LadybirdBrowser/ladybird
1
1
Fork 0
mirror of https://github.com/LadybirdBrowser/ladybird.git synced 2024-11-10 02:54:54 +03:00
Code Issues Packages Projects Releases Wiki Activity
7235ddfd98
ladybird/Kernel/SharedBuffer.h

101 lines
3.3 KiB
C
Raw Normal View History

Meta: Add license header to source files As suggested by Joshua, this commit adds the 2-clause BSD license as a comment block to the top of every source file. For the first pass, I've just added myself for simplicity. I encourage everyone to add themselves as copyright holders of any file they've added or modified in some significant way. If I've added myself in error somewhere, feel free to replace it with the appropriate copyright holder instead. Going forward, all new source files should include a license header.
2020-01-18 11:38:21 +03:00
/*
* Copyright (c) 2018-2020, Andreas Kling <kling@serenityos.org>
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* 1. Redistributions of source code must retain the above copyright notice, this
* list of conditions and the following disclaimer.
*
* 2. Redistributions in binary form must reproduce the above copyright notice,
* this list of conditions and the following disclaimer in the documentation
* and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
* AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
* DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
* SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
* CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
* OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
#pragma once
#include <AK/OwnPtr.h>
Kernel: Make Region weakable and use WeakPtr<Region> instead of Region* This turns use-after-free bugs into null pointer dereferences instead.
2020-02-24 15:24:30 +03:00
#include <AK/WeakPtr.h>
Kernel: Merge PurgeableVMObject into AnonymousVMObject This implements memory commitments and lazy-allocation of committed memory.
2020-09-06 00:52:14 +03:00
#include <Kernel/VM/AnonymousVMObject.h>
Kernel: Split VMObject into two classes: Anonymous- and InodeVMObject InodeVMObject is a VMObject with an underlying Inode in the filesystem. AnonymousVMObject has no Inode. I'm happy that InodeVMObject::inode() can now return Inode& instead of VMObject::inode() return Inode*. :^)
2019-08-07 19:06:17 +03:00
#include <Kernel/VM/MemoryManager.h>
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
Kernel: Move all code into the Kernel namespace
2020-02-16 03:27:42 +03:00
namespace Kernel {
class SharedBuffer {
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
private:
struct Reference {
Kernel: Fix SharedBuffer reference counting on fork We need to not only add a record for a reference, but we need to copy the reference count on fork as well, because the code in the fork assumes that it has the same amount of references, still. Also, once all references are dropped when a process is disowned, delete the shared buffer. Fixes #4076
2020-11-24 20:26:32 +03:00
Reference(ProcessID pid, unsigned count = 0)
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
: pid(pid)
Kernel: Fix SharedBuffer reference counting on fork We need to not only add a record for a reference, but we need to copy the reference count on fork as well, because the code in the fork assumes that it has the same amount of references, still. Also, once all references are dropped when a process is disowned, delete the shared buffer. Fixes #4076
2020-11-24 20:26:32 +03:00
, count(count)
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
{
}
Kernel: PID/TID typing This compiles, and contains exactly the same bugs as before. The regex 'FIXME: PID/' should reveal all markers that I left behind, including: - Incomplete conversion - Issues or things that look fishy - Actual bugs that will go wrong during runtime
2020-08-08 18:32:34 +03:00
ProcessID pid;
SharedBuffer: Fix a denial of service It's a very bad idea to increment the refcount on behalf of another process. That process may (for either benign or evil reasons) not reference the SharedBuffer, and then we'll be stuck with loads of SharedBuffers until we OOM. Instead, increment the refcount when the buffer is mapped. That way, a buffer is only kept if *someone* has explicitly requested it via get_shared_buffer. Fixes #341
2019-07-19 18:46:21 +03:00
unsigned count { 0 };
Kernel: Make Region weakable and use WeakPtr<Region> instead of Region* This turns use-after-free bugs into null pointer dereferences instead.
2020-02-24 15:24:30 +03:00
WeakPtr<Region> region;
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
};
Kernel: Split VMObject into two classes: Anonymous- and InodeVMObject InodeVMObject is a VMObject with an underlying Inode in the filesystem. AnonymousVMObject has no Inode. I'm happy that InodeVMObject::inode() can now return Inode& instead of VMObject::inode() return Inode*. :^)
2019-08-07 19:06:17 +03:00
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
public:
Kernel: Merge PurgeableVMObject into AnonymousVMObject This implements memory commitments and lazy-allocation of committed memory.
2020-09-06 00:52:14 +03:00
SharedBuffer(int id, NonnullRefPtr<AnonymousVMObject>&& vmobject)
Kernel+LibC: Rename shared buffer syscalls to use a prefix This feels a lot more consistent and Unixy: create_shared_buffer() => shbuf_create() share_buffer_with() => shbuf_allow_pid() share_buffer_globally() => shbuf_allow_all() get_shared_buffer() => shbuf_get() release_shared_buffer() => shbuf_release() seal_shared_buffer() => shbuf_seal() get_shared_buffer_size() => shbuf_get_size() Also, "shared_buffer_id" is shortened to "shbuf_id" all around.
2020-02-28 13:45:19 +03:00
: m_shbuf_id(id)
Kernel: Implement lazy committed page allocation By designating a committed page pool we can guarantee to have physical pages available for lazy allocation in mappings. However, when forking we will overcommit. The assumption is that worst-case it's better for the fork to die due to insufficient physical memory on COW access than the parent that created the region. If a fork wants to ensure that all memory is available (trigger a commit) then it can use madvise. This also means that fork now can gracefully fail if we don't have enough physical pages available.
2020-09-05 06:12:25 +03:00
, m_vmobject(move(vmobject))
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
{
#ifdef SHARED_BUFFER_DEBUG
Kernel: Implement lazy committed page allocation By designating a committed page pool we can guarantee to have physical pages available for lazy allocation in mappings. However, when forking we will overcommit. The assumption is that worst-case it's better for the fork to die due to insufficient physical memory on COW access than the parent that created the region. If a fork wants to ensure that all memory is available (trigger a commit) then it can use madvise. This also means that fork now can gracefully fail if we don't have enough physical pages available.
2020-09-05 06:12:25 +03:00
dbg() << "Created shared buffer " << m_shbuf_id << " of size " << m_vmobject->size();
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
#endif
}
~SharedBuffer()
{
#ifdef SHARED_BUFFER_DEBUG
Kernel+LibC: Rename shared buffer syscalls to use a prefix This feels a lot more consistent and Unixy: create_shared_buffer() => shbuf_create() share_buffer_with() => shbuf_allow_pid() share_buffer_globally() => shbuf_allow_all() get_shared_buffer() => shbuf_get() release_shared_buffer() => shbuf_release() seal_shared_buffer() => shbuf_seal() get_shared_buffer_size() => shbuf_get_size() Also, "shared_buffer_id" is shortened to "shbuf_id" all around.
2020-02-28 13:45:19 +03:00
dbg() << "Destroyed shared buffer " << m_shbuf_id << " of size " << size();
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
#endif
}
SharedBuffer: Amend commit 2d4d465206d I had the right cause of the SharedBuffer leak, but goofed the fix by desynching the per-pid refcount and the global refcount. Fix that, and add a generous sprinkle of asserts to make sure the two stay in sync. Fixes #341 (... for real this time)
2019-07-20 00:14:56 +03:00
void sanity_check(const char* what);
Kernel: PID/TID typing This compiles, and contains exactly the same bugs as before. The regex 'FIXME: PID/' should reveal all markers that I left behind, including: - Incomplete conversion - Issues or things that look fishy - Actual bugs that will go wrong during runtime
2020-08-08 18:32:34 +03:00
bool is_shared_with(ProcessID peer_pid) const;
SharedBuffer: Fix a denial of service It's a very bad idea to increment the refcount on behalf of another process. That process may (for either benign or evil reasons) not reference the SharedBuffer, and then we'll be stuck with loads of SharedBuffers until we OOM. Instead, increment the refcount when the buffer is mapped. That way, a buffer is only kept if *someone* has explicitly requested it via get_shared_buffer. Fixes #341
2019-07-19 18:46:21 +03:00
void* ref_for_process_and_get_address(Process& process);
Kernel: PID/TID typing This compiles, and contains exactly the same bugs as before. The regex 'FIXME: PID/' should reveal all markers that I left behind, including: - Incomplete conversion - Issues or things that look fishy - Actual bugs that will go wrong during runtime
2020-08-08 18:32:34 +03:00
void share_with(ProcessID peer_pid);
Kernel+ProcessManager: Let processes have an icon and show it in the table. Processes can now have an icon assigned, which is essentially a 16x16 RGBA32 bitmap exposed as a shared buffer ID. You set the icon ID by calling set_process_icon(int) and the icon ID will be exposed through /proc/all. To make this work, I added a mechanism for making shared buffers globally accessible. For safety reasons, each app seals the icon buffer before making it global. Right now the first call to GWindow::set_icon() is what determines the process icon. We'll probably change this in the future. :^)
2019-07-29 08:26:01 +03:00
void share_globally() { m_global = true; }
SharedBuffer: Fix a denial of service It's a very bad idea to increment the refcount on behalf of another process. That process may (for either benign or evil reasons) not reference the SharedBuffer, and then we'll be stuck with loads of SharedBuffers until we OOM. Instead, increment the refcount when the buffer is mapped. That way, a buffer is only kept if *someone* has explicitly requested it via get_shared_buffer. Fixes #341
2019-07-19 18:46:21 +03:00
void deref_for_process(Process& process);
Kernel: Fix SharedBuffer reference counting on fork We need to not only add a record for a reference, but we need to copy the reference count on fork as well, because the code in the fork assumes that it has the same amount of references, still. Also, once all references are dropped when a process is disowned, delete the shared buffer. Fixes #4076
2020-11-24 20:26:32 +03:00
bool disown(ProcessID pid);
Kernel: Inherit shared buffers when forking We need to create a reference for the new PID for each shared buffer that the process had a reference to. If the process subsequently get replaced through exec, those references will be dropped again. But if exec for some reason fails then other code, such as global destructors could still expect having access to them. Fixes #4076
2020-11-23 01:25:48 +03:00
static void share_all_shared_buffers(Process& from_process, Process& with_process);
Kernel: Rename "vmo" to "vmobject" everywhere
2019-09-04 12:27:14 +03:00
size_t size() const { return m_vmobject->size(); }
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
void destroy_if_unused();
void seal();
Kernel: Memory purging improvements This adds the ability for a Region to define volatile/nonvolatile areas within mapped memory using madvise(). This also means that memory purging takes into account all views of the PurgeableVMObject and only purges memory that is not needed by all of them. When calling madvise() to change an area to nonvolatile memory, return whether memory from that area was purged. At that time also try to remap all memory that is requested to be nonvolatile, and if insufficient pages are available notify the caller of that fact.
2020-09-03 07:57:09 +03:00
enum class SetVolatileError {
Success = 0,
NotPurgeable,
OutOfMemory,
NotMapped
};
SetVolatileError set_volatile_all(bool is_volatile, bool& was_purged);
Kernel: Merge PurgeableVMObject into AnonymousVMObject This implements memory commitments and lazy-allocation of committed memory.
2020-09-06 00:52:14 +03:00
AnonymousVMObject& vmobject() { return m_vmobject; }
const AnonymousVMObject& vmobject() const { return m_vmobject; }
Kernel+LibC: Rename shared buffer syscalls to use a prefix This feels a lot more consistent and Unixy: create_shared_buffer() => shbuf_create() share_buffer_with() => shbuf_allow_pid() share_buffer_globally() => shbuf_allow_all() get_shared_buffer() => shbuf_get() release_shared_buffer() => shbuf_release() seal_shared_buffer() => shbuf_seal() get_shared_buffer_size() => shbuf_get_size() Also, "shared_buffer_id" is shortened to "shbuf_id" all around.
2020-02-28 13:45:19 +03:00
int id() const { return m_shbuf_id; }
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
Kernel: Tidy up SharedBuffer interface Make is_shared_with() const and hide private members.
2020-05-16 04:01:34 +03:00
private:
Kernel+LibC: Rename shared buffer syscalls to use a prefix This feels a lot more consistent and Unixy: create_shared_buffer() => shbuf_create() share_buffer_with() => shbuf_allow_pid() share_buffer_globally() => shbuf_allow_all() get_shared_buffer() => shbuf_get() release_shared_buffer() => shbuf_release() seal_shared_buffer() => shbuf_seal() get_shared_buffer_size() => shbuf_get_size() Also, "shared_buffer_id" is shortened to "shbuf_id" all around.
2020-02-28 13:45:19 +03:00
int m_shbuf_id { -1 };
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
bool m_writable { true };
Kernel+ProcessManager: Let processes have an icon and show it in the table. Processes can now have an icon assigned, which is essentially a 16x16 RGBA32 bitmap exposed as a shared buffer ID. You set the icon ID by calling set_process_icon(int) and the icon ID will be exposed through /proc/all. To make this work, I added a mechanism for making shared buffers globally accessible. For safety reasons, each app seals the icon buffer before making it global. Right now the first call to GWindow::set_icon() is what determines the process icon. We'll probably change this in the future. :^)
2019-07-29 08:26:01 +03:00
bool m_global { false };
Kernel: Merge PurgeableVMObject into AnonymousVMObject This implements memory commitments and lazy-allocation of committed memory.
2020-09-06 00:52:14 +03:00
NonnullRefPtr<AnonymousVMObject> m_vmobject;
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
Vector<Reference, 2> m_refs;
SharedBuffer: Fix a denial of service It's a very bad idea to increment the refcount on behalf of another process. That process may (for either benign or evil reasons) not reference the SharedBuffer, and then we'll be stuck with loads of SharedBuffers until we OOM. Instead, increment the refcount when the buffer is mapped. That way, a buffer is only kept if *someone* has explicitly requested it via get_shared_buffer. Fixes #341
2019-07-19 18:46:21 +03:00
unsigned m_total_refs { 0 };
SharedBuffer: Fix deadlock on destroy We were locking the list of references, and then destroying the reference, which made things go a little crazy. It's more straightforward to just remove the per-reference lock: the syscalls all have to lock the full list anyway, so let's just do that and avoid the hassle. While I'm at it, also move the SharedBuffer code out to its own file as it's getting a little long and unwieldly, and Process.cpp is already huge.
2019-07-16 16:03:39 +03:00
};
Convert HashMap<Key, OwnPtr<T>> to HashMap<Key, NonnullOwnPtr<T>>. In every case I found, we never wanted to support null entry values. With NonnullOwnPtr, we can encode that at the type level. :^)
2019-07-24 09:42:55 +03:00
Lockable<HashMap<int, NonnullOwnPtr<SharedBuffer>>>& shared_buffers();
Kernel: Move all code into the Kernel namespace
2020-02-16 03:27:42 +03:00
}
Reference in New Issue Copy Permalink