| Message ID | 61c985987cae6571bd25b51d414b09496d80dbe5.1673457839.git.lorenzo@kernel.org (mailing list archive) |
|---|---|
| State | Superseded |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | [net-next] net: ethernet: mtk_wed: get rid of queue lock for tx queue | expand |
On Wed, 11 Jan 2023 18:26:29 +0100 Lorenzo Bianconi wrote:
> This patch is based on the following patch not applied yet:
The patchwork automation does not support any sort of patch dependency
tracking (it's probably quite doable, some special tag pointing at a
series ID in patchwork, but I ain't got the time..)
You'll have to repost.
> On Wed, 11 Jan 2023 18:26:29 +0100 Lorenzo Bianconi wrote: > > This patch is based on the following patch not applied yet: > > The patchwork automation does not support any sort of patch dependency > tracking (it's probably quite doable, some special tag pointing at a > series ID in patchwork, but I ain't got the time..) > You'll have to repost. ack, no worries. Regards, Lorenzo
diff --git a/drivers/net/ethernet/mediatek/mtk_wed_wo.c b/drivers/net/ethernet/mediatek/mtk_wed_wo.c index d32b86499896..69fba29055e9 100644 --- a/drivers/net/ethernet/mediatek/mtk_wed_wo.c +++ b/drivers/net/ethernet/mediatek/mtk_wed_wo.c @@ -258,7 +258,6 @@ mtk_wed_wo_queue_alloc(struct mtk_wed_wo *wo, struct mtk_wed_wo_queue *q, int n_desc, int buf_size, int index, struct mtk_wed_wo_queue_regs *regs) { - spin_lock_init(&q->lock); q->regs = *regs; q->n_desc = n_desc; q->buf_size = buf_size; @@ -290,7 +289,6 @@ mtk_wed_wo_queue_tx_clean(struct mtk_wed_wo *wo, struct mtk_wed_wo_queue *q) struct page *page; int i; - spin_lock_bh(&q->lock); for (i = 0; i < q->n_desc; i++) { struct mtk_wed_wo_queue_entry *entry = &q->entry[i]; @@ -299,7 +297,6 @@ mtk_wed_wo_queue_tx_clean(struct mtk_wed_wo *wo, struct mtk_wed_wo_queue *q) skb_free_frag(entry->buf); entry->buf = NULL; } - spin_unlock_bh(&q->lock); if (!q->cache.va) return; @@ -347,8 +344,6 @@ int mtk_wed_wo_queue_tx_skb(struct mtk_wed_wo *wo, struct mtk_wed_wo_queue *q, int ret = 0, index; u32 ctrl; - spin_lock_bh(&q->lock); - q->tail = mtk_wed_mmio_r32(wo, q->regs.dma_idx); index = (q->head + 1) % q->n_desc; if (q->tail == index) { @@ -379,8 +374,6 @@ int mtk_wed_wo_queue_tx_skb(struct mtk_wed_wo *wo, struct mtk_wed_wo_queue *q, mtk_wed_wo_queue_kick(wo, q, q->head); mtk_wed_wo_kickout(wo); out: - spin_unlock_bh(&q->lock); - dev_kfree_skb(skb); return ret; diff --git a/drivers/net/ethernet/mediatek/mtk_wed_wo.h b/drivers/net/ethernet/mediatek/mtk_wed_wo.h index c8fb85795864..dbcf42ce9173 100644 --- a/drivers/net/ethernet/mediatek/mtk_wed_wo.h +++ b/drivers/net/ethernet/mediatek/mtk_wed_wo.h @@ -211,7 +211,6 @@ struct mtk_wed_wo_queue { struct mtk_wed_wo_queue_regs regs; struct page_frag_cache cache; - spinlock_t lock; struct mtk_wed_wo_queue_desc *desc; dma_addr_t desc_dma;
Similar to MTK Wireless Ethernet Dispatcher (WED) MCU rx queue, we do not need to protect WED MCU tx queue with a spin lock since the tx queue is accessed in the two following routines: - mtk_wed_wo_queue_tx_skb(): it is run at initialization and during mt7915 normal operation. Moreover MCU messages are serialized through MCU mutex. - mtk_wed_wo_queue_tx_clean(): it runs just at mt7915 driver module unload when no more messages are sent to the MCU. Remove tx queue spinlock. Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org> --- This patch is based on the following patch not applied yet: https://lore.kernel.org/netdev/36ec3b729542ea60898471d890796f745479ba32.1673342990.git.lorenzo@kernel.org/ --- drivers/net/ethernet/mediatek/mtk_wed_wo.c | 7 ------- drivers/net/ethernet/mediatek/mtk_wed_wo.h | 1 - 2 files changed, 8 deletions(-)