Re: Linux 6.13-rc1

[Guenter Roeck]

On Sun, Dec 01, 2024 at 02:47:21PM -0800, Linus Torvalds wrote:
> So two weeks have passed, the merge window is over, and -rc1 is pushed out.
> 

Build results:
	total: 161 pass: 155 fail: 6
Failed builds:
	csky:allmodconfig
	i386:allyesconfig
	i386:allmodconfig
	openrisc:allmodconfig
	s390:allmodconfig
	xtensa:allmodconfig
Qemu test results:
	total: 544 pass: 528 fail: 16
Failed tests:
	arm:imx25-pdk:imx_v4_v5_defconfig:nonand:mem128:net=default:imx25-pdk:initrd
	arm:imx25-pdk:imx_v4_v5_defconfig:nonand:sd:mem128:net=default:imx25-pdk:ext2
	arm:imx25-pdk:imx_v4_v5_defconfig:nonand:usb0:mem128:net=default:imx25-pdk:ext2
	arm:imx25-pdk:imx_v4_v5_defconfig:nonand:usb1:mem128:net=default:imx25-pdk:ext2
	arm:mcimx6ul-evk:imx_v6_v7_defconfig:nodrm:mem256:net=nic:net=nic:imx6ul-14x14-evk:initrd
	arm:mcimx6ul-evk:imx_v6_v7_defconfig:nodrm:sd:mem256:net=nic:net=nic:imx6ul-14x14-evk:ext2
	arm:mcimx6ul-evk:imx_v6_v7_defconfig:nodrm:usb0:mem256:net=nic:net=nic:imx6ul-14x14-evk:ext2
	arm:mcimx6ul-evk:imx_v6_v7_defconfig:nodrm:usb1:mem256:net=nic:net=nic:imx6ul-14x14-evk:ext2
	arm:mcimx7d-sabre:imx_v6_v7_defconfig:nodrm:smp2:mem256:net=nic:imx7d-sdb:initrd
	arm:mcimx7d-sabre:imx_v6_v7_defconfig:nodrm:smp2:usb1:mem256:net=nic:imx7d-sdb:ext2
	arm:mcimx7d-sabre:imx_v6_v7_defconfig:nodrm:smp2:sd:mem256:net=nic:imx7d-sdb:ext2
	arm:sabrelite:imx_v6_v7_defconfig:nodrm:mem256:net=default:imx6dl-sabrelite:initrd
	arm:sabrelite:imx_v6_v7_defconfig:nodrm:mmc,b300:mem256:net=default:imx6dl-sabrelite:ext2
	arm:sabrelite:imx_v6_v7_defconfig:nodrm:usb0:mem256:net=default:imx6dl-sabrelite:ext2
	arm:sabrelite:imx_v6_v7_defconfig:nodrm:usb1:mem256:net=default:imx6dl-sabrelite:ext2
	openrisc:or1200:or1ksim_defconfig
Unit test results:
	pass: 464016 fail: 0

There are also spurious test failures on kudo-bmc as well as various
warning backtraces. I'll list the warnings as well as the known fixes
below. The fixes are available in my fixes branch at
git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging.git.

Note that I did not Cc: anyone; I hope I'll get to that next week.

Guenter

---
Known fixes. Please see
git://git.kernel.org/pub/scm/linux/kernel/git/groeck/linux-staging.git.
branch 'fixes' for details. All test builds pass with these fixes applied.

1eaa86ae9501 staging: gpib: Hack to fix x86 32-bit builds
34c5b4ea36f1 locking/lockdep: Enforce PROVE_RAW_LOCK_NESTING only if ARCH_SUPPORTS_RT
38fed97ad3db openrisc: place exception table at the head of vmlinux
fcbb9c518f5d staging: gpib: Make GPIB_NI_PCI_ISA depend on HAS_IOPORT
2622ab2d4ad4 timekeeping: Fix negative motion detection
aad00fb31380 ARM: imx: Re-introduce the PINCTRL selection

---
Remaining backtraces after above fixes have been applied. Unless mentioned,
the problems have been reported, but there is no fix known to me. I will
likely have missed some backtraces since there are so many.

========================================================
WARNING: possible irq lock inversion dependency detected
6.13.0-rc1-00045-ga7f3166624f6 #1 Tainted: G                 N
--------------------------------------------------------
ip/2616 just changed the state of lock:
ffffb03ce50b0d88 (napi_hash_lock){+...}-{3:3}, at: napi_disable+0x1a0/0x1b8

Seen during shutdown when the pcnet32 driver is loaded.
Caused by commit 86e25f40aa1e ("net: napi: Add napi_config").

===
WARNING: CPU: 0 PID: 0 at drivers/tty/serial/8250/8250_port.c:470 serial8250_set_defaults+0x148/0x1d8
Unsupported UART type 0

Seen with various architectures. Caused (or triggered) by commit
7c7e6c8924e7 ("tty: serial: handle HAS_IOPORT dependencies").

===
WARNING: CPU: 0 PID: 0 at include/trace/events/preemptirq.h:36 handle_int+0x11c/0x16c
RCU not watching for tracepoint

WARNING: CPU: 0 PID: 0 at include/trace/events/preemptirq.h:40 trace_hardirqs_on+0x184/0x228
RCU not watching for tracepoint

Seen with MIPS boot tests. Not analyzed.

===
WARNING: CPU: 0 PID: 0 at kernel/smp.c:807 smp_call_function_many_cond+0x4fc/0xa38
Modules linked in:
CPU: 0 UID: 0 PID: 0 Comm: swapper/0 Not tainted 6.13.0-rc1-00045-ga7f3166624f6 #1
Hardware name: MPC8544DS e500v2 0x80210030 MPC8544 DS
...
Call Trace:
[c268fe30] [c07aee40] __SCT__lsm_static_call_capable_0+0x0/0x20 (unreliable)
[c268fe90] [24002288] 0x24002288
[c268fea0] [c002355c] flush_tlb_kernel_range+0x2c/0x50
[c268feb0] [c0024b6c] patch_mem.constprop.0+0x108/0x1b0
[c268fed0] [c00198ac] arch_static_call_transform+0x10c/0x150
[c268fef0] [c2036ba0] security_add_hooks+0x138/0x24c
[c268ff20] [c20359a4] capability_init+0x24/0x38
[c268ff30] [c2035eb4] initialize_lsm+0x48/0x90
[c268ff50] [c2036794] security_init+0x2e0/0x5b4
[c268ffa0] [c20010ec] start_kernel+0x5d4/0x81c
[c268fff0] [c0000478] set_ivor+0x150/0x18c

Seen on powerpc. This was actually introduced with 6.12. Not analyzed.

===
[    0.107470][    T0] WARNING: suspicious RCU usage
[    0.107597][    T0] 6.13.0-rc1-00045-ga7f3166624f6 #1 Not tainted
[    0.107728][    T0] -----------------------------
[    0.107803][    T0] kernel/notifier.c:586 notify_die called but RCU thinks we're quiescent!
...
[    0.115756][    T0] WARNING: suspicious RCU usage
[    0.115832][    T0] 6.13.0-rc1-00045-ga7f3166624f6 #1 Not tainted
[    0.115923][    T0] -----------------------------
[    0.115996][    T0] include/linux/rcupdate.h:850 rcu_read_lock() used illegally while idle!

and more. Seen with ppc64 boot tests. Not analyzed.