On Fri, Sep 23, 2016 at 6:30 AM, Stefan Assmann <sassm...@kpanic.de> wrote:
> pf->msg_enable is a bitmask, therefore assigning the value of the
> "debug" parameter is wrong. It is initialized again later in
> i40e_sw_init() so it didn't cause any problem, except that we missed
> early debug messages. Moved the initialization and assigned
> pf->hw.debug_mask the bitmask as that's what the driver actually uses
> in i40e_debug(). Otherwise the debug parameter is just a noop.
>
> Fixes: 5b5faa4 ("i40e: enable debug earlier")
>
> Signed-off-by: Stefan Assmann <sassm...@kpanic.de>
> ---
> drivers/net/ethernet/intel/i40e/i40e_main.c | 16 +++++++---------
> 1 file changed, 7 insertions(+), 9 deletions(-)
>
> diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c
> b/drivers/net/ethernet/intel/i40e/i40e_main.c
> index 56369761..f972f0d 100644
> --- a/drivers/net/ethernet/intel/i40e/i40e_main.c
> +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
> @@ -8498,11 +8498,6 @@ static int i40e_sw_init(struct i40e_pf *pf)
> int err = 0;
> int size;
>
> - pf->msg_enable = netif_msg_init(debug,
> - NETIF_MSG_DRV |
> - NETIF_MSG_PROBE |
> - NETIF_MSG_LINK);
> -
> /* Set default capability flags */
> pf->flags = I40E_FLAG_RX_CSUM_ENABLED |
> I40E_FLAG_MSI_ENABLED |
> @@ -10812,10 +10807,13 @@ static int i40e_probe(struct pci_dev *pdev, const
> struct pci_device_id *ent)
> mutex_init(&hw->aq.asq_mutex);
> mutex_init(&hw->aq.arq_mutex);
>
> - if (debug != -1) {
> - pf->msg_enable = pf->hw.debug_mask;
> - pf->msg_enable = debug;
> - }
> + /* enable debug prints if requested */
> + pf->msg_enable = netif_msg_init(debug,
> + NETIF_MSG_DRV |
> + NETIF_MSG_PROBE |
> + NETIF_MSG_LINK);
> + if (debug != -1)
> + pf->hw.debug_mask = pf->msg_enable;
>
> /* do a special CORER for clearing PXE mode once at init */
> if (hw->revision_id == 0 &&
The patch is broken, mainly because the code was already broken. The
flags in pf->hw.debug_mask are in no way related to pf->msg_enable.
For now just use the default mask provided to populate
pf->hw.debug_mask and then the msg_enable portion is fine.
- Alex
