[PATCH v5 00/10] arm64: dts: socfpga: agilex5: enable network and add new board

[PATCH v5 00/10] arm64: dts: socfpga: agilex5: enable network and add new board

[Steffen Trumtrar]

Add an initial devicetree for a new board (Arrow AXE5-Eagle) and all
needed patches to support the network on current mainline.

Currently only QSPI and network are functional as all other hardware
currently lacks mainline support.

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
---
Changes in v5:
- remove binding conversion
- rebase to 6.18-rc1
- add sofpga_agilex5.dtsi patches
- add stmmac patches
- add dwxgmac compatible to have usable network
- Link to v4: https://lore.kernel.org/r/20250205-v6-12-topic-socfpga-agilex5-v4-0-ebf070e2075f@pengutronix.de

Changes in v4:
- extract gmii-to-sgmii-converter binding to pcs subdir
- fix dt_binding_check warnings
- rebase to v6.13-rc1
- Link to v3: https://lore.kernel.org/r/20241205-v6-12-topic-socfpga-agilex5-v3-0-2a8cdf73f50a@pengutronix.de

Changes in v3:
- add socfpga-stmmac-agilex5 compatible
- convert socfpga-dwmac.txt -> yaml
- add Acked-bys
- rebase to v6.13-rc1
- Link to v2: https://lore.kernel.org/r/20241125-v6-12-topic-socfpga-agilex5-v2-0-864256ecc7b2@pengutronix.de

Changes in v2:
- fix node names according to dtb_check
- remove gpio 'status = disabled'
- mdio0: remove setting of adi,[rt]x-internal-delay-ps. 2000 is the
  default value
- add Acked-by to dt-binding
- Link to v1: https://lore.kernel.org/r/20241030-v6-12-topic-socfpga-agilex5-v1-0-b2b67780e60e@pengutronix.de

---
Adrian Ng Ho Yin (1):
      arm64: dts: socfpga: agilex5: Add SMMU node

Austin Zhang (1):
      arm64: dts: socfpga: agilex5: smmu enablement

Steffen Trumtrar (6):
      net: stmmac: dwmac-socfpga: don't set has_gmac
      dt-bindings: net: altr,socfpga-stmmac: add generic dwxgmac compatible
      arm64: dts: socfpga: agilex5: add dwxgmac compatible
      dt-bindings: net: altr,socfpga-stmmac: allow dma-coherent property
      dt-bindings: intel: add agilex5-based Arrow AXE5-Eagle
      arm64: dts: socfpga: agilex5: initial support for Arrow AXE5-Eagle

Teoh Ji Sheng (1):
      net: stmmac: Use interrupt mode INTM=1 for per channel irq

Tham, Mun Yew (1):
      arm64: dts: socfpga: agilex5: dma coherent enablement for XGMACs

 .../devicetree/bindings/arm/intel,socfpga.yaml     |   1 +
 .../bindings/net/altr,socfpga-stmmac.yaml          |   4 +
 arch/arm64/boot/dts/intel/Makefile                 |   1 +
 arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi     |  31 ++++-
 .../boot/dts/intel/socfpga_agilex5_axe5_eagle.dts  | 146 +++++++++++++++++++++
 .../net/ethernet/stmicro/stmmac/dwmac-socfpga.c    |   1 -
 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 +
 10 files changed, 214 insertions(+), 5 deletions(-)
---
base-commit: f67859cf0b6d21bf3641e7dec9e99edba91e0829
change-id: 20241030-v6-12-topic-socfpga-agilex5-90fd3d8f980c

Best regards,
-- 
Steffen Trumtrar <s.trumtrar@pengutronix.de>

[PATCH v5 01/10] net: stmmac: dwmac-socfpga: don't set has_gmac

[Steffen Trumtrar]

Instead of setting the has_gmac or has_xgmac fields, let
stmmac_probe_config_dt()) fill these fields according to the more
generic compatibles.

Without setting the has_xgmac/has_gmac field correctly, even basic
functions will fail, because the register offsets are different.

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
---
 drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
index 354f01184e6cc..7ed125dcc73ea 100644
--- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
+++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
@@ -497,7 +497,6 @@ static int socfpga_dwmac_probe(struct platform_device *pdev)
 	plat_dat->pcs_init = socfpga_dwmac_pcs_init;
 	plat_dat->pcs_exit = socfpga_dwmac_pcs_exit;
 	plat_dat->select_pcs = socfpga_dwmac_select_pcs;
-	plat_dat->has_gmac = true;
 
 	plat_dat->riwt_off = 1;
 

-- 
2.51.0

Re: [PATCH v5 01/10] net: stmmac: dwmac-socfpga: don't set has_gmac

[Maxime Chevallier]

Hi Steffen

On 24/10/2025 13:49, Steffen Trumtrar wrote:
> Instead of setting the has_gmac or has_xgmac fields, let
> stmmac_probe_config_dt()) fill these fields according to the more
> generic compatibles.
> 
> Without setting the has_xgmac/has_gmac field correctly, even basic
> functions will fail, because the register offsets are different.
> 
> Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
> ---
>  drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
> index 354f01184e6cc..7ed125dcc73ea 100644
> --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
> +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
> @@ -497,7 +497,6 @@ static int socfpga_dwmac_probe(struct platform_device *pdev)
>  	plat_dat->pcs_init = socfpga_dwmac_pcs_init;
>  	plat_dat->pcs_exit = socfpga_dwmac_pcs_exit;
>  	plat_dat->select_pcs = socfpga_dwmac_select_pcs;
> -	plat_dat->has_gmac = true;

Note that this field is now gone as per :

  26ab9830beab ("net: stmmac: replace has_xxxx with core_type")

You'll need to rebase the series :)

Maxime


>  
>  	plat_dat->riwt_off = 1;
>  
> 

Re: [PATCH v5 01/10] net: stmmac: dwmac-socfpga: don't set has_gmac

[Steffen Trumtrar]

Hi Maxime,

On 2025-10-24 at 14:11 +02, Maxime Chevallier <maxime.chevallier@bootlin.com> wrote:

> Hi Steffen
> 
> On 24/10/2025 13:49, Steffen Trumtrar wrote:
> > Instead of setting the has_gmac or has_xgmac fields, let
> > stmmac_probe_config_dt()) fill these fields according to the more
> > generic compatibles.
> > 
> > Without setting the has_xgmac/has_gmac field correctly, even basic
> > functions will fail, because the register offsets are different.
> > 
> > Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
> > ---
> >  drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c | 1 -
> >  1 file changed, 1 deletion(-)
> > 
> > diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
> > index 354f01184e6cc..7ed125dcc73ea 100644
> > --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
> > +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
> > @@ -497,7 +497,6 @@ static int socfpga_dwmac_probe(struct platform_device *pdev)
> >  	plat_dat->pcs_init = socfpga_dwmac_pcs_init;
> >  	plat_dat->pcs_exit = socfpga_dwmac_pcs_exit;
> >  	plat_dat->select_pcs = socfpga_dwmac_select_pcs;
> > -	plat_dat->has_gmac = true;
> 
> Note that this field is now gone as per :
> 
>   26ab9830beab ("net: stmmac: replace has_xxxx with core_type")
> 
> You'll need to rebase the series :)
>

I see, bad timing, but luckily an easy patch :)


Best regards,
Steffen

-- 
Pengutronix e.K.                | Dipl.-Inform. Steffen Trumtrar |
Steuerwalder Str. 21            | https://www.pengutronix.de/    |
31137 Hildesheim, Germany       | Phone: +49-5121-206917-0       |
Amtsgericht Hildesheim, HRA 2686| Fax:   +49-5121-206917-5555    |

Re: [PATCH v5 01/10] net: stmmac: dwmac-socfpga: don't set has_gmac

[G Thomas, Rohan]

Hi Steffen,

On 10/24/2025 6:00 PM, Steffen Trumtrar wrote:
> 
> Hi Maxime,
> 
> On 2025-10-24 at 14:11 +02, Maxime Chevallier 
> <maxime.chevallier@bootlin.com> wrote:
> 
>> Hi Steffen
>>
>> On 24/10/2025 13:49, Steffen Trumtrar wrote:
>> > Instead of setting the has_gmac or has_xgmac fields, let
>> > stmmac_probe_config_dt()) fill these fields according to the more
>> > generic compatibles.
>> > > Without setting the has_xgmac/has_gmac field correctly, even basic
>> > functions will fail, because the register offsets are different.
>> > > Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
>> > ---
>> >  drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c | 1 -
>> >  1 file changed, 1 deletion(-)
>> > > diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c 
>> b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
>> > index 354f01184e6cc..7ed125dcc73ea 100644
>> > --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
>> > +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
>> > @@ -497,7 +497,6 @@ static int socfpga_dwmac_probe(struct 
>> platform_device *pdev)
>> >      plat_dat->pcs_init = socfpga_dwmac_pcs_init;
>> >      plat_dat->pcs_exit = socfpga_dwmac_pcs_exit;
>> >      plat_dat->select_pcs = socfpga_dwmac_select_pcs;
>> > -    plat_dat->has_gmac = true;

I was preparing a patch-set to enable additional features in Agilex5.
Since Agilex5 HPS EMAC has different configurations from the previous
platforms adding a separate platform setup callback for Agilex5 and
other platforms. Here is the link to this patch-set:
https://lore.kernel.org/all/20251029-agilex5_ext-v1-0-1931132d77d6@altera.com/

Could you please confirm if this patch set aligns with your commit?

>>
>> Note that this field is now gone as per :
>>
>>   26ab9830beab ("net: stmmac: replace has_xxxx with core_type")
>>
>> You'll need to rebase the series :)
>>
> 
> I see, bad timing, but luckily an easy patch :)
> 
> 
> Best regards,
> Steffen
> 
Best Regards,
Rohan

Re: [PATCH v5 01/10] net: stmmac: dwmac-socfpga: don't set has_gmac

[Russell King (Oracle)]

On Fri, Oct 24, 2025 at 01:49:53PM +0200, Steffen Trumtrar wrote:
> Instead of setting the has_gmac or has_xgmac fields, let
> stmmac_probe_config_dt()) fill these fields according to the more
> generic compatibles.
> 
> Without setting the has_xgmac/has_gmac field correctly, even basic
> functions will fail, because the register offsets are different.

net-next no longer has these has_xgmac/has_gmac fields. This changed
on 22nd October. Please ensure you test your patches against the
latest tree to which you wish them to be applied.

> diff --git a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
> index 354f01184e6cc..7ed125dcc73ea 100644
> --- a/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
> +++ b/drivers/net/ethernet/stmicro/stmmac/dwmac-socfpga.c
> @@ -497,7 +497,6 @@ static int socfpga_dwmac_probe(struct platform_device *pdev)
>  	plat_dat->pcs_init = socfpga_dwmac_pcs_init;
>  	plat_dat->pcs_exit = socfpga_dwmac_pcs_exit;
>  	plat_dat->select_pcs = socfpga_dwmac_select_pcs;
> -	plat_dat->has_gmac = true;
>  
>  	plat_dat->riwt_off = 1;

In net-next, this code currently looks like:

        plat_dat->pcs_init = socfpga_dwmac_pcs_init;
        plat_dat->pcs_exit = socfpga_dwmac_pcs_exit;
        plat_dat->select_pcs = socfpga_dwmac_select_pcs;
        plat_dat->core_type = DWMAC_CORE_GMAC;

        plat_dat->riwt_off = 1;

Thus, this patch will not apply to net-next.

-- 
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last!

[PATCH v5 02/10] net: stmmac: Use interrupt mode INTM=1 for per channel irq

[Steffen Trumtrar]

From: Teoh Ji Sheng <ji.sheng.teoh@intel.com>

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 <ji.sheng.teoh@intel.com>
Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
---
 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		24
diff --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);
+		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));
 
 	/* 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;
+	}
+
 	/* On some platforms e.g. SPEAr the wake up irq differs from the mac irq
 	 * The external wake up irq can be passed through the platform code
 	 * named as "eth_wake_irq"
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index fa1318bac06c4..a8b15b4e3c370 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -102,6 +102,7 @@ struct stmmac_dma_cfg {
 	bool aal;
 	bool eame;
 	bool multi_msi_en;
+	bool multi_irq_en;
 	bool dche;
 	bool atds;
 };
@@ -290,6 +291,7 @@ struct plat_stmmacenet_data {
 	u8 vlan_fail_q;
 	struct pci_dev *pdev;
 	int int_snapshot_num;
+	bool multi_irq_en;
 	int msi_mac_vec;
 	int msi_wol_vec;
 	int msi_lpi_vec;

-- 
2.51.0

Re: [PATCH v5 02/10] net: stmmac: Use interrupt mode INTM=1 for per channel irq

[Maxime Chevallier]

Hi Steffen,

On 24/10/2025 13:49, Steffen Trumtrar wrote:
> From: Teoh Ji Sheng <ji.sheng.teoh@intel.com>
> 
> 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 <ji.sheng.teoh@intel.com>
> Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
> ---
>  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		24
> diff --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);
> +		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");

You need to document this property in the binding

> +
>  	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));
>  
>  	/* 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;
> +	}

Same for these irq names

> +
>  	/* On some platforms e.g. SPEAr the wake up irq differs from the mac irq
>  	 * The external wake up irq can be passed through the platform code
>  	 * named as "eth_wake_irq"
> diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
> index fa1318bac06c4..a8b15b4e3c370 100644
> --- a/include/linux/stmmac.h
> +++ b/include/linux/stmmac.h
> @@ -102,6 +102,7 @@ struct stmmac_dma_cfg {
>  	bool aal;
>  	bool eame;
>  	bool multi_msi_en;
> +	bool multi_irq_en;
>  	bool dche;
>  	bool atds;
>  };
> @@ -290,6 +291,7 @@ struct plat_stmmacenet_data {
>  	u8 vlan_fail_q;
>  	struct pci_dev *pdev;
>  	int int_snapshot_num;
> +	bool multi_irq_en;

This seems to be unused ?

>  	int msi_mac_vec;
>  	int msi_wol_vec;
>  	int msi_lpi_vec;
> 
Thanks,

Maxime

Re: [PATCH v5 02/10] net: stmmac: Use interrupt mode INTM=1 for per channel irq

[Simon Horman]

On Fri, Oct 24, 2025 at 01:49:54PM +0200, Steffen Trumtrar wrote:

...

> diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c

...

> @@ -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;

nit: The two lines above abbove to be indented a bit too much.

> +	}
> +
>  	/* On some platforms e.g. SPEAr the wake up irq differs from the mac irq
>  	 * The external wake up irq can be passed through the platform code
>  	 * named as "eth_wake_irq"

...

Re: [PATCH v5 02/10] net: stmmac: Use interrupt mode INTM=1 for per channel irq

[Russell King (Oracle)]

On Fri, Oct 24, 2025 at 01:49:54PM +0200, Steffen Trumtrar wrote:
> From: Teoh Ji Sheng <ji.sheng.teoh@intel.com>
> 
> 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 <ji.sheng.teoh@intel.com>
> Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
> ---
>  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		24
> diff --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.

> +		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.

>  
>  	/* 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!

[PATCH v5 03/10] arm64: dts: socfpga: agilex5: Add SMMU node

[Steffen Trumtrar]

From: Adrian Ng Ho Yin <adrian.ho.yin.ng@intel.com>

Add SMMU node in Agilex5 device tree.

Signed-off-by: Adrian Ng Ho Yin <adrian.ho.yin.ng@intel.com>
Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
---
 arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi b/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
index c4ae09b242d07..4f7ed20749927 100644
--- a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
+++ b/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
@@ -354,6 +354,18 @@ rst: rstmgr@10d11000 {
 			#reset-cells = <1>;
 		};
 
+		smmu: iommu@16000000 {
+			compatible = "arm,smmu-v3";
+			reg = <0x16000000 0x30000>;
+			interrupts = <GIC_SPI 134 IRQ_TYPE_EDGE_RISING>,
+				     <GIC_SPI 129 IRQ_TYPE_EDGE_RISING>,
+				     <GIC_SPI 132 IRQ_TYPE_EDGE_RISING>;
+			interrupt-names = "eventq", "gerror", "priq";
+			dma-coherent;
+			#iommu-cells = <1>;
+			status = "disabled";
+		};
+
 		spi0: spi@10da4000 {
 			compatible = "snps,dw-apb-ssi";
 			reg = <0x10da4000 0x1000>;

-- 
2.51.0

[PATCH v5 04/10] arm64: dts: socfpga: agilex5: smmu enablement

[Steffen Trumtrar]

From: Austin Zhang <austin.zhang@intel.com>

Add iommu property for peripherals connected to TBU.

Signed-off-by: Adrian Ng Ho Yin <adrian.ho.yin.ng@intel.com>
Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
---
 arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi b/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
index 4f7ed20749927..4ccfebfd9d322 100644
--- a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
+++ b/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
@@ -303,6 +303,7 @@ nand: nand-controller@10b80000 {
 			interrupts = <GIC_SPI 97 IRQ_TYPE_LEVEL_HIGH>;
 			clocks = <&clkmgr AGILEX5_NAND_NF_CLK>;
 			cdns,board-delay-ps = <4830>;
+			iommus = <&smmu 4>;
 			status = "disabled";
 		};
 
@@ -329,6 +330,7 @@ dmac0: dma-controller@10db0000 {
 			snps,block-size = <32767 32767 32767 32767>;
 			snps,priority = <0 1 2 3>;
 			snps,axi-max-burst-len = <8>;
+			iommus = <&smmu 8>;
 		};
 
 		dmac1: dma-controller@10dc0000 {
@@ -346,6 +348,7 @@ dmac1: dma-controller@10dc0000 {
 			snps,block-size = <32767 32767 32767 32767>;
 			snps,priority = <0 1 2 3>;
 			snps,axi-max-burst-len = <8>;
+			iommus = <&smmu 9>;
 		};
 
 		rst: rstmgr@10d11000 {
@@ -468,6 +471,7 @@ usb0: usb@10b00000 {
 			reset-names = "dwc2", "dwc2-ecc";
 			clocks = <&clkmgr AGILEX5_USB2OTG_HCLK>;
 			clock-names = "otg";
+			iommus = <&smmu 6>;
 			status = "disabled";
 		};
 
@@ -553,6 +557,7 @@ gmac0: ethernet@10810000 {
 			snps,tso;
 			altr,sysmgr-syscon = <&sysmgr 0x44 0>;
 			snps,clk-csr = <0>;
+			iommus = <&smmu 1>;
 			status = "disabled";
 
 			stmmac_axi_emac0_setup: stmmac-axi-config {
@@ -665,6 +670,7 @@ gmac1: ethernet@10820000 {
 			snps,tso;
 			altr,sysmgr-syscon = <&sysmgr 0x48 0>;
 			snps,clk-csr = <0>;
+			iommus = <&smmu 2>;
 			status = "disabled";
 
 			stmmac_axi_emac1_setup: stmmac-axi-config {
@@ -777,6 +783,7 @@ gmac2: ethernet@10830000 {
 			snps,tso;
 			altr,sysmgr-syscon = <&sysmgr 0x4c 0>;
 			snps,clk-csr = <0>;
+			iommus = <&smmu 3>;
 			status = "disabled";
 
 			stmmac_axi_emac2_setup: stmmac-axi-config {

-- 
2.51.0

Re: [PATCH v5 04/10] arm64: dts: socfpga: agilex5: smmu enablement

[Dinh Nguyen]

On 10/24/25 06:49, Steffen Trumtrar wrote:
> From: Austin Zhang <austin.zhang@intel.com>
> 
> Add iommu property for peripherals connected to TBU.
> 
> Signed-off-by: Adrian Ng Ho Yin <adrian.ho.yin.ng@intel.com>
> Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
> ---
>   arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi | 7 +++++++
>   1 file changed, 7 insertions(+)
> 
> diff --git a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi b/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
> index 4f7ed20749927..4ccfebfd9d322 100644
> --- a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
> +++ b/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
> @@ -303,6 +303,7 @@ nand: nand-controller@10b80000 {
>   			interrupts = <GIC_SPI 97 IRQ_TYPE_LEVEL_HIGH>;
>   			clocks = <&clkmgr AGILEX5_NAND_NF_CLK>;
>   			cdns,board-delay-ps = <4830>;
> +			iommus = <&smmu 4>;
>   			status = "disabled";
>   		};
>   
> @@ -329,6 +330,7 @@ dmac0: dma-controller@10db0000 {
>   			snps,block-size = <32767 32767 32767 32767>;
>   			snps,priority = <0 1 2 3>;
>   			snps,axi-max-burst-len = <8>;
> +			iommus = <&smmu 8>;
>   		};
>   
>   		dmac1: dma-controller@10dc0000 {
> @@ -346,6 +348,7 @@ dmac1: dma-controller@10dc0000 {
>   			snps,block-size = <32767 32767 32767 32767>;
>   			snps,priority = <0 1 2 3>;
>   			snps,axi-max-burst-len = <8>;
> +			iommus = <&smmu 9>;
>   		};
>   
>   		rst: rstmgr@10d11000 {
> @@ -468,6 +471,7 @@ usb0: usb@10b00000 {
>   			reset-names = "dwc2", "dwc2-ecc";
>   			clocks = <&clkmgr AGILEX5_USB2OTG_HCLK>;
>   			clock-names = "otg";
> +			iommus = <&smmu 6>;
>   			status = "disabled";
>   		};
>   
> @@ -553,6 +557,7 @@ gmac0: ethernet@10810000 {
>   			snps,tso;
>   			altr,sysmgr-syscon = <&sysmgr 0x44 0>;
>   			snps,clk-csr = <0>;
> +			iommus = <&smmu 1>;
>   			status = "disabled";
>   
>   			stmmac_axi_emac0_setup: stmmac-axi-config {
> @@ -665,6 +670,7 @@ gmac1: ethernet@10820000 {
>   			snps,tso;
>   			altr,sysmgr-syscon = <&sysmgr 0x48 0>;
>   			snps,clk-csr = <0>;
> +			iommus = <&smmu 2>;
>   			status = "disabled";
>   
>   			stmmac_axi_emac1_setup: stmmac-axi-config {
> @@ -777,6 +783,7 @@ gmac2: ethernet@10830000 {
>   			snps,tso;
>   			altr,sysmgr-syscon = <&sysmgr 0x4c 0>;
>   			snps,clk-csr = <0>;
> +			iommus = <&smmu 3>;
>   			status = "disabled";
>   
>   			stmmac_axi_emac2_setup: stmmac-axi-config {
> 


I have a similar patches for this and 3/10 queued up for v6.19.

The DTS patches are staged in my repo under the branch 
socfpga_updates_for_v6.19

Dinh

Re: [PATCH v5 04/10] arm64: dts: socfpga: agilex5: smmu enablement

[Simon Horman]

On Fri, Oct 24, 2025 at 01:49:56PM +0200, Steffen Trumtrar wrote:
> From: Austin Zhang <austin.zhang@intel.com>
> 
> Add iommu property for peripherals connected to TBU.
> 
> Signed-off-by: Adrian Ng Ho Yin <adrian.ho.yin.ng@intel.com>
> Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>

As Austin Zhang is the author (as listed in the from line),
their Signed-off-by is needed.

...

[PATCH v5 05/10] dt-bindings: net: altr,socfpga-stmmac: add generic dwxgmac compatible

[Steffen Trumtrar]

Add the more generic "snps,dwxgmac" compatible to the list of
compatibles for the agilex5.

The snps,dwmac binding explicitly uses the "snps,dwxgmac-2.10",
"snps,dwxgmac" combination as example; the stmmac driver only checks on
dwmac and dwxgmac compatibles to decide what kind of gmac it is working
with.

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
---
 Documentation/devicetree/bindings/net/altr,socfpga-stmmac.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/net/altr,socfpga-stmmac.yaml b/Documentation/devicetree/bindings/net/altr,socfpga-stmmac.yaml
index fc445ad5a1f1a..b53847a2d9536 100644
--- a/Documentation/devicetree/bindings/net/altr,socfpga-stmmac.yaml
+++ b/Documentation/devicetree/bindings/net/altr,socfpga-stmmac.yaml
@@ -46,6 +46,7 @@ properties:
       - items:
           - const: altr,socfpga-stmmac-agilex5
           - const: snps,dwxgmac-2.10
+          - const: snps,dwxgmac
 
   clocks:
     minItems: 1

-- 
2.51.0

Re: [PATCH v5 05/10] dt-bindings: net: altr,socfpga-stmmac: add generic dwxgmac compatible

[Krzysztof Kozlowski]

On 24/10/2025 13:49, Steffen Trumtrar wrote:
> Add the more generic "snps,dwxgmac" compatible to the list of
> compatibles for the agilex5.
> 
> The snps,dwmac binding explicitly uses the "snps,dwxgmac-2.10",

Hm? No, binding says only dwxgmac-2.10, DTS only dwxgmac-2.10.

> "snps,dwxgmac" combination as example; the stmmac driver only checks on

Example does not matter.

> dwmac and dwxgmac compatibles to decide what kind of gmac it is working

Huh? I clearly see snps,dwxgmac-2.10 in dwmac-generic.c

> with.
This change really lacks proper rationale.

Best regards,
Krzysztof

[PATCH v5 06/10] arm64: dts: socfpga: agilex5: add dwxgmac compatible

[Steffen Trumtrar]

The gmac0/1/2 are also compatible to the more generic "snps,dwxgmac"
compatible. The platform code checks this to decide if it is a GMAC or
GMAC4 compatible IP core.

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
---
 arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi b/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
index 4ccfebfd9d322..d0c139f03541e 100644
--- a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
+++ b/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
@@ -536,7 +536,8 @@ qspi: spi@108d2000 {
 
 		gmac0: ethernet@10810000 {
 			compatible = "altr,socfpga-stmmac-agilex5",
-				     "snps,dwxgmac-2.10";
+				     "snps,dwxgmac-2.10",
+				     "snps,dwxgmac";
 			reg = <0x10810000 0x3500>;
 			interrupts = <GIC_SPI 190 IRQ_TYPE_LEVEL_HIGH>;
 			interrupt-names = "macirq";
@@ -649,7 +650,8 @@ queue7 {
 
 		gmac1: ethernet@10820000 {
 			compatible = "altr,socfpga-stmmac-agilex5",
-				     "snps,dwxgmac-2.10";
+				     "snps,dwxgmac-2.10",
+				     "snps,dwxgmac";
 			reg = <0x10820000 0x3500>;
 			interrupts = <GIC_SPI 207 IRQ_TYPE_LEVEL_HIGH>;
 			interrupt-names = "macirq";
@@ -762,7 +764,8 @@ queue7 {
 
 		gmac2: ethernet@10830000 {
 			compatible = "altr,socfpga-stmmac-agilex5",
-				     "snps,dwxgmac-2.10";
+				     "snps,dwxgmac-2.10",
+				     "snps,dwxgmac";
 			reg = <0x10830000 0x3500>;
 			interrupts = <GIC_SPI 224 IRQ_TYPE_LEVEL_HIGH>;
 			interrupt-names = "macirq";

-- 
2.51.0

Re: [PATCH v5 06/10] arm64: dts: socfpga: agilex5: add dwxgmac compatible

[Dinh Nguyen]

Hi Steffen,

On 10/24/25 06:49, Steffen Trumtrar wrote:
> The gmac0/1/2 are also compatible to the more generic "snps,dwxgmac"
> compatible. The platform code checks this to decide if it is a GMAC or
> GMAC4 compatible IP core.
> 
> Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
> ---
>   arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi | 9 ++++++---
>   1 file changed, 6 insertions(+), 3 deletions(-)
> 
> diff --git a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi b/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
> index 4ccfebfd9d322..d0c139f03541e 100644
> --- a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
> +++ b/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
> @@ -536,7 +536,8 @@ qspi: spi@108d2000 {
>   
>   		gmac0: ethernet@10810000 {
>   			compatible = "altr,socfpga-stmmac-agilex5",
> -				     "snps,dwxgmac-2.10";
> +				     "snps,dwxgmac-2.10",
> +				     "snps,dwxgmac";
>   			reg = <0x10810000 0x3500>;
>   			interrupts = <GIC_SPI 190 IRQ_TYPE_LEVEL_HIGH>;
>   			interrupt-names = "macirq";
> @@ -649,7 +650,8 @@ queue7 {
>   
>   		gmac1: ethernet@10820000 {
>   			compatible = "altr,socfpga-stmmac-agilex5",
> -				     "snps,dwxgmac-2.10";
> +				     "snps,dwxgmac-2.10",
> +				     "snps,dwxgmac";
>   			reg = <0x10820000 0x3500>;
>   			interrupts = <GIC_SPI 207 IRQ_TYPE_LEVEL_HIGH>;
>   			interrupt-names = "macirq";
> @@ -762,7 +764,8 @@ queue7 {
>   
>   		gmac2: ethernet@10830000 {
>   			compatible = "altr,socfpga-stmmac-agilex5",
> -				     "snps,dwxgmac-2.10";
> +				     "snps,dwxgmac-2.10",
> +				     "snps,dwxgmac";
>   			reg = <0x10830000 0x3500>;
>   			interrupts = <GIC_SPI 224 IRQ_TYPE_LEVEL_HIGH>;
>   			interrupt-names = "macirq";
> 

I just sent a patch for this yesterday:

https://lore.kernel.org/all/20251023214012.283600-1-dinguyen@kernel.org/

I'll make sure to include you on future submissions.

I didn't add it to the bindings document though.

Dinh

Re: [PATCH v5 06/10] arm64: dts: socfpga: agilex5: add dwxgmac compatible

[Steffen Trumtrar]

Hi Dinh,

On 2025-10-24 at 07:00 -05, Dinh Nguyen <dinguyen@kernel.org> wrote:

> Hi Steffen,
> 
> On 10/24/25 06:49, Steffen Trumtrar wrote:
> > The gmac0/1/2 are also compatible to the more generic "snps,dwxgmac"
> > compatible. The platform code checks this to decide if it is a GMAC or
> > GMAC4 compatible IP core.
> > Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
> > ---
> >   arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi | 9 ++++++---
> >   1 file changed, 6 insertions(+), 3 deletions(-)
> > diff --git a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
> > b/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
> > index 4ccfebfd9d322..d0c139f03541e 100644
> > --- a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
> > +++ b/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
> > @@ -536,7 +536,8 @@ qspi: spi@108d2000 {
> >     		gmac0: ethernet@10810000 {
> >   			compatible = "altr,socfpga-stmmac-agilex5",
> > -				     "snps,dwxgmac-2.10";
> > +				     "snps,dwxgmac-2.10",
> > +				     "snps,dwxgmac";
> >   			reg = <0x10810000 0x3500>;
> >   			interrupts = <GIC_SPI 190 IRQ_TYPE_LEVEL_HIGH>;
> >   			interrupt-names = "macirq";
> > @@ -649,7 +650,8 @@ queue7 {
> >     		gmac1: ethernet@10820000 {
> >   			compatible = "altr,socfpga-stmmac-agilex5",
> > -				     "snps,dwxgmac-2.10";
> > +				     "snps,dwxgmac-2.10",
> > +				     "snps,dwxgmac";
> >   			reg = <0x10820000 0x3500>;
> >   			interrupts = <GIC_SPI 207 IRQ_TYPE_LEVEL_HIGH>;
> >   			interrupt-names = "macirq";
> > @@ -762,7 +764,8 @@ queue7 {
> >     		gmac2: ethernet@10830000 {
> >   			compatible = "altr,socfpga-stmmac-agilex5",
> > -				     "snps,dwxgmac-2.10";
> > +				     "snps,dwxgmac-2.10",
> > +				     "snps,dwxgmac";
> >   			reg = <0x10830000 0x3500>;
> >   			interrupts = <GIC_SPI 224 IRQ_TYPE_LEVEL_HIGH>;
> >   			interrupt-names = "macirq";
> > 
> 
> I just sent a patch for this yesterday:
> 
> https://lore.kernel.org/all/20251023214012.283600-1-dinguyen@kernel.org/
>

ah, I missed that, than this patch is unneccessary.

> I'll make sure to include you on future submissions.
> 
> I didn't add it to the bindings document though.
>

As I always get complains from dt-check bot, I remembered to add it ;)


Best regards,
Steffen

-- 
Pengutronix e.K.                | Dipl.-Inform. Steffen Trumtrar |
Steuerwalder Str. 21            | https://www.pengutronix.de/    |
31137 Hildesheim, Germany       | Phone: +49-5121-206917-0       |
Amtsgericht Hildesheim, HRA 2686| Fax:   +49-5121-206917-5555    |

[PATCH v5 07/10] dt-bindings: net: altr,socfpga-stmmac: allow dma-coherent property

[Steffen Trumtrar]

The DMA operations on the SoCFPGA stmmac can be dma-coherent. Add as
optional property.

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
---
 Documentation/devicetree/bindings/net/altr,socfpga-stmmac.yaml | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/Documentation/devicetree/bindings/net/altr,socfpga-stmmac.yaml b/Documentation/devicetree/bindings/net/altr,socfpga-stmmac.yaml
index b53847a2d9536..c34f2ef0cf001 100644
--- a/Documentation/devicetree/bindings/net/altr,socfpga-stmmac.yaml
+++ b/Documentation/devicetree/bindings/net/altr,socfpga-stmmac.yaml
@@ -109,6 +109,9 @@ properties:
   txen-skew-ps:
     description: Skew control of TXC pad
 
+  dma-coherent:
+    description: Present if dma operations are coherent
+
   altr,emac-splitter:
     $ref: /schemas/types.yaml#/definitions/phandle
     description:

-- 
2.51.0

Re: [PATCH v5 07/10] dt-bindings: net: altr,socfpga-stmmac: allow dma-coherent property

[Krzysztof Kozlowski]

On 24/10/2025 13:49, Steffen Trumtrar wrote:
> The DMA operations on the SoCFPGA stmmac can be dma-coherent. Add as
> optional property.
> 
> Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>

There is little point in mixing two subsystems in one patchset.
Especially without marking this net-next how maintainers are supposed to
take it?

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>

[PATCH v5 08/10] arm64: dts: socfpga: agilex5: dma coherent enablement for XGMACs

[Steffen Trumtrar]

From: "Tham, Mun Yew" <mun.yew.tham@intel.com>

Add dma-coherent property to device tree for XGMACs to resolve kernel
panic error on dcache_inval_poc for XGMACs.

Signed-off-by: Tham, Mun Yew <mun.yew.tham@intel.com>
Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
---
 arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi b/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
index d0c139f03541e..63c0aa8550313 100644
--- a/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
+++ b/arch/arm64/boot/dts/intel/socfpga_agilex5.dtsi
@@ -558,6 +558,7 @@ gmac0: ethernet@10810000 {
 			snps,tso;
 			altr,sysmgr-syscon = <&sysmgr 0x44 0>;
 			snps,clk-csr = <0>;
+			dma-coherent;
 			iommus = <&smmu 1>;
 			status = "disabled";
 
@@ -672,6 +673,7 @@ gmac1: ethernet@10820000 {
 			snps,tso;
 			altr,sysmgr-syscon = <&sysmgr 0x48 0>;
 			snps,clk-csr = <0>;
+			dma-coherent;
 			iommus = <&smmu 2>;
 			status = "disabled";
 
@@ -786,6 +788,7 @@ gmac2: ethernet@10830000 {
 			snps,tso;
 			altr,sysmgr-syscon = <&sysmgr 0x4c 0>;
 			snps,clk-csr = <0>;
+			dma-coherent;
 			iommus = <&smmu 3>;
 			status = "disabled";
 

-- 
2.51.0

[PATCH v5 09/10] dt-bindings: intel: add agilex5-based Arrow AXE5-Eagle

[Steffen Trumtrar]

Add binding for the Arrow Agilex5-based AXE5-Eagle board.

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
---
 Documentation/devicetree/bindings/arm/intel,socfpga.yaml | 1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/arm/intel,socfpga.yaml b/Documentation/devicetree/bindings/arm/intel,socfpga.yaml
index c75cd7d29f1aa..3ab6058268e8c 100644
--- a/Documentation/devicetree/bindings/arm/intel,socfpga.yaml
+++ b/Documentation/devicetree/bindings/arm/intel,socfpga.yaml
@@ -24,6 +24,7 @@ properties:
       - description: Agilex5 boards
         items:
           - enum:
+              - arrow,socfpga-agilex5-axe5-eagle
               - intel,socfpga-agilex5-socdk
               - intel,socfpga-agilex5-socdk-nand
           - const: intel,socfpga-agilex5

-- 
2.51.0

[PATCH v5 10/10] arm64: dts: socfpga: agilex5: initial support for Arrow AXE5-Eagle

[Steffen Trumtrar]

The Arrow AXE5-Eagle is an Intel Agilex5 SoCFPGA based board with:

   - 1x PCIe Gen4.0 edge connector
   - 4-port USB HUB
   - 2x 1Gb Ethernet
   - microSD
   - HDMI output
   - 2x 10Gb SFP+ cages

As most devices aren't supported mainline yet, this is only the initial
support for the board: leds, gpios, gmac2, i2c, uart and qspi. Although
gmac1 is equipped on the board, it doesn't work yet as it needs
additional pinctrl settings in the bootloader to work.

Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
---
 arch/arm64/boot/dts/intel/Makefile                 |   1 +
 .../boot/dts/intel/socfpga_agilex5_axe5_eagle.dts  | 146 +++++++++++++++++++++
 2 files changed, 147 insertions(+)

diff --git a/arch/arm64/boot/dts/intel/Makefile b/arch/arm64/boot/dts/intel/Makefile
index 33f6d01266b1a..0a18ca306054b 100644
--- a/arch/arm64/boot/dts/intel/Makefile
+++ b/arch/arm64/boot/dts/intel/Makefile
@@ -2,6 +2,7 @@
 dtb-$(CONFIG_ARCH_INTEL_SOCFPGA) += socfpga_agilex_n6000.dtb \
 				socfpga_agilex_socdk.dtb \
 				socfpga_agilex_socdk_nand.dtb \
+				socfpga_agilex5_axe5_eagle.dtb \
 				socfpga_agilex5_socdk.dtb \
 				socfpga_agilex5_socdk_nand.dtb \
 				socfpga_n5x_socdk.dtb
diff --git a/arch/arm64/boot/dts/intel/socfpga_agilex5_axe5_eagle.dts b/arch/arm64/boot/dts/intel/socfpga_agilex5_axe5_eagle.dts
new file mode 100644
index 0000000000000..a2a262f3d07c6
--- /dev/null
+++ b/arch/arm64/boot/dts/intel/socfpga_agilex5_axe5_eagle.dts
@@ -0,0 +1,146 @@
+// SPDX-License-Identifier:     GPL-2.0
+/*
+ * Copyright (C) 2024, Arrow Electronics, Inc.
+ */
+#include "socfpga_agilex5.dtsi"
+
+/ {
+	model = "SoCFPGA Agilex5 Arrow AXE5-Eagle";
+	compatible = "arrow,socfpga-agilex5-axe5-eagle", "intel,socfpga-agilex5";
+
+	aliases {
+		serial0 = &uart0;
+	};
+
+	chosen {
+		stdout-path = "serial0:115200n8";
+	};
+
+	leds {
+		compatible = "gpio-leds";
+
+		led-0 {
+			label = "hps_led0";
+			gpios = <&porta 6 GPIO_ACTIVE_HIGH>;
+		};
+
+		led-1 {
+			label = "hps_led1";
+			gpios = <&porta 7 GPIO_ACTIVE_HIGH>;
+		};
+	};
+
+	gpio-keys {
+		compatible = "gpio-keys";
+
+		key-0 {
+			label = "hps_sw0";
+			gpios = <&porta 10 0>;
+			linux,input-type = <5>;	/* EV_SW */
+			linux,code = <0x0>;
+		};
+
+		key-1 {
+			label = "hps_sw1";
+			gpios = <&porta 1 0>;
+			linux,input-type = <5>;	/* EV_SW */
+			linux,code = <0x0>;
+		};
+
+		key-2 {
+			label = "hps_pb0";
+			gpios = <&porta 8 1>;
+			linux,code = <187>;		/* KEY_F17 */
+		};
+
+		key-3 {
+			label = "hps_pb1";
+			gpios = <&porta 9 1>;
+			linux,code = <188>;		/* KEY_F18 */
+		};
+	};
+
+	vdd: regulator-vdd {
+		compatible = "regulator-fixed";
+		regulator-name = "fixed-supply";
+		regulator-min-microvolt = <1800000>;
+		regulator-max-microvolt = <1800000>;
+		regulator-always-on;
+	};
+
+	vdd_3_3: regulator-vdd {
+		compatible = "regulator-fixed";
+		regulator-name = "fixed-supply";
+		regulator-min-microvolt = <3300000>;
+		regulator-max-microvolt = <3300000>;
+		regulator-always-on;
+	};
+};
+
+&gmac2 {
+	status = "okay";
+	phy-mode = "rgmii-id";
+	phy-handle = <&emac2_phy0>;
+
+	mdio0 {
+		compatible = "snps,dwmac-mdio";
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		emac2_phy0: ethernet-phy@1 {
+			compatible = "ethernet-phy-ieee802.3-c22";
+			reg = <0x1>;
+		};
+	};
+};
+
+&gpio0 {
+	status = "okay";
+};
+
+&i2c0 {
+	status = "okay";
+};
+
+&i2c1 {
+	status = "okay";
+
+	i2c-mux@70 {
+		compatible = "nxp,pca9544";
+		#address-cells = <1>;
+		#size-cells = <0>;
+		reg = <0x70>;
+		status = "okay";
+	};
+};
+
+&osc1 {
+	clock-frequency = <25000000>;
+};
+
+&qspi {
+	status = "okay";
+
+	flash@0 {
+		#address-cells = <1>;
+		#size-cells = <1>;
+		compatible = "micron,mt25qu02g", "jedec,spi-nor";
+		reg = <0>;
+		spi-max-frequency = <100000000>;
+
+		m25p,fast-read;
+		cdns,read-delay = <2>;
+		cdns,tshsl-ns = <50>;
+		cdns,tsd2d-ns = <50>;
+		cdns,tchsh-ns = <4>;
+		cdns,tslch-ns = <4>;
+	};
+};
+
+&smmu {
+	status = "okay";
+};
+
+&uart0 {
+	status = "okay";
+};

-- 
2.51.0

Re: [PATCH v5 00/10] arm64: dts: socfpga: agilex5: enable network and add new board

[Jakub Kicinski]

On Fri, 24 Oct 2025 13:49:52 +0200 Steffen Trumtrar wrote:
> Add an initial devicetree for a new board (Arrow AXE5-Eagle) and all
> needed patches to support the network on current mainline.
> 
> Currently only QSPI and network are functional as all other hardware
> currently lacks mainline support.

Please split out the drivers/net and bindings/net patches to a separate
series. As is this is unmergable to any single tree.

Re: [PATCH v5 00/10] arm64: dts: socfpga: agilex5: enable network and add new board

[Russell King (Oracle)]

On Fri, Oct 24, 2025 at 01:49:52PM +0200, Steffen Trumtrar wrote:
> Add an initial devicetree for a new board (Arrow AXE5-Eagle) and all
> needed patches to support the network on current mainline.
> 
> Currently only QSPI and network are functional as all other hardware
> currently lacks mainline support.
> 
> Signed-off-by: Steffen Trumtrar <s.trumtrar@pengutronix.de>

You haven't included which tree you are targetting with this series. I'm
assuming net-next for the following review. Please be explicit via
[PATCH net ...] or [PATCH net-next ...] in the subject line of the cover
message and each patch.

-- 
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last!