beenull/ladybird
1
0
Fork 0
mirror of https://github.com/LadybirdBrowser/ladybird.git synced 2024-11-27 01:50:24 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 7
ladybird/Kernel/Syscalls/socket.cpp

448 lines
17 KiB
C++
Raw Normal View History

Kernel: Add missing license headers to some syscall files
2020-07-30 23:01:41 +00:00
/*
* Copyright (c) 2018-2020, Andreas Kling <kling@serenityos.org>
*
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: Add missing license headers to some syscall files
2020-07-30 23:01:41 +00:00
*/
Kernel: Support sending filedescriptors with sendmsg(2) and SCM_RIGHTS This is necessary to support the wayland protocol. I also moved the CMSG_* macros to the kernel API since they are used in both kernel and userspace. this does not break ntpquery/SCM_TIMESTAMP.
2023-02-13 23:01:13 +00:00
#include <AK/ByteBuffer.h>
Kernel: Rename FileDescription => OpenFileDescription Dr. POSIX really calls these "open file description", not just "file description", so let's call them exactly that. :^)
2021-09-07 11:39:11 +00:00
#include <Kernel/FileSystem/OpenFileDescription.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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
#include <Kernel/Net/LocalSocket.h>
Kernel: Move all tasks-related code to the Tasks subdirectory
2023-02-24 17:45:37 +00:00
#include <Kernel/Tasks/Process.h>
Kernel+LibC+UserspaceEmulator: Mostly add recvmsg(), sendmsg() The implementation only supports a single iovec for now. Some might say having more than one iovec is the main point of recvmsg() and sendmsg(), but I'm interested in the control message bits.
2020-09-16 15:45:00 +00:00
#include <Kernel/UnixTypes.h>
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
namespace Kernel {
#define REQUIRE_PROMISE_FOR_SOCKET_DOMAIN(domain) \
do { \
if (domain == AF_INET) \
Kernel: Handle promise violations in the syscall handler Previously we would crash the process immediately when a promise violation was found during a syscall. This is error prone, as we don't unwind the stack. This means that in certain cases we can leak resources, like an OwnPtr / RefPtr tracked on the stack. Or even leak a lock acquired in a ScopeLockLocker. To remedy this situation we move the promise violation handling to the syscall handler, right before we return to user space. This allows the code to follow the normal unwind path, and grantees there is no longer any cleanup that needs to occur. The Process::require_promise() and Process::require_no_promises() functions were modified to return ErrorOr<void> so we enforce that the errors are always propagated by the caller.
2021-12-29 09:11:45 +00:00
TRY(require_promise(Pledge::inet)); \
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
else if (domain == AF_LOCAL) \
Kernel: Handle promise violations in the syscall handler Previously we would crash the process immediately when a promise violation was found during a syscall. This is error prone, as we don't unwind the stack. This means that in certain cases we can leak resources, like an OwnPtr / RefPtr tracked on the stack. Or even leak a lock acquired in a ScopeLockLocker. To remedy this situation we move the promise violation handling to the syscall handler, right before we return to user space. This allows the code to follow the normal unwind path, and grantees there is no longer any cleanup that needs to occur. The Process::require_promise() and Process::require_no_promises() functions were modified to return ErrorOr<void> so we enforce that the errors are always propagated by the caller.
2021-12-29 09:11:45 +00:00
TRY(require_promise(Pledge::unix)); \
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
} while (0)
Kernel: Use non-locking {Nonnull,}RefPtr for OpenFileDescription This patch switches away from {Nonnull,}LockRefPtr to the non-locking smart pointers throughout the kernel. I've looked at the handful of places where these were being persisted and I don't see any race situations. Note that the process file descriptor table (Process::m_fds) was already guarded via MutexProtected.
2023-03-06 18:29:25 +00:00
static void setup_socket_fd(Process::OpenFileDescriptions& fds, int fd, NonnullRefPtr<OpenFileDescription> description, int type)
Kernel+LibC: Implement the socketpair() syscall
2021-04-28 10:39:12 +00:00
{
description->set_readable(true);
description->set_writable(true);
unsigned flags = 0;
if (type & SOCK_CLOEXEC)
flags |= FD_CLOEXEC;
if (type & SOCK_NONBLOCK)
description->set_blocking(false);
Kernel: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
fds[fd].set(*description, flags);
Kernel+LibC: Implement the socketpair() syscall
2021-04-28 10:39:12 +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$socket(int domain, int type, int protocol)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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: Remove big lock from sys$socket With the implementation of the credentials object the socket syscall no longer needs the big lock.
2022-08-21 14:59:31 +00:00
VERIFY_NO_PROCESS_BIG_LOCK(this);
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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_FOR_SOCKET_DOMAIN(domain);
Kernel: Use Process::credentials() and remove user ID/group ID helpers Move away from using the group ID/user ID helpers in the process to allow for us to take advantage of the immutable credentials instead.
2022-08-20 22:21:01 +00:00
auto credentials = this->credentials();
if ((type & SOCK_TYPE_MASK) == SOCK_RAW && !credentials->is_superuser())
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 EACCES;
Kernel: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
Kernel: Switch process file descriptor table from spinlock to mutex There's no reason for this to use a spinlock. Instead, let's allow threads to block if someone else is using the descriptor table.
2022-01-29 00:29:07 +00:00
return m_fds.with_exclusive([&](auto& fds) -> ErrorOr<FlatPtr> {
Kernel: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
auto fd_allocation = TRY(fds.allocate());
auto socket = TRY(Socket::create(domain, type, protocol));
auto description = TRY(OpenFileDescription::try_create(socket));
setup_socket_fd(fds, fd_allocation.fd, move(description), type);
return fd_allocation.fd;
});
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
}
Everywhere: Run clang-format
2022-04-01 17:58:27 +00:00
ErrorOr<FlatPtr> Process::sys$bind(int sockfd, Userspace<sockaddr const*> address, socklen_t address_length)
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
{
Kernel: Require semicolon after VERIFY_{NO_,}PROCESS_BIG_LOCK_ACQUIRED This matches out general macro use, and specifically other verification macros like VERIFY(), VERIFY_NOT_REACHED(), VERIFY_INTERRUPTS_ENABLED(), and VERIFY_INTERRUPTS_DISABLED().
2022-08-17 20:03:04 +00:00
VERIFY_NO_PROCESS_BIG_LOCK(this);
Kernel: Don't hog file descriptor table lock in sys$bind() We don't need to hold the lock across the entire syscall. Once we've fetched the open file description we're interested in, we can let go.
2022-04-03 20:20:34 +00:00
auto description = TRY(open_file_description(sockfd));
if (!description->is_socket())
return ENOTSOCK;
auto& socket = *description->socket();
REQUIRE_PROMISE_FOR_SOCKET_DOMAIN(socket.domain());
Kernel: Make Socket::bind() take credentials as input
2022-08-21 14:33:09 +00:00
TRY(socket.bind(credentials(), address, address_length));
Kernel: Don't hog file descriptor table lock in sys$bind() We don't need to hold the lock across the entire syscall. Once we've fetched the open file description we're interested in, we can let go.
2022-04-03 20:20:34 +00:00
return 0;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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$listen(int sockfd, int backlog)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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: Require semicolon after VERIFY_{NO_,}PROCESS_BIG_LOCK_ACQUIRED This matches out general macro use, and specifically other verification macros like VERIFY(), VERIFY_NOT_REACHED(), VERIFY_INTERRUPTS_ENABLED(), and VERIFY_INTERRUPTS_DISABLED().
2022-08-17 20:03:04 +00:00
VERIFY_NO_PROCESS_BIG_LOCK(this);
Kernel: Treat a backlog argument of 0 to listen() as if it was 1 As per POSIX, the behavior of listen() with a backlog value of 0 is implementation defined: "A backlog argument of 0 may allow the socket to accept connections, in which case the length of the listen queue may be set to an implementation-defined minimum value." Since creating a socket that can't accept any connections seems relatively useless, and as other platforms (Linux, FreeBSD, etc) chose to support accepting connections with this backlog value, support it as well by normalizing it to 1.
2023-11-25 14:36:08 +00:00
// As per POSIX, the behavior of listen() with a backlog value of 0 is implementation defined:
// "A backlog argument of 0 may allow the socket to accept connections, in which case the length of the listen queue may be set to an implementation-defined minimum value."
// Since creating a socket that can't accept any connections seems relatively useless, and as other platforms (Linux, FreeBSD, etc) chose to support accepting connections
// with this backlog value, support it as well by normalizing it to 1.
// Also, as per POSIX, the behaviour of a negative backlog value is equivalent to a backlog value of 0:
// "If listen() is called with a backlog argument value that is less than 0, the function behaves as if it had been called with a backlog argument value of 0."
if (backlog <= 0)
backlog = 1;
Kernel: Don't hog file descriptor table lock in sys$listen() We don't need to hold the lock across the entire syscall. Once we've fetched the open file description we're interested in, we can let go.
2022-04-03 20:18:57 +00:00
auto description = TRY(open_file_description(sockfd));
if (!description->is_socket())
return ENOTSOCK;
auto& socket = *description->socket();
REQUIRE_PROMISE_FOR_SOCKET_DOMAIN(socket.domain());
if (socket.is_connected())
return EINVAL;
TRY(socket.listen(backlog));
return 0;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
}
Everywhere: Run clang-format
2022-04-01 17:58:27 +00:00
ErrorOr<FlatPtr> Process::sys$accept4(Userspace<Syscall::SC_accept4_params const*> user_params)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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: Require semicolon after VERIFY_{NO_,}PROCESS_BIG_LOCK_ACQUIRED This matches out general macro use, and specifically other verification macros like VERIFY(), VERIFY_NOT_REACHED(), VERIFY_INTERRUPTS_ENABLED(), and VERIFY_INTERRUPTS_DISABLED().
2022-08-17 20:03:04 +00:00
VERIFY_NO_PROCESS_BIG_LOCK(this);
Kernel: Handle promise violations in the syscall handler Previously we would crash the process immediately when a promise violation was found during a syscall. This is error prone, as we don't unwind the stack. This means that in certain cases we can leak resources, like an OwnPtr / RefPtr tracked on the stack. Or even leak a lock acquired in a ScopeLockLocker. To remedy this situation we move the promise violation handling to the syscall handler, right before we return to user space. This allows the code to follow the normal unwind path, and grantees there is no longer any cleanup that needs to occur. The Process::require_promise() and Process::require_no_promises() functions were modified to return ErrorOr<void> so we enforce that the errors are always propagated by the caller.
2021-12-29 09:11:45 +00:00
TRY(require_promise(Pledge::accept));
Kernel: Use copy_typed_from_user<T> for fetching syscall parameters
2021-09-05 15:51:37 +00:00
auto params = TRY(copy_typed_from_user(user_params));
Kernel+Userspace: Implement the accept4() system call Unlike accept() the new accept4() system call lets the caller specify flags for the newly accepted socket file descriptor, such as SOCK_CLOEXEC and SOCK_NONBLOCK.
2021-05-16 17:56:11 +00:00
int accepting_socket_fd = params.sockfd;
Userspace<sockaddr*> user_address((FlatPtr)params.addr);
Userspace<socklen_t*> user_address_size((FlatPtr)params.addrlen);
int flags = params.flags;
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
socklen_t address_size = 0;
Kernel: Remove big lock from `sys$accept4` The only thing we needed to check is whether `socket.accept()` returns a socket, and if not, we go back to blocking again.
2022-04-09 15:28:31 +00:00
if (user_address)
Everywhere: Run clang-format
2022-04-01 17:58:27 +00:00
TRY(copy_from_user(&address_size, static_ptr_cast<socklen_t const*>(user_address_size)));
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: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
ScopedDescriptionAllocation fd_allocation;
Kernel: Use non-locking {Nonnull,}RefPtr for OpenFileDescription This patch switches away from {Nonnull,}LockRefPtr to the non-locking smart pointers throughout the kernel. I've looked at the handful of places where these were being persisted and I don't see any race situations. Note that the process file descriptor table (Process::m_fds) was already guarded via MutexProtected.
2023-03-06 18:29:25 +00:00
RefPtr<OpenFileDescription> accepting_socket_description;
Kernel: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
Kernel: Switch process file descriptor table from spinlock to mutex There's no reason for this to use a spinlock. Instead, let's allow threads to block if someone else is using the descriptor table.
2022-01-29 00:29:07 +00:00
TRY(m_fds.with_exclusive([&](auto& fds) -> ErrorOr<void> {
Kernel: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
fd_allocation = TRY(fds.allocate());
accepting_socket_description = TRY(fds.open_file_description(accepting_socket_fd));
return {};
}));
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 (!accepting_socket_description->is_socket())
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 ENOTSOCK;
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
auto& socket = *accepting_socket_description->socket();
Kernel: Make self-contained locking smart pointers their own classes Until now, our kernel has reimplemented a number of AK classes to provide automatic internal locking: - RefPtr - NonnullRefPtr - WeakPtr - Weakable This patch renames the Kernel classes so that they can coexist with the original AK classes: - RefPtr => LockRefPtr - NonnullRefPtr => NonnullLockRefPtr - WeakPtr => LockWeakPtr - Weakable => LockWeakable The goal here is to eventually get rid of the Lock* classes in favor of using external locking.
2022-08-19 18:53:40 +00:00
LockRefPtr<Socket> accepted_socket;
Kernel: Remove big lock from `sys$accept4` The only thing we needed to check is whether `socket.accept()` returns a socket, and if not, we go back to blocking again.
2022-04-09 15:28:31 +00:00
for (;;) {
accepted_socket = socket.accept();
if (accepted_socket)
break;
if (!accepting_socket_description->is_blocking())
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 EAGAIN;
Kernel: Remove big lock from `sys$accept4` The only thing we needed to check is whether `socket.accept()` returns a socket, and if not, we go back to blocking again.
2022-04-09 15:28:31 +00:00
auto unblock_flags = Thread::FileBlocker::BlockFlags::None;
if (Thread::current()->block<Thread::AcceptBlocker>({}, *accepting_socket_description, unblock_flags).was_interrupted())
return EINTR;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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 (user_address) {
Kernel: Fix info leak from `sockaddr_un` in socket syscalls In `sys$accept4()` and `get_sock_or_peer_name()` we were not initializing the padding of the `sockaddr_un` struct, leading to an kernel information leak if the caller looked back at it's contents. Before Fix: 37.766 Clipboard(11:11): accept4 Bytes: 2f746d702f706f7274616c2f636c6970626f61726440eac130e7fbc1e8abbfc 19c10ffc18440eac15485bcc130e7fbc1549feaca6c9deaca549feaca1bb0bc 03efdf62c0e056eac1b402d7acd010ffc14602000001b0bc030100000050bf0 5c24602000001e7fbc1b402d7ac6bdc After Fix: 0.603 Clipboard(11:11): accept4 Bytes: 2f746d702f706f7274616c2f636c6970626f617264000000000000000000000 000000000000000000000000000000000000000000000000000000000000000 000000000000000000000000000000000000000000000000000000000000000 0000000000000000000000000000000
2021-12-29 10:54:25 +00:00
sockaddr_un address_buffer {};
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
address_size = min(sizeof(sockaddr_un), static_cast<size_t>(address_size));
Everywhere: Fix some alignment issues When creating uninitialized storage for variables, we need to make sure that the alignment is correct. Fixes a KUBSAN failure when running kernels compiled with Clang. In `Syscalls/socket.cpp`, we can simply use local variables, as `sockaddr_un` is a POD type. Along with moving the `alignas` specifier to the correct member, `AK::Optional`'s internal buffer has been made non-zeroed by default. GCC emitted bogus uninitialized memory access warnings, so we now use `__builtin_launder` to tell the compiler that we know what we are doing. This might disable some optimizations, but judging by how GCC failed to notice that the memory's initialization is dependent on `m_has_value`, I'm not sure that's a bad thing.
2021-07-01 09:29:28 +00:00
accepted_socket->get_peer_address((sockaddr*)&address_buffer, &address_size);
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_address, &address_buffer, address_size));
TRY(copy_to_user(user_address_size, &address_size));
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
}
Kernel: Rename FileDescription => OpenFileDescription Dr. POSIX really calls these "open file description", not just "file description", so let's call them exactly that. :^)
2021-09-07 11:39:11 +00:00
auto accepted_socket_description = TRY(OpenFileDescription::try_create(*accepted_socket));
Kernel: Improve ProcFS behavior in low memory conditions When ProcFS could no longer allocate KBuffer objects to serve calls to read, it would just return 0, indicating EOF. This then triggered parsing errors because code assumed it read the file. Because read isn't supposed to return ENOMEM, change ProcFS to populate the file data upon file open or seek to the beginning. This also means that calls to open can now return ENOMEM if needed. This allows the caller to either be able to successfully open the file and read it, or fail to open it in the first place.
2020-09-17 19:51:09 +00:00
Kernel: Use TRY() in sys$socket() and friends
2021-09-05 13:54:53 +00:00
accepted_socket_description->set_readable(true);
accepted_socket_description->set_writable(true);
Kernel+Userspace: Implement the accept4() system call Unlike accept() the new accept4() system call lets the caller specify flags for the newly accepted socket file descriptor, such as SOCK_CLOEXEC and SOCK_NONBLOCK.
2021-05-16 17:56:11 +00:00
if (flags & SOCK_NONBLOCK)
Kernel: Use TRY() in sys$socket() and friends
2021-09-05 13:54:53 +00:00
accepted_socket_description->set_blocking(false);
Kernel+Userspace: Implement the accept4() system call Unlike accept() the new accept4() system call lets the caller specify flags for the newly accepted socket file descriptor, such as SOCK_CLOEXEC and SOCK_NONBLOCK.
2021-05-16 17:56:11 +00:00
int fd_flags = 0;
if (flags & SOCK_CLOEXEC)
fd_flags |= FD_CLOEXEC;
Kernel: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
Kernel: Switch process file descriptor table from spinlock to mutex There's no reason for this to use a spinlock. Instead, let's allow threads to block if someone else is using the descriptor table.
2022-01-29 00:29:07 +00:00
TRY(m_fds.with_exclusive([&](auto& fds) -> ErrorOr<void> {
Kernel: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
fds[fd_allocation.fd].set(move(accepted_socket_description), fd_flags);
return {};
}));
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
// NOTE: Moving this state to Completed is what causes connect() to unblock on the client side.
accepted_socket->set_setup_state(Socket::SetupState::Completed);
Kernel: Use TRY() in sys$socket() and friends
2021-09-05 13:54:53 +00:00
return fd_allocation.fd;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
}
Everywhere: Run clang-format
2022-04-01 17:58:27 +00:00
ErrorOr<FlatPtr> Process::sys$connect(int sockfd, Userspace<sockaddr const*> user_address, socklen_t user_address_size)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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: Require semicolon after VERIFY_{NO_,}PROCESS_BIG_LOCK_ACQUIRED This matches out general macro use, and specifically other verification macros like VERIFY(), VERIFY_NOT_REACHED(), VERIFY_INTERRUPTS_ENABLED(), and VERIFY_INTERRUPTS_DISABLED().
2022-08-17 20:03:04 +00:00
VERIFY_NO_PROCESS_BIG_LOCK(this);
Kernel: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
auto description = TRY(open_file_description(sockfd));
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
if (!description->is_socket())
Kernel: Make all syscall functions return KResultOr<T> This makes it a lot easier to return errors since we no longer have to worry about negating EFOO errors and can just return them flat.
2021-03-01 12:49:16 +00:00
return ENOTSOCK;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
auto& socket = *description->socket();
REQUIRE_PROMISE_FOR_SOCKET_DOMAIN(socket.domain());
Kernel: Make Socket::connect() take credentials as input
2022-08-21 14:35:03 +00:00
TRY(socket.connect(credentials(), *description, user_address, user_address_size));
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
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
}
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
ErrorOr<FlatPtr> Process::sys$shutdown(int sockfd, int how)
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: Require semicolon after VERIFY_{NO_,}PROCESS_BIG_LOCK_ACQUIRED This matches out general macro use, and specifically other verification macros like VERIFY(), VERIFY_NOT_REACHED(), VERIFY_INTERRUPTS_ENABLED(), and VERIFY_INTERRUPTS_DISABLED().
2022-08-17 20:03:04 +00:00
VERIFY_NO_PROCESS_BIG_LOCK(this);
Kernel: Handle promise violations in the syscall handler Previously we would crash the process immediately when a promise violation was found during a syscall. This is error prone, as we don't unwind the stack. This means that in certain cases we can leak resources, like an OwnPtr / RefPtr tracked on the stack. Or even leak a lock acquired in a ScopeLockLocker. To remedy this situation we move the promise violation handling to the syscall handler, right before we return to user space. This allows the code to follow the normal unwind path, and grantees there is no longer any cleanup that needs to occur. The Process::require_promise() and Process::require_no_promises() functions were modified to return ErrorOr<void> so we enforce that the errors are always propagated by the caller.
2021-12-29 09:11:45 +00:00
TRY(require_promise(Pledge::stdio));
Kernel: Accept SHUT_RD and SHUT_WR as shutdown() how values The previous check for valid how values assumed this field was a bitmap and that SHUT_RDWR was simply a bitwise or of SHUT_RD and SHUT_WR, which is not the case.
2022-07-09 19:26:58 +00:00
if (how != SHUT_RD && how != SHUT_WR && how != SHUT_RDWR)
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: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
auto description = TRY(open_file_description(sockfd));
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 (!description->is_socket())
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 ENOTSOCK;
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
auto& socket = *description->socket();
REQUIRE_PROMISE_FOR_SOCKET_DOMAIN(socket.domain());
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
TRY(socket.shutdown(how));
return 0;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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$sendmsg(int sockfd, Userspace<const struct msghdr*> user_msg, int flags)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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: Require semicolon after VERIFY_{NO_,}PROCESS_BIG_LOCK_ACQUIRED This matches out general macro use, and specifically other verification macros like VERIFY(), VERIFY_NOT_REACHED(), VERIFY_INTERRUPTS_ENABLED(), and VERIFY_INTERRUPTS_DISABLED().
2022-08-17 20:03:04 +00:00
VERIFY_PROCESS_BIG_LOCK_ACQUIRED(this);
Kernel: Handle promise violations in the syscall handler Previously we would crash the process immediately when a promise violation was found during a syscall. This is error prone, as we don't unwind the stack. This means that in certain cases we can leak resources, like an OwnPtr / RefPtr tracked on the stack. Or even leak a lock acquired in a ScopeLockLocker. To remedy this situation we move the promise violation handling to the syscall handler, right before we return to user space. This allows the code to follow the normal unwind path, and grantees there is no longer any cleanup that needs to occur. The Process::require_promise() and Process::require_no_promises() functions were modified to return ErrorOr<void> so we enforce that the errors are always propagated by the caller.
2021-12-29 09:11:45 +00:00
TRY(require_promise(Pledge::stdio));
Kernel: Use copy_typed_from_user() in more places :^)
2021-12-17 08:12:20 +00:00
auto msg = TRY(copy_typed_from_user(user_msg));
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+UserspaceEmulator: Mostly add recvmsg(), sendmsg() The implementation only supports a single iovec for now. Some might say having more than one iovec is the main point of recvmsg() and sendmsg(), but I'm interested in the control message bits.
2020-09-16 15:45:00 +00:00
if (msg.msg_iovlen != 1)
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 ENOTSUP; // FIXME: Support this :)
Kernel+LibC+UserspaceEmulator: Mostly add recvmsg(), sendmsg() The implementation only supports a single iovec for now. Some might say having more than one iovec is the main point of recvmsg() and sendmsg(), but I'm interested in the control message bits.
2020-09-16 15:45:00 +00:00
Vector<iovec, 1> iovs;
AK: Make Vector::try_* functions return ErrorOr<void> Instead of signalling allocation failure with a bool return value (false), we now use ErrorOr<void> and return ENOMEM as appropriate. This allows us to use TRY() and MUST() with Vector. :^)
2021-11-10 10:55:37 +00:00
TRY(iovs.try_resize(msg.msg_iovlen));
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_n_from_user(iovs.data(), msg.msg_iov, msg.msg_iovlen));
Kernel: Remove various other uses of ssize_t
2021-06-16 14:44:15 +00:00
if (iovs[0].iov_len > NumericLimits<ssize_t>::max())
return EINVAL;
Kernel+LibC+UserspaceEmulator: Mostly add recvmsg(), sendmsg() The implementation only supports a single iovec for now. Some might say having more than one iovec is the main point of recvmsg() and sendmsg(), but I'm interested in the control message bits.
2020-09-16 15:45:00 +00:00
Everywhere: Run clang-format
2022-04-01 17:58:27 +00:00
Userspace<sockaddr const*> user_addr((FlatPtr)msg.msg_name);
Kernel+LibC+UserspaceEmulator: Mostly add recvmsg(), sendmsg() The implementation only supports a single iovec for now. Some might say having more than one iovec is the main point of recvmsg() and sendmsg(), but I'm interested in the control message bits.
2020-09-16 15:45:00 +00:00
socklen_t addr_length = msg.msg_namelen;
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: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
auto description = TRY(open_file_description(sockfd));
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
if (!description->is_socket())
Kernel: Make all syscall functions return KResultOr<T> This makes it a lot easier to return errors since we no longer have to worry about negating EFOO errors and can just return them flat.
2021-03-01 12:49:16 +00:00
return ENOTSOCK;
Kernel: Ensure socket is suitable for writing in sys$sendmsg Previously we would return a bytes written value of 0 if the writing end of the socket was full. Now we either exit with EAGAIN if the socket description is non-blocking, or block until the description can be written to. This is mostly a copy of the conditions in sys$write but with the "total nwritten" parts removed as sys$sendmsg does not have that.
2022-02-07 11:03:34 +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
auto& socket = *description->socket();
Kernel: Add support for MSG_NOSIGNAL and properly send SIGPIPE Previously we didn't send the SIGPIPE signal to processes when sendto()/sendmsg()/etc. returned EPIPE. And now we do. This also adds support for MSG_NOSIGNAL to suppress the signal.
2022-10-23 08:30:12 +00:00
if (socket.is_shut_down_for_writing()) {
if ((flags & MSG_NOSIGNAL) == 0)
Thread::current()->send_signal(SIGPIPE, &Process::current());
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 EPIPE;
Kernel: Add support for MSG_NOSIGNAL and properly send SIGPIPE Previously we didn't send the SIGPIPE signal to processes when sendto()/sendmsg()/etc. returned EPIPE. And now we do. This also adds support for MSG_NOSIGNAL to suppress the signal.
2022-10-23 08:30:12 +00:00
}
Kernel: Support sending filedescriptors with sendmsg(2) and SCM_RIGHTS This is necessary to support the wayland protocol. I also moved the CMSG_* macros to the kernel API since they are used in both kernel and userspace. this does not break ntpquery/SCM_TIMESTAMP.
2023-02-13 23:01:13 +00:00
if (msg.msg_controllen > 0) {
// Handle command messages.
auto cmsg_buffer = TRY(ByteBuffer::create_uninitialized(msg.msg_controllen));
TRY(copy_from_user(cmsg_buffer.data(), msg.msg_control, msg.msg_controllen));
msg.msg_control = cmsg_buffer.data();
for (struct cmsghdr* cmsg = CMSG_FIRSTHDR(&msg); cmsg != nullptr; cmsg = CMSG_NXTHDR(&msg, cmsg)) {
if (socket.is_local() && cmsg->cmsg_level == SOL_SOCKET && cmsg->cmsg_type == SCM_RIGHTS) {
auto& local_socket = static_cast<LocalSocket&>(socket);
int* fds = (int*)CMSG_DATA(cmsg);
size_t nfds = (cmsg->cmsg_len - CMSG_ALIGN(sizeof(struct cmsghdr))) / sizeof(int);
for (size_t i = 0; i < nfds; ++i) {
TRY(local_socket.sendfd(*description, TRY(open_file_description(fds[i]))));
}
}
}
}
Kernel: Make UserOrKernelBuffer::for_user_buffer() return ErrorOr<T> This simplifies EFAULT propagation with TRY(). :^)
2021-11-21 11:24:32 +00:00
auto data_buffer = TRY(UserOrKernelBuffer::for_user_buffer((u8*)iovs[0].iov_base, iovs[0].iov_len));
Kernel: Ensure socket is suitable for writing in sys$sendmsg Previously we would return a bytes written value of 0 if the writing end of the socket was full. Now we either exit with EAGAIN if the socket description is non-blocking, or block until the description can be written to. This is mostly a copy of the conditions in sys$write but with the "total nwritten" parts removed as sys$sendmsg does not have that.
2022-02-07 11:03:34 +00:00
while (true) {
while (!description->can_write()) {
if (!description->is_blocking()) {
return EAGAIN;
}
auto unblock_flags = Thread::FileBlocker::BlockFlags::None;
if (Thread::current()->block<Thread::WriteBlocker>({}, *description, unblock_flags).was_interrupted()) {
return EINTR;
}
// TODO: handle exceptions in unblock_flags
}
Kernel: Add support for MSG_NOSIGNAL and properly send SIGPIPE Previously we didn't send the SIGPIPE signal to processes when sendto()/sendmsg()/etc. returned EPIPE. And now we do. This also adds support for MSG_NOSIGNAL to suppress the signal.
2022-10-23 08:30:12 +00:00
auto bytes_sent_or_error = socket.sendto(*description, data_buffer, iovs[0].iov_len, flags, user_addr, addr_length);
if (bytes_sent_or_error.is_error()) {
if ((flags & MSG_NOSIGNAL) == 0 && bytes_sent_or_error.error().code() == EPIPE)
Thread::current()->send_signal(SIGPIPE, &Process::current());
return bytes_sent_or_error.release_error();
}
auto bytes_sent = bytes_sent_or_error.release_value();
Kernel: Ensure socket is suitable for writing in sys$sendmsg Previously we would return a bytes written value of 0 if the writing end of the socket was full. Now we either exit with EAGAIN if the socket description is non-blocking, or block until the description can be written to. This is mostly a copy of the conditions in sys$write but with the "total nwritten" parts removed as sys$sendmsg does not have that.
2022-02-07 11:03:34 +00:00
if (bytes_sent > 0)
return bytes_sent;
}
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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$recvmsg(int sockfd, Userspace<struct msghdr*> user_msg, int flags)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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: Require semicolon after VERIFY_{NO_,}PROCESS_BIG_LOCK_ACQUIRED This matches out general macro use, and specifically other verification macros like VERIFY(), VERIFY_NOT_REACHED(), VERIFY_INTERRUPTS_ENABLED(), and VERIFY_INTERRUPTS_DISABLED().
2022-08-17 20:03:04 +00:00
VERIFY_PROCESS_BIG_LOCK_ACQUIRED(this);
Kernel: Handle promise violations in the syscall handler Previously we would crash the process immediately when a promise violation was found during a syscall. This is error prone, as we don't unwind the stack. This means that in certain cases we can leak resources, like an OwnPtr / RefPtr tracked on the stack. Or even leak a lock acquired in a ScopeLockLocker. To remedy this situation we move the promise violation handling to the syscall handler, right before we return to user space. This allows the code to follow the normal unwind path, and grantees there is no longer any cleanup that needs to occur. The Process::require_promise() and Process::require_no_promises() functions were modified to return ErrorOr<void> so we enforce that the errors are always propagated by the caller.
2021-12-29 09:11:45 +00:00
TRY(require_promise(Pledge::stdio));
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
Kernel+LibC+UserspaceEmulator: Mostly add recvmsg(), sendmsg() The implementation only supports a single iovec for now. Some might say having more than one iovec is the main point of recvmsg() and sendmsg(), but I'm interested in the control message bits.
2020-09-16 15:45:00 +00:00
struct msghdr msg;
Kernel: Make copy_{from,to}_user() return KResult and use TRY() This makes EFAULT propagation flow much more naturally. :^)
2021-09-05 15:38:37 +00:00
TRY(copy_from_user(&msg, user_msg));
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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+UserspaceEmulator: Mostly add recvmsg(), sendmsg() The implementation only supports a single iovec for now. Some might say having more than one iovec is the main point of recvmsg() and sendmsg(), but I'm interested in the control message bits.
2020-09-16 15:45:00 +00:00
if (msg.msg_iovlen != 1)
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 ENOTSUP; // FIXME: Support this :)
Kernel+LibC+UserspaceEmulator: Mostly add recvmsg(), sendmsg() The implementation only supports a single iovec for now. Some might say having more than one iovec is the main point of recvmsg() and sendmsg(), but I'm interested in the control message bits.
2020-09-16 15:45:00 +00:00
Vector<iovec, 1> iovs;
AK: Make Vector::try_* functions return ErrorOr<void> Instead of signalling allocation failure with a bool return value (false), we now use ErrorOr<void> and return ENOMEM as appropriate. This allows us to use TRY() and MUST() with Vector. :^)
2021-11-10 10:55:37 +00:00
TRY(iovs.try_resize(msg.msg_iovlen));
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_n_from_user(iovs.data(), msg.msg_iov, msg.msg_iovlen));
Kernel+LibC+UserspaceEmulator: Mostly add recvmsg(), sendmsg() The implementation only supports a single iovec for now. Some might say having more than one iovec is the main point of recvmsg() and sendmsg(), but I'm interested in the control message bits.
2020-09-16 15:45:00 +00:00
Userspace<sockaddr*> user_addr((FlatPtr)msg.msg_name);
Userspace<socklen_t*> user_addr_length(msg.msg_name ? (FlatPtr)&user_msg.unsafe_userspace_ptr()->msg_namelen : 0);
Kernel: Make copy_to/from_user safe and remove unnecessary checks Since the CPU already does almost all necessary validation steps for us, we don't really need to attempt to do this. Doing it ourselves doesn't really work very reliably, because we'd have to account for other processors modifying virtual memory, and we'd have to account for e.g. pages not being able to be allocated due to insufficient resources. So change the copy_to/from_user (and associated helper functions) to use the new safe_memcpy, which will return whether it succeeded or not. The only manual validation step needed (which the CPU can't perform for us) is making sure the pointers provided by user mode aren't pointing to kernel mappings. To make it easier to read/write from/to either kernel or user mode data add the UserOrKernelBuffer helper class, which will internally either use copy_from/to_user or directly memcpy, or pass the data through directly using a temporary buffer on the stack. Last but not least we need to keep syscall params trivial as we need to copy them from/to user mode using copy_from/to_user.
2020-09-12 03:11:07 +00:00
Kernel: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
auto description = TRY(open_file_description(sockfd));
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
if (!description->is_socket())
Kernel: Make all syscall functions return KResultOr<T> This makes it a lot easier to return errors since we no longer have to worry about negating EFOO errors and can just return them flat.
2021-03-01 12:49:16 +00:00
return ENOTSOCK;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
auto& socket = *description->socket();
if (socket.is_shut_down_for_reading())
return 0;
Kernel: Make UserOrKernelBuffer::for_user_buffer() return ErrorOr<T> This simplifies EFAULT propagation with TRY(). :^)
2021-11-21 11:24:32 +00:00
auto data_buffer = TRY(UserOrKernelBuffer::for_user_buffer((u8*)iovs[0].iov_base, iovs[0].iov_len));
Kernel: Use UnixDateTime wherever applicable "Wherever applicable" = most places, actually :^), especially for networking and filesystem timestamps. This includes changes to unzip, which uses DOSPackedTime, since that is changed for the FAT file systems.
2023-03-13 21:11:13 +00:00
UnixDateTime timestamp {};
Kernel: Make sys$recvfrom() with MSG_DONTWAIT not so racy Instead of temporary changing the open file description's "blocking" flag while doing a non-waiting recvfrom, we instead plumb the currently wanted blocking behavior all the way through to the underlying socket.
2022-08-21 14:45:42 +00:00
bool blocking = (flags & MSG_DONTWAIT) ? false : description->is_blocking();
auto result = socket.recvfrom(*description, data_buffer, iovs[0].iov_len, flags, user_addr, user_addr_length, timestamp, blocking);
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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: Make File::write() and File::read() return KResultOr<size_t> Instead of returning a ssize_t where negative values mean error, we now return KResultOr<size_t> and use the error state to report errors exclusively.
2020-08-04 16:02:23 +00:00
if (result.is_error())
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
return result.release_error();
Kernel+LibC+UserspaceEmulator: Mostly add recvmsg(), sendmsg() The implementation only supports a single iovec for now. Some might say having more than one iovec is the main point of recvmsg() and sendmsg(), but I'm interested in the control message bits.
2020-09-16 15:45:00 +00:00
int msg_flags = 0;
if (result.value() > iovs[0].iov_len) {
Everywhere: Rename ASSERT => VERIFY (...and ASSERT_NOT_REACHED => VERIFY_NOT_REACHED) Since all of these checks are done in release builds as well, let's rename them to VERIFY to prevent confusion, as everyone is used to assertions being compiled out in release. We can introduce a new ASSERT macro that is specifically for debug checks, but I'm doing this wholesale conversion first since we've accumulated thousands of these already, and it's not immediately obvious which ones are suitable for ASSERT.
2021-02-23 19:42:32 +00:00
VERIFY(socket.type() != SOCK_STREAM);
Kernel+LibC+UserspaceEmulator: Mostly add recvmsg(), sendmsg() The implementation only supports a single iovec for now. Some might say having more than one iovec is the main point of recvmsg() and sendmsg(), but I'm interested in the control message bits.
2020-09-16 15:45:00 +00:00
msg_flags |= MSG_TRUNC;
}
Kernel: Support sending filedescriptors with sendmsg(2) and SCM_RIGHTS This is necessary to support the wayland protocol. I also moved the CMSG_* macros to the kernel API since they are used in both kernel and userspace. this does not break ntpquery/SCM_TIMESTAMP.
2023-02-13 23:01:13 +00:00
socklen_t current_cmsg_len = 0;
auto try_add_cmsg = [&](int level, int type, void const* data, socklen_t len) -> ErrorOr<bool> {
if (current_cmsg_len + len > msg.msg_controllen) {
Kernel+LibC+UserspaceEmulator: Add SO_TIMESTAMP, and cmsg definitions When SO_TIMESTAMP is set as an option on a SOCK_DGRAM socket, then recvmsg() will return a SCM_TIMESTAMP control message that contains a struct timeval with the system time that was current when the socket was received.
2020-09-16 16:32:45 +00:00
msg_flags |= MSG_CTRUNC;
Kernel: Support sending filedescriptors with sendmsg(2) and SCM_RIGHTS This is necessary to support the wayland protocol. I also moved the CMSG_* macros to the kernel API since they are used in both kernel and userspace. this does not break ntpquery/SCM_TIMESTAMP.
2023-02-13 23:01:13 +00:00
return false;
Kernel+LibC+UserspaceEmulator: Add SO_TIMESTAMP, and cmsg definitions When SO_TIMESTAMP is set as an option on a SOCK_DGRAM socket, then recvmsg() will return a SCM_TIMESTAMP control message that contains a struct timeval with the system time that was current when the socket was received.
2020-09-16 16:32:45 +00:00
}
Kernel: Support sending filedescriptors with sendmsg(2) and SCM_RIGHTS This is necessary to support the wayland protocol. I also moved the CMSG_* macros to the kernel API since they are used in both kernel and userspace. this does not break ntpquery/SCM_TIMESTAMP.
2023-02-13 23:01:13 +00:00
cmsghdr cmsg = { (socklen_t)CMSG_LEN(len), level, type };
cmsghdr* target = (cmsghdr*)(((char*)msg.msg_control) + current_cmsg_len);
TRY(copy_to_user(target, &cmsg));
TRY(copy_to_user(CMSG_DATA(target), data, len));
current_cmsg_len += CMSG_ALIGN(cmsg.cmsg_len);
return true;
};
if (socket.wants_timestamp()) {
timeval time = timestamp.to_timeval();
TRY(try_add_cmsg(SOL_SOCKET, SCM_TIMESTAMP, &time, sizeof(time)));
}
int space_for_fds = (msg.msg_controllen - current_cmsg_len - sizeof(struct cmsghdr)) / sizeof(int);
if (space_for_fds > 0 && socket.is_local()) {
auto& local_socket = static_cast<LocalSocket&>(socket);
auto descriptions = TRY(local_socket.recvfds(description, space_for_fds));
Vector<int> fdnums;
for (auto& description : descriptions) {
auto fd_allocation = TRY(m_fds.with_exclusive([](auto& fds) { return fds.allocate(); }));
Kernel: Stop using NonnullLockRefPtrVector
2023-03-06 16:56:28 +00:00
m_fds.with_exclusive([&](auto& fds) { fds[fd_allocation.fd].set(*description, 0); });
Kernel: Support sending filedescriptors with sendmsg(2) and SCM_RIGHTS This is necessary to support the wayland protocol. I also moved the CMSG_* macros to the kernel API since they are used in both kernel and userspace. this does not break ntpquery/SCM_TIMESTAMP.
2023-02-13 23:01:13 +00:00
fdnums.append(fd_allocation.fd);
}
TRY(try_add_cmsg(SOL_SOCKET, SCM_RIGHTS, fdnums.data(), fdnums.size() * sizeof(int)));
Kernel+LibC+UserspaceEmulator: Add SO_TIMESTAMP, and cmsg definitions When SO_TIMESTAMP is set as an option on a SOCK_DGRAM socket, then recvmsg() will return a SCM_TIMESTAMP control message that contains a struct timeval with the system time that was current when the socket was received.
2020-09-16 16:32:45 +00:00
}
Kernel: Support sending filedescriptors with sendmsg(2) and SCM_RIGHTS This is necessary to support the wayland protocol. I also moved the CMSG_* macros to the kernel API since they are used in both kernel and userspace. this does not break ntpquery/SCM_TIMESTAMP.
2023-02-13 23:01:13 +00:00
TRY(copy_to_user(&user_msg.unsafe_userspace_ptr()->msg_controllen, &current_cmsg_len));
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_msg.unsafe_userspace_ptr()->msg_flags, &msg_flags));
Kernel: Make File::write() and File::read() return KResultOr<size_t> Instead of returning a ssize_t where negative values mean error, we now return KResultOr<size_t> and use the error state to report errors exclusively.
2020-08-04 16:02:23 +00:00
return result.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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
}
Kernel: Make the getsockname/getpeername syscall helper a bit nicer Instead of templatizing on a bool parameter, use an enum for clarity.
2023-04-03 16:26:46 +00:00
template<Process::SockOrPeerName sock_or_peer_name, typename Params>
Everywhere: Run clang-format
2022-04-01 17:58:27 +00:00
ErrorOr<void> Process::get_sock_or_peer_name(Params const& params)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
{
socklen_t addrlen_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(&addrlen_value, params.addrlen, sizeof(socklen_t)));
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
if (addrlen_value <= 0)
Kernel: Make all syscall functions return KResultOr<T> This makes it a lot easier to return errors since we no longer have to worry about negating EFOO errors and can just return them flat.
2021-03-01 12:49:16 +00:00
return EINVAL;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
auto description = TRY(open_file_description(params.sockfd));
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
if (!description->is_socket())
Kernel: Make all syscall functions return KResultOr<T> This makes it a lot easier to return errors since we no longer have to worry about negating EFOO errors and can just return them flat.
2021-03-01 12:49:16 +00:00
return ENOTSOCK;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
auto& socket = *description->socket();
REQUIRE_PROMISE_FOR_SOCKET_DOMAIN(socket.domain());
Kernel: Fix info leak from `sockaddr_un` in socket syscalls In `sys$accept4()` and `get_sock_or_peer_name()` we were not initializing the padding of the `sockaddr_un` struct, leading to an kernel information leak if the caller looked back at it's contents. Before Fix: 37.766 Clipboard(11:11): accept4 Bytes: 2f746d702f706f7274616c2f636c6970626f61726440eac130e7fbc1e8abbfc 19c10ffc18440eac15485bcc130e7fbc1549feaca6c9deaca549feaca1bb0bc 03efdf62c0e056eac1b402d7acd010ffc14602000001b0bc030100000050bf0 5c24602000001e7fbc1b402d7ac6bdc After Fix: 0.603 Clipboard(11:11): accept4 Bytes: 2f746d702f706f7274616c2f636c6970626f617264000000000000000000000 000000000000000000000000000000000000000000000000000000000000000 000000000000000000000000000000000000000000000000000000000000000 0000000000000000000000000000000
2021-12-29 10:54:25 +00:00
sockaddr_un address_buffer {};
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
addrlen_value = min(sizeof(sockaddr_un), static_cast<size_t>(addrlen_value));
Kernel: Make the getsockname/getpeername syscall helper a bit nicer Instead of templatizing on a bool parameter, use an enum for clarity.
2023-04-03 16:26:46 +00:00
if constexpr (sock_or_peer_name == SockOrPeerName::SockName)
Everywhere: Fix some alignment issues When creating uninitialized storage for variables, we need to make sure that the alignment is correct. Fixes a KUBSAN failure when running kernels compiled with Clang. In `Syscalls/socket.cpp`, we can simply use local variables, as `sockaddr_un` is a POD type. Along with moving the `alignas` specifier to the correct member, `AK::Optional`'s internal buffer has been made non-zeroed by default. GCC emitted bogus uninitialized memory access warnings, so we now use `__builtin_launder` to tell the compiler that we know what we are doing. This might disable some optimizations, but judging by how GCC failed to notice that the memory's initialization is dependent on `m_has_value`, I'm not sure that's a bad thing.
2021-07-01 09:29:28 +00:00
socket.get_local_address((sockaddr*)&address_buffer, &addrlen_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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
else
Everywhere: Fix some alignment issues When creating uninitialized storage for variables, we need to make sure that the alignment is correct. Fixes a KUBSAN failure when running kernels compiled with Clang. In `Syscalls/socket.cpp`, we can simply use local variables, as `sockaddr_un` is a POD type. Along with moving the `alignas` specifier to the correct member, `AK::Optional`'s internal buffer has been made non-zeroed by default. GCC emitted bogus uninitialized memory access warnings, so we now use `__builtin_launder` to tell the compiler that we know what we are doing. This might disable some optimizations, but judging by how GCC failed to notice that the memory's initialization is dependent on `m_has_value`, I'm not sure that's a bad thing.
2021-07-01 09:29:28 +00:00
socket.get_peer_address((sockaddr*)&address_buffer, &addrlen_value);
Kernel: Make copy_{from,to}_user() return KResult and use TRY() This makes EFAULT propagation flow much more naturally. :^)
2021-09-05 15:38:37 +00:00
TRY(copy_to_user(params.addr, &address_buffer, addrlen_value));
return copy_to_user(params.addrlen, &addrlen_value);
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
}
Everywhere: Run clang-format
2022-04-01 17:58:27 +00:00
ErrorOr<FlatPtr> Process::sys$getsockname(Userspace<Syscall::SC_getsockname_params const*> user_params)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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: Require semicolon after VERIFY_{NO_,}PROCESS_BIG_LOCK_ACQUIRED This matches out general macro use, and specifically other verification macros like VERIFY(), VERIFY_NOT_REACHED(), VERIFY_INTERRUPTS_ENABLED(), and VERIFY_INTERRUPTS_DISABLED().
2022-08-17 20:03:04 +00:00
VERIFY_PROCESS_BIG_LOCK_ACQUIRED(this);
Kernel: Use copy_typed_from_user<T> for fetching syscall parameters
2021-09-05 15:51:37 +00:00
auto params = TRY(copy_typed_from_user(user_params));
Kernel: Make the getsockname/getpeername syscall helper a bit nicer Instead of templatizing on a bool parameter, use an enum for clarity.
2023-04-03 16:26:46 +00:00
TRY(get_sock_or_peer_name<SockOrPeerName::SockName>(params));
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
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
}
Everywhere: Run clang-format
2022-04-01 17:58:27 +00:00
ErrorOr<FlatPtr> Process::sys$getpeername(Userspace<Syscall::SC_getpeername_params const*> user_params)
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
{
Kernel: Require semicolon after VERIFY_{NO_,}PROCESS_BIG_LOCK_ACQUIRED This matches out general macro use, and specifically other verification macros like VERIFY(), VERIFY_NOT_REACHED(), VERIFY_INTERRUPTS_ENABLED(), and VERIFY_INTERRUPTS_DISABLED().
2022-08-17 20:03:04 +00:00
VERIFY_PROCESS_BIG_LOCK_ACQUIRED(this);
Kernel: Use copy_typed_from_user<T> for fetching syscall parameters
2021-09-05 15:51:37 +00:00
auto params = TRY(copy_typed_from_user(user_params));
Kernel: Make the getsockname/getpeername syscall helper a bit nicer Instead of templatizing on a bool parameter, use an enum for clarity.
2023-04-03 16:26:46 +00:00
TRY(get_sock_or_peer_name<SockOrPeerName::PeerName>(params));
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
return 0;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
}
Everywhere: Run clang-format
2022-04-01 17:58:27 +00:00
ErrorOr<FlatPtr> Process::sys$getsockopt(Userspace<Syscall::SC_getsockopt_params const*> user_params)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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: Require semicolon after VERIFY_{NO_,}PROCESS_BIG_LOCK_ACQUIRED This matches out general macro use, and specifically other verification macros like VERIFY(), VERIFY_NOT_REACHED(), VERIFY_INTERRUPTS_ENABLED(), and VERIFY_INTERRUPTS_DISABLED().
2022-08-17 20:03:04 +00:00
VERIFY_NO_PROCESS_BIG_LOCK(this);
Kernel: Use copy_typed_from_user<T> for fetching syscall parameters
2021-09-05 15:51:37 +00:00
auto params = TRY(copy_typed_from_user(user_params));
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: Use Userspace<T> for the getsockopt syscall and Socket interface The way getsockopt is implemented for socket types requires us to push down Userspace<T> using into those interfaces. This change does so, and utilizes proper copy implementations instead of the kind of haphazard pointer dereferencing that was occurring there before.
2020-08-07 09:29:05 +00:00
int sockfd = params.sockfd;
int level = params.level;
int option = params.option;
Kernel: Make copy_to/from_user safe and remove unnecessary checks Since the CPU already does almost all necessary validation steps for us, we don't really need to attempt to do this. Doing it ourselves doesn't really work very reliably, because we'd have to account for other processors modifying virtual memory, and we'd have to account for e.g. pages not being able to be allocated due to insufficient resources. So change the copy_to/from_user (and associated helper functions) to use the new safe_memcpy, which will return whether it succeeded or not. The only manual validation step needed (which the CPU can't perform for us) is making sure the pointers provided by user mode aren't pointing to kernel mappings. To make it easier to read/write from/to either kernel or user mode data add the UserOrKernelBuffer helper class, which will internally either use copy_from/to_user or directly memcpy, or pass the data through directly using a temporary buffer on the stack. Last but not least we need to keep syscall params trivial as we need to copy them from/to user mode using copy_from/to_user.
2020-09-12 03:11:07 +00:00
Userspace<void*> user_value((FlatPtr)params.value);
Userspace<socklen_t*> user_value_size((FlatPtr)params.value_size);
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: Use Userspace<T> for the getsockopt syscall and Socket interface The way getsockopt is implemented for socket types requires us to push down Userspace<T> using into those interfaces. This change does so, and utilizes proper copy implementations instead of the kind of haphazard pointer dereferencing that was occurring there before.
2020-08-07 09:29:05 +00:00
socklen_t value_size;
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(&value_size, params.value_size, sizeof(socklen_t)));
Kernel: Make copy_to/from_user safe and remove unnecessary checks Since the CPU already does almost all necessary validation steps for us, we don't really need to attempt to do this. Doing it ourselves doesn't really work very reliably, because we'd have to account for other processors modifying virtual memory, and we'd have to account for e.g. pages not being able to be allocated due to insufficient resources. So change the copy_to/from_user (and associated helper functions) to use the new safe_memcpy, which will return whether it succeeded or not. The only manual validation step needed (which the CPU can't perform for us) is making sure the pointers provided by user mode aren't pointing to kernel mappings. To make it easier to read/write from/to either kernel or user mode data add the UserOrKernelBuffer helper class, which will internally either use copy_from/to_user or directly memcpy, or pass the data through directly using a temporary buffer on the stack. Last but not least we need to keep syscall params trivial as we need to copy them from/to user mode using copy_from/to_user.
2020-09-12 03:11:07 +00:00
Kernel: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
auto description = TRY(open_file_description(sockfd));
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
if (!description->is_socket())
Kernel: Make all syscall functions return KResultOr<T> This makes it a lot easier to return errors since we no longer have to worry about negating EFOO errors and can just return them flat.
2021-03-01 12:49:16 +00:00
return ENOTSOCK;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
auto& socket = *description->socket();
Kernel: Remove pledge exception for sys$getsockopt() with SO_PEERCRED We had an exception that allowed SOL_SOCKET + SO_PEERCRED on local socket to support LibIPC's PID exchange mechanism. This is no longer needed so let's just remove the exception.
2021-01-31 08:27:36 +00:00
REQUIRE_PROMISE_FOR_SOCKET_DOMAIN(socket.domain());
Kernel: Replace KResult and KResultOr<T> with Error and ErrorOr<T> We now use AK::Error and AK::ErrorOr<T> in both kernel and userspace! This was a slightly tedious refactoring that took a long time, so it's not unlikely that some bugs crept in. Nevertheless, it does pass basic functionality testing, and it's just real nice to finally see the same pattern in all contexts. :^)
2021-11-07 23:51:39 +00:00
TRY(socket.getsockopt(*description, level, option, user_value, user_value_size));
return 0;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
}
Everywhere: Run clang-format
2022-04-01 17:58:27 +00:00
ErrorOr<FlatPtr> Process::sys$setsockopt(Userspace<Syscall::SC_setsockopt_params const*> user_params)
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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: Require semicolon after VERIFY_{NO_,}PROCESS_BIG_LOCK_ACQUIRED This matches out general macro use, and specifically other verification macros like VERIFY(), VERIFY_NOT_REACHED(), VERIFY_INTERRUPTS_ENABLED(), and VERIFY_INTERRUPTS_DISABLED().
2022-08-17 20:03:04 +00:00
VERIFY_NO_PROCESS_BIG_LOCK(this);
Kernel: Use copy_typed_from_user<T> for fetching syscall parameters
2021-09-05 15:51:37 +00:00
auto params = TRY(copy_typed_from_user(user_params));
Everywhere: Run clang-format
2022-04-01 17:58:27 +00:00
Userspace<void const*> user_value((FlatPtr)params.value);
Kernel: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
auto description = TRY(open_file_description(params.sockfd));
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
if (!description->is_socket())
Kernel: Make all syscall functions return KResultOr<T> This makes it a lot easier to return errors since we no longer have to worry about negating EFOO errors and can just return them flat.
2021-03-01 12:49:16 +00:00
return ENOTSOCK;
Kernel: Move syscall implementations out of Process.cpp This is something I've been meaning to do for a long time, and here we finally go. This patch moves all sys$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
auto& socket = *description->socket();
REQUIRE_PROMISE_FOR_SOCKET_DOMAIN(socket.domain());
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(socket.setsockopt(params.level, params.option, user_value, params.value_size));
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$foo functions out of Process.cpp and into files in Kernel/Syscalls/. It's not exactly one syscall per file (although it could be, but I got a bit tired of the repetitive work here..) This makes hacking on individual syscalls a lot less painful since you don't have to rebuild nearly as much code every time. I'm also hopeful that this makes it easier to understand individual syscalls. :^)
2020-07-30 21:38:15 +00:00
}
Everywhere: Run clang-format
2022-04-01 17:58:27 +00:00
ErrorOr<FlatPtr> Process::sys$socketpair(Userspace<Syscall::SC_socketpair_params const*> user_params)
Kernel+LibC: Implement the socketpair() syscall
2021-04-28 10:39:12 +00:00
{
Kernel: Require semicolon after VERIFY_{NO_,}PROCESS_BIG_LOCK_ACQUIRED This matches out general macro use, and specifically other verification macros like VERIFY(), VERIFY_NOT_REACHED(), VERIFY_INTERRUPTS_ENABLED(), and VERIFY_INTERRUPTS_DISABLED().
2022-08-17 20:03:04 +00:00
VERIFY_NO_PROCESS_BIG_LOCK(this);
Kernel: Use copy_typed_from_user<T> for fetching syscall parameters
2021-09-05 15:51:37 +00:00
auto params = TRY(copy_typed_from_user(user_params));
Kernel+LibC: Implement the socketpair() syscall
2021-04-28 10:39:12 +00:00
if (params.domain != AF_LOCAL)
return EINVAL;
if (params.protocol != 0 && params.protocol != PF_LOCAL)
return EINVAL;
Kernel: Use TRY() in sys$socket() and friends
2021-09-05 13:54:53 +00:00
auto pair = TRY(LocalSocket::try_create_connected_pair(params.type & SOCK_TYPE_MASK));
Kernel+LibC: Implement the socketpair() syscall
2021-04-28 10:39:12 +00:00
Kernel: Switch process file descriptor table from spinlock to mutex There's no reason for this to use a spinlock. Instead, let's allow threads to block if someone else is using the descriptor table.
2022-01-29 00:29:07 +00:00
return m_fds.with_exclusive([&](auto& fds) -> ErrorOr<FlatPtr> {
Kernel: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
auto fd_allocation0 = TRY(fds.allocate());
auto fd_allocation1 = TRY(fds.allocate());
int allocated_fds[2];
allocated_fds[0] = fd_allocation0.fd;
allocated_fds[1] = fd_allocation1.fd;
setup_socket_fd(fds, allocated_fds[0], pair.description0, params.type);
setup_socket_fd(fds, allocated_fds[1], pair.description1, params.type);
Kernel/Syscalls: Use copy_n_to_user when applicable copy_to_user() with bytes as the last argument could be changed to using copy_n_to_user() with a count.
2023-03-24 08:59:28 +00:00
if (copy_n_to_user(params.sv, allocated_fds, 2).is_error()) {
Kernel: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +00:00
// Avoid leaking both file descriptors on error.
fds[allocated_fds[0]] = {};
fds[allocated_fds[1]] = {};
return EFAULT;
}
return 0;
});
Kernel+LibC: Implement the socketpair() syscall
2021-04-28 10:39:12 +00:00
}
Kernel: Convert process file descriptor table to a SpinlockProtected Instead of manually locking in the various member functions of Process::OpenFileDescriptions, simply wrap it in a SpinlockProtected.
2022-01-29 00:22:28 +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
}
Reference in a new issue Copy permalink