Whaddaya Know? Update and It Goes Faster

[joncr]

Hey! I think my machine got faster after this morning's updates. Booting is definitely quicker. Cool.

'Course, it could be my imagination. Or, maybe the debug code is gone. Or, maybe the 3.7.2 kernel is a speedster. I dunno.

C'est le vie.

To my eyes, this is looking like a very good piece of work. Thanks and congrats to all who bang away on it. Have a nice day tomorrow.

[DBelton]

the kernel debug code had been disabled for quite awhile now, but other applications may still have had debug code enabled.

If you noticed an increase in performance it probably isn't your imagination, though. I haven't updated today, yet, so I can't tell what caused the boost you are seeing.

[nonamedotc]

Yup! Kernel-3.7.2 has been very good for me as well (nvidia, virtualbox, etc all work very well) - I am very happy about that!

[lsatenstein]

Is the update you guys are describing the RC4 version that is going to be the release version?

[nonamedotc]

No. This is the kernel version from updates-testing repo. It would probably in the updates after release.

[joncr]

Originally Posted by nonamedotc

No. This is the kernel version from updates-testing repo. It would probably in the updates after release.

Think it's in updates now. I don't have updates-testing enabled.

[nonamedotc]

Oops! Sorry. I have updates-testing enabled and guess I wasn't paying attention.

[expectATIon]

With kernel 3.7.2-201 my system crashes after about 5 minutes with kernel panic

[mikee]

Re F18 being faster....
I noticed that /tmp will be a "tmpfs" by default. That means it's a memory based filesystem rather than
disk, so that should speed up a lot of things!

https://fedoraproject.org/wiki/Features/tmp-on-tmpfs

[joncr]

Originally Posted by expectATIon

With kernel 3.7.2-201 my system crashes after about 5 minutes with kernel panic

Nvida card? If so, maybe do another update. I noticed RPMFusion posted a kmod-nvidia update that didn't come down when I got the updated kernel.

[AdamW]

Kernel 3.7 could improve speed in some cases on some hardware vs 3.6, I wouldn't expect it to make a really noticeable general difference though. Hard to know whether it's the kernel, something else, or your imagination without benchmarks and hardware details, really.

[lsatenstein]

Hi Adam
Reading all the kernal postings from LT, he mentioned the big gain is with btfrs.

Here is some ideas about improvements based on all updates testings applied to RC4
F17 on Netbook -- 4:02 hrs / minutes on batteries
F18 4:23 hrs / minutes on batteries.

263/242 * 100=9.5% improvement in performance.

[joncr]

Originally Posted by AdamW

Kernel 3.7 could improve speed in some cases on some hardware vs 3.6, I wouldn't expect it to make a really noticeable general difference though. Hard to know whether it's the kernel, something else, or your imagination without benchmarks and hardware details, really.

My imagination is often a wondrous thing.

Pretty sure, even without measurements, that the boot is faster, though.

Seriously, Adam, this is a slick release. Hope the release goes well and flack-free.

[Dutchy]

Haven't noticed performance improvements yet.
What I did notice is that now I can't re-enable nmi_watchdog because it will crash the kernel and my radeon still has memory and or corruption errors.
For the rest it looks pretty solid.