linaro-dev

linaro-dev@lists.linaro.org

4332 discussions

Re: cpuset configuration and measurements

by Amit Kucheria

Good start, but let's move this discussion public. On Tue, Aug 23, 2011 at 3:13 PM, Daniel Lezcano <daniel.lezcano(a)linaro.org> wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > > Hi all, > > I did the following cpuset configuration: > > if [ ! -d /sys/fs/cgroup ]; then > echo "cgroup not supported by the kernel" > else > mount -t cgroup cgroup /sys/fs/cgroup > > # needed to have the cgroup values inherited from > # the parent cgroup > echo 1 > /sys/fs/cgroup/cgroup.clone_children > > # create an empty cgroup 'cpu0' > mkdir /sys/fs/cgroup/cpu0 > > # create an empty cgroup 'cpu1' > mkdir /sys/fs/cgroup/cpu1 > > # assign cpu number '0' for cgroup cpu0 > echo 0 > /sys/fs/cgroup/cpu0/cpuset.cpus > > # move outself to this new cgroup > echo $$ > /sys/fs/cgroup/cpu0 > > # move all the tasks to the cpu0 cgroup > # kernel threads won't be moved, don't worry about the > # the errors. > for i in $(cat /sys/fs/cgroup/tasks); do > echo $i > /sys/fs/cgroup/cpu0/tasks || echo "failed to > add pid $i/$(cat /proc/$i/comm)" > done > > # assign a cpu for cgroup cpu1 > echo 1 > /sys/fs/cgroup/cpu1/cpuset.cpus > > # make cpu1 exclusive so processes belonging > # to the other cgroups won't be able to use it. > echo 1 > /sys/fs/cgroup/cpu1/cpuset.cpu_exclusive > fi > > > At this point, most of the processes have been moved to the 'cpu0' > cgroup and will use the cpu0 only. That does not prevent the cpu1 to be > used by the kernel threads. Any application launched will run on cpu0 only. > > After running 10 infinite loops, the cpu accounting gives: > > cat /sys/fs/cgroup/cpu0/cpuacct.usage_percpu > 418943579549 0 > > cat /sys/fs/cgroup/cpu1/cpuacct.usage_percpu > 0 0 > > But the system is: > > cat /sys/fs/cgroup/cpuacct.usage_percpu > 456741213041 30734572583 > > My guess is that reflects the kernel booted with the two cpus before we > created the cgroup. I am investigation the 'isolcpus' kernel parameter. > > I think it is a good starting point to do some consumption measurements. > > Mike, Vincent, what do you think ? Do you need more informations or > anything else ? > > Thanks > -- Daniel

13 years, 8 months

[CALL FOR TESTING] Linaro 11.08 Release Candidate (without QA Tracker!!!)

by Fathi Boudra

Hi, In preparation for the release of Linaro 11.08 images on 2011-08-25, we have collected candidate submissions for our Linaro Evaluation Build (LEB), Developers and Community images. All primary developement boards supported by Linaro Landing Teams have an Android LEB candidate image submitted and an Ubuntu LEB hwpack submitted. Note: Due to the early stage of the enablement, some images might not work. We don't want to promise too much but there are exciting plans that will finally make these calls for candidates a real joy by prepopulating results with data we get from our LAVA testing. For now as an intermediate step we moved our testing forms to something that allows us to request slighly more details while making filling out the questionaire far more comfortable than our previous QA Tracker solution. Based on your feedback, the Release Team will review the quality of those candidate builds and award builds that deliver a decent user experience with the "official LEB" badge. Builds not quite ready yet will continue to be published as Developers and Community builds for the final release until enough progress has been made. Note: Your feedback to this call for testing will provide the main source of data used by the Release Team to determine the user experience and hence is essential to this process. Linaro Evaluation Build (LEB) ============================= Please help our initiative by testing this month Linaro Evaluation Build (LEB) candidates for Android and Ubuntu: * i.MX53 Quick Start: Android: http://bit.ly/q4Lriu Ubuntu: http://bit.ly/nlRVIY * Origen: Android: http://bit.ly/pOV6WB Ubuntu: http://bit.ly/nIGG44 * PandaBoard: Android: http://bit.ly/pFnQQh Ubuntu: http://bit.ly/prrver * Snowball: Android: http://bit.ly/nBfAVc Ubuntu: (no submission) Developers and Community Build ============================== On top of our officially supported Linaro Evaluation Build candidate images from above, the Linaro Platform Team is proud to also provide additional images and additional board support bits that were prepared by Linaro developers and community for a more specific target audience. Note that the Developers and Community bits are provided on a best-effort basis and in the hope that they can be useful. We release last reported known to be working images. If you are interested in running Nano, ALIP or Developer image _or_ want to try our LEB on one of the board support packs maintained by our community, please help and test by following the download and installation instructions included in the questionaire linked from below your board: * BeagleBoard-xM: http://bit.ly/qzZz9U * EFIKA MX: http://bit.ly/ofl1M4 * IGEPv2: http://bit.ly/nE8B57 * i.MX51 Babbage: http://bit.ly/qIapvB * i.MX53 Quick Start: http://bit.ly/qDo8ef * Origen: http://bit.ly/qrK1nu * Overo: http://bit.ly/q1qXUL * PandaBoard: http://bit.ly/o3SjZJ * Versatile Express: http://bit.ly/nHVkj6 Resources ========= Note: Wonder what to download? Instructions including download URLs are documented as part of the feedback form reachable through the URLs above. Ubuntu: * Nano (think small): http://snapshots.linaro.org/11.05-daily/linaro-nano/20110823/1/images/tar/ * ALIP (Xfce desktop): http://snapshots.linaro.org/11.05-daily/linaro-alip/20110823/1/images/tar/ * Developer (text user interface + developer tools): http://snapshots.linaro.org/11.05-daily/linaro-developer/20110823/1/images/… * Ubuntu: http://snapshots.linaro.org/11.05-daily/linaro-ubuntu-desktop/20110823/3/im… * Hardware packs: http://snapshots.linaro.org/11.05-daily/linaro-hwpacks/${BOARD}/20110823/0/… * Installation Instructions: https://wiki.linaro.org/Platform/DevPlatform/Ubuntu/ImageInstallation Android: * Android LEB for i.MX53 Quick Start: https://android-build.linaro.org/builds/~linaro-android/leb-imx53/#5 * Android LEB for Origen: https://android-build.linaro.org/builds/~linaro-android/leb-origen-11.08-re… * Android LEB for PandaBoard: https://android-build.linaro.org/builds/~linaro-android/leb-panda-11.08-rel… * Android LEB for Snowball: https://android-build.linaro.org/builds/~linaro-android/leb-snowball-11.08-… * Installation Instructions: https://wiki.linaro.org/Platform/Android/ImageInstallation Known Issues ============ * ARM SMP scheduler performance bug (Ubuntu/PandaBoard) https://launchpad.net/bugs/709245 * Snowball will not boot with current version of startupfiles (Ubuntu/Snowball) https://launchpad.net/bugs/816530 * Pulseaudio consumes 100% of the cpu when trying to play a sound (Ubuntu/PandaBoard) https://launchpad.net/bugs/816638 * eth0 doesn't get a valid MAC address on startup (Ubuntu/i.MX51, i.MX53 LoCo) https://launchpad.net/bugs/817315 * fails to mount system and user partition interminttently (Android/Snowball) https://launchpad.net/bugs/823313 * linaro-media-create fails for snowball_emmc (Snowball) https://launchpad.net/bugs/829220 You can find the complete list of known issues for the 11.08 release at: https://launchpad.net/linaro-android/+milestone/11.08 https://launchpad.net/linaro-ubuntu/+milestone/11.08 When filling new bugs, please check if it's not yet reported. You can use: https://bugs.launchpad.net/linaro-android/+bugs?orderby=-datecreated https://bugs.launchpad.net/linaro-ubuntu/+bugs?orderby=-datecreated -- Fathi Boudra Linaro Release Manager | Platform Project Manager Linaro.org | Open source software for ARM SoCs

13 years, 8 months

[ACTIVITY] Multimedia WG weekly status report - wk 33.2011 (20110815-20110819)

by Ilias Biris

Meeting minutes :https://wiki.linaro.org/WorkingGroups/Middleware/Multimedia/Notes/2011-08-16 Status link: https://wiki.linaro.org/WorkingGroups/Middleware/Multimedia/WeeklyReport Summary of the highlights: For 11.08: libjpeg-turbo release 1.1.90: https://launchpad.net/libjpeg-turbo/1.2/1.1.90 Multimedia codec test content available from http://samplemedia.linaro.org/ NEON forum material available from https://wiki.linaro.org/WorkingGroups/Middleware/Multimedia/Specs/1111/NEON… OpenMAX highlighted survey: https://wiki.linaro.org/WorkingGroups/Middleware/Multimedia/Specs/1111/Open… Memory management related presentations : https://wiki.linaro.org/OfficeofCTO/MemoryManagement?action=AttachFile&do=g…, and https://wiki.linaro.org/OfficeofCTO/MemoryManagement?action=AttachFile&do=g… Team will be planning 11.09 during the next few days. BR, -- Ilias Biris ilias.biris(a)linaro.org Project Manager, Linaro M: +358504839608, IRC: ibiris Skype: ilias_biris Linaro.org│ Open source software for ARM SoCs

13 years, 8 months

Generic Linux cross toolchain for tests

by Marcin Juszkiewicz

Hi Some time ago we agreed that not everyone here uses Ubuntu distribution and decided to provide so called 'generic linux' cross toolchain. Recently I managed to get it done and now need brave testers to tell is it working or not. Get it here: http://people.linaro.org/~hrw/generic-linux/ (64bit only) Needed files are toolchain-11.07.tar.xz and init.sh script. Unpack tarball from / so /opt/linaro/11.07/ will be populated and put init.sh anywhere you want (it will be integrated into tarball later). How to use: $ source init.sh this will add cross toolchain into PATH and also set LD_LIBRARY_PATH to two directories: - one with binutils libraries - second with all extra libraries which may be needed Feel free to experiment with second dir by removing files from there and checking are system provided libs are fine too. So far I checked this toolchain under few distributions: - Ubuntu 10.04 'lucid' LTS - Ubuntu 11.04 'natty' - Fedora 14 - OpenSUSE 11.4 - CentOS 5.6 It failed only under CentOS (which was expected due to it's age). How did I checked? So far compilation of 'gpm' and 'zlib' were tested.

13 years, 8 months

Android gcc 4.6 1107 optimization benchmark

by Chao Yang

HI Zach, The BP ( https://blueprints.launchpad.net/linaro-android/+spec/linaro-android-use-gc…) and Bug <goog_24859600>#822113<https://bugs.launchpad.net/linaro-android/+bug/822113>aim at improving android performance. I think we also need to balance the size and the performance improvement. I used the gcc benchmark tool to benchmark the performance with different configuration: -O3 for arm files only, -O3 for both arm files and thumb files and -fstrict-aliasing. The results can be found at https://wiki.linaro.org/ChaoYang/Sandbox/gccoptimization. Please note, the results are based on linaro_android_2.3.4 for panda and toolchain-4.6-1107. I will benchmark linaro_android_2.3.5 and toolchain-4.6-1108 if necessary when they are stable enough. The image size increases significantly when -O3 is enabled for thumb files, however it does not look like performance has been improved as much as expected. Could you please let me know if you think it is worth building thumb files with -O3 regardless of size? Thanks. Regards -- Chao Yang Android Platform Team Linaro.org │ Open source software for ARM SoCs Follow Linaro: http://www.facebook.com/pages/Linaro/155974581091106 http://twitter.com/#!/linaroorg http://www.linaro.org/linaro-blog/

13 years, 8 months

Multistrap and required packages

by Tom Gall

Hi Wookey (and others that might be interested) With Mutlistrap and qemu we've gotten live-build to be able to cross build most of our images. The next step however in nano is to pull out a number of the packages and in some cases replace them with busybox components to get an even smaller rootfs. An important part of this is to not include recommends AND not take into account required packages. It would appear that multistrap is unable to ignore the required packages. There are a number of them. You will also note in the log below a number of the busybox replacements explicitly conflict with their non-busybox brethren. In the Multistrap package list, the non-busybox versions are not listed yet Multistrap is pulling them in (as well as others I had removed). Grepping through the Multistrap perl source I don't see a way forward but I know you're the expert here. I would appreciate your thoughts. Here's the log : multistrap 2.1.6ubuntu3 using config/multistrap.config Using foreign architecture: armel multistrap building armel multistrap on 'amd64' INF: Setting ./lib64 -> ./lib symbolic link. Getting package lists: apt-get -o Apt::Architecture=armel -o Dir::Etc::TrustedParts=/etc/apt/trusted.gpg.d -o Dir::Etc::Trusted=/etc/apt/trusted.gpg -o Apt::Get::Download-Only=true -o Apt::Install-Recommends=false -o Dir=/bld/nano-nano-cross/chroot/ -o Dir::Etc=/bld/nano-nano-cross/chroot/etc/apt/ -o Dir::State=/bld/nano-nano-cross/chroot/var/lib/apt/ -o Dir::State::Status=/bld/nano-nano-cross/chroot/var/lib/dpkg/status -o Dir::Cache=/bld/nano-nano-cross/chroot/var/cache/apt/ update [ERROR - 22.Aug.11 19:54:23]: Not on Blue network, aborting... Get:1 http://ppa.launchpad.net natty Release.gpg [316B] Ign http://ppa.launchpad.net/linaro-maintainers/overlay/ubuntu/ natty/main Translation-en Ign http://ppa.launchpad.net/linaro-maintainers/overlay/ubuntu/ natty/main Translation-en_US Get:2 http://ports.ubuntu.com natty Release.gpg [198B] Ign http://ports.ubuntu.com/ natty/main Translation-en Ign http://ports.ubuntu.com/ natty/main Translation-en_US Ign http://ports.ubuntu.com/ natty/universe Translation-en Ign http://ports.ubuntu.com/ natty/universe Translation-en_US Get:3 http://ports.ubuntu.com natty Release [39.8kB] Get:4 http://ppa.launchpad.net natty Release [9,753B] Get:5 http://ppa.launchpad.net natty/main armel Packages [42.8kB] Get:6 http://ports.ubuntu.com natty/main armel Packages [1,528kB] Get:7 http://ports.ubuntu.com natty/universe armel Packages [5,867kB] Fetched 7,488kB in 46s (162kB/s) Reading package lists... Done I: Calculating required packages. apt-get -y -o Apt::Architecture=armel -o Dir::Etc::TrustedParts=/etc/apt/trusted.gpg.d -o Dir::Etc::Trusted=/etc/apt/trusted.gpg -o Apt::Get::Download-Only=true -o Apt::Install-Recommends=false -o Dir=/bld/nano-nano-cross/chroot/ -o Dir::Etc=/bld/nano-nano-cross/chroot/etc/apt/ -o Dir::State=/bld/nano-nano-cross/chroot/var/lib/apt/ -o Dir::State::Status=/bld/nano-nano-cross/chroot/var/lib/dpkg/status -o Dir::Cache=/bld/nano-nano-cross/chroot/var/cache/apt/ install 2.7-minimal adduser apt apt-transport-https apt-utils base-files base-passwd bash bsdutils busybox-initramfs busybox-linaro-adduser busybox-linaro-findutils busybox-linaro-grep busybox-linaro-hostname busybox-linaro-mount busybox-linaro-nano busybox-linaro-tar busybox-linaro-udhcpc busybox-syslogd ca-certificates console-setup console-terminus coreutils cpio dash debconf debconf-i18n debianutils diffutils dpkg dropbear e2fslibs e2fsprogs findutils gcc-4.5-base gnupg gpgv grep gzip hostname ifupdown initramfs-tools initramfs-tools-bin initscripts insserv iproute iputils-ping kbd keyboard-configuration klibc-utils libacl1 libattr1 libblkid1 libbz2-1.0 libc-bin libc6 libcap2 libcomerr2 libcurl3-gnutls libdb4.8 libdbus-1-3 libdrm-intel1 libdrm-nouveau1a libdrm-radeon1 libdrm2 libgcc1 libgcrypt11 libglib2.0-0 libgnutls26 libgpg-error0 libgssapi-krb5-2 libk5crypto3 libkeyutils1 libklibc libkrb5-3 libkrb5support0 libldap-2.4-2 liblocale-gettext-perl liblzma2 libncurses5 libncursesw5 libnih-dbus1 libnih1 libpam-modules libpam-modules-bin libpam-runtime libpam0g libpcre3 libplymouth2 libpng12-0 libpopt0 libreadline6 libsasl2-2 libselinux1 libsepol1 libslang2 libss2 libssl0.9.8 libstdc++6 libtasn1-3 libtext-charwidth-perl libtext-iconv-perl libtext-wrapi18n-perl libudev0 libusb-0.1-4 libuuid1 linaro-overlay linaro-overlay-minimal locales login lsb-base makedev mawk module-init-tools mount mountall multiarch-support ncurses-base ncurses-bin net-tools netbase netcat-openbsd passwd perl-base plymouth procps python python-minimal python2.7-minimal readline-common sed sensible-utils sudo sysv-rc sysvinit-utils tar tzdata ubuntu-keyring udev upstart util-linux xkb-data xz-utils zlib1g Reading package lists... Done Building dependency tree... Done Note, selecting 'python2.7-minimal' for regex '2.7-minimal' Some packages could not be installed. This may mean that you have requested an impossible situation or if you are using the unstable distribution that some required packages have not yet been created or been moved out of Incoming. The following information may help to resolve the situation: The following packages have unmet dependencies: busybox-linaro-adduser : Conflicts: adduser busybox-linaro-grep : Conflicts: grep busybox-linaro-hostname : Conflicts: hostname busybox-linaro-mount : Conflicts: mount busybox-linaro-tar : Conflicts: cpio Conflicts: tar -- Regards, Tom "We want great men who, when fortune frowns will not be discouraged." - Colonel Henry Knox Linaro.org │ Open source software for ARM SoCs w) tom.gall att linaro.org w) tom_gall att vnet.ibm.com h) tom_gall att mac.com

13 years, 8 months

[ACTIVITY] OCTO status report - wk 33.2011 (20110815-20110819)

by Ilias Biris

Last meeting minutes : https://wiki.linaro.org/OfficeofCTO/2011-08-16 Status report link: https://wiki.linaro.org/OfficeofCTO/WeeklyReport Summary of the highlights: - work is continuing for ARMHF and by the end of the week we will have a clear picture of what was done from the original targets for August. - Device Tree: work is almost completed now from OCTO point of view, patches coming through to support DT - Boot architecture: ramping up after Connect. UEFI is a hot topic. - Arm server: Work carried out on network boot, installation and deployment of ARM servers. During August a Cobbler setup will be tested and documented. Best, -- Ilias Biris ilias.biris(a)linaro.org Project Manager, Linaro M: +358504839608, IRC: ibiris Skype: ilias_biris Linaro.org│ Open source software for ARM SoCs

13 years, 8 months

Re: QA tests for Android

by Zach Pfeffer

Sure. I didn't list it because Android doesn't really support generic USB (it will in Ice Cream Sandwich through a new USB device layer) Alexander, Would you add: * USB * Keyboard * Mouse On 17 August 2011 01:46, David Gilbert <david.gilbert(a)linaro.org> wrote: > On 16 August 2011 23:09, Zach Pfeffer <zach.pfeffer(a)linaro.org> wrote: >> Here's the proposed QA list. Comments and additions welcome. This list >> is based off the Ubuntu Origen list at: >> >> https://wiki.linaro.org/Cycles/1107/BoardSupportStatus/Samsung/Origen/Ubuntu > > Hi Zach, > I don't see USB testing in general in there; I'd have USB keyboard/mouse as > a basic test for boards that support it and I'd make sure I'd test > that on all USB > controllers on the board where the board has more than one. > > Dave >

13 years, 8 months

Re: issues benchmarking 11.08 toolchain

by Alexander Sack

On Mon, Aug 22, 2011 at 10:47 PM, Alexander Sack <asac(a)linaro.org> wrote: > That looks suspiciously close to the code that we patched out in platform a > while back to get 4.5/4.6 going in external/elfcopy/: > > > http://android.git.linaro.org/gitweb?p=platform/external/elfcopy.git;a=comm… > > This seems to be still used in our linaro_android_2.3.5 builds still. > > maybe the benchmark has a copy of that code and needs the same patch? or > its pulling in external/elfcopy and just needs to pull the > linaro_android_2.3.5 branch? > > turned out that benchmark.git has copies of soslim and elfcopy from a random AOSP eclair build that were build with an unpatched elfcopy.c. Replacing them with binaries from a fresh linaro build that has the patch from above applied makes this work again. Since we hit this problem with our platform builds earlier my guess would be that we just have been lucky when building benchmarks so far. andy will work around for now and will submit updated binaries to gerrit once release is out. > On Mon, Aug 22, 2011 at 10:33 PM, Andy Doan <andy.doan(a)linaro.org> wrote: > >> I'm hitting an issue building some of the tests in benchmark.git for >> this cycle. The python test is a good example: >> >> http://pastebin.com/e7XeLrMd: >> LINK SHARED LIBRARY: obj/libpython2.6.so >> obj/Modules/posixmodule.o: In function `posix_tempnam': >> >> /home/doanac/linaro/android/benchmarks/bmarks-bins-2011-08/linaro-4.6/benchmark/python/src/Modules/posixmodule.c:7167: >> warning: warning: tempnam() possibly used unsafely; consider using >> mkstemp() >> obj/Modules/posixmodule.o: In function `posix_tmpnam': >> >> /home/doanac/linaro/android/benchmarks/bmarks-bins-2011-08/linaro-4.6/benchmark/python/src/Modules/posixmodule.c:7214: >> warning: warning: tmpnam() possibly used unsafely; consider using >> mkstemp() >> MAP SHARED LIBRARY: out/SYMBOL/libpython2.6.so >> ASSERTION FAILURE external/elfcopy/elfcopy.c:2457: >> [!(shdr_info[sym->st_shndx].shdr.sh_flags & SHF_ALLOC)] >> make: *** [out/SYMBOL/libpython2.6.so] Error 1 >> >> >> anyone have a workaround so I can complete the benchmarking? >> >> -andy >> > > > > -- > > - Alexander > > -- - Alexander

13 years, 8 months

Re: Input for an "optimized" slide

by Zach Pfeffer

Thanks Bero. Sending this extremely useful information out to a wider audience. Alex, I think you're probably be very interested in this for your Mozilla work. >> -O3 >> * What is is, does, available on > > -O3 enables several additional compiler optimizations such as tree > vectorizing and loop unswitching, and optimizes for speed over code > size somewhat more aggressively than -O2, e.g. by inlining all calls > to small static functions. > It is available on any platform supported by gcc. > >> OpenMP >> * What is is, does, available on > > OpenMP is a simple API that makes it easier for a programmer to make > use of multi-core or multi-processor systems, e.g. by automatically > splitting marked loops into several threads. > Example: > > #pragma omp parallel for > for(int i=0; i<100; i++) > do_something(i); > > Would use up to 100 threads to do its job. > > > It is available on plaforms supported by gcc that can use libgomp, > gcc's OpenMP library. This includes most platforms that support POSIX > threads - but -- initially -- not Android. > > >> Loop parallelization >> * What is is, does, available on > > Loop parallelization takes OpenMP a step further by automatically > determining which loops are suitable for "#pragma omp parallel for" > and similar constructs. This allows code that was written without > multiprocessing in mind (such as most code written specifically for > ARM platforms - multicore/SMP ARM systems are quite new) to take > advantage of multicore/SMP systems (to some extent) without having to > modify the code. > > Compiler flag: -ftree-parallelize-loops=X (where X is the number of > threads to be optimized for - typically the number of CPU cores in the > target system) > > Available on anything supported by gcc that has both libgomp and > graphite (incl. CLooG, PPL or ISL) - the original Android toolchain > has neither of those. > >> ...and any other optimizations that you've done. > > None of the following is enabled yet (but the support in the toolchain > is there now), but I'm planning to enable them step by step once we > have systems built w/ the new toolchain that actually boot: > > binutils: --hash-style=gnu > By default, ld creates SysV style hash tables for function tables > in shared libraries. With --hash-style=gnu, we switch to GNU style > hashes, making symbol lookup a lot faster. (details: > http://sourceware.org/ml/binutils/2006-10/msg00377.html) > > binutils: -Bsymbolic-functions > Speed up the dynamic linker by binding references to global > functions in shared libraries where it is known that this doesn't > break things (it's safe for libraries that don't have any users trying > to override their symbols - it's probably safe to assume e.g. skia and > opengl could benefit). > (details: http://www.fkf.mpg.de/edv/docs/intel_composer/Documentation/en_US/compiler_…) > > binutils/gcc: -flto, -fwhole-program > Link-Time Optimization - causes code to be optimized again at link > time, when the compiler knows what functions are called form what > parts of the code, what functions are only called with constant > parameters, etc. > > gcc: -mtune=cortex-a9 (or whatever the actual target CPU is) > The Android build system uses -march=arm-v7a, which is good -- but > it doesn't do any tuning for the specifc CPU type (e.g. cortex-a8 vs. > cortex-a9). > > gcc: -fvisibility-inlines-hidden > Don't export C++ inline methods in shared libraries. Makes the > symbol table smaller, improving startup time and diskspace efficiency > > gcc: -fstrict-aliasing -Werror=strict-aliasing > Currently, Android uses -fno-strict-aliasing unconditionally for > thumb code, to work around some pieces of code that violate strict > aliasing rules. Using -Werror=strict-aliasing, we can determine what > pieces of code are affected, and fix them, or limit the use of > -fno-strict-aliasing to the specific files that need it - enabling the > rather useful strict-aliasing optimization for the rest of the build > > gcc: Investigate Graphite optimizations that aren't even enabled at -O3: > -fgraphite-identity -floop-block -floop-interchage > -floop-strip-mine -ftree-loop-distribution -ftree-loop-linear >

13 years, 8 months

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

linaro-dev