| Message ID | 20200903184850.53055-2-andriy.shevchenko@linux.intel.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [v1,1/4] Bluetooth: hci_intel: enable on new platform | expand |
Hi Andy, > Sparse rightfully complains: > > hci_intel.c:696:26: warning: cast to restricted __le16 > hci_intel.c:701:26: warning: cast to restricted __le16 > hci_intel.c:702:26: warning: cast to restricted __le16 > hci_intel.c:703:26: warning: cast to restricted __le16 > hci_intel.c:725:26: warning: cast to restricted __le16 > hci_intel.c:730:26: warning: cast to restricted __le16 > hci_intel.c:731:26: warning: cast to restricted __le16 > hci_intel.c:732:26: warning: cast to restricted __le16 > > because we access non-restricted types with le16_to_cpu(). > More confusion is added by using above against u8. On big-endian > architecture we will get all zeroes. I bet it's not what should be > in such case. > > Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> > --- > drivers/bluetooth/hci_intel.c | 14 ++++---------- > 1 file changed, 4 insertions(+), 10 deletions(-) patch has been applied to bluetooth-next tree. Regards Marcel
diff --git a/drivers/bluetooth/hci_intel.c b/drivers/bluetooth/hci_intel.c index 703d774be5a6..50e4fc6813c2 100644 --- a/drivers/bluetooth/hci_intel.c +++ b/drivers/bluetooth/hci_intel.c @@ -693,14 +693,11 @@ static int intel_setup(struct hci_uart *hu) case 0x0b: /* SfP */ case 0x0c: /* WsP */ snprintf(fwname, sizeof(fwname), "intel/ibt-%u-%u.sfi", - le16_to_cpu(ver.hw_variant), - le16_to_cpu(params.dev_revid)); + ver.hw_variant, le16_to_cpu(params.dev_revid)); break; case 0x12: /* ThP */ snprintf(fwname, sizeof(fwname), "intel/ibt-%u-%u-%u.sfi", - le16_to_cpu(ver.hw_variant), - le16_to_cpu(ver.hw_revision), - le16_to_cpu(ver.fw_revision)); + ver.hw_variant, ver.hw_revision, ver.fw_revision); break; default: bt_dev_err(hdev, "Unsupported Intel hardware variant (%u)", @@ -722,14 +719,11 @@ static int intel_setup(struct hci_uart *hu) case 0x0b: /* SfP */ case 0x0c: /* WsP */ snprintf(fwname, sizeof(fwname), "intel/ibt-%u-%u.ddc", - le16_to_cpu(ver.hw_variant), - le16_to_cpu(params.dev_revid)); + ver.hw_variant, le16_to_cpu(params.dev_revid)); break; case 0x12: /* ThP */ snprintf(fwname, sizeof(fwname), "intel/ibt-%u-%u-%u.ddc", - le16_to_cpu(ver.hw_variant), - le16_to_cpu(ver.hw_revision), - le16_to_cpu(ver.fw_revision)); + ver.hw_variant, ver.hw_revision, ver.fw_revision); break; default: bt_dev_err(hdev, "Unsupported Intel hardware variant (%u)",
Sparse rightfully complains: hci_intel.c:696:26: warning: cast to restricted __le16 hci_intel.c:701:26: warning: cast to restricted __le16 hci_intel.c:702:26: warning: cast to restricted __le16 hci_intel.c:703:26: warning: cast to restricted __le16 hci_intel.c:725:26: warning: cast to restricted __le16 hci_intel.c:730:26: warning: cast to restricted __le16 hci_intel.c:731:26: warning: cast to restricted __le16 hci_intel.c:732:26: warning: cast to restricted __le16 because we access non-restricted types with le16_to_cpu(). More confusion is added by using above against u8. On big-endian architecture we will get all zeroes. I bet it's not what should be in such case. Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> --- drivers/bluetooth/hci_intel.c | 14 ++++---------- 1 file changed, 4 insertions(+), 10 deletions(-)