* linux-next: build failure on powerpc 8xx with 16k pages
@ 2020-06-04 10:48 Christophe Leroy
2020-06-04 11:17 ` Will Deacon
0 siblings, 1 reply; 6+ messages in thread
From: Christophe Leroy @ 2020-06-04 10:48 UTC (permalink / raw)
To: Andrew Morton, Michael Ellerman, PowerPC, Will Deacon,
Thomas Gleixner, Stephen Rothwell
Cc: Linux Next Mailing List, Linux Kernel Mailing List
Hi all,
Using mpc885_ads_defconfig with CONFIG_PPC_16K_PAGES instead of
CONFIG_PPC_4K_PAGES, getting the following build failure:
CC mm/gup.o
In file included from ./include/linux/kernel.h:11:0,
from mm/gup.c:2:
In function 'gup_hugepte.constprop',
inlined from 'gup_huge_pd.isra.78' at mm/gup.c:2465:8:
./include/linux/compiler.h:392:38: error: call to
'__compiletime_assert_257' declared with attribute error: Unsupported
access size for {READ,WRITE}_ONCE().
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
^
./include/linux/compiler.h:373:4: note: in definition of macro
'__compiletime_assert'
prefix ## suffix(); \
^
./include/linux/compiler.h:392:2: note: in expansion of macro
'_compiletime_assert'
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
^
./include/linux/compiler.h:405:2: note: in expansion of macro
'compiletime_assert'
compiletime_assert(__native_word(t) || sizeof(t) == sizeof(long long), \
^
./include/linux/compiler.h:291:2: note: in expansion of macro
'compiletime_assert_rwonce_type'
compiletime_assert_rwonce_type(x); \
^
mm/gup.c:2428:8: note: in expansion of macro 'READ_ONCE'
pte = READ_ONCE(*ptep);
^
In function 'gup_get_pte',
inlined from 'gup_pte_range' at mm/gup.c:2228:9,
inlined from 'gup_pmd_range' at mm/gup.c:2613:15,
inlined from 'gup_pud_range' at mm/gup.c:2641:15,
inlined from 'gup_p4d_range' at mm/gup.c:2666:15,
inlined from 'gup_pgd_range' at mm/gup.c:2694:15,
inlined from 'internal_get_user_pages_fast' at mm/gup.c:2785:3:
./include/linux/compiler.h:392:38: error: call to
'__compiletime_assert_254' declared with attribute error: Unsupported
access size for {READ,WRITE}_ONCE().
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
^
./include/linux/compiler.h:373:4: note: in definition of macro
'__compiletime_assert'
prefix ## suffix(); \
^
./include/linux/compiler.h:392:2: note: in expansion of macro
'_compiletime_assert'
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
^
./include/linux/compiler.h:405:2: note: in expansion of macro
'compiletime_assert'
compiletime_assert(__native_word(t) || sizeof(t) == sizeof(long long), \
^
./include/linux/compiler.h:291:2: note: in expansion of macro
'compiletime_assert_rwonce_type'
compiletime_assert_rwonce_type(x); \
^
mm/gup.c:2199:9: note: in expansion of macro 'READ_ONCE'
return READ_ONCE(*ptep);
^
make[2]: *** [mm/gup.o] Error 1
Bisected to:
2ab3a0a02905 (HEAD, refs/bisect/bad) READ_ONCE: Enforce atomicity for
{READ,WRITE}_ONCE() memory accesses
Christophe
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: linux-next: build failure on powerpc 8xx with 16k pages
2020-06-04 10:48 linux-next: build failure on powerpc 8xx with 16k pages Christophe Leroy
@ 2020-06-04 11:17 ` Will Deacon
2020-06-04 12:00 ` Peter Zijlstra
2020-06-04 13:55 ` Christophe Leroy
0 siblings, 2 replies; 6+ messages in thread
From: Will Deacon @ 2020-06-04 11:17 UTC (permalink / raw)
To: Christophe Leroy
Cc: Andrew Morton, Michael Ellerman, PowerPC, Thomas Gleixner,
Stephen Rothwell, Linux Next Mailing List,
Linux Kernel Mailing List, peterz
Hi, [+Peter]
On Thu, Jun 04, 2020 at 10:48:03AM +0000, Christophe Leroy wrote:
> Using mpc885_ads_defconfig with CONFIG_PPC_16K_PAGES instead of
> CONFIG_PPC_4K_PAGES, getting the following build failure:
>
> CC mm/gup.o
> In file included from ./include/linux/kernel.h:11:0,
> from mm/gup.c:2:
> In function 'gup_hugepte.constprop',
> inlined from 'gup_huge_pd.isra.78' at mm/gup.c:2465:8:
> ./include/linux/compiler.h:392:38: error: call to '__compiletime_assert_257'
> declared with attribute error: Unsupported access size for
> {READ,WRITE}_ONCE().
> _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
> ^
> ./include/linux/compiler.h:373:4: note: in definition of macro
> '__compiletime_assert'
> prefix ## suffix(); \
> ^
> ./include/linux/compiler.h:392:2: note: in expansion of macro
> '_compiletime_assert'
> _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
> ^
> ./include/linux/compiler.h:405:2: note: in expansion of macro
> 'compiletime_assert'
> compiletime_assert(__native_word(t) || sizeof(t) == sizeof(long long), \
> ^
> ./include/linux/compiler.h:291:2: note: in expansion of macro
> 'compiletime_assert_rwonce_type'
> compiletime_assert_rwonce_type(x); \
> ^
> mm/gup.c:2428:8: note: in expansion of macro 'READ_ONCE'
> pte = READ_ONCE(*ptep);
> ^
> In function 'gup_get_pte',
> inlined from 'gup_pte_range' at mm/gup.c:2228:9,
> inlined from 'gup_pmd_range' at mm/gup.c:2613:15,
> inlined from 'gup_pud_range' at mm/gup.c:2641:15,
> inlined from 'gup_p4d_range' at mm/gup.c:2666:15,
> inlined from 'gup_pgd_range' at mm/gup.c:2694:15,
> inlined from 'internal_get_user_pages_fast' at mm/gup.c:2785:3:
At first glance, this looks like a real bug in the 16k page code -- you're
loading the pte non-atomically on the fast GUP path and so you're prone to
tearing, which probably isn't what you want. For a short-term hack, I'd
suggest having CONFIG_HAVE_FAST_GUP depend on !CONFIG_PPC_16K_PAGES, but if
you want to support this them you'll need to rework your pte_t so that it
can be loaded atomically.
Will
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: linux-next: build failure on powerpc 8xx with 16k pages
2020-06-04 11:17 ` Will Deacon
@ 2020-06-04 12:00 ` Peter Zijlstra
2020-06-04 14:35 ` Christophe Leroy
2020-06-04 13:55 ` Christophe Leroy
1 sibling, 1 reply; 6+ messages in thread
From: Peter Zijlstra @ 2020-06-04 12:00 UTC (permalink / raw)
To: Will Deacon
Cc: Christophe Leroy, Andrew Morton, Michael Ellerman, PowerPC,
Thomas Gleixner, Stephen Rothwell, Linux Next Mailing List,
Linux Kernel Mailing List
On Thu, Jun 04, 2020 at 12:17:23PM +0100, Will Deacon wrote:
> Hi, [+Peter]
>
> On Thu, Jun 04, 2020 at 10:48:03AM +0000, Christophe Leroy wrote:
> > Using mpc885_ads_defconfig with CONFIG_PPC_16K_PAGES instead of
> > CONFIG_PPC_4K_PAGES, getting the following build failure:
> >
> > CC mm/gup.o
> > In file included from ./include/linux/kernel.h:11:0,
> > from mm/gup.c:2:
> > In function 'gup_hugepte.constprop',
> > inlined from 'gup_huge_pd.isra.78' at mm/gup.c:2465:8:
> > ./include/linux/compiler.h:392:38: error: call to '__compiletime_assert_257'
> > declared with attribute error: Unsupported access size for
> > {READ,WRITE}_ONCE().
> > _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
> > ^
> > ./include/linux/compiler.h:373:4: note: in definition of macro
> > '__compiletime_assert'
> > prefix ## suffix(); \
> > ^
> > ./include/linux/compiler.h:392:2: note: in expansion of macro
> > '_compiletime_assert'
> > _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
> > ^
> > ./include/linux/compiler.h:405:2: note: in expansion of macro
> > 'compiletime_assert'
> > compiletime_assert(__native_word(t) || sizeof(t) == sizeof(long long), \
> > ^
> > ./include/linux/compiler.h:291:2: note: in expansion of macro
> > 'compiletime_assert_rwonce_type'
> > compiletime_assert_rwonce_type(x); \
> > ^
> > mm/gup.c:2428:8: note: in expansion of macro 'READ_ONCE'
> > pte = READ_ONCE(*ptep);
> > ^
> > In function 'gup_get_pte',
> > inlined from 'gup_pte_range' at mm/gup.c:2228:9,
> > inlined from 'gup_pmd_range' at mm/gup.c:2613:15,
> > inlined from 'gup_pud_range' at mm/gup.c:2641:15,
> > inlined from 'gup_p4d_range' at mm/gup.c:2666:15,
> > inlined from 'gup_pgd_range' at mm/gup.c:2694:15,
> > inlined from 'internal_get_user_pages_fast' at mm/gup.c:2785:3:
>
> At first glance, this looks like a real bug in the 16k page code -- you're
> loading the pte non-atomically on the fast GUP path and so you're prone to
> tearing, which probably isn't what you want. For a short-term hack, I'd
> suggest having CONFIG_HAVE_FAST_GUP depend on !CONFIG_PPC_16K_PAGES, but if
> you want to support this them you'll need to rework your pte_t so that it
> can be loaded atomically.
Looking at commit 55c8fc3f49302, they're all the exact same value, so
what they could do is grow another special gup_get_pte() variant that
just loads the first value.
Also, per that very same commit, there's a distinct lack of WRITE_ONCE()
in the pte_update() / __set_pte_at() paths for much of Power.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: linux-next: build failure on powerpc 8xx with 16k pages
2020-06-04 11:17 ` Will Deacon
2020-06-04 12:00 ` Peter Zijlstra
@ 2020-06-04 13:55 ` Christophe Leroy
1 sibling, 0 replies; 6+ messages in thread
From: Christophe Leroy @ 2020-06-04 13:55 UTC (permalink / raw)
To: Will Deacon
Cc: Andrew Morton, Michael Ellerman, PowerPC, Thomas Gleixner,
Stephen Rothwell, Linux Next Mailing List,
Linux Kernel Mailing List, peterz
On 06/04/2020 11:17 AM, Will Deacon wrote:
> Hi, [+Peter]
>
> On Thu, Jun 04, 2020 at 10:48:03AM +0000, Christophe Leroy wrote:
>> Using mpc885_ads_defconfig with CONFIG_PPC_16K_PAGES instead of
>> CONFIG_PPC_4K_PAGES, getting the following build failure:
>>
>> CC mm/gup.o
>> In file included from ./include/linux/kernel.h:11:0,
>> from mm/gup.c:2:
>> In function 'gup_hugepte.constprop',
>> inlined from 'gup_huge_pd.isra.78' at mm/gup.c:2465:8:
>> ./include/linux/compiler.h:392:38: error: call to '__compiletime_assert_257'
>> declared with attribute error: Unsupported access size for
>> {READ,WRITE}_ONCE().
>> _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
>> ^
>> ./include/linux/compiler.h:373:4: note: in definition of macro
>> '__compiletime_assert'
>> prefix ## suffix(); \
>> ^
>> ./include/linux/compiler.h:392:2: note: in expansion of macro
>> '_compiletime_assert'
>> _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
>> ^
>> ./include/linux/compiler.h:405:2: note: in expansion of macro
>> 'compiletime_assert'
>> compiletime_assert(__native_word(t) || sizeof(t) == sizeof(long long), \
>> ^
>> ./include/linux/compiler.h:291:2: note: in expansion of macro
>> 'compiletime_assert_rwonce_type'
>> compiletime_assert_rwonce_type(x); \
>> ^
>> mm/gup.c:2428:8: note: in expansion of macro 'READ_ONCE'
>> pte = READ_ONCE(*ptep);
>> ^
>> In function 'gup_get_pte',
>> inlined from 'gup_pte_range' at mm/gup.c:2228:9,
>> inlined from 'gup_pmd_range' at mm/gup.c:2613:15,
>> inlined from 'gup_pud_range' at mm/gup.c:2641:15,
>> inlined from 'gup_p4d_range' at mm/gup.c:2666:15,
>> inlined from 'gup_pgd_range' at mm/gup.c:2694:15,
>> inlined from 'internal_get_user_pages_fast' at mm/gup.c:2785:3:
>
> At first glance, this looks like a real bug in the 16k page code -- you're
> loading the pte non-atomically on the fast GUP path and so you're prone to
> tearing, which probably isn't what you want. For a short-term hack, I'd
> suggest having CONFIG_HAVE_FAST_GUP depend on !CONFIG_PPC_16K_PAGES, but if
> you want to support this them you'll need to rework your pte_t so that it
> can be loaded atomically.
What do you mean by *rework* pte_t ?
pte are 32 bits words in size and are spread every 4 words in memory.
Therefore pte_t has to be 128 bits because unlike huge_pte handling
which always use huge_pte_offset() in loops, many many places in the
kernel do pte++, so we need the pte type to be the size of the interval
from one pte to the next one.
Christophe
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: linux-next: build failure on powerpc 8xx with 16k pages
2020-06-04 12:00 ` Peter Zijlstra
@ 2020-06-04 14:35 ` Christophe Leroy
2020-06-04 16:10 ` Will Deacon
0 siblings, 1 reply; 6+ messages in thread
From: Christophe Leroy @ 2020-06-04 14:35 UTC (permalink / raw)
To: Peter Zijlstra, Will Deacon
Cc: Andrew Morton, Michael Ellerman, PowerPC, Thomas Gleixner,
Stephen Rothwell, Linux Next Mailing List,
Linux Kernel Mailing List
On 06/04/2020 12:00 PM, Peter Zijlstra wrote:
> On Thu, Jun 04, 2020 at 12:17:23PM +0100, Will Deacon wrote:
>> Hi, [+Peter]
>>
>> On Thu, Jun 04, 2020 at 10:48:03AM +0000, Christophe Leroy wrote:
>>> Using mpc885_ads_defconfig with CONFIG_PPC_16K_PAGES instead of
>>> CONFIG_PPC_4K_PAGES, getting the following build failure:
>>>
>>> CC mm/gup.o
>>> In file included from ./include/linux/kernel.h:11:0,
>>> from mm/gup.c:2:
>>> In function 'gup_hugepte.constprop',
>>> inlined from 'gup_huge_pd.isra.78' at mm/gup.c:2465:8:
>>> ./include/linux/compiler.h:392:38: error: call to '__compiletime_assert_257'
>>> declared with attribute error: Unsupported access size for
>>> {READ,WRITE}_ONCE().
>>> _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
>>> ^
>>> ./include/linux/compiler.h:373:4: note: in definition of macro
>>> '__compiletime_assert'
>>> prefix ## suffix(); \
>>> ^
>>> ./include/linux/compiler.h:392:2: note: in expansion of macro
>>> '_compiletime_assert'
>>> _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
>>> ^
>>> ./include/linux/compiler.h:405:2: note: in expansion of macro
>>> 'compiletime_assert'
>>> compiletime_assert(__native_word(t) || sizeof(t) == sizeof(long long), \
>>> ^
>>> ./include/linux/compiler.h:291:2: note: in expansion of macro
>>> 'compiletime_assert_rwonce_type'
>>> compiletime_assert_rwonce_type(x); \
>>> ^
>>> mm/gup.c:2428:8: note: in expansion of macro 'READ_ONCE'
>>> pte = READ_ONCE(*ptep);
>>> ^
>>> In function 'gup_get_pte',
>>> inlined from 'gup_pte_range' at mm/gup.c:2228:9,
>>> inlined from 'gup_pmd_range' at mm/gup.c:2613:15,
>>> inlined from 'gup_pud_range' at mm/gup.c:2641:15,
>>> inlined from 'gup_p4d_range' at mm/gup.c:2666:15,
>>> inlined from 'gup_pgd_range' at mm/gup.c:2694:15,
>>> inlined from 'internal_get_user_pages_fast' at mm/gup.c:2785:3:
>>
>> At first glance, this looks like a real bug in the 16k page code -- you're
>> loading the pte non-atomically on the fast GUP path and so you're prone to
>> tearing, which probably isn't what you want. For a short-term hack, I'd
>> suggest having CONFIG_HAVE_FAST_GUP depend on !CONFIG_PPC_16K_PAGES, but if
>> you want to support this them you'll need to rework your pte_t so that it
>> can be loaded atomically.
>
> Looking at commit 55c8fc3f49302, they're all the exact same value, so
> what they could do is grow another special gup_get_pte() variant that
> just loads the first value.
>
> Also, per that very same commit, there's a distinct lack of WRITE_ONCE()
> in the pte_update() / __set_pte_at() paths for much of Power.
>
Thanks for the idea.
Now I get the same issue at
CC mm/mincore.o
In file included from ./include/asm-generic/bug.h:5:0,
from ./arch/powerpc/include/asm/bug.h:109,
from ./include/linux/bug.h:5,
from ./include/linux/mmdebug.h:5,
from ./include/linux/mm.h:9,
from ./include/linux/pagemap.h:8,
from mm/mincore.c:11:
In function 'huge_ptep_get',
inlined from 'mincore_hugetlb' at mm/mincore.c:35:20:
./include/linux/compiler.h:392:38: error: call to
'__compiletime_assert_218' declared with attribute error: Unsupported
access size for {READ,WRITE}_ONCE().
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
^
./include/linux/compiler.h:373:4: note: in definition of macro
'__compiletime_assert'
prefix ## suffix(); \
^
./include/linux/compiler.h:392:2: note: in expansion of macro
'_compiletime_assert'
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
^
./include/linux/compiler.h:405:2: note: in expansion of macro
'compiletime_assert'
compiletime_assert(__native_word(t) || sizeof(t) == sizeof(long long), \
^
./include/linux/compiler.h:291:2: note: in expansion of macro
'compiletime_assert_rwonce_type'
compiletime_assert_rwonce_type(x); \
^
./include/asm-generic/hugetlb.h:125:9: note: in expansion of macro
'READ_ONCE'
return READ_ONCE(*ptep);
^
make[2]: *** [mm/mincore.o] Error 1
I guess for this one I have to implement platform specific huge_ptep_get()
Christophe
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: linux-next: build failure on powerpc 8xx with 16k pages
2020-06-04 14:35 ` Christophe Leroy
@ 2020-06-04 16:10 ` Will Deacon
0 siblings, 0 replies; 6+ messages in thread
From: Will Deacon @ 2020-06-04 16:10 UTC (permalink / raw)
To: Christophe Leroy
Cc: Peter Zijlstra, Andrew Morton, Michael Ellerman, PowerPC,
Thomas Gleixner, Stephen Rothwell, Linux Next Mailing List,
Linux Kernel Mailing List, arnd
[+Arnd since I think we spoke about this on IRC once]
On Thu, Jun 04, 2020 at 02:35:14PM +0000, Christophe Leroy wrote:
> Now I get the same issue at
>
> CC mm/mincore.o
> In file included from ./include/asm-generic/bug.h:5:0,
> from ./arch/powerpc/include/asm/bug.h:109,
> from ./include/linux/bug.h:5,
> from ./include/linux/mmdebug.h:5,
> from ./include/linux/mm.h:9,
> from ./include/linux/pagemap.h:8,
> from mm/mincore.c:11:
> In function 'huge_ptep_get',
> inlined from 'mincore_hugetlb' at mm/mincore.c:35:20:
> ./include/linux/compiler.h:392:38: error: call to '__compiletime_assert_218'
> declared with attribute error: Unsupported access size for
> {READ,WRITE}_ONCE().
> _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
> ^
> ./include/linux/compiler.h:373:4: note: in definition of macro
> '__compiletime_assert'
> prefix ## suffix(); \
> ^
> ./include/linux/compiler.h:392:2: note: in expansion of macro
> '_compiletime_assert'
> _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
> ^
> ./include/linux/compiler.h:405:2: note: in expansion of macro
> 'compiletime_assert'
> compiletime_assert(__native_word(t) || sizeof(t) == sizeof(long long), \
> ^
> ./include/linux/compiler.h:291:2: note: in expansion of macro
> 'compiletime_assert_rwonce_type'
> compiletime_assert_rwonce_type(x); \
> ^
> ./include/asm-generic/hugetlb.h:125:9: note: in expansion of macro
> 'READ_ONCE'
> return READ_ONCE(*ptep);
> ^
> make[2]: *** [mm/mincore.o] Error 1
>
> I guess for this one I have to implement platform specific huge_ptep_get()
Yeah, or bite the bullet and introduce proper accessors for all these
things:
pte_read()
pmd_read()
pud_read()
etc
with the default implementation pointing at READ_ONCE(), but allowing an
architecture override. It's a big job because mm/ would need repainting,
but it would have the benefit of being able to remove aggregate types from
READ_ONCE() entirely and using a special accessor just for the page-table
types.
That might also mean that we could have asm-generic versions of things
like ptep_get_and_clear() that work for architectures with hardware
update and need atomic rmw. But I'm getting ahead of myself.
Will
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2020-06-04 16:10 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-06-04 10:48 linux-next: build failure on powerpc 8xx with 16k pages Christophe Leroy
2020-06-04 11:17 ` Will Deacon
2020-06-04 12:00 ` Peter Zijlstra
2020-06-04 14:35 ` Christophe Leroy
2020-06-04 16:10 ` Will Deacon
2020-06-04 13:55 ` Christophe Leroy
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).