| Message ID | IA1PR20MB495351E19560D833EAC11BC8BBCDA@IA1PR20MB4953.namprd20.prod.outlook.com |
|---|---|
| State | Superseded |
| Headers | show |
| Series | lib: sbi: Add sub-regions check for sanitizing domain | expand |
On Tue, Oct 10, 2023 at 6:30 AM Inochi Amaoto <inochiama@outlook.com> wrote: > > make is_region_subset() check the bound of the domain region, so it can > handle the domain region with the same base and length. > > Signed-off-by: Inochi Amaoto <inochiama@outlook.com> Important fact here is that regB_start < regB_end and regA_start < regA_end because order == 0 region is of size 1 byte. When regB_start == regA_start and regB_end == regA_end, the following is the outcome of the expressions in the existing "if ()" check. (regB_start <= regA_start) == true (regA_start < regB_end) == true (regB_start < regA_end) == true (regA_end <= regB_end) == true Based on the above, I don't see how this patch helps. Regards, Anup > --- > lib/sbi/sbi_domain.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/lib/sbi/sbi_domain.c b/lib/sbi/sbi_domain.c > index 283b598..2b9d16f 100644 > --- a/lib/sbi/sbi_domain.c > +++ b/lib/sbi/sbi_domain.c > @@ -185,8 +185,8 @@ static bool is_region_subset(const struct sbi_domain_memregion *regA, > ulong regB_end = regB->base + (BIT(regB->order) - 1); > > if ((regB_start <= regA_start) && > - (regA_start < regB_end) && > - (regB_start < regA_end) && > + (regB_start <= regA_end) && > + (regA_start <= regB_end) && > (regA_end <= regB_end)) > return true; > > -- > 2.42.0 >
>On Tue, Oct 10, 2023 at 6:30 AM Inochi Amaoto <inochiama@outlook.com> wrote: >> >> make is_region_subset() check the bound of the domain region, so it can >> handle the domain region with the same base and length. >> >> Signed-off-by: Inochi Amaoto <inochiama@outlook.com> > >Important fact here is that regB_start < regB_end and >regA_start < regA_end because order == 0 region is of >size 1 byte. > >When regB_start == regA_start and regB_end == regA_end, >the following is the outcome of the expressions in the existing >"if ()" check. > >(regB_start <= regA_start) == true >(regA_start < regB_end) == true >(regB_start < regA_end) == true >(regA_end <= regB_end) == true > >Based on the above, I don't see how this patch helps. > Yes, this is my mistake, drop this patch is just OK. >Regards, >Anup > > > >> --- >> lib/sbi/sbi_domain.c | 4 ++-- >> 1 file changed, 2 insertions(+), 2 deletions(-) >> >> diff --git a/lib/sbi/sbi_domain.c b/lib/sbi/sbi_domain.c >> index 283b598..2b9d16f 100644 >> --- a/lib/sbi/sbi_domain.c >> +++ b/lib/sbi/sbi_domain.c >> @@ -185,8 +185,8 @@ static bool is_region_subset(const struct sbi_domain_memregion *regA, >> ulong regB_end = regB->base + (BIT(regB->order) - 1); >> >> if ((regB_start <= regA_start) && >> - (regA_start < regB_end) && >> - (regB_start < regA_end) && >> + (regB_start <= regA_end) && >> + (regA_start <= regB_end) && >> (regA_end <= regB_end)) >> return true; >> >> -- >> 2.42.0 >> >
diff --git a/lib/sbi/sbi_domain.c b/lib/sbi/sbi_domain.c index 283b598..2b9d16f 100644 --- a/lib/sbi/sbi_domain.c +++ b/lib/sbi/sbi_domain.c @@ -185,8 +185,8 @@ static bool is_region_subset(const struct sbi_domain_memregion *regA, ulong regB_end = regB->base + (BIT(regB->order) - 1); if ((regB_start <= regA_start) && - (regA_start < regB_end) && - (regB_start < regA_end) && + (regB_start <= regA_end) && + (regA_start <= regB_end) && (regA_end <= regB_end)) return true;
make is_region_subset() check the bound of the domain region, so it can handle the domain region with the same base and length. Signed-off-by: Inochi Amaoto <inochiama@outlook.com> --- lib/sbi/sbi_domain.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)