Anyone know when this will make it to testing ... the first build in koji (f12) was jan 4th - the last was today 2.6.32.3-27 ...
Good to take the later snapshot (2.6.32.3) but are there things holding back moving this to testing ?
thanks
gene/
On Mon, Jan 18, 2010 at 14:21:44 -0500, Mail Lists lists@sapience.com wrote:
Anyone know when this will make it to testing ... the first build in koji (f12) was jan 4th - the last was today 2.6.32.3-27 ...
Good to take the later snapshot (2.6.32.3) but are there things holding back moving this to testing ?
I tried out a couple of the 2.6.32 kernels on f12 and they seemed to work. If you just want to play, give it a try.
On 01/18/2010 04:07 PM, Bruno Wolff III wrote:
I tried out a couple of the 2.6.32 kernels on f12 and they seemed to work. If you just want to play, give it a try.
It fixes the Gnome log-in delay when compiz is enabled on nVidia hardware. In fact, 2.6.32 is overall faster to the naked eye.
On Tue, Jan 19, 2010 at 7:10 AM, Michael Cronenworth mike@cchtml.com wrote:
On 01/18/2010 04:07 PM, Bruno Wolff III wrote:
I tried out a couple of the 2.6.32 kernels on f12 and they seemed to work. If you just want to play, give it a try.
It fixes the Gnome log-in delay when compiz is enabled on nVidia hardware.
No it shouldn't this was an issue with the xserver; an updated xserver fixed it not the kernel.
On Mon, 2010-01-18 at 14:21 -0500, Mail Lists wrote:
Anyone know when this will make it to testing ... the first build in koji (f12) was jan 4th - the last was today 2.6.32.3-27 ...
Good to take the later snapshot (2.6.32.3) but are there things holding back moving this to testing ?
Updating a kernel version is a major and complex task and often takes quite a while. Just having a completed build is often a long way from even being in good enough state to go to -testing; there may be many patches that need porting to the new release or significant regressions that need to be addressed. There have been cases where it's taken months. So this isn't unusual and don't, necessarily, hold your breath :). You can always try the latest version from Koji and kill it if doesn't work, though.
On 01/19/2010 12:22 AM, Adam Williamson wrote:
Updating a kernel version is a major and complex task and often takes quite a while. Just having a completed build is often a long way from even being in good enough state to go to -testing; there may be many patches that need porting to the new release or significant regressions that need to be addressed. There have been cases where it's taken months. So this isn't unusual and don't, necessarily, hold your breath :). You can always try the latest version from Koji and kill it if doesn't work, though.
I do realize this, and good comments. There are some great improvements int this kernel - for one should finally fix the slow_path bug in iwl4965 wireless triggered by lot of data (eg backups) going through the wireless .. there are a lot of other good things too!
I will try the koji kernel ...
Thanks.
gene
On 01/19/2010 10:13 PM, Mail Lists wrote:
On 01/19/2010 12:22 AM, Adam Williamson wrote:
Updating a kernel version is a major and complex task and often takes quite a while. Just having a completed build is often a long way from even being in good enough state to go to -testing; there may be many patches that need porting to the new release or significant regressions that need to be addressed. There have been cases where it's taken months. So this isn't unusual and don't, necessarily, hold your breath :). You can always try the latest version from Koji and kill it if doesn't work, though.
I do realize this, and good comments. There are some great improvements int this kernel - for one should finally fix the slow_path bug in iwl4965 wireless triggered by lot of data (eg backups) going through the wireless .. there are a lot of other good things too!
I will try the koji kernel ...
I tried the 2.6.32.6 koji build - there were 4 warnings about missing things (firmware) for drivers I was not using (aic9xxx, qlaxxx, mmm forgoet what else) - i did add ati-firmware which I did find (even tho im using nvidia).
There was warning about nouveau during the rpm -i which I am using - but I crossed my fingers and it seems to work.
iwl4965 wifi is improved substantially (speed for N is once again visible) - the last test is the warn slowpath bug which I am pretty sure is fixed in this kernel too - it triggers when lots of packets go thru - so I will run a backup which usually triggers the bug to see.
So a few small tidy ups but so far working well for me (64 bit).
gene
On Thu, 2010-01-28 at 20:11 -0500, Mail Lists wrote:
I tried the 2.6.32.6 koji build - there were 4 warnings about missing things (firmware) for drivers I was not using (aic9xxx, qlaxxx, mmm forgoet what else)
This is normal for all kernels and can be ignored.
- i did add ati-firmware which I did find (even tho
im using nvidia).
There was warning about nouveau during the rpm -i which I am using - but I crossed my fingers and it seems to work.
Dunno what that was, more details?
On 01/29/2010 01:07 PM, Adam Williamson wrote:
There was warning about nouveau during the rpm -i which I am using - but I crossed my fingers and it seems to work.
Dunno what that was, more details?
If I recall correctly it was something like :
W: .... (some complaint firmware missing maybe) nouveau/????.ctxval W: ... (similar complainr) nouveau/ ?? ??.ctxprog
where ?? i dont recall but may have contained nv or something.
This happened when I did the rpm -i on kernel itself after updating other packages appropriately.
gene
On 01/29/2010 01:42 PM, Mail Lists wrote:
On 01/29/2010 01:07 PM, Adam Williamson wrote:
There was warning about nouveau during the rpm -i which I am using - but I crossed my fingers and it seems to work.
Dunno what that was, more details?
If I recall correctly it was something like :
W: .... (some complaint firmware missing maybe) nouveau/????.ctxval W: ... (similar complainr) nouveau/ ?? ??.ctxprog
where ?? i dont recall but may have contained nv or something.
This happened when I did the rpm -i on kernel itself after updating other packages appropriately.
gene
This problem seems to have been fixed in the 2.6.32.7 build.
gene/
-
Adam Williamson -
Bruno Wolff III -
drago01 -
Genes MailLists -
Michael Cronenworth