beenull/ladybird
1
0
Fork 0
mirror of https://github.com/LadybirdBrowser/ladybird.git synced 2024-09-30 08:41:15 +00:00
Code Issues Actions 5 Packages Projects Releases Wiki Activity
ladybird/Kernel/Syscalls/sigaction.cpp

311 lines
12 KiB
C++
Raw Normal View History

Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
/*
* Copyright (c) 2018-2020, Andreas Kling <kling@serenityos.org>
Kernel+LibC: Implement sigaltstack() This is required for compiling wine for serenity
2021-10-28 20:33:41 +00:00
* Copyright (c) 2021, Idan Horowitz <idan.horowitz@serenityos.org>
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
*
Everything: Move to SPDX license identifiers in all files. SPDX License Identifiers are a more compact / standardized way of representing file license information. See: https://spdx.dev/resources/use/#identifiers This was done with the `ambr` search and replace tool. ambr --no-parent-ignore --key-from-file --rep-from-file key.txt rep.txt *
2021-04-22 08:24:48 +00:00
* SPDX-License-Identifier: BSD-2-Clause
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
*/
Kernel: Split SmapDisabler so header is platform independent A new header file has been created in the Arch/ folder while the implementation has been moved into a CPP living in the X86 folder.
2021-10-14 21:53:48 +00:00
#include <Kernel/Arch/SmapDisabler.h>
Kernel: Pull apart CPU.h This does not add any functional changes
2021-06-21 15:34:09 +00:00
#include <Kernel/Arch/x86/InterruptDisabler.h>
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
#include <Kernel/Process.h>
namespace Kernel {
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
ErrorOr<FlatPtr> Process::sys$sigprocmask(int how, Userspace<const sigset_t*> set, Userspace<sigset_t*> old_set)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
{
Kernel: Annotate all syscalls with VERIFY_PROCESS_BIG_LOCK_ACQUIRED Before we start disabling acquisition of the big process lock for specific syscalls, make sure to document and assert that all the lock is held during all syscalls.
2021-07-18 18:20:12 +00:00
VERIFY_PROCESS_BIG_LOCK_ACQUIRED(this)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
REQUIRE_PROMISE(sigaction);
auto current_thread = Thread::current();
Kernel: Keep signal state in sync In c3d231616c1d20309b2b568f383fbcb736887dad we added the atomic variable m_have_any_unmasked_pending_signals tracking the state of pending signals. Add helper functions that automatically update this variable as needed.
2020-09-09 02:37:15 +00:00
u32 previous_signal_mask;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
if (set) {
sigset_t set_value;
Kernel: Make copy_{from,to}_user() return KResult and use TRY() This makes EFAULT propagation flow much more naturally. :^)
2021-09-05 15:38:37 +00:00
TRY(copy_from_user(&set_value, set));
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
switch (how) {
case SIG_BLOCK:
Kernel: Keep signal state in sync In c3d231616c1d20309b2b568f383fbcb736887dad we added the atomic variable m_have_any_unmasked_pending_signals tracking the state of pending signals. Add helper functions that automatically update this variable as needed.
2020-09-09 02:37:15 +00:00
previous_signal_mask = current_thread->signal_mask_block(set_value, true);
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
break;
case SIG_UNBLOCK:
Kernel: Keep signal state in sync In c3d231616c1d20309b2b568f383fbcb736887dad we added the atomic variable m_have_any_unmasked_pending_signals tracking the state of pending signals. Add helper functions that automatically update this variable as needed.
2020-09-09 02:37:15 +00:00
previous_signal_mask = current_thread->signal_mask_block(set_value, false);
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
break;
case SIG_SETMASK:
Kernel: Keep signal state in sync In c3d231616c1d20309b2b568f383fbcb736887dad we added the atomic variable m_have_any_unmasked_pending_signals tracking the state of pending signals. Add helper functions that automatically update this variable as needed.
2020-09-09 02:37:15 +00:00
previous_signal_mask = current_thread->update_signal_mask(set_value);
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
break;
default:
Kernel: Make all syscall functions return KResultOr<T> This makes it a lot easier to return errors since we no longer have to worry about negating EFOO errors and can just return them flat.
2021-03-01 12:49:16 +00:00
return EINVAL;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
}
Kernel: Keep signal state in sync In c3d231616c1d20309b2b568f383fbcb736887dad we added the atomic variable m_have_any_unmasked_pending_signals tracking the state of pending signals. Add helper functions that automatically update this variable as needed.
2020-09-09 02:37:15 +00:00
} else {
previous_signal_mask = current_thread->signal_mask();
}
Kernel: Make copy_{from,to}_user() return KResult and use TRY() This makes EFAULT propagation flow much more naturally. :^)
2021-09-05 15:38:37 +00:00
if (old_set) {
TRY(copy_to_user(old_set, &previous_signal_mask));
}
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
return 0;
}
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
ErrorOr<FlatPtr> Process::sys$sigpending(Userspace<sigset_t*> set)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
{
Kernel: Annotate all syscalls with VERIFY_PROCESS_BIG_LOCK_ACQUIRED Before we start disabling acquisition of the big process lock for specific syscalls, make sure to document and assert that all the lock is held during all syscalls.
2021-07-18 18:20:12 +00:00
VERIFY_PROCESS_BIG_LOCK_ACQUIRED(this)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
REQUIRE_PROMISE(stdio);
Kernel: Keep signal state in sync In c3d231616c1d20309b2b568f383fbcb736887dad we added the atomic variable m_have_any_unmasked_pending_signals tracking the state of pending signals. Add helper functions that automatically update this variable as needed.
2020-09-09 02:37:15 +00:00
auto pending_signals = Thread::current()->pending_signals();
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
TRY(copy_to_user(set, &pending_signals));
return 0;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
}
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
ErrorOr<FlatPtr> Process::sys$sigaction(int signum, Userspace<const sigaction*> user_act, Userspace<sigaction*> user_old_act)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
{
Kernel: Annotate all syscalls with VERIFY_PROCESS_BIG_LOCK_ACQUIRED Before we start disabling acquisition of the big process lock for specific syscalls, make sure to document and assert that all the lock is held during all syscalls.
2021-07-18 18:20:12 +00:00
VERIFY_PROCESS_BIG_LOCK_ACQUIRED(this)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
REQUIRE_PROMISE(sigaction);
if (signum < 1 || signum >= 32 || signum == SIGKILL || signum == SIGSTOP)
Kernel: Make all syscall functions return KResultOr<T> This makes it a lot easier to return errors since we no longer have to worry about negating EFOO errors and can just return them flat.
2021-03-01 12:49:16 +00:00
return EINVAL;
Kernel: Use Userspace<T> in sys$sigaction() fuzz-syscalls found a bunch of unaligned accesses into struct sigaction via this syscall. This patch fixes that issue by porting the syscall to Userspace<T> which we should have done anyway. :^) Fixes #5500.
2021-03-01 13:05:42 +00:00
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
InterruptDisabler disabler; // FIXME: This should use a narrower lock. Maybe a way to ignore signals temporarily?
auto& action = Thread::current()->m_signal_action_data[signum];
Kernel: Use Userspace<T> in sys$sigaction() fuzz-syscalls found a bunch of unaligned accesses into struct sigaction via this syscall. This patch fixes that issue by porting the syscall to Userspace<T> which we should have done anyway. :^) Fixes #5500.
2021-03-01 13:05:42 +00:00
if (user_old_act) {
sigaction old_act {};
old_act.sa_flags = action.flags;
old_act.sa_sigaction = reinterpret_cast<decltype(old_act.sa_sigaction)>(action.handler_or_sigaction.as_ptr());
Kernel: Make copy_{from,to}_user() return KResult and use TRY() This makes EFAULT propagation flow much more naturally. :^)
2021-09-05 15:38:37 +00:00
TRY(copy_to_user(user_old_act, &old_act));
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
}
Kernel: Support null act argument for sigaction syscall Userspace can provide a null argument for the `act` argument to the `sigaction` syscall to not set any new behavior. This is described here: https://pubs.opengroup.org/onlinepubs/007904875/functions/sigaction.html Without this fix, the `copy_from_user(...)` invocation on `user_act` fails and makes the syscall return early.
2021-04-24 00:17:22 +00:00
if (user_act) {
sigaction act {};
Kernel: Make copy_{from,to}_user() return KResult and use TRY() This makes EFAULT propagation flow much more naturally. :^)
2021-09-05 15:38:37 +00:00
TRY(copy_from_user(&act, user_act));
Kernel: Support null act argument for sigaction syscall Userspace can provide a null argument for the `act` argument to the `sigaction` syscall to not set any new behavior. This is described here: https://pubs.opengroup.org/onlinepubs/007904875/functions/sigaction.html Without this fix, the `copy_from_user(...)` invocation on `user_act` fails and makes the syscall return early.
2021-04-24 00:17:22 +00:00
action.flags = act.sa_flags;
action.handler_or_sigaction = VirtualAddress { reinterpret_cast<void*>(act.sa_sigaction) };
}
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
return 0;
}
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
ErrorOr<FlatPtr> Process::sys$sigreturn([[maybe_unused]] RegisterState& registers)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
{
Kernel: Annotate all syscalls with VERIFY_PROCESS_BIG_LOCK_ACQUIRED Before we start disabling acquisition of the big process lock for specific syscalls, make sure to document and assert that all the lock is held during all syscalls.
2021-07-18 18:20:12 +00:00
VERIFY_PROCESS_BIG_LOCK_ACQUIRED(this)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
REQUIRE_PROMISE(stdio);
SmapDisabler disabler;
Kernel: Move sys$sigaction() implementation inside ARCH(i386)
2021-02-25 10:33:06 +00:00
#if ARCH(I386)
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
// Here, we restore the state pushed by dispatch signal and asm_signal_trampoline.
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
u32* stack_ptr = (u32*)registers.userspace_esp;
u32 smuggled_eax = *stack_ptr;
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
// pop the stored eax, ebp, return address, handler and signal code
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
stack_ptr += 5;
Thread::current()->m_signal_mask = *stack_ptr;
stack_ptr++;
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
// pop edi, esi, ebp, esp, ebx, edx, ecx and eax
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
memcpy(&registers.edi, stack_ptr, 8 * sizeof(FlatPtr));
stack_ptr += 8;
registers.eip = *stack_ptr;
stack_ptr++;
Kernel: Don't allow modifying IOPL via sys$ptrace() or sys$sigreturn() It was possible to overwrite the entire EFLAGS register since we didn't do any masking in the ptrace and sigreturn syscalls. This made it trivial to gain IO privileges by raising IOPL to 3 and then you could talk to hardware to do all kinds of nasty things. Thanks to @allesctf for finding these issues! :^) Their exploit/write-up: https://github.com/allesctf/writeups/blob/master/2020/hxpctf/wisdom2/writeup.md
2020-12-22 17:23:34 +00:00
registers.eflags = (registers.eflags & ~safe_eflags_mask) | (*stack_ptr & safe_eflags_mask);
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
stack_ptr++;
registers.userspace_esp = registers.esp;
return smuggled_eax;
Kernel: Move sys$sigaction() implementation inside ARCH(i386)
2021-02-25 10:33:06 +00:00
#else
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
// Here, we restore the state pushed by dispatch signal and asm_signal_trampoline.
Kernel: Implement signal handling for x86_64
2021-06-29 08:31:25 +00:00
FlatPtr* stack_ptr = (FlatPtr*)registers.userspace_rsp;
FlatPtr smuggled_rax = *stack_ptr;
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
// pop the stored rax, rbp, return address, handler and signal code
Kernel: Implement signal handling for x86_64
2021-06-29 08:31:25 +00:00
stack_ptr += 5;
Thread::current()->m_signal_mask = *stack_ptr;
stack_ptr++;
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
// pop rdi, rsi, rbp, rsp, rbx, rdx, rcx, rax, r8, r9, r10, r11, r12, r13, r14 and r15
Kernel: Implement signal handling for x86_64
2021-06-29 08:31:25 +00:00
memcpy(&registers.rdi, stack_ptr, 16 * sizeof(FlatPtr));
stack_ptr += 16;
registers.rip = *stack_ptr;
stack_ptr++;
registers.rflags = (registers.rflags & ~safe_eflags_mask) | (*stack_ptr & safe_eflags_mask);
stack_ptr++;
registers.userspace_rsp = registers.rsp;
return smuggled_rax;
Kernel: Move sys$sigaction() implementation inside ARCH(i386)
2021-02-25 10:33:06 +00:00
#endif
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
}
Kernel+LibC: Implement sigaltstack() This is required for compiling wine for serenity
2021-10-28 20:33:41 +00:00
ErrorOr<void> Process::remap_range_as_stack(FlatPtr address, size_t size)
{
// FIXME: This duplicates a lot of logic from sys$mprotect, this should be abstracted out somehow
auto range_to_remap = TRY(Memory::expand_range_to_page_boundaries(address, size));
if (!range_to_remap.size())
return EINVAL;
if (!is_user_range(range_to_remap))
return EFAULT;
if (auto* whole_region = address_space().find_region_from_range(range_to_remap)) {
if (!whole_region->is_mmap())
return EPERM;
if (!whole_region->vmobject().is_anonymous() || whole_region->is_shared())
return EINVAL;
whole_region->unsafe_clear_access();
whole_region->set_readable(true);
whole_region->set_writable(true);
whole_region->set_stack(true);
whole_region->set_syscall_region(false);
whole_region->clear_to_zero();
whole_region->remap();
return {};
}
if (auto* old_region = address_space().find_region_containing(range_to_remap)) {
if (!old_region->is_mmap())
return EPERM;
if (!old_region->vmobject().is_anonymous() || old_region->is_shared())
return EINVAL;
// Remove the old region from our regions tree, since were going to add another region
// with the exact same start address, but do not deallocate it yet
auto region = address_space().take_region(*old_region);
// Unmap the old region here, specifying that we *don't* want the VM deallocated.
region->unmap(Memory::Region::ShouldDeallocateVirtualRange::No);
// This vector is the region(s) adjacent to our range.
// We need to allocate a new region for the range we wanted to change permission bits on.
auto adjacent_regions = TRY(address_space().try_split_region_around_range(*region, range_to_remap));
size_t new_range_offset_in_vmobject = region->offset_in_vmobject() + (range_to_remap.base().get() - region->range().base().get());
auto new_region = TRY(address_space().try_allocate_split_region(*region, range_to_remap, new_range_offset_in_vmobject));
new_region->unsafe_clear_access();
new_region->set_readable(true);
new_region->set_writable(true);
new_region->set_stack(true);
new_region->set_syscall_region(false);
new_region->clear_to_zero();
// Map the new regions using our page directory (they were just allocated and don't have one).
for (auto* adjacent_region : adjacent_regions) {
TRY(adjacent_region->map(address_space().page_directory()));
}
TRY(new_region->map(address_space().page_directory()));
return {};
}
if (const auto& regions = address_space().find_regions_intersecting(range_to_remap); regions.size()) {
size_t full_size_found = 0;
// Check that all intersecting regions are compatible.
for (const auto* region : regions) {
if (!region->is_mmap())
return EPERM;
if (!region->vmobject().is_anonymous() || region->is_shared())
return EINVAL;
full_size_found += region->range().intersect(range_to_remap).size();
}
if (full_size_found != range_to_remap.size())
return ENOMEM;
// Finally, iterate over each region, either updating its access flags if the range covers it wholly,
// or carving out a new subregion with the appropriate access flags set.
for (auto* old_region : regions) {
const auto intersection_to_remap = range_to_remap.intersect(old_region->range());
// If the region is completely covered by range, simply update the access flags
if (intersection_to_remap == old_region->range()) {
old_region->unsafe_clear_access();
old_region->set_readable(true);
old_region->set_writable(true);
old_region->set_stack(true);
old_region->set_syscall_region(false);
old_region->clear_to_zero();
old_region->remap();
continue;
}
// Remove the old region from our regions tree, since were going to add another region
// with the exact same start address, but dont deallocate it yet
auto region = address_space().take_region(*old_region);
// Unmap the old region here, specifying that we *don't* want the VM deallocated.
region->unmap(Memory::Region::ShouldDeallocateVirtualRange::No);
// This vector is the region(s) adjacent to our range.
// We need to allocate a new region for the range we wanted to change permission bits on.
auto adjacent_regions = TRY(address_space().try_split_region_around_range(*old_region, intersection_to_remap));
// Since the range is not contained in a single region, it can only partially cover its starting and ending region,
// therefore carving out a chunk from the region will always produce a single extra region, and not two.
VERIFY(adjacent_regions.size() == 1);
size_t new_range_offset_in_vmobject = old_region->offset_in_vmobject() + (intersection_to_remap.base().get() - old_region->range().base().get());
auto* new_region = TRY(address_space().try_allocate_split_region(*region, intersection_to_remap, new_range_offset_in_vmobject));
new_region->unsafe_clear_access();
new_region->set_readable(true);
new_region->set_writable(true);
new_region->set_stack(true);
new_region->set_syscall_region(false);
new_region->clear_to_zero();
// Map the new region using our page directory (they were just allocated and don't have one) if any.
if (adjacent_regions.size())
TRY(adjacent_regions[0]->map(address_space().page_directory()));
TRY(new_region->map(address_space().page_directory()));
}
return {};
}
return EINVAL;
}
ErrorOr<FlatPtr> Process::sys$sigaltstack(Userspace<const stack_t*> ss, Userspace<stack_t*> old_ss)
{
VERIFY_PROCESS_BIG_LOCK_ACQUIRED(this)
REQUIRE_PROMISE(sigaction);
if (old_ss) {
stack_t old_ss_value;
old_ss_value.ss_sp = (void*)Thread::current()->m_alternative_signal_stack;
old_ss_value.ss_size = Thread::current()->m_alternative_signal_stack_size;
old_ss_value.ss_flags = 0;
if (!Thread::current()->has_alternative_signal_stack())
old_ss_value.ss_flags = SS_DISABLE;
else if (Thread::current()->is_in_alternative_signal_stack())
old_ss_value.ss_flags = SS_ONSTACK;
TRY(copy_to_user(old_ss, &old_ss_value));
}
if (ss) {
stack_t ss_value;
TRY(copy_from_user(&ss_value, ss));
if (Thread::current()->is_in_alternative_signal_stack())
return EPERM;
if (ss_value.ss_flags == SS_DISABLE) {
Thread::current()->m_alternative_signal_stack_size = 0;
Thread::current()->m_alternative_signal_stack = 0;
} else if (ss_value.ss_flags == 0) {
if (ss_value.ss_size <= MINSIGSTKSZ)
return ENOMEM;
if (Checked<FlatPtr>::addition_would_overflow((FlatPtr)ss_value.ss_sp, ss_value.ss_size))
return ENOMEM;
// In order to preserve compatibility with our MAP_STACK, W^X and syscall region
// protections, sigaltstack ranges are carved out of their regions, zeroed, and
// turned into read/writable MAP_STACK-enabled regions.
// This is inspired by OpenBSD's solution: https://man.openbsd.org/sigaltstack.2
TRY(remap_range_as_stack((FlatPtr)ss_value.ss_sp, ss_value.ss_size));
Thread::current()->m_alternative_signal_stack = (FlatPtr)ss_value.ss_sp;
Thread::current()->m_alternative_signal_stack_size = ss_value.ss_size;
} else {
return EINVAL;
}
}
return 0;
}
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
}
Reference in a new issue Copy permalink