| Message ID | 1463312889-19351-5-git-send-email-Yuval.Mintz@qlogic.com |
|---|---|
| State | Accepted, archived |
| Delegated to: | David Miller |
| Headers | show |
diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c index a977f39..c325ee8 100644 --- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c +++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c @@ -3099,6 +3099,9 @@ static int qed_sriov_enable(struct qed_dev *cdev, int num) goto err; } + if (IS_MF_DEFAULT(hwfn)) + limit = MAX_NUM_VFS_BB / hwfn->num_funcs_on_engine; + memset(&sb_cnt_info, 0, sizeof(sb_cnt_info)); qed_int_get_num_sbs(hwfn, &sb_cnt_info); num_sbs = min_t(int, sb_cnt_info.sb_free_blk, limit);
In multi-function modes, PFs are currently limited to using 16 VFs - But that limitation would also currently apply in case there's a single PCI function exposed, where no such restriction should have existed. This lifts the restriction for the default mode; User should be able to start the maximum number of VFs as appear in the PCI config space. Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com> --- drivers/net/ethernet/qlogic/qed/qed_sriov.c | 3 +++ 1 file changed, 3 insertions(+)