| Message ID | 20220702151020.2524220-1-trix@redhat.com (mailing list archive) |
|---|---|
| State | Awaiting Upstream |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | iwlwifi: mvm: return an error if setting tbl_rev fails | expand |
| Context | Check | Description |
|---|---|---|
| netdev/tree_selection | success | Not a local patch |
On Sat, Jul 2, 2022 at 8:10 AM Tom Rix <trix@redhat.com> wrote: > > clang static analysis reports > drivers/net/wireless/intel/iwlwifi/fw/acpi.c:1048:17: warning: Assigned value is garbage or undefined [core.uninitialized.Assign] > fwrt->ppag_ver = tbl_rev; > ^ ~~~~~~~ > tbl_rev is optionaly set by a series of calls to iwl_acpi_get_wifi_pkg() > and then jumping to the read_table when a call is successful. The > error case when all the call fails is not handled. On all failed, > the code flow falls through to the read_table label. Add an error > handler for the all fail case. > > Fixes: e8e10a37c51c ("iwlwifi: acpi: move ppag code from mvm to fw/acpi") > Signed-off-by: Tom Rix <trix@redhat.com> > --- > drivers/net/wireless/intel/iwlwifi/fw/acpi.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c > index e6d64152c81a..1ef1e26c3206 100644 > --- a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c > +++ b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c > @@ -1044,6 +1044,9 @@ int iwl_acpi_get_ppag_table(struct iwl_fw_runtime *fwrt) > goto read_table; Thanks for the patch! I wonder why the pre-existing code had a goto to a label that was the subsequent statement? That's strange; maybe something was in between them before, was removed, and that wasn't cleaned up? I think the whole `if (!IS_ERR(wifi_pkg))` block would be clearer if this was: ... wifi_pkg = iwl_acpi_get_wifi_pkg(fwrt->dev, data, ACPI_PPAG_WIFI_DATA_SIZE_V1, &tbl_rev); if (IS_ERR(wifi_pkg) || tbl_rev != 0) { ret = -EINVAL; goto out_free; } num_sub_bands = IWL_NUM_SUB_BANDS_V1; IWL_DEBUG_RADIO(fwrt, "Reading PPAG table v1 (tbl_rev=0)\n"); read_table: ... rather than the existing spaghetti. Apologies for my formatting. > } > > + ret = -EINVAL; > + goto out_free; > + > read_table: > fwrt->ppag_ver = tbl_rev; > flags = &wifi_pkg->package.elements[1]; > -- > 2.27.0 >
diff --git a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c index e6d64152c81a..1ef1e26c3206 100644 --- a/drivers/net/wireless/intel/iwlwifi/fw/acpi.c +++ b/drivers/net/wireless/intel/iwlwifi/fw/acpi.c @@ -1044,6 +1044,9 @@ int iwl_acpi_get_ppag_table(struct iwl_fw_runtime *fwrt) goto read_table; } + ret = -EINVAL; + goto out_free; + read_table: fwrt->ppag_ver = tbl_rev; flags = &wifi_pkg->package.elements[1];
clang static analysis reports drivers/net/wireless/intel/iwlwifi/fw/acpi.c:1048:17: warning: Assigned value is garbage or undefined [core.uninitialized.Assign] fwrt->ppag_ver = tbl_rev; ^ ~~~~~~~ tbl_rev is optionaly set by a series of calls to iwl_acpi_get_wifi_pkg() and then jumping to the read_table when a call is successful. The error case when all the call fails is not handled. On all failed, the code flow falls through to the read_table label. Add an error handler for the all fail case. Fixes: e8e10a37c51c ("iwlwifi: acpi: move ppag code from mvm to fw/acpi") Signed-off-by: Tom Rix <trix@redhat.com> --- drivers/net/wireless/intel/iwlwifi/fw/acpi.c | 3 +++ 1 file changed, 3 insertions(+)