Comments

219 Comments

Those builds are for kernel test week, and not destined for F32 updates quite yet.

Right, with the logs you have posted (in almost every kernel update recently) this does not look kernel related. You will need to open up a bug in bugzilla either way. If it were kernel, we don't do bug tracking here. When you post here, and I don't see anything kernel related, I don't worry about it and move on. If it were in bugzilla, it could be assigned to the correct package, and all of the history goes with it.

What's the bug for that, and does it have logs showing this as kernel?

I think you can just give a +1 to cancel it out. So the general rule is, if it is a regression from what is currently on that branch, it would get negative karma. If it is not a new regression it would not. Particularly with kernel, we move quickly, and it may take a few updates before a bug is fixed, but denying an update because an older bug is not fixed just delays fixes for others. Especially as we are on a pre-beta test branch, on a pre-release kernel, the changeset posted each week contains hundreds of bug fixes.

These are not regressions from rc3

Right, I have brought in the patch bbca083de291, but the other needs a backport. I had expected Chris to do the backport, but it hasn't shown yet.

Are you recycling Phoronix or actually hitting an issue?

For the broken audio, what version of alsa-firmware is installed?

5.4.5-301 should have made testing, but no, 5.4.7 is due tomorrow and that will make testing

Paxtest failures are unfortunately common since spectre/meltdown fixes went in, I need to make it skip that test now.

"kernel-tools-5.4.5-200.fc31" is correct, in fact kerrnel-headers and kernel should have been 200 as well, this is the first rebase after F29 EOL, but they were already built. 5.4.6 kernel and headers will be 200.

The 301 build should have a fix for the too many files issue.

BZ#1758239 CVE-2019-17052 kernel: privilege escalation in ax25_create in net/ax25/af_ax25.c in the AF_AX25 network module
BZ#1758240 CVE-2019-17052 kernel: priviledge escalation in ax25_create in net/ax25/af_ax25.c in the AF_AX25 network module [fedora-all]
BZ#1758242 CVE-2019-17053 kernel: privilege escalation in ieee802154_create in net/ieee802154/socket.c in the AF_IEEE802154 network module
BZ#1758243 CVE-2019-17053 kernel: priviledge escalation in ieee802154_create in net/ieee802154/socket.c in the AF_IEEE802154 network module [fedora-all]
BZ#1758245 CVE-2019-17054 kernel: privilege escalation in atalk_create in net/appletalk/ddp.c in the AF_APPLETALK network module
BZ#1758246 CVE-2019-17054 kernel: privilege escalation in atalk_create in net/appletalk/ddp.c in the AF_APPLETALK network module [fedora-all]
BZ#1758248 CVE-2019-17055 kernel: privilege escalation in base_sock_create in drivers/isdn/mISDN/socket.c in the AF_ISDN network module
BZ#1758257 CVE-2019-17056 kernel: privilege escalation in llcp_sock_create in net/nfc/llcp_sock.c in the AF_NFC network module [fedora-all]

This update has been unpushed.

Sorry to edit your updates, but we need this to go out with the kernels for MDS mitigation.

/sys/devices/system/cpu/vulnerabilities/mds:Mitigation: Clear CPU buffers; SMT disabled

/sys/devices/system/cpu/vulnerabilities/mds:Mitigation: Clear CPU buffers; SMT disabled

/sys/devices/system/cpu/vulnerabilities/mds:Mitigation: Clear CPU buffers; SMT disabled

@mjg not particularly. CVEs come out frequently, and so do stable releases. it wasn't worth doing a special build for that one yesterday when 4.20.11 is out today.