Re: Linux 6.16-rc2

[Guenter Roeck]

On Sun, Jun 15, 2025 at 02:06:11PM -0700, Linus Torvalds wrote:
> Pretty quiet week, with a pretty small rc2 as a result. That's not
> uncommon, and things tend to pick up at rc3, but this is admittedly
> even smaller than usual.
> 
> It may be that people are taking a breather after a fairly sizable
> merge window, but it might also be seasonal, with Europe starting to
> see summer vacations... We'll see how this goes.
> 
> The diffstat looks somewhat unusual, with a lot of one-liners with
> both ARC and pincontrol having (presumably independently) ended up
> doing some unrelated trivial cleanups.
> 
> But even that is probably noticeable only because everything else is
> pretty small. That "everything else" is mostly network drivers (and
> bluetooth) and bcachefs, with some rust infrastructure and core
> networking changes thrown in.
> 
> And, as always, just random smaller stuff all over. Shortlog appended
> for people who like to see the details.
> 

Build results:
	total: 159 pass: 159 fail: 0
Qemu test results:
	total: 636 pass: 631 fail: 5
Failed tests:
	arm:supermicro-x11spi-bmc:aspeed_g5_defconfig:mtd32,0,6,1:net=nic:aspeed-bmc-supermicro-x11spi:sqf
	arm:ast2600-evb:aspeed_g5_defconfig:mtd64,0,6,1:net=nic:aspeed-ast2600-evb:ext2
	arm:fuji-bmc:aspeed_g5_defconfig:mem1G:mtd128,0,8,1:net=nic:aspeed-bmc-facebook-fuji:f2fs
	arm:g220a-bmc:aspeed_g5_defconfig:mtd32,0,12,2:net=nic:aspeed-bmc-bytedance-g220a:sqf
	arm:qcom-dc-scm-v1-bmc:aspeed_g5_defconfig:mtd64,0,12,2:net=nic:aspeed-bmc-qcom-dc-scm-v1:ext2
Unit test results:
	pass: 595297 fail: 1
Failed unit tests:
	x86_64:q35:iov_kunit_extract_pages_xarray
	x86_64:q35:iov_iter

Pretty much the same as last week. A revert for the mtd problem has been
submitted [1], so the arm test failures should hopefully be resolved soon.

There is also a compiler version dependent build failure seen when building
xtensa:allmodconfig.

tensa-linux-ld: drivers/clocksource/timer-tegra186.o: in function `tegra186_timer_remove':
timer-tegra186.c:(.text+0x350): undefined reference to `__udivdi3'

Some compiler versions don't translate
	(((u64)wdt->base.timeout * USEC_PER_SEC) / 5)
into a simple multiplication by (USEC_PER_SEC/5) but take the code
literally and multiply first, then divide. I have seen this with gcc 13.3
and 13.4. It is not seen with gcc 14.3 (which is why I missed it last week).
I submitted a fix [2].

The iov unit test failure is:

     # iov_kunit_extract_pages_xarray: ASSERTION FAILED at lib/tests/kunit_iov_iter.c:63
     Expected got == npages, but
         got == 1 (0x1)
         npages == 256 (0x100)
     # iov_kunit_extract_pages_xarray: pass:0 fail:1 skip:0 total:1
     not ok 12 iov_kunit_extract_pages_xarray
 # iov_iter: pass:11 fail:1 skip:0 total:12
 # Totals: pass:11 fail:1 skip:0 total:12
 not ok 26 iov_iter

This is not new; it is seen sometimes if PREEMPT_RT is enabled. The problem
is that alloc_pages_bulk() may return 1 if it can not allocate all pages,
and I guess that PREEMPT_RT randomly triggers that situation.

Guenter

---
[1] https://patchwork-proxy.ozlabs.org/project/linux-mtd/patch/20250613180722.1268159-1-miquel.raynal@xxxxxxxxxxx/
[2] https://lore.kernel.org/lkml/20250614175556.922159-1-linux@xxxxxxxxxxxx/