aboutsummaryrefslogtreecommitdiff
path: root/pgo (follow)
Commit message (Collapse)AuthorAgeLines
* pgo: stub get_current_modifiers()Daniel Eklöf2021-12-04-0/+5
|
* pgo: add stub for xcursor_for_csd_border()Daniel Eklöf2021-11-30-0/+6
|
* foot/client: implement xdga client activationRonan Pigott2021-11-01-1/+1
| | | | | | | | | | | This is an application of the xdg activation protocol that will allow compositors to associate new foot toplevels with the command that launched them. footclient receives an activation token from the launcher which the compositor can use to track application startup. It passes the token to the foot server, which then activates the new window with the token to complete the startup sequence.
* pgo: full-inner: add --override tweak.grapheme-shaping=noDaniel Eklöf2021-10-16-0/+1
| | | | | | | | | Grapheme shaping is now enabled by default in foot. However, when generating the profiling data in PGO builds, this results in skewed optimizations. The end result is worse benchmark results regardless of whether grapheme-shaping is enabled or not (when running the benchmarks).
* pgo: Use pixman renderer by default for PGO buildsStanislav Ochotnický2021-10-08-2/+2
| | | | | | | | | | Without pixman renderer override pgo builds in cage/sway require access to actual render devices. In Gentoo's case the builds are run as ordinary user and within a "sandbox" which prevents and logs access outside of predefined places during build. So let's default to pixman renderer for the PGO builds. Hopefully that won't matter for the pgo optimizations to do the right thing within foot.
* pgo: verify source dir really is a foot source dir before proceedingDaniel Eklöf2021-10-06-0/+5
|
* pgo: full-inner: do not set LC_CTYPE=en_US.UTF-8Daniel Eklöf2021-10-02-4/+4
| | | | | | | That locale may not exist. Instead, require the user/build script to explicitly set an UTF-8 locale. Document this in INSTALL.md, and in the bundled PKGBUILD.
* pgo.sh: set -x *after* verifying inputDaniel Eklöf2021-09-12-1/+3
|
* pgo: pgo.sh: remove both gcc and clang generated profile statsDaniel Eklöf2021-09-12-1/+8
|
* pgo: pgo.sh: add meson bld/src dir arguments lastDaniel Eklöf2021-09-12-1/+1
|
* pgo: pgo.sh: run ‘find’ in the build-directory, not cwdDaniel Eklöf2021-09-12-1/+1
|
* pgo: pgo.sh: allow a pre-existing build directoryDaniel Eklöf2021-09-12-4/+4
|
* pgo: pgo.sh: auto: don’t use full-headless-swayDaniel Eklöf2021-09-12-2/+2
| | | | | | | This requires Sway >= 1.6.2, which hasn’t been released yet. Thus, don’t auto-use it. Users can still request it explicitly.
* pgo: add set -x to all PGO scriptsDaniel Eklöf2021-09-12-8/+8
| | | | | Without this, debugging error reports from users is going to be _very_ difficult...
* pgo: full-headless-cage: remove references to SwayDaniel Eklöf2021-09-12-1/+0
|
* pgo: full-headless-cage: new headless variantDaniel Eklöf2021-09-12-2/+19
| | | | | | | Appears to work, but cage spams a lot of 00:00:08.026 [types/wlr_output.c:720] Basic output test failed for HEADLESS-1 00:00:08.036 [types/wlr_output.c:720] Basic output test failed for HEADLESS-1
* pgo: full-inner: quote path in trap’s “rm” commandDaniel Eklöf2021-09-12-1/+1
|
* pgo: full-inner: remove -o tweak.render-timer=logDaniel Eklöf2021-09-12-1/+0
|
* pgo: replace meson-pgo.sh with several script primitivesDaniel Eklöf2021-09-12-0/+202
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | All scripts are in the ‘pgo’ directory: * options: command line options for generate-alt-random-writes.py, sourced by other scripts. * pgo.sh: top-level script, generates a meson build directory, selects a PGO method, generates the profiling data, re-configures the meson build directory and does the final build. This script is intended to be used by end-users, and shows _how_ to integrate the script primitives. Build servers will most likely *not* want to use this script as-is. * partial.sh: generates alt-random-write data and runs “foot{,client} --version”, and then feeds the alt-random data to the PGO helper binary. Does not require a running Wayland session. Touches $blddir/pgo-ok on success. * full-inner.sh: runs “footclient --version”, and then a complex “foot” command that first generates alt-random-write data, and then “cat’s” it. Requires a running Wayland session, *but*, this script is usually not called directly (see below). Touches $blddir/pgo-ok onsucces.. * full-current-session.sh: runs full-inner.sh. That is, it runs foot in the currently running Wayland session. Note that this will pop up a foot window. * full-headless-sway.sh: generates a custom Sway configuration that exec’s foot-headless-sway-inner.sh (see below), and then executes a headless Sway. In other words, this script does a *full* PGO build, but *without* requiring a running Wayland session. Requires Sway >= 1.6.2. * full-headless-sway.sh: runs full-inner.sh + “swaymsg exit”. To do a custom PGO build, without using pgo.sh, you’d need to: CFLAGS=”$CFLAGS -O3” meson --buildtype=release -Db_lto=true meson configure -Db_pgo=generate ninja ninja test (only needed if tllist+fcft are built as subprojects) Run *one* of: - partial.sh - full-current-session.sh - full-headless-sway.sh meson configure -D b_pgo=use ninja
* pgo: add stub for search_selection_cancelled()Daniel Eklöf2021-07-23-0/+2
| | | | | | | Fixes PGO build failures caused by 251545203baa65364e491ddbe71ab3f8080bfe25 Closes #648
* pgo: fix PGO builds with clangDaniel Eklöf2021-07-22-0/+10
| | | | | | | | Add stubs for shm_chain_new(), shm_chain_free() and shm_unref(). This fixes ‘pgo’ linking failures in the ‘generate’ phase when doing a PGO build with clang. Closes #642
* extract: move ‘strip_trailing_empty’ parameter from extra_finish() to extract_begin()Daniel Eklöf2021-05-17-1/+1
|
* pgo: add wayl_win_set_urgent() stubDaniel Eklöf2021-05-14-0/+1
|
* ime: move preedit state from terminal struct to the seat structDaniel Eklöf2021-03-25-0/+1
| | | | | | | | | | | | | | | | | This ensures different seat’s don’t step on each others IME pre-edit state. It also removes most dependencies on having a valid term pointer for many IME operations. We’re still not all the way, since we support disabling IME with a private mode, which is per terminal, not seat. Thus, we still require the seat to have keyboard focus on one of our windows. Closes #324. But note that *rendering* of multiple seat’s IME pre-edit strings is still broken.
* fdm: add support for managing signalsDaniel Eklöf2021-02-11-1/+1
| | | | | | | | | | | | | | | | | | Add fdm_signal_add() and fdm_signal_del(). Signals added to the fdm will be monitored, and the provided callback called as “soon as possible” from the main context (i.e not from the signal handler context). Monitored signals are *blocked* by default. We use epoll_pwait() to unblock them while we’re polling. This allows us to do race-free signal detection. We use a single handler for all monitored signals; the handler simply updates the signal’s slot in a global array (sized to fit SIGRTMAX signals). When epoll_pwait() returns EINTR, we loop the global array. The callback associated with each signal that fired is called.
* urls: initial support for detecting URLs and rendering jump-labelsDaniel Eklöf2021-02-07-0/+1
| | | | The jump labels work, but is currently hardcoded to use xdg-open
* shm: unbreak build without memfd_createJan Beich2021-01-23-0/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | New FreeBSD versions have memfd_create but other BSDs don't. pgo/pgo.c:260:22: error: implicit declaration of function 'memfd_create' is invalid in C99 [-Werror,-Wimplicit-function-declaration] int mem_fd = memfd_create("foot-pgo-ptmx", MFD_CLOEXEC); ^ pgo/pgo.c:260:52: error: use of undeclared identifier 'MFD_CLOEXEC' int mem_fd = memfd_create("foot-pgo-ptmx", MFD_CLOEXEC); ^ shm.c:13:10: fatal error: 'linux/mman.h' file not found #include <linux/mman.h> ^~~~~~~~~~~~~~ shm.c:277:15: error: implicit declaration of function 'memfd_create' is invalid in C99 [-Werror,-Wimplicit-function-declaration] pool_fd = memfd_create("foot-wayland-shm-buffer-pool", MFD_CLOEXEC | MFD_ALLOW_SEALING); ^ shm.c:277:60: error: use of undeclared identifier 'MFD_CLOEXEC' pool_fd = memfd_create("foot-wayland-shm-buffer-pool", MFD_CLOEXEC | MFD_ALLOW_SEALING); ^ shm.c:277:74: error: use of undeclared identifier 'MFD_ALLOW_SEALING' pool_fd = memfd_create("foot-wayland-shm-buffer-pool", MFD_CLOEXEC | MFD_ALLOW_SEALING); ^ shm.c:339:15: error: use of undeclared identifier 'F_SEAL_GROW' F_SEAL_GROW | F_SEAL_SHRINK | /*F_SEAL_FUTURE_WRITE |*/ F_SEAL_SEAL) < 0) ^ shm.c:339:29: error: use of undeclared identifier 'F_SEAL_SHRINK' F_SEAL_GROW | F_SEAL_SHRINK | /*F_SEAL_FUTURE_WRITE |*/ F_SEAL_SEAL) < 0) ^ shm.c:339:71: error: use of undeclared identifier 'F_SEAL_SEAL' F_SEAL_GROW | F_SEAL_SHRINK | /*F_SEAL_FUTURE_WRITE |*/ F_SEAL_SEAL) < 0) ^ shm.c:338:24: error: use of undeclared identifier 'F_ADD_SEALS' if (fcntl(pool_fd, F_ADD_SEALS, ^
* pgo: stub implementation of reaper_del()Daniel Eklöf2021-01-12-0/+1
|
* terminal: shutdown (or --hold) when the client process terminatesDaniel Eklöf2020-12-26-0/+4
| | | | | | | | | Shutdown the terminal when the client process terminates, not when the ptmx file descriptor is closed. This fixes an issue where the terminal remains running after the client process has terminated, if it spawned child processes that inherited the ptmx file descriptor.
* notify: break out desktop notifications from osc.cDaniel Eklöf2020-12-10-0/+5
|
* ime: add functions to enable/disable IME, simplify code that enables IMEDaniel Eklöf2020-12-07-0/+3
| | | | | | | | | | | | | | | | | We may want to be able to enable/disable IME run-time, even though we have received an ‘enter’ IME event. This enables us to do that. Also add functions to enable/disable IME on a per-terminal instance basis. A terminal may have multiple seats focusing it, and enabling/disabling IME in a terminal instance enables/disables IME on all those seats. Finally, the code to enable IME is simplified; the *only* surface that can ever receive ‘enter’ IME events is the main grid. All other surfaces are sub-surfaces, without their own keyboard focus.
* sixel: change default max size to 10000x10000Daniel Eklöf2020-11-23-1/+7
| | | | | It used to be the size of the window. This caused images to be cropped when the application emitting them didn’t change the max size.
* pgo: feed VT data through fdm_ptmx(), not vt_from_slave()Daniel Eklöf2020-11-22-6/+61
| | | | | | | | | | | | | | | | | fdm_ptmx(), the FDM callback handler for ptmx data, is just as much in the hot path as vt_from_slave(). It is also slightly more complicated than a read() followed by a call to vt_from_slave(). As a result, some benchmarks performed significantly worse in a partial PGO build than in a full PGO build, since fdm_ptmx() wasn’t PGO:d. To be able to feed data through fdm_ptmx(), we need to set up the delayed rendering timer FDs, configure the timeout values, and provide a readable FD it can read the VT data from. The latter is done with a memory FD. This ensures *all* VT data is loaded into memory before we feed it to the parser.
* pgo: initialize (mouse) selectionDaniel Eklöf2020-11-22-0/+4
|
* pgo: initialize a dummy wayland backendDaniel Eklöf2020-11-22-0/+11
| | | | | | * Empty seat list * Empty monitor list * Add dummy terminal instance to terminal list
* remove pre-generated input corpus for PGODaniel Eklöf2020-11-22-1/+0
|
* pgo: updated stimuli (256-indexed colors)Daniel Eklöf2020-11-22-262144/+1
|
* pgo: pull in selection.c in pgolibDaniel Eklöf2020-11-22-63/+3
| | | | | Some of the selection functions are in the hot path, so this makes a huge difference!
* pgo: support for specifying multiple stimuli filesDaniel Eklöf2020-11-22-14/+56
|
* pgo: move pgo.c and pgo stimuli to pgo directoryDaniel Eklöf2020-11-22-0/+262377