Re: [PATCH v5 02/10] net: stmmac: Use interrupt mode INTM=1 for per channel irq
From: "Russell King (Oracle)" <linux@armlinux.org.uk>
Date: 2025-10-25 19:01:41
Also in:
linux-devicetree, lkml, netdev
On Fri, Oct 24, 2025 at 01:49:54PM +0200, Steffen Trumtrar wrote:
quoted hunk ↗ jump to hunk
From: Teoh Ji Sheng <redacted> commit 6ccf12ae111e ("net: stmmac: use interrupt mode INTM=1 for multi-MSI") is introduced for platform that uses MSI. Similar approach is taken to enable per channel interrupt that uses shared peripheral interrupt (SPI), so only per channel TX and RX intr (TI/RI) are handled by TX/RX ISR without calling common interrupt ISR. TX/RX NORMAL interrupts check is now decoupled, since NIS bit is not asserted for any TI/RI events when INTM=1. Signed-off-by: Teoh Ji Sheng <redacted> Signed-off-by: Steffen Trumtrar <redacted> --- drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h | 3 +++ drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c | 10 +++++++++- .../net/ethernet/stmicro/stmmac/stmmac_platform.c | 20 ++++++++++++++++++++ include/linux/stmmac.h | 2 ++ 4 files changed, 34 insertions(+), 1 deletion(-)diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h index 0d408ee17f337..64b533207e4a6 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h +++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2.h@@ -326,6 +326,9 @@ /* DMA Registers */ #define XGMAC_DMA_MODE 0x00003000 #define XGMAC_SWR BIT(0) +#define DMA_MODE_INTM_MASK GENMASK(13, 12) +#define DMA_MODE_INTM_SHIFT 12 +#define DMA_MODE_INTM_MODE1 0x1 #define XGMAC_DMA_SYSBUS_MODE 0x00003004 #define XGMAC_WR_OSR_LMT GENMASK(29, 24) #define XGMAC_WR_OSR_LMT_SHIFT 24diff --git a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c index 4d6bb995d8d84..1e9ee1f10f0ef 100644 --- a/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c +++ b/drivers/net/ethernet/stmicro/stmmac/dwxgmac2_dma.c@@ -31,6 +31,13 @@ static void dwxgmac2_dma_init(void __iomem *ioaddr, value |= XGMAC_EAME; writel(value, ioaddr + XGMAC_DMA_SYSBUS_MODE); + + if (dma_cfg->multi_irq_en) { + value = readl(ioaddr + XGMAC_DMA_MODE); + value &= ~DMA_MODE_INTM_MASK; + value |= (DMA_MODE_INTM_MODE1 << DMA_MODE_INTM_SHIFT);
No need for these parens. What is on the right hand side of |= is its own expression and can't be interpreted any other way.
quoted hunk ↗ jump to hunk
+ writel(value, ioaddr + XGMAC_DMA_MODE); + } } static void dwxgmac2_dma_init_chan(struct stmmac_priv *priv,@@ -359,13 +366,14 @@ static int dwxgmac2_dma_interrupt(struct stmmac_priv *priv, } } - /* TX/RX NORMAL interrupts */ + /* RX NORMAL interrupts */ if (likely(intr_status & XGMAC_RI)) { u64_stats_update_begin(&stats->syncp); u64_stats_inc(&stats->rx_normal_irq_n[chan]); u64_stats_update_end(&stats->syncp); ret |= handle_rx; } + /* TX NORMAL interrupts */ if (likely(intr_status & (XGMAC_TI | XGMAC_TBU))) { u64_stats_update_begin(&stats->syncp); u64_stats_inc(&stats->tx_normal_irq_n[chan]);diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c index 27bcaae07a7f2..cfa82b8e04b94 100644 --- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c +++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c@@ -607,6 +607,8 @@ stmmac_probe_config_dt(struct platform_device *pdev, u8 *mac) dma_cfg->fixed_burst = of_property_read_bool(np, "snps,fixed-burst"); dma_cfg->mixed_burst = of_property_read_bool(np, "snps,mixed-burst"); + dma_cfg->multi_irq_en = of_property_read_bool(np, "snps,multi-irq-en"); + plat->force_thresh_dma_mode = of_property_read_bool(np, "snps,force_thresh_dma_mode"); if (plat->force_thresh_dma_mode && plat->force_sf_dma_mode) { plat->force_sf_dma_mode = 0;@@ -737,6 +739,8 @@ EXPORT_SYMBOL_GPL(stmmac_pltfr_find_clk); int stmmac_get_platform_resources(struct platform_device *pdev, struct stmmac_resources *stmmac_res) { + char irq_name[11]; + int i; memset(stmmac_res, 0, sizeof(*stmmac_res));
We normally want to see a blank line between local variable declarations and code.
quoted hunk ↗ jump to hunk
/* Get IRQ information early to have an ability to ask for deferred@@ -746,6 +750,22 @@ int stmmac_get_platform_resources(struct platform_device *pdev, if (stmmac_res->irq < 0) return stmmac_res->irq; + /* For RX Channel */ + for (i = 0; i < MTL_MAX_RX_QUEUES; i++) { + sprintf(irq_name, "%s%d", "macirq_rx", i); + stmmac_res->rx_irq[i] = platform_get_irq_byname(pdev, irq_name); + if (stmmac_res->rx_irq[i] < 0) + break; + } + + /* For TX Channel */ + for (i = 0; i < MTL_MAX_TX_QUEUES; i++) { + sprintf(irq_name, "%s%d", "macirq_tx", i); + stmmac_res->tx_irq[i] = platform_get_irq_byname(pdev, irq_name); + if (stmmac_res->tx_irq[i] < 0) + break; + } +
It looks like multi-irq is a dwxgmac2 thing, should this be conditional on (a) multi_irq_en being set, and (b) should parsing multi_irq_en be conditional on dwxgmac2, (c) should the binding only allow snps,multi-irq-en if a dwxgmac2 compatible is indicated? -- RMK's Patch system: https://www.armlinux.org.uk/developer/patches/ FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last!