| Message ID | 1441619586-12098-2-git-send-email-emil.l.velikov@gmail.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Mon, Sep 07, 2015 at 10:53:06AM +0100, Emil Velikov wrote: > As going through the modetest patches for atomic support I've noticed > that if we pass NULL for the drmModeAtomicReqPtr argument we'll crash. > > So let's handle things appropriately if the user forgot to check the > return value of drmModeAtomicAlloc and drmModeAtomicDuplicate or made a > typo somewhere along the way. I'm not sure hand-holding the user to such an extent is actually useful. OTOH I guess one NULL check per function call isn't all that expensive either. > > Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> > Cc: Rob Clark <robclark@freedesktop.org> > Cc: Daniel Stone <daniels@collabora.com> > Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> > --- > xf86drmMode.c | 17 ++++++++++++++++- > 1 file changed, 16 insertions(+), 1 deletion(-) > > diff --git a/xf86drmMode.c b/xf86drmMode.c > index 23800dd..ab6b519 100644 > --- a/xf86drmMode.c > +++ b/xf86drmMode.c > @@ -1189,6 +1189,9 @@ drmModeAtomicReqPtr drmModeAtomicDuplicate(drmModeAtomicReqPtr old) > { > drmModeAtomicReqPtr new; > > + if (!old) > + return NULL; > + > new = drmMalloc(sizeof *new); > if (!new) > return NULL; > @@ -1213,6 +1216,9 @@ drmModeAtomicReqPtr drmModeAtomicDuplicate(drmModeAtomicReqPtr old) > > int drmModeAtomicMerge(drmModeAtomicReqPtr base, drmModeAtomicReqPtr augment) > { > + if (!base) > + return -EINVAL; > + > if (!augment || augment->cursor == 0) > return 0; > > @@ -1239,12 +1245,15 @@ int drmModeAtomicMerge(drmModeAtomicReqPtr base, drmModeAtomicReqPtr augment) > > int drmModeAtomicGetCursor(drmModeAtomicReqPtr req) > { > + if (!req) > + return -EINVAL; > return req->cursor; > } > > void drmModeAtomicSetCursor(drmModeAtomicReqPtr req, int cursor) > { > - req->cursor = cursor; > + if (req) > + req->cursor = cursor; > } > > int drmModeAtomicAddProperty(drmModeAtomicReqPtr req, > @@ -1252,6 +1261,9 @@ int drmModeAtomicAddProperty(drmModeAtomicReqPtr req, > uint32_t property_id, > uint64_t value) > { > + if (!req) > + return -EINVAL; > + > if (req->cursor >= req->size_items) { > drmModeAtomicReqItemPtr new; > > @@ -1309,6 +1321,9 @@ int drmModeAtomicCommit(int fd, drmModeAtomicReqPtr req, uint32_t flags, > int obj_idx = -1; > int ret = -1; > > + if (!req) > + return -EINVAL; > + > if (req->cursor == 0) > return 0; > > -- > 2.5.0
On 07/09/15 14:06, Ville Syrjälä wrote: > On Mon, Sep 07, 2015 at 10:53:06AM +0100, Emil Velikov wrote: >> As going through the modetest patches for atomic support I've noticed >> that if we pass NULL for the drmModeAtomicReqPtr argument we'll crash. >> >> So let's handle things appropriately if the user forgot to check the >> return value of drmModeAtomicAlloc and drmModeAtomicDuplicate or made a >> typo somewhere along the way. > > I'm not sure hand-holding the user to such an extent is actually useful. From what I've gathered "smoke-testing" is not meant to be smart. > OTOH I guess one NULL check per function call isn't all that expensive > either. > This is pretty much the reason why I bothered :) Although if anyone feels strongly against it I'll drop the patch. Thanks Emil
diff --git a/xf86drmMode.c b/xf86drmMode.c index 23800dd..ab6b519 100644 --- a/xf86drmMode.c +++ b/xf86drmMode.c @@ -1189,6 +1189,9 @@ drmModeAtomicReqPtr drmModeAtomicDuplicate(drmModeAtomicReqPtr old) { drmModeAtomicReqPtr new; + if (!old) + return NULL; + new = drmMalloc(sizeof *new); if (!new) return NULL; @@ -1213,6 +1216,9 @@ drmModeAtomicReqPtr drmModeAtomicDuplicate(drmModeAtomicReqPtr old) int drmModeAtomicMerge(drmModeAtomicReqPtr base, drmModeAtomicReqPtr augment) { + if (!base) + return -EINVAL; + if (!augment || augment->cursor == 0) return 0; @@ -1239,12 +1245,15 @@ int drmModeAtomicMerge(drmModeAtomicReqPtr base, drmModeAtomicReqPtr augment) int drmModeAtomicGetCursor(drmModeAtomicReqPtr req) { + if (!req) + return -EINVAL; return req->cursor; } void drmModeAtomicSetCursor(drmModeAtomicReqPtr req, int cursor) { - req->cursor = cursor; + if (req) + req->cursor = cursor; } int drmModeAtomicAddProperty(drmModeAtomicReqPtr req, @@ -1252,6 +1261,9 @@ int drmModeAtomicAddProperty(drmModeAtomicReqPtr req, uint32_t property_id, uint64_t value) { + if (!req) + return -EINVAL; + if (req->cursor >= req->size_items) { drmModeAtomicReqItemPtr new; @@ -1309,6 +1321,9 @@ int drmModeAtomicCommit(int fd, drmModeAtomicReqPtr req, uint32_t flags, int obj_idx = -1; int ret = -1; + if (!req) + return -EINVAL; + if (req->cursor == 0) return 0;
As going through the modetest patches for atomic support I've noticed that if we pass NULL for the drmModeAtomicReqPtr argument we'll crash. So let's handle things appropriately if the user forgot to check the return value of drmModeAtomicAlloc and drmModeAtomicDuplicate or made a typo somewhere along the way. Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Cc: Rob Clark <robclark@freedesktop.org> Cc: Daniel Stone <daniels@collabora.com> Signed-off-by: Emil Velikov <emil.l.velikov@gmail.com> --- xf86drmMode.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-)