Compare commits

...

2 Commits

9 changed files with 824 additions and 69 deletions

View File

@ -24,6 +24,7 @@ SET(src_files
${CMAKE_SOURCE_DIR}/chroma/system/memory/abstract_allocator.c
${CMAKE_SOURCE_DIR}/chroma/system/memory/physmem.c
${CMAKE_SOURCE_DIR}/chroma/system/drivers/keyboard.c
${CMAKE_SOURCE_DIR}/chroma/system/drivers/elf.c
)
SET(lib_files

View File

@ -11,6 +11,30 @@
extern "C" {
#endif
/* ELF headers hate me.
* Let's do this the hard way. */
#define ELF_IDENT_MAGIC_OFF 0x0
#define ELF_IDENT_CLASS_OFF 0x4
#define ELF_IDENT_DATA_OFF 0x5
#define ELF_IDENT_VERSION_OFF 0x6
#define ELF_IDENT_OSABI_OFF 0x7
#define ELF_IDENT_ABI_VERSION_OFF 0x8
#define ELF_IDENT_PAD_OFF 0x9
#define ELFTYPE_OFF 0x10
#define ELFMACHINE_OFF 0x12
#define ELFVERSION_OFF 0x14
#define ELFENTRY_OFF 0x18
#define ELFPHOFF_OFF 0x20
#define ELFSHOFF_OFF 0x28
#define ELFFLAGS_OFF 0x30
#define ELFEHSIZE_OFF 0x34
#define ELFPHENTSIZE_OFF 0x36
#define ELFPHNUM_OFF 0x38
#define ELFSHENTSIZE_OFF 0x3A
#define ELFSHNUM_OFF 0x3C
#define ELFSHSTRNDX_OFF 0x40
#define BOOT_MAGIC "BOOT"
/* minimum protocol level:
@ -108,30 +132,6 @@ typedef struct {
} __attribute__((packed)) bootinfo;
typedef struct {
uint32_t Magic;
uint8_t Class;
uint8_t Endianness;
uint8_t Version;
uint8_t ABI;
uint8_t ABIVersion;
uint8_t Unused[7];
uint16_t Type;
uint16_t TargetArchitecture;
uint32_t ELFVersion;
size_t EntryPoint;
size_t ProgramHeadersTable;
size_t SectionHeadersTable;
uint32_t Flags;
uint16_t ELFHeaderSize;
uint16_t ProgramHeadersEntrySize;
uint16_t ProgramHeaderEntries;
uint16_t SectionHeadersEntrySize;
uint16_t SectionHeaderEntries;
uint16_t SectionHeaderNameEntry;
uint8_t End;
} __attribute__((packed)) ELF64Header_t;
#ifdef __cplusplus
}

View File

@ -69,6 +69,8 @@ void InitPrint();
void SetupInitialGDT();
void SetupIDT();
int ParseKernelHeader(size_t InitrdPtr);
int Main();
void Exit();

View File

@ -8,8 +8,6 @@
*** Chroma ***
***********************/
size_t KernelLocation;
/************************************************
* C O N S T A N T S A N D M A C R O S
@ -30,6 +28,11 @@ size_t KernelLocation;
#define REINTERPRET_CAST(target, intermediate, value) ((target*)((intermediate*)value))
#define FIXENDIAN64(x) __builtin_bswap64(x)
#define FIXENDIAN32(x) __builtin_bswap32(x)
#define FIXENDIAN16(x) __builtin_bswap16(x)
#define CONCAT(x, y) x ## y
#define CONCAT2(x, y) CONCAT(x, y)
#define ASSERT(exp, error) \
@ -55,6 +58,7 @@ size_t KernelLocation;
#define ERR_INST 0x10
#define ELF64MAGIC 0x7F454c46
#define ELF64MAGICBE 0x464c457F
/*
@ -106,7 +110,7 @@ size_t KernelLocation;
#define MMIO_REGION 0xFFFFFFFFF8000000ull // Cannot move!
#define FB_REGION 0xFFFFFFFFFC000000ull // Cannot move!
#define FB_PHYSICAL 0x00000000E0000000ull // Physical location of the Framebuffer
#define FB_PHYSICAL 0x00000000FD000000ull // Physical location of the Framebuffer
#define KERNEL_REGION 0xFFFFFFFFFFE00000ull // -2MiB, from bootloader
#define USER_REGION 0x00007FFFFFFFFFFFull // Not needed yet, but we're higher half so we might as well be thorough

View File

@ -20,48 +20,13 @@ address_space_t KernelAddressSpace;
int Main(void) {
KernelAddressSpace = (address_space_t) {0};
KernelLocation = 0x112600;
SerialPrintf("\r\n[ boot] Booting Chroma..\r\n");
SerialPrintf("[ boot] Kernel loaded at 0x%p, ends at 0x%p, is %d bytes long.\r\n", KernelAddr, KernelEnd, KernelEnd - KernelAddr);
SerialPrintf("[ boot] Initrd is physically at 0x%p, and is %d bytes long.\r\n", bootldr.initrd_ptr, bootldr.initrd_size);
SerialPrintf("[ boot] Initrd's header is 0x%p\r\n", FIXENDIAN32(*((volatile uint32_t*)(bootldr.initrd_ptr))));
SerialPrintf("[ boot] Searching for kernel... Constants start at 0x%p\r\n", &_kernel_text_start);
// We stop at the constants in the kernel, otherwise we'll read the constant ELF64MAGIC which is stored inside the kernel...
size_t headerLoc = 0;
for(size_t i = KernelAddr; i < KernelEnd; i++) {
if(i < (size_t) (&_kernel_text_start) - KernelAddr) {
if(*((volatile uint32_t*)(i)) == ELF64MAGIC) {
SerialPrintf("[ boot] Matched kernel header at 0x%p.\r\n", i);
}
}
}
int flag = 0;
if(headerLoc) {
ELF64Header_t* PotentialKernelHeader = (ELF64Header_t*) &headerLoc;
SerialPrintf(
"[ boot] Considering ELF with:\r\n\tBitness %d\r\n\tEntry point 0x%p\r\n\tFile type %s : %d\r\n\tArchitecture %s : %d\r\n",
PotentialKernelHeader->Class == 2 ? 64 : 32, PotentialKernelHeader->EntryPoint, PotentialKernelHeader->Type == 0x02 ? "EXECUTABLE" : "OTHER", PotentialKernelHeader->Type, PotentialKernelHeader->TargetArchitecture == 0x3E ? "AMD64" : "OTHER", PotentialKernelHeader->TargetArchitecture);
if(PotentialKernelHeader->EntryPoint == KernelAddr) {
SerialPrintf("[ boot] Header at 0x%p matches kernel header.\r\n", headerLoc);
flag = 1;
}
if(!flag) {
for(size_t i = 0; i < 8; i++) {
SerialPrintf("[ boot] Header dump part %d: 0x%x\r\n", i, *((volatile uint32_t*)(headerLoc + i)));
}
}
}
if(!flag) {
SerialPrintf("[ boot] Unable to find kernel in memory. Fatal error.\r\n");
//for(;;) {}
}
ParseKernelHeader(bootldr.initrd_ptr);
SerialPrintf("[ boot] The bootloader has put the paging tables at 0x%p.\r\n", ReadControlRegister(3));

View File

@ -0,0 +1,91 @@
#include <kernel/chroma.h>
/************************
*** Team Kitty, 2020 ***
*** Chroma ***
***********************/
/*
* This file provides utility functions for parsing ELF headers.
* This exists so that the kernel can find itself for remapping,
* but I may end up using ELF as the kernel's executable format.
* Writing an ELF loader is on the to-do list, after all.
! This needs to be cleaned up.
! This creates a mess of numbers on the print
! This is hacky and hardcoded as heck and needs to be fixed
*/
extern size_t KernelLocation;
int ParseKernelHeader(size_t InitrdPtr) {
int flag = 0;
SerialPrintf("[ boot] Searching for kernel... Constants start at 0x%p / 0x%p\r\n", ((size_t) (&_kernel_text_start) - KernelAddr) + InitrdPtr, (size_t) (&_kernel_text_start));
// We stop at the constants in the kernel, otherwise we'll read the constant ELF64MAGIC which is stored inside the kernel...
size_t headerLoc = 0;
for(size_t i = InitrdPtr; i < ((size_t) (&_kernel_text_start) - KernelAddr) + InitrdPtr; i++) {
if(*((volatile uint32_t*)(i)) == ELF64MAGIC) {
SerialPrintf("[ boot] Matched kernel header at 0x%p.\r\n", i);
headerLoc = i;
}
if(FIXENDIAN32(*((volatile uint32_t*)(i))) == ELF64MAGIC) {
SerialPrintf("[ boot] Matched little-endian kernel header at 0x%p.\r\n", i);
headerLoc = i;
}
}
if(headerLoc) {
/* For whatever reason, reading a size_t here fails. The max that seems to work is uint16_t, so we read in the
* 64 bit address by constructing it from 4 individual reads.
* Note that these 4 reads are little endian, so we need to flip them around individually
*/
uint16_t EntryPoint0 = FIXENDIAN16(*((volatile uint16_t*)(headerLoc + ELFENTRY_OFF)));
uint16_t EntryPoint1 = FIXENDIAN16(*((volatile uint16_t*)(headerLoc + ELFENTRY_OFF + 2)));
uint16_t EntryPoint2 = FIXENDIAN16(*((volatile uint16_t*)(headerLoc + ELFENTRY_OFF + 4)));
uint16_t EntryPoint3 = FIXENDIAN16(*((volatile uint16_t*)(headerLoc + ELFENTRY_OFF + 6)));
size_t EntryPoint = ((size_t) EntryPoint0 << 48) | ((size_t) EntryPoint1 << 32) | ((size_t) EntryPoint2 << 16) | ((size_t) EntryPoint3);
/* At this point, EntryPoint is a little-endian 64 bit integer. That means we have to fix its endianness in order to read it */
SerialPrintf("[ boot] Fixing entry point from 0x%p to 0x%p\r\n", EntryPoint, FIXENDIAN64(EntryPoint));
EntryPoint = FIXENDIAN64(EntryPoint);
/* Now we can continue as normal */
uint8_t HeaderClass = *((volatile uint8_t*)(headerLoc + ELF_IDENT_CLASS_OFF));
uint16_t ExecutableType = (uint16_t) *((volatile uint8_t*)(headerLoc + ELFTYPE_OFF));
uint16_t MachineType = (uint16_t) *((volatile uint8_t*)(headerLoc + ELFMACHINE_OFF));
SerialPrintf(
"[ boot] ELF header at 0x%p.\r\n\tConsidering ELF with:\r\n\tBitness %d: %d\r\n\tEntry point 0x%p\r\n\tFile type %s : 0x%x\r\n\tArchitecture %s : 0x%x\r\n",
headerLoc,
HeaderClass == 2 ? 64 : 32,
HeaderClass,
EntryPoint,
ExecutableType == FIXENDIAN16(0x0200) ? "EXECUTABLE" : "OTHER",
FIXENDIAN16(ExecutableType),
MachineType == FIXENDIAN16(0x3E00) ? "AMD64" : "OTHER",
FIXENDIAN16(MachineType));
if(EntryPoint == (size_t) (&_kernel_text_start)) {
SerialPrintf("[ boot] Header at 0x%p matches kernel header.\r\n", headerLoc);
flag = 1;
// At this point, we've found the right ELF64 executable!
// Great, now we can map it into the proper place
KernelLocation = headerLoc;
}
if(!flag) {
for(char i = 0; i < 64; i++) {
SerialPrintf("[ boot] Header dump part %x: 0x%x\r\n", i, *((volatile uint8_t*)(headerLoc + i)));
}
}
}
return flag;
}

View File

@ -52,6 +52,8 @@
extern size_t _kernel_rodata_start;
extern size_t _kernel_data_start;
size_t KernelLocation;
//__attribute__((aligned(4096))) static size_t Pagetable[512] = {0};
#define LAST_ENTRY 0xFF8
@ -151,8 +153,9 @@ void InitPaging() {
//TODO: Disallow execution of rodata and data, and bootldr/environment
for(void* Address = CAST(void*, KERNEL_REGION);
Address < CAST(void*, KERNEL_REGION + (KernelEnd - KernelAddr)); // Lower half of Kernel
Address < CAST(void*, KERNEL_REGION + (KernelEnd - KernelAddr));
Address = CAST(void*, CAST(char*, Address) + PAGE_SIZE)) {
SerialPrintf("[ mem] Mapping 0x%p to 0x%p, relative to kernel at 0x%p\r\n", (CAST(size_t, Address) - KERNEL_REGION) + KernelLocation, Address, (CAST(size_t, Address) - KERNEL_REGION));
MapVirtualMemory(&KernelAddressSpace, Address, (CAST(size_t, Address) - KERNEL_REGION) + KernelLocation, MAP_EXEC);
}
@ -161,9 +164,9 @@ void InitPaging() {
Address = CAST(void*, CAST(char*, Address) + PAGE_SIZE)) {
MapVirtualMemory(&KernelAddressSpace, Address, (CAST(size_t, Address) - KERNEL_REGION) + KERNEL_PHYSICAL_2, MAP_EXEC);
}*/
SerialPrintf("[ mem] Framebuffer at 0x%p, is 0x%p long. Mapping to 0x%p.\r\n", bootldr.fb_ptr, bootldr.fb_size, FB_REGION);
for(void* Address = CAST(void*, FB_REGION);
Address < CAST(void*, 0x200000); // TODO: Turn this into a calculation with bootldr.fb_size
Address < CAST(void*, bootldr.fb_size + FB_REGION);
Address = CAST(void*, CAST(char*, Address) + PAGE_SIZE)) {
MapVirtualMemory(&KernelAddressSpace, Address, (CAST(size_t, Address) - FB_REGION) + FB_PHYSICAL, MAP_WRITE);
}

View File

@ -0,0 +1,310 @@
void InitPagingT() {
size_t* PML4 = (size_t*) 0xFFA000; // Layer 4
size_t* PDPE_RAM = (size_t*) 0xFFE000; // Layer 3, contains map for the first 4GB of RAM
size_t* PDE_RAM = (size_t*) 0xFFF000;
size_t* PDPE_KERNEL = (size_t*) 0xFFB000; // Layer 3, contains map for the Kernel and everything it needs to run.
size_t* PDE_KERNEL_FB = (size_t*) 0xFFC000; // Layer 2, contains map for the linear framebuffer.
size_t* PT_KERNEL = (size_t*) 0xFFD000; // Layer 1, the page table for the kernel itself.
size_t fb_ptr = (size_t) &fb;
SET_ADDRESS(PML4, PDPE_RAM); // 3rd Layer entry for RAM
SET_ADDRESS(PML4 + LAST_ENTRY, PDPE_KERNEL); // 3rd Layer entry for Kernel
SET_ADDRESS(PDPE_KERNEL + LAST_ENTRY, PDE_KERNEL_FB); // 2nd Layer entry for the framebuffer
// Set the 480th entry (PDE_KERNEL_FB + (480 * 8))
// To the framebuffer + flags
SET_ADDRESS(PDE_KERNEL_FB + 3840, USERWRITEABLE_FLAGS(fb_ptr));
// In 4 byte increments, we're gonna map 3840 (the framebuffer)
// Up to (4096 - 8) in the PDE_KERNEL_FB with 2MB paging.
size_t MappingIterations = 1;
for(size_t i = 3844; i < 4088; i += 4) {
SET_ADDRESS(PDE_KERNEL_FB + i, USERWRITEABLE_FLAGS(fb_ptr) + (MappingIterations * (2 * MiB)));
MappingIterations++;
}
// Now we map the last entry of PDE_KERNEL_FB to our Page Table
SET_ADDRESS(PDE_KERNEL_FB + LAST_ENTRY, PT_KERNEL);
// Mapping the kernel into the page tables....
SET_ADDRESS(PT_KERNEL, 0xFF8001); // bootldr, bootinfo
SET_ADDRESS(PT_KERNEL + 8, 0xFF9001); // environment
// Map the kernel itself
SET_ADDRESS(PT_KERNEL + 16, KernelAddr + 1);
// Iterate through the pages, identity mapping each one
MappingIterations = 1;
size_t MappingOffset = 0x14;
for(size_t i = 0; i < ((KernelEnd - KernelAddr) >> 12); i++) {
// Page Table + (0x10 increasing by 0x04 each time) = x * 4KiB
SET_ADDRESS(PT_KERNEL + MappingOffset, (MappingIterations * (4 * KiB)));
MappingOffset += 4;
MappingIterations++;
}
// Now we need to map the core stacks. Top-down, from 0xDFF8
// There's always at least one core, so we do that one fixed.
// TODO: Account for 0-core CPUs
SET_ADDRESS(PT_KERNEL + LAST_ENTRY, 0xF14003);
MappingIterations = 1;
// For every core:
for(size_t i = 0; i < (bootldr.numcores + 3U) >> 2; i++) {
// PT_KERNEL[512 - (iterations + 1)] = 0x14003 + (iterations * page-width)
SET_ADDRESS(PT_KERNEL + LAST_ENTRY - (MappingIterations * 8), 0xF14003 + (4096 * MappingIterations));
MappingIterations++;
}
SET_ADDRESS(PDPE_RAM, PDE_RAM + PAGE_PRESENT + PAGE_RW);
SET_ADDRESS(PDPE_RAM + 8, 0xF10000 + PAGE_PRESENT + PAGE_RW);
SET_ADDRESS(PDPE_RAM + 16, 0xF11000 + PAGE_PRESENT + PAGE_RW);
SET_ADDRESS(PDPE_RAM + 24, 0xF12000 + PAGE_PRESENT + PAGE_RW);
// Identity map 4GB of ram
// Each page table can only hold 512 entries, but we
// just set up 4 of them - overflowing PDE_RAM (0xF000)
// will take us into 0x10000, into 0x11000, into 0x120000.
for(size_t i = 0; i < 512 * 4/*GB*/; i++) {
// add PDE_RAM, 4
// mov eax, 0x83
// add eax, 2*1024*1024
SET_ADDRESS(PDE_RAM + (i * 4), USERWRITEABLE_FLAGS(i * (2 * MiB)));
}
// Map first 2MB of memory
SET_ADDRESS(PDE_RAM, 0xF13000 + PAGE_PRESENT + PAGE_RW);
for(size_t i = 0; i < 512; i++) {
SET_ADDRESS(0xF13000 + i * 4, i * (4 * KiB) + PAGE_PRESENT + PAGE_RW);
}
// 0xA000 should now contain our memory map.
}
void TraversePageTables() {
}
void InitPagingOldImpl() {
// Disable paging so that we can work with the pagetable
//size_t registerTemp = ReadControlRegister(0);
//UNSET_PGBIT(registerTemp);
//WriteControlRegister(0, registerTemp);
// Clear space for our pagetable
size_t PagetableDest = 0x1000;
memset((char*)PagetableDest, 0, 4096);
// Start setting pagetable indexes
*((size_t*)PagetableDest) = 0x2003; // PDP at 0x2000, present & r/w
*((size_t*)PagetableDest + 0x1000) = 0x3003; // PDT at 0x3000, present & r/w
*((size_t*)PagetableDest + 0x2000) = 0x4003; // PT at 0x4000, present & r/w
size_t value = 0x3;
size_t offset = 8;
for(size_t i = 0; i < 512; i++) { // 512 iterations (entries into the page table)
*((size_t*) PagetableDest + offset) = value; // We're setting 512 bytes with x003
// (identity mapping the first 4 megabytes of memory)
// (mapping the page table to itself)
value += 4096; // Point to start of next page
offset += 8; // + 8 bytes (next entry in list)
}
// Enable PAE paging
size_t reg = ReadControlRegister(4);
SET_PAEBIT(reg);
WriteControlRegister(4, reg);
WriteControlRegister(3, PagetableDest);
}
/* size_t registerTemp = ReadControlRegister(4);
if(registerTemp & (1 << 7)) {
TOGGLE_PGEBIT(registerTemp);
WriteControlRegister(4, registerTemp);
}
if(registerTemp & (1 << 7))
WriteControlRegister(4, registerTemp ^ (1 << 7));
size_t CPUIDReturn;
asm volatile("cpuid" : "=d" (CPUIDReturn) : "a" (0x80000001) : "%rbx", "%rcx");
if(CPUIDReturn & (1 << 26)) {
SerialPrintf("System supports 1GB pages.\r\n");
if(registerTemp & (1 << 12)) {
SerialPrintf("PML5 paging available - using that instead.\r\n");
if(MemorySize > (1ULL << 57))
SerialPrintf("System has over 128Petabytes of RAM. Please consider upgrading the OS on your supercomputer.\r\n");
size_t MaxPML5 = 1;
size_t MaxPML4 = 1;
size_t MaxPDP = 512;
size_t LastPML4Entry = 512;
size_t LastPDPEntry = 512;
size_t MemorySearchDepth = MemorySize;
while(MemorySearchDepth > (256ULL << 30)) {
MaxPML5++;
MemorySearchDepth -= (256ULL << 30);
}
if(MaxPML5 > 512)
MaxPML5 = 512;
if(MemorySearchDepth) {
LastPDPEntry = ( (MemorySearchDepth + ((1 << 30) - 1)) & (~0ULL << 30)) >> 30;
if(MaxPML5 > 512)
MaxPML5 = 512;
}
size_t PML4Size = PAGETABLE_SIZE * MaxPML5;
size_t PDPSize = PML4Size * MaxPML4;
size_t PML4Base = AllocatePagetable(PML4Size + PDPSize);
size_t PDPBase = PML4Base + PML4Size;
for(size_t PML5Entry = 0; PML5Entry < MaxPML5; PML5Entry++) {
Pagetable[PML5Entry] = PML4Base + (PML5Entry << 12);
if(PML5Entry == (MaxPML5 - 1))
MaxPML4 = LastPML4Entry;
for(size_t PML4Entry = 0; PML4Entry < MaxPML4; PML4Entry++) {
((size_t*) Pagetable[PML5Entry])[PML4Entry] = PDPBase + (((PML5Entry << 9) + PML5Entry) << 12);
if( (PML5Entry == (MaxPML5 - 1)) && (PML4Entry == (MaxPML4 -1)) )
MaxPDP = LastPDPEntry;
for(size_t PDPEntry = 0; PDPEntry < MaxPDP; PDPEntry++) {
((size_t* ) ((size_t* ) Pagetable[PML5Entry])[PML4Entry])[PDPEntry] = ( ((PML5Entry << 18) + (PML4Entry << 9) + PDPEntry) << 30) | (0x83);
}
((size_t* ) Pagetable[PML5Entry])[PML4Entry] |= 0x3;
}
Pagetable[PML5Entry] |= 0x3;
}
} else {
SerialPrintf("PML4 available - using that instead.\r\n");
size_t MemorySearchDepth = MemorySize;
if(MemorySearchDepth > (1ULL << 48))
SerialPrintf("RAM limited to 256TB.\r\n");
size_t MaxPML4 = 1;
size_t MaxPDP = 512;
size_t LastPDPEntry = 512;
while(MemorySearchDepth > (512ULL << 30)) {
MaxPML4++;
MemorySearchDepth -= (512ULL << 30);
}
if(MaxPML4 > 512)
MaxPML4 = 512;
if(MemorySearchDepth) {
LastPDPEntry = ( (MemorySearchDepth + ((1 << 30) - 1)) & (~0ULL << 30)) >> 30;
if(LastPDPEntry > 512)
LastPDPEntry = 512;
}
size_t PDPSize = PAGETABLE_SIZE * MaxPML4;
size_t PDPBase = AllocatePagetable(PDPSize);
for(size_t PML4Entry = 0; PML4Entry < MaxPML4; PML4Entry++) {
Pagetable[PML4Entry] = PDPBase + (PML4Entry << 12);
if(PML4Entry == (MaxPML4 - 1)) {
MaxPDP = LastPDPEntry;
}
for(size_t PDPEntry = 0; PDPEntry < MaxPDP; PDPEntry++) {
((size_t* ) Pagetable[PML4Entry])[PDPEntry] = (((PML4Entry << 9) + PDPEntry) << 30) | 0x83;
}
Pagetable[PML4Entry] |= 0x3;
}
}
} else {
SerialPrintf("System does not support 1GB pages - using 2MiB paging instead.\r\n");
size_t MemorySearchDepth = MemorySize;
if(MemorySearchDepth > (1ULL << 48)) {
SerialPrintf("Usable RAM is limited to 256TB, and the page table alone will use 1GB of space in memory.\r\n");
}
size_t MaxPML4 = 1, MaxPDP = 512, MaxPD = 512, LastPDPEntry = 1;
while(MemorySearchDepth > (512ULL << 30)) {
MaxPML4++;
MemorySearchDepth -= (512ULL << 30);
}
if(MaxPML4 > 512)
MaxPML4 = 512;
if(MemorySearchDepth) {
LastPDPEntry = ((MemorySearchDepth + ((1 << 30) - 1)) & (~0ULL << 30)) >> 30;
if(LastPDPEntry > 512)
LastPDPEntry = 512;
}
size_t PDPSize = PAGETABLE_SIZE * MaxPML4;
size_t PDSize = PDPSize * MaxPDP;
size_t PDPBase = AllocatePagetable(PDPSize + PDSize);
size_t PDBase = PDPBase + PDSize;
for(size_t PML4Entry = 0; PML4Entry < MaxPML4; PML4Entry++) {
Pagetable[PML4Entry] = PDBase + (PML4Entry << 12);
if(PML4Entry == (MaxPML4 - 1)) {
MaxPDP = LastPDPEntry;
}
for(size_t PDPEntry = 0; PDPEntry < MaxPDP; PDPEntry++) {
( (size_t* ) Pagetable[PML4Entry])[PDPEntry] = PDBase + (((PML4Entry << 9) + PDPEntry) << 12);
for(size_t PDEntry = 0; PDEntry < MaxPD; PDEntry++) {
( (size_t* ) ((size_t*) Pagetable[PML4Entry])[PDPEntry])[PDEntry] = (( (PML4Entry << 18) + (PDPEntry << 9) + PDPEntry) << 21) | 0x83;
}
( (size_t* ) Pagetable[PML4Entry])[PDPEntry] |= 0x3;
}
Pagetable[PML4Entry] |= 0x3;
}
}
WriteControlRegister(3, Pagetable);
registerTemp = ReadControlRegister(4);
if(!(registerTemp & (1 << 7))) {
TOGGLE_PGEBIT(registerTemp);
WriteControlRegister(4, registerTemp);
}*/

View File

@ -0,0 +1,379 @@
#include <kernel/chroma.h>
#include <lainlib/lainlib.h>
/************************
*** Team Kitty, 2020 ***
*** Chroma ***
***********************/
/****************************************
* W O R K I N P R O G R E S S *
****************************************
*
* This file contains functions for virtual memory management.
*
* Virtual Memory Management is still a work in progress.
* The functions here are hold-offs from old versions of the software implemented here, as well as from the EFI version of Chroma, called Sync.
*
* There, these functions worked, but here, under BIOS, it's a lot more difficult.
* It will take some time to get these functions working.
*
* The general plan, being that the BOOTBOOT loader has given us static addresses for all of our doodads,
* is to keep the core kernel where it is (FFFFFFFFFFE00000) and load in modules and libraries around it.
*
* We start in the higher half, so we'll dedicate the lower half (7FFFFFFFFFFF and below) to userspace.
*
* That means we have about 3 terabytes of RAM for the kernel.
* This will be identity mapped, always.
*
* Handily, since most modern processors ignore the highest 2 bytes of a virtual address, and the kernel
* is mapped to 0x80000000000 and above, we can use the nomenclature:
* * 0x00007FFFFFFFFFFF and below is user space.
* * 0xFFFF800000000000 and above is kernel space.
* The processor will ignore the first 4 chars, and this provides a great deal of readability for the
* future of the kernel.
*
* We'll have a kernel heap mapped into this kernel space, as well as a kernel stack (for task switching and error tracing).
* These will be 1GB each.
* We may have to increase this in the future, once Helix is fully integrated.
* Helix will take a lot of memory, as it is a fully featured 3D engine. We may have to implement things like
* texture streaming and mipmapping. Minimising RAM usage is NOT a priority for me, but it would be nice
* to have a minimum requirement above 32GB.
*
* // TODO: Expand Kernel Heap
*
*
* //TODO: there are lots of calls to AllocateFrame here, those need to be separated out into AllocateZeroFrame if necessary.
*
*
*/
extern size_t _kernel_text_start;
extern size_t _kernel_rodata_start;
extern size_t _kernel_data_start;
//__attribute__((aligned(4096))) static size_t Pagetable[512] = {0};
#define LAST_ENTRY 0xFF8
#define SET_ADDRESS(a,b) ((*(size_t*) (a)) = (size_t) b)
/*
* It turns out it's useful to have macros for the standard
* data size units.
*
* Who would've thoguht?
*/
#define KiB 1 * 1024
#define MiB 1 * 1024 * KiB
#define PAGE_PRESENT 1
#define PAGE_RW 2
#define PAGE_USER 4
#define PAGE_GLOBAL 8
#define USERWRITEABLE_FLAGS(a) ((a & 0xFFFFFF00) + 0x83)
// The AbstractAllocator control struct
static allocator_t Allocator = NULL;
// The AbstractAllocator Ticketlock.
static ticketlock_t AllocatorLock = {0};
// Entries to help allocate the Kernel Stack
static list_entry_t StackFreeList;
static ticketlock_t StackLock = {0};
static void* StackPointer = (void*) KERNEL_STACK_REGION;
// A temporary itoa function for better debugging..
const char* IntToAscii(int In) {
char* OutputBuffer = " ";
size_t Temp, i = 0, j = 0;
do {
Temp = In % 10;
OutputBuffer[i++] = (Temp < 10) ? (Temp + '0') : (Temp + 'a' - 10);
} while (In /= 10);
OutputBuffer[i--] = 0;
for(j = 0; j < i; j++, i--) {
Temp = OutputBuffer[j];
OutputBuffer[j] = OutputBuffer[i];
OutputBuffer[i] = Temp;
}
return OutputBuffer;
}
void InitPaging() {
StackFreeList = (list_entry_t) { &StackFreeList, &StackFreeList };
size_t Size = AlignUpwards(AllocatorSize(), PAGE_SIZE);
Allocator = PhysAllocateZeroMem(Size);
Allocator = CreateAllocatorWithPool(Allocator, Size);
SerialPrintf("[ Mem] Everything preallocated for paging.\n");
KernelAddressSpace = (address_space_t) {
.Lock = {0},
.PML4 = PhysAllocateZeroMem(PAGE_SIZE)
};
size_t* Pagetable = KernelAddressSpace.PML4;
//SerialPrintf("[ Mem] About to identity map the higher half.\n");
// Identity map the higher half
for(int i = 256; i < 512; i++) {
Pagetable[i] = (size_t)PhysAllocateZeroMem(PAGE_SIZE);
Pagetable[i] = (size_t)(((char*)Pagetable[i]) - DIRECT_REGION);
Pagetable[i] |= (PAGE_PRESENT | PAGE_RW);
//SerialPrintf("%d", i - 256);
}
SerialPrintf("[ Mem] Identity mapping higher half complete.\n");
MMapEnt* TopEntry = (MMapEnt*)(((&bootldr) + bootldr.size) - sizeof(MMapEnt));
size_t LargestAddress = TopEntry->ptr + TopEntry->size;
SerialPrintf("[ Mem] About to map lower memory into the Direct Region.\n");
for(size_t Address = 0; Address < AlignUpwards(LargestAddress, PAGE_SIZE); Address += PAGE_SIZE) {
MapVirtualMemory(&KernelAddressSpace, (size_t*)(((char*)Address) + DIRECT_REGION), Address, MAP_WRITE);
}
SerialPrintf("[ Mem] Lower half mapping complete.\n");
SerialPrintf("[ Mem] Mapping kernel into new memory map.\r\n");
//TODO: Disallow execution of rodata and data, and bootldr/environment
for(void* Address = CAST(void*, KERNEL_REGION);
Address < CAST(void*, KERNEL_REGION + 0x2000); // Lower half of Kernel
Address = CAST(void*, CAST(char*, Address) + PAGE_SIZE)) {
MapVirtualMemory(&KernelAddressSpace, Address, (CAST(size_t, Address) - KERNEL_REGION) + KERNEL_PHYSICAL, MAP_EXEC);
}
for(void* Address = CAST(void*, KERNEL_REGION + 0x2000);
Address < CAST(void*, KERNEL_REGION + 0x12000); // Higher half of kernel
Address = CAST(void*, CAST(char*, Address) + PAGE_SIZE)) {
MapVirtualMemory(&KernelAddressSpace, Address, (CAST(size_t, Address) - KERNEL_REGION) + KERNEL_PHYSICAL_2, MAP_EXEC);
}
for(void* Address = CAST(void*, FB_REGION);
Address < CAST(void*, 0x200000); // TODO: Turn this into a calculation with bootldr.fb_size
Address = CAST(void*, CAST(char*, Address) + PAGE_SIZE)) {
MapVirtualMemory(&KernelAddressSpace, Address, (CAST(size_t, Address) - FB_REGION) + FB_PHYSICAL, MAP_WRITE);
}
SerialPrintf("[ Mem] Kernel mapped into pagetables. New PML4 at 0x%p\r\n", KernelAddressSpace.PML4);
//ASSERT(Allocator != NULL);
}
static size_t GetCachingAttribute(pagecache_t Cache) {
switch (Cache) {
case CACHE_WRITE_BACK: return 0;
case CACHE_WRITE_THROUGH: return 1 << 2;
case CACHE_NONE: return 1 << 3;
case CACHE_WRITE_COMBINING: return 1 << 6;
}
return 1 << 3;
}
static bool ExpandAllocator(size_t NewSize) {
size_t AllocSize = AlignUpwards(AllocatorPoolOverhead() + sizeof(size_t) * 5 + NewSize, PAGE_SIZE);
void* Pool = PhysAllocateMem(AllocSize);
return AddPoolToAllocator(Allocator, Pool, AllocSize) != NULL;
}
static void GetPageFromTables(address_space_t* AddressSpace, size_t VirtualAddress, size_t** Page) {
//ASSERT(Page != NULL);
//ASSERT(AddressSpace != NULL);
size_t* Pagetable = AddressSpace->PML4;
for(int Level = 4; Level > 1; Level--) {
size_t* Entry = &Pagetable[(VirtualAddress >> (12u + 9u * (Level - 1))) & 0x1FFU];
ASSERT(*Entry & PAGE_PRESENT, "Page not present during retrieval");
Pagetable = (size_t*)((char*)(*Entry & 0x7ffffffffffff000ull) + DIRECT_REGION);
}
ASSERT(Pagetable[(VirtualAddress >> 12U) & 0x1FFU] & PAGE_PRESENT, "PDPE not present during retrieval");
*Page = &Pagetable[(VirtualAddress >> 12U) & 0x1FFU];
}
void SetAddressSpace(address_space_t* AddressSpace) {
//ASSERT(AddressSpace != NULL);
if((size_t)((char*)ReadControlRegister(3) + DIRECT_REGION) != (size_t) &AddressSpace->PML4) {
WriteControlRegister(3, CAST(size_t, &AddressSpace->PML4));
}
}
void MapVirtualMemory(address_space_t* AddressSpace, void* VirtualAddress, size_t PhysicalAddress, mapflags_t Flag) {
//bool MapGlobally = false;
size_t Virtual = (size_t)VirtualAddress;
//ASSERT(AddressSpace != NULL);
TicketAttemptLock(&AddressSpace->Lock);
size_t Flags = PAGE_PRESENT;
if(Flag & MAP_WRITE)
Flags |= MAP_WRITE;
if(Virtual < USER_REGION)
Flags |= PAGE_USER;
//TODO: Global mapping
size_t* Pagetable = AddressSpace->PML4;
for(int Level = 4; Level > 1; Level--) {
size_t* Entry = &Pagetable[(Virtual >> (12u + 9u * (Level - 1))) & 0x1FFu];
if(!(*Entry & PAGE_PRESENT)) {
directptr_t Pointer = PhysAllocateZeroMem(PAGE_SIZE);
*Entry = (size_t)(((char*)Pointer) + DIRECT_REGION);
}
*Entry |= Flags;
Pagetable = (size_t*)(((char*)(*Entry & 0x7ffffffffffff000ull) + DIRECT_REGION));
}
size_t* Entry = &Pagetable[(Virtual >> 12u) & 0x1FFu];
*Entry = Flags | PhysicalAddress;
if(AddressSpace != NULL) {
TicketUnlock(&AddressSpace->Lock);
}
}
void UnmapVirtualMemory(address_space_t* AddressSpace, void* VirtualAddress){
//ASSERT(AddressSpace != NULL);
TicketAttemptLock(&AddressSpace->Lock);
size_t* Entry;
GetPageFromTables(AddressSpace, (size_t)VirtualAddress, &Entry);
*Entry = 0;
InvalidatePage((size_t)VirtualAddress);
if(AddressSpace != NULL) {
TicketUnlock(&AddressSpace->Lock);
}
}
void CacheVirtualMemory(address_space_t* AddressSpace, void* VirtualAddress, pagecache_t Cache) {
//ASSERT(AddressSpace != NULL);
TicketAttemptLock(&AddressSpace->Lock);
size_t* Entry;
GetPageFromTables(AddressSpace, (size_t)VirtualAddress, &Entry);
*Entry &= ~((1 << 6) | (1 << 2) | (1 << 3));
*Entry |= GetCachingAttribute(Cache);
InvalidatePage((size_t)VirtualAddress);
if(AddressSpace != NULL) {
TicketUnlock(&AddressSpace->Lock);
}
}
void* AllocateMemory(size_t Bits) {
TicketAttemptLock(&AllocatorLock);
void* Result = AllocatorMalloc(Allocator, Bits);
if(Result == NULL) {
if(!ExpandAllocator(Bits)) {
TicketUnlock(&AllocatorLock);
return 0ULL;
}
Result = AllocatorMalloc(Allocator, Bits);
}
if(Result != NULL) {
memset(Result, 0, Bits);
}
TicketUnlock(&AllocatorLock);
return Result;
}
void* ReallocateMemory(void* Address, size_t NewSize) {
TicketAttemptLock(&AllocatorLock);
void* Result = AllocatorRealloc(Allocator, Address, NewSize);
if(Result == NULL) {
if(!ExpandAllocator(NewSize)) {
TicketUnlock(&AllocatorLock);
return 0ULL;
}
Result = AllocatorRealloc(Allocator, Address, NewSize);
}
TicketUnlock(&AllocatorLock);
return Result;
}
void FreeMemory(void* Address) {
TicketAttemptLock(&AllocatorLock);
AllocatorFree(Allocator, Address);
TicketUnlock(&AllocatorLock);
}
void* AllocateKernelStack() {
void* StackAddress = NULL;
size_t StackSize = PAGE_SIZE * 4;
TicketAttemptLock(&StackLock);
if(ListIsEmpty(&StackFreeList)) {
StackAddress = StackPointer;
StackPointer = (void*)(((char*)StackPointer) + (4*KiB) + StackSize);
for(size_t i = 0; i < (StackSize / PAGE_SIZE); i++) {
directptr_t NewStack;
NewStack = PhysAllocateZeroMem(PAGE_SIZE);
MapVirtualMemory(&KernelAddressSpace, (void*)((size_t)StackAddress + i * PAGE_SIZE), (size_t)((char*)NewStack) - DIRECT_REGION, MAP_WRITE);
}
} else {
list_entry_t* StackEntry = StackFreeList.Next;
ListRemove(StackEntry);
memset(StackEntry, 0, StackSize);
StackAddress = (void*)StackEntry;
}
TicketUnlock(&StackLock);
StackAddress = (void*)((size_t)StackAddress + StackSize);
StackAddress = (void*)((size_t)StackAddress - sizeof(size_t) * 2);
return StackAddress;
}
void FreeKernelStack(void* StackAddress) {
TicketAttemptLock(&StackLock);
list_entry_t* ListEntry = (list_entry_t*)(((size_t)(StackAddress) + (sizeof(size_t) * 2)) - (PAGE_SIZE * 4));
ListAdd(&StackFreeList, ListEntry);
TicketUnlock(&StackLock);
}