s390: Disable ARCH_WANT_OPTIMIZE_HUGETLB_VMEMMAP
Subsystems (2)
Heiko Carstens
maintainerPATCH 1/1 s390: Disable ARCH_WANT_OPTIMIZE_HUGETLB_VMEMMAP
As reported by Luiz Capitulino enabling HVO on s390 leads to reproducible crashes. The problem is that kernel page tables are modified without flushing corresponding TLB entries.
Even if it looks like the empty flush_tlb_all() implementation on s390 is the problem, it is actually a different problem: on s390 it is not allowed to replace an active/valid page table entry with another valid page table entry without the detour over an invalid entry. A direct replacement may lead to random crashes and/or data corruption.
In order to invalidate an entry special instructions have to be used (e.g. ipte or idte). Alternatively there are also special instructions available which allow to replace a valid entry with a different valid entry (e.g. crdte or cspg).
Given that the HVO code currently does not provide the hooks to allow for an implementation which is compliant with the s390 architecture requirements, disable ARCH_WANT_OPTIMIZE_HUGETLB_VMEMMAP again, which is basically a revert of the original patch which enabled it.
--- arch/s390/Kconfig | 1 - 1 file changed, 1 deletion(-)
arch/s390/Kconfig +0 -1
@@ -158,7 +158,6 @@ config S390 select ARCH_WANT_IRQS_OFF_ACTIVATE_MM select ARCH_WANT_KERNEL_PMD_MKWRITE select ARCH_WANT_LD_ORPHAN_WARN - select ARCH_WANT_OPTIMIZE_HUGETLB_VMEMMAP select ARCH_WANTS_THP_SWAP select BUILDTIME_TABLE_SORT select CLONE_BACKWARDS2
Signature · 1 line
2.48.1
Quoted 20 lines · Heiko Carstens
As reported by Luiz Capitulino enabling HVO on s390 leads to reproducible crashes. The problem is that kernel page tables are modified without flushing corresponding TLB entries.
Even if it looks like the empty
flush_tlb_all()implementation on s390 is the problem, it is actually a different problem: on s390 it is not allowed to replace an active/valid page table entry with another valid page table entry without the detour over an invalid entry. A direct replacement may lead to random crashes and/or data corruption.In order to invalidate an entry special instructions have to be used (e.g. ipte or idte). Alternatively there are also special instructions available which allow to replace a valid entry with a different valid entry (e.g. crdte or cspg).
Given that the HVO code currently does not provide the hooks to allow for an implementation which is compliant with the s390 architecture requirements, disable
ARCH_WANT_OPTIMIZE_HUGETLB_VMEMMAPagain, which is basically a revert of the original patch which enabled it.
Quoted 6 lines
Reported-by Luiz Capitulino <[email protected]>Fixes 00a34d5a99c0 ("s390: select ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP")Signed-off-by Heiko Carstens <[email protected]>---
Signature · 3 lines
Cheers David / dhildenb
Quoted 20 lines · On Thu, 30 Oct 2025 15:55:05 +0100 Heiko Carstens <[email protected]> wrote:
As reported by Luiz Capitulino enabling HVO on s390 leads to reproducible crashes. The problem is that kernel page tables are modified without flushing corresponding TLB entries.
Even if it looks like the empty
flush_tlb_all()implementation on s390 is the problem, it is actually a different problem: on s390 it is not allowed to replace an active/valid page table entry with another valid page table entry without the detour over an invalid entry. A direct replacement may lead to random crashes and/or data corruption.In order to invalidate an entry special instructions have to be used (e.g. ipte or idte). Alternatively there are also special instructions available which allow to replace a valid entry with a different valid entry (e.g. crdte or cspg).
Given that the HVO code currently does not provide the hooks to allow for an implementation which is compliant with the s390 architecture requirements, disable
ARCH_WANT_OPTIMIZE_HUGETLB_VMEMMAPagain, which is basically a revert of the original patch which enabled it.
Quoted 8 lines
Reported-by Luiz Capitulino <[email protected]>Fixes 00a34d5a99c0 ("s390: select ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP")Signed-off-by Heiko Carstens <[email protected]>--- arch/s390/Kconfig | 1 - 1 file changed, 1 deletion(-)
Thanks!
Quoted 20 lines · Heiko Carstens
As reported by Luiz Capitulino enabling HVO on s390 leads to reproducible crashes. The problem is that kernel page tables are modified without flushing corresponding TLB entries.
Even if it looks like the empty
flush_tlb_all()implementation on s390 is the problem, it is actually a different problem: on s390 it is not allowed to replace an active/valid page table entry with another valid page table entry without the detour over an invalid entry. A direct replacement may lead to random crashes and/or data corruption.In order to invalidate an entry special instructions have to be used (e.g. ipte or idte). Alternatively there are also special instructions available which allow to replace a valid entry with a different valid entry (e.g. crdte or cspg).
Given that the HVO code currently does not provide the hooks to allow for an implementation which is compliant with the s390 architecture requirements, disable
ARCH_WANT_OPTIMIZE_HUGETLB_VMEMMAPagain, which is basically a revert of the original patch which enabled it.
Quoted 5 lines
Reported-by Luiz Capitulino <[email protected]>Fixes 00a34d5a99c0 ("s390: select ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP")Signed-off-by Heiko Carstens <[email protected]>
Obvious enough change, but FWIW:
Quoted 16 lines
--- arch/s390/Kconfig | 1 - 1 file changed, 1 deletion(-)arch/s390/Kconfig +0
-1@@ -158,7 +158,6 @@ config S390 select ARCH_WANT_IRQS_OFF_ACTIVATE_MM select ARCH_WANT_KERNEL_PMD_MKWRITE select ARCH_WANT_LD_ORPHAN_WARN - select ARCH_WANT_OPTIMIZE_HUGETLB_VMEMMAP select ARCH_WANTS_THP_SWAP select BUILDTIME_TABLE_SORT select CLONE_BACKWARDS2
Quoted 20 lines · Heiko Carstens
As reported by Luiz Capitulino enabling HVO on s390 leads to reproducible crashes. The problem is that kernel page tables are modified without flushing corresponding TLB entries.
Even if it looks like the empty
flush_tlb_all()implementation on s390 is the problem, it is actually a different problem: on s390 it is not allowed to replace an active/valid page table entry with another valid page table entry without the detour over an invalid entry. A direct replacement may lead to random crashes and/or data corruption.In order to invalidate an entry special instructions have to be used (e.g. ipte or idte). Alternatively there are also special instructions available which allow to replace a valid entry with a different valid entry (e.g. crdte or cspg).
Given that the HVO code currently does not provide the hooks to allow for an implementation which is compliant with the s390 architecture requirements, disable
ARCH_WANT_OPTIMIZE_HUGETLB_VMEMMAPagain, which is basically a revert of the original patch which enabled it.
Quoted 8 lines
Reported-by Luiz Capitulino <[email protected]>Fixes 00a34d5a99c0 ("s390: select ARCH_WANT_HUGETLB_PAGE_OPTIMIZE_VMEMMAP")Signed-off-by Heiko Carstens <[email protected]>--- arch/s390/Kconfig | 1 - 1 file changed, 1 deletion(-)
Yup architectures like aarch64 and PPC64 keep it disable for the same reason. However patch [1] allows for enabling similar optimization for DAX vmemmap "ARCH_WANT_OPTIMIZE_DAX_VMEMMAP", if DAX is supported.
I see in 6.18-1 pull request s390 had "Add DAX support for DCSS memory block devices".
[1]: https://lore.kernel.org/all/[email protected]/T/#u
-ritesh
Quoted 15 lines
arch/s390/Kconfig +0
-1@@ -158,7 +158,6 @@ config S390 select ARCH_WANT_IRQS_OFF_ACTIVATE_MM select ARCH_WANT_KERNEL_PMD_MKWRITE select ARCH_WANT_LD_ORPHAN_WARN - select ARCH_WANT_OPTIMIZE_HUGETLB_VMEMMAP select ARCH_WANTS_THP_SWAP select BUILDTIME_TABLE_SORT select CLONE_BACKWARDS2Signature · 1 line
2.48.1