Oleg Vinichenko

downgrades are expected.
put package.mask for dev-util/pkgconfig for now. i'll look for this issue in details.

ego sync does not support this, if you changed branch manually, use git pull and git submodule update inside xorg-kit directory.

http://www.funtoo.org/News:Python_Multiplexing(And_another_Ego_update)

that quote was for not putting python targets entries into make.conf (by anything python related meant). Of course, you can put them, it's allowed, as well as adding any other system-wide settings. by default the make.conf is empty.

right, that's the idea of having automatically setting of python stuff. yes, automated process, which is also gives ability for clean update (or downgrade) of python. For example from 3.4 to 3.6 in one single step. This is step beyond known python switch everyone used to have in gentoo. By a bad fortune i accidentally enabled this feature little bit earlier than it's planned, so my apologies for that. 

and epro update is required to run.

http://www.funtoo.org/News:Kits_Are_Go_(And_Ego_Needs_a_Manual_Bump)

right, you can move the content of /usr/portage/distfiles into the new location /var/cache/portage/distfiles if the download is the concern (tarballs need to be refetched), other wise you shouldn't be much worry about it and delete old directories. eclean will work with the new location, then.

if you use only prime branch, then use "ego sync"
if you have some of kits with other branches enabled (such as master), do not use "ego sync"
to update kits with other branches enabled, use following sequence of commands:
 
cd /var/git/meta-repo
git pull
git submodule update
chown -R portage:portage /var/git/meta-repo
 
eix-sync does not update meta-repo, neither default branches nor any other branches. after manually updating meta-repo or if you use default branches and ego sync, you can use eix-update.

and from my own experience, i have never used stable builds anywhere and there was no really major problems (like disastrous ones), maybe 2-3 only in last 10 years. so, a key is - when you know what to do with stable or unstable - you could expect things to work. if you don't know what to do, likely you'll trap into troubles regardless stable or unstable used. Sometimes updates can be frustrating, i agree. And for newcomers, of course, it's hard as they mostly driven by existing documentation, web, irc -which can be really wrong. The "stable" is what is controlled by matter of letter, in ebuilds, i can put "amd64" instead "~amd64" and it will be considered stable by package manager, which is of course far from real definition of stable. For example, back into not so old case (about 4 monthes), xorg-server-1.19 stabilized by gentoo, due to multiple security vulnerabilities in xorg itself and in it's core parts. xorg-1.19, in turn, found to be very unstable in certain conditions, so the security fixes diminished by the fact you have no working xorg. Those security fixes can be easily, and they are in Funtoo, backported into good known 1.17 version. of course, if 1.19 were ok for the one, he could choose using 1.19. Just trying to describe that stabilization and stable by means of packages, is a volatile term.

and from my own experience, i have never used stable builds anywhere and there was no really major problems (like disastrous ones), maybe 2-3 only in last 10 years. so, a key is - when you know what to do with stable or unstable - you could expect things to work. if you don't know what to do, likely you'll trap into troubles regardless stable or unstable used. Sometimes updates can be frustrating, i agree. And for newcomers, of course, it's hard as they mostly driven by existing documentation, web, irc -which can be really wrong. The "stable" is what is controlled by matter of letter, in ebuilds, i can put "amd64" instead "~amd64" and it will be considered stable by package manager, which is of course far from real definition of stable. For example, back into not so old case (about 4 monthes), xorg-server-1.19 stabilized by gentoo, due to multiple security vulnerabilities in xorg itself and in it's core parts. xorg-1.19, in turn, found to be very unstable in certain conditions, so the security fixes diminished by the fact you have no working xorg. Those security fixes can be easily, and they are in Funtoo, backported into good known 1.17 version. of course, if 1.19 were ok for the one, he could choose using 1.19. Just trying to describe that stabilization and stable by means of packages, is a volatile term.

this link is about testing meta-repo, which will replace portage tree. it has some basic explanation of what kits are. master branches means they are have no direct control by Funtoo and pulling Gentoo ebuilds as they are. using master branch of curated kits, controlled kits, possibly can be provided, but it will not be actively supported.
More details (notice it's old news and ports-2017 is replaced with meta-repo) but it can probably give the idea: 
http://www.funtoo.org/News:New_Ports-2017_tree_and_Kits
It's not possible to give 100% information for now what will be production ready maintaining procedure as meta-repo is still beta. There will be announcement and a guide.
Stabilization of ebuilds is not performed -- if you mean adding stable keywords -- is what actually stable was. 

okey, they are modules. but initramfs which is shipped with debian-sources has no support for booting from NVMe drives. i will fix this.

its not able to run from root, because root is not in portage group by default
to fix this you need to add root into portage group by:
gpasswd -a root portage
and it will work

yes, you need revdep-rebuild command.

there is paludis already and which is even slower than original portage -- it doesn't require python but it does require boost -- just an example. I don't think this will fix anything you maybe think it will by porting to another language. The main problem in portage is not that it's written in python but that many parts of it are obscure, undocumented and doesn't look good, highly over-engineered and maybe also recent portage development only in maintenance mode - not in fixing code or development mode. Many blame portage for being slow, partially agreed with that, but, again it's not because it's in python. Remember that good code is good in any language. To summarize - cloning portage in other language will be only a clone, that's rewriting same framework and design in language that can be really bad experience and can result in much worse cases, like paludis. Ultimate fix, maybe, would be creating package manager from scratch.

BPF stands for Berkeley Packet Filter, with this target enabled, llvm will have specific code for this, which is used in some firewalls, etc. It's enabed for generic purposes, you can disable it. 
NVPTX is rather about nvidia-drivers, not nouveau. It enables specific code needed for CUDA, OpenCl and other code that can be used by gpu chips (nvidia in this particular case). Not sure nouveau does that.

http://www.funtoo.org/News:LLVM_Targets

new release with many fixes:

linux (4.8.15-1) unstable; urgency=medium


New upstream stable update: https://www.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.8.12

[x86] iommu/vt-d: Fix PASID table allocation
[x86] iommu/vt-d: Fix IOMMU lookup for SR-IOV Virtual Functions
[x86] KVM: fix out-of-bounds access in lapic
[x86] KVM: x86: drop error recovery in em_jmp_far and em_ret_far (CVE-2016-9756)
[x86] KVM: fix out-of-bounds accesses of rtc_eoi map (CVE-2016-9777)
[x86] KVM: check for pic and ioapic presence before use
[arm64, armhf] usb: chipidea: move the lock initialization to core file
USB: serial: cp210x: add ID for the Zone DPMX
USB: serial: ftdi_sio: add support for TI CC3200 LaunchPad
scsi: mpt3sas: Fix secure erase premature termination
cfg80211: limit scan results cache size
apparmor: fix change_hat not finding hat after policy replacement
NFSv4.x: hide array-bounds warning
[x86] fpu: Fix invalid FPU ptrace state after execve()
[x86] traps: Ignore high word of regs->cs in early_fixup_exception()
perf/core: Fix address filter parser
perf/x86/intel: Cure bogus unwind from PEBS entries
[x86] thermal/powerclamp: add back module device table
[hppa/parisc] Fix races in parisc_setup_cache_timing()
[hppa/parisc] Switch to generic sched_clock implementation
[hppa/parisc] Fix race in pci-dma.c
[hppa/parisc] Also flush data TLB in flush_icache_page_asm
mpi: Fix NULL ptr dereference in mpi_powm()
X.509: Fix double free in x509_cert_parse()
xc2028: Fix use-after-free bug properly
[powerpc] Set missing wakeup bit in LPCR on POWER9
[powerpc] mm: Fixup kernel read only mapping
[powerpc] boot: Fix the early OPAL console wrappers
can: bcm: fix support for CAN FD frames
mm, oom: stop pre-mature high-order OOM killer invocations
flow_dissect: call init_default_flow_dissectors() earlier
scsi: mpt3sas: Unblock device after controller reset https://www.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.8.13
libata-scsi: Fixup ata_gen_passthru_sense()
scsi: hpsa: use bus '3' for legacy HBA devices
scsi: libfc: fix seconds_since_last_reset miscalculation
mm, thp: propagation of conditional compilation in khugepaged.c
thp: fix corner case of munlock() of PTE-mapped THPs
zram: fix unbalanced idr management at hot removal
mm: fix false-positive WARN_ON() in truncate/invalidate for hugetlb
ovl: fix d_real() for stacked fs
Input: change KEY_DATA from 0x275 to 0x277
Input: psmouse - disable automatic probing of BYD touchpads
rcu: Fix soft lockup for rcu_nocb_kthread
mm: workingset: fix NULL ptr in count_shadow_nodes
PCI: Export pcie_find_root_port
PCI: Set Read Completion Boundary to 128 iff Root Port supports it (_HPX)
mwifiex: printk() overflow with 32-byte SSIDs
[arm64] KVM: vgic: Don't notify EOI for non-SPIs
[x86] drm/i915: Don't touch NULL sg on i915_gem_object_get_pages_gtt() error
[x86] drm/i915: drop the struct_mutex when wedged or trying to reset
[x86] drm/amdgpu: fix power state when port pm is unavailable
drm/radeon: fix power state when port pm is unavailable
[x86] drm/amdgpu: fix check for port PM availability
drm/radeon: fix check for port PM availability
[arm64] dts: juno: fix cluster sleep state entry latency on all SoC versions
KVM: use after free in kvm_ioctl_create_device()
pwm: Fix device reference leak
[x86] perf: Restore TASK_SIZE check on frame pointer
[armhf] clk: sunxi: Fix M factor computation for APB1
batman-adv: Detect missing primaryif during tp_send as error
[arm64] cpufeature: Schedule enable() calls instead of calling them via IPI
[arm64] mm: Set PSTATE.PAN from the cpu_enable_pan() call
[arm64] suspend: Reconfigure PSTATE after resume from idle https://www.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.8.14
gro_cells: mark napi struct as not busy poll candidates
virtio-net: add a missing synchronize_net()
[armhf] net: dsa: b53: Fix VLAN usage and how we treat CPU port
net: check dead netns for peernet2id_alloc()
ip6_tunnel: disable caching when the traffic class is inherited
net: sky2: Fix shutdown crash
af_unix: conditionally use freezable blocking calls in read
rtnetlink: fix FDB size computation
l2tp: fix racy SOCK_ZAPPED flag check in l2tp_ip{,6}_bind()
rtnl: fix the loop index update error in rtnl_dump_ifinfo()
ipv6: bump genid when the IFA_F_TENTATIVE flag is clear
udplite: call proper backlog handlers
[armhf] net: dsa: bcm_sf2: Ensure we re-negotiate EEE during after link change
net, sched: respect rcu grace period on cls destruction
[armhf] net: dsa: fix unbalanced dsa_switch_tree reference counting
net/sched: pedit: make sure that offset is valid
netlink: Call cb->done from a worker thread
netlink: Do not schedule work from sk_destruct
net/dccp: fix use-after-free in dccp_invalid_packet
GSO: Reload iph after pskb_may_pull
packet: fix race condition in packet_set_ring (CVE-2016-8655)
ip6_offload: check segs for NULL in ipv6_gso_segment.
cdc_ether: Fix handling connection notification
tipc: check minimum bearer MTU (CVE-2016-8632)
geneve: avoid use-after-free of skb->data
net: avoid signed overflows for SO_{SND|RCV}BUFFORCE (CVE-2016-9793)
net: ping: check minimum size on ICMP header length (CVE-2016-8399)
ipv4: Restore fib_trie_flush_external function and fix call ordering
ipv4: Fix memory leak in exception case for splitting tries
ipv4: Drop leaf from suffix pull/push functions
ipv4: Drop suffix update from resize code
[sparc64] Fix find_node warning if numa node cannot be found
[sparc64] fix compile warning section mismatch in find_node()
[sparc] Fix inverted invalid_frame_pointer checks on sigreturns
constify iov_iter_count() and iter_is_iovec()
Don't feed anything but regular iovec's to blk_rq_map_user_iov (CVE-2016-9576)
ipv6: Set skb->protocol properly for local output
ipv4: Set skb->protocol properly for local output
Revert: "ip6_tunnel: Update skb->protocol to ETH_P_IPV6 in ip6_tnl_xmit()"
flowcache: Increase threshold for refusing new allocations
esp4: Fix integrity verification when ESN are used
esp6: Fix integrity verification when ESN are used https://www.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.8.15
[powerpc] eeh: Fix deadlock when PE frozen state can't be cleared
[powerpc] mm: Fix lazy icache flush on pre-POWER5
[powerpc] boot: Fix build failure in 32-bit boot wrapper
fuse: fix clearing suid, sgid for chown()
[hppa/parisc] Purge TLB before setting PTE
[hppa/parisc] Remove unnecessary TLB purges from flush_dcache_page_asm and flush_icache_page_asm
[hppa/parisc] Fix TLB related boot crash on SMP machines
zram: restrict add/remove attributes to root only
locking/rtmutex: Prevent dequeue vs. unlock race
locking/rtmutex: Use READ_ONCE() in rt_mutex_owner()
device-dax: fix private mapping restriction, permit read-only
scsi: lpfc: fix oops/BUG in lpfc_sli_ringtxcmpl_put()
sched/autogroup: Fix 64-bit kernel nice level adjustment
[x86] perf: Fix full width counter, counter overflow
acpi, nfit: fix extended status translations for ACPI DSMs
acpi, nfit, libnvdimm: fix / harden ars_status output length handling
acpi, nfit: validate ars_status output buffer size
acpi, nfit: fix bus vs dimm confusion in xlat_status
[armel, armhf] crypto: marvell - Don't copy hash operation twice into the SRAM
crypto: caam - fix pointer size for AArch64 boot loader, AArch32 kernel
[armel, armhf] crypto: marvell - Don't corrupt state of an STD req for re-stepped ahash
can: raw: raw_setsockopt: limit number of can_filter that can be set
can: peak: fix bad memory access and free sequence
[armel] dts: orion5x: fix number of sata port for linkstation ls-gl (Closes: #845611)
ceph: don't set req->r_locked_dir in ceph_d_revalidate
[m68k] Fix ndelay() macro
batman-adv: Check for alloc errors when preparing TT local data
hotplug: Make register and unregister notifier API symmetric
[ Uwe Kleine-K?nig ]



[armhf] dts: armada-385: add support for Turris Omnia

[ Salvatore Bonaccorso ]



Add ABI reference for 4.8.0-2



Ignore ABI changes in KVM



net: handle no dst on skb in icmp6_send (CVE-2016-9919)



[rt] Update to 4.8.11-rt7



[rt] Update to 4.8.14-rt9



netfilter: ipv6: nf_defrag: drop mangled skb on ream error (CVE-2016-9755)



Ignore ABI changes in libnvdimm



docs: sphinx-extensions: make rstFlatTable work with docutils 0.13. Thanks to Dmitry Shachnev mitya57@debian.org (Closes: #848349)

[ Ben Hutchings ]



[amd64] Re-enable LEGACY_VSYSCALL_EMULATE instead of LEGACY_VSYSCALL_NONE. There are still binaries in stable that use vsyscall (via dietlibc). (Closes: #847154)



debian/rules.real: Exclude *.pyc from featureset diffs



debian/control: Fix build-dependency on flex to work with new versions that have M-A: foreign



debian/rules: Use dpkg-parsechangelog -S option to select fields



debian/rules: Tighten binNMU version matching, consistent with linux-latest

Hiya!
All stages that Funtoo/Linux offer now gpg-signed.
Please, follow this instructions on how to proceed.
http://www.funtoo.org/GPG_Signatures

http://www.funtoo.org/News:LLVM_Targets

this is known issue in latest firefox release
https://bugs.gentoo.org/show_bug.cgi?id=600344

Hiya!
gcc-5.3.0 now unmasked on both current and stable builds. Please, follow:
http://www.funtoo.org/News:GCC_Update 

this will not work before updating world as gnome packages that installed will try to merge good portion of gnome dependencies back.

sometime it takes more than 24 hours due to more complex updates. Notice, that Funtoo portage tree updates are having QA tests before updates and this involves manual intervention. This ensures there are no significant problems with updates for regular users. It's not 100% polished but it works.