RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-24 22:27:42 +00:00
Code Issues Activity Actions

Kernel: Lock target process when generating core dump

Browse source 
Dumping core can happen at the end of a profiling run, and in that case
we have to protect the target process and take the lock while iterating
over its region map.

Fixes #4509.
This commit is contained in:
Andreas Kling 2020-12-27 00:51:48 +01:00
parent 4395e1b240
commit 87492e723b
2 changed files with 17 additions and 17 deletions
Download patch file Download diff file Expand all files Collapse all files

28
Kernel/CoreDump.cpp
View file

@ -40,23 +40,23 @@
namespace Kernel {
OwnPtr<CoreDump> CoreDump::create(Process& process, const String& output_path)
OwnPtr<CoreDump> CoreDump::create(NonnullRefPtr<Process> process, const String& output_path)
{
if (!process.is_dumpable()) {
dbgln("Refusing to generate CoreDump for non-dumpable process {}", process.pid().value());
if (!process->is_dumpable()) {
dbgln("Refusing to generate CoreDump for non-dumpable process {}", process->pid().value());
return nullptr;
}
auto fd = create_target_file(process, output_path);
if (!fd)
return nullptr;
return adopt_own(*new CoreDump(process, fd.release_nonnull()));
return adopt_own(*new CoreDump(move(process), fd.release_nonnull()));
}
CoreDump::CoreDump(Process& process, NonnullRefPtr<FileDescription>&& fd)
: m_process(process)
CoreDump::CoreDump(NonnullRefPtr<Process> process, NonnullRefPtr<FileDescription>&& fd)
: m_process(move(process))
, m_fd(move(fd))
, m_num_program_headers(process.m_regions.size() + 1) // +1 for NOTE segment
, m_num_program_headers(m_process->m_regions.size() + 1) // +1 for NOTE segment
{
}
@ -130,7 +130,7 @@ KResult CoreDump::write_elf_header()
KResult CoreDump::write_program_headers(size_t notes_size)
{
size_t offset = sizeof(Elf32_Ehdr) + m_num_program_headers * sizeof(Elf32_Phdr);
for (auto& region : m_process.m_regions) {
for (auto& region : m_process->m_regions) {
Elf32_Phdr phdr {};
phdr.p_type = PT_LOAD;
@ -171,16 +171,15 @@ KResult CoreDump::write_program_headers(size_t notes_size)
KResult CoreDump::write_regions()
{
for (auto& region : m_process.m_regions) {
for (auto& region : m_process->m_regions) {
if (region.is_kernel())
continue;
region.set_readable(true);
region.remap();
auto& vmobj = region.vmobject();
for (size_t i = 0; i < region.page_count(); i++) {
PhysicalPage* page = vmobj.physical_pages()[region.first_page_index() + i];
auto* page = region.physical_page(i);
uint8_t zero_buffer[PAGE_SIZE] = {};
Optional<UserOrKernelBuffer> src_buffer;
@ -213,7 +212,7 @@ ByteBuffer CoreDump::create_notes_threads_data() const
{
ByteBuffer threads_data;
m_process.for_each_thread([&](Thread& thread) {
m_process->for_each_thread([&](Thread& thread) {
ByteBuffer entry_buff;
ELF::Core::ThreadInfo info {};
@ -233,13 +232,13 @@ ByteBuffer CoreDump::create_notes_threads_data() const
ByteBuffer CoreDump::create_notes_regions_data() const
{
ByteBuffer regions_data;
for (size_t region_index = 0; region_index < m_process.m_regions.size(); ++region_index) {
for (size_t region_index = 0; region_index < m_process->m_regions.size(); ++region_index) {
ByteBuffer memory_region_info_buffer;
ELF::Core::MemoryRegionInfo info {};
info.header.type = ELF::Core::NotesEntryHeader::Type::MemoryRegionInfo;
auto& region = m_process.m_regions[region_index];
auto& region = m_process->m_regions[region_index];
info.region_start = reinterpret_cast<uint32_t>(region.vaddr().as_ptr());
info.region_end = reinterpret_cast<uint32_t>(region.vaddr().as_ptr() + region.size());
info.program_header_index = region_index;
@ -272,6 +271,7 @@ ByteBuffer CoreDump::create_notes_segment_data() const
KResult CoreDump::write()
{
ScopedSpinLock lock(m_process->get_lock());
ProcessPagingScope scope(m_process);
ByteBuffer notes_segment = create_notes_segment_data();