Re: [PATCH] headers: untangle kmemleak.h from mm.h

[Randy Dunlap]

On 02/11/2018 11:27 PM, Ingo Molnar wrote:
> 
> * Randy Dunlap  wrote:
> 
>> From: Randy Dunlap 
>>
>> Currently  #includes  for no obvious
>> reason. It looks like it's only a convenience, so remove kmemleak.h
>> from slab.h and add  to any users of kmemleak_*
>> that don't already #include it.
>> Also remove  from source files that do not use it.
>>
>> This is tested on i386 allmodconfig and x86_64 allmodconfig. It
>> would be good to run it through the 0day bot for other $ARCHes.
>> I have neither the horsepower nor the storage space for the other
>> $ARCHes.
>>
>> [slab.h is the second most used header file after module.h; kernel.h
>> is right there with slab.h. There could be some minor error in the
>> counting due to some #includes having comments after them and I
>> didn't combine all of those.]
>>
>> This is Lingchi patch #1 (death by a thousand cuts, applied to kernel
>> header files).
>>
>> Signed-off-by: Randy Dunlap 
> 
> Nice find:
> 
> Reviewed-by: Ingo Molnar 
> 
> I agree that it needs to go through 0-day to find any hidden dependencies we 
> might 
> have grown due to this.

Andrew,

This patch has mostly survived both 0day and ozlabs multi-arch testing with
2 build errors being reported by both of them.  I have posted patches for
those separately. (and are attached here)

other-patch-1:
lkml.kernel.org/r/5664ced1-a0cd-7e4e-71b6-9c3a97d68...@infradead.org
"lib/test_firmware: add header file to prevent build errors"

other-patch-2:
lkml.kernel.org/r/b3b7eebb-0e9f-f175-94a8-379c5ddca...@infradead.org
"integrity/security: fix digsig.c build error"

Will you see that these are merged or do you want me to repost them?

thanks,
-- 
~Randy
From: Randy Dunlap 

security/integrity/digsig.c has build errors on some $ARCH due to a
missing header file, so add it.

  security/integrity/digsig.c:146:2: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]

Reported-by: Michael Ellerman 
Signed-off-by: Randy Dunlap 
Cc: Mimi Zohar 
Cc: linux-integr...@vger.kernel.org
Link: http://kisskb.ellerman.id.au/kisskb/head/13396/
---
 security/integrity/digsig.c |1 +
 1 file changed, 1 insertion(+)

--- lnx-416-rc1.orig/security/integrity/digsig.c
+++ lnx-416-rc1/security/integrity/digsig.c
@@ -18,6 +18,7 @@
 #include 
 #include 
 #include 
+#include 
 #include 
 #include 
 



From: Randy Dunlap 

lib/test_firmware.c has build errors on some $ARCH due to a
missing header file, so add it.

  lib/test_firmware.c:134:2: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]
  lib/test_firmware.c:620:25: error: implicit declaration of function 'vzalloc' [-Werror=implicit-function-declaration]

Reported-by: Michael Ellerman 
Signed-off-by: Randy Dunlap 
Cc: Wei Yongjun 
Cc: Luis R. Rodriguez 
Cc: Greg Kroah-Hartman 
Link: http://kisskb.ellerman.id.au/kisskb/head/13396/
---
 lib/test_firmware.c |1 +
 1 file changed, 1 insertion(+)

--- lnx-416-rc1.orig/lib/test_firmware.c
+++ lnx-416-rc1/lib/test_firmware.c
@@ -21,6 +21,7 @@
 #include 
 #include 
 #include 
+#include 
 
 #define TEST_FIRMWARE_NAME	"test-firmware.bin"
 #define TEST_FIRMWARE_NUM_REQS	4

Re: [PATCH] headers: untangle kmemleak.h from mm.h

[Randy Dunlap]

On 02/13/2018 02:09 AM, Michael Ellerman wrote:
> Randy Dunlap  writes:
> 
>> On 02/12/2018 04:28 AM, Michael Ellerman wrote:
>>> Randy Dunlap  writes:
>>>
 From: Randy Dunlap 

 Currently  #includes  for no obvious
 reason. It looks like it's only a convenience, so remove kmemleak.h
 from slab.h and add  to any users of kmemleak_*
 that don't already #include it.
 Also remove  from source files that do not use it.

 This is tested on i386 allmodconfig and x86_64 allmodconfig. It
 would be good to run it through the 0day bot for other $ARCHes.
 I have neither the horsepower nor the storage space for the other
 $ARCHes.

 [slab.h is the second most used header file after module.h; kernel.h
 is right there with slab.h. There could be some minor error in the
 counting due to some #includes having comments after them and I
 didn't combine all of those.]

 This is Lingchi patch #1 (death by a thousand cuts, applied to kernel
 header files).

 Signed-off-by: Randy Dunlap 
>>>
>>> I threw it at a random selection of configs and so far the only failures
>>> I'm seeing are:
>>>
>>>   lib/test_firmware.c:134:2: error: implicit declaration of function 
>>> 'vfree' [-Werror=implicit-function-declaration] 
>>> 
>>>  
>>>   lib/test_firmware.c:620:25: error: implicit declaration of function 
>>> 'vzalloc' [-Werror=implicit-function-declaration]
>>>   lib/test_firmware.c:620:2: error: implicit declaration of function 
>>> 'vzalloc' [-Werror=implicit-function-declaration]
>>>   security/integrity/digsig.c:146:2: error: implicit declaration of 
>>> function 'vfree' [-Werror=implicit-function-declaration]
>>
>> Both of those source files need to #include .
> 
> Yep, I added those and rebuilt. I don't see any more failures that look
> related to your patch.

Great, thanks.

I also sent patches for both of those.

>   http://kisskb.ellerman.id.au/kisskb/head/13399/
> 
> 
> I haven't gone through the defconfigs I have enabled for a while, so
> it's possible I have some missing but it's still a reasonable cross
> section.


-- 
~Randy

Re: [PATCH] headers: untangle kmemleak.h from mm.h

[Michael Ellerman]

Randy Dunlap  writes:

> On 02/12/2018 04:28 AM, Michael Ellerman wrote:
>> Randy Dunlap  writes:
>> 
>>> From: Randy Dunlap 
>>>
>>> Currently  #includes  for no obvious
>>> reason. It looks like it's only a convenience, so remove kmemleak.h
>>> from slab.h and add  to any users of kmemleak_*
>>> that don't already #include it.
>>> Also remove  from source files that do not use it.
>>>
>>> This is tested on i386 allmodconfig and x86_64 allmodconfig. It
>>> would be good to run it through the 0day bot for other $ARCHes.
>>> I have neither the horsepower nor the storage space for the other
>>> $ARCHes.
>>>
>>> [slab.h is the second most used header file after module.h; kernel.h
>>> is right there with slab.h. There could be some minor error in the
>>> counting due to some #includes having comments after them and I
>>> didn't combine all of those.]
>>>
>>> This is Lingchi patch #1 (death by a thousand cuts, applied to kernel
>>> header files).
>>>
>>> Signed-off-by: Randy Dunlap 
>> 
>> I threw it at a random selection of configs and so far the only failures
>> I'm seeing are:
>> 
>>   lib/test_firmware.c:134:2: error: implicit declaration of function 'vfree' 
>> [-Werror=implicit-function-declaration]  
>> 
>>   lib/test_firmware.c:620:25: error: implicit declaration of function 
>> 'vzalloc' [-Werror=implicit-function-declaration]
>>   lib/test_firmware.c:620:2: error: implicit declaration of function 
>> 'vzalloc' [-Werror=implicit-function-declaration]
>>   security/integrity/digsig.c:146:2: error: implicit declaration of function 
>> 'vfree' [-Werror=implicit-function-declaration]
>
> Both of those source files need to #include .

Yep, I added those and rebuilt. I don't see any more failures that look
related to your patch.

  http://kisskb.ellerman.id.au/kisskb/head/13399/


I haven't gone through the defconfigs I have enabled for a while, so
it's possible I have some missing but it's still a reasonable cross
section.

cheers

Re: [PATCH] headers: untangle kmemleak.h from mm.h

[Randy Dunlap]

On 02/12/2018 04:28 AM, Michael Ellerman wrote:
> Randy Dunlap  writes:
> 
>> From: Randy Dunlap 
>>
>> Currently  #includes  for no obvious
>> reason. It looks like it's only a convenience, so remove kmemleak.h
>> from slab.h and add  to any users of kmemleak_*
>> that don't already #include it.
>> Also remove  from source files that do not use it.
>>
>> This is tested on i386 allmodconfig and x86_64 allmodconfig. It
>> would be good to run it through the 0day bot for other $ARCHes.
>> I have neither the horsepower nor the storage space for the other
>> $ARCHes.
>>
>> [slab.h is the second most used header file after module.h; kernel.h
>> is right there with slab.h. There could be some minor error in the
>> counting due to some #includes having comments after them and I
>> didn't combine all of those.]
>>
>> This is Lingchi patch #1 (death by a thousand cuts, applied to kernel
>> header files).
>>
>> Signed-off-by: Randy Dunlap 
> 
> I threw it at a random selection of configs and so far the only failures
> I'm seeing are:
> 
>   lib/test_firmware.c:134:2: error: implicit declaration of function 'vfree' 
> [-Werror=implicit-function-declaration]   
>
>   lib/test_firmware.c:620:25: error: implicit declaration of function 
> 'vzalloc' [-Werror=implicit-function-declaration]
>   lib/test_firmware.c:620:2: error: implicit declaration of function 
> 'vzalloc' [-Werror=implicit-function-declaration]
>   security/integrity/digsig.c:146:2: error: implicit declaration of function 
> 'vfree' [-Werror=implicit-function-declaration]
> 
> Full results trickling in here, not all the failures there are caused by
> this patch, ie. some configs are broken in mainline:
> 
>   http://kisskb.ellerman.id.au/kisskb/head/13396/

That's very useful, thanks.

I'll send a few patches for those.

-- 
~Randy

Re: [PATCH] headers: untangle kmemleak.h from mm.h

[Randy Dunlap]

On 02/12/2018 04:28 AM, Michael Ellerman wrote:
> Randy Dunlap  writes:
> 
>> From: Randy Dunlap 
>>
>> Currently  #includes  for no obvious
>> reason. It looks like it's only a convenience, so remove kmemleak.h
>> from slab.h and add  to any users of kmemleak_*
>> that don't already #include it.
>> Also remove  from source files that do not use it.
>>
>> This is tested on i386 allmodconfig and x86_64 allmodconfig. It
>> would be good to run it through the 0day bot for other $ARCHes.
>> I have neither the horsepower nor the storage space for the other
>> $ARCHes.
>>
>> [slab.h is the second most used header file after module.h; kernel.h
>> is right there with slab.h. There could be some minor error in the
>> counting due to some #includes having comments after them and I
>> didn't combine all of those.]
>>
>> This is Lingchi patch #1 (death by a thousand cuts, applied to kernel
>> header files).
>>
>> Signed-off-by: Randy Dunlap 
> 
> I threw it at a random selection of configs and so far the only failures
> I'm seeing are:
> 
>   lib/test_firmware.c:134:2: error: implicit declaration of function 'vfree' 
> [-Werror=implicit-function-declaration]   
>
>   lib/test_firmware.c:620:25: error: implicit declaration of function 
> 'vzalloc' [-Werror=implicit-function-declaration]
>   lib/test_firmware.c:620:2: error: implicit declaration of function 
> 'vzalloc' [-Werror=implicit-function-declaration]
>   security/integrity/digsig.c:146:2: error: implicit declaration of function 
> 'vfree' [-Werror=implicit-function-declaration]
> 

Both of those source files need to #include .

> Full results trickling in here, not all the failures there are caused by
> this patch, ie. some configs are broken in mainline:
> 
>   http://kisskb.ellerman.id.au/kisskb/head/13396/
> 
> cheers

:)

-- 
~Randy

Re: [PATCH] headers: untangle kmemleak.h from mm.h

[Michael Ellerman]

Randy Dunlap  writes:

> From: Randy Dunlap 
>
> Currently  #includes  for no obvious
> reason. It looks like it's only a convenience, so remove kmemleak.h
> from slab.h and add  to any users of kmemleak_*
> that don't already #include it.
> Also remove  from source files that do not use it.
>
> This is tested on i386 allmodconfig and x86_64 allmodconfig. It
> would be good to run it through the 0day bot for other $ARCHes.
> I have neither the horsepower nor the storage space for the other
> $ARCHes.
>
> [slab.h is the second most used header file after module.h; kernel.h
> is right there with slab.h. There could be some minor error in the
> counting due to some #includes having comments after them and I
> didn't combine all of those.]
>
> This is Lingchi patch #1 (death by a thousand cuts, applied to kernel
> header files).
>
> Signed-off-by: Randy Dunlap 

I threw it at a random selection of configs and so far the only failures
I'm seeing are:

  lib/test_firmware.c:134:2: error: implicit declaration of function 'vfree' 
[-Werror=implicit-function-declaration] 
 
  lib/test_firmware.c:620:25: error: implicit declaration of function 'vzalloc' 
[-Werror=implicit-function-declaration]
  lib/test_firmware.c:620:2: error: implicit declaration of function 'vzalloc' 
[-Werror=implicit-function-declaration]
  security/integrity/digsig.c:146:2: error: implicit declaration of function 
'vfree' [-Werror=implicit-function-declaration]

Full results trickling in here, not all the failures there are caused by
this patch, ie. some configs are broken in mainline:

  http://kisskb.ellerman.id.au/kisskb/head/13396/

cheers

[PATCH] headers: untangle kmemleak.h from mm.h

[Randy Dunlap]

From: Randy Dunlap 

Currently  #includes  for no obvious
reason. It looks like it's only a convenience, so remove kmemleak.h
from slab.h and add  to any users of kmemleak_*
that don't already #include it.
Also remove  from source files that do not use it.

This is tested on i386 allmodconfig and x86_64 allmodconfig. It
would be good to run it through the 0day bot for other $ARCHes.
I have neither the horsepower nor the storage space for the other
$ARCHes.

[slab.h is the second most used header file after module.h; kernel.h
is right there with slab.h. There could be some minor error in the
counting due to some #includes having comments after them and I
didn't combine all of those.]

This is Lingchi patch #1 (death by a thousand cuts, applied to kernel
header files).

Signed-off-by: Randy Dunlap 
---

Fengguang, can you have this patch run thru 0day builds, please?

 arch/powerpc/sysdev/dart_iommu.c  |1 +
 arch/powerpc/sysdev/msi_bitmap.c  |1 +
 arch/s390/kernel/nmi.c|2 +-
 arch/s390/kernel/smp.c|1 -
 arch/sparc/kernel/irq_64.c|1 -
 arch/x86/kernel/pci-dma.c |1 -
 drivers/iommu/exynos-iommu.c  |1 +
 drivers/iommu/mtk_iommu_v1.c  |1 -
 drivers/net/ethernet/ti/cpsw.c|1 +
 drivers/net/wireless/realtek/rtlwifi/pci.c|1 -
 drivers/net/wireless/realtek/rtlwifi/rtl8192c/fw_common.c |1 -
 drivers/staging/rtl8188eu/hal/fw.c|2 +-
 drivers/staging/rtlwifi/pci.c |1 -
 drivers/virtio/virtio_ring.c  |1 -
 include/linux/slab.h  |1 -
 kernel/ucount.c   |1 +
 mm/cma.c  |1 +
 mm/memblock.c |1 +
 net/core/sysctl_net_core.c|1 -
 net/ipv4/route.c  |1 -
 security/apparmor/lsm.c   |1 -
 21 files changed, 9 insertions(+), 14 deletions(-)

--- lnx-416-rc1.orig/include/linux/slab.h
+++ lnx-416-rc1/include/linux/slab.h
@@ -125,7 +125,6 @@
 #define ZERO_OR_NULL_PTR(x) ((unsigned long)(x) <= \
(unsigned long)ZERO_SIZE_PTR)
 
-#include 
 #include 
 
 struct mem_cgroup;
--- lnx-416-rc1.orig/kernel/ucount.c
+++ lnx-416-rc1/kernel/ucount.c
@@ -10,6 +10,7 @@
 #include 
 #include 
 #include 
+#include 
 #include 
 
 #define UCOUNTS_HASHTABLE_BITS 10
--- lnx-416-rc1.orig/mm/memblock.c
+++ lnx-416-rc1/mm/memblock.c
@@ -17,6 +17,7 @@
 #include 
 #include 
 #include 
+#include 
 #include 
 #include 
 
--- lnx-416-rc1.orig/mm/cma.c
+++ lnx-416-rc1/mm/cma.c
@@ -35,6 +35,7 @@
 #include 
 #include 
 #include 
+#include 
 #include 
 
 #include "cma.h"
--- lnx-416-rc1.orig/drivers/staging/rtl8188eu/hal/fw.c
+++ lnx-416-rc1/drivers/staging/rtl8188eu/hal/fw.c
@@ -30,7 +30,7 @@
 #include "rtl8188e_hal.h"
 
 #include 
-#include 
+#include 
 
 static void _rtl88e_enable_fw_download(struct adapter *adapt, bool enable)
 {
--- lnx-416-rc1.orig/drivers/iommu/exynos-iommu.c
+++ lnx-416-rc1/drivers/iommu/exynos-iommu.c
@@ -17,6 +17,7 @@
 #include 
 #include 
 #include 
+#include 
 #include 
 #include 
 #include 
--- lnx-416-rc1.orig/arch/s390/kernel/nmi.c
+++ lnx-416-rc1/arch/s390/kernel/nmi.c
@@ -15,7 +15,7 @@
 #include 
 #include 
 #include 
-#include 
+#include 
 #include 
 #include 
 #include 
--- lnx-416-rc1.orig/arch/powerpc/sysdev/dart_iommu.c
+++ lnx-416-rc1/arch/powerpc/sysdev/dart_iommu.c
@@ -38,6 +38,7 @@
 #include 
 #include 
 #include 
+#include 
 #include 
 #include 
 #include 
--- lnx-416-rc1.orig/arch/powerpc/sysdev/msi_bitmap.c
+++ lnx-416-rc1/arch/powerpc/sysdev/msi_bitmap.c
@@ -10,6 +10,7 @@
 
 #include 
 #include 
+#include 
 #include 
 #include 
 #include 
--- lnx-416-rc1.orig/drivers/net/ethernet/ti/cpsw.c
+++ lnx-416-rc1/drivers/net/ethernet/ti/cpsw.c
@@ -35,6 +35,7 @@
 #include 
 #include 
 #include 
+#include 
 
 #include 
 
--- lnx-416-rc1.orig/drivers/virtio/virtio_ring.c
+++ lnx-416-rc1/drivers/virtio/virtio_ring.c
@@ -23,7 +23,6 @@
 #include 
 #include 
 #include 
-#include 
 #include 
 #include 
 
--- lnx-416-rc1.orig/security/apparmor/lsm.c
+++ lnx-416-rc1/security/apparmor/lsm.c
@@ -23,7 +23,6 @@
 #include 
 #include 
 #include 
-#include 
 #include 
 
 #include "include/apparmor.h"
--- lnx-416-rc1.orig/drivers/iommu/mtk_iommu_v1.c
+++ lnx-416-rc1/drivers/iommu/mtk_iommu_v1.c
@@ -25,7 +25,6 @@
 #include 
 #include 
 #include 
-#include 
 #include 
 #include 
 #include 
--- lnx-416-rc1.orig/drivers/staging/rtlwifi/pci.c
+++ 

Re: [PATCH] headers: untangle kmemleak.h from mm.h

[Ingo Molnar]

* Randy Dunlap  wrote:

> From: Randy Dunlap 
> 
> Currently  #includes  for no obvious
> reason. It looks like it's only a convenience, so remove kmemleak.h
> from slab.h and add  to any users of kmemleak_*
> that don't already #include it.
> Also remove  from source files that do not use it.
> 
> This is tested on i386 allmodconfig and x86_64 allmodconfig. It
> would be good to run it through the 0day bot for other $ARCHes.
> I have neither the horsepower nor the storage space for the other
> $ARCHes.
> 
> [slab.h is the second most used header file after module.h; kernel.h
> is right there with slab.h. There could be some minor error in the
> counting due to some #includes having comments after them and I
> didn't combine all of those.]
> 
> This is Lingchi patch #1 (death by a thousand cuts, applied to kernel
> header files).
> 
> Signed-off-by: Randy Dunlap 

Nice find:

Reviewed-by: Ingo Molnar 

I agree that it needs to go through 0-day to find any hidden dependencies we 
might 
have grown due to this.

Thanks,

Ingo