diff options
author | Scott Murray <scott.murray@konsulko.com> | 2019-02-08 10:53:08 -0500 |
---|---|---|
committer | Stephane Desneux <stephane.desneux@iot.bzh> | 2019-04-04 18:02:11 +0200 |
commit | 7faccb97d69c7581e338f88ce3a2153cdd69fd16 (patch) | |
tree | 57dd664e04593af6eed43cb6ecffab438d93d860 /meta-agl-profile-core/recipes-multimedia/pulseaudio/pulseaudio-10.0/0003-volume-ramp-adding-volume-ramping-to-sink-input.patch | |
parent | e978a20f40916eac57a5e1af8f65b6ed9f719e50 (diff) |
Upgrade to thud
Changes include:
- Add LAYERSERIES_COMPAT definitions to layer.conf files
- Remove now unnecessary SECURITY_*FLAGS over-rides from distro
configuration
- Set intel-corei7-64 preferred kernel version to 4.19 to match
latest linux-intel kernel available in meta-intel
- Update qemuarm preferred kernel version to 4.18 to match latest
linux-yocto
- Update firmware package and devicetree file names for raspberrypi3
- Remove linux-firmware bbappend specific to raspberrypi, it seems no
longer required and breaks the cross SDK build
- Update linux-intel bbappend to 4.19, remove now unnecessary patch
- Remove now unnecessary lttng-modules backport
- Update linux-raspberrypi bbappend to 4.14 kernel
- Added kernel configuration fragment for raspberrypi to disable
Kprobes. This is required until linux-raspberrypi is updated to
greater than 4.14.104 to avoid a build failure in lttng-modules
related to a check for known breakage in the kernel CONFIG_OPTPROBES
code.
- Replace obsolete base_conditional usage with oe.utils.conditional
- Add gstreamer1.0-plugins-bad bbappend for raspberrypi3 to disable
faad PACKAGECONFIG to avoid commercial license issues
- Remove unused and unbuildable Vayu gstreamer recipes
- Update linux-ti-staging bbappend for new BSP kernel
- Regen dcan2_pinmux_enable.patch for linux-ti-staging to remove fuzz
warning, and remove upstreamed fix_dcan_addresses.patch
- Remove ipumm-fw from meta-agl-bsp/meta-ti, as newer version is
available in the upstream BSP
- Update meta-agl-bsp/meta-ti weston patch to apply against 5.0.0
- Update meta-agl-bsp/meta-ti wayland-ivi-extension patch to apply
against 2.2.0
- Add ti-sgx-ddk-km patch to add AGL toolchain configuration file
- Remove now unnecessary fdtoverlay recipe
- Update core.cfg and ivishell.cfg in weston-ini-conf recipe to handle
move of ivi-controller.so configuration in Weston 5.0.0
- Update connman-ncurses patch to remove fuzz warning
- Add installation of systemd over-ride file for run-postinsts.service
in run-postinsts bbappend to workaround race condition between
ldconfig.service and the /sbin/ldconfig invocations in the
post-install scripts run by run-postinsts.service. The observed
failure was cynara's post-install script failing and its database
not being created.
- Remove now unnecessary valgrind backport
- Add patches to fix most driver compilation against newer kernels
- Update libmicrohttpd bbappend
- Remove libssp-dev from agl-image-graphical-qt5-crosssdk and
agl-demo-platform-html5-crosssdk, upstream have removed it from
non-mingw32 platform SDKs
- Update wayland-ivi-extension recipe to build 2.2.0, and update
local patches
- Update weston patches for 5.0.0. Patches:
0016-ivi-shell_add_screen_remove_layer_api.patch
0017-ivi-shell-register-ivi_layout_interface.patch
have been removed as they have been applied upstream and are no longer
necessary. Patches:
0018-compositor-add-output-type-to-weston_output.patch
0019-compositor-drm-introduce-drm_get_dmafd_from_view.patch
(both related to Waltham) have been disabled for now as they need
significant rework.
- Remove weston-conf RRECOMMENDS in weston bbappend to avoid conflict
with weston-ini-conf
- Add OECMAKE_GENERATOR = "Unix Makefiles" to aglwgt.bbclass to work
around CMake+ninja issue in cmake-apps-module
- Update dbus cynara patches for 1.12.10
- Add do_install_append in cynara recipe to remove /var/cynara from
cynara package so the directory creation and labelling in the
post-install scriptlet will function as intended
- Remove now unnecessary e2fsprogs backport
- Remove now unnecessary libcap-ng backport
- Update pulseaudio patches to remove fuzz warnings
- Update neardal patch to remove fuzz warning
- Update freetype patch to remove fuzz warning
- Rename opencv bbappend to 3.% to handle 3.x backports in upstream
- Updated qtwayland patch to remove fuzz warning
Changes from Stephane Desneux <stephane.desneux@iot.bzh>:
- Remove wayland-ivi-extension PREFERRED_VERSION
- Remove now unnecessary nativesdk-cmake patch
- Remove now unnecessary ptest-runner patches
- Remove now unnecessary harfbuzz patches
- Disable waltham-transmitter as it does not build against weston 5.0.0
- Update af-main, cynara, and security-manager to use pkg_postinst_ontarget
- Bump connman-ncurses revision to avoid deprecated ncurses functions
- Update libva package usage with new intel-vaapi-driver name
- Add patches to security-manager to fix compilation with gcc8
- Updated systemd bbappend
Changes from Jan-Simon Möller <jsmoeller@linuxfoundation.org>:
- Remove meta-agl-bsp/ROCKO.FIXMEs
- Remove linux-yocto_4.12.bbappend and now unnecessary associated
patch
- Remove now unneeded kern-tools-native patch
- Bump gstreamer PREFERRED_VERSIONs to 1.14.x
- Remove latencytop from packagegroup-agl-core-devel, it has been
dropped by upstream
- Remove now unnecessary rpm patches
- Update pulseaudio bbappend to 12.2
- Update opencv bbappend to 3.4
- Update freetype bbappend to 2.9.1
- Update dbus bbappend to 1.12.10
- Update weston bbappend to 5.0.0
- Update cynara patches to remove fuzz warnings
- Add patch to cynara to fix compilation with gcc8
- Add xmlsec1 bbappend to clear EXTRA_OECONF to fix compilation on
sumo or newer
Changes from Ronan Le Martet <ronan.lemartet@iot.bzh>:
- Update meta-rcar-gen3-adas layer gstreamer1.0-plugin-vspfilter
bbappend to version 1.0.1
Known issues (marked with FIXME):
- CMake+ninja issue in cmake-apps-module has been worked around with
OECMAKE_GENERATOR
- waltham-transmitter and the patches to weston related to it have been
disabled
- Currently unclear if patch to libcap-native is actually required or
not
Bug-AGL: SPEC-1837
Change-Id: I7b8b9ef667aec2d229952eace6663dfc761654d0
Signed-off-by: Scott Murray <scott.murray@konsulko.com>
Diffstat (limited to 'meta-agl-profile-core/recipes-multimedia/pulseaudio/pulseaudio-10.0/0003-volume-ramp-adding-volume-ramping-to-sink-input.patch')
-rw-r--r-- | meta-agl-profile-core/recipes-multimedia/pulseaudio/pulseaudio-10.0/0003-volume-ramp-adding-volume-ramping-to-sink-input.patch | 189 |
1 files changed, 0 insertions, 189 deletions
diff --git a/meta-agl-profile-core/recipes-multimedia/pulseaudio/pulseaudio-10.0/0003-volume-ramp-adding-volume-ramping-to-sink-input.patch b/meta-agl-profile-core/recipes-multimedia/pulseaudio/pulseaudio-10.0/0003-volume-ramp-adding-volume-ramping-to-sink-input.patch deleted file mode 100644 index eb485ca7c..000000000 --- a/meta-agl-profile-core/recipes-multimedia/pulseaudio/pulseaudio-10.0/0003-volume-ramp-adding-volume-ramping-to-sink-input.patch +++ /dev/null @@ -1,189 +0,0 @@ -From e4469df7c69316b49cad93dd288badc98fa1cad5 Mon Sep 17 00:00:00 2001 -From: Sangchul Lee <sangchul1011@gmail.com> -Date: Sat, 27 Aug 2016 21:33:17 +0900 -Subject: [PATCH 3/6] volume ramp: adding volume ramping to sink-input - -The original patch is - - https://review.tizen.org/git/?p=platform/upstream/pulseaudio.git;a=commit;h=98042248fd67ce0ab3807c5c472c0d5d8b0f99d3 - - by Jaska Uimonen <jaska.uimonen <at> helsinki.fi> - -Signed-off-by: Sangchul Lee <sc11.lee@samsung.com> ---- - src/pulsecore/sink-input.c | 61 ++++++++++++++++++++++++++++++++++++++++++++++ - src/pulsecore/sink-input.h | 11 ++++++++- - 2 files changed, 71 insertions(+), 1 deletion(-) - -diff --git a/src/pulsecore/sink-input.c b/src/pulsecore/sink-input.c -index 8ec63b5..cc8953f 100644 ---- a/src/pulsecore/sink-input.c -+++ b/src/pulsecore/sink-input.c -@@ -526,6 +526,11 @@ int pa_sink_input_new( - reset_callbacks(i); - i->userdata = NULL; - -+ if (data->flags & PA_SINK_INPUT_START_RAMP_MUTED) -+ pa_cvolume_ramp_int_init(&i->ramp, PA_VOLUME_MUTED, data->sample_spec.channels); -+ else -+ pa_cvolume_ramp_int_init(&i->ramp, PA_VOLUME_NORM, data->sample_spec.channels); -+ - i->thread_info.state = i->state; - i->thread_info.attached = false; - pa_atomic_store(&i->thread_info.drained, 1); -@@ -542,6 +547,8 @@ int pa_sink_input_new( - i->thread_info.playing_for = 0; - i->thread_info.direct_outputs = pa_hashmap_new(pa_idxset_trivial_hash_func, pa_idxset_trivial_compare_func); - -+ i->thread_info.ramp = i->ramp; -+ - pa_assert_se(pa_idxset_put(core->sink_inputs, i, &i->index) == 0); - pa_assert_se(pa_idxset_put(i->sink->inputs, pa_sink_input_ref(i), NULL) == 0); - -@@ -923,6 +930,8 @@ void pa_sink_input_peek(pa_sink_input *i, size_t slength /* in sink bytes */, pa - while (tchunk.length > 0) { - pa_memchunk wchunk; - bool nvfs = need_volume_factor_sink; -+ pa_cvolume target; -+ pa_bool_t tmp; - - wchunk = tchunk; - pa_memblock_ref(wchunk.memblock); -@@ -959,6 +968,16 @@ void pa_sink_input_peek(pa_sink_input *i, size_t slength /* in sink bytes */, pa - pa_volume_memchunk(&wchunk, &i->sink->sample_spec, &i->volume_factor_sink); - } - -+ /* check for possible volume ramp */ -+ if (pa_cvolume_ramp_active(&i->thread_info.ramp)) { -+ pa_memchunk_make_writable(&wchunk, 0); -+ pa_volume_ramp_memchunk(&wchunk, &i->sink->sample_spec, &(i->thread_info.ramp)); -+ } else if ((tmp = pa_cvolume_ramp_target_active(&(i->thread_info.ramp)))) { -+ pa_memchunk_make_writable(&wchunk, 0); -+ pa_cvolume_ramp_get_targets(&i->thread_info.ramp, &target); -+ pa_volume_memchunk(&wchunk, &i->sink->sample_spec, &target); -+ } -+ - pa_memblockq_push_align(i->thread_info.render_memblockq, &wchunk); - } else { - pa_memchunk rchunk; -@@ -975,6 +994,16 @@ void pa_sink_input_peek(pa_sink_input *i, size_t slength /* in sink bytes */, pa - pa_volume_memchunk(&rchunk, &i->sink->sample_spec, &i->volume_factor_sink); - } - -+ /* check for possible volume ramp */ -+ if (pa_cvolume_ramp_active(&(i->thread_info.ramp))) { -+ pa_memchunk_make_writable(&rchunk, 0); -+ pa_volume_ramp_memchunk(&rchunk, &i->sink->sample_spec, &(i->thread_info.ramp)); -+ } else if (pa_cvolume_ramp_target_active(&(i->thread_info.ramp))) { -+ pa_memchunk_make_writable(&rchunk, 0); -+ pa_cvolume_ramp_get_targets(&i->thread_info.ramp, &target); -+ pa_volume_memchunk(&rchunk, &i->sink->sample_spec, &target); -+ } -+ - pa_memblockq_push_align(i->thread_info.render_memblockq, &rchunk); - pa_memblock_unref(rchunk.memblock); - } -@@ -1339,6 +1368,31 @@ int pa_sink_input_remove_volume_factor(pa_sink_input *i, const char *key) { - return 0; - } - -+/* Called from main thread */ -+void pa_sink_input_set_volume_ramp( -+ pa_sink_input *i, -+ const pa_cvolume_ramp *ramp, -+ pa_bool_t send_msg, -+ pa_bool_t save) { -+ -+ pa_sink_input_assert_ref(i); -+ pa_assert_ctl_context(); -+ pa_assert(PA_SINK_INPUT_IS_LINKED(i->state)); -+ pa_assert(ramp); -+ -+ pa_cvolume_ramp_convert(ramp, &i->ramp, i->sample_spec.rate); -+ -+ pa_log_debug("setting volume ramp with target vol:%d and length:%ld", -+ i->ramp.ramps[0].target, -+ i->ramp.ramps[0].length); -+ -+ -+ /* This tells the sink that volume ramp changed */ -+ if (send_msg) -+ pa_assert_se(pa_asyncmsgq_send(i->sink->asyncmsgq, PA_MSGOBJECT(i), PA_SINK_INPUT_MESSAGE_SET_VOLUME_RAMP, -+ NULL, 0, NULL) == 0); -+} -+ - /* Called from main context */ - static void set_real_ratio(pa_sink_input *i, const pa_cvolume *v) { - pa_sink_input_assert_ref(i); -@@ -1932,6 +1986,13 @@ int pa_sink_input_process_msg(pa_msgobject *o, int code, void *userdata, int64_t - } - return 0; - -+ case PA_SINK_INPUT_MESSAGE_SET_VOLUME_RAMP: -+ /* we have ongoing ramp where we take current start values */ -+ pa_cvolume_ramp_start_from(&i->thread_info.ramp, &i->ramp); -+ i->thread_info.ramp = i->ramp; -+ pa_sink_input_request_rewind(i, 0, true, false, false); -+ return 0; -+ - case PA_SINK_INPUT_MESSAGE_SET_SOFT_MUTE: - if (i->thread_info.muted != i->muted) { - i->thread_info.muted = i->muted; -diff --git a/src/pulsecore/sink-input.h b/src/pulsecore/sink-input.h -index 86deab2..6e1b211 100644 ---- a/src/pulsecore/sink-input.h -+++ b/src/pulsecore/sink-input.h -@@ -32,6 +32,7 @@ - #include <pulsecore/client.h> - #include <pulsecore/sink.h> - #include <pulsecore/core.h> -+#include <pulsecore/mix.h> - - typedef enum pa_sink_input_state { - PA_SINK_INPUT_INIT, /*< The stream is not active yet, because pa_sink_input_put() has not been called yet */ -@@ -58,7 +59,8 @@ typedef enum pa_sink_input_flags { - PA_SINK_INPUT_DONT_INHIBIT_AUTO_SUSPEND = 256, - PA_SINK_INPUT_NO_CREATE_ON_SUSPEND = 512, - PA_SINK_INPUT_KILL_ON_SUSPEND = 1024, -- PA_SINK_INPUT_PASSTHROUGH = 2048 -+ PA_SINK_INPUT_PASSTHROUGH = 2048, -+ PA_SINK_INPUT_START_RAMP_MUTED = 4096, - } pa_sink_input_flags_t; - - struct pa_sink_input { -@@ -121,6 +123,9 @@ struct pa_sink_input { - * this.*/ - bool save_sink:1, save_volume:1, save_muted:1; - -+ /* for volume ramps */ -+ pa_cvolume_ramp_int ramp; -+ - pa_resample_method_t requested_resample_method, actual_resample_method; - - /* Returns the chunk of audio data and drops it from the -@@ -249,6 +254,8 @@ struct pa_sink_input { - pa_usec_t requested_sink_latency; - - pa_hashmap *direct_outputs; -+ -+ pa_cvolume_ramp_int ramp; - } thread_info; - - void *userdata; -@@ -265,6 +272,7 @@ enum { - PA_SINK_INPUT_MESSAGE_SET_STATE, - PA_SINK_INPUT_MESSAGE_SET_REQUESTED_LATENCY, - PA_SINK_INPUT_MESSAGE_GET_REQUESTED_LATENCY, -+ PA_SINK_INPUT_MESSAGE_SET_VOLUME_RAMP, - PA_SINK_INPUT_MESSAGE_MAX - }; - -@@ -370,6 +378,7 @@ void pa_sink_input_set_volume(pa_sink_input *i, const pa_cvolume *volume, bool s - void pa_sink_input_add_volume_factor(pa_sink_input *i, const char *key, const pa_cvolume *volume_factor); - int pa_sink_input_remove_volume_factor(pa_sink_input *i, const char *key); - pa_cvolume *pa_sink_input_get_volume(pa_sink_input *i, pa_cvolume *volume, bool absolute); -+void pa_sink_input_set_volume_ramp(pa_sink_input *i, const pa_cvolume_ramp *ramp, pa_bool_t send_msg, pa_bool_t save); - - void pa_sink_input_set_mute(pa_sink_input *i, bool mute, bool save); - --- -1.9.1 - |