Der BOINC-Entwicklungsfred

Zuletzt bearbeitet:
Moin,

die Boinc Installer 7.2.31 sind für:

windows_intel86
windows_x86_64

windows_intel86_vbox
windows_x86_64_vbox

sind draussen!

es stehen auch noch 7.2.30 Versionen für Linux zur Verfügung
x86_64-pc-linux-gnu.sh
x686-pc-linux.gnu.sh

und eine Zwischerversion 7.2.28.1 für:

windows_intel86
windows_x86_64

windows_intel86_vbox
windows_x86_64_vbox


Download: http://boinc.berkeley.edu/dl/?C=M;O=D

Changelog:
Preliminary Change Log 7.2.28 --> 7.2.29:

client: Give boinc_master full control of any process spawned using the service token created by the client for boinc_project.

This will allow the core client to kill VirtualBox VM's launched indirectly by vboxwrapper. Vboxwrapper launches vboxsvc.exe which launches vboxheadless.exe. This should also take care of the core client being able to kill child processes of the regular wrapper as well. I don't know the full scope of this type of issue? Maybe the default ACLs for a process changed within the last couple of versions of Windows.

WINBUILD: Fix build break from last commit.
MGR: Disable downloading image files in notices.
client: AMD GPU naming, from Jord.
client: add is_wrapper flag to APP_VERSION. If set, the main program uses little CPU and can be run at elevated priority.
client: parse (but ignore) <min_rpc_time> in state file.
client: don't delete global prefs file if we detach the project it came from.
client: message tweak.
LIB: Re-introduce code to get a host's MAC address.
MGR: Add a VirtualBox icon to the list of supported platforms for any given project that uses VirtualBox.
MGR: Change the VirtualBox icon color pallet so there are no artifacts around the image after conversion.
BOINC lib: change get_mac_address to return just one address; prefer Ethernet.
client: generate host CPID as a function of MAC address

This makes the host CPID stable; if you repeatedly install BOINC on a particular node, it will get the same host CPID each time, and your host table won't get lots of redundant entries. A host can have multiple NICs; we use the MAC address of the first Ethernet controller we find, or the last NIC if there is none. Of course, this will create problems if we get the same MAC address for different hosts; in principle this shouldn't happen.

Remove the unused file hostinfo_network.h
compile fix.
client: On Macintosh, return MAC address for first Ethernet controller only.
client: fix error returns for get_mac_address() on Unix.
WINBUILD: Add mac_address.cpp, .h to the boinc_cli project.
client: preliminary implementation (commented out) of sub-second throttling.
client: debug sub-second CPU throttling.
client: Initialize diagnostics framework for the sub-second CPU throttler.


Preliminary Change Log 7.2.29 --> 7.2.30:

client: allow <app_version> elements in app_config.xml, allowing users to override the parameters of particular app versions.
client and manager: the set_cc_config and get_cc_config GUI RPCs had nonstandard replies, which weren't parsed correctly by the calling code.

The reply from a GUI RPC that don't return info is one of:
<success/>
<error>error message</error>
<status>N</status>

LIB: Special case a value of 0.0 in time_to_string()
LIB: Fix build break.

Preliminary Change Log 7.2.28 --> 7.2.28.1:

LIB: Disable the ability to download images within notices.

Preliminary Change Log 7.2.30 --> 7.2.31:

client: Give a visual indication of a hot fix.
MGR: Disable downloading of all file types. POGS is also using a script to generate an image.


---------- Beitrag hinzugefügt um 12:30 ---------- Vorheriger Beitrag um 08:23 ----------

die Boinc 7.2.32 Installer für:

macOSX_SymbolTablets
macOSX_i686
i686-macOS-apple-darwin

Download:http://boinc.berkeley.edu/dl/?C=M;O=D
 
7.2.29 ist da

Windows 32 (vbox) / 64 (vbox)

Gibt es eine Version wo schon Linux und der Boinc Manager mit an Board sind?
Ich hatte mir auf dem Läppi die Win 64 vbox Version runtergeladen.
Aber da passiert ja nichts weiter als das vbox "nackig" installiert wird.
Oder mache ich da einen Fehler?
 
Nee VBox-Boinc ist nur Boinc + VBOX.

Die jeweiligen Boincprojekte laden eine eigene VM runter und starten die jeweiligen Apps in der eigenen VM:
- RNA@Home
- LHC@Home (T4)
 
So wer nicht warten möchte, die Installer 7.2.32 für:


windows_intel86
windows_x86_64

windows_intel86_vbox
windows_x86_64_vbox sind auch veröffentlicht! Download:http://boinc.berkeley.edu/dl/?C=M;O=D

Changelog kommt sobald er da ist.

Edit:

die Linux Installer 7.2.32 können jetzt auch geladen werden!

x86_64-pc-linux-gnu.sh
x686-pc-linux.gnu.sh
 
Zuletzt bearbeitet:
Laut Wiki ist der Changelog für die 7.2.32 wie folgt:

LIB: Fix build break.
WINBUILD: Updates for 7.2.30
Move thread.cpp,h from lib/ to client/, to remove dependency on pthread in libboinc
MGR: Disable the download of all embedded content within a notice.
 
Hi,

Ihr könnt ab sofort die 7.2.33 Installer für :

windows_intel86
windows_x86_64

windows_intel86_vbox
windows_x86_64_vbox

x86_64-pc-linux-gnu.sh
x686-pc-linux.gnu.sh

macOSX_SymbolTablets
macOSX_i686
macOS_apple_darwin

jetzt Downloaden und zwar hier: http://boinc.berkeley.edu/dl/?C=M;O=D


Changelog:

Preliminary Change Log 7.2.28 --> 7.2.33:

MGR: Disable downloading image files in notices.
LIB: Disable the ability to download images within notices.
client: Give a visual indication of a hot fix.
MGR: Disable downloading of all file types. POGS is also using a script to generate an image.
client and manager: the set_cc_config and get_cc_config GUI RPCs had nonstandard replies, which weren't parsed correctly by the calling code.

The reply from a GUI RPC that don't return info is one of:
<success/>
<error>error message</error>
<status>N</status>
 
Zuletzt bearbeitet:
So, es ist wieder soweit. Die neuen Boinc Installer 7.2.34 für:

macOSX_SymbolTablets
macOSX_i686
macOS_apple_darwin


windows_intel86
windows_x86_64

windows_intel86_vbox
windows_x86_64_vbox

x86_64-pc-linux-gnu.sh
x686-pc-linux.gnu.sh

sind eingetrudelt. Download hier--> http://boinc.berkeley.edu/dl/?C=M;O=D

es können auch neue vboxwrapper geladen werden!
 
Zuletzt bearbeitet:
So es hat lange gedauert, aber hier ist er:


Preliminary Change Log 7.2.33 --> 7.2.34:

client: fix bugs with CPU throttling and GPU apps.

Various bad things could happen when CPU throttling was used together w/ GPU apps.
Examples:

-> on a multi-GPU system, several GPU tasks are assigned to the same GPU.
-> a suspended GPU task remains in memory (tying up its GPU resources) while other tasks try to use the GPU.

The problem was that parts of the code assumed that suspended GPU processes don't exist - i.e. that when a GPU task is suspended it's always removed from memory.
This isn't true in the presence of CPU throttling.

So David made the following changes:

-> When assigning GPUs to tasks, treat suspended tasks like running tasks (i.e. reserve their GPUs).
-> At the end of the CPU-scheduling logic, if there are any GPU tasks that are suspended and not scheduled, remove them from memory, and trigger a reschedule so we can reallocate their GPUs.

Also, a cosmetic change: in the resource usage string shown in the GUI, include "(device X)" even if the task is suspended (i.e. because of throttling).
Also: zero out COPROC::opencl_device_indexes[] so we don't write a garbage number to init_data.xml for non-OpenCL jobs.

MGR: Store the ISO language ID for future use.
Usage: wxString strLang = wxGetApp().GetISOLanguageID();
MGR: Bug fix to my previous commit, initialize string after we initialize the wxLocale object with the proper locale information.
Client and Manager: make notices translatable

-> Add a GUI RPC ("set_language") that lets the Manager communicate the user's selected language code to the client at startup.
-> The client stores the language code in the client state file.
-> The client appends a "lang=X" GET argument to the URLs from which notices are fetched.
-> The next steps (not done) are 1) to change the get_notices.php script to parse the argument and do translation, and 2) extend our Pootle system to allow volunteer translation of notices by all projects.



Gleichzeitig wünsche ich allen ein "Frohes Fest und einen Guten Rutsch ins Neue Jahr". :w_feiern:
 
So, nach der Wrapper Orgie hier mal wieder was Handfestes. ;D

Die Boinc Installer 7.2.36 für:

windows_intel86
windows_x86_64

windows_intel86_vbox
windows_x86_64_vbox

x686-pc-linux.gnu.sh
x86_64-pc-linux-gnu.sh

macOSX_SymbolTablets
macOSX_i686_apple_darwin
macOS_i686

und die Boinc.apk für Android sind erschienen.


Download: http://boinc.berkeley.edu/dl/?C=M;O=D
 
Zuletzt bearbeitet:
So, hier ist der fehlende Changelog:


Preliminary Change Log 7.2.34 --> 7.2.35:

MGR: work around a problem sizing the URL TextEdit field in the Attach Wizard Project Info page when a larger default font size is selected on Windows.
I don’t understand why this “fixes” it and I suspect there is a better solution.
MGR: Fix drawing of Task Control on Windows.

Preliminary Change Log 7.2.35 --> 7.2.36:

MGR: Prevent zero-size window in a rare situation where registry values have accidentally been set to zero.
MGR: Reverse commit a1fb470 (Fix drawing of Task Control on Windows) to fix build break; that change was for use with wxWidgets 3.0 only and should not have been ported to this branch.
client: Fix typo which had no actual effect in practice.
client: fix bug that caused idle GPUs when CPU throttling used.
client: if have a non-English language, set ACCEPT_LANGUAGE in all HTTP requests.
client: message tweak.
client: fix bugs involving CPU throttling and GPU apps.

Suspended tasks can be either left in memory (LIM) or removed from memory (RFM). CPU throttling always uses LIM. Other types of suspension (e.g. user request) use LIM or RFM depending on user prefs, except that RFM is always used for GPU tasks.

There was a bug: if tasks were suspended because of CPU throttling, and then the user suspended activity, GPU apps would remain LIM. They need to be RFM.

client: allow <file> in app_info.xml

In a checkin on 20 July 2011, I changed <file_info> to <file> in client_state.xml. However, the code that parses app_info.xml still requires <file_info>. This means you can't copy and paste stuff from client_state.xml to app_info.xml. Fix: parse either <file> or <file_info> in app_info.xml

client: when write file upload error XML to stderr, include user-friendly string.
client: report tasks within 1 hour of completion... instead of 24. Better user experience at the expense of slightly increased server load.
 
Moin Moin,

wenn ich die changelogs richtig interpretiere kann der Windows Boinc client immer noch nicht den AVX Befehlssatz der neueren Prozessoren erkennen, oder?
Bei Asteroids@home hole ich mir deren avx app per app_info aber es wäre ja schön wenn das mit Bordmitteln ginge.
 
die 7.2.37 für:

windows_intel86
windows_x86_64

windows_intel86_vbox
windows_x86_64_vbox

x86_64-pc-linux-gnu.sh
x686-pc-linux.gnu.sh


macOSX_SymbolTablets
macOSX_i686_apple_darwin
macOS_i686

Boink.apk für Android

ist da!

Download: http://boinc.berkeley.edu/dl/?C=M;O=D
 
Zuletzt bearbeitet:
Es hat etwas gedauert, aber hier ist der Changelog:7.2.36 --> 7.2.37


Preliminary Change Log:7.2.36 --> 7.2.37

MGR: Remove outdated comments.
client: fix bug that caused lots of spurious "no shared memory segment" msgs.

ACTIVE_TASK_SET::suspend_all() originally skipped tasks in states other than PROCESS_EXECUTING. I took this out in commit 47b4d6b because - for example - a GPU task might be suspended due to CPU throttling, and therefore left in memory, but if it's then suspended for some other reason, it must be removed from memory.

However, this change was overkill - it causes tasks for which no process exists to be suspended, resulting in the spurious msgs.

Solution: skip tasks in states other than PROCESS_EXECUTING and PROCESS_SUSPENDED.

lib: make run_on_batteries false for all platforms by default.
 
Changelog: 7.2.37-->7.2.38


Preliminary Change Log 7.2.37 --> 7.2.38:

client: job scheduler tweaks to avoid idle CPUs
-> Allow overcommitment by > 1 CPU. E.g. If there are two 6-CPU jobs on an 8 CPU machine, run them both.
-> Prefer MT jobs to ST jobs in general. When reorder the run list (i.e. converting "preliminary" to "final" list), prefer job J1 to J2 if:
1) J1 is EDF and J2 isn't.
2) J1 uses GPUs and J2 doesn't.
3) J1 is in the middle of a timeslice and J2 isn't.
4) J1 uses more CPUs than J2.
5) J1's project has higher scheduling priority than J2's ... in that order.

4) is new; it replaces the function promote_multi_thread_jobs(), which did something similar but didn't work in some cases.

client: fix problems related to CPU throttling.
-> Don't throttle GPU apps. GPU apps spend all their time in a critical section, during which they can't be suspended. They length of these critical sections (i.e. of GPU kernels) may be a significant part of a second, or more, so sub-second throttling isn't possible.
-> Account elapsed time correctly when throttling is used.
-> Also (not related to throttling) don't schedule a job in QUIT_PENDING or ABORT_PENDING state. Doing so results in 2 processes in the slot dir, and can cause the job to fail.

client: Fix compiler warning (unused static function).
client: if app doesn't report fraction done, estimate it.
client: if app doesn't report fraction done, estimate fraction done in a way that converges to but never reaches 100%.
 
Neue Installer 7.2.39 für:


windows_intel86
windows_x86_64

windows_intel86_vbox
windows_x86_64_vbox

macOSX_SymbolTablets
macOSX_i686_apple_darwin
macOS_i686

x86_64-pc-linux-gnu.sh
x686-pc-linux.gnu.sh

und die Boinc_7.2.39.apk

sind da!


Download: http://boinc.berkeley.edu/dl/?C=M;O=D
 
Zuletzt bearbeitet:
Mein Tip:

Erstma Finger weg vom 38er und 39er - es gibt erhebliche Probleme mit CPU-P-States - Boinc verhindert offenbar, dass die CPUs runtertakten, wenn keine Arbeit erfolgt !?
 
Japp - ist erheblicher Emailverkehr dazu im Gange (Maillists)
 
so trotz des Bug im 7.2.39 Installer, noch der Changelog:


Preliminary Change Log 7.2.38 --> 7.2.39:

client: <cpu_sched> shouldn't show suspend/resume msgs for CPU throttling.
client: fix bugs in accounting elapsed time.
client: don't use sub-second CPU throttling.

The developer forgot that the wrapper has a 1-second poll for suspend and resume, so sub-second throttling won't work properly for wrapper apps. Revert to a variant of the old scheme, in which the min of the suspended and resumed periods is 1 sec. Also, fix task start/suspend/resume log messages.
 
Die Boinc Version 7.3.2 für:

windows_intel86
windows_x86_64

windows_intel86_vbox
windows_x86_64_vbox


ist da!

Download:http://boinc.berkeley.edu/dl/?C=M;O=D




BOINC 7.2.39 available for the Public for Windows, Linux and Macintosh.

---------- Beitrag hinzugefügt um 09:02 ---------- Vorheriger Beitrag um 00:14 ----------

ich habe mir die "neue" Version 7.3.2 installiert, jedoch wenn ich Boinc öffne wird mir die Version 7.3.1 angezeigt.

Ist das ein Fehler oder ist das so gewollt?
 
Zuletzt bearbeitet:
Naja der Manager ist 7.3.2 aber der Core nur 7.3.1 - bekannter "Fehler" ....

Lustiger ist der Fred bzgl. dass man nicht 7.4.0/7.3.2 unter Fedora buildn kann ...
 
Zurück
Oben Unten