ladybird/Kernel/Storage/IDEChannel.cpp
Liav A 531037db7e Kernel: Remove support for CHS addressing
An article about IDE limits states that:
"Hard drives over 8.4 GB are supposed to report their geometry as
16383/16/63. This in effect means that the `geometry' is obsolete, and
the total disk size can no longer be computed from the geometry, but is
found in the LBA capacity field returned by the IDENTIFY command.
Hard drives over 137.4 GB are supposed to report an LBA capacity of
0xfffffff = 268435455 sectors (137438952960 bytes). Now the actual disk
size is found in the new 48-capacity field."
(https://tldp.org/HOWTO/Large-Disk-HOWTO-4.html) which is the main
reason to not support CHS as harddrives with less than 8.4 GB capacity
are completely obsolete.

Another good reason is that virtually any harddrive in the last 20 years
or so, supports LBA mode. Therefore, it's probably OK to just ignore CHS
as it's unlikely to encounter a harddrive that doesn't support LBA.

This is somewhat simplifying the IDE initialization and access code.
Also, we should use the ATAIdentifyBlock structure if possible,
so now we do it instead of using macros to calculate offsets.

With the usage of the ATAIdentifyBlock structure, we now use the
48-bit LBA max count if the drive indicates it supports 48-bit LBA mode.
2021-03-27 16:40:16 +01:00

584 lines
22 KiB
C++

/*
* Copyright (c) 2018-2021, Andreas Kling <kling@serenityos.org>
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* 1. Redistributions of source code must retain the above copyright notice, this
* list of conditions and the following disclaimer.
*
* 2. Redistributions in binary form must reproduce the above copyright notice,
* this list of conditions and the following disclaimer in the documentation
* and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
* DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
* SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
* CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
* OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <AK/ByteBuffer.h>
#include <AK/Singleton.h>
#include <AK/StringView.h>
#include <Kernel/FileSystem/ProcFS.h>
#include <Kernel/IO.h>
#include <Kernel/Process.h>
#include <Kernel/Storage/ATA.h>
#include <Kernel/Storage/IDEChannel.h>
#include <Kernel/Storage/IDEController.h>
#include <Kernel/Storage/PATADiskDevice.h>
#include <Kernel/VM/MemoryManager.h>
#include <Kernel/WorkQueue.h>
namespace Kernel {
#define PATA_PRIMARY_IRQ 14
#define PATA_SECONDARY_IRQ 15
#define PCI_Mass_Storage_Class 0x1
#define PCI_IDE_Controller_Subclass 0x1
UNMAP_AFTER_INIT NonnullOwnPtr<IDEChannel> IDEChannel::create(const IDEController& controller, IOAddressGroup io_group, ChannelType type, bool force_pio)
{
return make<IDEChannel>(controller, io_group, type, force_pio);
}
RefPtr<StorageDevice> IDEChannel::master_device() const
{
return m_master;
}
RefPtr<StorageDevice> IDEChannel::slave_device() const
{
return m_slave;
}
UNMAP_AFTER_INIT IDEChannel::IDEChannel(const IDEController& controller, IOAddressGroup io_group, ChannelType type, bool force_pio)
: IRQHandler(type == ChannelType::Primary ? PATA_PRIMARY_IRQ : PATA_SECONDARY_IRQ)
, m_channel_type(type)
, m_io_group(io_group)
, m_parent_controller(controller)
{
disable_irq();
// FIXME: The device may not be capable of DMA.
m_dma_enabled.resource() = !force_pio;
ProcFS::add_sys_bool("ide_dma", m_dma_enabled);
initialize(force_pio);
detect_disks();
// Note: calling to detect_disks could generate an interrupt, clear it if that's the case
clear_pending_interrupts();
enable_irq();
}
void IDEChannel::clear_pending_interrupts() const
{
m_io_group.io_base().offset(ATA_REG_STATUS).in<u8>();
}
UNMAP_AFTER_INIT IDEChannel::~IDEChannel()
{
}
void IDEChannel::start_request(AsyncBlockDeviceRequest& request, bool use_dma, bool is_slave, u16 capabilities)
{
ScopedSpinLock lock(m_request_lock);
dbgln_if(PATA_DEBUG, "IDEChannel::start_request");
m_current_request = &request;
m_current_request_block_index = 0;
m_current_request_uses_dma = use_dma;
m_current_request_flushing_cache = false;
if (request.request_type() == AsyncBlockDeviceRequest::Read) {
if (use_dma)
ata_read_sectors_with_dma(is_slave, capabilities);
else
ata_read_sectors(is_slave, capabilities);
} else {
if (use_dma)
ata_write_sectors_with_dma(is_slave, capabilities);
else
ata_write_sectors(is_slave, capabilities);
}
}
void IDEChannel::complete_current_request(AsyncDeviceRequest::RequestResult result)
{
// NOTE: this may be called from the interrupt handler!
VERIFY(m_current_request);
VERIFY(m_request_lock.is_locked());
// Now schedule reading back the buffer as soon as we leave the irq handler.
// This is important so that we can safely write the buffer back,
// which could cause page faults. Note that this may be called immediately
// before Processor::deferred_call_queue returns!
g_io_work->queue([this, result]() {
dbgln_if(PATA_DEBUG, "IDEChannel::complete_current_request result: {}", (int)result);
ScopedSpinLock lock(m_request_lock);
VERIFY(m_current_request);
auto& request = *m_current_request;
m_current_request = nullptr;
if (m_current_request_uses_dma) {
if (result == AsyncDeviceRequest::Success) {
if (request.request_type() == AsyncBlockDeviceRequest::Read) {
if (!request.write_to_buffer(request.buffer(), m_dma_buffer_page->paddr().offset(0xc0000000).as_ptr(), 512 * request.block_count())) {
lock.unlock();
request.complete(AsyncDeviceRequest::MemoryFault);
return;
}
}
// I read somewhere that this may trigger a cache flush so let's do it.
m_io_group.bus_master_base().offset(2).out<u8>(m_io_group.bus_master_base().offset(2).in<u8>() | 0x6);
}
}
lock.unlock();
request.complete(result);
});
}
UNMAP_AFTER_INIT void IDEChannel::initialize(bool force_pio)
{
m_parent_controller->enable_pin_based_interrupts();
dbgln_if(PATA_DEBUG, "IDEChannel: {} IO base: {}", channel_type_string(), m_io_group.io_base());
dbgln_if(PATA_DEBUG, "IDEChannel: {} control base: {}", channel_type_string(), m_io_group.control_base());
dbgln_if(PATA_DEBUG, "IDEChannel: {} bus master base: {}", channel_type_string(), m_io_group.bus_master_base());
if (force_pio) {
dbgln("IDEChannel: Requested to force PIO mode; not setting up DMA");
return;
}
// Let's try to set up DMA transfers.
PCI::enable_bus_mastering(m_parent_controller->pci_address());
m_prdt_page = MM.allocate_supervisor_physical_page();
prdt().end_of_table = 0x8000;
m_dma_buffer_page = MM.allocate_supervisor_physical_page();
}
static void print_ide_status(u8 status)
{
dbgln("IDEChannel: print_ide_status: DRQ={} BSY={}, DRDY={}, DSC={}, DF={}, CORR={}, IDX={}, ERR={}",
(status & ATA_SR_DRQ) != 0,
(status & ATA_SR_BSY) != 0,
(status & ATA_SR_DRDY) != 0,
(status & ATA_SR_DSC) != 0,
(status & ATA_SR_DF) != 0,
(status & ATA_SR_CORR) != 0,
(status & ATA_SR_IDX) != 0,
(status & ATA_SR_ERR) != 0);
}
void IDEChannel::try_disambiguate_error()
{
dbgln("IDEChannel: Error cause:");
switch (m_device_error) {
case ATA_ER_BBK:
dbgln("IDEChannel: - Bad block");
break;
case ATA_ER_UNC:
dbgln("IDEChannel: - Uncorrectable data");
break;
case ATA_ER_MC:
dbgln("IDEChannel: - Media changed");
break;
case ATA_ER_IDNF:
dbgln("IDEChannel: - ID mark not found");
break;
case ATA_ER_MCR:
dbgln("IDEChannel: - Media change request");
break;
case ATA_ER_ABRT:
dbgln("IDEChannel: - Command aborted");
break;
case ATA_ER_TK0NF:
dbgln("IDEChannel: - Track 0 not found");
break;
case ATA_ER_AMNF:
dbgln("IDEChannel: - No address mark");
break;
default:
dbgln("IDEChannel: - No one knows");
break;
}
}
void IDEChannel::handle_irq(const RegisterState&)
{
u8 status = m_io_group.io_base().offset(ATA_REG_STATUS).in<u8>();
m_entropy_source.add_random_event(status);
u8 bstatus = m_io_group.bus_master_base().offset(2).in<u8>();
if (!(bstatus & 0x4)) {
// interrupt not from this device, ignore
dbgln_if(PATA_DEBUG, "IDEChannel: ignore interrupt");
return;
}
ScopedSpinLock lock(m_request_lock);
dbgln_if(PATA_DEBUG, "IDEChannel: interrupt: DRQ={}, BSY={}, DRDY={}",
(status & ATA_SR_DRQ) != 0,
(status & ATA_SR_BSY) != 0,
(status & ATA_SR_DRDY) != 0);
if (!m_current_request) {
dbgln("IDEChannel: IRQ but no pending request!");
return;
}
if (status & ATA_SR_ERR) {
print_ide_status(status);
m_device_error = m_io_group.io_base().offset(ATA_REG_ERROR).in<u8>();
dbgln("IDEChannel: Error {:#02x}!", (u8)m_device_error);
try_disambiguate_error();
complete_current_request(AsyncDeviceRequest::Failure);
return;
}
m_device_error = 0;
if (m_current_request_uses_dma) {
complete_current_request(AsyncDeviceRequest::Success);
return;
}
// Now schedule reading/writing the buffer as soon as we leave the irq handler.
// This is important so that we can safely access the buffers, which could
// trigger page faults
Processor::deferred_call_queue([this]() {
ScopedSpinLock lock(m_request_lock);
if (m_current_request->request_type() == AsyncBlockDeviceRequest::Read) {
dbgln_if(PATA_DEBUG, "IDEChannel: Read block {}/{}", m_current_request_block_index, m_current_request->block_count());
if (ata_do_read_sector()) {
if (++m_current_request_block_index >= m_current_request->block_count()) {
complete_current_request(AsyncDeviceRequest::Success);
return;
}
// Wait for the next block
enable_irq();
}
} else {
if (!m_current_request_flushing_cache) {
dbgln_if(PATA_DEBUG, "IDEChannel: Wrote block {}/{}", m_current_request_block_index, m_current_request->block_count());
if (++m_current_request_block_index >= m_current_request->block_count()) {
// We read the last block, flush cache
VERIFY(!m_current_request_flushing_cache);
m_current_request_flushing_cache = true;
m_io_group.io_base().offset(ATA_REG_COMMAND).out<u8>(ATA_CMD_CACHE_FLUSH);
} else {
// Read next block
ata_do_write_sector();
}
} else {
complete_current_request(AsyncDeviceRequest::Success);
}
}
});
}
static void io_delay()
{
for (int i = 0; i < 4; ++i)
IO::in8(0x3f6);
}
void IDEChannel::wait_until_not_busy()
{
while (m_io_group.control_base().in<u8>() & ATA_SR_BSY)
;
}
String IDEChannel::channel_type_string() const
{
if (m_channel_type == ChannelType::Primary)
return "Primary";
return "Secondary";
}
UNMAP_AFTER_INIT void IDEChannel::detect_disks()
{
auto channel_string = [](u8 i) -> const char* {
if (i == 0)
return "master";
return "slave";
};
// There are only two possible disks connected to a channel
for (auto i = 0; i < 2; i++) {
m_io_group.io_base().offset(ATA_REG_HDDEVSEL).out<u8>(0xA0 | (i << 4)); // First, we need to select the drive itself
m_io_group.io_base().offset(ATA_REG_COMMAND).out<u8>(ATA_CMD_IDENTIFY); // Send the ATA_IDENTIFY command
// Wait for the BSY flag to be reset
while (m_io_group.control_base().in<u8>() & ATA_SR_BSY)
;
if (m_io_group.control_base().in<u8>() == 0x00) {
dbgln_if(PATA_DEBUG, "IDEChannel: No {} {} disk detected!", channel_type_string().to_lowercase(), channel_string(i));
continue;
}
bool check_for_atapi = false;
PATADiskDevice::InterfaceType interface_type = PATADiskDevice::InterfaceType::ATA;
for (;;) {
u8 status = m_io_group.control_base().in<u8>();
if (status & ATA_SR_ERR) {
dbgln_if(PATA_DEBUG, "IDEChannel: {} {} device is not ATA. Will check for ATAPI.", channel_type_string(), channel_string(i));
check_for_atapi = true;
break;
}
if (!(status & ATA_SR_BSY) && (status & ATA_SR_DRQ)) {
dbgln_if(PATA_DEBUG, "IDEChannel: {} {} device appears to be ATA.", channel_type_string(), channel_string(i));
interface_type = PATADiskDevice::InterfaceType::ATA;
break;
}
}
if (check_for_atapi) {
u8 cl = m_io_group.io_base().offset(ATA_REG_LBA1).in<u8>();
u8 ch = m_io_group.io_base().offset(ATA_REG_LBA2).in<u8>();
if ((cl == 0x14 && ch == 0xEB) || (cl == 0x69 && ch == 0x96)) {
interface_type = PATADiskDevice::InterfaceType::ATAPI;
dbgln("IDEChannel: {} {} device appears to be ATAPI. We're going to ignore it for now as we don't support it.", channel_type_string(), channel_string(i));
continue;
} else {
dbgln("IDEChannel: {} {} device doesn't appear to be ATA or ATAPI. Ignoring it.", channel_type_string(), channel_string(i));
continue;
}
}
ByteBuffer wbuf = ByteBuffer::create_uninitialized(512);
ByteBuffer bbuf = ByteBuffer::create_uninitialized(512);
u8* b = bbuf.data();
u16* w = (u16*)wbuf.data();
for (u32 i = 0; i < 256; ++i) {
u16 data = m_io_group.io_base().offset(ATA_REG_DATA).in<u16>();
*(w++) = data;
*(b++) = MSB(data);
*(b++) = LSB(data);
}
// "Unpad" the device name string.
for (u32 i = 93; i > 54 && bbuf[i] == ' '; --i)
bbuf[i] = 0;
volatile ATAIdentifyBlock& identify_block = (volatile ATAIdentifyBlock&)(*wbuf.data());
u16 capabilities = identify_block.capabilites[0];
// If the drive is so old that it doesn't support LBA, ignore it.
if (!(capabilities & ATA_CAP_LBA))
continue;
u64 max_addressable_block = identify_block.max_28_bit_addressable_logical_sector;
// if we support 48-bit LBA, use that value instead.
if (identify_block.commands_and_feature_sets_supported[1] & (1 << 10))
max_addressable_block = identify_block.user_addressable_logical_sectors_count;
dbgln("IDEChannel: {} {} {} device found: Name={}, Capacity={}, Capabilities=0x{:04x}", channel_type_string(), channel_string(i), interface_type == PATADiskDevice::InterfaceType::ATA ? "ATA" : "ATAPI", ((char*)bbuf.data() + 54), max_addressable_block * 512, capabilities);
if (i == 0) {
m_master = PATADiskDevice::create(m_parent_controller, *this, PATADiskDevice::DriveType::Master, interface_type, capabilities, max_addressable_block);
} else {
m_slave = PATADiskDevice::create(m_parent_controller, *this, PATADiskDevice::DriveType::Slave, interface_type, capabilities, max_addressable_block);
}
}
}
void IDEChannel::ata_access(Direction direction, bool slave_request, u64 lba, u8 block_count, u16 capabilities, bool use_dma)
{
LBAMode lba_mode;
u8 head = 0;
VERIFY(capabilities & ATA_CAP_LBA);
if (lba >= 0x10000000) {
lba_mode = LBAMode::FortyEightBit;
head = 0;
} else {
lba_mode = LBAMode::TwentyEightBit;
head = (lba & 0xF000000) >> 24;
}
wait_until_not_busy();
m_io_group.io_base().offset(ATA_REG_HDDEVSEL).out<u8>(0xE0 | (static_cast<u8>(slave_request) << 4) | head);
if (lba_mode == LBAMode::FortyEightBit) {
m_io_group.io_base().offset(ATA_REG_SECCOUNT1).out<u8>(0);
m_io_group.io_base().offset(ATA_REG_LBA3).out<u8>((lba & 0xFF000000) >> 24);
m_io_group.io_base().offset(ATA_REG_LBA4).out<u8>((lba & 0xFF00000000ull) >> 32);
m_io_group.io_base().offset(ATA_REG_LBA5).out<u8>((lba & 0xFF0000000000ull) >> 40);
}
m_io_group.io_base().offset(ATA_REG_SECCOUNT0).out<u8>(block_count);
m_io_group.io_base().offset(ATA_REG_LBA0).out<u8>((lba & 0x000000FF) >> 0);
m_io_group.io_base().offset(ATA_REG_LBA1).out<u8>((lba & 0x0000FF00) >> 8);
m_io_group.io_base().offset(ATA_REG_LBA2).out<u8>((lba & 0x00FF0000) >> 16);
for (;;) {
auto status = m_io_group.control_base().in<u8>();
if (!(status & ATA_SR_BSY) && (status & ATA_SR_DRDY))
break;
}
if (lba_mode != LBAMode::FortyEightBit) {
if (use_dma)
m_io_group.io_base().offset(ATA_REG_COMMAND).out<u8>(direction == Direction::Read ? ATA_CMD_READ_DMA : ATA_CMD_WRITE_DMA);
else
m_io_group.io_base().offset(ATA_REG_COMMAND).out<u8>(direction == Direction::Read ? ATA_CMD_READ_PIO : ATA_CMD_WRITE_PIO);
} else {
if (use_dma)
m_io_group.io_base().offset(ATA_REG_COMMAND).out<u8>(direction == Direction::Read ? ATA_CMD_READ_DMA_EXT : ATA_CMD_WRITE_DMA_EXT);
else
m_io_group.io_base().offset(ATA_REG_COMMAND).out<u8>(direction == Direction::Read ? ATA_CMD_READ_PIO_EXT : ATA_CMD_WRITE_PIO_EXT);
}
enable_irq();
}
void IDEChannel::ata_read_sectors_with_dma(bool slave_request, u16 capabilities)
{
auto& request = *m_current_request;
auto lba = request.block_index();
dbgln_if(PATA_DEBUG, "IDEChannel::ata_read_sectors_with_dma ({} x {})", lba, request.block_count());
prdt().offset = m_dma_buffer_page->paddr();
prdt().size = 512 * request.block_count();
VERIFY(prdt().size <= PAGE_SIZE);
// Stop bus master
m_io_group.bus_master_base().out<u8>(0);
// Write the PRDT location
m_io_group.bus_master_base().offset(4).out(m_prdt_page->paddr().get());
// Turn on "Interrupt" and "Error" flag. The error flag should be cleared by hardware.
m_io_group.bus_master_base().offset(2).out<u8>(m_io_group.bus_master_base().offset(2).in<u8>() | 0x6);
// Set transfer direction
m_io_group.bus_master_base().out<u8>(0x8);
ata_access(Direction::Read, slave_request, lba, request.block_count(), capabilities, true);
// Start bus master
m_io_group.bus_master_base().out<u8>(0x9);
}
bool IDEChannel::ata_do_read_sector()
{
dbgln_if(PATA_DEBUG, "IDEChannel::ata_do_read_sector");
auto& request = *m_current_request;
auto out_buffer = request.buffer().offset(m_current_request_block_index * 512);
ssize_t nwritten = request.write_to_buffer_buffered<512>(out_buffer, 512, [&](u8* buffer, size_t buffer_bytes) {
for (size_t i = 0; i < buffer_bytes; i += sizeof(u16))
*(u16*)&buffer[i] = IO::in16(m_io_group.io_base().offset(ATA_REG_DATA).get());
return (ssize_t)buffer_bytes;
});
if (nwritten < 0) {
// TODO: Do we need to abort the PATA read if this wasn't the last block?
complete_current_request(AsyncDeviceRequest::MemoryFault);
return false;
}
return true;
}
// FIXME: This doesn't quite work and locks up reading LBA 3.
void IDEChannel::ata_read_sectors(bool slave_request, u16 capabilities)
{
auto& request = *m_current_request;
VERIFY(request.block_count() <= 256);
dbgln_if(PATA_DEBUG, "IDEChannel::ata_read_sectors");
auto lba = request.block_index();
dbgln_if(PATA_DEBUG, "IDEChannel: Reading {} sector(s) @ LBA {}", request.block_count(), lba);
ata_access(Direction::Read, slave_request, lba, request.block_count(), capabilities, false);
}
void IDEChannel::ata_write_sectors_with_dma(bool slave_request, u16 capabilities)
{
auto& request = *m_current_request;
auto lba = request.block_index();
dbgln_if(PATA_DEBUG, "IDEChannel::ata_write_sectors_with_dma ({} x {})", lba, request.block_count());
prdt().offset = m_dma_buffer_page->paddr();
prdt().size = 512 * request.block_count();
if (!request.read_from_buffer(request.buffer(), m_dma_buffer_page->paddr().offset(0xc0000000).as_ptr(), 512 * request.block_count())) {
complete_current_request(AsyncDeviceRequest::MemoryFault);
return;
}
VERIFY(prdt().size <= PAGE_SIZE);
// Stop bus master
m_io_group.bus_master_base().out<u8>(0);
// Write the PRDT location
m_io_group.bus_master_base().offset(4).out<u32>(m_prdt_page->paddr().get());
// Turn on "Interrupt" and "Error" flag. The error flag should be cleared by hardware.
m_io_group.bus_master_base().offset(2).out<u8>(m_io_group.bus_master_base().offset(2).in<u8>() | 0x6);
ata_access(Direction::Write, slave_request, lba, request.block_count(), capabilities, true);
// Start bus master
m_io_group.bus_master_base().out<u8>(0x1);
}
void IDEChannel::ata_do_write_sector()
{
auto& request = *m_current_request;
io_delay();
while ((m_io_group.control_base().in<u8>() & ATA_SR_BSY) || !(m_io_group.control_base().in<u8>() & ATA_SR_DRQ))
;
u8 status = m_io_group.control_base().in<u8>();
VERIFY(status & ATA_SR_DRQ);
auto in_buffer = request.buffer().offset(m_current_request_block_index * 512);
dbgln_if(PATA_DEBUG, "IDEChannel: Writing 512 bytes (part {}) (status={:#02x})...", m_current_request_block_index, status);
ssize_t nread = request.read_from_buffer_buffered<512>(in_buffer, 512, [&](const u8* buffer, size_t buffer_bytes) {
for (size_t i = 0; i < buffer_bytes; i += sizeof(u16))
IO::out16(m_io_group.io_base().offset(ATA_REG_DATA).get(), *(const u16*)&buffer[i]);
return (ssize_t)buffer_bytes;
});
if (nread < 0)
complete_current_request(AsyncDeviceRequest::MemoryFault);
}
// FIXME: I'm assuming this doesn't work based on the fact PIO read doesn't work.
void IDEChannel::ata_write_sectors(bool slave_request, u16 capabilities)
{
auto& request = *m_current_request;
VERIFY(request.block_count() <= 256);
auto start_sector = request.block_index();
u32 count = request.block_count();
dbgln_if(PATA_DEBUG, "IDEChannel: Writing {} sector(s) @ LBA {}", count, start_sector);
ata_access(Direction::Write, slave_request, start_sector, request.block_count(), capabilities, false);
ata_do_write_sector();
}
}