5.3.14 kernel available

Forum Forums News Announcements 5.3.14 kernel available

  • This topic has 11 replies, 5 voices, and was last updated Dec 15-4:23 am by fungalnet.
Viewing 12 posts - 1 through 12 (of 12 total)
  • Author
    Posts
  • #30189
    Forum Admin
    anticapitalista

    I have a 5.3.14 kernel available to antiX-19 64 bit users (buster, testing and sid).

    This may be useful for those with new hardware.

    Note: Virtualbox, broadcom-sta, ndiswrapper (and probably nvidia) drivers may not build for the buster version, but should build for testing and sid ones.

    Philosophers have interpreted the world in many ways; the point is to change it.

    antiX with runit - leaner and meaner.

    #30190
    Member
    Koo
    Helpful
    Up
    0

    Installed antix19 5.3.14-antix.1-amd64-smp from (5.2.21-antix.2-amd64-smp) this afternoon had to rerun sgfxi nvidia 440.36 again on boot up to into modules for the new kernel everything is running just fine. Thanks

    T430 i7-3632QM 16gb , antiX-19.2.1-runit_x64-base Hannie Schaft 29 March 2020 , 5.8.16-antix.1-amd64-smp

    #30220
    Member
    Koo
    Helpful
    Up
    0

    Tried running inxi this morning just get errors.

    Unmatched right curly bracket at /usr/local/bin/inxi line 20018, at end of line
    syntax error at /usr/local/bin/inxi line 20018, near "}"
    Unmatched right curly bracket at /usr/local/bin/inxi line 20019, at end of line
    Can't redeclare "my" in "my" at /usr/local/bin/inxi line 20022, near "my"
    Global symbol "$b_gcc" requires explicit package name (did you forget to declare "my $b_gcc"?) at /usr/local/bin/inxi line 20028.
    Global symbol "%data" requires explicit package name (did you forget to declare "my %data"?) at /usr/local/bin/inxi line 20029.
    

    and so on..

    T430 i7-3632QM 16gb , antiX-19.2.1-runit_x64-base Hannie Schaft 29 March 2020 , 5.8.16-antix.1-amd64-smp

    #30221
    Forum Admin
    anticapitalista
    Helpful
    Up
    0

    Are you running testing/sid repos?

    Even if you are or aren’t, I can’t replicate the error

    Philosophers have interpreted the world in many ways; the point is to change it.

    antiX with runit - leaner and meaner.

    #30223
    Member
    Koo
    Helpful
    Up
    0

    @ anticapitalista

    This was a full install of antiX19 all is standard apart from sgfxi install of nvidia 440.36 and i3.

    No (antiX19 buster) just tried with Debian 10 and it inxi work fine.

    After the preinstalled version on anitX19 gave errors I tried reinstall installing with cd /usr/local/bin && wget -Nc smxi.org/inxi && chmod +x inxi and not installing giving errors.

    Yet with Debian 10 I install inxi with cd /usr/local/bin && wget -Nc smxi.org/inxi && chmod +x inxi & it installs plus works.

    I will try uninstall inxi all together from antiX19 then reinstalling.

    Thanks for your reply.

    T430 i7-3632QM 16gb , antiX-19.2.1-runit_x64-base Hannie Schaft 29 March 2020 , 5.8.16-antix.1-amd64-smp

    #30224
    Forum Admin
    anticapitalista
    Helpful
    Up
    0

    inxi version in antiX-19 is via the smx-inxi deb shown here:

    https://repo.antixlinux.com/buster/pool/main/s/smxi-inxi-antix/

    So, try apt-get purge smxi-inxi-antix && apt-get install smxi-inxi-antix

    Philosophers have interpreted the world in many ways; the point is to change it.

    antiX with runit - leaner and meaner.

    #30225
    Member
    Koo
    Helpful
    Up
    0

    Ok I get this apt-get purge smxi-inxi-antix && apt-get install smxi-inxi-antix command ago.

    I find something interesting but as I see from your purge command and synaptic its self that the inxi script has been written for antiX and is not the full inxi install. Plus the script is part of control-centre as in smxi-inxi-anti 0.4.18 and inxi-gui-antix 0.3.7 so if you try to remove inxi from synaptic it removes control-centre as well. Also you can’t install inxi 3.032-1-1 from synaptic as it complains about over riding the smxi-inxi-antix script and will remove control-panel also.

    And yes your purse command worked Thank You

    T430 i7-3632QM 16gb , antiX-19.2.1-runit_x64-base Hannie Schaft 29 March 2020 , 5.8.16-antix.1-amd64-smp

    #30226
    Member
    Koo
    Helpful
    Up
    0

    Sorry about the typos on the first line and I hope the screenshot is not to big.

    T430 i7-3632QM 16gb , antiX-19.2.1-runit_x64-base Hannie Schaft 29 March 2020 , 5.8.16-antix.1-amd64-smp

    #30253
    Member
    olsztyn
    Helpful
    Up
    0

    Referring to issue of memory usage in 5.3.14 anti brought up in another thread (New Kernels Available) I have a question to experts (I am not one at all):
    Running ps_mem.py for antiX with 4.9 and 5.3.14 shows twice total memory used, which is reported neither by conky mem nor Htop. I think ps_mem was designed to more accurately report memory as used by running processes, with a split between private and shared.
    Inspecting memory use values for processes between antiX 19 kernel 4.9 and 5.3.14 seems to show roughly the same private memory use but way much higher shared memory. This seems what is happening across all processes…
    So my question is: What is causing that processes in antiX with kernel 5.3.14 are using multiple times the amount of shared memory than in case of kernel 4.9 or kernel 5.2 (as ps_mem shows reported by anti)?
    Such radical shift in shared memory use seems between antiX with kernel 5.2 and 5.3. It seems that a point release difference of kernel would not be expected to cause such radical change…

    #30675
    Member
    fungalnet
    Helpful
    Up
    0

    5.3.14 works very well on my machine, thanks anti-

    #30693
    Member
    skidoo
    Helpful
    Up
    0

    Such radical shift in shared memory use seems between antiX with kernel 5.2 and 5.3

    websearch:
    memory usage 5.2 5.3 linux kernel

    various quirks have been widely reported against 5.3.x
    Reputedly, most of ’em have been addressed (resolved) in the 5.4.x branch

    #30725
    Member
    fungalnet
    Helpful
    Up
    0

    Are you saying that the fixes implemented on 5.4 are not backported on 5.3? In my brief experience with 5.4 5.3 appeared as quicker, more responsive, lighter. Ever since 5.3.1 it appeared fast. I kept 4.20 for a while when 5.1 and 5.2 were coming out.
    I also take memory leak reports with a grain of salt. If one was to start every process in a terminal and keep track visibly of what they spit out during their life (as processes) what sometimes is reported as memory leak is just output of the program running, and the whole thing combined appears to be growing in memory use as that output has to stay somewhere for you to discover.
    I run as much as I can without dbus, so when software that make the assumption that dbus is running can’t find it they spit out warnings of “failure to communicate” (cool hand luke). But even when the damn thing is running it doesn’t mean it is successful in what it needs to do, so more trash being thrown out and not cleaned.
    Memory leaks are dangerous things and should be investigated and diagnosed, don’t let me discourage anyone from hunting.

    It is like some scumbag waiting outside a school and saying “hey, hi Mary” – “I am not Mary, I am Suzie Quo, I live on 234 5th street, my daddy is rich” – “that’s what I wanted to know!”.

Viewing 12 posts - 1 through 12 (of 12 total)
  • You must be logged in to reply to this topic.