Felix Kühling wrote:
I don't think this is a matter of the chip revision but rather what the
card manufacturer puts around the chip, like DDR vs. SDR RAM and 64-bit
vs. 128-bit memory interface and at which frequencies the chip and
memory are clocked.
Speaking of ram width, would it be possible to fix the radeon_driver.c so it doesn't always report 64bit?

RADEONRAMRec RADEONRAM[] = {    /* Memory Specifications
                                   From Radeon Manual */
    { 4, 4, 1, 2, 1, 2, 1, 16, 12, "64-bit SDR SDRAM" },
    { 4, 4, 3, 3, 2, 3, 1, 16, 12, "64-bit DDR SDRAM" },


info->MemCntl = INREG(RADEON_SDRAM_MODE_REG); info->BusCntl = INREG(RADEON_BUS_CNTL);

                                /* RAM */
    switch (info->MemCntl >> 30) {
    case 0:  offset = 0; break; /*  64-bit SDR SDRAM */
    case 1:  offset = 1; break; /*  64-bit DDR SDRAM */
    default: offset = 0;
    }
    info->ram = &RADEONRAM[offset];

and those cas latency etc. values are likely very wrong with anything else than oldest plain old radeons anyways (though luckily I don't think anyone uses that information in that info record?)

If we're lucky maybe the width of the interface is also encoded in that sdram_mode_reg? And AFAIK it should be possible to probe latency values too (at least some windows tweakers can do it and even change them).

Roland



-------------------------------------------------------
This SF.net email is sponsored by Dice.com.
Did you know that Dice has over 25,000 tech jobs available today? From
careers in IT to Engineering to Tech Sales, Dice has tech jobs from the
best hiring companies. http://www.dice.com/index.epl?rel_code4
_______________________________________________
Dri-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/dri-devel

Reply via email to