Re: Linux 6.14-rc1

[Guenter Roeck]

On Sun, Feb 02, 2025 at 04:26:53PM -0800, Linus Torvalds wrote:
> Two weeks have passed, and the merge window is over.
> 
...
> 
> Let's hope that a small release also ends up meaning smooth sailing
> during the stabilization phase. That's obviously guaranteed, but
> fingers crossed...
> 

Build results:
	total: 161 pass: 156 fail: 5
Failed builds:
	i386:allyesconfig
	i386:allmodconfig
	parisc:allmodconfig
	sparc64:allmodconfig
	sparc64:defconfig
Qemu test results:
	total: 544 pass: 517 fail: 27
Failed tests:
	arm:npcm750-evb:multi_v7_defconfig:npcm:net=nic,npcm-gmac:nuvoton-npcm750-evb:initrd
	arm:npcm750-evb:multi_v7_defconfig:npcm:mtd32,6,5:net=nic,npcm-gmac:nuvoton-npcm750-evb:sqf
	arm:npcm750-evb:multi_v7_defconfig:npcm:usb0.1:net=nic,npcm-gmac:nuvoton-npcm750-evb:ext2
	arm:quanta-gsj:multi_v7_defconfig:npcm:net=nic,npcm-gmac:nuvoton-npcm730-gsj:initrd
	arm:quanta-gsj:multi_v7_defconfig:npcm:mtd32:net=nic,npcm-gmac:nuvoton-npcm730-gsj:ext2
	arm:quanta-gsj:multi_v7_defconfig:npcm:usb0.1:net=nic,npcm-gmac:nuvoton-npcm730-gsj:ext2
	arm:kudo-bmc:multi_v7_defconfig:npcm:net=nic,npcm-gmac:nuvoton-npcm730-kudo:initrd
	arm:kudo-bmc:multi_v7_defconfig:npcm:mtd64,8,3:net=nic,npcm-gmac:nuvoton-npcm730-kudo:cramfs
	arm:kudo-bmc:multi_v7_defconfig:npcm:usb0.1:net=nic,npcm-gmac:nuvoton-npcm730-kudo:ext2
	sheb:rts7751r2dplus_defconfig:ext2
	sparc64:sun4u:smp:net=default:initrd
	sparc64:sun4u:smp:ata:net=rtl8139:hd
	sparc64:sun4u:smp:ata:net=e1000:cd
	sparc64:sun4u:smp:ata:net=e1000-82544gc:hd
	sparc64:sun4u:smp:sdhci-mmc:net=rtl8139:hd
	sparc64:sun4u:smp:nvme:net=tulip:hd
	sparc64:sun4u:smp:scsi[DC395]:net=i82559c:cd
	sparc64:sun4u:smp:scsi[MEGASAS]:net=i82559a:hd
	sparc64:sun4u:smp:scsi[AM53C974]:net=usb-net:hd
	sparc64:sun4u:smp:usb-xhci:net=virtio-net-pci:hd
	sparc64:sun4u:smp:usb-uas-xhci:net=i82801:hd
	sparc64:sun4u:smp:virtio-pci:net=i82559er:hd
	sparc64:sun4v:smp:ata:net=i82562:hd
	sparc64:sun4v:smp:ata:net=e1000-82545em:cd
	sparc64:sun4v:smp:nvme:net=default:hd
	sparc64:sun4u:nosmp:ata:net=e1000:hd
	sparc64:sun4v:nosmp:ata:net=pcnet:hd
Unit test results:
	pass: 472629 fail: 39

Most failures are known with fixes submitted. Patches with more details
are available in the 'fixes' branch at
git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging.git.

950432b79a5d net: stmmac: Allow zero for [tr]x_fifo_size
	Fixes arm boot failures
9d7a89e83e06 dmaengine: tegra210-adma: Fix build error due to 64-by-32 division
	Fixes i386 build failures
eadf076115eb drm/i915/backlight: Return immediately when scale() finds invalid parameters
	Fixes parisc build failure

The sparc64 build failures are due to commit 1e1a9cecfab3 ("block: force
noio scope in blk_mq_freeze_queue"), which changed some APIs but not all
drivers using it.

The sheb boot failure is a problem in the sh4eb emulation, seen when
booting from flash. I just mention it here for reference and will likely
stop testing that going forward.

There is also a parisc64 boot failure seen if KFENCE and SLUB_DEBUG are
enabled. That is also an emulation bug, mentioned here for reference.

The failed unit tests are

	m68k:q800:stackinit

The patches introducing the failures are:

e71a29db79da stackinit: Add union initialization to selftests
ad9f265c7328 stackinit: Add old-style zero-init syntax to struct tests

Similar to 1e1a9cecfab3, those are all last minute patches, so I did
not have time to analyze the problems.

Guenter