greybus: fix an allocation flag bug

We allocate message buffers with GFP_KERNEL allocation flags if
possible.  However when an incoming request message is received we
can be in interrupt context, so we must use GFP_ATOMIC in that case.

The computation of gfp_flags in gb_operation_message_init() is
wrong.  It is needlessly using GFP_ATOMIC when allocating outbound
response buffers.  Fix the flawed logic.

Change the name of "data_out" to be "outbound" to be consistent with
usage elsewhere.  (Data/messages are "inbound" or "outbound";
requests are "incoming" or "outgoing".)

Signed-off-by: Alex Elder <elder@linaro.org>
Signed-off-by: Greg Kroah-Hartman <greg@kroah.com>
This commit is contained in:
Alex Elder
2014-11-18 13:26:38 -06:00
committed by Greg Kroah-Hartman
parent 374e6a269c
commit 4f83b8d349

View File

@@ -213,14 +213,14 @@ static void operation_timeout(struct work_struct *work)
*/
static int gb_operation_message_init(struct gb_operation *operation,
u8 type, size_t size,
bool request, bool data_out)
bool request, bool outbound)
{
struct gb_connection *connection = operation->connection;
struct greybus_host_device *hd = connection->hd;
struct gb_message *message;
struct gb_operation_msg_hdr *header;
struct gbuf *gbuf;
gfp_t gfp_flags = data_out ? GFP_KERNEL : GFP_ATOMIC;
gfp_t gfp_flags = request && !outbound ? GFP_ATOMIC : GFP_KERNEL;
u16 dest_cport_id;
int ret;
@@ -236,7 +236,7 @@ static int gb_operation_message_init(struct gb_operation *operation,
}
gbuf = &message->gbuf;
if (data_out)
if (outbound)
dest_cport_id = connection->interface_cport_id;
else
dest_cport_id = CPORT_ID_BAD;