f_ changed the topic of ##raspberrypi-internals to: The inner workings of the Raspberry Pi (Low level VPU/HW) -- for general queries please visit #raspberrypi -- open firmware: https://librerpi.github.io/ -- VC4 VPU Programmers Manual: https://github.com/hermanhermitage/videocoreiv/wiki -- chat logs: https://libera.irclog.whitequark.org/~h~raspberrypi-internals -- bridged to matrix and discord
Stromeko has quit [Ping timeout: 256 seconds]
Stromeko has joined ##raspberrypi-internals
jcea has quit [Ping timeout: 268 seconds]
<f_ridge> <x​2x6_/D> Hi, clever. So the problem is not that I can not understand arguments or the output - problem Is I can not receive 512 block at all
<f_ridge> <x​2x6_/D> I have some CRC errors in the interrupt or status register instead
<f_ridge> <x​2x6_/D> And that's even without seeting bin 31. Ony the check part. So I am not doing any kind of function chang
<f_ridge> <x​2x6_/D> And that's even without seeting bin 31. Ony the check part. So I am not doing any kind of function change(edited)
<f_ridge> <x​2x6_/D> I think BCM2835 emmc controller is just not capable to process this command for some reason
<f_ridge> <x​2x6_/D> when I send cmd6 , i receive CMD done, DATA CRC err
<f_ridge> <x​2x6_/D> when I send cmd6 , i read interrupt register value as CMD_DON | ERR | DATA_CRC_ERR(edited)
jcea has joined ##raspberrypi-internals
dolphinana has joined ##raspberrypi-internals
dolphinana has quit [Quit: Leaving]
jcea has quit [Ping timeout: 246 seconds]
Stromeko has quit [Quit: Going… gone.]
Stromeko has joined ##raspberrypi-internals
clever has joined ##raspberrypi-internals
<f_ridge> <c​lever___/D> @x2x6_ does it give an R1 response?
indy_ has joined ##raspberrypi-internals
<f_ridge> <x​2x6_/D> Yes, its 900
<f_ridge> <c​lever___/D> @x2x6_ does emmc require you to program in the expected block size and block count?
jcea has joined ##raspberrypi-internals
<f_ridge> <x​2x6_/D> Yep, I do this the same way I program for reads/writes
<f_ridge> <c​lever___/D> and what byte count do you program it for?
<f_ridge> <x​2x6_/D> 512 sector size / 1 sector
<f_ridge> <c​lever___/D> thats incorrect, CMD6 returns a 64 byte block, 512 bits
<f_ridge> <c​lever___/D> that would explain the checksum errors
<f_ridge> <x​2x6_/D> You told me this, I've read this, I always saw 512 bits at 512 bytes. I had some "mental overlay".
<f_ridge> <x​2x6_/D> Thanks.
<f_ridge> <x​2x6_/D> I will try now
<f_ridge> <c​lever___/D> basically, the SD controller will read `$size` bytes from the DAT pins, then it will expect the checksum to follow
<f_ridge> <c​lever___/D> so youve configured it to read the full 64 byte reply, then keep reading another 448 bytes of junk (including the true checksum)
<f_ridge> <c​lever___/D> then it will expect the checksum, where there is none, and that idle bus fails the checksum
<f_ridge> <x​2x6_/D> So blocksize should be 64 and num sectors 1
<f_ridge> <c​lever___/D> yep
<f_ridge> <G​itHub Lines/D> ```cpp
<f_ridge> <G​itHub Lines/D> *REG32(SH_HBCT) = 64;
<f_ridge> <G​itHub Lines/D> *REG32(SH_HBLC) = 1;
<f_ridge> <G​itHub Lines/D> send_raw(SD_SEND_SWITCH_FUNC | SH_CMD_READ_CMD_SET, arg);
<f_ridge> <G​itHub Lines/D> ```
<f_ridge> <x​2x6_/D> Unbelievable. I've been looking at this for at least one week. Now it worked.
<f_ridge> <c​lever___/D> `0x00c88001 0x80018001 0x80018001 0x80030000 0x01000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000 0x00000000`
<f_ridge> <c​lever___/D> in my case, i get a list of bytes like this
<f_ridge> <x​2x6_/D> Thanks a lot, Clever, looks like my brain is completely wasted)
<f_ridge> <x​2x6_/D> Yes, I get something like this now
<f_ridge> <c​lever___/D> what numbers exactly are you getting?
<f_ridge> <x​2x6_/D> I need to add now the processing , but raw debug shows this readings from data register
<f_ridge> <x​2x6_/D> data:01803200,intr:00000021
<f_ridge> <x​2x6_/D> data:01800180,intr:00000000
<f_ridge> <x​2x6_/D> data:01800180,intr:00000000
<f_ridge> <x​2x6_/D> data:00000380,intr:00000000
<f_ridge> <x​2x6_/D> data:00000001,intr:00000000
<f_ridge> <x​2x6_/D> data:00000000,intr:00000000
<f_ridge> <x​2x6_/D> data:00000000,intr:00000000
<f_ridge> <x​2x6_/D> data:00000000,intr:00000000
<f_ridge> <x​2x6_/D> data:00000000,intr:00000000
<f_ridge> <x​2x6_/D> data:00000000,intr:00000000
<f_ridge> <c​lever___/D> looks like it needs a byte-order swap, same as i did
<f_ridge> <G​itHub Lines/D> ```cpp
<f_ridge> <G​itHub Lines/D> for (unsigned int i=0; i<(64/4); i++) {
<f_ridge> <G​itHub Lines/D> printf("0x%08x ", htonl(switch_reply[i]));
<f_ridge> <G​itHub Lines/D> }
<f_ridge> <G​itHub Lines/D> ```
<f_ridge> <x​2x6_/D> Ok.
<f_ridge> <c​lever___/D> the `3200` in that first line, after byte swap, is `0x0032`, or 50
<f_ridge> <c​lever___/D> which coresponds to these 16 bits, and means the card will draw 50mA
<f_ridge> <c​lever___/D> which coresponds to these 16 bits, and means the card will draw 50mA from the vcc rail(edited)
Herc has quit [Remote host closed the connection]
Herc has joined ##raspberrypi-internals
Ad0 has quit [Ping timeout: 252 seconds]
vriska has quit [Ping timeout: 252 seconds]
vriska has joined ##raspberrypi-internals
Ad0 has joined ##raspberrypi-internals