Quantcast
Channel: ARM Connected Community: Message List
Viewing all 8551 articles
Browse latest View live

Re: Reg Optimization

$
0
0

Pavi,

Kindly post your code snippet here.


Re: Armulator integration with DS-5

$
0
0

Hi,

 

DS-5 includes fixed virtual platforms, which give you a way of testing bare metal and Linux applications on your host machine. As Martin says, ARMulator was discontinued a while ago.

Re: Can't compile any shader with the OpenGL ES 2 Emulator

$
0
0

Thanks for your help reporting this,

 

We've also reproduced this, after updating our nvidia drivers so it's unrelated to the CUDA SDK and is being provoked by a recent change in the nvidia driver. This will be fixed in an upcoming release. For now the work around is to use an older nvidia driver, or follow your steps with the hex editor for the GLES2 emulator.

 

Thanks,

Chris

Re: OpenGL ES 2.0 Emulator: Strange behavior when using a varying

$
0
0

Hi degenerated,

 

you're right it shouldn't corrupt the texture coordinate data - I was assuming (incorrectly it seems) that the texture coordinates were being modified wrongly by offsets calculated from buggy normal values. Without the fragment shader source that was my best guess :-)

 

Glad you made progress, as Chris says attribute location seems a likely cause - I've been caught out by the same issue when the compiler spots a uniform or attribute isn't actually used and optimizes it away...

 

Cheers, Pete

Re: Re: OpenGL ES 2.0 Emulator: Strange behavior when using a varying

$
0
0

I've been caught out by the same issue when the compiler spots a uniform or attribute isn't actually used and optimizes it away...

Not optimal, but a conservative way of doing this might be:

 

for each (attribute) {     GLuint attribLoc = glGetAttribLocation(attribute);     if (attribLoc != -1)          glVertexAttribPointer(blah blah);
}

 

Attrib data always gets pushed to the right place, or not at all if optimized out.

Re: Can't compile any shader with the OpenGL ES 2 Emulator

$
0
0

Hi degenerated,

 

The workaround is - as Chris has already suggested - to use older NVIDIA Graphics Driver.

In my environment (NVIDIA NVS 300, Windows 7 64bit) the most recent driver that still works with OpenGL ES 3.0 Emulator v1.3.1 is 332.76.


Hope that helps,

Jacek

commercial development boards for armv8

$
0
0

Hi experts,

 

Can you name few commercial ARMv8 development boards at least possible price for hobby stuffs.

unable to add contact in jive sw

$
0
0

HI experts,

 

when i try to add contact in the android version of jive software, error message is displayed with "unable to add contact"


Re: OpenGL ES 2.0 Emulator: Strange behavior when using a varying

$
0
0

Yeah, that sort of what I did last night to fix the issue. Now it's working fine!

 

Thanks for all your support!

What is Cortex-M3?

arm 汇编 dcb指令

$
0
0

Error   : armlink: L6204:  Entry point (0x00008000) does not point to an instruction.这个错误的产生,我觉得是我用了dcb指令,放在了程序的开头,它开辟了一个连续的存储空间,都是用来放数据的,所以无法找到指令。然后就有了一个新的问题,这个dcb指令,以及其他的内存分配指令,应该怎样使用,他们在程序中的位置应该在哪,比如说我像开辟一个以字节为单位的连续存储空间用来存放abcdef,应该把它放在哪里,放在程序中间感觉不太合适。

DG535 是用什么硬件开发的?

$
0
0

DG535 是用什么硬件开发的?

双口ram读数据的速度太慢

$
0
0

系统结构与功能:

       lpc3131外接8k*8双口ram与8G Flash,从双口ram读取数据(来自FPGA),并存至Flash中。

问题:

       丢数据,系统速度远低于ram的读速与Flash的写速度。

 

  双口ram读数据直接采用指针,如下:

        for(i=0;i<tmp;i++)

      {

        nandflash.databuf[nandflash.ptr_buf] = *(volatile UNS_8 *)(EXT_SRAM0_0_BASE + ram.ptr);

        nandflash.ptr_buf++;

        ram.ptr++;

      }

 

Flash按页写的程序如下:

INT_32 nand_page_write(INT_32 devid,

                       void * buffer,

                       INT_32 bank_num,

                       INT_32 block_num,

                       INT_32 page_num)

{

  NAND_DRV_T *nanddrv = (NAND_DRV_T *) devid;

  NAND_FLASH_CTRL_REGS_T *nandreg = nanddrv->regptr;

  UNS_32 * pData;

  UNS_32 * pnfbuf[2];

  UNS_32 i,j,tmp;

  UNS_8 addr1 = 0;

  UNS_8 addr2 = 0;

  UNS_8 addr3 = (UNS_8)(((block_num<<6)&0x0C0)+(page_num&0x03F));

  UNS_8 addr4 = (UNS_8)((block_num>>2)&0x0FF);

  UNS_8 addr5 = (UNS_8)((block_num>>10)&0x07);

  pnfbuf[0] = (UNS_32 *) NANDFLASH_CTRL_S0_BASE;

  pnfbuf[1] = (UNS_32 *) (NANDFLASH_CTRL_S0_BASE + 0x400);

  nandreg->set_ce = NAND_SETCE_WP | NAND_SETCE_CV(bank_num);

  nandreg->set_cmd = NAND_CMD_SDIN;

  nandreg->set_addr = addr1;

  nandreg->set_addr = addr2;

  nandreg->set_addr = addr3;

  nandreg->set_addr = addr4;

  nandreg->set_addr = addr5;

  for(j=0;j<8;j++)

  {

    pData = (UNS_32 *)((UNS_8 *)buffer + j*512);

    tmp = j&0x01;

    for(i=0;i<128;i++)

    {

      *(pnfbuf[tmp]+i) = *(pData + i);

    }

    if(j>0)

    {

      while(!(nand_ioctl(devid, NAND_IRQ_STATUS, 3-tmp)));

      while((nand_ioctl(devid, NAND_IRQ_STATUS, 3-tmp)))nandreg->irq_status_raw = 0xFFFFFFFF;

    }

    nandreg->control_flow = NAND_CTRL_RD_RAM(4+tmp);

  }

  while(!(nand_ioctl(devid, NAND_IRQ_STATUS, 3)));

  while((nand_ioctl(devid, NAND_IRQ_STATUS, 3)))nandreg->irq_status_raw = 0xFFFFFFFF;

  nandreg->set_cmd = NAND_CMD_PAGEPROG;

  while((nand_ioctl(devid, NAND_RB_STATUS, bank_num)));

  while(!(nand_ioctl(devid, NAND_RB_STATUS, bank_num)));

  return 0;

}

 

 

为什么速度远低于芯片说明文档中的读写速度???

求指点,非常感谢!

Re: 中文社区积极用户奖励方案(2014年3月1日至2014年6月30日)

$
0
0

请问意思是说要到6月30号,最后的结果才是吗?不是每个月一次吧?

Re: How to disable the instruction caching (prefetch unit)?

$
0
0

Hi Volker,

in the Cortex-M3 and Cortex-M4 processors there is no programmable register for disabling prefetch behavior. In almost all pipelined processors, when a instruction enter the execution stage, some of the subsequent instructions would already been fetched.

 

could you point us to the document that implies prefetch can be disabled? It might be a documentation error.

 

note: if you are talking about prefetching of instructions in device specific flash accelerators, that's a different thing.

regard,

joseph


Re: Flash programming in ASM for Cortex M4

$
0
0

FLash programming mechanism is device specific. So you need to tell us which chip you are using, then maybe we can point you to the right direction.

Re: What is Cortex-M3?

Re: How to jumpstart an ARM9 board from its SD slot?

$
0
0

You most likely need to have a device name to download the image for.

For instance, if you use CubieBoard2, you could use this search phrase with Google:

CubieBoard2 Linux

...It would give you a lot of results, and among those, you would find "Cubian" (which is the easiest one to get working with CubieBoard).

If you decided on the $47 board, you need the name of that board and use the above search-phrase with the CubieBoard2 name replaced.

Let's say you decided on the $45 CuBox-i. On Google, the phrase...

CuBox-i Linux

gives you a number of results related to installing Linux on this box.

First link I get is this: Linux Distros Ready: Ubuntu, Fedora, Debian | The CuBox i From SolidRun. The page has a few download links.

Clicking the FAQ takes you to this page, which shows a step-by-step guide how to install the downloaded files onto a Micro SD card.

The CuBox-i is also $2 lower than the board you mentioned, and it has an enclosure included, a very interesting little device.

New Partner Rewards Program

$
0
0

Thank you for your contributions in 2013, we are delighted to see the increasing number of participation from our partners. 

 

In 2014, we will be hosting a Partner Rewards Program– our way of saying ‘thank you’ for your loyal participation in the ARM Connected Community. 

 

Partner Rewards Program Details:

The Partner Space with the most views by December 31, 2014 will receive a prize for your office. Click Vote for Partner Rewards Program Prize to vote on what the prize should be!

 

From previous stats, the Partner Spaces with the most views are the ones that publish the most content, active in the community and share their contributions via their marketing channels.

 

We value your participation and thank you again for joining the new ARM Connected Community.

 

Cheers to a successful 2014!

Re: Design Start ARM Cortex-M0

$
0
0

Hi,

 

My professor asked me to do offline processing for signal. I think for that i would only require a processor core, memory and probably a UART.

  I have already built those hardware  from design kit.

 

I wanted to write  application code(in Keil) for my algorithm which takes data from memory  process it and store it  back into memory.

Since most of the design example(EDK) are in assembly but I m  more familiar with C  but addressing of  peripheral and  starting reset handler routine. I  do not know  how to write in C.   I did not use CMSIS also:(

 

 

Could you suggest some  thing which makes it easier

Viewing all 8551 articles
Browse latest View live