Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1625084 > unrolled thread

[PATCH] powerpc/32: Move entry_32 functions just after HEAD functions.

Back to article view | Back to linux.kernel

Contents

  [PATCH] powerpc/32: Move entry_32 functions just after HEAD  functions. Christophe Leroy <christophe.leroy@c-s.fr> - 2017-04-18 09:10 +0200
    RE: [PATCH] powerpc/32: Move entry_32 functions just after HEAD  functions. David Laight <David.Laight@ACULAB.COM> - 2017-04-19 17:20 +0200
      RE: [PATCH] powerpc/32: Move entry_32 functions just after HEAD functions. Michael Ellerman <mpe@ellerman.id.au> - 2017-04-20 13:30 +0200
        Re: [PATCH] powerpc/32: Move entry_32 functions just after HEAD  functions. Christophe LEROY <christophe.leroy@c-s.fr> - 2017-04-20 13:50 +0200

#1625084 — [PATCH] powerpc/32: Move entry_32 functions just after HEAD functions.

By default, PPC8xx PINs an ITLB on the first 8M of memory in order
to avoid any ITLB miss on kernel code.
However, with some debug functions like DEBUG_PAGEALLOC and
(soon to come) DEBUG_RODATA, the PINned TLB is invalidated soon
after startup so ITLB missed start to happen also on the kernel code.

In order to avoid any ITLB miss in a critical section, we have to
ensure that their is no page boundary crossed between the setup of
a new value in SRR0/SRR1 and the associated RFI. This cannot be done
easily if entry_32 functions sits in the middle of other .text
functions. By placing entry_32 just after the .head section (as already
done for entry_64 on PPC64), we can more easily ensure the issue
doesn't happen.

Ex: with this change, all entry_32 functions playing with srr0/srr1
appears between 0xc0002000 and 0xc0002fff on PPC_8xx.

Signed-off-by: Christophe Leroy <christophe.leroy@c-s.fr>
---
 arch/powerpc/Makefile        | 2 +-
 arch/powerpc/kernel/Makefile | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/powerpc/Makefile b/arch/powerpc/Makefile
index 19b0d1a81959..8aa1dafb14eb 100644
--- a/arch/powerpc/Makefile
+++ b/arch/powerpc/Makefile
@@ -243,7 +243,7 @@ head-$(CONFIG_40x)		:= arch/powerpc/kernel/head_40x.o
 head-$(CONFIG_44x)		:= arch/powerpc/kernel/head_44x.o
 head-$(CONFIG_FSL_BOOKE)	:= arch/powerpc/kernel/head_fsl_booke.o
 
-head-$(CONFIG_PPC64)		+= arch/powerpc/kernel/entry_64.o
+head-y				+= arch/powerpc/kernel/entry_$(BITS).o
 head-$(CONFIG_PPC_FPU)		+= arch/powerpc/kernel/fpu.o
 head-$(CONFIG_ALTIVEC)		+= arch/powerpc/kernel/vector.o
 head-$(CONFIG_PPC_OF_BOOT_TRAMPOLINE)  += arch/powerpc/kernel/prom_init.o
diff --git a/arch/powerpc/kernel/Makefile b/arch/powerpc/kernel/Makefile
index 811f441a125f..eb2db44c5fe9 100644
--- a/arch/powerpc/kernel/Makefile
+++ b/arch/powerpc/kernel/Makefile
@@ -90,7 +90,7 @@ extra-y				+= vmlinux.lds
 
 obj-$(CONFIG_RELOCATABLE)	+= reloc_$(BITS).o
 
-obj-$(CONFIG_PPC32)		+= entry_32.o setup_32.o
+obj-$(CONFIG_PPC32)		+= setup_32.o
 obj-$(CONFIG_PPC64)		+= dma-iommu.o iommu.o
 obj-$(CONFIG_KGDB)		+= kgdb.o
 obj-$(CONFIG_BOOTX_TEXT)	+= btext.o
@@ -154,7 +154,7 @@ UBSAN_SANITIZE_vdso.o := n
 
 extra-$(CONFIG_PPC_FPU)		+= fpu.o
 extra-$(CONFIG_ALTIVEC)		+= vector.o
-extra-$(CONFIG_PPC64)		+= entry_64.o
+extra-y				+= entry_$(BITS).o
 extra-$(CONFIG_PPC_OF_BOOT_TRAMPOLINE)	+= prom_init.o
 
 extra-y				+= systbl_chk.i
-- 
2.12.0

[toc] | [next] | [standalone]

#1626333

From: Christophe Leroy
> By default, PPC8xx PINs an ITLB on the first 8M of memory in order
> to avoid any ITLB miss on kernel code.
> However, with some debug functions like DEBUG_PAGEALLOC and
> (soon to come) DEBUG_RODATA, the PINned TLB is invalidated soon
> after startup so ITLB missed start to happen also on the kernel code.
> 
> In order to avoid any ITLB miss in a critical section, we have to
> ensure that their is no page boundary crossed between the setup of
> a new value in SRR0/SRR1 and the associated RFI. This cannot be done
> easily if entry_32 functions sits in the middle of other .text
> functions. By placing entry_32 just after the .head section (as already
> done for entry_64 on PPC64), we can more easily ensure the issue
> doesn't happen.

Shouldn't this be done by putting all the functions that 'matter'
into a named section instead of relying on the order of the input files?
(Which is what I think this is doing.)

	David

[toc] | [prev] | [next] | [standalone]

#1627373 — RE: [PATCH] powerpc/32: Move entry_32 functions just after HEAD functions.

David Laight <David.Laight@ACULAB.COM> writes:

> From: Christophe Leroy
>> By default, PPC8xx PINs an ITLB on the first 8M of memory in order
>> to avoid any ITLB miss on kernel code.
>> However, with some debug functions like DEBUG_PAGEALLOC and
>> (soon to come) DEBUG_RODATA, the PINned TLB is invalidated soon
>> after startup so ITLB missed start to happen also on the kernel code.
>> 
>> In order to avoid any ITLB miss in a critical section, we have to
>> ensure that their is no page boundary crossed between the setup of
>> a new value in SRR0/SRR1 and the associated RFI. This cannot be done
>> easily if entry_32 functions sits in the middle of other .text
>> functions. By placing entry_32 just after the .head section (as already
>> done for entry_64 on PPC64), we can more easily ensure the issue
>> doesn't happen.
>
> Shouldn't this be done by putting all the functions that 'matter'
> into a named section instead of relying on the order of the input files?
> (Which is what I think this is doing.)

Yeah that is fragile if there's nothing more to it.

I'm not sure if we need a special section. If the functions that must
not cross a page boundary are aligned to a page boundary (with .align)
then that would also work wouldn't it?

cheers

[toc] | [prev] | [next] | [standalone]

#1627393

Le 20/04/2017 à 13:25, Michael Ellerman a écrit :
> David Laight <David.Laight@ACULAB.COM> writes:
>
>> From: Christophe Leroy
>>> By default, PPC8xx PINs an ITLB on the first 8M of memory in order
>>> to avoid any ITLB miss on kernel code.
>>> However, with some debug functions like DEBUG_PAGEALLOC and
>>> (soon to come) DEBUG_RODATA, the PINned TLB is invalidated soon
>>> after startup so ITLB missed start to happen also on the kernel code.
>>>
>>> In order to avoid any ITLB miss in a critical section, we have to
>>> ensure that their is no page boundary crossed between the setup of
>>> a new value in SRR0/SRR1 and the associated RFI. This cannot be done
>>> easily if entry_32 functions sits in the middle of other .text
>>> functions. By placing entry_32 just after the .head section (as already
>>> done for entry_64 on PPC64), we can more easily ensure the issue
>>> doesn't happen.
>>
>> Shouldn't this be done by putting all the functions that 'matter'
>> into a named section instead of relying on the order of the input files?
>> (Which is what I think this is doing.)
>
> Yeah that is fragile if there's nothing more to it.
>
> I'm not sure if we need a special section. If the functions that must
> not cross a page boundary are aligned to a page boundary (with .align)
> then that would also work wouldn't it?

Yes it would work, but we would also 'loose' some space (up to one page 
?), as I don't think 'ld' would fill the blanks with other functions.

Maybe we can find a lower alignment than a page boundary, I'll 
investigate that.

Christophe

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web