looking forward to your reply

Fail to run simulations

i actually cant able to generate .mem file.kindly can anyone provide any solutions for it.

i am using windows gnu complier for this.does all scripts written will support?

,mem file

root@jayanth:/home/jayanth/proc_paper/amber_down/trunk/hw/isim# ./run.sh hello-world
Test hello-world, type 4
make -s -C ../mini-libc MIN_SIZE=1
arm-none-linux-gnueabi-ld -Bstatic -Map boot-loader-serial.map  --strip-debug --fix-v4bx -o boot-loader-serial.elf -T sections.lds boot-loader-serial.o start.o crc16.o xmodem.o elfsplitter.o ../mini-libc/printf.o ../mini-libc/libc_asm.o ../mini-libc/memcpy.o
arm-none-linux-gnueabi-ld: warning: crc16.o uses 32-bit enums yet the output is to use variable-size enums; use of enum values across objects may fail
arm-none-linux-gnueabi-ld: warning: xmodem.o uses 32-bit enums yet the output is to use variable-size enums; use of enum values across objects may fail
arm-none-linux-gnueabi-ld: warning: elfsplitter.o uses 32-bit enums yet the output is to use variable-size enums; use of enum values across objects may fail
arm-none-linux-gnueabi-objcopy -R .comment -R .note boot-loader-serial.elf
../tools/amber-elfsplitter boot-loader-serial.elf > boot-loader-serial.mem
../tools/amber-memparams32.sh boot-loader-serial.mem boot-loader-serial_memparams32.v
../tools/amber-memparams128.sh boot-loader-serial.mem boot-loader-serial_memparams128.v
arm-none-linux-gnueabi-objdump -C -S -EL boot-loader-serial.elf > boot-loader-serial.dis
../tools/check_mem_size.sh boot-loader-serial.mem "@000020"
make -s -C ../mini-libc MIN_SIZE=1
arm-none-linux-gnueabi-ld -Bstatic -Map hello-world.map  --strip-debug --fix-v4bx -elf2flt=-v -elf2flt=-k -o hello-world.flt -T sections.lds hello-world.o start.o ../mini-libc/printf.o ../mini-libc/libc_asm.o ../mini-libc/memcpy.o
arm-none-linux-gnueabi-ld: warning: ../mini-libc/printf.o uses variable-size enums yet the output is to use 32-bit enums; use of enum values across objects may fail
arm-none-linux-gnueabi-ld: warning: ../mini-libc/memcpy.o uses variable-size enums yet the output is to use 32-bit enums; use of enum values across objects may fail
arm-none-linux-gnueabi-ld: warning: cannot find entry symbol lf2flt=-k; defaulting to 00008000
hello-world.o: In function `main':
hello-world.c:(.text.startup+0xc): undefined reference to `__printf_chk'
hello-world.c:(.text.startup+0x18): undefined reference to `__printf_chk'
../include/common.mk:111: recipe for target 'hello-world.elf' failed
make: *** [hello-world.elf] Error 1

hello-world.elf failed

The Code Sourcery package mentioned in "2.2 Installing the Compiler" seems to be not for free anymore?

Greeting,
Benjamin

User Guide Update

I'm using mentor's code sourcery tool chain to compile c code for simulation.
When I do divide, I use __aeabi_idiv in mini-libc. It works really fine.
But with arm-none-linux-eabi library, the program doesn't work.
i.e. When I do divide using __aeabi_idiv which is in arm-none-linux-eabi library, it doesn't work at the simulation process.

When I use division function in mini-libc, simulation works and calculation results are fine.
However, when I use one in arm-none-linux-eaba library instead, simulation doesn't work and calculation results are incorrect.

Can I use function in arm-none-linux-eabi library?

Please let me know.

Really thanks for all your works.
Best and regards.
Dave

Division using gcc library

Hi,

For those of you that have an FPGA board without any UART connector available, you may be interested in the ALTERA JTAG UART feature and the nios2-terminal.
This can be achieved very simply by modifying  the file "uart.v" ( Revision 82 ) located into directory "hw\vlog\system\".

The procedure is the following:

1/ Insert the following line after line 171:

  wire tx_fifo_pop;
  
2/ Modify line 370 (formerly 369) accordingly:

  assign   tx_fifo_pop_not_empty = tx_fifo_pop && !tx_fifo_empty;

3/ Insert these two lines after line 426 ( formerly 425 )

  `ifndef ALTERA_JTAG_UART
   assign tx_fifo_pop = txd_state == TXD_STOP3 && tx_bit_pulse == 1'd1;

4/ Finally, add these lines after line 720 ( formerly 717 )

  `else
      
  `ifdef AMBER_SIM_CTRL
    reg r_ena = 1'b0;
    always@( posedge i_clk ) 
      r_ena <= ~r_ena; // dummy stuff for simulation
    assign rx_fifo_push = 1'b0;
  `else
    wire t_pause;
    wire r_ena;  
    alt_jtag_atlantic jtag_uart_alt_jtag_atlantic
      (
        .clk     (i_clk),
        .r_dat   (tx_byte),          // <- data to send
        .r_ena   (r_ena),            // -> allow to send data
        .r_val   (r_val),            // <- data is valid
        .rst_n   (1'b1),
        .t_dat   (rx_byte),          // -> received data 
        .t_dav   (~rx_fifo_full),    // <- data can be sent
        .t_ena   (rx_fifo_push),     // -> data is sent
        .t_pause (t_pause)
      );
  
    defparam jtag_uart_alt_jtag_atlantic.INSTANCE_ID = 0, // 0 for first instance, 1 for second instance
             jtag_uart_alt_jtag_atlantic.LOG2_RXFIFO_DEPTH = 6,
             jtag_uart_alt_jtag_atlantic.LOG2_TXFIFO_DEPTH = 6,
             jtag_uart_alt_jtag_atlantic.SLD_AUTO_INSTANCE_INDEX = "YES";
  `endif //AMBER_SIM_CTRL
    reg r_val;        
  	assign tx_fifo_pop = r_val;
    always@( posedge i_clk )
  	  r_val <= ~tx_fifo_empty & r_ena;	         
  `endif //ALTERA_JTAG_UART
  
5/ That's all folks! And there is no modification on software side...
Don't forget to define ALTERA_JTAG_UART in your verilog compiler settings.

6/ Invoking nios2-terminal
Open the nios2-shell ( menu ALTERA 15.0.0.145 Web Edition -> NIOS II EDS 15.0.0.145 -> NIOS II 15.0 Command Shell ) 
and type 'nios2-terminal'. If you have compiled your design with the 'boot-loader-serial' application, you are then able to send commands and see messages.

Cyril

Modification of "uart.v" to be able to use Altera JTAG UART

I'm trying to use example "boot-loader-ethmac".

In start.S file

        /* Switch to IRQ Mode */
        mov     r0, #0x00000002
        teqp    pc, r0
        /* Set IRQ Mode stack pointer */
        ldr     sp, AdrIRQStack

There is teqp opcode.

But I compile this start.S file, I get different disassemble result.

  78:	e3a00002 	mov	r0, #2
  7c:	e13ff000 	teq	pc, r0
  80:	e59fd26c 	ldr	sp, [pc, #620]	; 2f4 <AdrIRQStack>

I traced that operation in simulation. teqp operation has to do Exclusive OR operation. But it does operates like only teq in simulation.

I'm really confused about those issues.

Please let me know if you have any knowledge about this.
Thanks.

Opcode TEQP

Hello,

I'm trying to use interrupt controller with amber23.

I use whole source code you uploaded. So interrupt controller I use is included in hw source code.

I have few questions.
First, does your interrupt controller can control nested interrupts?
Second, like I said, I'm trying to use interrupt feature. But there are too less information about amber23 processor. Is there any reference design, code or documents?

I'm really digging to have processor working.

Thanks for all your works.
Dave.
Thanks.

Interrupt controller

Following the amber-user-guide.pdf, when I try to compile the boot-loader-serial I run into this error:

make -s -C ../mini-libc MIN_SIZE=1
arm-none-linux-gnueabi-gcc -c -Os -march=armv2a -mno-thumb-interwork -ffreestanding -I../include   -c -o boot-loader-serial.o boot-loader-serial.c
arm-none-linux-gnueabi-gcc -I../include    -c -o start.o start.S
arm-none-linux-gnueabi-gcc -c -Os -march=armv2a -mno-thumb-interwork -ffreestanding -I../include   -c -o crc16.o crc16.c
arm-none-linux-gnueabi-gcc -c -Os -march=armv2a -mno-thumb-interwork -ffreestanding -I../include   -c -o xmodem.o xmodem.c
arm-none-linux-gnueabi-gcc -c -Os -march=armv2a -mno-thumb-interwork -ffreestanding -I../include   -c -o elfsplitter.o elfsplitter.c
arm-none-linux-gnueabi-ld -Bstatic -Map boot-loader-serial.map  --strip-debug --fix-v4bx -o boot-loader-serial.elf -T sections.lds boot-loader-serial.o start.o crc16.o xmodem.o elfsplitter.o ../mini-libc/printf.o ../mini-libc/libc_asm.o ../mini-libc/memcpy.o
arm-none-linux-gnueabi-objcopy -R .comment -R .note boot-loader-serial.elf
../tools/amber-elfsplitter boot-loader-serial.elf > boot-loader-serial.mem
../tools/amber-memparams32.sh boot-loader-serial.mem boot-loader-serial_memparams32.v
../tools/amber-memparams128.sh boot-loader-serial.mem boot-loader-serial_memparams128.v
arm-none-linux-gnueabi-objdump -C -S -EL boot-loader-serial.elf > boot-loader-serial.dis
../tools/check_mem_size.sh boot-loader-serial.mem "@000020"
Mem file boot-loader-serial.mem too big
make: *** [all] Error 1


The size of the generated .mem is 40K. 


Thanks, 
Ciprian

mem file too big

Re-opening this issue since I've identified a simple configuration to make it appear.

Considering the last Amber revision (86),

if we modify line 192 of file "main_mem.v" from: 
  assign o_wb_ack = i_wb_stb && ( start_write || start_read_d2 );

to:
  assign o_wb_ack = i_wb_stb && ( wr_en       || start_read_d2 );

This has the effect of delaying the wb acknowledge of one clock cycle, generating a wait state when writing to the main memory.
In this configuration, when the "boot-loader-ethmac" software is run in simulation for a23, Amber enters into an infinite loop when executing the instruction "push	{r3, lr}" at address 0x1003788. 
This is the first instruction of the "main" function.  
This issue does not appear when running this test with a25.

Browse

Amber ARM-compatible core