Pale Moon forum

After the Meltdown patches are applied, clearly the system is going to see some performance drop due to more context switches and TLB flushes. This drop can be upto 30% for some workloads where program is making a lot of system calls. I think Pale Moon also falls in such a category as it does heavy network (and file) IO which is delegated to the OS. So, does this mean that after the patches, Pale Moon is going to get significantly slower? Am I missing something here?

By the time firmware patches are installed (maybe only for the more recent series?), probable revised OS patches, plus a likely series of patches for 'net facing software/VM's etc., there's most likely going to be impact(s) somewhere, we really won't know the full extent for months and even then, some of the later patches/revisions might recover some of the performance loss created by the early 'knee-jerk' patches.

From the other side of the fence: My Windows (v7 x64) has been patched, and I see no discernible slowdown, in Pale Moon or elsewhere.

Likewise, my testing/benchmarking comparisons/gaming/normal usage hasn't shown up any performance hits since patching my own W7x64 on Saturday am.

I did get some long waits while trying to login to online gaming servers however, both on the patched W7 and an unpatched one, possibly indicating that the servers had also undergone patching. All back to normal when I tested ~10 hours ago, maybe they've tweaked the servers now.

I likewise see no slow down on my system. I am using an Intel processor and I found this concerning AMD Athlon X2, Opteron, and Turion model processors.

https://www.ghacks.net/2018/01/08/fix-w ... kb4056894/

This is probably why they didn't publish the patch yet -- it may not have had proper testing on old-old hardware.

The slow down is going to happen to the cloud computing industry more so than anything. Cloudflare, Amazon, Azure, etc. Maybe even a HTPC or a NAS. But 30% is hardly anything to worry about if you have those.

Where did this magic number of 30% come from anyway? I've read it's up to 30%. That's not a number set in stone.

Not only do you need a Windows patch, but you also need a BIOS/UEFI update as well to mitigate this flaw.

Techspot has performed two rounds of testing, but the BIOS/hardware available to test with is pretty limited.

https://www.techspot.com/article/1554-m ... e-windows/

https://www.techspot.com/article/1556-m ... e-windows/

Certain SSD benchmarks take a 40% or more hit, but again, very small pool to work with here. Most games, office software, etc sees very little impact.

For the record, our sync server which sees the most consistent heavy loads in terms of I/O and CPU hasn't shown any signs of performance loss. And yes, our VPS provider was on the ball getting that patched asap.

Be prepared for new patches soon, Nvidia are preparing some:

NVIDIA's response to speculative side channels CVE-2017-5753, CVE-2017-5715, and CVE-2017-5754

For their main GPU drivers:

The vulnerability has three known variants (my bold):

Variant 1 (CVE-2017-5753): Mitigations are provided with the security update included in this bulletin. NVIDIA expects to work together with its ecosystem partners on future updates to further strengthen mitigations.
Variant 2 (CVE-2017-5715): NVIDIA’s initial analysis indicates that the NVIDIA GPU Display Driver is potentially affected by this variant. NVIDIA expects to work together with its ecosystem partners on future updates for this variant.
Variant 3 (CVE-2017-5754): At this time, NVIDIA has no reason to believe that the NVIDIA GPU Display Driver is vulnerable to this variant.

It says, "local user access ." Does that mean like an old maid approach or what?

John connor wrote:It says, "local user access ." Does that mean like an old maid approach or what?

Not sure what you mean with "old maid approach" but if it requires local user access, it won't be exploitable remotely and needs someone with physical access to your machine or potentially a remote desktop with local user login (in which case you tend to have bigger issues).

Oops! It''s Evil Maid. HAHAHA

https://www.schneier.com/blog/archives/ ... attac.html


Yeah, local access means game over anyway. In the case of true crypt, if they already have physical access to your machine then it's game over. This applies to any encryption scheme. I use true crypt since it's been audited, despite the few flaws. But with this CPU vulnerability crap and now this, one has to wonder. Hopefully Sandboxie doesn't get breached, or Pale Moon for that matter. But like clock work, there seems to be a knew CVE every flipping day.

Patch installed on a win 8.1 (x64) HP machine running an Intel core i5 5200u.

No noticeable slowdown on this machine (as of yet).

On one of my main sites, Deviantart (which always imposes a load on the browser) there has always been a noticeable delay of a second or two when showing the main user menus. This has always been the case using Firefox and when I migrated to Palemoon, browsing felt in general as if it had received a turbo upgrade. As an example the menu on Deviantart showed immediately. The day after the upgrade, the delay on the menu returned and in general it felt as if the turbo upgrade had been switched off.

Core2duos definitely seem slower, I have two 2.5ghz laptops that received the patch from MS and prior to the patch they were surprisingly nimble machines that made me question why I ever needed to upgrade. Palemoon flew! Since the upgrade they are noticeably slower, with add-ons or in safe mode. They are still very usable, it is just that snappy 'edge' that seems to have gone.