lib: sbi: Synchronize PMP settings with virtual memory system
PATCH 1/1 lib: sbi: Synchronize PMP settings with virtual memory system
As per section 3.7.2 of RISC-V Privileged Specification, PMP settings must be synchronized with the virtual memory system after PMP settings have been written.
Signed-off-by Himanshu Chauhan <[email protected]>
--- lib/sbi/sbi_hart.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+)
lib/sbi/sbi_hart.c +22 -0
@@ -22,6 +22,7 @@#include <sbi/sbi_pmu.h> #include <sbi/sbi_string.h> #include <sbi/sbi_trap.h> +#include <sbi/sbi_hfence.h> extern void __sbi_expected_trap (void );extern void __sbi_expected_trap_hext (void );
@@ -321,6 +322,27 @@ int sbi_hart_pmp_configure(struct sbi_scratch *scratch) } } +/* +* As per section 3.7.2 of privileged specification v1.12, +* virtual address translations can be speculatively performed +* (even before actual access). These, along with PMP traslations, +* can be cached. This can pose a problem with CPU hotplug +* and non-retentive suspend scenario because PMP states are +* not preserved. +* It is advisable to flush the caching structures under such +* conditions. +*/ +if (misa_extension ('S' )) { + __asm__ __volatile__("sfence.vma" ); + +/* +* If hypervisor mode is supported, flush caching +* structures in guest mode too. +*/ +if (misa_extension ('H' )) +__sbi_hfence_gvma_all (); + } +return 0 ; }
Signature · 1 line
2.38.1
Anup Patel
AppliedPATCH 1/1 lib: sbi: Synchronize PMP settings with virtual memory system
Quoted 6 lines · Himanshu Chauhan
As per section 3.7.2 of RISC-V Privileged Specification, PMP settings must be synchronized with the virtual memory system after PMP settings have been written.
Signed-off-by Himanshu Chauhan <[email protected]>
Looks good to me.
Reviewed-by Anup Patel <[email protected]>
Applied this patch to the riscv/opensbi repo.
Thanks,
Anup
Quoted 52 lines
--- lib/sbi/sbi_hart.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+)lib/sbi/sbi_hart.c +22
-1@@ -22,6 +22,7 @@#include <sbi/sbi_pmu.h> #include <sbi/sbi_string.h> #include <sbi/sbi_trap.h> +#include <sbi/sbi_hfence.h> extern void __sbi_expected_trap (void );extern void __sbi_expected_trap_hext (void );@@ -321,6 +322,27 @@ int sbi_hart_pmp_configure(struct sbi_scratch *scratch) } } +/* +* As per section 3.7.2 of privileged specification v1.12, +* virtual address translations can be speculatively performed +* (even before actual access). These, along with PMP traslations, +* can be cached. This can pose a problem with CPU hotplug +* and non-retentive suspend scenario because PMP states are +* not preserved. +* It is advisable to flush the caching structures under such +* conditions. +*/ +if (misa_extension ('S' )) { + __asm__ __volatile__("sfence.vma" ); + +/* +* If hypervisor mode is supported, flush caching +* structures in guest mode too. +*/ +if (misa_extension ('H' )) +__sbi_hfence_gvma_all (); + } +return 0 ; } --2.38.1
--
opensbi mailing list
opensbi at lists.infradead.org
http://lists.infradead.org/mailman/listinfo/opensbi