On 04/26/2017 04:38 PM, Iyappan Subramanian wrote: > From: Quan Nguyen <qngu...@apm.com> > > This patch adds lock to protect indirect mac access sequence. > > Signed-off-by: Quan Nguyen <qngu...@apm.com> > Signed-off-by: Iyappan Subramanian <isubraman...@apm.com> > --- > drivers/net/ethernet/apm/xgene/xgene_enet_hw.c | 2 ++ > drivers/net/ethernet/apm/xgene/xgene_enet_main.c | 1 + > drivers/net/ethernet/apm/xgene/xgene_enet_main.h | 1 + > drivers/net/ethernet/apm/xgene/xgene_enet_sgmac.c | 7 ++++++- > drivers/net/ethernet/apm/xgene/xgene_enet_xgmac.c | 2 ++ > 5 files changed, 12 insertions(+), 1 deletion(-) > > diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c > b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c > index 2a835e0..3697ba7 100644 > --- a/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c > +++ b/drivers/net/ethernet/apm/xgene/xgene_enet_hw.c > @@ -365,9 +365,11 @@ static void xgene_enet_rd_mcx_mac(struct > xgene_enet_pdata *pdata, > cmd = pdata->mcx_mac_addr + MAC_COMMAND_REG_OFFSET; > cmd_done = pdata->mcx_mac_addr + MAC_COMMAND_DONE_REG_OFFSET; > > + spin_lock(&pdata->mac_lock); > if (!xgene_enet_rd_indirect(addr, rd, cmd, cmd_done, rd_addr, rd_data)) > netdev_err(pdata->ndev, "MCX mac read failed, addr: %04x\n", > rd_addr); > + spin_unlock(&pdata->mac_lock);
Why not fold the spinlock manipulation within xgenet_enet_rd_indirect()? That way the caller does not have to know that acquiring the spinlock is required and you avoid potential bugs in the future where you are missing the spinlock on indirect accesses? -- Florian
