net: mana: Cap MSI-X vectors to the device MSI-X table size
Subsystems (5)
PATCH 1/1 net: mana: Cap MSI-X vectors to the device MSI-X table size
mana_gd_query_max_resources() sizes gc->num_msix_usable from resp.max_msix and the CPU count, but never from the device MSI-X table. On a 1792 vCPU M-series VM that yields 1793 while the table has 1024 entries, and mana_gd_setup_remaining_irqs() then walks indices 1..1792, running off the end of the region mapped by msix_map_region():
BUG: unable to handle page fault for address: ff8e347f8b99800c RIP: 0010:msix_prepare_msi_desc+0x7a/0x90 RAX: 0000000000004000 RBX: ff4330cb164ea780 RCX: ff8e347f8b998000 Call Trace: <TASK> __msi_domain_alloc_irqs+0x13a/0x440 msi_domain_alloc_irq_at+0x149/0x1b0 mana_gd_setup+0x351/0x890 mana_gd_probe+0x274/0x390 </TASK>
RAX is index 1024 * PCI_MSIX_ENTRY_SIZE, one entry past the table.
msi_insert_desc() does range check the index, but only against the MSI domain hwsize, which matches the table only for devices on an MSI parent domain. With a global PCI/MSI domain hwsize is MSI_XA_DOMAIN_SIZE, so nothing bounds the request.
Cap num_msix_usable with pci_msix_vec_count().
--- .../net/ethernet/microsoft/mana/gdma_main.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+)
drivers/net/ethernet/microsoft/mana/gdma_main.c +19 -0
@@ -182,6 +182,7 @@ static int mana_gd_query_max_resources(struct pci_dev *pdev)struct gdma_query_max_resources_resp resp = {};struct gdma_general_req req = {};unsigned int max_num_queues; +unsigned int msix_vec_count;u8 bm_hostmode;u16 num_ports;int err;
@@ -218,6 +219,24 @@ static int mana_gd_query_max_resources(struct pci_dev *pdev) gc->num_msix_usable =min (resp.max_msix,num_online_cpus () +1 ); } +/* MSI-X vectors are allocated by index into the device MSI-X table, so +* never ask for more than the table holds. It can be smaller than both +* resp.max_msix and the CPU count. +*/ + err =pci_msix_vec_count (pdev); +if (err <=0 ) { +dev_err (gc->dev,"Failed to query MSI-X table size: %d\n" , err); +return err <0 ? err : -ENOSPC ; + } + msix_vec_count = err; + +if (gc->num_msix_usable > msix_vec_count) { +dev_info (gc->dev, +"Limiting MSI-X vectors from %u to table size %u\n" , + gc->num_msix_usable, msix_vec_count); + gc->num_msix_usable = msix_vec_count; + } +if (gc->num_msix_usable <=1 )return -ENOSPC ;
base-commit: 746fc0787f616da418ffc04a110296fe95d53491
Signature · 1 line
2.43.0
Quoted 24 lines · Long Li
mana_gd_query_max_resources()sizesgc->num_msix_usablefrom resp.max_msix and the CPU count, but never from the device MSI-X table. On a 1792 vCPU M-series VM that yields 1793 while the table has 1024 entries, andmana_gd_setup_remaining_irqs()then walks indices 1..1792, running off the end of the region mapped bymsix_map_region():BUG: unable to handle page fault for address: ff8e347f8b99800c RIP: 0010:msix_prepare_msi_desc+0x7a/0x90 RAX: 0000000000004000 RBX: ff4330cb164ea780 RCX: ff8e347f8b998000 Call Trace: <TASK> __msi_domain_alloc_irqs+0x13a/0x440 msi_domain_alloc_irq_at+0x149/0x1b0 mana_gd_setup+0x351/0x890 mana_gd_probe+0x274/0x390 </TASK>RAX is index 1024 *
PCI_MSIX_ENTRY_SIZE, one entry past the table.
msi_insert_desc()does range check the index, but only against the MSI domain hwsize, which matches the table only for devices on an MSI parent domain. With a global PCI/MSI domain hwsize isMSI_XA_DOMAIN_SIZE, so nothing bounds the request.
Quoted 4 lines
Cap
num_msix_usablewithpci_msix_vec_count().Fixes 755391121038 ("net: mana: Allocate MSI-X vectors dynamically")Signed-off-by Long Li <[email protected]>
This patch was applied to netdev/net.git (main) by Jakub Kicinski <[email protected]>:
Hello:
This patch was applied to netdev/net.git (main)
by Jakub Kicinski <[email protected]>:
Quoted 17 lines · On Fri, 21 Aug 2026 11:37:36 -0700 you wrote:
mana_gd_query_max_resources()sizesgc->num_msix_usablefrom resp.max_msix and the CPU count, but never from the device MSI-X table. On a 1792 vCPU M-series VM that yields 1793 while the table has 1024 entries, andmana_gd_setup_remaining_irqs()then walks indices 1..1792, running off the end of the region mapped bymsix_map_region():BUG: unable to handle page fault for address: ff8e347f8b99800c RIP: 0010:msix_prepare_msi_desc+0x7a/0x90 RAX: 0000000000004000 RBX: ff4330cb164ea780 RCX: ff8e347f8b998000 Call Trace: <TASK> __msi_domain_alloc_irqs+0x13a/0x440 msi_domain_alloc_irq_at+0x149/0x1b0 mana_gd_setup+0x351/0x890 mana_gd_probe+0x274/0x390 </TASK>
Quoted 1 line
[...]
Here is the summary with links:
- [net] net: mana: Cap MSI-X vectors to the device MSI-X table size
https://git.kernel.org/netdev/net/c/2c7493f98014
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html