Message ID | 20230630132159.376995-9-richard.henderson@linaro.org |
---|---|
State | Superseded |
Headers | show |
Series | linux-user: mmap range fixes | expand |
On Fri, Jun 30, 2023 at 7:22 AM Richard Henderson < richard.henderson@linaro.org> wrote: > The previous check returned -1 when any page within > [start, start+len) is unmapped, not when all are unmapped. > > Cc: Warner Losh <imp@bsdimp.com> > Cc: Kyle Evans <kevans@freebsd.org> > Signed-off-by: Richard Henderson <richard.henderson@linaro.org> > --- > bsd-user/mmap.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/bsd-user/mmap.c b/bsd-user/mmap.c > index 565b9f97ed..07b5b8055e 100644 > --- a/bsd-user/mmap.c > +++ b/bsd-user/mmap.c > @@ -609,7 +609,7 @@ abi_long target_mmap(abi_ulong start, abi_ulong len, > int prot, > } > > /* Reject the mapping if any page within the range is mapped */ > - if ((flags & MAP_EXCL) && page_check_range(start, len, 0) < 0) { > + if ((flags & MAP_EXCL) && !page_check_range_empty(start, end - > 1)) { > Won't computing end overflow if start is 1<<32 - len? Then subtracting one here would be too late. Warner > errno = EINVAL; > goto fail; > } > -- > 2.34.1 > >
On 6/30/23 20:46, Warner Losh wrote: > > > On Fri, Jun 30, 2023 at 7:22 AM Richard Henderson <richard.henderson@linaro.org > <mailto:richard.henderson@linaro.org>> wrote: > > The previous check returned -1 when any page within > [start, start+len) is unmapped, not when all are unmapped. > > Cc: Warner Losh <imp@bsdimp.com <mailto:imp@bsdimp.com>> > Cc: Kyle Evans <kevans@freebsd.org <mailto:kevans@freebsd.org>> > Signed-off-by: Richard Henderson <richard.henderson@linaro.org > <mailto:richard.henderson@linaro.org>> > --- > bsd-user/mmap.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/bsd-user/mmap.c b/bsd-user/mmap.c > index 565b9f97ed..07b5b8055e 100644 > --- a/bsd-user/mmap.c > +++ b/bsd-user/mmap.c > @@ -609,7 +609,7 @@ abi_long target_mmap(abi_ulong start, abi_ulong len, int prot, > } > > /* Reject the mapping if any page within the range is mapped */ > - if ((flags & MAP_EXCL) && page_check_range(start, len, 0) < 0) { > + if ((flags & MAP_EXCL) && !page_check_range_empty(start, end - 1)) { > > > Won't computing end overflow if start is 1<<32 - len? Then subtracting one here would be > too late. -fwrapv, so fully twos-compliment. So start per above, end == 0, end - 1 == UINT32_MAX. See also the rest of the linux-user code, wherein I work with "last == end - 1". Which has the advantage of preserving comparison order: start < last. r~
diff --git a/bsd-user/mmap.c b/bsd-user/mmap.c index 565b9f97ed..07b5b8055e 100644 --- a/bsd-user/mmap.c +++ b/bsd-user/mmap.c @@ -609,7 +609,7 @@ abi_long target_mmap(abi_ulong start, abi_ulong len, int prot, } /* Reject the mapping if any page within the range is mapped */ - if ((flags & MAP_EXCL) && page_check_range(start, len, 0) < 0) { + if ((flags & MAP_EXCL) && !page_check_range_empty(start, end - 1)) { errno = EINVAL; goto fail; }
The previous check returned -1 when any page within [start, start+len) is unmapped, not when all are unmapped. Cc: Warner Losh <imp@bsdimp.com> Cc: Kyle Evans <kevans@freebsd.org> Signed-off-by: Richard Henderson <richard.henderson@linaro.org> --- bsd-user/mmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)