[PATCH 1/2] RDMA/ionic: Cap eq_count to the eth driver's interrupt vector budget
Abhijit Gangurde <[email protected]>
| Newsgroups | org.kernel.vger.linux-rdma,org.kernel.vger.linux-kernel,org.kernel.vger.stable |
|---|---|
| Message-ID | <[email protected]> |
From: Brett Creeley <[email protected]> ionic_fill_lif_cfg() reads eq_count from firmware uncapped, but the eth driver only reserves ionic->neqs_per_lif MSI-X vectors for RDMA event queues. Since ionic_rdma probes via the auxiliary bus before the netdev is brought up, it can exhaust the shared interrupt bitmap, causing ionic_open() to fail with -ENOSPC when allocating rx/tx interrupts. Cap RDMA eq_count to neqs_per_lif, which is populated by ionic_lif_size() at PCI probe before the RDMA aux device registers. Fixes: 8d765af51a09 ("RDMA/ionic: Register auxiliary module for ionic ethernet adapter") Cc: [email protected] Signed-off-by: Brett Creeley <[email protected]> Signed-off-by: Abhijit Gangurde <[email protected]> --- drivers/infiniband/hw/ionic/ionic_lif_cfg.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/ionic/ionic_lif_cfg.c b/drivers/infiniband/hw/ionic/ionic_lif_cfg.c index f3cd281c3a2f..45846dc2e89b 100644 --- a/drivers/infiniband/hw/ionic/ionic_lif_cfg.c +++ b/drivers/infiniband/hw/ionic/ionic_lif_cfg.c @@ -70,7 +70,7 @@ void ionic_fill_lif_cfg(struct ionic_lif *lif, struct ionic_lif_cfg *cfg) * eq_count is tunable; see ionic_eq_count */ cfg->aq_count = le32_to_cpu(ident->rdma.aq_qtype.qid_count); - cfg->eq_count = le32_to_cpu(ident->rdma.eq_qtype.qid_count); + cfg->eq_count = lif->ionic->neqs_per_lif; cfg->cq_count = le32_to_cpu(ident->rdma.cq_qtype.qid_count); cfg->qp_count = le32_to_cpu(ident->rdma.sq_qtype.qid_count); cfg->dbid_count = le32_to_cpu(lif->ionic->ident.dev.ndbpgs_per_lif); -- 2.43.0