Is it just me, or is Manjaro XFCE using more RAM lately?

@winnie

Hello. I’m a beginner, so please bear with me.

First of all, I suppose I should rephrase my question? When I say “ram usage” I was referring to the number I read from htop. Typically when I ran htop right after I boot, the “Mem” meter showed 680MB/31.2G

It was only fairly recently that the 680MB became 1.1 GB, and I was just wondering about the change.

1 Like

I wasn’t being technical when I implore people to mentally “reframe” the question / concern in their minds.

Old paradigms don’t always fit well into modern times. In the past it was all about squeezing down every last bit of RAM “usage” due to the smaller breathing room our systems had; lower hardware resources and capacities.

Now it’s different. As the performance-to-price ratio continues to rise, it’s a matter of using your hardware, RAM included.

“More sophisticated applications? More complex themes? Greater modularity? Caching data for near-instant access? Lessening the burden on non-volatile storage? Let’s use everything we’ve got!”

Otherwise, what’s the point in having 32GB of RAM on your personal computer, if it’s not even being used?

What’s the point of using a top-tier CPU with the highest benchmarks, if all you do is browse the web and check email, and constantly monitor your CPU temps to keep it as low as possible?


Look at games, web browsers, operating systems, and software in general. They require more disk space since they are larger in size than their predecessors. Hence, more to load up into RAM.

6 Likes

just boot and reports

inxi -Fza

what is your video card ?

You’re sure this trend doesn’t have more to do with cheap/lazy development (not that I’m calling the XFCE devs lazy, but more big tech and the games industry) because there is an expectation for a fast SSD, fast and lots of empty RAM and a new generation fast CPU and to use those resources frivolously?

It’s all well and good if you have a new computer but my 8GB 1333Mhz DDR2 RAM is really starting to be come nowhere near suitable for running exactly the same stuff it ran 7 years ago.

If I play Dota 2 and open Firefox or Signal (or any other electron app) my system has to aggressively swap from ram and if I’m not careful my whole system can become quickly unusable.

1 Like

You can download an older image of Manjaro Xfce and compare ram usage of different processes.

1 Like

I think a lot of the comments so far have no real evidence to support some of the speculations

RAM usage on htop is shown in binary units Gibibytes and Mebibytes instead of decimal units Gigabytes and Megabytes

  • 1 GiB = 29 bytes = 1,073,741,824 bytes
  • 1 GB = 109 bytes = 1,000,000,000 bytes

My system has a few modifications from stock Manjaro and has been rolling for over 5 years

$ free -h
               total        used        free      shared  buff/cache   available
Mem:           7.7Gi       538Mi       6.7Gi       4.0Mi       505Mi       7.0Gi
Swap:          3.9Gi          0B       3.9Gi

My partner’s Xfce has not been modified much from standard Manjaro install and has only been rolling for 18 months

$ free -h
               total        used        free      shared  buff/cache   available
Mem:           7.8Gi       493Mi       6.7Gi        57Mi       578Mi       7.0Gi
Swap:          3.9Gi          0B       3.9Gi

Both systems have 8GB (7.45 GiB) RAM installed

So these systems are using approximately 6% of available RAM when booted
But your system is only using about 3.5% of available RAM

I do not think that having a bit more RAM used when OS is first booted will impair performance of Xfce when in use

2 Likes

Hello, no it’s not just you, thanks for posting I was wondering where to post then found this thread. I have been religiously running HTOP after every reboot for about a year now then keeping it open to monitor, my ram usage jumped exactly after the last stable update. No other changes. Previously it would always start fresh around 600MB (or I now gather it’s actually MiB,

it now starts at 1000+. This is my long-running messed with home XFCE system, I have a barely touched backup/road XFCE system that went from 500+ to 700+, again right after the last stable update.

I too am a beginner and will be studying this thread, from reading I understand that this may not be a ‘problem’ (I have 32MB of ram as well, 8MB on my backup) but just wanted to post because it definitely presented itself right after the last stable update (2021-09-24)

output from my barely tweaked XFCE road machine, I will add my long time home XFCE later after I reboot, thanks

[tgz@roadm6500 ~]$ inxi -Fza
System:
  Kernel: 5.10.68-1-MANJARO x86_64 bits: 64 compiler: gcc v: 11.1.0 
  parameters: BOOT_IMAGE=/boot/vmlinuz-5.10-x86_64 
  root=UUID=376164d0-ad70-4b70-b0c4-3aacf4cb1aba rw quiet apparmor=1 
  security=apparmor udev.log_priority=3 
  Desktop: Xfce 4.16.0 tk: Gtk 3.24.29 info: xfce4-panel wm: xfwm 4.16.1 
  vt: 7 dm: LightDM 1.30.0 Distro: Manjaro Linux base: Arch Linux 
Machine:
  Type: Portable System: Dell product: Precision M6300 v: N/A 
  serial: <filter> Chassis: type: 8 serial: <filter> 
  Mobo: Dell model: 0JM680 serial: <filter> BIOS: Dell v: A15 
  date: 06/05/2013 
Battery:
  ID-1: BAT0 charge: 48.1 Wh (100.0%) condition: 48.1/79.9 Wh (60.2%) 
  volts: 12.2 min: 11.1 model: Sanyo DELL C54478 type: Li-ion 
  serial: <filter> status: Full 
CPU:
  Info: Dual Core model: Intel Core2 Duo T9300 bits: 64 type: MCP 
  arch: Penryn family: 6 model-id: 17 (23) stepping: 6 microcode: 60F cache: 
  L2: 6 MiB 
  flags: lm nx pae sse sse2 sse3 sse4_1 ssse3 bogomips: 9977 
  Speed: 798 MHz min/max: 800/2501 MHz boost: enabled Core speeds (MHz): 
  1: 798 2: 798 
  Vulnerabilities: Type: itlb_multihit status: KVM: VMX unsupported 
  Type: l1tf mitigation: PTE Inversion 
  Type: mds 
  status: Vulnerable: Clear CPU buffers attempted, no microcode; SMT disabled 
  Type: meltdown mitigation: PTI 
  Type: spec_store_bypass status: Vulnerable 
  Type: spectre_v1 
  mitigation: usercopy/swapgs barriers and __user pointer sanitization 
  Type: spectre_v2 
  mitigation: Full generic retpoline, STIBP: disabled, RSB filling 
  Type: srbds status: Not affected 
  Type: tsx_async_abort status: Not affected 
Graphics:
  Device-1: AMD RV515/M54 [Mobility Radeon X1400] vendor: Dell 
  driver: radeon v: kernel bus-ID: 01:00.0 chip-ID: 1002:7145 class-ID: 0300 
  Display: x11 server: X.Org 1.20.13 compositor: xfwm4 v: 4.16.1 driver: 
  loaded: ati,radeon unloaded: modesetting alternate: fbdev,vesa 
  display-ID: :0.0 screens: 1 
  Screen-1: 0 s-res: 1920x1200 s-dpi: 96 s-size: 508x318mm (20.0x12.5") 
  s-diag: 599mm (23.6") 
  Monitor-1: LVDS res: 1920x1200 hz: 60 dpi: 133 size: 367x230mm (14.4x9.1") 
  diag: 433mm (17.1") 
  Monitor-2: DVI-0 res: 1920x1200 hz: 60 dpi: 94 
  size: 519x324mm (20.4x12.8") diag: 612mm (24.1") 
  OpenGL: renderer: ATI RV515 v: 2.1 Mesa 21.2.2 direct render: Yes 
Audio:
  Device-1: Intel 82801H HD Audio vendor: Dell driver: snd_hda_intel 
  v: kernel bus-ID: 00:1b.0 chip-ID: 8086:284b class-ID: 0403 
  Sound Server-1: ALSA v: k5.10.68-1-MANJARO running: yes 
  Sound Server-2: JACK v: 1.9.19 running: no 
  Sound Server-3: PulseAudio v: 15.0 running: yes 
  Sound Server-4: PipeWire v: 0.3.37 running: no 
Network:
  Device-1: Broadcom NetXtreme BCM5756ME Gigabit Ethernet PCI Express 
  vendor: Dell driver: tg3 v: kernel port: ee00 bus-ID: 09:00.0 
  chip-ID: 14e4:1674 class-ID: 0200 
  IF: enp9s0 state: up speed: 1000 Mbps duplex: full mac: <filter> 
Drives:
  Local Storage: total: 149.05 GiB used: 13.2 GiB (8.9%) 
  SMART Message: Required tool smartctl not installed. Check --recommends 
  ID-1: /dev/sda maj-min: 8:0 vendor: Intel model: SSDSA2M160G2GN 
  size: 149.05 GiB block-size: physical: 512 B logical: 512 B 
  speed: 3.0 Gb/s type: SSD serial: <filter> rev: 02M3 scheme: MBR 
Partition:
  ID-1: / raw-size: 149.05 GiB size: 145.71 GiB (97.76%) 
  used: 13.2 GiB (9.1%) fs: ext4 dev: /dev/sda1 maj-min: 8:1 
Swap:
  Alert: No swap data was found. 
Sensors:
  System Temperatures: cpu: 52.0 C mobo: N/A sodimm: SODIMM C 
  Fan Speeds (RPM): cpu: 0 
Info:
  Processes: 158 Uptime: 27m wakeups: 2 Memory: 7.77 GiB 
  used: 807.6 MiB (10.2%) Init: systemd v: 249 tool: systemctl Compilers: 
  gcc: 11.1.0 Packages: pacman: 1207 lib: 393 flatpak: 0 Shell: Bash 
  v: 5.1.8 running-in: xfce4-terminal inxi: 3.3.06

output from immediately after booting my long time home XFCE system, went from 600 to 1000+ MiB after update

[tgz@xfce-1 ~]$ inxi -Fza
System:
  Kernel: 5.10.68-1-MANJARO x86_64 bits: 64 compiler: gcc v: 11.1.0 
  parameters: BOOT_IMAGE=/boot/vmlinuz-5.10-x86_64 
  root=UUID=cf9d3d26-6bb5-40df-bf67-b4729a325742 rw quiet apparmor=1 
  security=apparmor udev.log_priority=3 
  Desktop: Xfce 4.16.0 tk: Gtk 3.24.29 info: xfce4-panel wm: xfwm 4.16.1 
  vt: 7 dm: LightDM 1.30.0 Distro: Manjaro Linux base: Arch Linux 
Machine:
  Type: Portable System: Dell product: Precision M6500 v: N/A 
  serial: <filter> Chassis: type: 8 serial: <filter> 
  Mobo: Dell model: 0R1203 serial: <filter> BIOS: Dell v: A10 
  date: 06/04/2013 
Battery:
  ID-1: BAT0 charge: 88.0 Wh (100.0%) condition: 88.0/93.2 Wh (94.4%) 
  volts: 12.3 min: 11.1 model: Samsung SDI DELL 3M19008 type: Li-ion 
  serial: <filter> status: Full 
CPU:
  Info: Quad Core model: Intel Core i7 Q 740 bits: 64 type: MT MCP 
  arch: Nehalem family: 6 model-id: 1E (30) stepping: 5 microcode: A cache: 
  L2: 6 MiB 
  flags: lm nx pae sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx bogomips: 27672 
  Speed: 931 MHz min/max: 933/1734 MHz boost: enabled Core speeds (MHz): 
  1: 931 2: 930 3: 931 4: 931 5: 931 6: 931 7: 931 8: 931 
  Vulnerabilities: Type: itlb_multihit status: KVM: VMX disabled 
  Type: l1tf 
  mitigation: PTE Inversion; VMX: conditional cache flushes, SMT vulnerable 
  Type: mds status: Vulnerable: Clear CPU buffers attempted, no microcode; 
  SMT vulnerable 
  Type: meltdown mitigation: PTI 
  Type: spec_store_bypass 
  mitigation: Speculative Store Bypass disabled via prctl and seccomp 
  Type: spectre_v1 
  mitigation: usercopy/swapgs barriers and __user pointer sanitization 
  Type: spectre_v2 mitigation: Full generic retpoline, IBPB: conditional, 
  IBRS_FW, STIBP: conditional, RSB filling 
  Type: srbds status: Not affected 
  Type: tsx_async_abort status: Not affected 
Graphics:
  Device-1: AMD Broadway XT [Mobility Radeon HD 5870] 
  vendor: Dell FirePro M7820 driver: radeon v: kernel bus-ID: 01:00.0 
  chip-ID: 1002:68a0 class-ID: 0300 
  Device-2: Logic3 / SpectraVideo plc G-720 Keyboard type: USB 
  driver: hid-generic,usbhid bus-ID: 2-1.4.1.1.4.4:24 chip-ID: 1267:0103 
  class-ID: 0300 
  Display: x11 server: X.Org 1.20.13 compositor: xfwm4 v: 4.16.1 driver: 
  loaded: ati,radeon unloaded: modesetting alternate: fbdev,vesa 
  display-ID: :0.0 screens: 1 
  Screen-1: 0 s-res: 1920x1200 s-dpi: 96 s-size: 508x317mm (20.0x12.5") 
  s-diag: 599mm (23.6") 
  Monitor-1: LVDS res: 1920x1200 hz: 60 
  Monitor-2: DisplayPort-0 res: 1920x1200 hz: 60 dpi: 94 
  size: 519x324mm (20.4x12.8") diag: 612mm (24.1") 
  OpenGL: renderer: AMD JUNIPER (DRM 2.50.0 / 5.10.68-1-MANJARO LLVM 12.0.1) 
  v: 3.3 Mesa 21.2.2 compat-v: 3.1 direct render: Yes 
Audio:
  Device-1: Intel 5 Series/3400 Series High Definition Audio vendor: Dell 
  driver: snd_hda_intel v: kernel bus-ID: 00:1b.0 chip-ID: 8086:3b56 
  class-ID: 0403 
  Device-2: AMD Juniper HDMI Audio [Radeon HD 5700 Series] vendor: Dell 
  driver: snd_hda_intel v: kernel bus-ID: 01:00.1 chip-ID: 1002:aa58 
  class-ID: 0403 
  Sound Server-1: ALSA v: k5.10.68-1-MANJARO running: yes 
  Sound Server-2: JACK v: 1.9.19 running: no 
  Sound Server-3: PulseAudio v: 15.0 running: yes 
  Sound Server-4: PipeWire v: 0.3.37 running: no 
Network:
  Device-1: Broadcom NetXtreme BCM5761e Gigabit Ethernet PCIe vendor: Dell 
  driver: tg3 v: kernel port: de00 bus-ID: 09:00.0 chip-ID: 14e4:1680 
  class-ID: 0200 
  IF: eno1 state: up speed: 1000 Mbps duplex: full mac: <filter> 
RAID:
  Hardware-1: Intel 82801 Mobile SATA Controller [RAID mode] driver: ahci 
  v: 3.0 port: 6ea0 bus-ID: 00:1f.2 chip-ID: 8086.282a rev: 05 
  class-ID: 0104 
Drives:
  Local Storage: total: 447.14 GiB used: 32.91 GiB (7.4%) 
  SMART Message: Required tool smartctl not installed. Check --recommends 
  ID-1: /dev/sda maj-min: 8:0 vendor: Intel model: SSDSA2CW160G3 
  size: 149.05 GiB block-size: physical: 512 B logical: 512 B 
  speed: 3.0 Gb/s type: SSD serial: <filter> rev: 0362 scheme: MBR 
  ID-2: /dev/sdb maj-min: 8:16 type: USB model: Mass Storage Device 
  size: 298.09 GiB block-size: physical: 512 B logical: 512 B type: N/A 
  serial: <filter> scheme: MBR 
Partition:
  ID-1: / raw-size: 149.05 GiB size: 145.71 GiB (97.76%) 
  used: 32.91 GiB (22.6%) fs: ext4 dev: /dev/sda1 maj-min: 8:1 
Swap:
  Alert: No swap data was found. 
Sensors:
  System Temperatures: cpu: 51.0 C mobo: N/A sodimm: SODIMM C gpu: radeon 
  temp: 51.0 C 
  Fan Speeds (RPM): cpu: 2635 
Info:
  Processes: 253 Uptime: 1m wakeups: 2 Memory: 31.28 GiB 
  used: 1.01 GiB (3.2%) Init: systemd v: 249 tool: systemctl Compilers: 
  gcc: 11.1.0 Packages: pacman: 1302 lib: 398 flatpak: 0 Shell: Bash 
  v: 5.1.8 running-in: xfce4-terminal inxi: 3.3.06 

thanks

1 Like

After I read your post, I’ve checked what is going on my rig and looks like RAM usage is higher than it should be. I’ve switched kernels and on 5.14 the result is something around 632 and on 5.10 is 596 (I’ve trimmed my setup to something around 300-320 MiB).

Never paid attention to this, but for example, xfce4-power-manager have 3 outputs on 5.14 and 2 on 5.10 kernel in htop.

gnome-keyring-deamon have 4, xfce4-session have 3 and so on.

1 Like

Thanks I’ll give 5.14 a shot later just to see what happens, as others mentioned the RAM usage itself might not be an issue in itself overall. I’ve stuck with LTS kernels only for some time now after some troubles I had way back. Can’t remember what, but I’m running 10-14 year old laptops… once the top models in their class, but now long in the tooth and bought cheap :slight_smile:

I don’t have the knowledge to understand the repercussions of this off the top of my head, but thanks for the tip!

I’ve checked this on Xfce installed on VM, and it looks the same, those are sub-processes (I’m just stupid). However, RAM usage is normal, around 600 MiB (didn’t update VM yet).

I’ve checked 5.4 and 4.19 kernels, and they also have different RAM usage. Lowest is 5.10, at least on my potato.

heh, I misread this and your earlier post, I thought 5.14 was lowest. As I have plenty of RAM and saw a few glitches with 5.14 mentioned in the stable update thread I think I’ll just leave everything as is for now. 5.10 LTS. thanks again!

:smile:

Data posted for the ‘road machine’ shows an uptime of 27m and may not be relevant to RAM use when first booted. But this data also shows only 158 active processes

Data for the ‘long time’ system is showing an uptime of 1m, but is also showing 2 wakeups
and 253 active processes

I rebooted my main system to get more data from free and inxi

               total        used        free      shared  buff/cache   available
Mem:           7.7Gi       536Mi       6.7Gi       4.0Mi       510Mi       7.0Gi
Swap:          3.9Gi          0B       3.9Gi

No packages were updated or changed since previous boot but system reports 2 MiB less than before

  Processes: 222 Uptime: 1m wakeups: 0 Memory: 7.75 GiB used: 853 MiB (10.7%)

@tgz @mr_glitch @nikgnomic

Based on the comments here and my own experimentation, it is clear that tools like htop/task manager understand “ram usage” differently compared to free

After making this post, I downloaded the ISO and did a clean reinstall so that I can compare the differences before and after the latest update. Here is a summary of what I discovered:

  1. After the reinstall, htop reported 680MB initial RAM usage before I ran a system update.
  2. After running the system update, htop reported 1.1GB RAM usage.
  3. The results of the free command were consistent both before and after the system update. free -h reported around 590MiB RAM usage for both cases.

I think the best way to resolve this is to find out what htop considers as “ram usage.” Since free returned the same result before and after the latest stable update, it’s likely that the new update doesn’t use more RAM, just that somehow htop is reporting that it is using more.

1 Like

I think this is the best possible answer based on limited data available

To get a more specific answer would require a lot of testing
User would have to start with an old ISO running v5.10 kernel, as already suggested
system would have to be cold-booted about 10 times to get an average of values from inxi and free
(values for RAM used can vary even when no changes are made to system)
Then system can be updated and repeat tests to get new data values
And for a thorough job the kernel should be updated to v5.14 and repeat tests for the later kernel

But considering that none of the comments here mention the usual consequences of having additional processes loaded and more RAM used (increase in time taken to boot or less responsive desktop) this seems like a lot of effort for not much gain

Used memory is what free reports.
The bar memory counter in htop is not the same.
Add a detailed text memory counter in htop now, and come again.

Waiting for your results.

:popcorn:

1 Like

@omano Yes, that appears to be the case. Before this, I assumed that the memory reported in htop is the same as the one in free. But I was clearly mistaken. I am still rather curious about the difference though, even though that knowledge might not have any practical value other than to save beginners some unnecessary confusion about RAM usage.

@nikgnomic A good way to start is to just find out how htop and task manager define “ram usage.”

After all the comments above, essentially, the gist of this post becomes: “Why does htop’s numbers differ from those of free?”

It helps to understand the meaning behind the numbers htop reports.

And one quick google answered that question.

According to this answer on AskUbuntu:

Main difference is htop counts cached memory and buffers into free memory, whereas free command does not.

1 Like

If you want more to think about, explain me this now lol (this is why I posted my previous reply, the following image is from another thread where I found out this):

https://forum.manjaro.org/uploads/default/original/3X/c/2/c2019389a63f33287c8bf69cb6fb3930dc5f73f8.jpeg

Different tools, different results, most of them not making sense (especially the KDE tools, my conclusion was that they are garbage, and opening tickets on KDE seems to be useless, based on my short experience where they ‘fixed’ another bug I reported regarding the hard drive space used, but it didn’t fix anything, and they got silenced when I tried to have this looked at again so I don’t bother I count on people more motivated than me to get things fixed).

free seems to be reliable, other tools are approximative.

1 Like

And for a more detailed answer about the difference between htop and free, read this:
https://unix.stackexchange.com/questions/65835/htop-reporting-much-higher-memory-usage-than-free-or-top

1 Like