RfidResearchGroup/proxmark3
1
1
Fork 0
mirror of https://github.com/RfidResearchGroup/proxmark3.git synced 2025-02-13 10:43:01 +08:00
Code Issues Projects Releases Packages Wiki Activity

HF_COLIN/STANDALONE: FIX Simulation/Writing Tag to flash/Reading back from flash

Browse source 
* Simulation was broke by the 4/7/10 UID lenght pseudo-support (which, in regard of the whole, will never be of any use unless complete reimplementaiton). Kept this part as respect, but enforced UID_IN_EMUL, since it will _ALWAYS_ be the case, be it direct simulation after tag breaking or from reading back tag.
* Writing tag was broken when flashmem driver had SPI settings changed, enforcing 48MHz speed as necessity for some work (like in hf_bog), so now it is
* Reading back, however, should absolutely be done at 24Mhz to avoid those unecessary and unstable [On this chip] FAST_READS.

Should worsk flawlessly now.
If one has spare time, handling getting new keys or defining whole Schemes in Flashmem would be perfect. Just has to be a conditional and a 8/16key ordered answer.
This commit is contained in:
Colin J. Brigato 2019-07-14 13:29:40 +02:00 committed by GitHub
parent 55602bf833
commit 2fc26fbf35
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 16 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

19
armsrc/Standalone/hf_colin.c
View file

@ -101,6 +101,8 @@ void ReadLastTagFromFlash() {
size_t size = len;
uint8_t *mem = BigBuf_malloc(size);
FlashmemSetSpiBaudrate(24000000);
if (!FlashInit()) {
return;
}
@ -148,6 +150,9 @@ void WriteTagToFlash(uint8_t index, size_t size) {
emlGetMem(data, 0, (size * 64) / 1024);
FlashmemSetSpiBaudrate(48000000);
if (!FlashInit()) {
return;
}
@ -424,7 +429,7 @@ failtag:
//-----------------------------------------------------------------------------
// also we could avoid first UID check for every block
// then let’s expose this “optimal case” of “well known vigik schemes” :
// then let's expose this “optimal case” of “well known vigik schemes” :
for (uint8_t type = 0; type < 2 && !err && !trapped; type++) {
for (int sec = 0; sec < sectorsCnt && !err && !trapped; ++sec) {
key = cjat91_saMifareChkKeys(sec * 4, type, NULL, size, &keyBlock[0], &key64);
@ -783,10 +788,18 @@ readysim:
case 7:
flags = FLAG_7B_UID_IN_DATA;
break;
default:
case 4:
flags = FLAG_4B_UID_IN_DATA;
break;
default:
flags = FLAG_UID_IN_EMUL;
break;
}
// Use UID, SAK, ATQA from EMUL, if uid not defined
//if ((flags & (FLAG_4B_UID_IN_DATA | FLAG_7B_UID_IN_DATA | FLAG_10B_UID_IN_DATA)) == 0) {
flags |= FLAG_UID_IN_EMUL;
//}
Mifare1ksim(flags | FLAG_MF_1K, 0, cjuid);
LED_C_OFF();
SpinOff(50);
@ -902,7 +915,7 @@ int e_MifareECardLoad(uint32_t numofsectors, uint8_t keytype) {
return (isOK) ? PM3_SUCCESS : PM3_EUNDEF;
}
/* the chk function is a piwi’ed(tm) check that will try all keys for
/* the chk function is a piwi'ed(tm) check that will try all keys for
a particular sector. also no tracing no dbg */
int cjat91_saMifareChkKeys(uint8_t blockNo, uint8_t keyType, bool clearTrace, uint8_t keyCount, uint8_t *datain, uint64_t *key) {
DBGLEVEL = DBG_NONE;