Posts Tagged arm

Testing FreeSWITCH performance on Scaleway C1

The dedicated ARM hosting servers at Scaleway appear to be a decent platform for a mid-sized PBX.

In short, the platform displays the following results in performance tests:

  • OPUS<->PCMA transcoding: 16 simultaneous calls with  at about 95% total CPU load and no noticeable distortions.
  • SILK<->PCMA transcoding: 72 simultaneous calls were going without distortions, with average total CPU load at 63%. Higher number of calls resulted in noticeable distortions.
  • G722<->PCMA transcoding: 96 simultaneous calls without distortions, at 76% CPU load, and noticeable distortions for higher numbers.

Read the rest of this entry »

, , , ,

3 Comments

Installing FreeSWITCH on Scaleway C1

Scaleway (a cloud service by online.net) offers ARM-based dedicated servers for EUR9.99/month, and the first month free. The platform is powerful enough to run a small or FreeSWITCH server, and it shows nice results in voice quality tests.

These instructions are for Debian Wheezy distribution.

By default, the server is created with Linux kernel 3.2.34, and this kernel version does not have a high-resolution timer. You need to choose 3.19.3 in server settings.

At Scaleway, you get a dedicated public IP address and 1:1 NAT to a private IP address on your server. So, FreeSWITCH SIP profiles need to be updated (“ext-rtp-ip” and “ext-sip-ip” to point to you rpublic IP address).

FreeSWITCH compiles and links “mpg123-1.13.2” library, which fails to compile on ARM architecture.  You need to edit the corresponding files to point to “mpg123-1.19.0” and commit back to Git, because the build scripts check if any modified and uncommitted files exist in the source tree. Also the patch forces to use gcc-4.7, as 4.6 is known with some problems on ARM architecture. Read the rest of this entry »

, , , ,

8 Comments

CVJI-E212, one more cheap Allwinner A10 media player

I started looking into a number of devices based on Allwinner A10 SoC — this appears to be a very popular chip in China, and there are endless varieties of cheap tablets and HDMI media players. Among the most known are Mele A1000/A2000, MK802, MK802+, MiniX media players — they all provide HDMI interface, one or two USB ports, and some of them come with a built-in WiFi adapter.  The chip also supports OTG USB interface, and some of those devices also provide it. Also MK802 can be powered from the USB port.

Those devices come usually with pre-installed Android 4.0. Typically the OS supports only a limited range of screen resolutions and frequencies (I couldn’t get a decent picture on my Sony Bravia TV and on a computer monitor via a HDMI-to-DVI adapter).

Anyway, I’m a server guy, and my primary interest is in “server” applications, even on such small appliances. So, I don’t really care about graphics.

Read the rest of this entry »

, ,

28 Comments

Big Endian for i.MX27: bug test code

a piece of assembler code which proves the bug in big-endian mode:

 Read the rest of this entry »

, ,

Leave a comment

Big Endian for i.MX27: found the problem

Found the big endian problem: it looks like the memory controller bug.

Read the rest of this entry »

, , ,

2 Comments

Big Endian for i.MX27: looks like it’s not working

I managed to launch the CPU in big endian mode, and it even executes
some part of u-boot code. But the serial is silent, and MTD driver can’t find
any NAND chips.

The UART control registers receive correct values, I verified this.
So, most probably the UART hardware fails to get those values from a big-endian CPU.

, , ,

Leave a comment

Moving to big endian: first steps

Here’s my message at Armadeus mailing list

Here’s the link to the u-boot patch to be studied.

UPD: Currently the nand_spl loader is compiled in LE, and then it prepares the u-boot image in the memory and switches to BE mode. Then u-boot starts quite normally, but then it hangs up without printing anything to the serial console. Looks like some initialization code breaks internal memory. Need to solder a JTAG connector for further debugging…

, ,

Leave a comment

Mounting the LCD touchscreen on top of Armadeus devboard

img_0878Assembled all the Armadeus parts together. Now it’s much easier to move them around the table 🙂

The Armadeus devboard comes without any cables. The LCD and touchscreen are equipped with two cables for connecting to the devboard. The devboard needs a 2.5/5.5 power jack, and the power supply should support at least 12W power draw.

Read the rest of this entry »

,

2 Comments

Armadeus apf27 hardware ordered

Ordered the following hardware from Armadeus.com:

  • APF27-M128F256-EUF200 (i.MX27,  128 Mb RAM, 256MB NAND Flash, FPGA spartan3A (200Kgates))
  • APF27_Dev_Light
    • DAC+ADC
    • CAN
    • backuped RTC
  • LQ043_adapt kit
    • Touch screen FJT N010-0554-T241

, ,

Leave a comment

Big Endian mode on Armadeus apf27 (i.MX27)

I started to port the Armadeus BSP to Big Endian mode. So far, only 5 or 6 files needed to be patched in order to compile everything. But I haven’t started the field tests, as my Armadeus board is still in the ordering process 🙂

, , ,

Leave a comment