Hi all, guys!

Been working on the release in the weekend. The snapshot it is built against is 2009-04-03 evening. I'm aware of no critical feature update since then.
There are some improvements: some visible, some other invisible.

About the former:

• xfsprogs has been updated to version 3.0.0 + xfsdump has been included
• btrfs support has been included as well (with btrfs-progs)
• ext4 support reaches maturity: marked stable upstream, works flawlessly. Compatibility mode has been disabled, since there should be no use for it (no static hardcoded "ext4dev" entries in the liveCD libs)
• madwifi-ng does not compile: at first, it seemed like an ebuild issue: updating the checks for CONFIG_MODULE instead of CONFIG_KMOD should have done the trick (and it did). In the end it showed problems compiling, which are due to internal kernel structures changes/renames. Since I'm no madwifi developer, instead of collecting patches here and there and ending up with a "might work" driver, I chose to "suspend" it. Until no official update becomes available through gentoo or the sf.net repository, expect no update. When it becomes available, I may opt for a "bugfix" release instead of rebuilding everything from scratch. Note that ath5k is getting better and better, if you find out your card doesn't work try the in-kernel one before asking a revbump for madwifi-ng :)
• unfortunately, there's no truecrypt-6.1a ebuild available. Tried bumping, compiling, but it complains about missing pkcs11.h. The included version is 6.0a. Might get updated on the next revision.

About the latter: since kernel includes squashfs now, the underlying filesystem has been migrated to 4.x, built with squashfs-tools-4.0_pre20090324. This is because otherwise the CD would fail to boot with mount errors. From the performance perspective you should expect almost no gains and benefits. It's just an underground change. Building with squashfs-tools-3.3 and squashfs-tools4.0_pre showed almost 1:1 size matching between the images. Maybe there's some optimization when decompressing related to in-kernel routines being used, I can't tell that. Anyway, booting process doesn't seem to show any speed changes. To me it looks like squashfs 3.x code was merged and renamed to 4.x, nothing more, nothing less. Just wanted to tell =)

Enjoy the new release and have fun! ;)

Cheers,
Fabio Scaccabarozzi

Hi all!

Just a quick update to notify about the status of the works.
2.6.29 kernel has been released last week, and reiser4 patches are finally available. I've still got three main reasons, though, to delay a bit more this release:

• if you look at the the kernel Changelog, the git one, you'll find out that in the last week a huge amount of fixes have been merged concerning the networking, like entire new schemas for handling packets/connection, not to mention important updates to the hardware drivers and still some others. So, hoping it'll be the next week, I'm still delaying to wait for 2.6.29.1 to come out
• I'm stumbling upon some places were information about gcc and glibc release engineering should be diffused, so to know if I may end up missing some critical feature for just some days' advance
• I've got very busy schedules at university, which allow me to spend only 1 hour per day on my PC, that means I don't know how fast I will be able to work on the medias, especially the part concerning testing and troubleshooting. If 2.6.29.1 is out before Saturday, expect a release timed for Sunday evening. Else, for the week after (yes, we're probably going for an April release anyway).

Apart from the reasons above, I think I'll be releasing the stages first with some advance from the CDs, if I find no critical toolchain update is to be published before Saturday.

Cheers,
Fabio Scaccabarozzi

Hi all! It's been quite long since last post.

Anyway, exams at university have ended, and I finally have some time to build the new release. Lately my main concerns were headed to gcc-4.3.3, which had to be re-emerged twice on my currently working systems to function properly. Luckily everything went fine, and the errors I got when trying to build stage2 went away. For this release, I'm going to wait for 2.6.29 to come out and have the "official" reiser4 patches available (see http://www.kernel.org/pub/linux/kernel/people/edward/), so that we'll have a brand new shining kernel. Following the kernel release times, hopefully 2.6.29 will be out for about the 2nd week of March, allowing me to meet the published release schedule. Major features will be linux-headers-2.6.28-r1, glibc-2.9_p20081201-r2 and gcc-4.3.3.

As for the stages release, judging from the actual amount of data downloaded from the server, there seems to be no need for one-month updates (afterall in 3 months we hardly have a "core" feature update). If that's the case, I'll rebuild the stages.

I'm also planning to implement two wordpress blogs on subdomains of faskatech.net, where I will transfer my actual italian blog and open a new english one (mainly about gentoo experiences), so that I'll centralize everything.

Stay tuned ;)

Cheers,
Fabio Scaccabarozzi