linuxdc-dev

[linuxdc-dev] DMA issue

[Adrian M. <ad...@mc...>]

I suspect this is more to do with how the kernel works than SH coding, bu=
t=20
I'll ask anayway...

Writing some maple driver code on the Dreamcast which is called by the ke=
rnel=20
at start up ...

=09for (i=3D0; i<(1<<MAPLE_DMA_PAGES); i++)
=09=09dma_cache_wback_inv(maple_sendbuf+i*PAGE_SIZE, PAGE_SIZE);
=09ctrl_outl(1, MAPLE_STATE); /*Start maple DMA transfer */
=09do {
=09}while (wait_event_interruptible(maple_dma_wait, maple_dma_done())!=3D=
 0);=20


with...

int maple_dma_done(void)
{
//=09pr_info("Testing DMA\n");
=09return (ctrl_inl(MAPLE_STATE) & 1) =3D=3D 0;
}

When this is run on start up it hangs the box. Uncomment the pr_info line=
 (as=20
I will do in the patch I am about to post) and it works as I expect it (i=
e=20
waits till maple dma is done and then reads off the reply).

What have I got wrong here?

Adrian

[linuxdc-dev] problem loading snapgear

[R V. <lnx...@ya...>]

Hi,
 
I am using IPL 0.11 on MS7751SE01 (solution engine sh 7751) board.
 
I am using minicom with line dealy 10 ms. i am facing following problem while loading kernel 2.4.x (snapgear dist.), while loading it using sh-linux-gdb...
 
 i got the following message

Loading section .empty_zero_page, size 0x1000 lma 0x8c001000

Loading section .text, size 0x9407c lma 0x8c002000

Loading section .rodata, size 0x9086 lma 0x8c09607c

Loading section .kstrtab, size 0x515c lma 0x8c09f104

Loading section __ex_table, size 0xba0 lma 0x8c0a4260

Loading section __ksymtab, size 0x18b8 lma 0x8c0a4e00

Loading section .data, size 0x520c lma 0x8c0a66b8

Loading section .data.init_task, size 0x2000 lma 0x8c0ac000

Loading section .text.init, size 0x6ec0 lma 0x8c0ae000

Loading section .data.init, size 0x6f0 lma 0x8c0b4ec0

Loading section .setup.init, size 0xb0 lma 0x8c0b55b0

Loading section .initcall.init, size 0x54 lma 0x8c0b5660

Loading section .data.cacheline_aligned, size 0xa80 lma 0x8c0b6000

Start address 0x8c002000, load size 739830

Transfer rate: 59784 bits/sec, 149 bytes/write.

(gdb)cont

Linux version 2.4.24-uc0 (root@dev) (gcc version 3.2.3) #3 Tue Jun 15 19:46:25 IST 2004

initrd extends beyond end of memory (0x80400000 > 0x10000000)

disabling initrd

On node 0 totalpages: 16384

zone(0): 16384 pages.

zone(1): 0 pages.

zone(2): 0 pages.

Kernel command line:

Setting GDB trap vector to 0x80000100

SH RTC: invalid value, resetting to 1 Jan 2000

CPU clock: 162.04MHz

Bus clock: 27.00MHz

Module clock: 27.00MHz

 

----------------------------------

Thanks in advance

Vijay

		
---------------------------------
Do you Yahoo!?
Yahoo! Mail is new and improved - Check it out!