Linux 2.6-10.rc3 8xx: debugging (over-writing) content of bd_in fo structure in the kernel booting code

Linux 2.6-10.rc3 8xx: debugging (over-writing) content of bd_in fo structure in the kernel booting code

[Povolotsky, Alexander]

I am now suspecting that the content of my bd_info structure is set or is
handled incorrectly,
thus causing incorrect behavior of the serial driver upon kernel booting
(only 3 characters are printed correctly and the rest is the garbage - as I
reported before), 
so I am trying (for debugging purposes) to over-write the values, passed
from bd_info
during the kernel booting, by the hard-coded "correct" values.

So far I tried forcing baud, bits, parity, flow directly in in
cpm_uart_console_setup()
( drivers/serial/cpm_uart/cpm_uar t_core.c ) - see below - it didn't help
...

Where in the kernel booting code the clock speed is handled (it also comes
from the bd_info)
 - so I could try to force set clock speed there ?

Thanks,
Alex
****************************************************************
static int __init cpm_uart_console_setup(struct console *co, char *options)
{
        struct uart_port *port;
        struct uart_cpm_port *pinfo;
        int baud = 38400;
        int bits = 8;
        int parity = 'n';
        int flow = 'n';
        int ret;

        port =
            (struct uart_port
*)&cpm_uart_ports[cpm_uart_port_map[co->index]];
        pinfo = (struct uart_cpm_port *)port;

        pinfo->flags |= FLAG_CONSOLE;

        if (options) {
                uart_parse_options(options, &baud, &parity, &bits, &flow);
        } else {
                bd_t *bd = (bd_t *) __res;

                if (bd->bi_baudrate)
                        baud = bd->bi_baudrate;
                else
                        baud = 9600;
        }
        baud = 38400;
<<===========================================
        parity = 'n';
<<===========================================
        flow = 'n';
<<===========================================
        bits = 8;
<<===========================================
        /*
         * Setup any port IO, connect any baud rate generators,
         * etc.  This is expected to be handled by board
         * dependant code
         */
        if (pinfo->set_lineif)
                pinfo->set_lineif(pinfo);

        if (IS_SMC(pinfo)) {
                pinfo->smcp->smc_smcm &= ~(SMCM_RX | SMCM_TX);
                pinfo->smcp->smc_smcmr &= ~(SMCMR_REN | SMCMR_TEN);
        } else {
                pinfo->sccp->scc_sccm &= ~(UART_SCCM_TX | UART_SCCM_RX);
                pinfo->sccp->scc_gsmrl &= ~(SCC_GSMRL_ENR | SCC_GSMRL_ENT);
        }

        ret = cpm_uart_allocbuf(pinfo, 1);

        if (ret)
                return ret;

        cpm_uart_initbd(pinfo);

        if (IS_SMC(pinfo))
                cpm_uart_init_smc(pinfo);
        else
                cpm_uart_init_scc(pinfo);

        uart_set_options(port, co, baud, parity, bits, flow);

        return 0;
}

Re: Linux 2.6-10.rc3 8xx: debugging (over-writing) content of bd_in fo structure in the kernel booting code

[Hans Schillstrom]

Hi Alex,
On Tue, 2005-01-18 at 12:45, Povolotsky, Alexander wrote:
> I am now suspecting that the content of my bd_info structure is set or is
> handled incorrectly,
> thus causing incorrect behavior of the serial driver upon kernel booting
> (only 3 characters are printed correctly and the rest is the garbage - as I
> reported before), 
> so I am trying (for debugging purposes) to over-write the values, passed
> from bd_info
> during the kernel booting, by the hard-coded "correct" values.
> 
> So far I tried forcing baud, bits, parity, flow directly in in
> cpm_uart_console_setup()
> ( drivers/serial/cpm_uart/cpm_uar t_core.c ) - see below - it didn't help
> ...
> 
> Where in the kernel booting code the clock speed is handled (it also comes
> from the bd_info)
>  - so I could try to force set clock speed there ?

Have a look into ppc/8xx_io/commproc.c
There you have the baudrate dependency - "bi_intfreq"

#define BRG_INT_CLK		(((bd_t *)__res)->bi_intfreq)
#define BRG_UART_CLK		(BRG_INT_CLK/16)
#define BRG_UART_CLK_DIV16	(BRG_UART_CLK/16)

void
cpm_setbrg(uint brg, uint rate)
{
 ...

/Hans

> 
> Thanks,
> Alex
> ****************************************************************
> static int __init cpm_uart_console_setup(struct console *co, char *options)
> {
>         struct uart_port *port;
>         struct uart_cpm_port *pinfo;
>         int baud = 38400;
>         int bits = 8;
>         int parity = 'n';
>         int flow = 'n';
>         int ret;
> 
>         port =
>             (struct uart_port
> *)&cpm_uart_ports[cpm_uart_port_map[co->index]];
>         pinfo = (struct uart_cpm_port *)port;
> 
>         pinfo->flags |= FLAG_CONSOLE;
> 
>         if (options) {
>                 uart_parse_options(options, &baud, &parity, &bits, &flow);
>         } else {
>                 bd_t *bd = (bd_t *) __res;
> 
>                 if (bd->bi_baudrate)
>                         baud = bd->bi_baudrate;
>                 else
>                         baud = 9600;
>         }
>         baud = 38400;
> <<===========================================
>         parity = 'n';
> <<===========================================
>         flow = 'n';
> <<===========================================
>         bits = 8;
> <<===========================================
>         /*
>          * Setup any port IO, connect any baud rate generators,
>          * etc.  This is expected to be handled by board
>          * dependant code
>          */
>         if (pinfo->set_lineif)
>                 pinfo->set_lineif(pinfo);
> 
>         if (IS_SMC(pinfo)) {
>                 pinfo->smcp->smc_smcm &= ~(SMCM_RX | SMCM_TX);
>                 pinfo->smcp->smc_smcmr &= ~(SMCMR_REN | SMCMR_TEN);
>         } else {
>                 pinfo->sccp->scc_sccm &= ~(UART_SCCM_TX | UART_SCCM_RX);
>                 pinfo->sccp->scc_gsmrl &= ~(SCC_GSMRL_ENR | SCC_GSMRL_ENT);
>         }
> 
>         ret = cpm_uart_allocbuf(pinfo, 1);
> 
>         if (ret)
>                 return ret;
> 
>         cpm_uart_initbd(pinfo);
> 
>         if (IS_SMC(pinfo))
>                 cpm_uart_init_smc(pinfo);
>         else
>                 cpm_uart_init_scc(pinfo);
> 
>         uart_set_options(port, co, baud, parity, bits, flow);
> 
>         return 0;
> }
> 

RE: Linux 2.6-10.rc3 8xx: debugging (over-writing) content of b d_in fo structure in the kernel booting code

[Povolotsky, Alexander]

Hi,

my "custom" bd_info structure (passed from my custom bootloader) looks as
following:

typedef struct bd_info {
        unsigned long   bi_memstart;    /* start of DRAM memory */
        unsigned long   bi_memsize;     /* size  of DRAM memory in bytes */
        unsigned long   bi_flashstart;  /* start of FLASH memory */
        unsigned long   bi_flashsize;   /* size  of FLASH memory */
        unsigned long   bi_flashoffset; /* reserved area for startup monitor
*/
        unsigned long   bi_sramstart;   /* start of SRAM memory */
        unsigned long   bi_sramsize;    /* size  of SRAM memory */
        unsigned long   bi_immr_base;   /* base of IMMR register */
        unsigned long   bi_bootflags;   /* boot / reboot flag (for LynxOS)
*/
        unsigned long   bi_ip_addr;     /* IP Address */
        unsigned char   bi_enetaddr[6]; /* Ethernet adress */
        unsigned short  bi_ethspeed;    /* Ethernet speed in Mbps */
        unsigned long   bi_intfreq;     /* Internal Freq, in MHz */
        unsigned long   bi_busfreq;     /* Bus Freq, in MHz */
        unsigned long   bi_baudrate;    /* Console Baudrate */
        unsigned char   bi_run_bank;    /* Running Bank */

} bd_t;

My embed_config(bd_t **bdp) function (in
arch/ppc/boot/simple/embed_config.c)
does the following (using above structure) :

 bd->bi_baudrate = 38400; /* changed from 115200 for debug - Alex */
 bd->bi_memstart = 0;
 bd->bi_memsize = (32 * 1024 * 1024);
 bd->bi_intfreq = 50000000;
 bd->bi_busfreq = 50000000;

Then we have, as Hans noted to me, the cpm_setbrg() function
 (in arch/ppc/8xx_io/commproc.c):

/* Set a baud rate generator.  This needs lots of work.  There are
 * four BRGs, any of which can be wired to any channel.
 * The internal baud rate clock is the system clock divided by 16.
 * This assumes the baudrate is 16x oversampled by the uart.
 */
#define BRG_INT_CLK             (((bd_t *)__res)->bi_intfreq)
#define BRG_UART_CLK            (BRG_INT_CLK/16)
#define BRG_UART_CLK_DIV16      (BRG_UART_CLK/16)

void
cpm_setbrg(uint brg, uint rate)
{
        volatile uint   *bp;

        /* This is good enough to get SMCs running.....
        */
        bp = (uint *)&cpmp->cp_brgc1;
        bp += brg;
        /* The BRG has a 12-bit counter.  For really slow baud rates (or
         * really fast processors), we may have to further divide by 16.
         */
        if (((BRG_UART_CLK / rate) - 1) < 4096)
                *bp = (((BRG_UART_CLK / rate) - 1) << 1) | CPM_BRG_EN;
        else
                *bp = (((BRG_UART_CLK_DIV16 / rate) - 1) << 1) |
                                                CPM_BRG_EN | CPM_BRG_DIV16;
}

So what could go wrong here in terms of setting the baudrate for the serial
uart 
(which "presumably" causes printing 3 good characters and then garbage
during kernel booting) ?

Re: Linux 2.6-10.rc3 8xx: debugging (over-writing) content of bd_in fo structure in the kernel booting code

[Mark Chambers]

>
> So what could go wrong here in terms of setting the baudrate for the
serial
> uart
> (which "presumably" causes printing 3 good characters and then garbage
> during kernel booting) ?
>

If you can get your hands on an oscilloscope and look at the TxD from your
board, you can get an exact measurement of your baudrate and make sure
you are on the right track. (Baudrate = 1/shortest-high-or-low-time)

Mark Chambers