Instead of a 32-bit/64-bit dichotomy, check the MSI address against
msi_addr_mask.

This allows platforms with MSI doorbell above 32-bit address space to
work with devices without full 64-bit MSI address support, as long as
the doorbell is within addressable range of MSI of the device.

Signed-off-by: Vivian Wang <wangruikang@iscas.ac.cn>

---
v2: No changes
---
 drivers/pci/msi/msi.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/pci/msi/msi.c b/drivers/pci/msi/msi.c
index 48f5f03d1479..2ecbcd6c436a 100644
--- a/drivers/pci/msi/msi.c
+++ b/drivers/pci/msi/msi.c
@@ -321,14 +321,17 @@ static int msi_setup_msi_desc(struct pci_dev *dev, int nvec,
 static int msi_verify_entries(struct pci_dev *dev)
 {
 	struct msi_desc *entry;
+	u64 address;
 
 	if (dev->msi_addr_mask == DMA_BIT_MASK(64))
 		return 0;
 
 	msi_for_each_desc(entry, &dev->dev, MSI_DESC_ALL) {
-		if (entry->msg.address_hi) {
-			pci_err(dev, "arch assigned 64-bit MSI address %#x%08x but device only supports 32 bits\n",
-				entry->msg.address_hi, entry->msg.address_lo);
+		address = (u64)entry->msg.address_hi << 32 |
+			  entry->msg.address_lo;
+		if (address & ~dev->msi_addr_mask) {
+			pci_err(dev, "arch assigned 64-bit MSI address %llx above device MSI address mask %llx\n",
+				address, dev->msi_addr_mask);
 			break;
 		}
 	}

-- 
2.52.0

[+cc Thomas, thread at https://lore.kernel.org/r/20260121-pci-msi-addr-mask-v2-0-f42593168989@iscas.ac.cn]

On Wed, Jan 21, 2026 at 11:49:38AM +0800, Vivian Wang wrote:
> Instead of a 32-bit/64-bit dichotomy, check the MSI address against
> msi_addr_mask.
> 
> This allows platforms with MSI doorbell above 32-bit address space to
> work with devices without full 64-bit MSI address support, as long as
> the doorbell is within addressable range of MSI of the device.
> 
> Signed-off-by: Vivian Wang <wangruikang@iscas.ac.cn>
> 
> ---
> v2: No changes
> ---
>  drivers/pci/msi/msi.c | 9 ++++++---
>  1 file changed, 6 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/pci/msi/msi.c b/drivers/pci/msi/msi.c
> index 48f5f03d1479..2ecbcd6c436a 100644
> --- a/drivers/pci/msi/msi.c
> +++ b/drivers/pci/msi/msi.c
> @@ -321,14 +321,17 @@ static int msi_setup_msi_desc(struct pci_dev *dev, int nvec,
>  static int msi_verify_entries(struct pci_dev *dev)
>  {
>  	struct msi_desc *entry;
> +	u64 address;
>  
>  	if (dev->msi_addr_mask == DMA_BIT_MASK(64))
>  		return 0;
>  
>  	msi_for_each_desc(entry, &dev->dev, MSI_DESC_ALL) {
> -		if (entry->msg.address_hi) {
> -			pci_err(dev, "arch assigned 64-bit MSI address %#x%08x but device only supports 32 bits\n",
> -				entry->msg.address_hi, entry->msg.address_lo);
> +		address = (u64)entry->msg.address_hi << 32 |
> +			  entry->msg.address_lo;
> +		if (address & ~dev->msi_addr_mask) {
> +			pci_err(dev, "arch assigned 64-bit MSI address %llx above device MSI address mask %llx\n",

Use %#llx so it's clear these addresses are hex.  The previous message
did that, not sure why you dropped it.

> +				address, dev->msi_addr_mask);
>  			break;
>  		}
>  	}
> 
> -- 
> 2.52.0
>

On 1/22/26 07:38, Bjorn Helgaas wrote:
> [+cc Thomas, thread at https://lore.kernel.org/r/20260121-pci-msi-addr-mask-v2-0-f42593168989@iscas.ac.cn]
>
> On Wed, Jan 21, 2026 at 11:49:38AM +0800, Vivian Wang wrote:
>> Instead of a 32-bit/64-bit dichotomy, check the MSI address against
>> msi_addr_mask.
>>
>> This allows platforms with MSI doorbell above 32-bit address space to
>> work with devices without full 64-bit MSI address support, as long as
>> the doorbell is within addressable range of MSI of the device.
>>
>> Signed-off-by: Vivian Wang <wangruikang@iscas.ac.cn>
>>
>> ---
>> v2: No changes
>> ---
>>  drivers/pci/msi/msi.c | 9 ++++++---
>>  1 file changed, 6 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/pci/msi/msi.c b/drivers/pci/msi/msi.c
>> index 48f5f03d1479..2ecbcd6c436a 100644
>> --- a/drivers/pci/msi/msi.c
>> +++ b/drivers/pci/msi/msi.c
>> @@ -321,14 +321,17 @@ static int msi_setup_msi_desc(struct pci_dev *dev, int nvec,
>>  static int msi_verify_entries(struct pci_dev *dev)
>>  {
>>  	struct msi_desc *entry;
>> +	u64 address;
>>  
>>  	if (dev->msi_addr_mask == DMA_BIT_MASK(64))
>>  		return 0;
>>  
>>  	msi_for_each_desc(entry, &dev->dev, MSI_DESC_ALL) {
>> -		if (entry->msg.address_hi) {
>> -			pci_err(dev, "arch assigned 64-bit MSI address %#x%08x but device only supports 32 bits\n",
>> -				entry->msg.address_hi, entry->msg.address_lo);
>> +		address = (u64)entry->msg.address_hi << 32 |
>> +			  entry->msg.address_lo;
>> +		if (address & ~dev->msi_addr_mask) {
>> +			pci_err(dev, "arch assigned 64-bit MSI address %llx above device MSI address mask %llx\n",
> Use %#llx so it's clear these addresses are hex.  The previous message
> did that, not sure why you dropped it.

Thanks for catching. I misunderstood the purpose of the original message
formatting.

I will fix this in v3.

Vivian "dramforever" Wang