On Tue, Jul 21, 2026 at 11:28 AM Ed Maste <[email protected]> wrote: > > The branch main has been updated by emaste: > > URL: > https://cgit.FreeBSD.org/src/commit/?id=caabdb3aefdc45cae90203210034086801fa9005 > > commit caabdb3aefdc45cae90203210034086801fa9005 > Author: Ed Maste <[email protected]> > AuthorDate: 2026-07-21 18:10:00 +0000 > Commit: Ed Maste <[email protected]> > CommitDate: 2026-07-21 18:27:47 +0000 > > contigmalloc.9: Note that M_WAITOK may still return NULL > > Reviewed by: markj, bapt > Sponsored by: The FreeBSD Foundation > Differential Revision: https://reviews.freebsd.org/D58382 > --- > share/man/man9/contigmalloc.9 | 11 ++++++++++- > 1 file changed, 10 insertions(+), 1 deletion(-) > > diff --git a/share/man/man9/contigmalloc.9 b/share/man/man9/contigmalloc.9 > index 2e5d55ae8ba1..a9ebaf100eb9 100644 > --- a/share/man/man9/contigmalloc.9 > +++ b/share/man/man9/contigmalloc.9 > @@ -23,7 +23,7 @@ > .\" ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF > THE > .\" POSSIBILITY OF SUCH DAMAGE. > .\" > -.Dd July 26, 2024 > +.Dd July 21, 2026 > .Dt CONTIGMALLOC 9 > .Os > .Sh NAME > @@ -124,6 +124,15 @@ function returns a kernel virtual address if allocation > succeeds, > or > .Dv NULL > otherwise. > +Note that in contrast with > +.Xr malloc 9 , > +.Fn contigmalloc > +may return > +.Dv NULL > +even if > +.Dv M_WAITOK > +is specified, if no physically congiguous range is available that meets the > +specified constraints. > .Sh EXAMPLES > .Bd -literal > void *p; >
But is this desired and intentional behavior, or simply what the code does? Paging through grep of contigmalloc with M_WAITOK, it appears to be a mix of callers that handle and do not handle a NULL return. Some of them just handle it by panicking. Is someone planning to audit M_WAITOK callers? > +is specified, if no physically congiguous range is available that meets the Typo: "congiguous". Ryan
