Skip to contentSkip to conversation

hugetlb: error: ‘cpu_has_pse’ undeclared

-mmots 2016-04-25 Sergey Senozhatsky 4 participants 4 messages Apr 26, 2016 – Apr 27, 2016

lore
Share & tools
lore
#0 lore raw

Hello,

v4.6-rc5-mmots-2016-04-25-17-33

In file included from include/linux/hugetlb.h:418:0,
from fs/hugetlbfs/inode.c:28:
fs/hugetlbfs/inode.c: In function ‘init_hugetlbfs_fs’:
./arch/x86/include/asm/hugetlb.h:7:31: error: ‘cpu_has_pse’ undeclared (first use in this function)

 #define hugepages_supported() cpu_has_pse
                               ^
fs/hugetlbfs/inode.c:1325:7: note: in expansion of macro ‘hugepages_supported’
  if (!hugepages_supported()) {
       ^~~~~~~~~~~~~~~~~~~
./arch/x86/include/asm/hugetlb.h:7:31: note: each undeclared identifier is reported only once for each function it appears in
 #define hugepages_supported() cpu_has_pse
                               ^
fs/hugetlbfs/inode.c:1325:7: note: in expansion of macro ‘hugepages_supported’
  if (!hugepages_supported()) {
       ^~~~~~~~~~~~~~~~~~~

scripts/Makefile.build:289: recipe for target 'fs/hugetlbfs/inode.o' failed make[2]: *** [fs/hugetlbfs/inode.o] Error 1 scripts/Makefile.build:440: recipe for target 'fs/hugetlbfs' failed make[1]: *** [fs/hugetlbfs] Error 2
make[1]: *** Waiting for unfinished jobs....

git grep cpu_has_pse
arch/x86/include/asm/hugetlb.h:#define hugepages_supported() cpu_has_pse

	-ss
#1 lore raw
Quoted 11 lines · On Tue, 26 Apr 2016 23:14:35 +0900 Sergey Senozhatsky…

Hello,

v4.6-rc5-mmots-2016-04-25-17-33

In file included from include/linux/hugetlb.h:418:0,
from fs/hugetlbfs/inode.c:28:
fs/hugetlbfs/inode.c: In function 'init_hugetlbfs_fs':
./arch/x86/include/asm/hugetlb.h:7:31: error: 'cpu_has_pse' undeclared (first use in this function)

 #define hugepages_supported() cpu_has_pse
                               ^

hm, how did that happen. I had some issues with cpu_has_pse a number of days ago but they later went away.

In my current tree I have, in arch/x86/include/asm/hugetlb.h:

#define hugepages_supported() boot_cpu_has(X86_FEATURE_PSE)

and that came in from linux-next.patch. I wonder why your tree is different.

<looks>

OK, http://ozlabs.org/~akpm/mmots/broken-out/linux-next.patch has no changes to arch/x86/include/asm/hugetlb.h at all. Maybe I fat-fingered something. Odd.

I have just uploaded a new snapshot - it should be good now. Sorry about that.

#2 lore raw
Quoted 20 lines · Andrew Morton
Quoted 9 lines

v4.6-rc5-mmots-2016-04-25-17-33

In file included from include/linux/hugetlb.h:418:0,
from fs/hugetlbfs/inode.c:28:
fs/hugetlbfs/inode.c: In function 'init_hugetlbfs_fs':
./arch/x86/include/asm/hugetlb.h:7:31: error: 'cpu_has_pse' undeclared (first use in this function)

 #define hugepages_supported() cpu_has_pse
                               ^

hm, how did that happen. I had some issues with cpu_has_pse a number of days ago but they later went away.

In my current tree I have, in arch/x86/include/asm/hugetlb.h:

#define hugepages_supported() boot_cpu_has(X86_FEATURE_PSE)

and that came in from linux-next.patch. I wonder why your tree is different.

Quoted 8 lines

<looks>

OK, http://ozlabs.org/~akpm/mmots/broken-out/linux-next.patch has no changes to arch/x86/include/asm/hugetlb.h at all. Maybe I fat-fingered something. Odd.

I have just uploaded a new snapshot - it should be good now. Sorry about that.

no worries. thanks!

	-ss
#3 lore raw
In reply to Andrew Morton
Quoted 21 lines · Andrew Morton
Quoted 11 lines · On Tue, 26 Apr 2016 23:14:35 +0900 Sergey Senozhatsky…

Hello,

v4.6-rc5-mmots-2016-04-25-17-33

In file included from include/linux/hugetlb.h:418:0,
from fs/hugetlbfs/inode.c:28:
fs/hugetlbfs/inode.c: In function 'init_hugetlbfs_fs':
./arch/x86/include/asm/hugetlb.h:7:31: error: 'cpu_has_pse' undeclared (first use in this function)

 #define hugepages_supported() cpu_has_pse
                               ^

hm, how did that happen. I had some issues with cpu_has_pse a number of days ago but they later went away.

In my current tree I have, in arch/x86/include/asm/hugetlb.h:

#define hugepages_supported() boot_cpu_has(X86_FEATURE_PSE)

Quoted 8 lines

and that came in from linux-next.patch. I wonder why your tree is different.

<looks>

OK, http://ozlabs.org/~akpm/mmots/broken-out/linux-next.patch has no changes to arch/x86/include/asm/hugetlb.h at all. Maybe I fat-fingered something. Odd.

So I think the reason is that cpu_has_pse is gone from the x86 devel tree, please use this instead:

boot_cpu_has(X86_FEATURE_PSE)

Thanks,

Ingo