Most glue driver for PCI-based DWMAC controllers utilize similar
platform suspend/resume routines. Add a generic implementation to reduce
duplicated code.
Signed-off-by: Yao Zi <ziyao@disroot.org>
---
drivers/net/ethernet/stmicro/stmmac/stmmac.h | 2 +
.../net/ethernet/stmicro/stmmac/stmmac_main.c | 37 +++++++++++++++++++
2 files changed, 39 insertions(+)
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac.h b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
index d5af9344dfb0..baa4ff14bdfe 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac.h
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac.h
@@ -395,6 +395,8 @@ int stmmac_xdp_open(struct net_device *dev);
void stmmac_xdp_release(struct net_device *dev);
int stmmac_resume(struct device *dev);
int stmmac_suspend(struct device *dev);
+int stmmac_pci_plat_suspend(struct device *dev, void *bsp_priv);
+int stmmac_pci_plat_resume(struct device *dev, void *bsp_priv);
void stmmac_dvr_remove(struct device *dev);
int stmmac_dvr_probe(struct device *device,
struct plat_stmmacenet_data *plat_dat,
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index fd5106880192..3bd284019ca7 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -28,6 +28,7 @@
#include <linux/if_vlan.h>
#include <linux/dma-mapping.h>
#include <linux/slab.h>
+#include <linux/pci.h>
#include <linux/pm_runtime.h>
#include <linux/pm_wakeirq.h>
#include <linux/prefetch.h>
@@ -7938,6 +7939,42 @@ EXPORT_SYMBOL_GPL(stmmac_resume);
DEFINE_SIMPLE_DEV_PM_OPS(stmmac_simple_pm_ops, stmmac_suspend, stmmac_resume);
EXPORT_SYMBOL_GPL(stmmac_simple_pm_ops);
+#ifdef CONFIG_PCI
+int stmmac_pci_plat_suspend(struct device *dev, void *bsp_priv)
+{
+ struct pci_dev *pdev = to_pci_dev(dev);
+ int ret;
+
+ ret = pci_save_state(pdev);
+ if (ret)
+ return ret;
+
+ pci_disable_device(pdev);
+ pci_wake_from_d3(pdev, true);
+
+ return 0;
+}
+EXPORT_SYMBOL_GPL(stmmac_pci_plat_suspend);
+
+int stmmac_pci_plat_resume(struct device *dev, void *bsp_priv)
+{
+ struct pci_dev *pdev = to_pci_dev(dev);
+ int ret;
+
+ pci_restore_state(pdev);
+ pci_set_power_state(pdev, PCI_D0);
+
+ ret = pci_enable_device(pdev);
+ if (ret)
+ return ret;
+
+ pci_set_master(pdev);
+
+ return 0;
+}
+EXPORT_SYMBOL_GPL(stmmac_pci_plat_resume);
+#endif /* CONFIG_PCI */
+
#ifndef MODULE
static int __init stmmac_cmdline_opt(char *str)
{
--
2.50.1On Tue, Oct 28, 2025 at 03:43:30PM +0000, Yao Zi wrote: > Most glue driver for PCI-based DWMAC controllers utilize similar > platform suspend/resume routines. Add a generic implementation to reduce > duplicated code. > > Signed-off-by: Yao Zi <ziyao@disroot.org> > --- > drivers/net/ethernet/stmicro/stmmac/stmmac.h | 2 + > .../net/ethernet/stmicro/stmmac/stmmac_main.c | 37 +++++++++++++++++++ I would prefer not to make stmmac_main.c even larger by including bus specific helpers there. We already have stmmac_pltfm.c for those which use struct platform_device. The logical name would be stmmac_pci.c, but that's already taken by a driver. One way around that would be to rename stmmac_pci.c to dwmac-pci.c (glue drivers tend to be named dwmac-foo.c) and then re-use stmmac_pci.c for PCI-related stuff in the same way that stmmac_pltfm.c is used. Another idea would be stmmac_libpci.c. -- RMK's Patch system: https://www.armlinux.org.uk/developer/patches/ FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last!
在 2025/10/28 下午11:59, Russell King (Oracle) 写道: > On Tue, Oct 28, 2025 at 03:43:30PM +0000, Yao Zi wrote: >> Most glue driver for PCI-based DWMAC controllers utilize similar >> platform suspend/resume routines. Add a generic implementation to reduce >> duplicated code. >> >> Signed-off-by: Yao Zi <ziyao@disroot.org> >> --- >> drivers/net/ethernet/stmicro/stmmac/stmmac.h | 2 + >> .../net/ethernet/stmicro/stmmac/stmmac_main.c | 37 +++++++++++++++++++ > I would prefer not to make stmmac_main.c even larger by including bus > specific helpers there. We already have stmmac_pltfm.c for those which > use struct platform_device. The logical name would be stmmac_pci.c, but > that's already taken by a driver. > > One way around that would be to rename stmmac_pci.c to dwmac-pci.c > (glue drivers tend to be named dwmac-foo.c) and then re-use > stmmac_pci.c for PCI-related stuff in the same way that stmmac_pltfm.c > is used. > > Another idea would be stmmac_libpci.c. I also don't want stmmac_main.c to grow larger, and I prefer stmmac_libpci.c instead. Another approach - maybe we can keep these helper functions in stmmac_pci.c and just declare them as extern where needed? Thanks, Yanteng >
On Wed, Oct 29, 2025 at 10:27:18AM +0800, Yanteng Si wrote: > 在 2025/10/28 下午11:59, Russell King (Oracle) 写道: > > On Tue, Oct 28, 2025 at 03:43:30PM +0000, Yao Zi wrote: > > > Most glue driver for PCI-based DWMAC controllers utilize similar > > > platform suspend/resume routines. Add a generic implementation to reduce > > > duplicated code. > > > > > > Signed-off-by: Yao Zi <ziyao@disroot.org> > > > --- > > > drivers/net/ethernet/stmicro/stmmac/stmmac.h | 2 + > > > .../net/ethernet/stmicro/stmmac/stmmac_main.c | 37 +++++++++++++++++++ > > I would prefer not to make stmmac_main.c even larger by including bus > > specific helpers there. We already have stmmac_pltfm.c for those which > > use struct platform_device. The logical name would be stmmac_pci.c, but > > that's already taken by a driver. > > > > One way around that would be to rename stmmac_pci.c to dwmac-pci.c > > (glue drivers tend to be named dwmac-foo.c) and then re-use > > stmmac_pci.c for PCI-related stuff in the same way that stmmac_pltfm.c > > is used. > > > > Another idea would be stmmac_libpci.c. > > I also don't want stmmac_main.c to grow larger, and I prefer > stmmac_libpci.c instead. Another approach - maybe we can > keep these helper functions in stmmac_pci.c and just declare > them as extern where needed? stmmac_pci.c is itself a glue driver, supporting PCI IDs: 0x0700, 0x1108 - synthetic ID 0x104a, 0xcc09 - ST Micro MAC 0x16c3, 0x7102 - Synopsys GMAC5 I don't think we should try to turn a glue driver into a library, even though it would be the easier option (we could reuse CONFIG_STMMAC_PCI.) -- RMK's Patch system: https://www.armlinux.org.uk/developer/patches/ FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last!
在 2025/10/29 下午5:00, Russell King (Oracle) 写道: > On Wed, Oct 29, 2025 at 10:27:18AM +0800, Yanteng Si wrote: >> 在 2025/10/28 下午11:59, Russell King (Oracle) 写道: >>> On Tue, Oct 28, 2025 at 03:43:30PM +0000, Yao Zi wrote: >>>> Most glue driver for PCI-based DWMAC controllers utilize similar >>>> platform suspend/resume routines. Add a generic implementation to reduce >>>> duplicated code. >>>> >>>> Signed-off-by: Yao Zi<ziyao@disroot.org> >>>> --- >>>> drivers/net/ethernet/stmicro/stmmac/stmmac.h | 2 + >>>> .../net/ethernet/stmicro/stmmac/stmmac_main.c | 37 +++++++++++++++++++ >>> I would prefer not to make stmmac_main.c even larger by including bus >>> specific helpers there. We already have stmmac_pltfm.c for those which >>> use struct platform_device. The logical name would be stmmac_pci.c, but >>> that's already taken by a driver. >>> >>> One way around that would be to rename stmmac_pci.c to dwmac-pci.c >>> (glue drivers tend to be named dwmac-foo.c) and then re-use >>> stmmac_pci.c for PCI-related stuff in the same way that stmmac_pltfm.c >>> is used. >>> >>> Another idea would be stmmac_libpci.c. >> I also don't want stmmac_main.c to grow larger, and I prefer >> stmmac_libpci.c instead. Another approach - maybe we can >> keep these helper functions in stmmac_pci.c and just declare >> them as extern where needed? > stmmac_pci.c is itself a glue driver, supporting PCI IDs: > > 0x0700, 0x1108 - synthetic ID > 0x104a, 0xcc09 - ST Micro MAC > 0x16c3, 0x7102 - Synopsys GMAC5 > > I don't think we should try to turn a glue driver into a library, > even though it would be the easier option (we could reuse > CONFIG_STMMAC_PCI.) I agree with your opinion; let's build a new library. Thanks, Yanteng
On Wed, Oct 29, 2025 at 10:27:18AM +0800, Yanteng Si wrote: > > 在 2025/10/28 下午11:59, Russell King (Oracle) 写道: > > On Tue, Oct 28, 2025 at 03:43:30PM +0000, Yao Zi wrote: > > > Most glue driver for PCI-based DWMAC controllers utilize similar > > > platform suspend/resume routines. Add a generic implementation to reduce > > > duplicated code. > > > > > > Signed-off-by: Yao Zi <ziyao@disroot.org> > > > --- > > > drivers/net/ethernet/stmicro/stmmac/stmmac.h | 2 + > > > .../net/ethernet/stmicro/stmmac/stmmac_main.c | 37 +++++++++++++++++++ > > I would prefer not to make stmmac_main.c even larger by including bus > > specific helpers there. We already have stmmac_pltfm.c for those which > > use struct platform_device. The logical name would be stmmac_pci.c, but > > that's already taken by a driver. > > > > One way around that would be to rename stmmac_pci.c to dwmac-pci.c > > (glue drivers tend to be named dwmac-foo.c) and then re-use > > stmmac_pci.c for PCI-related stuff in the same way that stmmac_pltfm.c > > is used. > > > > Another idea would be stmmac_libpci.c. > > I also don't want stmmac_main.c to grow larger, and I prefer > > stmmac_libpci.c instead. Okay, then I'll separate the code into stmmac_libpci.c instead. This also avoids moving code around, making it easier to track git log in the future. > Another approach - maybe we can > > keep these helper functions in stmmac_pci.c and just declare > > them as extern where needed? stmmac_pci.c is a standalone DWMAC glue driver, none of its symbols are for external usage. I don't think it's appropriate to put these helpers in the driver. Furthermore, this will also require PCI-based glues making use of these helpers to depend on an unrelated driver, introducing unnecessary code size, which doesn't sound like a good idea to me. I'd still like to introduce stmmac_libpci.c for these helpers. > Thanks, > > Yanteng > > > Best regards, Yao Zi
© 2016 - 2026 Red Hat, Inc.