blob: 881ea74c624f5d01e0eb8d0bc3dff81f391868ac (
plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
|
# template start; name=nosystemd-git; version=1.0;
# template start; name=base; version=1.0;
# Maintainer: Xenhat Hex (aur@xenh.at)
# Contributor: Justin Jagieniak <justin@jagieniak.net>
# Contributor: Rye Mutt
#
# shellcheck disable=2034,3030,2154
pkgdesc="A Second Life client with focus on performance and code correctness - Git build."
arch=('x86_64')
license=('LGPL')
options=('!buildflags' '!lto' '!strip')
url="https://www.alchemyviewer.org"
install=alchemy.install
# template end;
# template start; name=deps; version=1.0;
depends=('glu' 'libgl' 'libiconv' 'libidn' 'libjpeg-turbo' 'libpng' 'libxml2' 'libxss' 'mesa' 'nss' 'openal' 'sdl2' 'vlc' 'zlib')
makedepends=('gcc' 'python-pip' 'git')
optdepends=(
'alsa-lib: ALSA support'
'freealut: OpenAL support'
'gamemode: Gamemode support'
'lib32-libidn11: SLVoice support'
'lib32-libsndfile: SLVoice support'
'lib32-util-linux: SLVoice support'
'lib32-gstreamer0.10: SLVoice support'
'libpulse: PulseAudio support'
'mesa-libgl: Intel, Radeon, Nouveau support'
'nvidia-libgl: NVIDIA support'
'nvidia-utils: NVIDIA support'
'wine: More up-to-date, less buggy SLVoice support'
'xdg-desktop-portal: File picker portal')
# template end;
pkgname=alchemy-next-viewer-nosystemd-git
pkgver=7.1.3.55266.2a64c7b773
pkgrel=1
provides=('alchemy-viewer')
# template start; name=source; version=1.0;
source=("${pkgname}"::'git+https://git.alchemyviewer.org/alchemy/viewer.git#branch='"${AL_BRANCH_OVERRIDE:-main}")
sha256sums=('SKIP')
# template end;
# Not included as a dependency as it is impossible to update the package otherwise
#depends+=('libelogind')
conflicts+=("alchemy-next-viewer-systemd-git")
# template start; name=pkgver; version=1.0;
pkgver() {
cd "${pkgname}" || exit 1
(
set -o pipefail
vwr_version=$(cat indra/newview/VIEWER_VERSION.txt)
rev=$(git rev-list --count HEAD)
short=$(git rev-parse --short HEAD)
printf "%s.%s.%s" "${vwr_version}" "${rev}" "${short}"
)
}
# template end;
# template start; name=prepare; version=1.0;
prepare() {
cd ${srcdir} || exit
echo "Creating virtual environment in $PWD"
virtualenv ".venv" -p python3
source "${srcdir}/.venv/bin/activate"
echo "Installing build toolchain..."
pip3 install --upgrade cmake llbase llsd certifi autobuild ninja
}
# template end;
# template start; name=build; version=1.0;
build() {
cd "${pkgname}" || exit 1
source "${srcdir}/.venv/bin/activate"
echo ""
echo "Configuring the build"
_logfile="build.${CARCH}.$(date +%s).log"
build_jobs=$(nproc)
AL_CMAKE_CONFIG=(
-DLL_TESTS:BOOL=${ENABLE_TESTS:-ON}
-DDISABLE_FATAL_WARNINGS=ON
-DUSE_LTO:BOOL=OFF
-DVIEWER_CHANNEL="Alchemy Test"
)
if [[ " ${BUILDENV[*]} " =~ ' ccache ' ]] && command -v ccache >/dev/null 2>&1; then
AL_CMAKE_CONFIG+=("-DCMAKE_CXX_COMPILER_LAUNCHER=$(which ccache)")
echo "ccache was found and will be used"
fi
if [[ -z "$NO_CLANG" ]] && command -v clang++ >/dev/null 2>&1; then
AL_CMAKE_CONFIG+=("-DCMAKE_C_COMPILER=$(which clang)")
AL_CMAKE_CONFIG+=("-DCMAKE_CXX_COMPILER=$(which clang++)")
echo "clang was found and will be used instead of gcc"
fi
if [[ -z "$NO_SMART_JOB_COUNT" ]]; then
if [[ ${build_jobs} -gt 1 ]]; then
# The viewer requires an average of 2GB of memory per core to link
# Note: Behaviour change compared to the previous versions:
# This script will no longer try to allocate build memory into swap
# This is bad practice, and swap should be reserved to evict process
# memory from physical ram to make place for the current workset.
# This script will now try to check if swap is present and sufficent
# for the current used memory to be stored in swap before allocating,
# and will fallback to conservative allocation if swap is not available
mempercorekb=$((1048576))
requiredmemorykb=$(($(nproc) * mempercorekb))
free_output="$(free --kilo --total | tail -n+2 | tr -s ' ')"
physical_output=$(grep "Mem:" <<<"$free_output")
totalmemorykbphysical=$(cut -d ' ' -f 2 <<<"$physical_output")
usedmemorykbphysical=$(cut -d ' ' -f 3 <<<"$physical_output")
# Don't factor in the caches, these will be flushed as needed
#freememorykbphysical=$(cut -d ' ' -f 4 <<<"$physical_output")
availablememorykbphysical=$(cut -d ' ' -f 7 <<<"$free_output")
total_output=$(grep "Total:" <<<"$free_output")
totalmemorykbcombined=$(cut -d ' ' -f 2 <<<"$total_output")
usedmemorytotal=$(cut -d ' ' -f 2 <<<"$total_output")
freememorytotal=$(cut -d ' ' -f 4 <<<"$total_output")
swap_output=$(grep Swap: <<<"$free_output")
# Determine available swap space
availableswapkb=0
if [[ -n "$swap_output" ]]; then
availableswapkb=$(cut -d ' ' -f 4 <<<"$swap_output")
fi
echo "Required memory at $(nproc) jobs: $((requiredmemorykb/1024/1024))GB"
echo "Available memory (counting swap): $((totalmemorykbcombined/1024/1024))GB"
echo "Total RAM: $((totalmemorykbphysical/1024/1024))GB"
if [[ ${requiredmemorykb} -gt ${totalmemorykbphysical} ]]; then
echo "Not enough physical memory to use all cores"
if [[ ${usedmemorykbphysical} -lt ${availableswapkb} ]]; then
# There is enough swap to fit all the used memory. Use all physical ram as swap will do its job
echo "Using swap memory to store current processes memory"
# We do not want to compile in swap, so adjust accordingly
jobs=$(((totalmemorykbphysical) / mempercorekb))
else
# TODO: Verify this logic on low-ram systems
# Not enough swap to hold ram contents, calculate manually
jobs=1
echo "${jobs} job would consume $(((jobs * mempercorekb) / 1024 / 1024))GB"
while [[ $((jobs * mempercorekb)) -le ${availablememorykbphysical} ]]; do
((jobs++))
echo "${jobs} jobs would consume $(((jobs * mempercorekb) / 1024 / 1024))GB"
done
# Back off one job count. Not sure why I have to do this but
fi
build_jobs=${jobs}
fi
fi
echo "Adjusted job count: ${build_jobs}"
fi
export AUTOBUILD_CPU_COUNT=$build_jobs
# And now we configure and build the viewer with our adjusted configuration
autobuild configure -A 64 -c ReleaseOS -- "${AL_CMAKE_CONFIG[@]}" > >(tee -a "$_logfile") 2> >(tee -a "$_logfile" >&2)
echo "Building with ${AUTOBUILD_CPU_COUNT} jobs"
autobuild build -A 64 -c ReleaseOS --no-configure > >(tee -a "$_logfile") 2> >(tee -a "$_logfile" >&2)
}
# template end;
# template start; name=package; version=1.0;
package() {
mkdir -p "${pkgdir}/opt"
mkdir -p "${pkgdir}/usr/local/share/applications"
# Patch shortcut maker
sed -i 's;Name=Alchemy;Name=Alchemy (git build);' "${pkgname}/build-linux-64/newview/packaged/etc/refresh_desktop_app_entry.sh"
sed -i 's;alchemy-viewer\.desktop;'"${pkgname}\.desktop"';' "${pkgname}/build-linux-64/newview/packaged/etc/refresh_desktop_app_entry.sh"
mv "${pkgname}/build-linux-64/newview/packaged" "${pkgdir}/opt/${pkgname}"
}
# template end;
# template end;
|