* [PATCH v6 2/3] hyperv: Use reboot notifier to configure sleep state
From: Praveen K Paladugu @ 2025-11-26 21:49 UTC (permalink / raw)
To: kys, haiyangz, wei.liu, decui, tglx, mingo, linux-hyperv,
linux-kernel, bp, dave.hansen, x86, hpa, arnd
Cc: anbelski, prapal, easwar.hariharan, nunodasneves, skinsburskii
In-Reply-To: <20251126215013.11549-1-prapal@linux.microsoft.com>
Configure sleep state information in mshv hypervisor. This sleep state
information from ACPI will be used by hypervisor to poweroff the host.
Signed-off-by: Praveen K Paladugu <prapal@linux.microsoft.com>
Co-developed-by: Anatol Belski <anbelski@linux.microsoft.com>
Signed-off-by: Anatol Belski <anbelski@linux.microsoft.com>
Reviewed-by: Easwar Hariharan <easwar.hariharan@linux.microsoft.com>
---
arch/x86/hyperv/hv_init.c | 1 +
arch/x86/include/asm/mshyperv.h | 2 +
drivers/hv/mshv_common.c | 80 +++++++++++++++++++++++++++++++++
3 files changed, 83 insertions(+)
diff --git a/arch/x86/hyperv/hv_init.c b/arch/x86/hyperv/hv_init.c
index e28737ec7054..daf97a984b78 100644
--- a/arch/x86/hyperv/hv_init.c
+++ b/arch/x86/hyperv/hv_init.c
@@ -555,6 +555,7 @@ void __init hyperv_init(void)
hv_remap_tsc_clocksource();
hv_root_crash_init();
+ hv_sleep_notifiers_register();
} else {
hypercall_msr.guest_physical_address = vmalloc_to_pfn(hv_hypercall_pg);
wrmsrq(HV_X64_MSR_HYPERCALL, hypercall_msr.as_uint64);
diff --git a/arch/x86/include/asm/mshyperv.h b/arch/x86/include/asm/mshyperv.h
index 10037125099a..166053df0484 100644
--- a/arch/x86/include/asm/mshyperv.h
+++ b/arch/x86/include/asm/mshyperv.h
@@ -182,8 +182,10 @@ int hyperv_fill_flush_guest_mapping_list(
void hv_apic_init(void);
void __init hv_init_spinlocks(void);
bool hv_vcpu_is_preempted(int vcpu);
+void hv_sleep_notifiers_register(void);
#else
static inline void hv_apic_init(void) {}
+static inline void hv_sleep_notifiers_register(void) {};
#endif
struct irq_domain *hv_create_pci_msi_domain(void);
diff --git a/drivers/hv/mshv_common.c b/drivers/hv/mshv_common.c
index aa2be51979fd..f1d4e81107ee 100644
--- a/drivers/hv/mshv_common.c
+++ b/drivers/hv/mshv_common.c
@@ -14,6 +14,9 @@
#include <asm/mshyperv.h>
#include <linux/resume_user_mode.h>
#include <linux/export.h>
+#include <linux/acpi.h>
+#include <linux/notifier.h>
+#include <linux/reboot.h>
#include "mshv.h"
@@ -138,3 +141,80 @@ int hv_call_get_partition_property(u64 partition_id,
return 0;
}
EXPORT_SYMBOL_GPL(hv_call_get_partition_property);
+
+#ifdef CONFIG_X86_64
+/*
+ * Corresponding sleep states have to be initialized in order for a subsequent
+ * HVCALL_ENTER_SLEEP_STATE call to succeed. Currently only S5 state as per
+ * ACPI 6.4 chapter 7.4.2 is relevant, while S1, S2 and S3 can be supported.
+ *
+ * In order to pass proper PM values to mshv, ACPI should be initialized and
+ * should support S5 sleep state when this method is invoked.
+ */
+static int hv_initialize_sleep_states(void)
+{
+ u64 status;
+ unsigned long flags;
+ struct hv_input_set_system_property *in;
+ acpi_status acpi_status;
+ u8 sleep_type_a, sleep_type_b;
+
+ if (!acpi_sleep_state_supported(ACPI_STATE_S5)) {
+ pr_err("%s: S5 sleep state not supported.\n", __func__);
+ return -ENODEV;
+ }
+
+ acpi_status = acpi_get_sleep_type_data(ACPI_STATE_S5, &sleep_type_a,
+ &sleep_type_b);
+ if (ACPI_FAILURE(acpi_status))
+ return -ENODEV;
+
+ local_irq_save(flags);
+ in = *this_cpu_ptr(hyperv_pcpu_input_arg);
+ memset(in, 0, sizeof(*in));
+
+ in->property_id = HV_SYSTEM_PROPERTY_SLEEP_STATE;
+ in->set_sleep_state_info.sleep_state = HV_SLEEP_STATE_S5;
+ in->set_sleep_state_info.pm1a_slp_typ = sleep_type_a;
+ in->set_sleep_state_info.pm1b_slp_typ = sleep_type_b;
+
+ status = hv_do_hypercall(HVCALL_SET_SYSTEM_PROPERTY, in, NULL);
+ local_irq_restore(flags);
+
+ if (!hv_result_success(status)) {
+ hv_status_err(status, "\n");
+ return hv_result_to_errno(status);
+ }
+
+ return 0;
+}
+
+/*
+ * This notifier initializes sleep states in mshv hypervisor which will be
+ * used during power off.
+ */
+static int hv_reboot_notifier_handler(struct notifier_block *this,
+ unsigned long code, void *another)
+{
+ int ret = 0;
+
+ if (code == SYS_HALT || code == SYS_POWER_OFF)
+ ret = hv_initialize_sleep_states();
+
+ return ret ? NOTIFY_DONE : NOTIFY_OK;
+}
+
+static struct notifier_block hv_reboot_notifier = {
+ .notifier_call = hv_reboot_notifier_handler,
+};
+
+void hv_sleep_notifiers_register(void)
+{
+ int ret;
+
+ ret = register_reboot_notifier(&hv_reboot_notifier);
+ if (ret)
+ pr_err("%s: cannot register reboot notifier %d\n", __func__,
+ ret);
+}
+#endif
--
2.51.0
^ permalink raw reply related
* [PATCH v6 1/3] hyperv: Add definitions for MSHV sleep state configuration
From: Praveen K Paladugu @ 2025-11-26 21:49 UTC (permalink / raw)
To: kys, haiyangz, wei.liu, decui, tglx, mingo, linux-hyperv,
linux-kernel, bp, dave.hansen, x86, hpa, arnd
Cc: anbelski, prapal, easwar.hariharan, nunodasneves, skinsburskii
In-Reply-To: <20251126215013.11549-1-prapal@linux.microsoft.com>
Add the definitions required to configure sleep states in mshv hypervsior.
Signed-off-by: Praveen K Paladugu <prapal@linux.microsoft.com>
Co-developed-by: Anatol Belski <anbelski@linux.microsoft.com>
Signed-off-by: Anatol Belski <anbelski@linux.microsoft.com>
Reviewed-by: Easwar Hariharan <easwar.hariharan@linux.microsoft.com>
Reviewed-by: Nuno Das Neves <nunodasneves@linux.microsoft.com>
---
include/hyperv/hvgdk_mini.h | 4 +++-
include/hyperv/hvhdk_mini.h | 40 +++++++++++++++++++++++++++++++++++++
2 files changed, 43 insertions(+), 1 deletion(-)
diff --git a/include/hyperv/hvgdk_mini.h b/include/hyperv/hvgdk_mini.h
index 1d5ce11be8b6..04b18d0e37af 100644
--- a/include/hyperv/hvgdk_mini.h
+++ b/include/hyperv/hvgdk_mini.h
@@ -465,19 +465,21 @@ union hv_vp_assist_msr_contents { /* HV_REGISTER_VP_ASSIST_PAGE */
#define HVCALL_RESET_DEBUG_SESSION 0x006b
#define HVCALL_MAP_STATS_PAGE 0x006c
#define HVCALL_UNMAP_STATS_PAGE 0x006d
+#define HVCALL_SET_SYSTEM_PROPERTY 0x006f
#define HVCALL_ADD_LOGICAL_PROCESSOR 0x0076
#define HVCALL_GET_SYSTEM_PROPERTY 0x007b
#define HVCALL_MAP_DEVICE_INTERRUPT 0x007c
#define HVCALL_UNMAP_DEVICE_INTERRUPT 0x007d
#define HVCALL_RETARGET_INTERRUPT 0x007e
#define HVCALL_NOTIFY_PARTITION_EVENT 0x0087
+#define HVCALL_ENTER_SLEEP_STATE 0x0084
#define HVCALL_NOTIFY_PORT_RING_EMPTY 0x008b
#define HVCALL_REGISTER_INTERCEPT_RESULT 0x0091
#define HVCALL_ASSERT_VIRTUAL_INTERRUPT 0x0094
#define HVCALL_CREATE_PORT 0x0095
#define HVCALL_CONNECT_PORT 0x0096
#define HVCALL_START_VP 0x0099
-#define HVCALL_GET_VP_INDEX_FROM_APIC_ID 0x009a
+#define HVCALL_GET_VP_INDEX_FROM_APIC_ID 0x009a
#define HVCALL_FLUSH_GUEST_PHYSICAL_ADDRESS_SPACE 0x00af
#define HVCALL_FLUSH_GUEST_PHYSICAL_ADDRESS_LIST 0x00b0
#define HVCALL_SIGNAL_EVENT_DIRECT 0x00c0
diff --git a/include/hyperv/hvhdk_mini.h b/include/hyperv/hvhdk_mini.h
index f2d7b50de7a4..41a29bf8ec14 100644
--- a/include/hyperv/hvhdk_mini.h
+++ b/include/hyperv/hvhdk_mini.h
@@ -140,6 +140,7 @@ enum hv_snp_status {
enum hv_system_property {
/* Add more values when needed */
+ HV_SYSTEM_PROPERTY_SLEEP_STATE = 3,
HV_SYSTEM_PROPERTY_SCHEDULER_TYPE = 15,
HV_DYNAMIC_PROCESSOR_FEATURE_PROPERTY = 21,
HV_SYSTEM_PROPERTY_CRASHDUMPAREA = 47,
@@ -155,6 +156,19 @@ union hv_pfn_range { /* HV_SPA_PAGE_RANGE */
} __packed;
};
+enum hv_sleep_state {
+ HV_SLEEP_STATE_S1 = 1,
+ HV_SLEEP_STATE_S2 = 2,
+ HV_SLEEP_STATE_S3 = 3,
+ HV_SLEEP_STATE_S4 = 4,
+ HV_SLEEP_STATE_S5 = 5,
+ /*
+ * After hypervisor has received this, any follow up sleep
+ * state registration requests will be rejected.
+ */
+ HV_SLEEP_STATE_LOCK = 6
+};
+
enum hv_dynamic_processor_feature_property {
/* Add more values when needed */
HV_X64_DYNAMIC_PROCESSOR_FEATURE_MAX_ENCRYPTED_PARTITIONS = 13,
@@ -184,6 +198,32 @@ struct hv_output_get_system_property {
};
} __packed;
+struct hv_sleep_state_info {
+ u32 sleep_state; /* enum hv_sleep_state */
+ u8 pm1a_slp_typ;
+ u8 pm1b_slp_typ;
+} __packed;
+
+struct hv_input_set_system_property {
+ u32 property_id; /* enum hv_system_property */
+ u32 reserved;
+ union {
+ /* More fields to be filled in when needed */
+ struct hv_sleep_state_info set_sleep_state_info;
+
+ /*
+ * Add a reserved field to ensure the union is 8-byte aligned as
+ * existing members may not be. This is a temporary measure
+ * until all remaining members are added.
+ */
+ u64 reserved0[8];
+ };
+} __packed;
+
+struct hv_input_enter_sleep_state { /* HV_INPUT_ENTER_SLEEP_STATE */
+ u32 sleep_state; /* enum hv_sleep_state */
+} __packed;
+
struct hv_input_map_stats_page {
u32 type; /* enum hv_stats_object_type */
u32 padding;
--
2.51.0
^ permalink raw reply related
* [PATCH v6 0/3] Add support for clean shutdown with MSHV
From: Praveen K Paladugu @ 2025-11-26 21:49 UTC (permalink / raw)
To: kys, haiyangz, wei.liu, decui, tglx, mingo, linux-hyperv,
linux-kernel, bp, dave.hansen, x86, hpa, arnd
Cc: anbelski, prapal, easwar.hariharan, nunodasneves, skinsburskii
Add support for clean shutdown of the root partition when running on
MSHV Hypervisor.
v6:
- Fixed build errors, by adding CONFIG_X86_64 guard
- Moved machine_ops hook definition to ms_hyperv_init_platform
- Addressed review comments in v5
v5:
- Fixed build errors
- Padded struct hv_input_set_system_property for alignment
- Dropped CONFIG_ACPI stub
v4:
- Adopted machine_ops to order invoking HV_ENTER_SLEEP_STATE as the
last step in shutdown sequence.
- This ensures rest of the cleanups are done before powering off
v3:
- Dropped acpi_sleep handlers as they are not used on mshv
- Applied ordering for hv_reboot_notifier
- Fixed build issues on i386, arm64 architectures
v2:
- Addressed review comments from v1.
- Moved all sleep state handling methods under CONFIG_ACPI stub
- - This fixes build issues on non-x86 architectures.
Praveen K Paladugu (3):
hyperv: Add definitions for MSHV sleep state configuration
hyperv: Use reboot notifier to configure sleep state
hyperv: Cleanly shutdown root partition with MSHV
arch/x86/hyperv/hv_init.c | 1 +
arch/x86/include/asm/mshyperv.h | 4 ++
arch/x86/kernel/cpu/mshyperv.c | 2 +
drivers/hv/mshv_common.c | 98 +++++++++++++++++++++++++++++++++
include/hyperv/hvgdk_mini.h | 4 +-
include/hyperv/hvhdk_mini.h | 40 ++++++++++++++
6 files changed, 148 insertions(+), 1 deletion(-)
--
2.51.0
^ permalink raw reply
* RE: [EXTERNAL] Re: [Patch net-next v3] net: mana: Handle hardware recovery events when probing the device
From: Long Li @ 2025-11-26 21:46 UTC (permalink / raw)
To: Jakub Kicinski
Cc: Simon Horman, longli@linux.microsoft.com, KY Srinivasan,
Haiyang Zhang, Wei Liu, Dexuan Cui, David S. Miller, Eric Dumazet,
Paolo Abeni, Shradha Gupta, Konstantin Taranov,
Souradeep Chakrabarti, Erick Archer, linux-hyperv@vger.kernel.org,
netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-rdma@vger.kernel.org
In-Reply-To: <20251125172410.124e2eba@kernel.org>
> -----Original Message-----
> From: Jakub Kicinski <kuba@kernel.org>
> Sent: Tuesday, November 25, 2025 5:24 PM
> To: Long Li <longli@microsoft.com>
> Cc: Simon Horman <horms@kernel.org>; longli@linux.microsoft.com; KY
> Srinivasan <kys@microsoft.com>; Haiyang Zhang <haiyangz@microsoft.com>;
> Wei Liu <wei.liu@kernel.org>; Dexuan Cui <DECUI@microsoft.com>; David S.
> Miller <davem@davemloft.net>; Eric Dumazet <edumazet@google.com>;
> Paolo Abeni <pabeni@redhat.com>; Shradha Gupta
> <shradhagupta@linux.microsoft.com>; Konstantin Taranov
> <kotaranov@microsoft.com>; Souradeep Chakrabarti
> <schakrabarti@linux.microsoft.com>; Erick Archer
> <erick.archer@outlook.com>; linux-hyperv@vger.kernel.org;
> netdev@vger.kernel.org; linux-kernel@vger.kernel.org; linux-
> rdma@vger.kernel.org
> Subject: Re: [EXTERNAL] Re: [Patch net-next v3] net: mana: Handle hardware
> recovery events when probing the device
>
> On Wed, 26 Nov 2025 00:40:08 +0000 Long Li wrote:
> > > tags (with a blank line in between).
> >
> > Can we keep the "Fixes" tag and queue the patch to net-next?
>
> No.
>
> > There are other MANA patches pending in net-next and they will
> > conflict with each other when merging.
>
> We merge the trees every Thursday.
I have sent v4, with "Fixes" tag removed.
Thank you,
Long
^ permalink raw reply
* [Patch net-next v4] net: mana: Handle hardware recovery events when probing the device
From: longli @ 2025-11-26 21:45 UTC (permalink / raw)
To: K. Y. Srinivasan, Haiyang Zhang, Wei Liu, Dexuan Cui,
David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
Shradha Gupta, Simon Horman, Konstantin Taranov,
Souradeep Chakrabarti, Erick Archer, linux-hyperv, netdev,
linux-kernel, linux-rdma
Cc: Long Li
From: Long Li <longli@microsoft.com>
When MANA is being probed, it's possible that hardware is in recovery
mode and the device may get GDMA_EQE_HWC_RESET_REQUEST over HWC in the
middle of the probe. Detect such condition and go through the recovery
service procedure.
Signed-off-by: Long Li <longli@microsoft.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
---
Changes
v2: Use a list for handling multiple devices.
Use disable_delayed_work_sync() on driver exit.
Replace atomic_t with flags to detect if interrupt happens before probe finishes
v3: Rebase to latest net-next. Change list_for_each_entry_safe() to while(!list_empty()).
v4: Remove "Fixes:" tag.
.../net/ethernet/microsoft/mana/gdma_main.c | 176 ++++++++++++++++--
include/net/mana/gdma.h | 12 +-
2 files changed, 170 insertions(+), 18 deletions(-)
diff --git a/drivers/net/ethernet/microsoft/mana/gdma_main.c b/drivers/net/ethernet/microsoft/mana/gdma_main.c
index 8fd70b34807a..efb4e412ec7e 100644
--- a/drivers/net/ethernet/microsoft/mana/gdma_main.c
+++ b/drivers/net/ethernet/microsoft/mana/gdma_main.c
@@ -15,6 +15,20 @@
struct dentry *mana_debugfs_root;
+struct mana_dev_recovery {
+ struct list_head list;
+ struct pci_dev *pdev;
+ enum gdma_eqe_type type;
+};
+
+static struct mana_dev_recovery_work {
+ struct list_head dev_list;
+ struct delayed_work work;
+
+ /* Lock for dev_list above */
+ spinlock_t lock;
+} mana_dev_recovery_work;
+
static u32 mana_gd_r32(struct gdma_context *g, u64 offset)
{
return readl(g->bar0_va + offset);
@@ -387,6 +401,25 @@ EXPORT_SYMBOL_NS(mana_gd_ring_cq, "NET_MANA");
#define MANA_SERVICE_PERIOD 10
+static void mana_serv_rescan(struct pci_dev *pdev)
+{
+ struct pci_bus *parent;
+
+ pci_lock_rescan_remove();
+
+ parent = pdev->bus;
+ if (!parent) {
+ dev_err(&pdev->dev, "MANA service: no parent bus\n");
+ goto out;
+ }
+
+ pci_stop_and_remove_bus_device(pdev);
+ pci_rescan_bus(parent);
+
+out:
+ pci_unlock_rescan_remove();
+}
+
static void mana_serv_fpga(struct pci_dev *pdev)
{
struct pci_bus *bus, *parent;
@@ -419,9 +452,12 @@ static void mana_serv_reset(struct pci_dev *pdev)
{
struct gdma_context *gc = pci_get_drvdata(pdev);
struct hw_channel_context *hwc;
+ int ret;
if (!gc) {
- dev_err(&pdev->dev, "MANA service: no GC\n");
+ /* Perform PCI rescan on device if GC is not set up */
+ dev_err(&pdev->dev, "MANA service: GC not setup, rescanning\n");
+ mana_serv_rescan(pdev);
return;
}
@@ -440,9 +476,18 @@ static void mana_serv_reset(struct pci_dev *pdev)
msleep(MANA_SERVICE_PERIOD * 1000);
- mana_gd_resume(pdev);
+ ret = mana_gd_resume(pdev);
+ if (ret == -ETIMEDOUT || ret == -EPROTO) {
+ /* Perform PCI rescan on device if we failed on HWC */
+ dev_err(&pdev->dev, "MANA service: resume failed, rescanning\n");
+ mana_serv_rescan(pdev);
+ goto out;
+ }
- dev_info(&pdev->dev, "MANA reset cycle completed\n");
+ if (ret)
+ dev_info(&pdev->dev, "MANA reset cycle failed err %d\n", ret);
+ else
+ dev_info(&pdev->dev, "MANA reset cycle completed\n");
out:
gc->in_service = false;
@@ -454,18 +499,9 @@ struct mana_serv_work {
enum gdma_eqe_type type;
};
-static void mana_serv_func(struct work_struct *w)
+static void mana_do_service(enum gdma_eqe_type type, struct pci_dev *pdev)
{
- struct mana_serv_work *mns_wk;
- struct pci_dev *pdev;
-
- mns_wk = container_of(w, struct mana_serv_work, serv_work);
- pdev = mns_wk->pdev;
-
- if (!pdev)
- goto out;
-
- switch (mns_wk->type) {
+ switch (type) {
case GDMA_EQE_HWC_FPGA_RECONFIG:
mana_serv_fpga(pdev);
break;
@@ -475,12 +511,48 @@ static void mana_serv_func(struct work_struct *w)
break;
default:
- dev_err(&pdev->dev, "MANA service: unknown type %d\n",
- mns_wk->type);
+ dev_err(&pdev->dev, "MANA service: unknown type %d\n", type);
break;
}
+}
+
+static void mana_recovery_delayed_func(struct work_struct *w)
+{
+ struct mana_dev_recovery_work *work;
+ struct mana_dev_recovery *dev;
+ unsigned long flags;
+
+ work = container_of(w, struct mana_dev_recovery_work, work.work);
+
+ spin_lock_irqsave(&work->lock, flags);
+
+ while (!list_empty(&work->dev_list)) {
+ dev = list_first_entry(&work->dev_list,
+ struct mana_dev_recovery, list);
+ list_del(&dev->list);
+ spin_unlock_irqrestore(&work->lock, flags);
+
+ mana_do_service(dev->type, dev->pdev);
+ pci_dev_put(dev->pdev);
+ kfree(dev);
+
+ spin_lock_irqsave(&work->lock, flags);
+ }
+
+ spin_unlock_irqrestore(&work->lock, flags);
+}
+
+static void mana_serv_func(struct work_struct *w)
+{
+ struct mana_serv_work *mns_wk;
+ struct pci_dev *pdev;
+
+ mns_wk = container_of(w, struct mana_serv_work, serv_work);
+ pdev = mns_wk->pdev;
+
+ if (pdev)
+ mana_do_service(mns_wk->type, pdev);
-out:
pci_dev_put(pdev);
kfree(mns_wk);
module_put(THIS_MODULE);
@@ -541,6 +613,17 @@ static void mana_gd_process_eqe(struct gdma_queue *eq)
case GDMA_EQE_HWC_RESET_REQUEST:
dev_info(gc->dev, "Recv MANA service type:%d\n", type);
+ if (!test_and_set_bit(GC_PROBE_SUCCEEDED, &gc->flags)) {
+ /*
+ * Device is in probe and we received a hardware reset
+ * event, the probe function will detect that the flag
+ * has changed and perform service procedure.
+ */
+ dev_info(gc->dev,
+ "Service is to be processed in probe\n");
+ break;
+ }
+
if (gc->in_service) {
dev_info(gc->dev, "Already in service\n");
break;
@@ -1938,8 +2021,19 @@ static int mana_gd_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
if (err)
goto cleanup_mana;
+ /*
+ * If a hardware reset event has occurred over HWC during probe,
+ * rollback and perform hardware reset procedure.
+ */
+ if (test_and_set_bit(GC_PROBE_SUCCEEDED, &gc->flags)) {
+ err = -EPROTO;
+ goto cleanup_mana_rdma;
+ }
+
return 0;
+cleanup_mana_rdma:
+ mana_rdma_remove(&gc->mana_ib);
cleanup_mana:
mana_remove(&gc->mana, false);
cleanup_gd:
@@ -1963,6 +2057,35 @@ static int mana_gd_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
disable_dev:
pci_disable_device(pdev);
dev_err(&pdev->dev, "gdma probe failed: err = %d\n", err);
+
+ /*
+ * Hardware could be in recovery mode and the HWC returns TIMEDOUT or
+ * EPROTO from mana_gd_setup(), mana_probe() or mana_rdma_probe(), or
+ * we received a hardware reset event over HWC interrupt. In this case,
+ * perform the device recovery procedure after MANA_SERVICE_PERIOD
+ * seconds.
+ */
+ if (err == -ETIMEDOUT || err == -EPROTO) {
+ struct mana_dev_recovery *dev;
+ unsigned long flags;
+
+ dev_info(&pdev->dev, "Start MANA recovery mode\n");
+
+ dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+ if (!dev)
+ return err;
+
+ dev->pdev = pci_dev_get(pdev);
+ dev->type = GDMA_EQE_HWC_RESET_REQUEST;
+
+ spin_lock_irqsave(&mana_dev_recovery_work.lock, flags);
+ list_add_tail(&dev->list, &mana_dev_recovery_work.dev_list);
+ spin_unlock_irqrestore(&mana_dev_recovery_work.lock, flags);
+
+ schedule_delayed_work(&mana_dev_recovery_work.work,
+ secs_to_jiffies(MANA_SERVICE_PERIOD));
+ }
+
return err;
}
@@ -2067,6 +2190,10 @@ static int __init mana_driver_init(void)
{
int err;
+ INIT_LIST_HEAD(&mana_dev_recovery_work.dev_list);
+ spin_lock_init(&mana_dev_recovery_work.lock);
+ INIT_DELAYED_WORK(&mana_dev_recovery_work.work, mana_recovery_delayed_func);
+
mana_debugfs_root = debugfs_create_dir("mana", NULL);
err = pci_register_driver(&mana_driver);
@@ -2080,6 +2207,21 @@ static int __init mana_driver_init(void)
static void __exit mana_driver_exit(void)
{
+ struct mana_dev_recovery *dev;
+ unsigned long flags;
+
+ disable_delayed_work_sync(&mana_dev_recovery_work.work);
+
+ spin_lock_irqsave(&mana_dev_recovery_work.lock, flags);
+ while (!list_empty(&mana_dev_recovery_work.dev_list)) {
+ dev = list_first_entry(&mana_dev_recovery_work.dev_list,
+ struct mana_dev_recovery, list);
+ list_del(&dev->list);
+ pci_dev_put(dev->pdev);
+ kfree(dev);
+ }
+ spin_unlock_irqrestore(&mana_dev_recovery_work.lock, flags);
+
pci_unregister_driver(&mana_driver);
debugfs_remove(mana_debugfs_root);
diff --git a/include/net/mana/gdma.h b/include/net/mana/gdma.h
index a4cf307859f8..eaa27483f99b 100644
--- a/include/net/mana/gdma.h
+++ b/include/net/mana/gdma.h
@@ -382,6 +382,10 @@ struct gdma_irq_context {
char name[MANA_IRQ_NAME_SZ];
};
+enum gdma_context_flags {
+ GC_PROBE_SUCCEEDED = 0,
+};
+
struct gdma_context {
struct device *dev;
struct dentry *mana_pci_debugfs;
@@ -430,6 +434,8 @@ struct gdma_context {
u64 pf_cap_flags1;
struct workqueue_struct *service_wq;
+
+ unsigned long flags;
};
static inline bool mana_gd_is_mana(struct gdma_dev *gd)
@@ -600,6 +606,9 @@ enum {
/* Driver can send HWC periodically to query stats */
#define GDMA_DRV_CAP_FLAG_1_PERIODIC_STATS_QUERY BIT(21)
+/* Driver can handle hardware recovery events during probe */
+#define GDMA_DRV_CAP_FLAG_1_PROBE_RECOVERY BIT(22)
+
#define GDMA_DRV_CAP_FLAGS1 \
(GDMA_DRV_CAP_FLAG_1_EQ_SHARING_MULTI_VPORT | \
GDMA_DRV_CAP_FLAG_1_NAPI_WKDONE_FIX | \
@@ -611,7 +620,8 @@ enum {
GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE | \
GDMA_DRV_CAP_FLAG_1_HW_VPORT_LINK_AWARE | \
GDMA_DRV_CAP_FLAG_1_PERIODIC_STATS_QUERY | \
- GDMA_DRV_CAP_FLAG_1_SKB_LINEARIZE)
+ GDMA_DRV_CAP_FLAG_1_SKB_LINEARIZE | \
+ GDMA_DRV_CAP_FLAG_1_PROBE_RECOVERY)
#define GDMA_DRV_CAP_FLAGS2 0
--
2.43.0
^ permalink raw reply related
* Re: [PATCH 2/3] irqchip/gic-v3: allocate one SGI for MSHV
From: Marc Zyngier @ 2025-11-26 21:27 UTC (permalink / raw)
To: Anirudh Rayabharam
Cc: kys, haiyangz, wei.liu, decui, longli, catalin.marinas, will,
tglx, Arnd Bergmann, akpm, agordeev, guoweikang.kernel, osandov,
bsz, linux-hyperv, linux-arm-kernel, linux-kernel, linux-arch
In-Reply-To: <aSbahzqu_3GN-PPJ@anirudh-surface.localdomain>
On Wed, 26 Nov 2025 10:46:31 +0000,
Anirudh Rayabharam <anirudh@anirudhrb.com> wrote:
>
> On Wed, Nov 26, 2025 at 09:02:30AM +0000, Marc Zyngier wrote:
> > On Wed, 26 Nov 2025 08:51:59 +0000,
> > Anirudh Rayabharam <anirudh@anirudhrb.com> wrote:
> > >
> > > On Tue, Nov 25, 2025 at 06:01:38PM +0000, Marc Zyngier wrote:
> > > > On Tue, 25 Nov 2025 17:01:23 +0000,
> > > > Anirudh Raybharam <anirudh@anirudhrb.com> wrote:
> > > > >
> > > > > From: Anirudh Rayabharam <anirudh@anirudhrb.com>
> > > > >
> > > > > From: Anirudh Rayabharam (Microsoft) <anirudh@anirudhrb.com>
> > > > >
> > > > > Currently SGIs are allocated only for the smp subsystem. The MSHV
> > > > > (Microsoft Hypervisor aka Hyper-V) code also needs an SGI that can be
> > > > > programmed into the SYNIC to receive intercepts from the hypervisor. The
> > > > > hypervisor would then assert this SGI whenever there is a guest
> > > > > VMEXIT.
> > > > >
> > > > > Allocate one SGI for MSHV use in addition to the SGIs allocated for
> > > > > IPIs. When running under MSHV, the full SGI range can be used i.e. no
> > > > > need to reserve SGIs 8-15 for the secure firmware.
> > > > >
> > > > > Since this SGI is needed only when running as a parent partition (i.e.
> > > > > we can create guest partitions), check for it before allocating an SGI.
> > > >
> > > > Sorry, but that's not an acceptable situation.
> > > >
> > > > SGIs are for Linux to use, nobody else, and that allocation must be
> > >
> > > Why does this restriction exist? In the code SGIs 8-15 are left for
> > > secure firmware. So, things other than Linux can use SGIs. Why not MSHV
> > > then?
> >
> > Because SGIs are for *internal* usage. Not usage from another random
> > piece of SW. The ACPI tables explicitly don't describe SGIs. DT
> > explicitly don't describe SGIs. Do you get the clue?
>
> The name Software Generated Interrupts suggests that it is supposed to be
> used by pieces of SW.
I'm so glad you spell it out for me, I had no idea. I can't help but
notice that it is not called SGIFRPOSIDKA (Software Generated
Interrupt From Random Piece Of Software I Don't Know About).
Linux owns the SGIs, full stop. If you want to generate interrupts
from outside of Linux, use a standard interrupts. Not rocket science.
> Yes, ACPI/DT don't describe SGIs because they're not supposed to be used
> by devices. SW has full control over SGIs and it is up to the SW to
> assign meaning to them, isn't it?
No. It means that a single *consistent* software agent *owns* these
interrupts and doesn't have to let *anyone* else use them from outer
space.
> > > > the same irrespective of whether Linux runs virtualised or not. This
> > > > also won't work with GICv5 (there are no SGIs at all), so this is
> > > > doomed from the very start, and would immediately create technical
> > > > debt.
> > >
> > > Hyper-V always presents a GICv3 so we don't need to worry about GICv5.
> >
> > Well, that's pretty short sighted of you, and eventually you'll have
> > to support it, or just die. So do the right thing from the beginning.
>
> Well, we don't when or if that will happen. But if it does happen, we
> can solve it in a way that makes sense for GICv5. If there are no SGIs
> at all, great, maybe we'll have a nicer solution then.
Great. See you then. In the meantime, I have no interest in this sort
of sorry hacks polluting the stuff I maintain.
> > > > If you want to signal an interrupt to Linux, expose a device with an
> > > > interrupt in a firmware table (i.e. not an SGI), and use that in your
> > > > driver.
> > >
> > > You mean in the ACPI tables? That would require us to modify the
> > > firmware to expose this virtual device right?
> >
> > Yes. How is that surprising?
>
> It's not ideal that we would need some custom firmware to run Linux on
> MSHV (as a parent). Do you think there could be some other possible solution
> for handling this in the kernel? Maybe by thinking of it as a platform specific
> quirk or something?
You either do it the *correct* way, by exposing a virtual device, with
an edge-triggered PPI, just like other hypervisors have done, or you
keep your toy to yourself. It is that simple. We don't have to accept
ugly crap in Linux just for the sake of you not having to do the right
thing in your firmware.
Feel free to post a new series once you have something that matches
the above expectations.
M.
--
Jazz isn't dead. It just smells funny.
^ permalink raw reply
* [PATCH v3 9/9] efi: libstub: Simplify interfaces for primary_display
From: Thomas Zimmermann @ 2025-11-26 16:03 UTC (permalink / raw)
To: ardb, javierm, arnd, richard.lyu, helgaas
Cc: x86, linux-arm-kernel, linux-kernel, linux-efi, loongarch,
linux-riscv, dri-devel, linux-hyperv, linux-pci, linux-fbdev,
Thomas Zimmermann
In-Reply-To: <20251126160854.553077-1-tzimmermann@suse.de>
Rename alloc_primary_display() and __alloc_primary_display(), clarify
free semantics to make interfaces easier to understand.
Rename alloc_primary_display() to lookup_primary_display() as it
does not necessarily allocate. Then rename __alloc_primary_display()
to the new alloc_primary_display(). The helper belongs to
free_primary_display), so it should be named without underscores.
The lookup helper does not necessarily allocate, so the output
parameter needs_free to indicate when free should be called. Pass
an argument through the calls to track this state. Put the free
handling into release_primary_display() for simplificy.
Also move the comment fro primary_display.c to efi-stub-entry.c,
where it now describes lookup_primary_display().
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
---
drivers/firmware/efi/libstub/efi-stub-entry.c | 23 +++++++++++++++++--
drivers/firmware/efi/libstub/efi-stub.c | 22 ++++++++++++------
drivers/firmware/efi/libstub/efistub.h | 2 +-
.../firmware/efi/libstub/primary_display.c | 17 +-------------
drivers/firmware/efi/libstub/zboot.c | 6 +++--
5 files changed, 42 insertions(+), 28 deletions(-)
diff --git a/drivers/firmware/efi/libstub/efi-stub-entry.c b/drivers/firmware/efi/libstub/efi-stub-entry.c
index aa85e910fe59..3077b51fe0b2 100644
--- a/drivers/firmware/efi/libstub/efi-stub-entry.c
+++ b/drivers/firmware/efi/libstub/efi-stub-entry.c
@@ -14,10 +14,29 @@ static void *kernel_image_addr(void *addr)
return addr + kernel_image_offset;
}
-struct sysfb_display_info *alloc_primary_display(void)
+/*
+ * There are two ways of populating the core kernel's sysfb_primary_display
+ * via the stub:
+ *
+ * - using a configuration table, which relies on the EFI init code to
+ * locate the table and copy the contents; or
+ *
+ * - by linking directly to the core kernel's copy of the global symbol.
+ *
+ * The latter is preferred because it makes the EFIFB earlycon available very
+ * early, but it only works if the EFI stub is part of the core kernel image
+ * itself. The zboot decompressor can only use the configuration table
+ * approach.
+ */
+
+struct sysfb_display_info *lookup_primary_display(bool *needs_free)
{
+ *needs_free = true;
+
if (IS_ENABLED(CONFIG_ARM))
- return __alloc_primary_display();
+ return alloc_primary_display();
+
+ *needs_free = false;
if (IS_ENABLED(CONFIG_X86) ||
IS_ENABLED(CONFIG_EFI_EARLYCON) ||
diff --git a/drivers/firmware/efi/libstub/efi-stub.c b/drivers/firmware/efi/libstub/efi-stub.c
index 42d6073bcd06..dc545f62c62b 100644
--- a/drivers/firmware/efi/libstub/efi-stub.c
+++ b/drivers/firmware/efi/libstub/efi-stub.c
@@ -51,14 +51,14 @@ static bool flat_va_mapping = (EFI_RT_VIRTUAL_OFFSET != 0);
void __weak free_primary_display(struct sysfb_display_info *dpy)
{ }
-static struct sysfb_display_info *setup_primary_display(void)
+static struct sysfb_display_info *setup_primary_display(bool *dpy_needs_free)
{
struct sysfb_display_info *dpy;
struct screen_info *screen = NULL;
struct edid_info *edid = NULL;
efi_status_t status;
- dpy = alloc_primary_display();
+ dpy = lookup_primary_display(dpy_needs_free);
if (!dpy)
return NULL;
screen = &dpy->screen;
@@ -68,15 +68,22 @@ static struct sysfb_display_info *setup_primary_display(void)
status = efi_setup_graphics(screen, edid);
if (status != EFI_SUCCESS)
- goto err_free_primary_display;
+ goto err___free_primary_display;
return dpy;
-err_free_primary_display:
- free_primary_display(dpy);
+err___free_primary_display:
+ if (*dpy_needs_free)
+ free_primary_display(dpy);
return NULL;
}
+static void release_primary_display(struct sysfb_display_info *dpy, bool dpy_needs_free)
+{
+ if (dpy && dpy_needs_free)
+ free_primary_display(dpy);
+}
+
static void install_memreserve_table(void)
{
struct linux_efi_memreserve *rsv;
@@ -156,13 +163,14 @@ efi_status_t efi_stub_common(efi_handle_t handle,
char *cmdline_ptr)
{
struct sysfb_display_info *dpy;
+ bool dpy_needs_free;
efi_status_t status;
status = check_platform_features();
if (status != EFI_SUCCESS)
return status;
- dpy = setup_primary_display();
+ dpy = setup_primary_display(&dpy_needs_free);
efi_retrieve_eventlog();
@@ -182,7 +190,7 @@ efi_status_t efi_stub_common(efi_handle_t handle,
status = efi_boot_kernel(handle, image, image_addr, cmdline_ptr);
- free_primary_display(dpy);
+ release_primary_display(dpy, dpy_needs_free);
return status;
}
diff --git a/drivers/firmware/efi/libstub/efistub.h b/drivers/firmware/efi/libstub/efistub.h
index 979a21818cc1..1503ffb82903 100644
--- a/drivers/firmware/efi/libstub/efistub.h
+++ b/drivers/firmware/efi/libstub/efistub.h
@@ -1176,8 +1176,8 @@ efi_enable_reset_attack_mitigation(void) { }
void efi_retrieve_eventlog(void);
+struct sysfb_display_info *lookup_primary_display(bool *needs_free);
struct sysfb_display_info *alloc_primary_display(void);
-struct sysfb_display_info *__alloc_primary_display(void);
void free_primary_display(struct sysfb_display_info *dpy);
void efi_cache_sync_image(unsigned long image_base,
diff --git a/drivers/firmware/efi/libstub/primary_display.c b/drivers/firmware/efi/libstub/primary_display.c
index cdaebab26514..34c54ac1e02a 100644
--- a/drivers/firmware/efi/libstub/primary_display.c
+++ b/drivers/firmware/efi/libstub/primary_display.c
@@ -7,24 +7,9 @@
#include "efistub.h"
-/*
- * There are two ways of populating the core kernel's sysfb_primary_display
- * via the stub:
- *
- * - using a configuration table, which relies on the EFI init code to
- * locate the table and copy the contents; or
- *
- * - by linking directly to the core kernel's copy of the global symbol.
- *
- * The latter is preferred because it makes the EFIFB earlycon available very
- * early, but it only works if the EFI stub is part of the core kernel image
- * itself. The zboot decompressor can only use the configuration table
- * approach.
- */
-
static efi_guid_t primary_display_guid = LINUX_EFI_PRIMARY_DISPLAY_TABLE_GUID;
-struct sysfb_display_info *__alloc_primary_display(void)
+struct sysfb_display_info *alloc_primary_display(void)
{
struct sysfb_display_info *dpy;
efi_status_t status;
diff --git a/drivers/firmware/efi/libstub/zboot.c b/drivers/firmware/efi/libstub/zboot.c
index 4b76f74c56da..c1fd1fdbcb08 100644
--- a/drivers/firmware/efi/libstub/zboot.c
+++ b/drivers/firmware/efi/libstub/zboot.c
@@ -26,9 +26,11 @@ void __weak efi_cache_sync_image(unsigned long image_base,
// executable code loaded into memory to be safe for execution.
}
-struct sysfb_display_info *alloc_primary_display(void)
+struct sysfb_display_info *lookup_primary_display(bool *needs_free)
{
- return __alloc_primary_display();
+ *needs_free = true;
+
+ return alloc_primary_display();
}
asmlinkage efi_status_t __efiapi
--
2.51.1
^ permalink raw reply related
* [PATCH v3 8/9] efi: Support EDID information
From: Thomas Zimmermann @ 2025-11-26 16:03 UTC (permalink / raw)
To: ardb, javierm, arnd, richard.lyu, helgaas
Cc: x86, linux-arm-kernel, linux-kernel, linux-efi, loongarch,
linux-riscv, dri-devel, linux-hyperv, linux-pci, linux-fbdev,
Thomas Zimmermann
In-Reply-To: <20251126160854.553077-1-tzimmermann@suse.de>
In the EFI config table, rename LINUX_EFI_SCREEN_INFO_TABLE_GUID to
LINUX_EFI_PRIMARY_DISPLAY_TABLE_GUID. Read sysfb_primary_display
from the entry. In addition to the screen_info, the entry now also
contains EDID information.
In libstub, replace struct screen_info with struct sysfb_display_info
from the kernel's sysfb_primary_display and rename functions accordingly.
Transfer it to the runtime kernel using the kernel's global state or
the config-table entry LINUX_EFI_PRIMARY_DISPLAY_TABLE_GUID.
With CONFIG_FIRMWARE_EDID=y, libstub now transfers the GOP device's
EDID information to the kernel. If CONFIG_FIRMWARE_EDID=n, EDID
information is disabled. Make the Kconfig symbol CONFIG_FIRMWARE_EDID
available with EFI. Setting the value to 'n' disables EDID support.
Also rename screen_info.c to primary_display.c and adapt the contained
comment according to the changes.
v3:
- replace SCREEN_INFO table (Ard)
- merge libstub changes
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
---
arch/loongarch/kernel/efi.c | 17 +++---
drivers/firmware/efi/efi-init.c | 17 +++---
drivers/firmware/efi/efi.c | 4 +-
drivers/firmware/efi/libstub/Makefile | 2 +-
drivers/firmware/efi/libstub/efi-stub-entry.c | 11 ++--
drivers/firmware/efi/libstub/efi-stub.c | 41 +++++++++-----
drivers/firmware/efi/libstub/efistub.h | 7 ++-
.../firmware/efi/libstub/primary_display.c | 56 +++++++++++++++++++
drivers/firmware/efi/libstub/screen_info.c | 53 ------------------
drivers/firmware/efi/libstub/zboot.c | 4 +-
drivers/video/Kconfig | 8 ++-
include/linux/efi.h | 9 +--
12 files changed, 123 insertions(+), 106 deletions(-)
create mode 100644 drivers/firmware/efi/libstub/primary_display.c
delete mode 100644 drivers/firmware/efi/libstub/screen_info.c
diff --git a/arch/loongarch/kernel/efi.c b/arch/loongarch/kernel/efi.c
index 1ef38036e8ae..342aca00d42c 100644
--- a/arch/loongarch/kernel/efi.c
+++ b/arch/loongarch/kernel/efi.c
@@ -72,7 +72,7 @@ bool efi_poweroff_required(void)
(acpi_gbl_reduced_hardware || acpi_no_s5);
}
-unsigned long __initdata screen_info_table = EFI_INVALID_TABLE_ADDR;
+unsigned long __initdata primary_display_table = EFI_INVALID_TABLE_ADDR;
#if defined(CONFIG_SYSFB) || defined(CONFIG_EFI_EARLYCON)
struct sysfb_display_info sysfb_primary_display __section(".data");
@@ -81,16 +81,17 @@ EXPORT_SYMBOL_GPL(sysfb_primary_display);
static void __init init_primary_display(void)
{
- if (screen_info_table == EFI_INVALID_TABLE_ADDR) {
- struct screen_info *si = early_memremap(screen_info_table, sizeof(*si));
+ if (primary_display_table != EFI_INVALID_TABLE_ADDR) {
+ struct sysfb_display_info *dpy =
+ early_memremap(primary_display_table, sizeof(*dpy));
- if (!si) {
- pr_err("Could not map screen_info config table\n");
+ if (!dpy) {
+ pr_err("Could not map primary_display config table\n");
return;
}
- sysfb_primary_display.screen = *si;
- memset(si, 0, sizeof(*si));
- early_memunmap(si, sizeof(*si));
+ sysfb_primary_display = *dpy;
+ memset(dpy, 0, sizeof(*dpy));
+ early_memunmap(dpy, sizeof(*dpy));
} else {
return;
}
diff --git a/drivers/firmware/efi/efi-init.c b/drivers/firmware/efi/efi-init.c
index ca697d485116..b22e208a1aae 100644
--- a/drivers/firmware/efi/efi-init.c
+++ b/drivers/firmware/efi/efi-init.c
@@ -23,7 +23,7 @@
#include <asm/efi.h>
-unsigned long __initdata screen_info_table = EFI_INVALID_TABLE_ADDR;
+unsigned long __initdata primary_display_table = EFI_INVALID_TABLE_ADDR;
static int __init is_memory(efi_memory_desc_t *md)
{
@@ -67,16 +67,17 @@ EXPORT_SYMBOL_GPL(sysfb_primary_display);
static void __init init_primary_display(void)
{
- if (screen_info_table != EFI_INVALID_TABLE_ADDR) {
- struct screen_info *si = early_memremap(screen_info_table, sizeof(*si));
+ if (primary_display_table != EFI_INVALID_TABLE_ADDR) {
+ struct sysfb_display_info *dpy =
+ early_memremap(primary_display_table, sizeof(*dpy));
- if (!si) {
- pr_err("Could not map screen_info config table\n");
+ if (!dpy) {
+ pr_err("Could not map primary_display config table\n");
return;
}
- sysfb_primary_display.screen = *si;
- memset(si, 0, sizeof(*si));
- early_memunmap(si, sizeof(*si));
+ sysfb_primary_display = *dpy;
+ memset(dpy, 0, sizeof(*dpy));
+ early_memunmap(dpy, sizeof(*dpy));
} else {
return;
}
diff --git a/drivers/firmware/efi/efi.c b/drivers/firmware/efi/efi.c
index a9070d00b833..955193691f10 100644
--- a/drivers/firmware/efi/efi.c
+++ b/drivers/firmware/efi/efi.c
@@ -63,7 +63,7 @@ static unsigned long __initdata mem_reserve = EFI_INVALID_TABLE_ADDR;
static unsigned long __initdata rt_prop = EFI_INVALID_TABLE_ADDR;
static unsigned long __initdata initrd = EFI_INVALID_TABLE_ADDR;
-extern unsigned long screen_info_table;
+extern unsigned long primary_display_table;
struct mm_struct efi_mm = {
.mm_mt = MTREE_INIT_EXT(mm_mt, MM_MT_FLAGS, efi_mm.mmap_lock),
@@ -641,7 +641,7 @@ static const efi_config_table_type_t common_tables[] __initconst = {
{LINUX_EFI_UNACCEPTED_MEM_TABLE_GUID, &efi.unaccepted, "Unaccepted" },
#endif
#ifdef CONFIG_EFI_GENERIC_STUB
- {LINUX_EFI_SCREEN_INFO_TABLE_GUID, &screen_info_table },
+ {LINUX_EFI_PRIMARY_DISPLAY_TABLE_GUID, &primary_display_table },
#endif
{},
};
diff --git a/drivers/firmware/efi/libstub/Makefile b/drivers/firmware/efi/libstub/Makefile
index 7d15a85d579f..e386ffd009b7 100644
--- a/drivers/firmware/efi/libstub/Makefile
+++ b/drivers/firmware/efi/libstub/Makefile
@@ -80,7 +80,7 @@ $(obj)/lib-%.o: $(srctree)/lib/%.c FORCE
$(call if_changed_rule,cc_o_c)
lib-$(CONFIG_EFI_GENERIC_STUB) += efi-stub.o string.o intrinsics.o systable.o \
- screen_info.o efi-stub-entry.o
+ primary_display.o efi-stub-entry.o
lib-$(CONFIG_ARM) += arm32-stub.o
lib-$(CONFIG_ARM64) += kaslr.o arm64.o arm64-stub.o smbios.o
diff --git a/drivers/firmware/efi/libstub/efi-stub-entry.c b/drivers/firmware/efi/libstub/efi-stub-entry.c
index 401ecbbdf331..aa85e910fe59 100644
--- a/drivers/firmware/efi/libstub/efi-stub-entry.c
+++ b/drivers/firmware/efi/libstub/efi-stub-entry.c
@@ -14,18 +14,15 @@ static void *kernel_image_addr(void *addr)
return addr + kernel_image_offset;
}
-struct screen_info *alloc_screen_info(void)
+struct sysfb_display_info *alloc_primary_display(void)
{
if (IS_ENABLED(CONFIG_ARM))
- return __alloc_screen_info();
+ return __alloc_primary_display();
if (IS_ENABLED(CONFIG_X86) ||
IS_ENABLED(CONFIG_EFI_EARLYCON) ||
- IS_ENABLED(CONFIG_SYSFB)) {
- struct sysfb_display_info *dpy = kernel_image_addr(&sysfb_primary_display);
-
- return &dpy->screen;
- }
+ IS_ENABLED(CONFIG_SYSFB))
+ return kernel_image_addr(&sysfb_primary_display);
return NULL;
}
diff --git a/drivers/firmware/efi/libstub/efi-stub.c b/drivers/firmware/efi/libstub/efi-stub.c
index 9cb814c5ba1b..42d6073bcd06 100644
--- a/drivers/firmware/efi/libstub/efi-stub.c
+++ b/drivers/firmware/efi/libstub/efi-stub.c
@@ -10,7 +10,7 @@
*/
#include <linux/efi.h>
-#include <linux/screen_info.h>
+#include <linux/sysfb.h>
#include <asm/efi.h>
#include "efistub.h"
@@ -48,23 +48,33 @@
static u64 virtmap_base = EFI_RT_VIRTUAL_BASE;
static bool flat_va_mapping = (EFI_RT_VIRTUAL_OFFSET != 0);
-void __weak free_screen_info(struct screen_info *si)
-{
-}
+void __weak free_primary_display(struct sysfb_display_info *dpy)
+{ }
-static struct screen_info *setup_graphics(void)
+static struct sysfb_display_info *setup_primary_display(void)
{
- struct screen_info *si, tmp = {};
+ struct sysfb_display_info *dpy;
+ struct screen_info *screen = NULL;
+ struct edid_info *edid = NULL;
+ efi_status_t status;
- if (efi_setup_graphics(&tmp, NULL) != EFI_SUCCESS)
+ dpy = alloc_primary_display();
+ if (!dpy)
return NULL;
+ screen = &dpy->screen;
+#if defined(CONFIG_FIRMWARE_EDID)
+ edid = &dpy->edid;
+#endif
- si = alloc_screen_info();
- if (!si)
- return NULL;
+ status = efi_setup_graphics(screen, edid);
+ if (status != EFI_SUCCESS)
+ goto err_free_primary_display;
- *si = tmp;
- return si;
+ return dpy;
+
+err_free_primary_display:
+ free_primary_display(dpy);
+ return NULL;
}
static void install_memreserve_table(void)
@@ -145,14 +155,14 @@ efi_status_t efi_stub_common(efi_handle_t handle,
unsigned long image_addr,
char *cmdline_ptr)
{
- struct screen_info *si;
+ struct sysfb_display_info *dpy;
efi_status_t status;
status = check_platform_features();
if (status != EFI_SUCCESS)
return status;
- si = setup_graphics();
+ dpy = setup_primary_display();
efi_retrieve_eventlog();
@@ -172,7 +182,8 @@ efi_status_t efi_stub_common(efi_handle_t handle,
status = efi_boot_kernel(handle, image, image_addr, cmdline_ptr);
- free_screen_info(si);
+ free_primary_display(dpy);
+
return status;
}
diff --git a/drivers/firmware/efi/libstub/efistub.h b/drivers/firmware/efi/libstub/efistub.h
index b2fb0c3fa721..979a21818cc1 100644
--- a/drivers/firmware/efi/libstub/efistub.h
+++ b/drivers/firmware/efi/libstub/efistub.h
@@ -36,6 +36,7 @@
struct edid_info;
struct screen_info;
+struct sysfb_display_info;
extern bool efi_no5lvl;
extern bool efi_nochunk;
@@ -1175,9 +1176,9 @@ efi_enable_reset_attack_mitigation(void) { }
void efi_retrieve_eventlog(void);
-struct screen_info *alloc_screen_info(void);
-struct screen_info *__alloc_screen_info(void);
-void free_screen_info(struct screen_info *si);
+struct sysfb_display_info *alloc_primary_display(void);
+struct sysfb_display_info *__alloc_primary_display(void);
+void free_primary_display(struct sysfb_display_info *dpy);
void efi_cache_sync_image(unsigned long image_base,
unsigned long alloc_size);
diff --git a/drivers/firmware/efi/libstub/primary_display.c b/drivers/firmware/efi/libstub/primary_display.c
new file mode 100644
index 000000000000..cdaebab26514
--- /dev/null
+++ b/drivers/firmware/efi/libstub/primary_display.c
@@ -0,0 +1,56 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/efi.h>
+#include <linux/sysfb.h>
+
+#include <asm/efi.h>
+
+#include "efistub.h"
+
+/*
+ * There are two ways of populating the core kernel's sysfb_primary_display
+ * via the stub:
+ *
+ * - using a configuration table, which relies on the EFI init code to
+ * locate the table and copy the contents; or
+ *
+ * - by linking directly to the core kernel's copy of the global symbol.
+ *
+ * The latter is preferred because it makes the EFIFB earlycon available very
+ * early, but it only works if the EFI stub is part of the core kernel image
+ * itself. The zboot decompressor can only use the configuration table
+ * approach.
+ */
+
+static efi_guid_t primary_display_guid = LINUX_EFI_PRIMARY_DISPLAY_TABLE_GUID;
+
+struct sysfb_display_info *__alloc_primary_display(void)
+{
+ struct sysfb_display_info *dpy;
+ efi_status_t status;
+
+ status = efi_bs_call(allocate_pool, EFI_ACPI_RECLAIM_MEMORY,
+ sizeof(*dpy), (void **)&dpy);
+
+ if (status != EFI_SUCCESS)
+ return NULL;
+
+ memset(dpy, 0, sizeof(*dpy));
+
+ status = efi_bs_call(install_configuration_table,
+ &primary_display_guid, dpy);
+ if (status == EFI_SUCCESS)
+ return dpy;
+
+ efi_bs_call(free_pool, dpy);
+ return NULL;
+}
+
+void free_primary_display(struct sysfb_display_info *dpy)
+{
+ if (!dpy)
+ return;
+
+ efi_bs_call(install_configuration_table, &primary_display_guid, NULL);
+ efi_bs_call(free_pool, dpy);
+}
diff --git a/drivers/firmware/efi/libstub/screen_info.c b/drivers/firmware/efi/libstub/screen_info.c
deleted file mode 100644
index 5d3a1e32d177..000000000000
--- a/drivers/firmware/efi/libstub/screen_info.c
+++ /dev/null
@@ -1,53 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-
-#include <linux/efi.h>
-#include <linux/screen_info.h>
-
-#include <asm/efi.h>
-
-#include "efistub.h"
-
-/*
- * There are two ways of populating the core kernel's struct screen_info via the stub:
- * - using a configuration table, like below, which relies on the EFI init code
- * to locate the table and copy the contents;
- * - by linking directly to the core kernel's copy of the global symbol.
- *
- * The latter is preferred because it makes the EFIFB earlycon available very
- * early, but it only works if the EFI stub is part of the core kernel image
- * itself. The zboot decompressor can only use the configuration table
- * approach.
- */
-
-static efi_guid_t screen_info_guid = LINUX_EFI_SCREEN_INFO_TABLE_GUID;
-
-struct screen_info *__alloc_screen_info(void)
-{
- struct screen_info *si;
- efi_status_t status;
-
- status = efi_bs_call(allocate_pool, EFI_ACPI_RECLAIM_MEMORY,
- sizeof(*si), (void **)&si);
-
- if (status != EFI_SUCCESS)
- return NULL;
-
- memset(si, 0, sizeof(*si));
-
- status = efi_bs_call(install_configuration_table,
- &screen_info_guid, si);
- if (status == EFI_SUCCESS)
- return si;
-
- efi_bs_call(free_pool, si);
- return NULL;
-}
-
-void free_screen_info(struct screen_info *si)
-{
- if (!si)
- return;
-
- efi_bs_call(install_configuration_table, &screen_info_guid, NULL);
- efi_bs_call(free_pool, si);
-}
diff --git a/drivers/firmware/efi/libstub/zboot.c b/drivers/firmware/efi/libstub/zboot.c
index c47ace06f010..4b76f74c56da 100644
--- a/drivers/firmware/efi/libstub/zboot.c
+++ b/drivers/firmware/efi/libstub/zboot.c
@@ -26,9 +26,9 @@ void __weak efi_cache_sync_image(unsigned long image_base,
// executable code loaded into memory to be safe for execution.
}
-struct screen_info *alloc_screen_info(void)
+struct sysfb_display_info *alloc_primary_display(void)
{
- return __alloc_screen_info();
+ return __alloc_primary_display();
}
asmlinkage efi_status_t __efiapi
diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index d51777df12d1..f452fac90a9f 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -63,11 +63,13 @@ endif # HAS_IOMEM
config FIRMWARE_EDID
bool "Enable firmware EDID"
- depends on X86
+ depends on EFI || X86
help
This enables access to the EDID transferred from the firmware.
- On x86, this is from the VESA BIOS. DRM display drivers will
- be able to export the information to userspace.
+ On EFI systems, the EDID comes from the same device as the
+ primary GOP. On x86 with BIOS, it comes from the VESA BIOS.
+ DRM display drivers will be able to export the information
+ to userspace.
Also enable this if DDC/I2C transfers do not work for your driver
and if you are using nvidiafb, i810fb or savagefb.
diff --git a/include/linux/efi.h b/include/linux/efi.h
index 2a43094e23f7..664898d09ff5 100644
--- a/include/linux/efi.h
+++ b/include/linux/efi.h
@@ -406,11 +406,12 @@ void efi_native_runtime_setup(void);
#define EFI_CC_FINAL_EVENTS_TABLE_GUID EFI_GUID(0xdd4a4648, 0x2de7, 0x4665, 0x96, 0x4d, 0x21, 0xd9, 0xef, 0x5f, 0xb4, 0x46)
/*
- * This GUID is used to pass to the kernel proper the struct screen_info
- * structure that was populated by the stub based on the GOP protocol instance
- * associated with ConOut
+ * This GUIDs are used to pass to the kernel proper the primary
+ * display that has been populated by the stub based on the GOP
+ * instance associated with ConOut.
*/
-#define LINUX_EFI_SCREEN_INFO_TABLE_GUID EFI_GUID(0xe03fc20a, 0x85dc, 0x406e, 0xb9, 0x0e, 0x4a, 0xb5, 0x02, 0x37, 0x1d, 0x95)
+#define LINUX_EFI_PRIMARY_DISPLAY_TABLE_GUID EFI_GUID(0xe03fc20a, 0x85dc, 0x406e, 0xb9, 0x0e, 0x4a, 0xb5, 0x02, 0x37, 0x1d, 0x95)
+
#define LINUX_EFI_ARM_CPU_STATE_TABLE_GUID EFI_GUID(0xef79e4aa, 0x3c3d, 0x4989, 0xb9, 0x02, 0x07, 0xa9, 0x43, 0xe5, 0x50, 0xd2)
#define LINUX_EFI_LOADER_ENTRY_GUID EFI_GUID(0x4a67b082, 0x0a4c, 0x41cf, 0xb6, 0xc7, 0x44, 0x0b, 0x29, 0xbb, 0x8c, 0x4f)
#define LINUX_EFI_RANDOM_SEED_TABLE_GUID EFI_GUID(0x1ce1e5bc, 0x7ceb, 0x42f2, 0x81, 0xe5, 0x8a, 0xad, 0xf1, 0x80, 0xf5, 0x7b)
--
2.51.1
^ permalink raw reply related
* [PATCH v3 7/9] efi: Refactor init_primary_display() helpers
From: Thomas Zimmermann @ 2025-11-26 16:03 UTC (permalink / raw)
To: ardb, javierm, arnd, richard.lyu, helgaas
Cc: x86, linux-arm-kernel, linux-kernel, linux-efi, loongarch,
linux-riscv, dri-devel, linux-hyperv, linux-pci, linux-fbdev,
Thomas Zimmermann
In-Reply-To: <20251126160854.553077-1-tzimmermann@suse.de>
Rework the kernel's init_primary_display() helpers to allow for later
support of additional config-table entries and EDID information. No
functional changes.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
---
arch/loongarch/kernel/efi.c | 22 +++++++++++-----------
drivers/firmware/efi/efi-init.c | 19 ++++++++++---------
2 files changed, 21 insertions(+), 20 deletions(-)
diff --git a/arch/loongarch/kernel/efi.c b/arch/loongarch/kernel/efi.c
index 638a392d2cd2..1ef38036e8ae 100644
--- a/arch/loongarch/kernel/efi.c
+++ b/arch/loongarch/kernel/efi.c
@@ -81,19 +81,19 @@ EXPORT_SYMBOL_GPL(sysfb_primary_display);
static void __init init_primary_display(void)
{
- struct screen_info *si;
-
- if (screen_info_table == EFI_INVALID_TABLE_ADDR)
- return;
-
- si = early_memremap(screen_info_table, sizeof(*si));
- if (!si) {
- pr_err("Could not map screen_info config table\n");
+ if (screen_info_table == EFI_INVALID_TABLE_ADDR) {
+ struct screen_info *si = early_memremap(screen_info_table, sizeof(*si));
+
+ if (!si) {
+ pr_err("Could not map screen_info config table\n");
+ return;
+ }
+ sysfb_primary_display.screen = *si;
+ memset(si, 0, sizeof(*si));
+ early_memunmap(si, sizeof(*si));
+ } else {
return;
}
- sysfb_primary_display.screen = *si;
- memset(si, 0, sizeof(*si));
- early_memunmap(si, sizeof(*si));
memblock_reserve(__screen_info_lfb_base(&sysfb_primary_display.screen),
sysfb_primary_display.screen.lfb_size);
diff --git a/drivers/firmware/efi/efi-init.c b/drivers/firmware/efi/efi-init.c
index d1d418a34407..ca697d485116 100644
--- a/drivers/firmware/efi/efi-init.c
+++ b/drivers/firmware/efi/efi-init.c
@@ -67,10 +67,9 @@ EXPORT_SYMBOL_GPL(sysfb_primary_display);
static void __init init_primary_display(void)
{
- struct screen_info *si;
-
if (screen_info_table != EFI_INVALID_TABLE_ADDR) {
- si = early_memremap(screen_info_table, sizeof(*si));
+ struct screen_info *si = early_memremap(screen_info_table, sizeof(*si));
+
if (!si) {
pr_err("Could not map screen_info config table\n");
return;
@@ -78,14 +77,16 @@ static void __init init_primary_display(void)
sysfb_primary_display.screen = *si;
memset(si, 0, sizeof(*si));
early_memunmap(si, sizeof(*si));
+ } else {
+ return;
+ }
- if (memblock_is_map_memory(sysfb_primary_display.screen.lfb_base))
- memblock_mark_nomap(sysfb_primary_display.screen.lfb_base,
- sysfb_primary_display.screen.lfb_size);
+ if (memblock_is_map_memory(sysfb_primary_display.screen.lfb_base))
+ memblock_mark_nomap(sysfb_primary_display.screen.lfb_base,
+ sysfb_primary_display.screen.lfb_size);
- if (IS_ENABLED(CONFIG_EFI_EARLYCON))
- efi_earlycon_reprobe();
- }
+ if (IS_ENABLED(CONFIG_EFI_EARLYCON))
+ efi_earlycon_reprobe();
}
static int __init uefi_init(u64 efi_system_table)
--
2.51.1
^ permalink raw reply related
* [PATCH v3 3/9] sysfb: Add struct sysfb_display_info
From: Thomas Zimmermann @ 2025-11-26 16:03 UTC (permalink / raw)
To: ardb, javierm, arnd, richard.lyu, helgaas
Cc: x86, linux-arm-kernel, linux-kernel, linux-efi, loongarch,
linux-riscv, dri-devel, linux-hyperv, linux-pci, linux-fbdev,
Thomas Zimmermann
In-Reply-To: <20251126160854.553077-1-tzimmermann@suse.de>
Add struct sysfb_display_info to wrap display-related state. For now
it contains only the screen's video mode. Later EDID will be added as
well.
This struct will be helpful for passing display state to sysfb drivers
or from the EFI stub library.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
---
include/linux/sysfb.h | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/include/linux/sysfb.h b/include/linux/sysfb.h
index 8527a50a5290..8b37247528bf 100644
--- a/include/linux/sysfb.h
+++ b/include/linux/sysfb.h
@@ -8,6 +8,7 @@
*/
#include <linux/err.h>
+#include <linux/screen_info.h>
#include <linux/types.h>
#include <linux/platform_data/simplefb.h>
@@ -60,6 +61,10 @@ struct efifb_dmi_info {
int flags;
};
+struct sysfb_display_info {
+ struct screen_info screen;
+};
+
#ifdef CONFIG_SYSFB
void sysfb_disable(struct device *dev);
--
2.51.1
^ permalink raw reply related
* [PATCH v3 6/9] sysfb: Move edid_info into sysfb_primary_display
From: Thomas Zimmermann @ 2025-11-26 16:03 UTC (permalink / raw)
To: ardb, javierm, arnd, richard.lyu, helgaas
Cc: x86, linux-arm-kernel, linux-kernel, linux-efi, loongarch,
linux-riscv, dri-devel, linux-hyperv, linux-pci, linux-fbdev,
Thomas Zimmermann
In-Reply-To: <20251126160854.553077-1-tzimmermann@suse.de>
Move x86's edid_info into sysfb_primary_display as a new field named
edid. Adapt all users.
An instance of edid_info has only been defined on x86. With the move
into sysfb_primary_display, it becomes available on all architectures.
Therefore remove this contraint from CONFIG_FIRMWARE_EDID.
x86 fills the EDID data from boot_params.edid_info. DRM drivers pick
up the raw data and make it available to DRM clients. Replace the
drivers' references to edid_info and instead use the sysfb_display_info
as passed from sysfb.
v2:
- drop changes to CONFIG_FIRMWARE_EDID
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
---
arch/x86/kernel/setup.c | 6 +-----
drivers/gpu/drm/sysfb/efidrm.c | 5 ++---
drivers/gpu/drm/sysfb/vesadrm.c | 5 ++---
drivers/video/fbdev/core/fbmon.c | 8 +++++---
include/linux/sysfb.h | 6 ++++++
include/video/edid.h | 4 ----
6 files changed, 16 insertions(+), 18 deletions(-)
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 675e4b9deb1f..d9bfe2032cd9 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -215,10 +215,6 @@ arch_initcall(init_x86_sysctl);
struct sysfb_display_info sysfb_primary_display;
EXPORT_SYMBOL(sysfb_primary_display);
-#if defined(CONFIG_FIRMWARE_EDID)
-struct edid_info edid_info;
-EXPORT_SYMBOL_GPL(edid_info);
-#endif
extern int root_mountflags;
@@ -530,7 +526,7 @@ static void __init parse_boot_params(void)
ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev);
sysfb_primary_display.screen = boot_params.screen_info;
#if defined(CONFIG_FIRMWARE_EDID)
- edid_info = boot_params.edid_info;
+ sysfb_primary_display.edid = boot_params.edid_info;
#endif
#ifdef CONFIG_X86_32
apm_info.bios = boot_params.apm_bios_info;
diff --git a/drivers/gpu/drm/sysfb/efidrm.c b/drivers/gpu/drm/sysfb/efidrm.c
index 29533ae8fbbf..50e0aeef709c 100644
--- a/drivers/gpu/drm/sysfb/efidrm.c
+++ b/drivers/gpu/drm/sysfb/efidrm.c
@@ -24,7 +24,6 @@
#include <drm/drm_print.h>
#include <drm/drm_probe_helper.h>
-#include <video/edid.h>
#include <video/pixel_format.h>
#include "drm_sysfb_helper.h"
@@ -207,8 +206,8 @@ static struct efidrm_device *efidrm_device_create(struct drm_driver *drv,
&format->format, width, height, stride);
#if defined(CONFIG_FIRMWARE_EDID)
- if (drm_edid_header_is_valid(edid_info.dummy) == 8)
- sysfb->edid = edid_info.dummy;
+ if (drm_edid_header_is_valid(dpy->edid.dummy) == 8)
+ sysfb->edid = dpy->edid.dummy;
#endif
sysfb->fb_mode = drm_sysfb_mode(width, height, 0, 0);
sysfb->fb_format = format;
diff --git a/drivers/gpu/drm/sysfb/vesadrm.c b/drivers/gpu/drm/sysfb/vesadrm.c
index 16fc223f8c5b..0680638b8131 100644
--- a/drivers/gpu/drm/sysfb/vesadrm.c
+++ b/drivers/gpu/drm/sysfb/vesadrm.c
@@ -25,7 +25,6 @@
#include <drm/drm_print.h>
#include <drm/drm_probe_helper.h>
-#include <video/edid.h>
#include <video/pixel_format.h>
#include <video/vga.h>
@@ -474,8 +473,8 @@ static struct vesadrm_device *vesadrm_device_create(struct drm_driver *drv,
}
#if defined(CONFIG_FIRMWARE_EDID)
- if (drm_edid_header_is_valid(edid_info.dummy) == 8)
- sysfb->edid = edid_info.dummy;
+ if (drm_edid_header_is_valid(dpy->edid.dummy) == 8)
+ sysfb->edid = dpy->edid.dummy;
#endif
sysfb->fb_mode = drm_sysfb_mode(width, height, 0, 0);
sysfb->fb_format = format;
diff --git a/drivers/video/fbdev/core/fbmon.c b/drivers/video/fbdev/core/fbmon.c
index 0a65bef01e3c..07df7e98f8a3 100644
--- a/drivers/video/fbdev/core/fbmon.c
+++ b/drivers/video/fbdev/core/fbmon.c
@@ -32,11 +32,13 @@
#include <linux/module.h>
#include <linux/pci.h>
#include <linux/slab.h>
-#include <video/edid.h>
+#include <linux/string_choices.h>
+#include <linux/sysfb.h>
+
#include <video/of_videomode.h>
#include <video/videomode.h>
+
#include "../edid.h"
-#include <linux/string_choices.h>
/*
* EDID parser
@@ -1504,7 +1506,7 @@ const unsigned char *fb_firmware_edid(struct device *device)
res = &dev->resource[PCI_ROM_RESOURCE];
if (res && res->flags & IORESOURCE_ROM_SHADOW)
- edid = edid_info.dummy;
+ edid = sysfb_primary_display.edid.dummy;
return edid;
}
diff --git a/include/linux/sysfb.h b/include/linux/sysfb.h
index e8bde392c690..5226efde9ad4 100644
--- a/include/linux/sysfb.h
+++ b/include/linux/sysfb.h
@@ -12,6 +12,8 @@
#include <linux/screen_info.h>
#include <linux/types.h>
+#include <video/edid.h>
+
struct device;
struct platform_device;
struct screen_info;
@@ -62,6 +64,10 @@ struct efifb_dmi_info {
struct sysfb_display_info {
struct screen_info screen;
+
+#if defined(CONFIG_FIRMWARE_EDID)
+ struct edid_info edid;
+#endif
};
extern struct sysfb_display_info sysfb_primary_display;
diff --git a/include/video/edid.h b/include/video/edid.h
index c2b186b1933a..52aabb706032 100644
--- a/include/video/edid.h
+++ b/include/video/edid.h
@@ -4,8 +4,4 @@
#include <uapi/video/edid.h>
-#if defined(CONFIG_FIRMWARE_EDID)
-extern struct edid_info edid_info;
-#endif
-
#endif /* __linux_video_edid_h__ */
--
2.51.1
^ permalink raw reply related
* [PATCH v3 2/9] efi: sysfb_efi: Reduce number of references to global screen_info
From: Thomas Zimmermann @ 2025-11-26 16:03 UTC (permalink / raw)
To: ardb, javierm, arnd, richard.lyu, helgaas
Cc: x86, linux-arm-kernel, linux-kernel, linux-efi, loongarch,
linux-riscv, dri-devel, linux-hyperv, linux-pci, linux-fbdev,
Thomas Zimmermann
In-Reply-To: <20251126160854.553077-1-tzimmermann@suse.de>
Replace usage of global screen_info with local pointers. This will
later reduce churn when screen_info is being moved.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Richard Lyu <richard.lyu@suse.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
---
drivers/firmware/efi/sysfb_efi.c | 81 ++++++++++++++++---------------
drivers/firmware/sysfb.c | 4 +-
drivers/firmware/sysfb_simplefb.c | 2 +-
include/linux/sysfb.h | 9 ++--
4 files changed, 51 insertions(+), 45 deletions(-)
diff --git a/drivers/firmware/efi/sysfb_efi.c b/drivers/firmware/efi/sysfb_efi.c
index 1e509595ac03..8e0f9d08397f 100644
--- a/drivers/firmware/efi/sysfb_efi.c
+++ b/drivers/firmware/efi/sysfb_efi.c
@@ -92,7 +92,7 @@ void efifb_setup_from_dmi(struct screen_info *si, const char *opt)
})
#ifdef CONFIG_EFI
-static int __init efifb_set_system(const struct dmi_system_id *id)
+static int __init efifb_set_system(struct screen_info *si, const struct dmi_system_id *id)
{
struct efifb_dmi_info *info = id->driver_data;
@@ -101,14 +101,14 @@ static int __init efifb_set_system(const struct dmi_system_id *id)
return 0;
/* Trust the bootloader over the DMI tables */
- if (screen_info.lfb_base == 0) {
+ if (si->lfb_base == 0) {
#if defined(CONFIG_PCI)
struct pci_dev *dev = NULL;
int found_bar = 0;
#endif
if (info->base) {
- screen_info.lfb_base = choose_value(info->base,
- screen_info.lfb_base, OVERRIDE_BASE,
+ si->lfb_base = choose_value(info->base,
+ si->lfb_base, OVERRIDE_BASE,
info->flags);
#if defined(CONFIG_PCI)
@@ -135,49 +135,53 @@ static int __init efifb_set_system(const struct dmi_system_id *id)
start = pci_resource_start(dev, i);
end = pci_resource_end(dev, i);
- if (screen_info.lfb_base >= start &&
- screen_info.lfb_base < end) {
+ if (si->lfb_base >= start && si->lfb_base < end) {
found_bar = 1;
break;
}
}
}
if (!found_bar)
- screen_info.lfb_base = 0;
+ si->lfb_base = 0;
#endif
}
}
- if (screen_info.lfb_base) {
- screen_info.lfb_linelength = choose_value(info->stride,
- screen_info.lfb_linelength, OVERRIDE_STRIDE,
+ if (si->lfb_base) {
+ si->lfb_linelength = choose_value(info->stride,
+ si->lfb_linelength, OVERRIDE_STRIDE,
info->flags);
- screen_info.lfb_width = choose_value(info->width,
- screen_info.lfb_width, OVERRIDE_WIDTH,
+ si->lfb_width = choose_value(info->width,
+ si->lfb_width, OVERRIDE_WIDTH,
info->flags);
- screen_info.lfb_height = choose_value(info->height,
- screen_info.lfb_height, OVERRIDE_HEIGHT,
+ si->lfb_height = choose_value(info->height,
+ si->lfb_height, OVERRIDE_HEIGHT,
info->flags);
- if (screen_info.orig_video_isVGA == 0)
- screen_info.orig_video_isVGA = VIDEO_TYPE_EFI;
+ if (si->orig_video_isVGA == 0)
+ si->orig_video_isVGA = VIDEO_TYPE_EFI;
} else {
- screen_info.lfb_linelength = 0;
- screen_info.lfb_width = 0;
- screen_info.lfb_height = 0;
- screen_info.orig_video_isVGA = 0;
+ si->lfb_linelength = 0;
+ si->lfb_width = 0;
+ si->lfb_height = 0;
+ si->orig_video_isVGA = 0;
return 0;
}
printk(KERN_INFO "efifb: dmi detected %s - framebuffer at 0x%08x "
"(%dx%d, stride %d)\n", id->ident,
- screen_info.lfb_base, screen_info.lfb_width,
- screen_info.lfb_height, screen_info.lfb_linelength);
+ si->lfb_base, si->lfb_width,
+ si->lfb_height, si->lfb_linelength);
return 1;
}
+static int __init efifb_set_system_callback(const struct dmi_system_id *id)
+{
+ return efifb_set_system(&screen_info, id);
+}
+
#define EFIFB_DMI_SYSTEM_ID(vendor, name, enumid) \
{ \
- efifb_set_system, \
+ efifb_set_system_callback, \
name, \
{ \
DMI_MATCH(DMI_BIOS_VENDOR, vendor), \
@@ -284,12 +288,13 @@ static const struct dmi_system_id efifb_dmi_swap_width_height[] __initconst = {
{},
};
-static bool efifb_overlaps_pci_range(const struct of_pci_range *range)
+static bool efifb_overlaps_pci_range(const struct screen_info *si,
+ const struct of_pci_range *range)
{
- u64 fb_base = screen_info.lfb_base;
+ u64 fb_base = si->lfb_base;
- if (screen_info.capabilities & VIDEO_CAPABILITY_64BIT_BASE)
- fb_base |= (u64)(unsigned long)screen_info.ext_lfb_base << 32;
+ if (si->capabilities & VIDEO_CAPABILITY_64BIT_BASE)
+ fb_base |= (u64)(unsigned long)si->ext_lfb_base << 32;
return fb_base >= range->cpu_addr &&
fb_base < (range->cpu_addr + range->size);
@@ -311,7 +316,7 @@ static struct device_node *find_pci_overlap_node(void)
}
for_each_of_pci_range(&parser, &range)
- if (efifb_overlaps_pci_range(&range))
+ if (efifb_overlaps_pci_range(&screen_info, &range))
return np;
}
return NULL;
@@ -349,25 +354,25 @@ static const struct fwnode_operations efifb_fwnode_ops = {
static struct fwnode_handle efifb_fwnode;
-__init void sysfb_apply_efi_quirks(void)
+__init void sysfb_apply_efi_quirks(struct screen_info *si)
{
- if (screen_info.orig_video_isVGA != VIDEO_TYPE_EFI ||
- !(screen_info.capabilities & VIDEO_CAPABILITY_SKIP_QUIRKS))
+ if (si->orig_video_isVGA != VIDEO_TYPE_EFI ||
+ !(si->capabilities & VIDEO_CAPABILITY_SKIP_QUIRKS))
dmi_check_system(efifb_dmi_system_table);
- if (screen_info.orig_video_isVGA == VIDEO_TYPE_EFI &&
+ if (si->orig_video_isVGA == VIDEO_TYPE_EFI &&
dmi_check_system(efifb_dmi_swap_width_height)) {
- u16 temp = screen_info.lfb_width;
+ u16 temp = si->lfb_width;
- screen_info.lfb_width = screen_info.lfb_height;
- screen_info.lfb_height = temp;
- screen_info.lfb_linelength = 4 * screen_info.lfb_width;
+ si->lfb_width = si->lfb_height;
+ si->lfb_height = temp;
+ si->lfb_linelength = 4 * si->lfb_width;
}
}
-__init void sysfb_set_efifb_fwnode(struct platform_device *pd)
+__init void sysfb_set_efifb_fwnode(const struct screen_info *si, struct platform_device *pd)
{
- if (screen_info.orig_video_isVGA == VIDEO_TYPE_EFI && IS_ENABLED(CONFIG_PCI)) {
+ if (si->orig_video_isVGA == VIDEO_TYPE_EFI && IS_ENABLED(CONFIG_PCI)) {
fwnode_init(&efifb_fwnode, &efifb_fwnode_ops);
pd->dev.fwnode = &efifb_fwnode;
}
diff --git a/drivers/firmware/sysfb.c b/drivers/firmware/sysfb.c
index 889e5b05c739..916b28538a29 100644
--- a/drivers/firmware/sysfb.c
+++ b/drivers/firmware/sysfb.c
@@ -155,7 +155,7 @@ static __init int sysfb_init(void)
if (disabled)
goto unlock_mutex;
- sysfb_apply_efi_quirks();
+ sysfb_apply_efi_quirks(si);
parent = sysfb_parent_dev(si);
if (IS_ERR(parent)) {
@@ -200,7 +200,7 @@ static __init int sysfb_init(void)
pd->dev.parent = parent;
- sysfb_set_efifb_fwnode(pd);
+ sysfb_set_efifb_fwnode(si, pd);
ret = platform_device_add_data(pd, si, sizeof(*si));
if (ret)
diff --git a/drivers/firmware/sysfb_simplefb.c b/drivers/firmware/sysfb_simplefb.c
index 592d8a644619..71f542e37732 100644
--- a/drivers/firmware/sysfb_simplefb.c
+++ b/drivers/firmware/sysfb_simplefb.c
@@ -117,7 +117,7 @@ __init struct platform_device *sysfb_create_simplefb(const struct screen_info *s
pd->dev.parent = parent;
- sysfb_set_efifb_fwnode(pd);
+ sysfb_set_efifb_fwnode(si, pd);
ret = platform_device_add_resources(pd, &res, 1);
if (ret)
diff --git a/include/linux/sysfb.h b/include/linux/sysfb.h
index b449665c686a..8527a50a5290 100644
--- a/include/linux/sysfb.h
+++ b/include/linux/sysfb.h
@@ -82,16 +82,17 @@ static inline bool sysfb_handles_screen_info(void)
#ifdef CONFIG_EFI
extern struct efifb_dmi_info efifb_dmi_list[];
-void sysfb_apply_efi_quirks(void);
-void sysfb_set_efifb_fwnode(struct platform_device *pd);
+void sysfb_apply_efi_quirks(struct screen_info *si);
+void sysfb_set_efifb_fwnode(const struct screen_info *si, struct platform_device *pd);
#else /* CONFIG_EFI */
-static inline void sysfb_apply_efi_quirks(void)
+static inline void sysfb_apply_efi_quirks(struct screen_info *si)
{
}
-static inline void sysfb_set_efifb_fwnode(struct platform_device *pd)
+static inline void sysfb_set_efifb_fwnode(const struct screen_info *si,
+ struct platform_device *pd)
{
}
--
2.51.1
^ permalink raw reply related
* [PATCH v3 5/9] sysfb: Pass sysfb_primary_display to devices
From: Thomas Zimmermann @ 2025-11-26 16:03 UTC (permalink / raw)
To: ardb, javierm, arnd, richard.lyu, helgaas
Cc: x86, linux-arm-kernel, linux-kernel, linux-efi, loongarch,
linux-riscv, dri-devel, linux-hyperv, linux-pci, linux-fbdev,
Thomas Zimmermann
In-Reply-To: <20251126160854.553077-1-tzimmermann@suse.de>
Instead of screen_info, store a copy of sysfb_primary_display as
device data. Pick it up in drivers. Later changes will add additional
data to the display info, such as EDID information.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
---
drivers/firmware/sysfb.c | 5 +++--
drivers/gpu/drm/sysfb/efidrm.c | 9 ++++++---
drivers/gpu/drm/sysfb/vesadrm.c | 9 ++++++---
drivers/video/fbdev/efifb.c | 10 ++++++----
drivers/video/fbdev/vesafb.c | 10 ++++++----
drivers/video/fbdev/vga16fb.c | 8 +++++---
6 files changed, 32 insertions(+), 19 deletions(-)
diff --git a/drivers/firmware/sysfb.c b/drivers/firmware/sysfb.c
index 1f671f9219b0..8833582c1883 100644
--- a/drivers/firmware/sysfb.c
+++ b/drivers/firmware/sysfb.c
@@ -141,7 +141,8 @@ static struct device *sysfb_parent_dev(const struct screen_info *si)
static __init int sysfb_init(void)
{
- struct screen_info *si = &sysfb_primary_display.screen;
+ struct sysfb_display_info *dpy = &sysfb_primary_display;
+ struct screen_info *si = &dpy->screen;
struct device *parent;
unsigned int type;
struct simplefb_platform_data mode;
@@ -202,7 +203,7 @@ static __init int sysfb_init(void)
sysfb_set_efifb_fwnode(si, pd);
- ret = platform_device_add_data(pd, si, sizeof(*si));
+ ret = platform_device_add_data(pd, dpy, sizeof(*dpy));
if (ret)
goto err;
diff --git a/drivers/gpu/drm/sysfb/efidrm.c b/drivers/gpu/drm/sysfb/efidrm.c
index 1b683d55d6ea..29533ae8fbbf 100644
--- a/drivers/gpu/drm/sysfb/efidrm.c
+++ b/drivers/gpu/drm/sysfb/efidrm.c
@@ -4,7 +4,7 @@
#include <linux/efi.h>
#include <linux/limits.h>
#include <linux/platform_device.h>
-#include <linux/screen_info.h>
+#include <linux/sysfb.h>
#include <drm/clients/drm_client_setup.h>
#include <drm/drm_atomic.h>
@@ -141,6 +141,7 @@ static const struct drm_mode_config_funcs efidrm_mode_config_funcs = {
static struct efidrm_device *efidrm_device_create(struct drm_driver *drv,
struct platform_device *pdev)
{
+ const struct sysfb_display_info *dpy;
const struct screen_info *si;
const struct drm_format_info *format;
int width, height, stride;
@@ -160,9 +161,11 @@ static struct efidrm_device *efidrm_device_create(struct drm_driver *drv,
size_t nformats;
int ret;
- si = dev_get_platdata(&pdev->dev);
- if (!si)
+ dpy = dev_get_platdata(&pdev->dev);
+ if (!dpy)
return ERR_PTR(-ENODEV);
+ si = &dpy->screen;
+
if (screen_info_video_type(si) != VIDEO_TYPE_EFI)
return ERR_PTR(-ENODEV);
diff --git a/drivers/gpu/drm/sysfb/vesadrm.c b/drivers/gpu/drm/sysfb/vesadrm.c
index 7b7b5ba26317..16fc223f8c5b 100644
--- a/drivers/gpu/drm/sysfb/vesadrm.c
+++ b/drivers/gpu/drm/sysfb/vesadrm.c
@@ -4,7 +4,7 @@
#include <linux/ioport.h>
#include <linux/limits.h>
#include <linux/platform_device.h>
-#include <linux/screen_info.h>
+#include <linux/sysfb.h>
#include <drm/clients/drm_client_setup.h>
#include <drm/drm_atomic.h>
@@ -391,6 +391,7 @@ static const struct drm_mode_config_funcs vesadrm_mode_config_funcs = {
static struct vesadrm_device *vesadrm_device_create(struct drm_driver *drv,
struct platform_device *pdev)
{
+ const struct sysfb_display_info *dpy;
const struct screen_info *si;
const struct drm_format_info *format;
int width, height, stride;
@@ -410,9 +411,11 @@ static struct vesadrm_device *vesadrm_device_create(struct drm_driver *drv,
size_t nformats;
int ret;
- si = dev_get_platdata(&pdev->dev);
- if (!si)
+ dpy = dev_get_platdata(&pdev->dev);
+ if (!dpy)
return ERR_PTR(-ENODEV);
+ si = &dpy->screen;
+
if (screen_info_video_type(si) != VIDEO_TYPE_VLFB)
return ERR_PTR(-ENODEV);
diff --git a/drivers/video/fbdev/efifb.c b/drivers/video/fbdev/efifb.c
index 0e1bd3dba255..47ebc0107209 100644
--- a/drivers/video/fbdev/efifb.c
+++ b/drivers/video/fbdev/efifb.c
@@ -15,7 +15,7 @@
#include <linux/fb.h>
#include <linux/platform_device.h>
#include <linux/printk.h>
-#include <linux/screen_info.h>
+#include <linux/sysfb.h>
#include <video/vga.h>
#include <asm/efi.h>
#include <drm/drm_utils.h> /* For drm_get_panel_orientation_quirk */
@@ -345,6 +345,7 @@ ATTRIBUTE_GROUPS(efifb);
static int efifb_probe(struct platform_device *dev)
{
+ struct sysfb_display_info *dpy;
struct screen_info *si;
struct fb_info *info;
struct efifb_par *par;
@@ -360,10 +361,11 @@ static int efifb_probe(struct platform_device *dev)
* driver. We get a copy of the attached screen_info, so that we can
* modify its values without affecting later drivers.
*/
- si = dev_get_platdata(&dev->dev);
- if (!si)
+ dpy = dev_get_platdata(&dev->dev);
+ if (!dpy)
return -ENODEV;
- si = devm_kmemdup(&dev->dev, si, sizeof(*si), GFP_KERNEL);
+
+ si = devm_kmemdup(&dev->dev, &dpy->screen, sizeof(*si), GFP_KERNEL);
if (!si)
return -ENOMEM;
diff --git a/drivers/video/fbdev/vesafb.c b/drivers/video/fbdev/vesafb.c
index f135033c22fb..f84f4db244bf 100644
--- a/drivers/video/fbdev/vesafb.c
+++ b/drivers/video/fbdev/vesafb.c
@@ -20,7 +20,7 @@
#include <linux/ioport.h>
#include <linux/init.h>
#include <linux/platform_device.h>
-#include <linux/screen_info.h>
+#include <linux/sysfb.h>
#include <linux/io.h>
#include <video/vga.h>
@@ -243,6 +243,7 @@ static int vesafb_setup(char *options)
static int vesafb_probe(struct platform_device *dev)
{
+ struct sysfb_display_info *dpy;
struct screen_info *si;
struct fb_info *info;
struct vesafb_par *par;
@@ -257,10 +258,11 @@ static int vesafb_probe(struct platform_device *dev)
* driver. We get a copy of the attached screen_info, so that we can
* modify its values without affecting later drivers.
*/
- si = dev_get_platdata(&dev->dev);
- if (!si)
+ dpy = dev_get_platdata(&dev->dev);
+ if (!dpy)
return -ENODEV;
- si = devm_kmemdup(&dev->dev, si, sizeof(*si), GFP_KERNEL);
+
+ si = devm_kmemdup(&dev->dev, &dpy->screen, sizeof(*si), GFP_KERNEL);
if (!si)
return -ENOMEM;
diff --git a/drivers/video/fbdev/vga16fb.c b/drivers/video/fbdev/vga16fb.c
index 6b81337a4909..22085d3668e8 100644
--- a/drivers/video/fbdev/vga16fb.c
+++ b/drivers/video/fbdev/vga16fb.c
@@ -21,7 +21,7 @@
#include <linux/ioport.h>
#include <linux/init.h>
#include <linux/platform_device.h>
-#include <linux/screen_info.h>
+#include <linux/sysfb.h>
#include <asm/io.h>
#include <video/vga.h>
@@ -1305,15 +1305,17 @@ static const struct fb_ops vga16fb_ops = {
static int vga16fb_probe(struct platform_device *dev)
{
+ struct sysfb_display_info *dpy;
struct screen_info *si;
struct fb_info *info;
struct vga16fb_par *par;
int i;
int ret = 0;
- si = dev_get_platdata(&dev->dev);
- if (!si)
+ dpy = dev_get_platdata(&dev->dev);
+ if (!dpy)
return -ENODEV;
+ si = &dpy->screen;
ret = check_mode_supported(si);
if (ret)
--
2.51.1
^ permalink raw reply related
* [PATCH v3 1/9] efi: earlycon: Reduce number of references to global screen_info
From: Thomas Zimmermann @ 2025-11-26 16:03 UTC (permalink / raw)
To: ardb, javierm, arnd, richard.lyu, helgaas
Cc: x86, linux-arm-kernel, linux-kernel, linux-efi, loongarch,
linux-riscv, dri-devel, linux-hyperv, linux-pci, linux-fbdev,
Thomas Zimmermann
In-Reply-To: <20251126160854.553077-1-tzimmermann@suse.de>
Replace usage of global screen_info with local pointers. This will
later reduce churn when screen_info is being moved.
v2:
- fix coding style
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Reviewed-by: Richard Lyu <richard.lyu@suse.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
---
drivers/firmware/efi/earlycon.c | 40 ++++++++++++++++-----------------
1 file changed, 20 insertions(+), 20 deletions(-)
diff --git a/drivers/firmware/efi/earlycon.c b/drivers/firmware/efi/earlycon.c
index d18a1a5de144..42e3a173dac1 100644
--- a/drivers/firmware/efi/earlycon.c
+++ b/drivers/firmware/efi/earlycon.c
@@ -32,12 +32,13 @@ static void *efi_fb;
*/
static int __init efi_earlycon_remap_fb(void)
{
+ const struct screen_info *si = &screen_info;
+
/* bail if there is no bootconsole or it was unregistered already */
if (!earlycon_console || !console_is_registered(earlycon_console))
return 0;
- efi_fb = memremap(fb_base, screen_info.lfb_size,
- fb_wb ? MEMREMAP_WB : MEMREMAP_WC);
+ efi_fb = memremap(fb_base, si->lfb_size, fb_wb ? MEMREMAP_WB : MEMREMAP_WC);
return efi_fb ? 0 : -ENOMEM;
}
@@ -71,12 +72,12 @@ static __ref void efi_earlycon_unmap(void *addr, unsigned long len)
early_memunmap(addr, len);
}
-static void efi_earlycon_clear_scanline(unsigned int y)
+static void efi_earlycon_clear_scanline(unsigned int y, const struct screen_info *si)
{
unsigned long *dst;
u16 len;
- len = screen_info.lfb_linelength;
+ len = si->lfb_linelength;
dst = efi_earlycon_map(y*len, len);
if (!dst)
return;
@@ -85,7 +86,7 @@ static void efi_earlycon_clear_scanline(unsigned int y)
efi_earlycon_unmap(dst, len);
}
-static void efi_earlycon_scroll_up(void)
+static void efi_earlycon_scroll_up(const struct screen_info *si)
{
unsigned long *dst, *src;
u16 maxlen = 0;
@@ -99,8 +100,8 @@ static void efi_earlycon_scroll_up(void)
}
maxlen *= 4;
- len = screen_info.lfb_linelength;
- height = screen_info.lfb_height;
+ len = si->lfb_linelength;
+ height = si->lfb_height;
for (i = 0; i < height - font->height; i++) {
dst = efi_earlycon_map(i*len, len);
@@ -120,7 +121,8 @@ static void efi_earlycon_scroll_up(void)
}
}
-static void efi_earlycon_write_char(u32 *dst, unsigned char c, unsigned int h)
+static void efi_earlycon_write_char(u32 *dst, unsigned char c, unsigned int h,
+ const struct screen_info *si)
{
const u32 color_black = 0x00000000;
const u32 color_white = 0x00ffffff;
@@ -145,13 +147,12 @@ static void efi_earlycon_write_char(u32 *dst, unsigned char c, unsigned int h)
static void
efi_earlycon_write(struct console *con, const char *str, unsigned int num)
{
- struct screen_info *si;
+ const struct screen_info *si = &screen_info;
u32 cur_efi_x = efi_x;
unsigned int len;
const char *s;
void *dst;
- si = &screen_info;
len = si->lfb_linelength;
while (num) {
@@ -174,7 +175,7 @@ efi_earlycon_write(struct console *con, const char *str, unsigned int num)
x = efi_x;
while (n-- > 0) {
- efi_earlycon_write_char(dst + x*4, *s, h);
+ efi_earlycon_write_char(dst + x * 4, *s, h, si);
x += font->width;
s++;
}
@@ -207,10 +208,10 @@ efi_earlycon_write(struct console *con, const char *str, unsigned int num)
cur_line_y = (cur_line_y + 1) % max_line_y;
efi_y -= font->height;
- efi_earlycon_scroll_up();
+ efi_earlycon_scroll_up(si);
for (i = 0; i < font->height; i++)
- efi_earlycon_clear_scanline(efi_y + i);
+ efi_earlycon_clear_scanline(efi_y + i, si);
}
}
}
@@ -226,22 +227,21 @@ void __init efi_earlycon_reprobe(void)
static int __init efi_earlycon_setup(struct earlycon_device *device,
const char *opt)
{
- struct screen_info *si;
+ const struct screen_info *si = &screen_info;
u16 xres, yres;
u32 i;
fb_wb = opt && !strcmp(opt, "ram");
- if (screen_info.orig_video_isVGA != VIDEO_TYPE_EFI) {
+ if (si->orig_video_isVGA != VIDEO_TYPE_EFI) {
fb_probed = true;
return -ENODEV;
}
- fb_base = screen_info.lfb_base;
- if (screen_info.capabilities & VIDEO_CAPABILITY_64BIT_BASE)
- fb_base |= (u64)screen_info.ext_lfb_base << 32;
+ fb_base = si->lfb_base;
+ if (si->capabilities & VIDEO_CAPABILITY_64BIT_BASE)
+ fb_base |= (u64)si->ext_lfb_base << 32;
- si = &screen_info;
xres = si->lfb_width;
yres = si->lfb_height;
@@ -266,7 +266,7 @@ static int __init efi_earlycon_setup(struct earlycon_device *device,
efi_y -= font->height;
for (i = 0; i < (yres - efi_y) / font->height; i++)
- efi_earlycon_scroll_up();
+ efi_earlycon_scroll_up(si);
device->con->write = efi_earlycon_write;
earlycon_console = device->con;
--
2.51.1
^ permalink raw reply related
* [PATCH v3 4/9] sysfb: Replace screen_info with sysfb_primary_display
From: Thomas Zimmermann @ 2025-11-26 16:03 UTC (permalink / raw)
To: ardb, javierm, arnd, richard.lyu, helgaas
Cc: x86, linux-arm-kernel, linux-kernel, linux-efi, loongarch,
linux-riscv, dri-devel, linux-hyperv, linux-pci, linux-fbdev,
Thomas Zimmermann, Bjorn Helgaas
In-Reply-To: <20251126160854.553077-1-tzimmermann@suse.de>
Replace the global screen_info with sysfb_primary_display of type
struct sysfb_display_info. Adapt all users of screen_info.
Instances of screen_info are defined for x86, loongarch and EFI,
with only one instance compiled into a specific build. Replace all
of them with sysfb_primary_display.
All existing users of screen_info are updated by pointing them to
sysfb_primary_display.screen instead. This introduces some churn to
the code, but has no impact on functionality.
Boot parameters and EFI config tables are unchanged. They transfer
screen_info as before. The logic in EFI's alloc_screen_info() changes
slightly, as it now returns the screen field of sysfb_primary_display.
v2:
- update comment
- rename init_screen_info() to init_primary_display()
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
Acked-by: Bjorn Helgaas <bhelgaas@google.com> # drivers/pci/
---
arch/arm64/kernel/image-vars.h | 2 +-
arch/loongarch/kernel/efi.c | 15 +++++++------
arch/loongarch/kernel/image-vars.h | 2 +-
arch/riscv/kernel/image-vars.h | 2 +-
arch/x86/kernel/kexec-bzimage64.c | 4 +++-
arch/x86/kernel/setup.c | 10 +++++----
arch/x86/video/video-common.c | 4 ++--
drivers/firmware/efi/earlycon.c | 8 +++----
drivers/firmware/efi/efi-init.c | 22 +++++++++----------
drivers/firmware/efi/libstub/efi-stub-entry.c | 18 ++++++++++-----
drivers/firmware/efi/sysfb_efi.c | 4 ++--
drivers/firmware/sysfb.c | 6 ++---
drivers/hv/vmbus_drv.c | 6 ++---
drivers/pci/vgaarb.c | 4 ++--
drivers/video/screen_info_pci.c | 5 +++--
include/linux/screen_info.h | 2 --
include/linux/sysfb.h | 5 +++--
17 files changed, 66 insertions(+), 53 deletions(-)
diff --git a/arch/arm64/kernel/image-vars.h b/arch/arm64/kernel/image-vars.h
index 85bc629270bd..d7b0d12b1015 100644
--- a/arch/arm64/kernel/image-vars.h
+++ b/arch/arm64/kernel/image-vars.h
@@ -38,7 +38,7 @@ PROVIDE(__efistub__end = _end);
PROVIDE(__efistub___inittext_end = __inittext_end);
PROVIDE(__efistub__edata = _edata);
#if defined(CONFIG_EFI_EARLYCON) || defined(CONFIG_SYSFB)
-PROVIDE(__efistub_screen_info = screen_info);
+PROVIDE(__efistub_sysfb_primary_display = sysfb_primary_display);
#endif
PROVIDE(__efistub__ctype = _ctype);
diff --git a/arch/loongarch/kernel/efi.c b/arch/loongarch/kernel/efi.c
index 860a3bc030e0..638a392d2cd2 100644
--- a/arch/loongarch/kernel/efi.c
+++ b/arch/loongarch/kernel/efi.c
@@ -18,7 +18,7 @@
#include <linux/kobject.h>
#include <linux/memblock.h>
#include <linux/reboot.h>
-#include <linux/screen_info.h>
+#include <linux/sysfb.h>
#include <linux/uaccess.h>
#include <asm/early_ioremap.h>
@@ -75,11 +75,11 @@ bool efi_poweroff_required(void)
unsigned long __initdata screen_info_table = EFI_INVALID_TABLE_ADDR;
#if defined(CONFIG_SYSFB) || defined(CONFIG_EFI_EARLYCON)
-struct screen_info screen_info __section(".data");
-EXPORT_SYMBOL_GPL(screen_info);
+struct sysfb_display_info sysfb_primary_display __section(".data");
+EXPORT_SYMBOL_GPL(sysfb_primary_display);
#endif
-static void __init init_screen_info(void)
+static void __init init_primary_display(void)
{
struct screen_info *si;
@@ -91,11 +91,12 @@ static void __init init_screen_info(void)
pr_err("Could not map screen_info config table\n");
return;
}
- screen_info = *si;
+ sysfb_primary_display.screen = *si;
memset(si, 0, sizeof(*si));
early_memunmap(si, sizeof(*si));
- memblock_reserve(__screen_info_lfb_base(&screen_info), screen_info.lfb_size);
+ memblock_reserve(__screen_info_lfb_base(&sysfb_primary_display.screen),
+ sysfb_primary_display.screen.lfb_size);
}
void __init efi_init(void)
@@ -127,7 +128,7 @@ void __init efi_init(void)
set_bit(EFI_CONFIG_TABLES, &efi.flags);
if (IS_ENABLED(CONFIG_EFI_EARLYCON) || IS_ENABLED(CONFIG_SYSFB))
- init_screen_info();
+ init_primary_display();
if (boot_memmap == EFI_INVALID_TABLE_ADDR)
return;
diff --git a/arch/loongarch/kernel/image-vars.h b/arch/loongarch/kernel/image-vars.h
index 41ddcf56d21c..e557ebd46c2b 100644
--- a/arch/loongarch/kernel/image-vars.h
+++ b/arch/loongarch/kernel/image-vars.h
@@ -12,7 +12,7 @@ __efistub_kernel_entry = kernel_entry;
__efistub_kernel_asize = kernel_asize;
__efistub_kernel_fsize = kernel_fsize;
#if defined(CONFIG_EFI_EARLYCON) || defined(CONFIG_SYSFB)
-__efistub_screen_info = screen_info;
+__efistub_sysfb_primary_display = sysfb_primary_display;
#endif
#endif
diff --git a/arch/riscv/kernel/image-vars.h b/arch/riscv/kernel/image-vars.h
index 3df30dd1c458..3bd9d06a8b8f 100644
--- a/arch/riscv/kernel/image-vars.h
+++ b/arch/riscv/kernel/image-vars.h
@@ -29,7 +29,7 @@ __efistub__end = _end;
__efistub__edata = _edata;
__efistub___init_text_end = __init_text_end;
#if defined(CONFIG_EFI_EARLYCON) || defined(CONFIG_SYSFB)
-__efistub_screen_info = screen_info;
+__efistub_sysfb_primary_display = sysfb_primary_display;
#endif
#endif
diff --git a/arch/x86/kernel/kexec-bzimage64.c b/arch/x86/kernel/kexec-bzimage64.c
index c3244ac680d1..7508d0ccc740 100644
--- a/arch/x86/kernel/kexec-bzimage64.c
+++ b/arch/x86/kernel/kexec-bzimage64.c
@@ -20,6 +20,7 @@
#include <linux/of_fdt.h>
#include <linux/efi.h>
#include <linux/random.h>
+#include <linux/sysfb.h>
#include <asm/bootparam.h>
#include <asm/setup.h>
@@ -303,7 +304,8 @@ setup_boot_parameters(struct kimage *image, struct boot_params *params,
params->hdr.hardware_subarch = boot_params.hdr.hardware_subarch;
/* Copying screen_info will do? */
- memcpy(¶ms->screen_info, &screen_info, sizeof(struct screen_info));
+ memcpy(¶ms->screen_info, &sysfb_primary_display.screen,
+ sizeof(sysfb_primary_display.screen));
/* Fill in memsize later */
params->screen_info.ext_mem_k = 0;
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index 1b2edd07a3e1..675e4b9deb1f 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -22,6 +22,7 @@
#include <linux/random.h>
#include <linux/root_dev.h>
#include <linux/static_call.h>
+#include <linux/sysfb.h>
#include <linux/swiotlb.h>
#include <linux/tboot.h>
#include <linux/usb/xhci-dbgp.h>
@@ -211,8 +212,9 @@ arch_initcall(init_x86_sysctl);
/*
* Setup options
*/
-struct screen_info screen_info;
-EXPORT_SYMBOL(screen_info);
+
+struct sysfb_display_info sysfb_primary_display;
+EXPORT_SYMBOL(sysfb_primary_display);
#if defined(CONFIG_FIRMWARE_EDID)
struct edid_info edid_info;
EXPORT_SYMBOL_GPL(edid_info);
@@ -526,7 +528,7 @@ static void __init parse_setup_data(void)
static void __init parse_boot_params(void)
{
ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev);
- screen_info = boot_params.screen_info;
+ sysfb_primary_display.screen = boot_params.screen_info;
#if defined(CONFIG_FIRMWARE_EDID)
edid_info = boot_params.edid_info;
#endif
@@ -1254,7 +1256,7 @@ void __init setup_arch(char **cmdline_p)
#ifdef CONFIG_VT
#if defined(CONFIG_VGA_CONSOLE)
if (!efi_enabled(EFI_BOOT) || (efi_mem_type(0xa0000) != EFI_CONVENTIONAL_MEMORY))
- vgacon_register_screen(&screen_info);
+ vgacon_register_screen(&sysfb_primary_display.screen);
#endif
#endif
x86_init.oem.banner();
diff --git a/arch/x86/video/video-common.c b/arch/x86/video/video-common.c
index e0aeee99bc99..152789f00fcd 100644
--- a/arch/x86/video/video-common.c
+++ b/arch/x86/video/video-common.c
@@ -9,7 +9,7 @@
#include <linux/module.h>
#include <linux/pci.h>
-#include <linux/screen_info.h>
+#include <linux/sysfb.h>
#include <linux/vgaarb.h>
#include <asm/video.h>
@@ -29,7 +29,7 @@ EXPORT_SYMBOL(pgprot_framebuffer);
bool video_is_primary_device(struct device *dev)
{
#ifdef CONFIG_SCREEN_INFO
- struct screen_info *si = &screen_info;
+ struct screen_info *si = &sysfb_primary_display.screen;
struct resource res[SCREEN_INFO_MAX_RESOURCES];
ssize_t i, numres;
#endif
diff --git a/drivers/firmware/efi/earlycon.c b/drivers/firmware/efi/earlycon.c
index 42e3a173dac1..3d060d59968c 100644
--- a/drivers/firmware/efi/earlycon.c
+++ b/drivers/firmware/efi/earlycon.c
@@ -9,7 +9,7 @@
#include <linux/io.h>
#include <linux/kernel.h>
#include <linux/serial_core.h>
-#include <linux/screen_info.h>
+#include <linux/sysfb.h>
#include <linux/string.h>
#include <asm/early_ioremap.h>
@@ -32,7 +32,7 @@ static void *efi_fb;
*/
static int __init efi_earlycon_remap_fb(void)
{
- const struct screen_info *si = &screen_info;
+ const struct screen_info *si = &sysfb_primary_display.screen;
/* bail if there is no bootconsole or it was unregistered already */
if (!earlycon_console || !console_is_registered(earlycon_console))
@@ -147,7 +147,7 @@ static void efi_earlycon_write_char(u32 *dst, unsigned char c, unsigned int h,
static void
efi_earlycon_write(struct console *con, const char *str, unsigned int num)
{
- const struct screen_info *si = &screen_info;
+ const struct screen_info *si = &sysfb_primary_display.screen;
u32 cur_efi_x = efi_x;
unsigned int len;
const char *s;
@@ -227,7 +227,7 @@ void __init efi_earlycon_reprobe(void)
static int __init efi_earlycon_setup(struct earlycon_device *device,
const char *opt)
{
- const struct screen_info *si = &screen_info;
+ const struct screen_info *si = &sysfb_primary_display.screen;
u16 xres, yres;
u32 i;
diff --git a/drivers/firmware/efi/efi-init.c b/drivers/firmware/efi/efi-init.c
index a65c2d5b9e7b..d1d418a34407 100644
--- a/drivers/firmware/efi/efi-init.c
+++ b/drivers/firmware/efi/efi-init.c
@@ -19,7 +19,7 @@
#include <linux/of_address.h>
#include <linux/of_fdt.h>
#include <linux/platform_device.h>
-#include <linux/screen_info.h>
+#include <linux/sysfb.h>
#include <asm/efi.h>
@@ -57,15 +57,15 @@ static phys_addr_t __init efi_to_phys(unsigned long addr)
extern __weak const efi_config_table_type_t efi_arch_tables[];
/*
- * x86 defines its own screen_info and uses it even without EFI,
- * everything else can get it from here.
+ * x86 defines its own instance of sysfb_primary_display and uses
+ * it even without EFI, everything else can get them from here.
*/
#if !defined(CONFIG_X86) && (defined(CONFIG_SYSFB) || defined(CONFIG_EFI_EARLYCON))
-struct screen_info screen_info __section(".data");
-EXPORT_SYMBOL_GPL(screen_info);
+struct sysfb_display_info sysfb_primary_display __section(".data");
+EXPORT_SYMBOL_GPL(sysfb_primary_display);
#endif
-static void __init init_screen_info(void)
+static void __init init_primary_display(void)
{
struct screen_info *si;
@@ -75,13 +75,13 @@ static void __init init_screen_info(void)
pr_err("Could not map screen_info config table\n");
return;
}
- screen_info = *si;
+ sysfb_primary_display.screen = *si;
memset(si, 0, sizeof(*si));
early_memunmap(si, sizeof(*si));
- if (memblock_is_map_memory(screen_info.lfb_base))
- memblock_mark_nomap(screen_info.lfb_base,
- screen_info.lfb_size);
+ if (memblock_is_map_memory(sysfb_primary_display.screen.lfb_base))
+ memblock_mark_nomap(sysfb_primary_display.screen.lfb_base,
+ sysfb_primary_display.screen.lfb_size);
if (IS_ENABLED(CONFIG_EFI_EARLYCON))
efi_earlycon_reprobe();
@@ -274,5 +274,5 @@ void __init efi_init(void)
if (IS_ENABLED(CONFIG_X86) ||
IS_ENABLED(CONFIG_SYSFB) ||
IS_ENABLED(CONFIG_EFI_EARLYCON))
- init_screen_info();
+ init_primary_display();
}
diff --git a/drivers/firmware/efi/libstub/efi-stub-entry.c b/drivers/firmware/efi/libstub/efi-stub-entry.c
index a6c049835190..401ecbbdf331 100644
--- a/drivers/firmware/efi/libstub/efi-stub-entry.c
+++ b/drivers/firmware/efi/libstub/efi-stub-entry.c
@@ -1,13 +1,18 @@
// SPDX-License-Identifier: GPL-2.0-only
#include <linux/efi.h>
-#include <linux/screen_info.h>
+#include <linux/sysfb.h>
#include <asm/efi.h>
#include "efistub.h"
-static unsigned long screen_info_offset;
+static unsigned long kernel_image_offset;
+
+static void *kernel_image_addr(void *addr)
+{
+ return addr + kernel_image_offset;
+}
struct screen_info *alloc_screen_info(void)
{
@@ -16,8 +21,11 @@ struct screen_info *alloc_screen_info(void)
if (IS_ENABLED(CONFIG_X86) ||
IS_ENABLED(CONFIG_EFI_EARLYCON) ||
- IS_ENABLED(CONFIG_SYSFB))
- return (void *)&screen_info + screen_info_offset;
+ IS_ENABLED(CONFIG_SYSFB)) {
+ struct sysfb_display_info *dpy = kernel_image_addr(&sysfb_primary_display);
+
+ return &dpy->screen;
+ }
return NULL;
}
@@ -73,7 +81,7 @@ efi_status_t __efiapi efi_pe_entry(efi_handle_t handle,
return status;
}
- screen_info_offset = image_addr - (unsigned long)image->image_base;
+ kernel_image_offset = image_addr - (unsigned long)image->image_base;
status = efi_stub_common(handle, image, image_addr, cmdline_ptr);
diff --git a/drivers/firmware/efi/sysfb_efi.c b/drivers/firmware/efi/sysfb_efi.c
index 8e0f9d08397f..46ad95084b50 100644
--- a/drivers/firmware/efi/sysfb_efi.c
+++ b/drivers/firmware/efi/sysfb_efi.c
@@ -176,7 +176,7 @@ static int __init efifb_set_system(struct screen_info *si, const struct dmi_syst
static int __init efifb_set_system_callback(const struct dmi_system_id *id)
{
- return efifb_set_system(&screen_info, id);
+ return efifb_set_system(&sysfb_primary_display.screen, id);
}
#define EFIFB_DMI_SYSTEM_ID(vendor, name, enumid) \
@@ -316,7 +316,7 @@ static struct device_node *find_pci_overlap_node(void)
}
for_each_of_pci_range(&parser, &range)
- if (efifb_overlaps_pci_range(&screen_info, &range))
+ if (efifb_overlaps_pci_range(&sysfb_primary_display.screen, &range))
return np;
}
return NULL;
diff --git a/drivers/firmware/sysfb.c b/drivers/firmware/sysfb.c
index 916b28538a29..1f671f9219b0 100644
--- a/drivers/firmware/sysfb.c
+++ b/drivers/firmware/sysfb.c
@@ -66,7 +66,7 @@ static bool sysfb_unregister(void)
*/
void sysfb_disable(struct device *dev)
{
- struct screen_info *si = &screen_info;
+ struct screen_info *si = &sysfb_primary_display.screen;
struct device *parent;
mutex_lock(&disable_lock);
@@ -92,7 +92,7 @@ EXPORT_SYMBOL_GPL(sysfb_disable);
*/
bool sysfb_handles_screen_info(void)
{
- const struct screen_info *si = &screen_info;
+ const struct screen_info *si = &sysfb_primary_display.screen;
return !!screen_info_video_type(si);
}
@@ -141,7 +141,7 @@ static struct device *sysfb_parent_dev(const struct screen_info *si)
static __init int sysfb_init(void)
{
- struct screen_info *si = &screen_info;
+ struct screen_info *si = &sysfb_primary_display.screen;
struct device *parent;
unsigned int type;
struct simplefb_platform_data mode;
diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c
index a53af6fe81a6..9c937190be81 100644
--- a/drivers/hv/vmbus_drv.c
+++ b/drivers/hv/vmbus_drv.c
@@ -29,7 +29,7 @@
#include <linux/delay.h>
#include <linux/panic_notifier.h>
#include <linux/ptrace.h>
-#include <linux/screen_info.h>
+#include <linux/sysfb.h>
#include <linux/efi.h>
#include <linux/random.h>
#include <linux/kernel.h>
@@ -2340,8 +2340,8 @@ static void __maybe_unused vmbus_reserve_fb(void)
if (efi_enabled(EFI_BOOT)) {
/* Gen2 VM: get FB base from EFI framebuffer */
if (IS_ENABLED(CONFIG_SYSFB)) {
- start = screen_info.lfb_base;
- size = max_t(__u32, screen_info.lfb_size, 0x800000);
+ start = sysfb_primary_display.screen.lfb_base;
+ size = max_t(__u32, sysfb_primary_display.screen.lfb_size, 0x800000);
}
} else {
/* Gen1 VM: get FB base from PCI */
diff --git a/drivers/pci/vgaarb.c b/drivers/pci/vgaarb.c
index 436fa7f4c387..805be9ea4a34 100644
--- a/drivers/pci/vgaarb.c
+++ b/drivers/pci/vgaarb.c
@@ -26,7 +26,7 @@
#include <linux/poll.h>
#include <linux/miscdevice.h>
#include <linux/slab.h>
-#include <linux/screen_info.h>
+#include <linux/sysfb.h>
#include <linux/vt.h>
#include <linux/console.h>
#include <linux/acpi.h>
@@ -557,7 +557,7 @@ EXPORT_SYMBOL(vga_put);
static bool vga_is_firmware_default(struct pci_dev *pdev)
{
#if defined CONFIG_X86
- return pdev == screen_info_pci_dev(&screen_info);
+ return pdev == screen_info_pci_dev(&sysfb_primary_display.screen);
#else
return false;
#endif
diff --git a/drivers/video/screen_info_pci.c b/drivers/video/screen_info_pci.c
index 66bfc1d0a6dc..8f34d8a74f09 100644
--- a/drivers/video/screen_info_pci.c
+++ b/drivers/video/screen_info_pci.c
@@ -4,6 +4,7 @@
#include <linux/printk.h>
#include <linux/screen_info.h>
#include <linux/string.h>
+#include <linux/sysfb.h>
static struct pci_dev *screen_info_lfb_pdev;
static size_t screen_info_lfb_bar;
@@ -26,7 +27,7 @@ static bool __screen_info_relocation_is_valid(const struct screen_info *si, stru
void screen_info_apply_fixups(void)
{
- struct screen_info *si = &screen_info;
+ struct screen_info *si = &sysfb_primary_display.screen;
if (screen_info_lfb_pdev) {
struct resource *pr = &screen_info_lfb_pdev->resource[screen_info_lfb_bar];
@@ -75,7 +76,7 @@ static void screen_info_fixup_lfb(struct pci_dev *pdev)
.flags = IORESOURCE_MEM,
};
const struct resource *pr;
- const struct screen_info *si = &screen_info;
+ const struct screen_info *si = &sysfb_primary_display.screen;
if (screen_info_lfb_pdev)
return; // already found
diff --git a/include/linux/screen_info.h b/include/linux/screen_info.h
index 1690706206e8..c022403c599a 100644
--- a/include/linux/screen_info.h
+++ b/include/linux/screen_info.h
@@ -151,6 +151,4 @@ static inline struct pci_dev *screen_info_pci_dev(const struct screen_info *si)
}
#endif
-extern struct screen_info screen_info;
-
#endif /* _SCREEN_INFO_H */
diff --git a/include/linux/sysfb.h b/include/linux/sysfb.h
index 8b37247528bf..e8bde392c690 100644
--- a/include/linux/sysfb.h
+++ b/include/linux/sysfb.h
@@ -8,11 +8,10 @@
*/
#include <linux/err.h>
+#include <linux/platform_data/simplefb.h>
#include <linux/screen_info.h>
#include <linux/types.h>
-#include <linux/platform_data/simplefb.h>
-
struct device;
struct platform_device;
struct screen_info;
@@ -65,6 +64,8 @@ struct sysfb_display_info {
struct screen_info screen;
};
+extern struct sysfb_display_info sysfb_primary_display;
+
#ifdef CONFIG_SYSFB
void sysfb_disable(struct device *dev);
--
2.51.1
^ permalink raw reply related
* [PATCH v3 0/9] arch,sysfb,efi: Support EDID on non-x86 EFI systems
From: Thomas Zimmermann @ 2025-11-26 16:03 UTC (permalink / raw)
To: ardb, javierm, arnd, richard.lyu, helgaas
Cc: x86, linux-arm-kernel, linux-kernel, linux-efi, loongarch,
linux-riscv, dri-devel, linux-hyperv, linux-pci, linux-fbdev,
Thomas Zimmermann
Replace screen_info and edid_info with sysfb_primary_device of type
struct sysfb_display_info. Update all users. Then implement EDID support
in the kernel EFI code.
Sysfb DRM drivers currently fetch the global edid_info directly, when
they should get that information together with the screen_info from their
device. Wrapping screen_info and edid_info in sysfb_primary_display and
passing this to drivers enables this.
Replacing both with sysfb_primary_display has been motivate by the EFI
stub. EFI wants to transfer EDID via config table in a single entry.
Using struct sysfb_display_info this will become easily possible. Hence
accept some churn in architecture code for the long-term improvements.
Patches 1 and 2 reduce the exposure of screen_info in EFI-related code.
Patch 3 adds struct sysfb_display_info.
Patch 4 replaces scren_info with sysfb_primary_display. This results in
several changes throught the kernel, but is really just a refactoring.
Patch 5 updates sysfb to transfer sysfb_primary_display to the related
drivers.
Patch 6 moves edid_info into sysfb_primary_display. This resolves some
drivers' reference to the global edid_info, but also makes the EDID data
available on non-x86 architectures.
Patches 7 and 8 add support for EDID transfers on non-x86 EFI systems.
Patch 9 cleans up the config-table allocation to be easier to understand.
v3:
- replace SCREEN_INFO table entry (Ard)
- merge libstub patch into kernel patch
v2:
- combine v1 of the series at [1] plus changes from [2] and [3].
[1] https://lore.kernel.org/dri-devel/20251121135624.494768-1-tzimmermann@suse.de/
[2] https://lore.kernel.org/dri-devel/20251015160816.525825-1-tzimmermann@suse.de/
[3] https://lore.kernel.org/linux-efi/20251119123011.1187249-5-ardb+git@google.com/
Thomas Zimmermann (9):
efi: earlycon: Reduce number of references to global screen_info
efi: sysfb_efi: Reduce number of references to global screen_info
sysfb: Add struct sysfb_display_info
sysfb: Replace screen_info with sysfb_primary_display
sysfb: Pass sysfb_primary_display to devices
sysfb: Move edid_info into sysfb_primary_display
efi: Refactor init_primary_display() helpers
efi: Support EDID information
efi: libstub: Simplify interfaces for primary_display
arch/arm64/kernel/image-vars.h | 2 +-
arch/loongarch/kernel/efi.c | 38 ++++-----
arch/loongarch/kernel/image-vars.h | 2 +-
arch/riscv/kernel/image-vars.h | 2 +-
arch/x86/kernel/kexec-bzimage64.c | 4 +-
arch/x86/kernel/setup.c | 16 ++--
arch/x86/video/video-common.c | 4 +-
drivers/firmware/efi/earlycon.c | 42 +++++-----
drivers/firmware/efi/efi-init.c | 46 ++++++-----
drivers/firmware/efi/efi.c | 4 +-
drivers/firmware/efi/libstub/Makefile | 2 +-
drivers/firmware/efi/libstub/efi-stub-entry.c | 36 +++++++--
drivers/firmware/efi/libstub/efi-stub.c | 49 +++++++----
drivers/firmware/efi/libstub/efistub.h | 7 +-
.../firmware/efi/libstub/primary_display.c | 41 ++++++++++
drivers/firmware/efi/libstub/screen_info.c | 53 ------------
drivers/firmware/efi/libstub/zboot.c | 6 +-
drivers/firmware/efi/sysfb_efi.c | 81 ++++++++++---------
drivers/firmware/sysfb.c | 13 +--
drivers/firmware/sysfb_simplefb.c | 2 +-
drivers/gpu/drm/sysfb/efidrm.c | 14 ++--
drivers/gpu/drm/sysfb/vesadrm.c | 14 ++--
drivers/hv/vmbus_drv.c | 6 +-
drivers/pci/vgaarb.c | 4 +-
drivers/video/Kconfig | 8 +-
drivers/video/fbdev/core/fbmon.c | 8 +-
drivers/video/fbdev/efifb.c | 10 ++-
drivers/video/fbdev/vesafb.c | 10 ++-
drivers/video/fbdev/vga16fb.c | 8 +-
drivers/video/screen_info_pci.c | 5 +-
include/linux/efi.h | 9 ++-
include/linux/screen_info.h | 2 -
include/linux/sysfb.h | 23 ++++--
include/video/edid.h | 4 -
34 files changed, 321 insertions(+), 254 deletions(-)
create mode 100644 drivers/firmware/efi/libstub/primary_display.c
delete mode 100644 drivers/firmware/efi/libstub/screen_info.c
base-commit: d724c6f85e80a23ed46b7ebc6e38b527c09d64f5
--
2.51.1
^ permalink raw reply
* Re: [PATCH 2/3] irqchip/gic-v3: allocate one SGI for MSHV
From: Anirudh Rayabharam @ 2025-11-26 10:46 UTC (permalink / raw)
To: Marc Zyngier
Cc: kys, haiyangz, wei.liu, decui, longli, catalin.marinas, will,
tglx, Arnd Bergmann, akpm, agordeev, guoweikang.kernel, osandov,
bsz, linux-hyperv, linux-arm-kernel, linux-kernel, linux-arch
In-Reply-To: <86a509qi8p.wl-maz@kernel.org>
On Wed, Nov 26, 2025 at 09:02:30AM +0000, Marc Zyngier wrote:
> On Wed, 26 Nov 2025 08:51:59 +0000,
> Anirudh Rayabharam <anirudh@anirudhrb.com> wrote:
> >
> > On Tue, Nov 25, 2025 at 06:01:38PM +0000, Marc Zyngier wrote:
> > > On Tue, 25 Nov 2025 17:01:23 +0000,
> > > Anirudh Raybharam <anirudh@anirudhrb.com> wrote:
> > > >
> > > > From: Anirudh Rayabharam <anirudh@anirudhrb.com>
> > > >
> > > > From: Anirudh Rayabharam (Microsoft) <anirudh@anirudhrb.com>
> > > >
> > > > Currently SGIs are allocated only for the smp subsystem. The MSHV
> > > > (Microsoft Hypervisor aka Hyper-V) code also needs an SGI that can be
> > > > programmed into the SYNIC to receive intercepts from the hypervisor. The
> > > > hypervisor would then assert this SGI whenever there is a guest
> > > > VMEXIT.
> > > >
> > > > Allocate one SGI for MSHV use in addition to the SGIs allocated for
> > > > IPIs. When running under MSHV, the full SGI range can be used i.e. no
> > > > need to reserve SGIs 8-15 for the secure firmware.
> > > >
> > > > Since this SGI is needed only when running as a parent partition (i.e.
> > > > we can create guest partitions), check for it before allocating an SGI.
> > >
> > > Sorry, but that's not an acceptable situation.
> > >
> > > SGIs are for Linux to use, nobody else, and that allocation must be
> >
> > Why does this restriction exist? In the code SGIs 8-15 are left for
> > secure firmware. So, things other than Linux can use SGIs. Why not MSHV
> > then?
>
> Because SGIs are for *internal* usage. Not usage from another random
> piece of SW. The ACPI tables explicitly don't describe SGIs. DT
> explicitly don't describe SGIs. Do you get the clue?
The name Software Generated Interrupts suggests that it is supposed to be
used by pieces of SW.
Yes, ACPI/DT don't describe SGIs because they're not supposed to be used
by devices. SW has full control over SGIs and it is up to the SW to
assign meaning to them, isn't it?
>
> > > the same irrespective of whether Linux runs virtualised or not. This
> > > also won't work with GICv5 (there are no SGIs at all), so this is
> > > doomed from the very start, and would immediately create technical
> > > debt.
> >
> > Hyper-V always presents a GICv3 so we don't need to worry about GICv5.
>
> Well, that's pretty short sighted of you, and eventually you'll have
> to support it, or just die. So do the right thing from the beginning.
Well, we don't when or if that will happen. But if it does happen, we
can solve it in a way that makes sense for GICv5. If there are no SGIs
at all, great, maybe we'll have a nicer solution then.
>
> > >
> > > If you want to signal an interrupt to Linux, expose a device with an
> > > interrupt in a firmware table (i.e. not an SGI), and use that in your
> > > driver.
> >
> > You mean in the ACPI tables? That would require us to modify the
> > firmware to expose this virtual device right?
>
> Yes. How is that surprising?
It's not ideal that we would need some custom firmware to run Linux on
MSHV (as a parent). Do you think there could be some other possible solution
for handling this in the kernel? Maybe by thinking of it as a platform specific
quirk or something?
Thanks,
Anirudh.
>
> M.
>
> --
> Without deviation from the norm, progress is not possible.
^ permalink raw reply
* Re: [PATCH 2/3] irqchip/gic-v3: allocate one SGI for MSHV
From: Marc Zyngier @ 2025-11-26 9:02 UTC (permalink / raw)
To: Anirudh Rayabharam
Cc: kys, haiyangz, wei.liu, decui, longli, catalin.marinas, will,
tglx, Arnd Bergmann, akpm, agordeev, guoweikang.kernel, osandov,
bsz, linux-hyperv, linux-arm-kernel, linux-kernel, linux-arch
In-Reply-To: <aSa_rxG80LDXDlhr@anirudh-surface.localdomain>
On Wed, 26 Nov 2025 08:51:59 +0000,
Anirudh Rayabharam <anirudh@anirudhrb.com> wrote:
>
> On Tue, Nov 25, 2025 at 06:01:38PM +0000, Marc Zyngier wrote:
> > On Tue, 25 Nov 2025 17:01:23 +0000,
> > Anirudh Raybharam <anirudh@anirudhrb.com> wrote:
> > >
> > > From: Anirudh Rayabharam <anirudh@anirudhrb.com>
> > >
> > > From: Anirudh Rayabharam (Microsoft) <anirudh@anirudhrb.com>
> > >
> > > Currently SGIs are allocated only for the smp subsystem. The MSHV
> > > (Microsoft Hypervisor aka Hyper-V) code also needs an SGI that can be
> > > programmed into the SYNIC to receive intercepts from the hypervisor. The
> > > hypervisor would then assert this SGI whenever there is a guest
> > > VMEXIT.
> > >
> > > Allocate one SGI for MSHV use in addition to the SGIs allocated for
> > > IPIs. When running under MSHV, the full SGI range can be used i.e. no
> > > need to reserve SGIs 8-15 for the secure firmware.
> > >
> > > Since this SGI is needed only when running as a parent partition (i.e.
> > > we can create guest partitions), check for it before allocating an SGI.
> >
> > Sorry, but that's not an acceptable situation.
> >
> > SGIs are for Linux to use, nobody else, and that allocation must be
>
> Why does this restriction exist? In the code SGIs 8-15 are left for
> secure firmware. So, things other than Linux can use SGIs. Why not MSHV
> then?
Because SGIs are for *internal* usage. Not usage from another random
piece of SW. The ACPI tables explicitly don't describe SGIs. DT
explicitly don't describe SGIs. Do you get the clue?
> > the same irrespective of whether Linux runs virtualised or not. This
> > also won't work with GICv5 (there are no SGIs at all), so this is
> > doomed from the very start, and would immediately create technical
> > debt.
>
> Hyper-V always presents a GICv3 so we don't need to worry about GICv5.
Well, that's pretty short sighted of you, and eventually you'll have
to support it, or just die. So do the right thing from the beginning.
> >
> > If you want to signal an interrupt to Linux, expose a device with an
> > interrupt in a firmware table (i.e. not an SGI), and use that in your
> > driver.
>
> You mean in the ACPI tables? That would require us to modify the
> firmware to expose this virtual device right?
Yes. How is that surprising?
M.
--
Without deviation from the norm, progress is not possible.
^ permalink raw reply
* Re: [PATCH 2/3] irqchip/gic-v3: allocate one SGI for MSHV
From: Anirudh Rayabharam @ 2025-11-26 8:51 UTC (permalink / raw)
To: Marc Zyngier
Cc: kys, haiyangz, wei.liu, decui, longli, catalin.marinas, will,
tglx, Arnd Bergmann, akpm, agordeev, guoweikang.kernel, osandov,
bsz, linux-hyperv, linux-arm-kernel, linux-kernel, linux-arch
In-Reply-To: <86bjkqq9dp.wl-maz@kernel.org>
On Tue, Nov 25, 2025 at 06:01:38PM +0000, Marc Zyngier wrote:
> On Tue, 25 Nov 2025 17:01:23 +0000,
> Anirudh Raybharam <anirudh@anirudhrb.com> wrote:
> >
> > From: Anirudh Rayabharam <anirudh@anirudhrb.com>
> >
> > From: Anirudh Rayabharam (Microsoft) <anirudh@anirudhrb.com>
> >
> > Currently SGIs are allocated only for the smp subsystem. The MSHV
> > (Microsoft Hypervisor aka Hyper-V) code also needs an SGI that can be
> > programmed into the SYNIC to receive intercepts from the hypervisor. The
> > hypervisor would then assert this SGI whenever there is a guest
> > VMEXIT.
> >
> > Allocate one SGI for MSHV use in addition to the SGIs allocated for
> > IPIs. When running under MSHV, the full SGI range can be used i.e. no
> > need to reserve SGIs 8-15 for the secure firmware.
> >
> > Since this SGI is needed only when running as a parent partition (i.e.
> > we can create guest partitions), check for it before allocating an SGI.
>
> Sorry, but that's not an acceptable situation.
>
> SGIs are for Linux to use, nobody else, and that allocation must be
Why does this restriction exist? In the code SGIs 8-15 are left for
secure firmware. So, things other than Linux can use SGIs. Why not MSHV
then?
> the same irrespective of whether Linux runs virtualised or not. This
> also won't work with GICv5 (there are no SGIs at all), so this is
> doomed from the very start, and would immediately create technical
> debt.
Hyper-V always presents a GICv3 so we don't need to worry about GICv5.
>
> If you want to signal an interrupt to Linux, expose a device with an
> interrupt in a firmware table (i.e. not an SGI), and use that in your
> driver.
You mean in the ACPI tables? That would require us to modify the
firmware to expose this virtual device right?
Anirudh.
>
> Thanks,
>
> M.
>
> --
> Without deviation from the norm, progress is not possible.
^ permalink raw reply
* [PATCH 5.4-6.6 v2] uio_hv_generic: Enable user space to manage interrupt_mask for subchannels
From: Naman Jain @ 2025-11-26 6:19 UTC (permalink / raw)
To: Wei Liu, Dexuan Cui, Greg Kroah-Hartman, Michael Kelley, Long Li,
Saurabh Sengar
Cc: K . Y . Srinivasan, Haiyang Zhang, Tianyu Lan, Naman Jain,
linux-hyperv, linux-kernel, stable, Salvatore Bonaccorso,
Peter Morrow
From: Long Li <longli@microsoft.com>
Enable the user space to manage interrupt_mask for subchannels through
irqcontrol interface for uio device. Also remove the memory barrier
when monitor bit is enabled as it is not necessary.
This is a backport of the upstream
commit d062463edf17 ("uio_hv_generic: Set event for all channels on the device")
with some modifications to resolve merge conflicts and take care of
missing support for slow devices on older kernels.
Original change was not a fix, but it needs to be backported to fix a
NULL pointer crash resulting from missing interrupt mask setting.
Commit b15b7d2a1b09 ("uio_hv_generic: Let userspace take care of interrupt mask")
removed the default setting of interrupt_mask for channels (including
subchannels) in the uio_hv_generic driver, as it relies on the user space
to take care of managing it. This approach works fine when user space
can control this setting using the irqcontrol interface provided for uio
devices. Support for setting the interrupt mask through this interface for
subchannels came only after commit d062463edf17 ("uio_hv_generic: Set event
for all channels on the device"). On older kernels, this change is not
present. With uio_hv_generic no longer setting the interrupt_mask, and
userspace not having the capability to set it, it remains unset,
and interrupts can come for the subchannels, which can result in a crash
in hv_uio_channel_cb. Backport the change to older kernels, where this
change was not present, to allow userspace to set the interrupt mask
properly for subchannels. Additionally, this patch also adds certain
checks for primary vs subchannels in the hv_uio_channel_cb, which can
gracefully handle these two cases and prevent the NULL pointer crashes.
Signed-off-by: Long Li <longli@microsoft.com>
Fixes: b15b7d2a1b09 ("uio_hv_generic: Let userspace take care of interrupt mask")
Closes: https://bugs.debian.org/1120602
Cc: stable@vger.kernel.org # all LTS kernels from 5.4 till 6.6
Signed-off-by: Naman Jain <namjain@linux.microsoft.com>
---
Changes since v1:
https://lore.kernel.org/all/20251115085937.2237-1-namjain@linux.microsoft.com/
* Changed commit being fixed, to reflect upstream commit
* Added stable kernel versions info in Stable tag and email subject.
This needs to be ported to all the stable kernels, where the Fixes patch
went in - 5.4.x, 5.10.x, 5.15.x, 6.1.x, 6.6.x.
Previous notes:
Remove reviewed-by tags since the original code has changed quite a bit
while backporting.
Backported change for 6.12 kernel is sent separately.
---
drivers/uio/uio_hv_generic.c | 21 +++++++++++++++++----
1 file changed, 17 insertions(+), 4 deletions(-)
diff --git a/drivers/uio/uio_hv_generic.c b/drivers/uio/uio_hv_generic.c
index 137109f5f69b..95c1f08028a3 100644
--- a/drivers/uio/uio_hv_generic.c
+++ b/drivers/uio/uio_hv_generic.c
@@ -80,9 +80,15 @@ hv_uio_irqcontrol(struct uio_info *info, s32 irq_state)
{
struct hv_uio_private_data *pdata = info->priv;
struct hv_device *dev = pdata->device;
+ struct vmbus_channel *primary, *sc;
- dev->channel->inbound.ring_buffer->interrupt_mask = !irq_state;
- virt_mb();
+ primary = dev->channel;
+ primary->inbound.ring_buffer->interrupt_mask = !irq_state;
+
+ mutex_lock(&vmbus_connection.channel_mutex);
+ list_for_each_entry(sc, &primary->sc_list, sc_list)
+ sc->inbound.ring_buffer->interrupt_mask = !irq_state;
+ mutex_unlock(&vmbus_connection.channel_mutex);
return 0;
}
@@ -93,11 +99,18 @@ hv_uio_irqcontrol(struct uio_info *info, s32 irq_state)
static void hv_uio_channel_cb(void *context)
{
struct vmbus_channel *chan = context;
- struct hv_device *hv_dev = chan->device_obj;
- struct hv_uio_private_data *pdata = hv_get_drvdata(hv_dev);
+ struct hv_device *hv_dev;
+ struct hv_uio_private_data *pdata;
virt_mb();
+ /*
+ * The callback may come from a subchannel, in which case look
+ * for the hv device in the primary channel
+ */
+ hv_dev = chan->primary_channel ?
+ chan->primary_channel->device_obj : chan->device_obj;
+ pdata = hv_get_drvdata(hv_dev);
uio_event_notify(&pdata->info);
}
--
2.43.0
^ permalink raw reply related
* Re: [PATCH 6.6 and older] uio_hv_generic: Enable user space to manage interrupt_mask for subchannels
From: Naman Jain @ 2025-11-26 6:15 UTC (permalink / raw)
To: Salvatore Bonaccorso, Greg Kroah-Hartman
Cc: Wei Liu, Dexuan Cui, Michael Kelley, Long Li, Saurabh Sengar,
K . Y . Srinivasan, Haiyang Zhang, Tianyu Lan, linux-hyperv,
linux-kernel, stable, Peter Morrow
In-Reply-To: <aSBeVyYD2HQ5zNbC@eldamar.lan>
On 11/21/2025 6:13 PM, Salvatore Bonaccorso wrote:
> Hi,
>
> On Fri, Nov 21, 2025 at 11:10:43AM +0100, Greg Kroah-Hartman wrote:
>> On Sat, Nov 15, 2025 at 02:29:37PM +0530, Naman Jain wrote:
>>> From: Long Li <longli@microsoft.com>
>>>
>>> Enable the user space to manage interrupt_mask for subchannels through
>>> irqcontrol interface for uio device. Also remove the memory barrier
>>> when monitor bit is enabled as it is not necessary.
>>>
>>> This is a backport of the upstream commit
>>> d062463edf17 ("uio_hv_generic: Set event for all channels on the device")
>>> with some modifications to resolve merge conflicts and take care of
>>> missing support for slow devices on older kernels.
>>> Original change was not a fix, but it needs to be backported to fix a
>>> NULL pointer crash resulting from missing interrupt mask setting.
>>>
>>> Commit 37bd91f22794 ("uio_hv_generic: Let userspace take care of interrupt mask")
>>> removed the default setting of interrupt_mask for channels (including
>>> subchannels) in the uio_hv_generic driver, as it relies on the user space
>>> to take care of managing it. This approach works fine when user space
>>> can control this setting using the irqcontrol interface provided for uio
>>> devices. Support for setting the interrupt mask through this interface for
>>> subchannels came only after commit d062463edf17 ("uio_hv_generic: Set event
>>> for all channels on the device"). On older kernels, this change is not
>>> present. With uio_hv_generic no longer setting the interrupt_mask, and
>>> userspace not having the capability to set it, it remains unset,
>>> and interrupts can come for the subchannels, which can result in a crash
>>> in hv_uio_channel_cb. Backport the change to older kernels, where this
>>> change was not present, to allow userspace to set the interrupt mask
>>> properly for subchannels. Additionally, this patch also adds certain
>>> checks for primary vs subchannels in the hv_uio_channel_cb, which can
>>> gracefully handle these two cases and prevent the NULL pointer crashes.
>>>
>>> Signed-off-by: Long Li <longli@microsoft.com>
>>> Fixes: 37bd91f22794 ("uio_hv_generic: Let userspace take care of interrupt mask")
>>
>> This is a 6.12.y commit id, so a fix for 6.6.y does not make sense :(
>
> Should maybe be updated to reflect the original upstream commit. In
> fact b15b7d2a1b09 ("uio_hv_generic: Let userspace take care of
> interrupt mask") was backported to various stable series:
>
> v5.4.301: 540aac117eaea5723cef5e4cbf3035c4ac654d92 uio_hv_generic: Let userspace take care of interrupt mask
> v5.10.246: 65d40acd911c7011745cbbd2aaac34eb5266d11e uio_hv_generic: Let userspace take care of interrupt mask
> v5.15.195: a44f61f878f32071d6378e8dd7c2d47f9490c8f7 uio_hv_generic: Let userspace take care of interrupt mask
> v6.1.156: 01ce972e6f9974a7c76943bcb7e93746917db83a uio_hv_generic: Let userspace take care of interrupt mask
> v6.6.112: 2af39ab5e6dc46b835a52e80a22d0cad430985e3 uio_hv_generic: Let userspace take care of interrupt mask
> v6.12.53: 37bd91f22794dc05436130d6983302cb90ecfe7e uio_hv_generic: Let userspace take care of interrupt mask
> v6.17.3: e29587c07537929684faa365027f4b0d87521e1b uio_hv_generic: Let userspace take care of interrupt mask
>
> And Peter just confirmed in
> https://lore.kernel.org/stable/CAFcZKTyOcDqDJRB4sgN7Q-dabBU0eg7KKs=yBJhB=CNDyy7scQ@mail.gmail.com/
> that he is seeing the problem now as well after updating from
> 6.1.153-1 to 6.1.158-1 in Debian.
>
>>> Closes: https://bugs.debian.org/1120602
>>> Cc: <stable@vger.kernel.org> # 6.6.x and older
>>
>> How "old" do you want this? Can you fix the Fixes: line up and resend
>> with this info?
Hi Greg,
Sorry for replying late, as I was away for personal reasons since last week.
I'll change the commit to reflect upstream commit id and resend the
patch and also include the exact older kernels info in the stable tag.
>
> It is at least relevant for back in 6.1.y now, but I'm not sure about
> the older series. I will let Naman speak up.
>
> I guess the proper fixes tracking is a bit "tricky" because it only
> affected some of the stable series, namely those which had a backport
> of b15b7d2a1b09 ("uio_hv_generic: Let userspace take care of interrupt
> mask") done before the including a backport of d062463edf17
> ("uio_hv_generic: Set event for all channels on the device"). So this
> is the reason why we seeing it first in 6.12.y stable series (but now
> as well on olders), but not a problem on 6.17.y.
>
> Hope this explanation helps, please keep in mind that I'm no expert
> here by no means, just helping to report it from downstream Debian up
> here.
>
> Regards,
> Salvatore
Thanks Salvatore for chiming in. I was under the impression that the
commit ids remain same across different kernel versions for the same
commit, but that was completely wrong.
Regards,
Naman
^ permalink raw reply
* Re: Bug#1120602: [REGRESSION 6.12.y] hyper-v: BUG: kernel NULL pointer dereference, address: 00000000000000a0: RIP: 0010:hv_uio_channel_cb+0xd/0x20 [uio_hv_generic]
From: Naman Jain @ 2025-11-26 6:15 UTC (permalink / raw)
To: Peter Morrow
Cc: Salvatore Bonaccorso, 1120602, Long Li, linux-hyperv,
linux-kernel, regressions, stable, John Starks, Michael Kelley,
Tianyu Lan, K. Y. Srinivasan, Haiyang Zhang, Wei Liu, Dexuan Cui,
Greg Kroah-Hartman
In-Reply-To: <CAFcZKTyOcDqDJRB4sgN7Q-dabBU0eg7KKs=yBJhB=CNDyy7scQ@mail.gmail.com>
On 11/21/2025 3:34 PM, Peter Morrow wrote:
> Hi Naman/Salvatore,
>
> Is it possible to get this fixed in the 6.1 LTS series too? I just ran
> into this crash when moving from bookworm based Debian kernel
> 6.1.153-1 to 6.1.158-1. I saw that "uio_hv_generic: Let userspace take
> care of interrupt mask" appeared in 6.1.156.
>
> Thanks,
> Peter.
>
Hi Peter,
Yes, I have sent a patch for older kernel versions as well.
I am working to fix the review comments and send new revisions.
Here is the link:
https://lore.kernel.org/all/20251115085937.2237-1-namjain@linux.microsoft.com/
Regards,
Naman
^ permalink raw reply
* Re: [PATCH v7 1/9] x86/acpi: Add functions to setup and access the wakeup mailbox
From: Ricardo Neri @ 2025-11-26 2:20 UTC (permalink / raw)
To: Rafael J. Wysocki
Cc: x86, Krzysztof Kozlowski, Conor Dooley, Rob Herring,
K. Y. Srinivasan, Haiyang Zhang, Wei Liu, Dexuan Cui,
Michael Kelley, Saurabh Sengar, Chris Oo, Kirill A. Shutemov,
linux-hyperv, devicetree, linux-acpi, linux-kernel, Ricardo Neri
In-Reply-To: <CAJZ5v0gd_6b6s4aEpSvdfb4-+AULTWkqQqM3OE1eg5XzYaxQFQ@mail.gmail.com>
On Mon, Nov 24, 2025 at 05:01:11PM +0100, Rafael J. Wysocki wrote:
> On Mon, Nov 17, 2025 at 6:04 PM Ricardo Neri
> <ricardo.neri-calderon@linux.intel.com> wrote:
> >
> > Systems that describe hardware using DeviceTree graphs may enumerate and
> > implement the wakeup mailbox as defined in the ACPI specification but do
> > not otherwise depend on ACPI. Expose functions to setup and access the
> > location of the wakeup mailbox from outside ACPI code.
> >
> > The function acpi_setup_mp_wakeup_mailbox() stores the physical address of
> > the mailbox and updates the wakeup_secondary_cpu_64() APIC callback.
> >
> > The function acpi_madt_multiproc_wakeup_mailbox() returns a pointer to the
> > mailbox.
> >
> > Signed-off-by: Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
>
> Acked-by: Rafael J. Wysocki (Intel) <rafael@kernel.org>
Thanks Rafael!
^ permalink raw reply
* [PATCH v7 7/7] Drivers: hv: Add support for movable memory regions
From: Stanislav Kinsburskii @ 2025-11-26 2:09 UTC (permalink / raw)
To: kys, haiyangz, wei.liu, decui; +Cc: linux-hyperv, linux-kernel
In-Reply-To: <176412196000.447063.4256335030026363827.stgit@skinsburskii-cloud-desktop.internal.cloudapp.net>
Introduce support for movable memory regions in the Hyper-V root partition
driver to improve memory management flexibility and enable advanced use
cases such as dynamic memory remapping.
Mirror the address space between the Linux root partition and guest VMs
using HMM. The root partition owns the memory, while guest VMs act as
devices with page tables managed via hypercalls. MSHV handles VP intercepts
by invoking hmm_range_fault() and updating SLAT entries. When memory is
reclaimed, HMM invalidates the relevant regions, prompting MSHV to clear
SLAT entries; guest VMs will fault again on access.
Integrate mmu_interval_notifier for movable regions, implement handlers for
HMM faults and memory invalidation, and update memory region mapping logic
to support movable regions.
While MMU notifiers are commonly used in virtualization drivers, this
implementation leverages HMM (Heterogeneous Memory Management) for its
specialized functionality. HMM provides a framework for mirroring,
invalidation, and fault handling, reducing boilerplate and improving
maintainability compared to generic MMU notifiers.
Signed-off-by: Stanislav Kinsburskii <skinsburskii@linux.microsoft.com>
---
drivers/hv/Kconfig | 2
drivers/hv/mshv_regions.c | 215 ++++++++++++++++++++++++++++++++++++++++++-
drivers/hv/mshv_root.h | 17 +++
drivers/hv/mshv_root_main.c | 139 +++++++++++++++++++++++-----
4 files changed, 343 insertions(+), 30 deletions(-)
diff --git a/drivers/hv/Kconfig b/drivers/hv/Kconfig
index d4a8d349200c..7937ac0cbd0f 100644
--- a/drivers/hv/Kconfig
+++ b/drivers/hv/Kconfig
@@ -76,6 +76,8 @@ config MSHV_ROOT
depends on PAGE_SIZE_4KB
select EVENTFD
select VIRT_XFER_TO_GUEST_WORK
+ select HMM_MIRROR
+ select MMU_NOTIFIER
default n
help
Select this option to enable support for booting and running as root
diff --git a/drivers/hv/mshv_regions.c b/drivers/hv/mshv_regions.c
index 6450a7ed8493..d7b0f012c3be 100644
--- a/drivers/hv/mshv_regions.c
+++ b/drivers/hv/mshv_regions.c
@@ -7,6 +7,8 @@
* Authors: Microsoft Linux virtualization team
*/
+#include <linux/hmm.h>
+#include <linux/hyperv.h>
#include <linux/kref.h>
#include <linux/mm.h>
#include <linux/vmalloc.h>
@@ -15,6 +17,8 @@
#include "mshv_root.h"
+#define MSHV_MAP_FAULT_IN_PAGES PTRS_PER_PMD
+
/**
* mshv_region_process_chunk - Processes a contiguous chunk of memory pages
* in a region.
@@ -152,9 +156,6 @@ struct mshv_mem_region *mshv_region_create(u64 guest_pfn, u64 nr_pages,
if (flags & BIT(MSHV_SET_MEM_BIT_EXECUTABLE))
region->hv_map_flags |= HV_MAP_GPA_EXECUTABLE;
- if (!is_mmio)
- region->flags.range_pinned = true;
-
kref_init(®ion->refcount);
return region;
@@ -239,7 +240,7 @@ int mshv_region_map(struct mshv_mem_region *region)
static void mshv_region_invalidate_pages(struct mshv_mem_region *region,
u64 page_offset, u64 page_count)
{
- if (region->flags.range_pinned)
+ if (region->type == MSHV_REGION_TYPE_MEM_PINNED)
unpin_user_pages(region->pages + page_offset, page_count);
memset(region->pages + page_offset, 0,
@@ -313,6 +314,9 @@ static void mshv_region_destroy(struct kref *ref)
struct mshv_partition *partition = region->partition;
int ret;
+ if (region->type == MSHV_REGION_TYPE_MEM_MOVABLE)
+ mshv_region_movable_fini(region);
+
if (mshv_partition_encrypted(partition)) {
ret = mshv_region_share(region);
if (ret) {
@@ -339,3 +343,206 @@ int mshv_region_get(struct mshv_mem_region *region)
{
return kref_get_unless_zero(®ion->refcount);
}
+
+/**
+ * mshv_region_hmm_fault_and_lock - Handle HMM faults and lock the memory region
+ * @region: Pointer to the memory region structure
+ * @range: Pointer to the HMM range structure
+ *
+ * This function performs the following steps:
+ * 1. Reads the notifier sequence for the HMM range.
+ * 2. Acquires a read lock on the memory map.
+ * 3. Handles HMM faults for the specified range.
+ * 4. Releases the read lock on the memory map.
+ * 5. If successful, locks the memory region mutex.
+ * 6. Verifies if the notifier sequence has changed during the operation.
+ * If it has, releases the mutex and returns -EBUSY to match with
+ * hmm_range_fault() return code for repeating.
+ *
+ * Return: 0 on success, a negative error code otherwise.
+ */
+static int mshv_region_hmm_fault_and_lock(struct mshv_mem_region *region,
+ struct hmm_range *range)
+{
+ int ret;
+
+ range->notifier_seq = mmu_interval_read_begin(range->notifier);
+ mmap_read_lock(region->mni.mm);
+ ret = hmm_range_fault(range);
+ mmap_read_unlock(region->mni.mm);
+ if (ret)
+ return ret;
+
+ mutex_lock(®ion->mutex);
+
+ if (mmu_interval_read_retry(range->notifier, range->notifier_seq)) {
+ mutex_unlock(®ion->mutex);
+ cond_resched();
+ return -EBUSY;
+ }
+
+ return 0;
+}
+
+/**
+ * mshv_region_range_fault - Handle memory range faults for a given region.
+ * @region: Pointer to the memory region structure.
+ * @page_offset: Offset of the page within the region.
+ * @page_count: Number of pages to handle.
+ *
+ * This function resolves memory faults for a specified range of pages
+ * within a memory region. It uses HMM (Heterogeneous Memory Management)
+ * to fault in the required pages and updates the region's page array.
+ *
+ * Return: 0 on success, negative error code on failure.
+ */
+static int mshv_region_range_fault(struct mshv_mem_region *region,
+ u64 page_offset, u64 page_count)
+{
+ struct hmm_range range = {
+ .notifier = ®ion->mni,
+ .default_flags = HMM_PFN_REQ_FAULT | HMM_PFN_REQ_WRITE,
+ };
+ unsigned long *pfns;
+ int ret;
+ u64 i;
+
+ pfns = kmalloc_array(page_count, sizeof(unsigned long), GFP_KERNEL);
+ if (!pfns)
+ return -ENOMEM;
+
+ range.hmm_pfns = pfns;
+ range.start = region->start_uaddr + page_offset * HV_HYP_PAGE_SIZE;
+ range.end = range.start + page_count * HV_HYP_PAGE_SIZE;
+
+ do {
+ ret = mshv_region_hmm_fault_and_lock(region, &range);
+ } while (ret == -EBUSY);
+
+ if (ret)
+ goto out;
+
+ for (i = 0; i < page_count; i++)
+ region->pages[page_offset + i] = hmm_pfn_to_page(pfns[i]);
+
+ ret = mshv_region_remap_pages(region, region->hv_map_flags,
+ page_offset, page_count);
+
+ mutex_unlock(®ion->mutex);
+out:
+ kfree(pfns);
+ return ret;
+}
+
+bool mshv_region_handle_gfn_fault(struct mshv_mem_region *region, u64 gfn)
+{
+ u64 page_offset, page_count;
+ int ret;
+
+ /* Align the page offset to the nearest MSHV_MAP_FAULT_IN_PAGES. */
+ page_offset = ALIGN_DOWN(gfn - region->start_gfn,
+ MSHV_MAP_FAULT_IN_PAGES);
+
+ /* Map more pages than requested to reduce the number of faults. */
+ page_count = min(region->nr_pages - page_offset,
+ MSHV_MAP_FAULT_IN_PAGES);
+
+ ret = mshv_region_range_fault(region, page_offset, page_count);
+
+ WARN_ONCE(ret,
+ "p%llu: GPA intercept failed: region %#llx-%#llx, gfn %#llx, page_offset %llu, page_count %llu\n",
+ region->partition->pt_id, region->start_uaddr,
+ region->start_uaddr + (region->nr_pages << HV_HYP_PAGE_SHIFT),
+ gfn, page_offset, page_count);
+
+ return !ret;
+}
+
+/**
+ * mshv_region_interval_invalidate - Invalidate a range of memory region
+ * @mni: Pointer to the mmu_interval_notifier structure
+ * @range: Pointer to the mmu_notifier_range structure
+ * @cur_seq: Current sequence number for the interval notifier
+ *
+ * This function invalidates a memory region by remapping its pages with
+ * no access permissions. It locks the region's mutex to ensure thread safety
+ * and updates the sequence number for the interval notifier. If the range
+ * is blockable, it uses a blocking lock; otherwise, it attempts a non-blocking
+ * lock and returns false if unsuccessful.
+ *
+ * NOTE: Failure to invalidate a region is a serious error, as the pages will
+ * be considered freed while they are still mapped by the hypervisor.
+ * Any attempt to access such pages will likely crash the system.
+ *
+ * Return: true if the region was successfully invalidated, false otherwise.
+ */
+static bool mshv_region_interval_invalidate(struct mmu_interval_notifier *mni,
+ const struct mmu_notifier_range *range,
+ unsigned long cur_seq)
+{
+ struct mshv_mem_region *region = container_of(mni,
+ struct mshv_mem_region,
+ mni);
+ u64 page_offset, page_count;
+ unsigned long mstart, mend;
+ int ret = -EPERM;
+
+ if (mmu_notifier_range_blockable(range))
+ mutex_lock(®ion->mutex);
+ else if (!mutex_trylock(®ion->mutex))
+ goto out_fail;
+
+ mmu_interval_set_seq(mni, cur_seq);
+
+ mstart = max(range->start, region->start_uaddr);
+ mend = min(range->end, region->start_uaddr +
+ (region->nr_pages << HV_HYP_PAGE_SHIFT));
+
+ page_offset = HVPFN_DOWN(mstart - region->start_uaddr);
+ page_count = HVPFN_DOWN(mend - mstart);
+
+ ret = mshv_region_remap_pages(region, HV_MAP_GPA_NO_ACCESS,
+ page_offset, page_count);
+ if (ret)
+ goto out_fail;
+
+ mshv_region_invalidate_pages(region, page_offset, page_count);
+
+ mutex_unlock(®ion->mutex);
+
+ return true;
+
+out_fail:
+ WARN_ONCE(ret,
+ "Failed to invalidate region %#llx-%#llx (range %#lx-%#lx, event: %u, pages %#llx-%#llx, mm: %#llx): %d\n",
+ region->start_uaddr,
+ region->start_uaddr + (region->nr_pages << HV_HYP_PAGE_SHIFT),
+ range->start, range->end, range->event,
+ page_offset, page_offset + page_count - 1, (u64)range->mm, ret);
+ return false;
+}
+
+static const struct mmu_interval_notifier_ops mshv_region_mni_ops = {
+ .invalidate = mshv_region_interval_invalidate,
+};
+
+void mshv_region_movable_fini(struct mshv_mem_region *region)
+{
+ mmu_interval_notifier_remove(®ion->mni);
+}
+
+bool mshv_region_movable_init(struct mshv_mem_region *region)
+{
+ int ret;
+
+ ret = mmu_interval_notifier_insert(®ion->mni, current->mm,
+ region->start_uaddr,
+ region->nr_pages << HV_HYP_PAGE_SHIFT,
+ &mshv_region_mni_ops);
+ if (ret)
+ return false;
+
+ mutex_init(®ion->mutex);
+
+ return true;
+}
diff --git a/drivers/hv/mshv_root.h b/drivers/hv/mshv_root.h
index 4249534ba900..9cd76076d490 100644
--- a/drivers/hv/mshv_root.h
+++ b/drivers/hv/mshv_root.h
@@ -15,6 +15,7 @@
#include <linux/hashtable.h>
#include <linux/dev_printk.h>
#include <linux/build_bug.h>
+#include <linux/mmu_notifier.h>
#include <uapi/linux/mshv.h>
/*
@@ -70,6 +71,12 @@ do { \
#define vp_info(v, fmt, ...) vp_devprintk(info, v, fmt, ##__VA_ARGS__)
#define vp_dbg(v, fmt, ...) vp_devprintk(dbg, v, fmt, ##__VA_ARGS__)
+enum mshv_region_type {
+ MSHV_REGION_TYPE_MEM_PINNED,
+ MSHV_REGION_TYPE_MEM_MOVABLE,
+ MSHV_REGION_TYPE_MMIO
+};
+
struct mshv_mem_region {
struct hlist_node hnode;
struct kref refcount;
@@ -77,11 +84,10 @@ struct mshv_mem_region {
u64 start_gfn;
u64 start_uaddr;
u32 hv_map_flags;
- struct {
- u64 range_pinned: 1;
- u64 reserved: 63;
- } flags;
struct mshv_partition *partition;
+ enum mshv_region_type type;
+ struct mmu_interval_notifier mni;
+ struct mutex mutex; /* protects region pages remapping */
struct page *pages[];
};
@@ -324,5 +330,8 @@ void mshv_region_invalidate(struct mshv_mem_region *region);
int mshv_region_pin(struct mshv_mem_region *region);
void mshv_region_put(struct mshv_mem_region *region);
int mshv_region_get(struct mshv_mem_region *region);
+bool mshv_region_handle_gfn_fault(struct mshv_mem_region *region, u64 gfn);
+void mshv_region_movable_fini(struct mshv_mem_region *region);
+bool mshv_region_movable_init(struct mshv_mem_region *region);
#endif /* _MSHV_ROOT_H_ */
diff --git a/drivers/hv/mshv_root_main.c b/drivers/hv/mshv_root_main.c
index 1ef2a28beb17..6003fb4477bc 100644
--- a/drivers/hv/mshv_root_main.c
+++ b/drivers/hv/mshv_root_main.c
@@ -594,14 +594,98 @@ static long mshv_run_vp_with_root_scheduler(struct mshv_vp *vp)
static_assert(sizeof(struct hv_message) <= MSHV_RUN_VP_BUF_SZ,
"sizeof(struct hv_message) must not exceed MSHV_RUN_VP_BUF_SZ");
+static struct mshv_mem_region *
+mshv_partition_region_by_gfn(struct mshv_partition *partition, u64 gfn)
+{
+ struct mshv_mem_region *region;
+
+ hlist_for_each_entry(region, &partition->pt_mem_regions, hnode) {
+ if (gfn >= region->start_gfn &&
+ gfn < region->start_gfn + region->nr_pages)
+ return region;
+ }
+
+ return NULL;
+}
+
+#ifdef CONFIG_X86_64
+static struct mshv_mem_region *
+mshv_partition_region_by_gfn_get(struct mshv_partition *p, u64 gfn)
+{
+ struct mshv_mem_region *region;
+
+ spin_lock(&p->pt_mem_regions_lock);
+ region = mshv_partition_region_by_gfn(p, gfn);
+ if (!region || !mshv_region_get(region)) {
+ spin_unlock(&p->pt_mem_regions_lock);
+ return NULL;
+ }
+ spin_unlock(&p->pt_mem_regions_lock);
+
+ return region;
+}
+
+/**
+ * mshv_handle_gpa_intercept - Handle GPA (Guest Physical Address) intercepts.
+ * @vp: Pointer to the virtual processor structure.
+ *
+ * This function processes GPA intercepts by identifying the memory region
+ * corresponding to the intercepted GPA, aligning the page offset, and
+ * mapping the required pages. It ensures that the region is valid and
+ * handles faults efficiently by mapping multiple pages at once.
+ *
+ * Return: true if the intercept was handled successfully, false otherwise.
+ */
+static bool mshv_handle_gpa_intercept(struct mshv_vp *vp)
+{
+ struct mshv_partition *p = vp->vp_partition;
+ struct mshv_mem_region *region;
+ struct hv_x64_memory_intercept_message *msg;
+ bool ret;
+ u64 gfn;
+
+ msg = (struct hv_x64_memory_intercept_message *)
+ vp->vp_intercept_msg_page->u.payload;
+
+ gfn = HVPFN_DOWN(msg->guest_physical_address);
+
+ region = mshv_partition_region_by_gfn_get(p, gfn);
+ if (!region)
+ return false;
+
+ /* Only movable memory ranges are supported for GPA intercepts */
+ if (region->type == MSHV_REGION_TYPE_MEM_MOVABLE)
+ ret = mshv_region_handle_gfn_fault(region, gfn);
+ else
+ ret = false;
+
+ mshv_region_put(region);
+
+ return ret;
+}
+#else /* CONFIG_X86_64 */
+static bool mshv_handle_gpa_intercept(struct mshv_vp *vp) { return false; }
+#endif /* CONFIG_X86_64 */
+
+static bool mshv_vp_handle_intercept(struct mshv_vp *vp)
+{
+ switch (vp->vp_intercept_msg_page->header.message_type) {
+ case HVMSG_GPA_INTERCEPT:
+ return mshv_handle_gpa_intercept(vp);
+ }
+ return false;
+}
+
static long mshv_vp_ioctl_run_vp(struct mshv_vp *vp, void __user *ret_msg)
{
long rc;
- if (hv_scheduler_type == HV_SCHEDULER_TYPE_ROOT)
- rc = mshv_run_vp_with_root_scheduler(vp);
- else
- rc = mshv_run_vp_with_hyp_scheduler(vp);
+ do {
+ if (hv_scheduler_type == HV_SCHEDULER_TYPE_ROOT)
+ rc = mshv_run_vp_with_root_scheduler(vp);
+ else
+ rc = mshv_run_vp_with_hyp_scheduler(vp);
+ } while (rc == 0 && mshv_vp_handle_intercept(vp));
if (rc)
return rc;
@@ -1059,20 +1143,6 @@ static void mshv_async_hvcall_handler(void *data, u64 *status)
*status = partition->async_hypercall_status;
}
-static struct mshv_mem_region *
-mshv_partition_region_by_gfn(struct mshv_partition *partition, u64 gfn)
-{
- struct mshv_mem_region *region;
-
- hlist_for_each_entry(region, &partition->pt_mem_regions, hnode) {
- if (gfn >= region->start_gfn &&
- gfn < region->start_gfn + region->nr_pages)
- return region;
- }
-
- return NULL;
-}
-
/*
* NB: caller checks and makes sure mem->size is page aligned
* Returns: 0 with regionpp updated on success, or -errno
@@ -1100,6 +1170,14 @@ static int mshv_partition_create_region(struct mshv_partition *partition,
if (IS_ERR(rg))
return PTR_ERR(rg);
+ if (is_mmio)
+ rg->type = MSHV_REGION_TYPE_MMIO;
+ else if (mshv_partition_encrypted(partition) ||
+ !mshv_region_movable_init(rg))
+ rg->type = MSHV_REGION_TYPE_MEM_PINNED;
+ else
+ rg->type = MSHV_REGION_TYPE_MEM_MOVABLE;
+
rg->partition = partition;
*regionpp = rg;
@@ -1215,11 +1293,28 @@ mshv_map_user_memory(struct mshv_partition *partition,
if (ret)
return ret;
- if (is_mmio)
- ret = hv_call_map_mmio_pages(partition->pt_id, mem.guest_pfn,
- mmio_pfn, HVPFN_DOWN(mem.size));
- else
+ switch (region->type) {
+ case MSHV_REGION_TYPE_MEM_PINNED:
ret = mshv_prepare_pinned_region(region);
+ break;
+ case MSHV_REGION_TYPE_MEM_MOVABLE:
+ /*
+ * For movable memory regions, remap with no access to let
+ * the hypervisor track dirty pages, enabling pre-copy live
+ * migration.
+ */
+ ret = hv_call_map_gpa_pages(partition->pt_id,
+ region->start_gfn,
+ region->nr_pages,
+ HV_MAP_GPA_NO_ACCESS, NULL);
+ break;
+ case MSHV_REGION_TYPE_MMIO:
+ ret = hv_call_map_mmio_pages(partition->pt_id,
+ region->start_gfn,
+ mmio_pfn,
+ region->nr_pages);
+ break;
+ }
if (ret)
goto errout;
^ permalink raw reply related
* [PATCH v7 6/7] Drivers: hv: Add refcount and locking to mem regions
From: Stanislav Kinsburskii @ 2025-11-26 2:09 UTC (permalink / raw)
To: kys, haiyangz, wei.liu, decui; +Cc: linux-hyperv, linux-kernel
In-Reply-To: <176412196000.447063.4256335030026363827.stgit@skinsburskii-cloud-desktop.internal.cloudapp.net>
Introduce kref-based reference counting and spinlock protection for
memory regions in Hyper-V partition management. This change improves
memory region lifecycle management and ensures thread-safe access to the
region list.
Also improves the check for overlapped memory regions during region
creation, preventing duplicate or conflicting mappings.
Previously, the regions list was protected by the partition mutex.
However, this approach is too heavy for frequent fault and invalidation
operations. Finer grained locking is now used to improve efficiency and
concurrency.
This is a precursor to supporting movable memory regions. Fault and
invalidation handling for movable regions will require safe traversal of
the region list and holding a region reference while performing
invalidation or fault operations.
Signed-off-by: Stanislav Kinsburskii <skinsburskii@linux.microsoft.com>
---
drivers/hv/mshv_regions.c | 19 ++++++++++++++++---
drivers/hv/mshv_root.h | 6 +++++-
drivers/hv/mshv_root_main.c | 34 ++++++++++++++++++++++++++--------
3 files changed, 47 insertions(+), 12 deletions(-)
diff --git a/drivers/hv/mshv_regions.c b/drivers/hv/mshv_regions.c
index d535d2e3e811..6450a7ed8493 100644
--- a/drivers/hv/mshv_regions.c
+++ b/drivers/hv/mshv_regions.c
@@ -7,6 +7,7 @@
* Authors: Microsoft Linux virtualization team
*/
+#include <linux/kref.h>
#include <linux/mm.h>
#include <linux/vmalloc.h>
@@ -154,6 +155,8 @@ struct mshv_mem_region *mshv_region_create(u64 guest_pfn, u64 nr_pages,
if (!is_mmio)
region->flags.range_pinned = true;
+ kref_init(®ion->refcount);
+
return region;
}
@@ -303,13 +306,13 @@ static int mshv_region_unmap(struct mshv_mem_region *region)
mshv_region_chunk_unmap);
}
-void mshv_region_destroy(struct mshv_mem_region *region)
+static void mshv_region_destroy(struct kref *ref)
{
+ struct mshv_mem_region *region =
+ container_of(ref, struct mshv_mem_region, refcount);
struct mshv_partition *partition = region->partition;
int ret;
- hlist_del(®ion->hnode);
-
if (mshv_partition_encrypted(partition)) {
ret = mshv_region_share(region);
if (ret) {
@@ -326,3 +329,13 @@ void mshv_region_destroy(struct mshv_mem_region *region)
vfree(region);
}
+
+void mshv_region_put(struct mshv_mem_region *region)
+{
+ kref_put(®ion->refcount, mshv_region_destroy);
+}
+
+int mshv_region_get(struct mshv_mem_region *region)
+{
+ return kref_get_unless_zero(®ion->refcount);
+}
diff --git a/drivers/hv/mshv_root.h b/drivers/hv/mshv_root.h
index ff3374f13691..4249534ba900 100644
--- a/drivers/hv/mshv_root.h
+++ b/drivers/hv/mshv_root.h
@@ -72,6 +72,7 @@ do { \
struct mshv_mem_region {
struct hlist_node hnode;
+ struct kref refcount;
u64 nr_pages;
u64 start_gfn;
u64 start_uaddr;
@@ -97,6 +98,8 @@ struct mshv_partition {
u64 pt_id;
refcount_t pt_ref_count;
struct mutex pt_mutex;
+
+ spinlock_t pt_mem_regions_lock;
struct hlist_head pt_mem_regions; // not ordered
u32 pt_vp_count;
@@ -319,6 +322,7 @@ int mshv_region_unshare(struct mshv_mem_region *region);
int mshv_region_map(struct mshv_mem_region *region);
void mshv_region_invalidate(struct mshv_mem_region *region);
int mshv_region_pin(struct mshv_mem_region *region);
-void mshv_region_destroy(struct mshv_mem_region *region);
+void mshv_region_put(struct mshv_mem_region *region);
+int mshv_region_get(struct mshv_mem_region *region);
#endif /* _MSHV_ROOT_H_ */
diff --git a/drivers/hv/mshv_root_main.c b/drivers/hv/mshv_root_main.c
index ae600b927f49..1ef2a28beb17 100644
--- a/drivers/hv/mshv_root_main.c
+++ b/drivers/hv/mshv_root_main.c
@@ -1086,9 +1086,13 @@ static int mshv_partition_create_region(struct mshv_partition *partition,
u64 nr_pages = HVPFN_DOWN(mem->size);
/* Reject overlapping regions */
+ spin_lock(&partition->pt_mem_regions_lock);
if (mshv_partition_region_by_gfn(partition, mem->guest_pfn) ||
- mshv_partition_region_by_gfn(partition, mem->guest_pfn + nr_pages - 1))
+ mshv_partition_region_by_gfn(partition, mem->guest_pfn + nr_pages - 1)) {
+ spin_unlock(&partition->pt_mem_regions_lock);
return -EEXIST;
+ }
+ spin_unlock(&partition->pt_mem_regions_lock);
rg = mshv_region_create(mem->guest_pfn, nr_pages,
mem->userspace_addr, mem->flags,
@@ -1220,8 +1224,9 @@ mshv_map_user_memory(struct mshv_partition *partition,
if (ret)
goto errout;
- /* Install the new region */
+ spin_lock(&partition->pt_mem_regions_lock);
hlist_add_head(®ion->hnode, &partition->pt_mem_regions);
+ spin_unlock(&partition->pt_mem_regions_lock);
return 0;
@@ -1240,17 +1245,27 @@ mshv_unmap_user_memory(struct mshv_partition *partition,
if (!(mem.flags & BIT(MSHV_SET_MEM_BIT_UNMAP)))
return -EINVAL;
+ spin_lock(&partition->pt_mem_regions_lock);
+
region = mshv_partition_region_by_gfn(partition, mem.guest_pfn);
- if (!region)
- return -EINVAL;
+ if (!region) {
+ spin_unlock(&partition->pt_mem_regions_lock);
+ return -ENOENT;
+ }
/* Paranoia check */
if (region->start_uaddr != mem.userspace_addr ||
region->start_gfn != mem.guest_pfn ||
- region->nr_pages != HVPFN_DOWN(mem.size))
+ region->nr_pages != HVPFN_DOWN(mem.size)) {
+ spin_unlock(&partition->pt_mem_regions_lock);
return -EINVAL;
+ }
+
+ hlist_del(®ion->hnode);
- mshv_region_destroy(region);
+ spin_unlock(&partition->pt_mem_regions_lock);
+
+ mshv_region_put(region);
return 0;
}
@@ -1653,8 +1668,10 @@ static void destroy_partition(struct mshv_partition *partition)
remove_partition(partition);
hlist_for_each_entry_safe(region, n, &partition->pt_mem_regions,
- hnode)
- mshv_region_destroy(region);
+ hnode) {
+ hlist_del(®ion->hnode);
+ mshv_region_put(region);
+ }
/* Withdraw and free all pages we deposited */
hv_call_withdraw_memory(U64_MAX, NUMA_NO_NODE, partition->pt_id);
@@ -1852,6 +1869,7 @@ mshv_ioctl_create_partition(void __user *user_arg, struct device *module_dev)
INIT_HLIST_HEAD(&partition->pt_devices);
+ spin_lock_init(&partition->pt_mem_regions_lock);
INIT_HLIST_HEAD(&partition->pt_mem_regions);
mshv_eventfd_init(partition);
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox