summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2020-02-23Kernel: Commit the entire region up front in KBuffer::copy()Andreas Kling
Since we know exactly how much physical memory we'll need, we might as well commit it up front instead of letting page faults drive it.
2020-02-23LibGUI: Don't require passing a parent to widget constructorsAndreas Kling
This is a step towards using Core::Object::add<T> more, which takes care of parenting the newly created child automatically.
2020-02-23LibCore: Add Core::Object::add<T> helper for creating child objectsAndreas Kling
Consider the old pattern for creating a Core::Object parent and child: auto parent = Core::Object::construct(...); auto child = Core::Object::construct(..., parent); The above was an artifact of the pre-reference-counting Object era. Now that objects have less esoteric lifetime management, we can replace the old pattern with something more expressive: auto parent = Core::Object::construct(...); auto child = parent->add<Core::Object>(...); This reads a lot more naturally, and it also means we can get rid of all the parent pointer arguments to Core::Object subclass constructors.
2020-02-23LibGUI: Add some missing widget classes to Forward.hAndreas Kling
2020-02-23LibGfx: Add a way to construct an empty Font with arbitrary metricsAndreas Kling
2020-02-23Userland: Delete redundant code in truncateShannon Booth
Fixes #1262
2020-02-22Shell: Make some functions constShannon Booth
2020-02-22Shell: Add basic tilde expansionShannon Booth
This does not work with shell completion yet, but the basics of being able a cd being able to expand "~" to the current user's home directory, and "~foo" to the home directory of user "foo" is added in this commit Work towards: #115
2020-02-22LibC: Implement strchrnul()Shannon Booth
2020-02-22AK: Add StringBuilder::is_empty()Shannon Booth
2020-02-22AK: Add StringView::starts_with(char) & StringView::ends_with(char)Shannon Booth
This is simply meant to be a more efficient implementation in the case that we only need to check a single character.
2020-02-22LibGUI: Remove debug spam when resizing table columnsAndreas Kling
2020-02-22Kernel: Build without debugging symbols by defaultAndreas Kling
Compiling with -g adds roughly 30% to kernel build times. Anyone who wants this can turn it on locally instead.
2020-02-22SystemMonitor: Fix display of file system size column.Brian Gianforcaro
The Size column in the "File systems" tab of SystemMonitor had a rendering artifact where the bounding box outline would be drawn in the same location as the text. This makes the text look strange and hard to read. Pad the size with a signle space character on either side to give the text a gap on each side.
2020-02-22TTYServer: Use unveil()Brian Gianforcaro
2020-02-22head: Use pledge()Brian Gianforcaro
2020-02-22SystemMenu: Silence debug spam on startupAndreas Kling
2020-02-22LibCore: Log a more helpful message when Socket::connect() failsAndreas Kling
Fixes #1272.
2020-02-22LibCore: Fix wrong return value in Core::Socket::destination_address()Andreas Kling
2020-02-22Ext2FS: Add Missing HashMap.h includeAndreas Kling
2020-02-22Kernel: Remove unnecessary allocation metadata from kmalloc() chunksAndreas Kling
Each allocation header was tracking its index into the chunk bitmap, but that index can be computed from the allocation address anyway. Removing this means that each allocation gets 4 more bytes of memory and this avoids allocating an extra chunk in many cases. :^)
2020-02-22LibGUI: Take scroll offset into account when manipulating table headersAndreas Kling
Fixes #1271.
2020-02-22Kernel: Make FileDescription slab-allocatedAndreas Kling
2020-02-22Kernel: Tweak SlabAllocator's slab sizesAndreas Kling
Nobody was using the 8-byte slab size, so get rid of it and move all of its capacity to the new 64-byte slab size (which replaces 48-byte.)
2020-02-22Kernel: Make Custody slab-allocatedAndreas Kling
2020-02-22Kernel: Increase kmalloc chunk size from 8 bytes to 32 bytesAndreas Kling
This gives a huge speedup when running "git status" in a SerenityOS repository directory. Most of the time was spent allocating strings.
2020-02-22Kernel: Disown shared buffers on sys$execve()Andreas Kling
When committing to a new executable, disown any shared buffers that the process was previously co-owning. Otherwise accessing the same shared buffer ID from the new program would cause the kernel to find a cached (and stale!) reference to the previous program's VM region corresponding to that shared buffer, leading to a Region* use-after-free. Fixes #1270.
2020-02-22ProfileViewer: Fix treeview selection looking unselected on Left keyAndreas Kling
When pressing the Left arrow key, we now travel to the parent_index() of the currently selected index. Our implementation of parent_index() was always returning an index with column 0, instead of using the same column as the current index. This prevented the selected item from looking selected.
2020-02-22ProfileViewer: Remove loading code for the old file formatAndreas Kling
We're still jumping through all kinds of silly hoops to load the new format, and this commit only gets rid of the API for loading old files.
2020-02-22LibELF: Avoid unnecessarily recomputing loop boundaries over and overAndreas Kling
2020-02-22ProfileViewer: Symbolicate unknown addresses as "??" for nowAndreas Kling
This makes unknown addresses accumulate their children together in the treeview, which turns out to be a bit more useful than having hundreds of unique garbage addresses each with their own subtree.
2020-02-22Kernel: Disable profiling during the critical section of sys$execve()Andreas Kling
Since we're gonna throw away these stacks at the end of exec anyway, we might as well disable profiling before starting to mess with the process page tables. One less weird situation to worry about in the sampling code.
2020-02-22profile: Allow launching a command with profiling enabledAndreas Kling
You can now profile a program from start to finish by doing: $ profile -c "cat /etc/passwd" The old "enable or disable profiling for a PID" mode is accessible via: $ profile -p <PID> -e # Enable profiling for PID $ profile -p <PID> -d # Disable profiling for PID The generated profile is available via /proc/profile like before. This is far from perfect, but it at least makes profiling a lot nicer to use since you don't have to hurry and attach to something when you want to profile the whole thing anyway.
2020-02-22Kernel: Reset profiling state on exec() (but keep it going)Andreas Kling
We now log the new executable on exec() and throw away all the samples we've accumulated so far. But profiling keeps going.
2020-02-22ProfileViewer: Symbolicate kernel addresses when possibleAndreas Kling
ProfileViewer will now attempt to open /boot/kernel and use that to symbolicate kernel addresses (anything above the 3GB mark.) In other words, if you run ProfileViewer as root, on a profile that was generated by root, you can now see kernel functions properly as well. This is not available to non-privileged users.
2020-02-22Kernel+ProfileViewer: Move symbolication to userspace for time profilesAndreas Kling
This makes the time profiles look like the memory profiles so we can use the userspace symbolication code in ProfileViewer.
2020-02-22Kernel: Fully validate pointers when walking stack during profilingAndreas Kling
It's not enough to just check that things wouldn't page fault, we also need to verify that addresses are accessible to the profiled thread.
2020-02-22Kernel: Put "Couldn't find user region" spam behind MM_DEBUGAndreas Kling
This basically never tells us anything actionable anyway, and it's a real annoyance when doing something validation-heavy like profiling.
2020-02-21WindowServer: Use system theme for the drag and drop popupsTibor Nagy
2020-02-21LibGUI: Improve TreeView keyboard navigationTibor Nagy
This commit adds two new behaviour to the key event handler of the TreeView widget: Pressing left now jumps to the parent node if the current treenode is closed or has no children. Pressing right now jumps to the first children node if the current treenode is open.
2020-02-21Ext2FS: The max current block count of a file is size/block_sizeAndreas Kling
Turns out that i_blocks does not take block list holes into account.
2020-02-21Kernel: Log instead of crashing when getting a page fault during IRQAndreas Kling
This is definitely a bug, but it seems to happen randomly every now and then and we need more info to track it down, so let's log for now.
2020-02-21Toolchain: Build demangling into LibC except during toolchain buildAndreas Kling
2020-02-21Ext2FS: Allow holes in block listsAndreas Kling
Linux creates holes in block lists for all-zero content. This is very reasonable and we can now handle that situation as well. Note that we're not smart enough to generate these holes ourselves yet, but now we can at least read from such files.
2020-02-21LibELF: Use the ELF_STRTAB string constant instead of hard-codingAndreas Kling
2020-02-21Kernel: Fix crash when reading /proc/PID/vmobjectsAndreas Kling
InodeVMObjects can have nulled-out physical page slots. That just means we haven't cached that page from disk right now.
2020-02-21Kernel: Don't trigger page faults during profiling stack walkAndreas Kling
The kernel sampling profiler will walk thread stacks during the timer tick handler. Since it's not safe to trigger page faults during IRQ's, we now avoid this by checking the page tables manually before accessing each stack location.
2020-02-21Kernel: Commit the profiling sample buffer memory up frontAndreas Kling
This avoids getting page faults while storing samples in the timer IRQ.
2020-02-21Kernel: Expose the underlying Region of a KBufferAndreas Kling
2020-02-21Kernel: Assert on page fault during IRQAndreas Kling
We're not equipped to deal with page faults during an IRQ handler, so add an assertion so we can immediately tell what's wrong. This is why profiling sometimes hangs the system -- walking the stack of the profiled thread causes a page fault and things fall apart.
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-15 11:59:32 +0000