OpenSUSE's Spectre Mitigation Approach Is One Of The Reasons For Its Slower Performance

[Leopard]

I don't like how performance driven some people think and how also tests on Phoronix are tending a lot toward only performance.

I am missing an explanation between IBRS and Retpoline in the article as much as I missed information about why several distributions were faster or slower in the past.

But it was clear that this kind of "pressure" would be applied on more secure but slower approaches since the default behavior of the Kernel is one of the riskier options.

One shouldn't forget that there is more than just raw performance in terms of results per time. Performance also includes how risky it is to use a certain product for a certain purpose.

Username and post.

Just hilarious.

[milkylainen]

x86 is garbage not useful beyond arch-tied proprietary PC applications, thankfully it is now obsolete and it is dying along with Intel, we will have ARM as de facto arch in Apple macbooks by 2020, the last slice of x86 market share. AMD is OK and x86 in consoles is OK.

You sure you want to get in the line of people who has been speculating the hard demise of x86 since, well yes, its inception?
I haven't seen any believable proof that x86 is going anywhere real soon. Heck, even Intel tried it's best to kill it off. Atleast twice!
Leaving them nothing but a sour taste every time and a burning hole in their wallet.

[hax0r]

so opensuse is full of bullshit too, just like openbsd

I myself can't name a single thing OpenSuse is trying to accomplish, any reason behind why it exists. They could call quits and join efforts of other distros.

On the other hand
Debian.......the standard
Gentoo.......the god, meta based distro, limitless flexbility and possibilities, nothing gets in your way
Ununtu......bringing usefulness of Debian to the mass, decent OOBE
Archlinux...among things, rolling distro, easy to install and easy to stay up to date, however user has no choice of how software was compiled, hence he/she goes back to gentoo
Fedora......among things, integration of bleeding edge software and new innovations, vast amount of time and resources spent in debugging, every distro downstream benfits from Fedora's effort
RHEL........server standard, main value proposition of RHEL is that Red Hat provides ABI stability and backports bug fixes, security and selectively new features
CentOS......err I can't afford support plan
Slackware...KISS, really makes you think, developers could actually spend time developing better software instead of 10s of distros repackaging and fragmenting same software in 10s different ways
TAILS.......escaping botnet

Edit: I don't want to get into OpenBSD, Theo de Raadt was kicked out of NetBSD for a good reason. NetBSD is the best BSD.

[Hugh]

I myself can't name a single thing OpenSuse is trying to accomplish, any reason behind why it exists. They could call quits and join efforts of other distros.

I don't use SuSE but I respect it. It's kind of like Red Hat but doesn't quite have the same heft. Competition is good!

We especially need it now that IBM owns Red Hat. I haven't seen any damage (good!) but that may not be how things evolve.

[ms178]

You sure you want to get in the line of people who has been speculating the hard demise of x86 since, well yes, its inception?
I haven't seen any believable proof that x86 is going anywhere real soon. Heck, even Intel tried it's best to kill it off. Atleast twice!
Leaving them nothing but a sour taste every time and a burning hole in their wallet.

True, but let's wait and see how this battle of the ISAs and their implementations will turn out. Intel assembled a great team to fight back. But they are still vulnerable. Apple, Microsoft, Google are all investing into the ARM notebook / desktop ecosystem. The hurdles are still high for success though. It all comes down to native software support and value for us consumers. If the latter is not compelling enough, people won't buy it. I wonder if betting on Qualcomm alone is the right strategy here, I would hope that other ARM vendors will join that race as well (e.g. Nvidia and Samsung could pull it off from a technological standpoint) to have some competition.

[Hugh]

From https://lists.opensuse.org/opensuse-.../msg00144.html :

retpolines are not complete protection on skylake+. Coffee lake has
EIBRS which should restore the performance a bit. Perhaps, one day,
Intel will add EIBRS support also to Skylake, if possible (I don't
remember the details).

[dremon_nl]

I myself can't name a single thing OpenSuse is trying to accomplish, any reason behind why it exists. They could call quits and join efforts of other distros.

On the other hand
Debian.......the standard
Gentoo.......the god, meta based distro, limitless flexbility and possibilities, nothing gets in your way
Ununtu......bringing usefulness of Debian to the mass, decent OOBE
Archlinux...among things, rolling distro, easy to install and easy to stay up to date, however user has no choice of how software was compiled, hence he/she goes back to gentoo
Fedora......among things, integration of bleeding edge software and new innovations, vast amount of time and resources spent in debugging, every distro downstream benfits from Fedora's effort
RHEL........server standard, main value proposition of RHEL is that Red Hat provides ABI stability and backports bug fixes, security and selectively new features
CentOS......err I can't afford support plan
Slackware...KISS, really makes you think, developers could actually spend time developing better software instead of 10s of distros repackaging and fragmenting same software in 10s different ways
TAILS.......escaping botnet

Edit: I don't want to get into OpenBSD, Theo de Raadt was kicked out of NetBSD for a good reason. NetBSD is the best BSD.

openSUSE has two flavors: Leap for the enterprise-level stability for free (same sources as their SLES product) and Tumbleweed for tested and stable rolling release. Nothing from your list provides that under the same umbrella. Tumbleweed also provides the best rolling experience with Plasma desktop.

[andyprough]

I myself can't name a single thing OpenSuse is trying to accomplish, any reason behind why it exists. They could call quits and join efforts of other distros.

Well, if you had bothered to look you would find a large number of pretty unique things openSUSE is doing. All that you've let us know with this post is that you didn't bother to look, and that you have no intention to educate yourself further.

[andyprough]

The kernels I compiled are using Mitigation: Full generic retpoline .

The other changes I made from the stock OpenSUSE kernel config to match what I typically run:

OpenSUSE Leap default value --> changed value

CONFIG_HZ_250=y --> CONFIG_HZ_1000=y
CONFIG_NO_HZ_IDLE=y --> CONFIG_NO_HZ_FULL=y
CONFIG_PREEMPT_NONE=y --> CONFIG_PREEMPT_VOLUNTARY=y
# CONFIG_SCHED_AUTOGROUP is not set --> CONFIG_SCHED_AUTOGROUP=y



With these changes, I haven't seen any perceivable performance loss running OpenSUSE as my primary desktop and on some servers.

I also run a Laptop with openSUSE Tumbleweed, a custom Kernel config and with the Spectre/Meltdown mitigations disabled. I'd also turn off debug symbols and change the CPU governor to performance and use blk-mq as I/O scheduler.

Good points I missed to call out! -- I use blk-mq (standard now with a 5.0.x kernel), set schedutil as the governor, and set CONFIG_DEBUG_KERNEL=n .

These are excellent points, thank you both for posting, I'm going to try out these changes.

[starshipeleven]

I myself can't name a single thing OpenSuse is trying to accomplish, any reason behind why it exists.

Considering that the list you posted is bullshit (Debian a standard? Arch is easy to install?, Slackware what? just to cite the most important points ), I'm not really surprised.