linux-next: build failure after merge of the hyperv tree

linux-next: build failure after merge of the hyperv tree

[broonie]

Hi all,

After merging the hyperv tree, today's linux-next build (x86 allmodconfig)
failed like this:

/tmp/next/build/arch/x86/kernel/x86_init.c:36:12: error: static declaration of 'set_rtc_noop' follows non-static declaration
   36 | static int set_rtc_noop(const struct timespec64 *now) { return -EINVAL; }
      |            ^~~~~~~~~~~~
In file included from /tmp/next/build/arch/x86/include/asm/mpspec.h:7,
                 from /tmp/next/build/arch/x86/include/asm/topology.h:38,
                 from /tmp/next/build/include/linux/topology.h:36,
                 from /tmp/next/build/include/linux/gfp.h:8,
                 from /tmp/next/build/include/linux/xarray.h:15,
                 from /tmp/next/build/include/linux/radix-tree.h:21,
                 from /tmp/next/build/include/linux/idr.h:15,
                 from /tmp/next/build/include/linux/kernfs.h:12,
                 from /tmp/next/build/include/linux/sysfs.h:16,
                 from /tmp/next/build/include/linux/kobject.h:20,
                 from /tmp/next/build/include/linux/pci.h:35,
                 from /tmp/next/build/arch/x86/kernel/x86_init.c:9:
/tmp/next/build/arch/x86/include/asm/x86_init.h:333:12: note: previous declaration of 'set_rtc_noop' with type 'int(const struct timespec64 *)'
  333 | extern int set_rtc_noop(const struct timespec64 *now);
      |            ^~~~~~~~~~~~
/tmp/next/build/arch/x86/kernel/x86_init.c:37:13: error: static declaration of 'get_rtc_noop' follows non-static declaration
   37 | static void get_rtc_noop(struct timespec64 *now) { }
      |             ^~~~~~~~~~~~
In file included from /tmp/next/build/arch/x86/include/asm/mpspec.h:7,
                 from /tmp/next/build/arch/x86/include/asm/topology.h:38,
                 from /tmp/next/build/include/linux/topology.h:36,
                 from /tmp/next/build/include/linux/gfp.h:8,
                 from /tmp/next/build/include/linux/xarray.h:15,
                 from /tmp/next/build/include/linux/radix-tree.h:21,
                 from /tmp/next/build/include/linux/idr.h:15,
                 from /tmp/next/build/include/linux/kernfs.h:12,
                 from /tmp/next/build/include/linux/sysfs.h:16,
                 from /tmp/next/build/include/linux/kobject.h:20,
                 from /tmp/next/build/include/linux/pci.h:35,
                 from /tmp/next/build/arch/x86/kernel/x86_init.c:9:
/tmp/next/build/arch/x86/include/asm/x86_init.h:334:13: note: previous declaration of 'get_rtc_noop' with type 'void(struct timespec64 *)'
  334 | extern void get_rtc_noop(struct timespec64 *now);
      |             ^~~~~~~~~~~~

Caused by commit

  25dcc7316ef7def25c ("x86/init: Make get/set_rtc_noop() public")

I will use the hyperv tree from yesterday instead.

Re: linux-next: build failure after merge of the hyperv tree

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 3057 bytes --]

On Tue, Apr 18, 2023 at 06:00:21PM +0100, broonie@kernel.org wrote:
> Hi all,
> 
> After merging the hyperv tree, today's linux-next build (x86 allmodconfig)
> failed like this:

Adding more Hyper-V CCs, sorry the capitialisation in MAINTAINERS
confused my search.

> 
> /tmp/next/build/arch/x86/kernel/x86_init.c:36:12: error: static declaration of 'set_rtc_noop' follows non-static declaration
>    36 | static int set_rtc_noop(const struct timespec64 *now) { return -EINVAL; }
>       |            ^~~~~~~~~~~~
> In file included from /tmp/next/build/arch/x86/include/asm/mpspec.h:7,
>                  from /tmp/next/build/arch/x86/include/asm/topology.h:38,
>                  from /tmp/next/build/include/linux/topology.h:36,
>                  from /tmp/next/build/include/linux/gfp.h:8,
>                  from /tmp/next/build/include/linux/xarray.h:15,
>                  from /tmp/next/build/include/linux/radix-tree.h:21,
>                  from /tmp/next/build/include/linux/idr.h:15,
>                  from /tmp/next/build/include/linux/kernfs.h:12,
>                  from /tmp/next/build/include/linux/sysfs.h:16,
>                  from /tmp/next/build/include/linux/kobject.h:20,
>                  from /tmp/next/build/include/linux/pci.h:35,
>                  from /tmp/next/build/arch/x86/kernel/x86_init.c:9:
> /tmp/next/build/arch/x86/include/asm/x86_init.h:333:12: note: previous declaration of 'set_rtc_noop' with type 'int(const struct timespec64 *)'
>   333 | extern int set_rtc_noop(const struct timespec64 *now);
>       |            ^~~~~~~~~~~~
> /tmp/next/build/arch/x86/kernel/x86_init.c:37:13: error: static declaration of 'get_rtc_noop' follows non-static declaration
>    37 | static void get_rtc_noop(struct timespec64 *now) { }
>       |             ^~~~~~~~~~~~
> In file included from /tmp/next/build/arch/x86/include/asm/mpspec.h:7,
>                  from /tmp/next/build/arch/x86/include/asm/topology.h:38,
>                  from /tmp/next/build/include/linux/topology.h:36,
>                  from /tmp/next/build/include/linux/gfp.h:8,
>                  from /tmp/next/build/include/linux/xarray.h:15,
>                  from /tmp/next/build/include/linux/radix-tree.h:21,
>                  from /tmp/next/build/include/linux/idr.h:15,
>                  from /tmp/next/build/include/linux/kernfs.h:12,
>                  from /tmp/next/build/include/linux/sysfs.h:16,
>                  from /tmp/next/build/include/linux/kobject.h:20,
>                  from /tmp/next/build/include/linux/pci.h:35,
>                  from /tmp/next/build/arch/x86/kernel/x86_init.c:9:
> /tmp/next/build/arch/x86/include/asm/x86_init.h:334:13: note: previous declaration of 'get_rtc_noop' with type 'void(struct timespec64 *)'
>   334 | extern void get_rtc_noop(struct timespec64 *now);
>       |             ^~~~~~~~~~~~
> 
> Caused by commit
> 
>   25dcc7316ef7def25c ("x86/init: Make get/set_rtc_noop() public")
> 
> I will use the hyperv tree from yesterday instead.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: build failure after merge of the hyperv tree

[Wei Liu]

On Tue, Apr 18, 2023 at 06:06:58PM +0100, Mark Brown wrote:
> On Tue, Apr 18, 2023 at 06:00:21PM +0100, broonie@kernel.org wrote:
> > Hi all,
> > 
> > After merging the hyperv tree, today's linux-next build (x86 allmodconfig)
> > failed like this:
> 
> Adding more Hyper-V CCs, sorry the capitialisation in MAINTAINERS
> confused my search.

I will fix this. Thanks for the heads up and sorry for the breakage.

Wei.

Re: linux-next: build failure after merge of the hyperv tree

[Wei Liu]

On Tue, Mar 12, 2024 at 03:18:07PM +1100, Stephen Rothwell wrote:
> Hi all,
> 
> After merging the hyperv tree, today's linux-next build (i386 defconfig)
> failed like this:
> 
> arch/x86/kernel/cpu/mshyperv.c:355:5: error: no previous prototype for 'hv_get_hypervisor_version' [-Werror=missing-prototypes]
>   355 | int hv_get_hypervisor_version(union hv_hypervisor_version_info *info)
>       |     ^~~~~~~~~~~~~~~~~~~~~~~~~
> cc1: all warnings being treated as errors

We noticed this as well.  A fix has been sent out. I will update the tree
shortly.

Thanks,
Wei.

> 
> Caused by commit
> 
>   1634df06ea12 ("mshyperv: Introduce hv_get_hypervisor_version function")
> 
> I have reverted that commit for today.
> 
> -- 
> Cheers,
> Stephen Rothwell

linux-next: build failure after merge of the hyperv tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 569 bytes --]

Hi all,

After merging the hyperv tree, today's linux-next build (i386 defconfig)
failed like this:

arch/x86/kernel/cpu/mshyperv.c:355:5: error: no previous prototype for 'hv_get_hypervisor_version' [-Werror=missing-prototypes]
  355 | int hv_get_hypervisor_version(union hv_hypervisor_version_info *info)
      |     ^~~~~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors

Caused by commit

  1634df06ea12 ("mshyperv: Introduce hv_get_hypervisor_version function")

I have reverted that commit for today.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: build failure after merge of the hyperv tree

[Dexuan-Linux Cui]

On Sun, Aug 6, 2023 at 10:39 PM Wei Liu <wei.liu@kernel.org> wrote:
>
> On Mon, Aug 07, 2023 at 02:22:38PM +1000, Stephen Rothwell wrote:
> > Hi all,
> >
> > After merging the hyperv tree, today's linux-next build (x86_64
> > allmodconfig) failed like this:
> >
> > drivers/pci/controller/pci-hyperv.o: warning: objtool: .altinstr_replacement+0x0: weirdly overlapping alternative! 11 != 6
> > drivers/hv/hv_balloon.o: warning: objtool: .altinstr_replacement+0x0: weirdly overlapping alternative! 11 != 6
> > drivers/hv/hv_vmbus.o: warning: objtool: .altinstr_replacement+0x3: weirdly overlapping alternative! 10 != 5
> > vmlinux.o: warning: objtool: .altinstr_replacement+0x633: weirdly overlapping alternative! 11 != 6
> > incomplete ORC unwind tables in file: vmlinux
> > Failed to sort kernel tables
> >
> > I have no idea what caused this.

Hi Stephen,
My kernel source was checked out from the Hyper-V tree's hyperv-next
branch last Friday, which had Tianyu's 8 patches.

I'm able to repro the same build failure in a Ubuntu 22.04 VM with
gcc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4:

I run the below commands:
make mrproper
make allmodconfig
make -j128
make -j128

...
  LD [M]  net/qrtr/qrtr-tun.ko
  LD [M]  net/qrtr/qrtr-mhi.ko
  LD [M]  virt/lib/irqbypass.ko
  LD [M]  arch/x86/video/fbdev.ko
make: *** [Makefile:234: __sub-make] Error 2

make -j128
  DESCEND objtool
  INSTALL libsubcmd_headers
  CALL    scripts/checksyscalls.sh
  CHK     kernel/kheaders_data.tar.xz
  UPD     include/generated/utsversion.h
  CC      init/version-timestamp.o
  LD      .tmp_vmlinux.kallsyms1
  NM      .tmp_vmlinux.kallsyms1.syms
  KSYMS   .tmp_vmlinux.kallsyms1.S
  AS      .tmp_vmlinux.kallsyms1.S
  LD      .tmp_vmlinux.kallsyms2
  NM      .tmp_vmlinux.kallsyms2.syms
  KSYMS   .tmp_vmlinux.kallsyms2.S
  AS      .tmp_vmlinux.kallsyms2.S
  LD      vmlinux
  NM      System.map
  SORTTAB vmlinux
incomplete ORC unwind tables in file: vmlinux
Failed to sort kernel tables
make[2]: *** [scripts/Makefile.vmlinux:36: vmlinux] Error 1
make[2]: *** Deleting file 'vmlinux'
make[1]: *** [/opt/linux-next/Makefile:1250: vmlinux] Error 2
make[1]: *** Waiting for unfinished jobs....
make: *** [Makefile:234: __sub-make] Error 2

If I revert Tianyu's patches by "git diff v6.5-rc4 | patch -Rp1",
the build can succeed.

If I apply the the patches, I get the same failure:

drivers/hv/hv_balloon.o: warning: objtool: .altinstr_replacement+0x0:
weirdly overlapping alternative! 11 != 6
...
drivers/pci/controller/pci-hyperv.o: warning: objtool:
.altinstr_replacement+0x0: weirdly overlapping alternative! 11 != 6
...
drivers/hv/hv_vmbus.o: warning: objtool: .altinstr_replacement+0x3:
weirdly overlapping alternative! 10 != 5
...
  AR      built-in.a
  AR      vmlinux.a
  LD      vmlinux.o
vmlinux.o: warning: objtool: .altinstr_replacement+0x633: weirdly
overlapping alternative! 11 != 6
  OBJCOPY modules.builtin.modinfo
  GEN     modules.builtin
  GEN     .vmlinux.objs
  MODPOST Module.symvers
  CC      .vmlinux.export.o
  UPD     include/generated/utsversion.h
  CC      init/version-timestamp.o
  LD      .tmp_vmlinux.kallsyms1
  NM      .tmp_vmlinux.kallsyms1.syms
  KSYMS   .tmp_vmlinux.kallsyms1.S
  AS      .tmp_vmlinux.kallsyms1.S
  LD      .tmp_vmlinux.kallsyms2
  NM      .tmp_vmlinux.kallsyms2.syms
  KSYMS   .tmp_vmlinux.kallsyms2.S
  AS      .tmp_vmlinux.kallsyms2.S
  LD      vmlinux
  NM      System.map
  SORTTAB vmlinux
incomplete ORC unwind tables in file: vmlinux
Failed to sort kernel tables
make[2]: *** [scripts/Makefile.vmlinux:36: vmlinux] Error 1
make[2]: *** Deleting file 'vmlinux'
make[1]: *** [/opt/linux-next/Makefile:1250: vmlinux] Error 2
make[1]: *** Waiting for unfinished jobs....
...
make: *** [Makefile:234: __sub-make] Error 2

It turns out that I already made a fix
https://github.com/dcui/linux/commit/c4db45f6256248435b2a303b264ecbb41320c41d
which can fix the build failure and fix a crash issue (see
https://lwn.net/ml/linux-kernel/SA1PR21MB1335A21D5D037FE88D9CF820BF0EA@SA1PR21MB1335.namprd21.prod.outlook.com/
)

Thanks,
Dexuan

> /me scratches head.
>
> > I have used the hyperv tree from next-20230804 for today.
>
> I've reverted that series from my tree. Thanks for the heads up.
>
> Thanks,
> Wei.

Re: linux-next: build failure after merge of the hyperv tree

[Wei Liu]

On Mon, Aug 07, 2023 at 02:22:38PM +1000, Stephen Rothwell wrote:
> Hi all,
> 
> After merging the hyperv tree, today's linux-next build (x86_64
> allmodconfig) failed like this:
> 
> drivers/pci/controller/pci-hyperv.o: warning: objtool: .altinstr_replacement+0x0: weirdly overlapping alternative! 11 != 6
> drivers/hv/hv_balloon.o: warning: objtool: .altinstr_replacement+0x0: weirdly overlapping alternative! 11 != 6
> drivers/hv/hv_vmbus.o: warning: objtool: .altinstr_replacement+0x3: weirdly overlapping alternative! 10 != 5
> vmlinux.o: warning: objtool: .altinstr_replacement+0x633: weirdly overlapping alternative! 11 != 6
> incomplete ORC unwind tables in file: vmlinux
> Failed to sort kernel tables
> 
> I have no idea what caused this.
> 

/me scratches head.

> I have used the hyperv tree from next-20230804 for today.

I've reverted that series from my tree. Thanks for the heads up.

Thanks,
Wei.

linux-next: build failure after merge of the hyperv tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 765 bytes --]

Hi all,

After merging the hyperv tree, today's linux-next build (x86_64
allmodconfig) failed like this:

drivers/pci/controller/pci-hyperv.o: warning: objtool: .altinstr_replacement+0x0: weirdly overlapping alternative! 11 != 6
drivers/hv/hv_balloon.o: warning: objtool: .altinstr_replacement+0x0: weirdly overlapping alternative! 11 != 6
drivers/hv/hv_vmbus.o: warning: objtool: .altinstr_replacement+0x3: weirdly overlapping alternative! 10 != 5
vmlinux.o: warning: objtool: .altinstr_replacement+0x633: weirdly overlapping alternative! 11 != 6
incomplete ORC unwind tables in file: vmlinux
Failed to sort kernel tables

I have no idea what caused this.

I have used the hyperv tree from next-20230804 for today.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: build failure after merge of the hyperv tree

[Wei Liu]

On Tue, Apr 18, 2023 at 06:31:57PM +0100, broonie@kernel.org wrote:
> Hi all,
> 
> After merging the hyperv tree, today's linux-next build (x86 allmodconfig)
> failed like this:
> 
> /tmp/next/build/drivers/hv/connection.c: In function 'vmbus_connect':
> /tmp/next/build/drivers/hv/connection.c:228:53: error: implicit
> declaration of function 'hv_alloc_hyperv_page'; did you mean 'hv_free_hyperv_page'? [-Werror=implicit-function-declaration]
>   228 |         vmbus_connection.monitor_pages[0] = (void *)hv_alloc_hyperv_page();
>       |                                                     ^~~~~~~~~~~~~~~~~~~~
>       |                                                     hv_free_hyperv_page
> /tmp/next/build/drivers/hv/connection.c:228:45: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast]
>   228 |         vmbus_connection.monitor_pages[0] = (void *)hv_alloc_hyperv_page();
>       |                                             ^
> /tmp/next/build/drivers/hv/connection.c:229:45: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast]
>   229 |         vmbus_connection.monitor_pages[1] = (void *)hv_alloc_hyperv_page();
>       |                                             ^
> cc1: all warnings being treated as errors
> 
> Caused by commit
> 
>   4dbdcfe1c5db80edca ("Drivers: hv: move panic report code from vmbus to hv early init code")
> 
> I will use the hyperv tree from yesterday instead.

This has been fixed by removing the problematic commit.

Thanks,
Wei.

linux-next: build failure after merge of the hyperv tree

[broonie]

Hi all,

After merging the hyperv tree, today's linux-next build (x86 allmodconfig)
failed like this:

/tmp/next/build/drivers/hv/connection.c: In function 'vmbus_connect':
/tmp/next/build/drivers/hv/connection.c:228:53: error: implicit
declaration of function 'hv_alloc_hyperv_page'; did you mean 'hv_free_hyperv_page'? [-Werror=implicit-function-declaration]
  228 |         vmbus_connection.monitor_pages[0] = (void *)hv_alloc_hyperv_page();
      |                                                     ^~~~~~~~~~~~~~~~~~~~
      |                                                     hv_free_hyperv_page
/tmp/next/build/drivers/hv/connection.c:228:45: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast]
  228 |         vmbus_connection.monitor_pages[0] = (void *)hv_alloc_hyperv_page();
      |                                             ^
/tmp/next/build/drivers/hv/connection.c:229:45: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast]
  229 |         vmbus_connection.monitor_pages[1] = (void *)hv_alloc_hyperv_page();
      |                                             ^
cc1: all warnings being treated as errors

Caused by commit

  4dbdcfe1c5db80edca ("Drivers: hv: move panic report code from vmbus to hv early init code")

I will use the hyperv tree from yesterday instead.

linux-next: build failure after merge of the hyperv tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1730 bytes --]

Hi all,

After merging the hyperv tree, today's linux-next build (x86_64
allmodconfig) failed like this:

arch/x86/kvm/hyperv.c: In function 'kvm_vcpu_ioctl_get_hv_cpuid':
arch/x86/kvm/hyperv.c:2020:16: error: 'HV_X64_DEBUGGING' undeclared (first use in this function); did you mean 'HV_DEBUGGING'?
 2020 |    ent->ebx |= HV_X64_DEBUGGING;
      |                ^~~~~~~~~~~~~~~~
      |                HV_DEBUGGING
arch/x86/kvm/hyperv.c:2020:16: note: each undeclared identifier is reported only once for each function it appears in

Caused by commit

  c55a844f46f9 ("x86/hyperv: Split hyperv-tlfs.h into arch dependent and independent files")

interacting with commit

  f97f5a56f597 ("x86/kvm/hyper-v: Add support for synthetic debugger interface")

from the kvm tree.

I have applied this patch for today:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 2 Jun 2020 17:31:06 +1000
Subject: [PATCH] x86/hyperv: merge fix for HV_X64_DEBUGGING name change

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 arch/x86/kvm/hyperv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/kvm/hyperv.c b/arch/x86/kvm/hyperv.c
index abde638548e0..af9cdb426dd2 100644
--- a/arch/x86/kvm/hyperv.c
+++ b/arch/x86/kvm/hyperv.c
@@ -2017,7 +2017,7 @@ int kvm_vcpu_ioctl_get_hv_cpuid(struct kvm_vcpu *vcpu, struct kvm_cpuid2 *cpuid,
 			ent->edx |= HV_FEATURE_FREQUENCY_MSRS_AVAILABLE;
 			ent->edx |= HV_FEATURE_GUEST_CRASH_MSR_AVAILABLE;
 
-			ent->ebx |= HV_X64_DEBUGGING;
+			ent->ebx |= HV_DEBUGGING;
 			ent->edx |= HV_X64_GUEST_DEBUGGING_AVAILABLE;
 			ent->edx |= HV_FEATURE_DEBUG_MSRS_AVAILABLE;
 
-- 
2.26.2

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: linux-next: build failure after merge of the hyperv tree

[Wei Liu]

On Tue, Mar 31, 2020 at 05:23:35PM +1100, Stephen Rothwell wrote:
> Hi all,
> 
> After merging the hyperv tree, today's linux-next build (x86_64
> allmodconfig) failed like this:
> 
> ERROR: modpost: "panic_on_oops" [drivers/hv/hv_vmbus.ko] undefined!
> 
> Caused by commit
> 
>   a6a5aa4b0179 ("x86/Hyper-V: Report crash data in die() when panic_on_oops is set")
> 
> I have reverted that commit for today.

I will remove that patch from hyperv-next. I have been busy with other
things on Monday and Tuesday, so sorry for not paying close attention.

Wei.

> 
> -- 
> Cheers,
> Stephen Rothwell

linux-next: build failure after merge of the hyperv tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 364 bytes --]

Hi all,

After merging the hyperv tree, today's linux-next build (x86_64
allmodconfig) failed like this:

ERROR: modpost: "panic_on_oops" [drivers/hv/hv_vmbus.ko] undefined!

Caused by commit

  a6a5aa4b0179 ("x86/Hyper-V: Report crash data in die() when panic_on_oops is set")

I have reverted that commit for today.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]