* [PATCH v10 01/18] PCI/P2PDMA: Document the TLP attribute assumptions
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:11 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 02/18] PCI/P2PDMA: Derive routing from directional ACS controls Leon Romanovsky
` (16 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
P2PDMA selects a mapping without receiving the Request's ordering or
Address Type attributes. Its ACS handles only strictly ordered Requests
carrying an Untranslated address.
Document that the result is not defined for Relaxed Ordering or
ATS-translated Requests because those TLP attributes can select different
routes through the fabric.
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
Documentation/driver-api/pci/p2pdma.rst | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/Documentation/driver-api/pci/p2pdma.rst b/Documentation/driver-api/pci/p2pdma.rst
index 63cff9e4d2c9..80f8fec9b0e9 100644
--- a/Documentation/driver-api/pci/p2pdma.rst
+++ b/Documentation/driver-api/pci/p2pdma.rst
@@ -15,6 +15,13 @@ then based on the ACS settings the transaction can route entirely within
the PCIe hierarchy and never reach the root port. The kernel will evaluate
the PCIe topology and always permit P2P in these well-defined cases.
+This evaluation assumes clients issue strictly ordered Requests carrying an
+Untranslated address. Its result is not defined when clients use Relaxed
+Ordering or issue ATS-translated Requests because those TLP attributes can
+select different routes through the fabric. Unless ACS Translation Blocking
+is enabled, a Port with ACS Direct Translated P2P enabled routes a
+Translated Request directly to the peer regardless of the redirect controls.
+
However, if the P2P transaction reaches the host bridge then it might have to
hairpin back out the same root port, be routed inside the CPU SOC to another
PCIe root port, or routed internally to the SOC.
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* [PATCH v10 02/18] PCI/P2PDMA: Derive routing from directional ACS controls
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
2026-10-12 13:07 ` [PATCH v10 01/18] PCI/P2PDMA: Document the TLP attribute assumptions Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:16 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 03/18] PCI: Reject unreadable ACS controls in isolation checks Leon Romanovsky
` (15 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
pci_bridge_has_acs_redir() treats Request and Completion Redirect as
interchangeable. On asymmetric fabrics, a control for only the reverse TLP
direction can unnecessarily force P2PDMA through the host bridge.
Evaluate Request Redirect for the Requester's Requests and Completion
Redirect for the Completer's read Completions. Continue treating enabled
Egress Control conservatively as a Request redirect.
Fixes: 52916982af48 ("PCI/P2PDMA: Support peer-to-peer memory")
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
drivers/pci/p2pdma.c | 75 ++++++++++++++++++++++++++++++++++++++++------------
1 file changed, 58 insertions(+), 17 deletions(-)
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index 45260423dfb0..db54865b0be9 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -21,6 +21,8 @@
#include <linux/seq_buf.h>
#include <linux/xarray.h>
+#include "pci.h"
+
struct pci_p2pdma {
struct gen_pool *pool;
bool p2pmem_published;
@@ -490,26 +492,56 @@ static struct pci_dev *find_parent_pci_dev(struct device *dev)
return NULL;
}
+enum pci_acs_p2pdma_state {
+ PCI_ACS_P2PDMA_DIRECT,
+ PCI_ACS_P2PDMA_REDIRECT,
+};
+
/*
- * Check if a PCI bridge has its ACS redirection bits set to redirect P2P
- * TLPs upstream via ACS. Returns 1 if the packets will be redirected
- * upstream, 0 otherwise.
+ * Decide how a peer-to-peer Request at an ACS-capable ingress port routes,
+ * from that port's ACS Control register.
+ *
+ * Linux does not read the Egress Control Vector, so Egress Control is treated
+ * conservatively as a redirect. Per PCIe r7.0 Table 6-11 the outcomes it
+ * selects are a direct route and an ACS Violation, and neither one lets peer
+ * bus addressing be assumed.
*/
-static int pci_bridge_has_acs_redir(struct pci_dev *pdev)
+static enum pci_acs_p2pdma_state
+pci_acs_p2pdma_request(u16 ctrl)
{
- int pos;
- u16 ctrl;
+ return ctrl & (PCI_ACS_RR | PCI_ACS_EC) ?
+ PCI_ACS_P2PDMA_REDIRECT : PCI_ACS_P2PDMA_DIRECT;
+}
- pos = pdev->acs_cap;
- if (!pos)
- return 0;
+/*
+ * Decide how a peer-to-peer Completion at an ACS-capable ingress port routes.
+ * PCIe r7.0 sec 6.12.1.1: no ACS control other than P2P Completion Redirect
+ * affects a Completion.
+ */
+static enum pci_acs_p2pdma_state
+pci_acs_p2pdma_completion(u16 ctrl)
+{
+ return ctrl & PCI_ACS_CR ? PCI_ACS_P2PDMA_REDIRECT :
+ PCI_ACS_P2PDMA_DIRECT;
+}
- pci_read_config_word(pdev, pos + PCI_ACS_CTRL, &ctrl);
+/*
+ * Read @pdev's ACS Control register. A device without an ACS capability has
+ * no peer-to-peer controls at all, which routes the same as having them all
+ * clear. Returns false when the register is present but cannot be read; @ctrl
+ * is then meaningless.
+ */
+static bool pci_acs_p2pdma_ctrl(struct pci_dev *pdev, u16 *ctrl)
+{
+ int pos;
- if (ctrl & (PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_EC))
- return 1;
+ pos = pdev->acs_cap;
+ if (!pos) {
+ *ctrl = 0;
+ return true;
+ }
- return 0;
+ return !pci_read_config_word(pdev, pos + PCI_ACS_CTRL, ctrl);
}
static void seq_buf_print_bus_devfn(struct seq_buf *buf, struct pci_dev *pdev)
@@ -704,6 +736,10 @@ static unsigned long map_types_idx(struct pci_dev *client)
* then to Device B. The mapping type returned depends on the ACS
* redirection setting of the ports along the path.
*
+ * The client is the Requester and the provider the Completer. Check Request
+ * Redirect on the Requester's path and Completion Redirect for read
+ * Completions on the Completer's path.
+ *
* If ACS redirect is set on any port in the path, traffic between the
* devices will go through the host bridge, so return
* PCI_P2PDMA_MAP_THRU_HOST_BRIDGE; otherwise return
@@ -727,6 +763,7 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
int dist_a = 0;
int dist_b = 0;
char buf[128];
+ u16 ctrl;
seq_buf_init(&acs_list, buf, sizeof(buf));
@@ -738,7 +775,9 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
while (a) {
dist_b = 0;
- if (pci_bridge_has_acs_redir(a)) {
+ if (!pci_acs_p2pdma_ctrl(a, &ctrl) ||
+ pci_acs_p2pdma_completion(ctrl) ==
+ PCI_ACS_P2PDMA_REDIRECT) {
seq_buf_print_bus_devfn(&acs_list, a);
acs_cnt++;
}
@@ -767,7 +806,9 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
if (a == bb)
break;
- if (pci_bridge_has_acs_redir(bb)) {
+ if (!pci_acs_p2pdma_ctrl(bb, &ctrl) ||
+ pci_acs_p2pdma_request(ctrl) ==
+ PCI_ACS_P2PDMA_REDIRECT) {
seq_buf_print_bus_devfn(&acs_list, bb);
acs_cnt++;
}
@@ -1116,10 +1157,10 @@ EXPORT_SYMBOL_GPL(pci_p2pmem_publish);
/**
* pci_p2pdma_map_type - Determine the mapping type for P2PDMA transfers
* @provider: P2PDMA provider structure
- * @dev: Target device for the transfer
+ * @dev: Client device that initiates the transfer
*
* Determines how peer-to-peer DMA transfers should be mapped between
- * the provider and the target device. The mapping type indicates whether
+ * the provider and the client device. The mapping type indicates whether
* the transfer can be done directly through PCI switches or must go
* through the host bridge.
*/
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* [PATCH v10 03/18] PCI: Reject unreadable ACS controls in isolation checks
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
2026-10-12 13:07 ` [PATCH v10 01/18] PCI/P2PDMA: Document the TLP attribute assumptions Leon Romanovsky
2026-10-12 13:07 ` [PATCH v10 02/18] PCI/P2PDMA: Derive routing from directional ACS controls Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:20 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 04/18] PCI/P2PDMA: Evaluate ACS controls at the path divergence Leon Romanovsky
` (14 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
pci_acs_flags_enabled() and the Intel SPT PCH quirk use ACS registers
without checking config-space read errors. A failed read may leave control
state indeterminate yet allow the device to satisfy requested isolation
controls.
Return false when either ACS capability or control state cannot be read.
An unknown state cannot prove isolation.
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
drivers/pci/pci.c | 3 ++-
drivers/pci/quirks.c | 6 ++++--
2 files changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index b2879a6be5f8..f7d94ecf9157 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -3594,7 +3594,8 @@ static bool pci_acs_flags_enabled(struct pci_dev *pdev, u16 acs_flags)
*/
acs_flags &= (pdev->acs_capabilities | PCI_ACS_EC);
- pci_read_config_word(pdev, pos + PCI_ACS_CTRL, &ctrl);
+ if (pci_read_config_word(pdev, pos + PCI_ACS_CTRL, &ctrl))
+ return false;
return (ctrl & acs_flags) == acs_flags;
}
diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index de9bbccda21f..d5c3e6802840 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -4992,10 +4992,12 @@ static int pci_quirk_intel_spt_pch_acs(struct pci_dev *dev, u16 acs_flags)
return -ENOTTY;
/* see pci_acs_flags_enabled() */
- pci_read_config_dword(dev, pos + PCI_ACS_CAP, &cap);
+ if (pci_read_config_dword(dev, pos + PCI_ACS_CAP, &cap))
+ return 0;
acs_flags &= (cap | PCI_ACS_EC);
- pci_read_config_dword(dev, pos + INTEL_SPT_ACS_CTRL, &ctrl);
+ if (pci_read_config_dword(dev, pos + INTEL_SPT_ACS_CTRL, &ctrl))
+ return 0;
return pci_acs_ctrl_enabled(acs_flags, ctrl);
}
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* [PATCH v10 04/18] PCI/P2PDMA: Evaluate ACS controls at the path divergence
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (2 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 03/18] PCI: Reject unreadable ACS controls in isolation checks Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:15 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 05/18] PCI/P2PDMA: Document directional ACS routing Leon Romanovsky
` (13 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
ACS redirect controls choose between peer and upstream routes only at the
path divergence. Applying them below that point rejects valid nested
topologies because traffic already has only an upstream route.
Host bridge 1 Host bridge 2
| |
PCIe switch 1 PCIe switch 2
| | |
A B C
A's and B's paths diverge in switch 1. A and C share no upstream bridge,
so no direct route exists for ACS to gate.
Evaluate Request controls on the Requester-side divergence port and
Completion Redirect on the Completer-side port and reject an unreadable ACS
Control register.
Fixes: 52916982af48 ("PCI/P2PDMA: Support peer-to-peer memory")
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
drivers/pci/p2pdma.c | 102 +++++++++++++++++++++++++++++----------------
include/linux/pci-p2pdma.h | 8 ++--
2 files changed, 71 insertions(+), 39 deletions(-)
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index db54865b0be9..8af11a8449f1 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -493,6 +493,7 @@ static struct pci_dev *find_parent_pci_dev(struct device *dev)
}
enum pci_acs_p2pdma_state {
+ PCI_ACS_P2PDMA_NOT_SUPPORTED,
PCI_ACS_P2PDMA_DIRECT,
PCI_ACS_P2PDMA_REDIRECT,
};
@@ -736,13 +737,14 @@ static unsigned long map_types_idx(struct pci_dev *client)
* then to Device B. The mapping type returned depends on the ACS
* redirection setting of the ports along the path.
*
- * The client is the Requester and the provider the Completer. Check Request
- * Redirect on the Requester's path and Completion Redirect for read
- * Completions on the Completer's path.
+ * The client is the Requester and the provider the Completer. At the path
+ * divergence, check Request Redirect and Egress Control on the Requester-side
+ * port, and Completion Redirect for read Completions on the Completer-side
+ * port.
*
- * If ACS redirect is set on any port in the path, traffic between the
- * devices will go through the host bridge, so return
- * PCI_P2PDMA_MAP_THRU_HOST_BRIDGE; otherwise return
+ * If ACS redirects traffic at either divergence port, return
+ * PCI_P2PDMA_MAP_THRU_HOST_BRIDGE. If the ACS Control register cannot be
+ * read, return PCI_P2PDMA_MAP_NOT_SUPPORTED. Otherwise, return
* PCI_P2PDMA_MAP_BUS_ADDR.
*
* Any two devices that have a data path that goes through the host bridge
@@ -756,10 +758,13 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
int *dist, bool verbose)
{
enum pci_p2pdma_map_type map_type = PCI_P2PDMA_MAP_THRU_HOST_BRIDGE;
+ enum pci_acs_p2pdma_state state = PCI_ACS_P2PDMA_NOT_SUPPORTED;
struct pci_dev *a = provider, *b = client, *bb;
+ struct pci_dev *a_child = NULL, *b_child = NULL;
+ struct pci_dev *acs_unreadable = NULL;
struct pci_p2pdma *p2pdma;
struct seq_buf acs_list;
- int acs_cnt = 0;
+ int acs_redirect_cnt = 0;
int dist_a = 0;
int dist_b = 0;
char buf[128];
@@ -774,51 +779,67 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
*/
while (a) {
dist_b = 0;
-
- if (!pci_acs_p2pdma_ctrl(a, &ctrl) ||
- pci_acs_p2pdma_completion(ctrl) ==
- PCI_ACS_P2PDMA_REDIRECT) {
- seq_buf_print_bus_devfn(&acs_list, a);
- acs_cnt++;
- }
-
+ b_child = NULL;
bb = b;
while (bb) {
if (a == bb)
- goto check_b_path_acs;
+ goto check_paths_acs;
+ b_child = bb;
bb = pci_upstream_bridge(bb);
dist_b++;
}
+ a_child = a;
a = pci_upstream_bridge(a);
dist_a++;
}
+ /*
+ * The paths share no upstream bridge, so there is no direct path for
+ * ACS to gate: PCI_P2PDMA_MAP_BUS_ADDR is not reachable here and the
+ * request can only get to the peer through the host bridge.
+ */
*dist = dist_a + dist_b;
goto map_through_host_bridge;
-check_b_path_acs:
- bb = b;
-
- while (bb) {
- if (a == bb)
- break;
+check_paths_acs:
+ *dist = dist_a + dist_b;
- if (!pci_acs_p2pdma_ctrl(bb, &ctrl) ||
- pci_acs_p2pdma_request(ctrl) ==
- PCI_ACS_P2PDMA_REDIRECT) {
- seq_buf_print_bus_devfn(&acs_list, bb);
- acs_cnt++;
+ /*
+ * ACS P2P routing controls apply where a TLP can route toward the peer
+ * or upstream. Below that divergence, its only route toward the other
+ * branch is upstream, so redirect controls do not affect the path.
+ */
+ if (a_child && b_child) {
+ if (pci_acs_p2pdma_ctrl(a_child, &ctrl))
+ state = pci_acs_p2pdma_completion(ctrl);
+ if (state != PCI_ACS_P2PDMA_DIRECT) {
+ seq_buf_print_bus_devfn(&acs_list, a_child);
+ if (state == PCI_ACS_P2PDMA_REDIRECT)
+ acs_redirect_cnt++;
+ else if (!acs_unreadable)
+ acs_unreadable = a_child;
}
- bb = pci_upstream_bridge(bb);
+ state = PCI_ACS_P2PDMA_NOT_SUPPORTED;
+ if (pci_acs_p2pdma_ctrl(b_child, &ctrl))
+ state = pci_acs_p2pdma_request(ctrl);
+ if (state != PCI_ACS_P2PDMA_DIRECT) {
+ seq_buf_print_bus_devfn(&acs_list, b_child);
+ if (state == PCI_ACS_P2PDMA_REDIRECT)
+ acs_redirect_cnt++;
+ else if (!acs_unreadable)
+ acs_unreadable = b_child;
+ }
}
- *dist = dist_a + dist_b;
-
- if (!acs_cnt) {
+ /*
+ * Below a shared upstream bridge, a path whose divergence ports do not
+ * redirect routes the request directly.
+ */
+ if (!acs_unreadable && !acs_redirect_cnt) {
map_type = PCI_P2PDMA_MAP_BUS_ADDR;
goto done;
}
@@ -827,10 +848,21 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
/* Drop the final semicolon; the list is not empty here */
if (!seq_buf_has_overflowed(&acs_list))
acs_list.buffer[acs_list.len - 1] = '\0';
- pci_warn(client, "ACS redirect is set between the client and provider (%s)\n",
- pci_name(provider));
- pci_warn(client, "to disable ACS redirect for this path, add the kernel parameter: pci=disable_acs_redir=%s\n",
- seq_buf_str(&acs_list));
+ if (acs_unreadable)
+ pci_warn(client, "ACS Control is unreadable for provider %s at %s\n",
+ pci_name(provider), pci_name(acs_unreadable));
+ else {
+ pci_warn(client, "ACS redirect is set between the client and provider (%s)\n",
+ pci_name(provider));
+ pci_warn(client, "to disable ACS controls for this path, add the kernel parameter: pci=disable_acs_redir=%s\n",
+ seq_buf_str(&acs_list));
+ }
+ }
+
+ /* An unreadable control does not establish an upstream redirect. */
+ if (acs_unreadable) {
+ map_type = PCI_P2PDMA_MAP_NOT_SUPPORTED;
+ goto done;
}
map_through_host_bridge:
diff --git a/include/linux/pci-p2pdma.h b/include/linux/pci-p2pdma.h
index 873de20a2247..dd17501ba1b6 100644
--- a/include/linux/pci-p2pdma.h
+++ b/include/linux/pci-p2pdma.h
@@ -42,10 +42,10 @@ enum pci_p2pdma_map_type {
PCI_P2PDMA_MAP_NONE,
/*
- * PCI_P2PDMA_MAP_NOT_SUPPORTED: Indicates the transaction will
- * traverse the host bridge and the host bridge is not in the
- * allowlist. DMA Mapping routines should return an error when
- * this is returned.
+ * PCI_P2PDMA_MAP_NOT_SUPPORTED: Indicates no safe mapping is available,
+ * for example because ACS blocks the direct path or the required host
+ * bridge is not in the allowlist. DMA Mapping routines should return an
+ * error when this is returned.
*/
PCI_P2PDMA_MAP_NOT_SUPPORTED,
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* [PATCH v10 05/18] PCI/P2PDMA: Document directional ACS routing
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (3 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 04/18] PCI/P2PDMA: Evaluate ACS controls at the path divergence Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:15 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 06/18] PCI/P2PDMA: Collect the path's ACS controls before deciding Leon Romanovsky
` (12 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
P2PDMA documentation describes ACS controls as path-wide, although Request
and Completion controls apply to different transaction directions and only
affect peer-versus-upstream decisions at the path divergence.
Document the client as the Requester and the provider as the Completer,
the divergence port checked for each TLP direction, and the conservative
handling of unreadable ACS state. Clarify which controls disable_acs_redir
changes.
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
Documentation/admin-guide/kernel-parameters.txt | 11 ++++++-----
Documentation/driver-api/pci/p2pdma.rst | 18 ++++++++++++++++++
2 files changed, 24 insertions(+), 5 deletions(-)
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 68647ff4bdd2..8507ac79a919 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -5292,11 +5292,12 @@ Kernel parameters
Specify one or more PCI devices (in the format
specified above) separated by semicolons.
Each device specified will have the PCI ACS
- redirect capabilities forced off which will
- allow P2P traffic between devices through
- bridges without forcing it upstream. Note:
- this removes isolation between devices and
- may put more devices in an IOMMU group.
+ P2P Request Redirect, Completion Redirect, and
+ Egress Control features forced off. This may
+ allow P2P traffic through bridges that would
+ otherwise be redirected upstream. Note: this
+ removes isolation between devices and may
+ cause affected devices to share an IOMMU group.
config_acs=
Format:
<ACS flags>@<pci_dev>[; ...]
diff --git a/Documentation/driver-api/pci/p2pdma.rst b/Documentation/driver-api/pci/p2pdma.rst
index 80f8fec9b0e9..40f5268ba9db 100644
--- a/Documentation/driver-api/pci/p2pdma.rst
+++ b/Documentation/driver-api/pci/p2pdma.rst
@@ -15,6 +15,24 @@ then based on the ACS settings the transaction can route entirely within
the PCIe hierarchy and never reach the root port. The kernel will evaluate
the PCIe topology and always permit P2P in these well-defined cases.
+The client is the PCIe Requester and the provider is the Completer, whether
+the client reads or writes provider memory. If both are below the same Root
+Port, their paths toward it first meet in a Switch, where the paths diverge.
+A Request enters that Switch through the Downstream Port above the
+Requester, the Requester-side port. If ACS P2P Request Redirect is enabled
+there, the Request is routed upstream; otherwise it is routed back
+downstream through the Downstream Port leading to the Completer, the
+Completer-side port. The path of Completions from a read likewise depends on
+ACS P2P Completion Redirect in the Completer-side port. An enabled Egress
+Control is conservatively treated as a Request redirect.
+
+Below the divergence, the route toward the other branch is already upstream,
+so those P2P redirect controls do not affect it. Redirect controls for the
+reverse transaction directions do not affect the mapping. P2P DMA is routed
+through the host bridge when either applicable port redirects. If an ACS
+Control register cannot be read, P2P DMA is rejected because the kernel cannot
+establish a usable route.
+
This evaluation assumes clients issue strictly ordered Requests carrying an
Untranslated address. Its result is not defined when clients use Relaxed
Ordering or issue ATS-translated Requests because those TLP attributes can
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* [PATCH v10 06/18] PCI/P2PDMA: Collect the path's ACS controls before deciding
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (4 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 05/18] PCI/P2PDMA: Document directional ACS routing Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:15 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 07/18] PCI/P2PDMA: Answer routing per TLP class Leon Romanovsky
` (11 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
calc_map_type_and_dist() reads each divergence port's ACS Control register
and folds the result into running counters as it goes. Any routing property
that depends on the kind of TLP being routed would have to be threaded
through that code, so there is nowhere to put one without reading the
registers again for each kind.
Collect the two ports' ACS Control values into struct pci_p2pdma_acs_path
first, then decide from it. pci_p2pdma_route() applies the same rule as
before: a path routes directly only when both directions do.
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
drivers/pci/p2pdma.c | 148 +++++++++++++++++++++++++++++++++------------------
1 file changed, 96 insertions(+), 52 deletions(-)
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index 8af11a8449f1..ced177bf4c11 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -553,6 +553,80 @@ static void seq_buf_print_bus_devfn(struct seq_buf *buf, struct pci_dev *pdev)
seq_buf_printf(buf, "%s;", pci_name(pdev));
}
+/*
+ * What the topology walk found out about one provider/client path. Producing
+ * this costs a walk and one config read per divergence port, none of which
+ * depends on the TLP being routed.
+ *
+ * @req_ctrl: ACS Control of the Requester-side divergence port. That is the
+ * first port at which a Request can route toward the peer rather
+ * than upstream, so it is where the Request controls apply.
+ * @cpl_ctrl: ACS Control of the Completer-side divergence port, likewise for
+ * the Completions travelling back.
+ * @unreadable: First port whose ACS Control could not be read, if any.
+ */
+struct pci_p2pdma_acs_path {
+ u16 req_ctrl;
+ u16 cpl_ctrl;
+ struct pci_dev *unreadable;
+};
+
+/*
+ * Combine both directions into a mapping type. Only a path that routes the
+ * Request and the Completions it generates directly can be programmed with
+ * the peer's bus addresses.
+ */
+static enum pci_p2pdma_map_type
+pci_p2pdma_route(const struct pci_p2pdma_acs_path *path)
+{
+ if (path->unreadable)
+ return PCI_P2PDMA_MAP_NOT_SUPPORTED;
+
+ if (pci_acs_p2pdma_request(path->req_ctrl) == PCI_ACS_P2PDMA_DIRECT &&
+ pci_acs_p2pdma_completion(path->cpl_ctrl) == PCI_ACS_P2PDMA_DIRECT)
+ return PCI_P2PDMA_MAP_BUS_ADDR;
+
+ return PCI_P2PDMA_MAP_THRU_HOST_BRIDGE;
+}
+
+/*
+ * Name the ports that keep this path off a direct route, so that the admin
+ * can hand them to pci=disable_acs_redir=.
+ */
+static void pci_p2pdma_warn_path(struct pci_dev *client,
+ struct pci_dev *provider,
+ const struct pci_p2pdma_acs_path *path,
+ struct pci_dev *a_child,
+ struct pci_dev *b_child)
+{
+ struct seq_buf acs_list;
+ char buf[128];
+
+ if (path->unreadable) {
+ pci_warn(client,
+ "ACS Control is unreadable for provider %s at %s\n",
+ pci_name(provider), pci_name(path->unreadable));
+ return;
+ }
+
+ seq_buf_init(&acs_list, buf, sizeof(buf));
+ if (pci_acs_p2pdma_completion(path->cpl_ctrl) != PCI_ACS_P2PDMA_DIRECT)
+ seq_buf_print_bus_devfn(&acs_list, a_child);
+ if (pci_acs_p2pdma_request(path->req_ctrl) != PCI_ACS_P2PDMA_DIRECT)
+ seq_buf_print_bus_devfn(&acs_list, b_child);
+
+ /* Drop the final semicolon; the list is not empty here */
+ if (!seq_buf_has_overflowed(&acs_list))
+ acs_list.buffer[acs_list.len - 1] = '\0';
+
+ pci_warn(client,
+ "ACS redirect is set between the client and provider (%s)\n",
+ pci_name(provider));
+ pci_warn(client,
+ "to disable ACS controls for this path, add the kernel parameter: pci=disable_acs_redir=%s\n",
+ seq_buf_str(&acs_list));
+}
+
static bool cpu_supports_p2pdma(void)
{
#ifdef CONFIG_X86
@@ -758,19 +832,13 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
int *dist, bool verbose)
{
enum pci_p2pdma_map_type map_type = PCI_P2PDMA_MAP_THRU_HOST_BRIDGE;
- enum pci_acs_p2pdma_state state = PCI_ACS_P2PDMA_NOT_SUPPORTED;
struct pci_dev *a = provider, *b = client, *bb;
struct pci_dev *a_child = NULL, *b_child = NULL;
- struct pci_dev *acs_unreadable = NULL;
+ struct pci_p2pdma_acs_path path = {};
struct pci_p2pdma *p2pdma;
- struct seq_buf acs_list;
- int acs_redirect_cnt = 0;
+ bool cpu_p2pdma, host_whitelisted = false;
int dist_a = 0;
int dist_b = 0;
- char buf[128];
- u16 ctrl;
-
- seq_buf_init(&acs_list, buf, sizeof(buf));
/*
* Note, we don't need to take references to devices returned by
@@ -813,61 +881,35 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
* branch is upstream, so redirect controls do not affect the path.
*/
if (a_child && b_child) {
- if (pci_acs_p2pdma_ctrl(a_child, &ctrl))
- state = pci_acs_p2pdma_completion(ctrl);
- if (state != PCI_ACS_P2PDMA_DIRECT) {
- seq_buf_print_bus_devfn(&acs_list, a_child);
- if (state == PCI_ACS_P2PDMA_REDIRECT)
- acs_redirect_cnt++;
- else if (!acs_unreadable)
- acs_unreadable = a_child;
- }
-
- state = PCI_ACS_P2PDMA_NOT_SUPPORTED;
- if (pci_acs_p2pdma_ctrl(b_child, &ctrl))
- state = pci_acs_p2pdma_request(ctrl);
- if (state != PCI_ACS_P2PDMA_DIRECT) {
- seq_buf_print_bus_devfn(&acs_list, b_child);
- if (state == PCI_ACS_P2PDMA_REDIRECT)
- acs_redirect_cnt++;
- else if (!acs_unreadable)
- acs_unreadable = b_child;
- }
+ if (!pci_acs_p2pdma_ctrl(a_child, &path.cpl_ctrl))
+ path.unreadable = a_child;
+ if (!pci_acs_p2pdma_ctrl(b_child, &path.req_ctrl) &&
+ !path.unreadable)
+ path.unreadable = b_child;
}
/*
* Below a shared upstream bridge, a path whose divergence ports do not
* redirect routes the request directly.
*/
- if (!acs_unreadable && !acs_redirect_cnt) {
- map_type = PCI_P2PDMA_MAP_BUS_ADDR;
+ map_type = pci_p2pdma_route(&path);
+ if (map_type == PCI_P2PDMA_MAP_BUS_ADDR)
goto done;
- }
- if (verbose) {
- /* Drop the final semicolon; the list is not empty here */
- if (!seq_buf_has_overflowed(&acs_list))
- acs_list.buffer[acs_list.len - 1] = '\0';
- if (acs_unreadable)
- pci_warn(client, "ACS Control is unreadable for provider %s at %s\n",
- pci_name(provider), pci_name(acs_unreadable));
- else {
- pci_warn(client, "ACS redirect is set between the client and provider (%s)\n",
- pci_name(provider));
- pci_warn(client, "to disable ACS controls for this path, add the kernel parameter: pci=disable_acs_redir=%s\n",
- seq_buf_str(&acs_list));
- }
- }
+ if (verbose)
+ pci_p2pdma_warn_path(client, provider, &path, a_child, b_child);
/* An unreadable control does not establish an upstream redirect. */
- if (acs_unreadable) {
- map_type = PCI_P2PDMA_MAP_NOT_SUPPORTED;
+ if (path.unreadable)
goto done;
- }
map_through_host_bridge:
- if (!cpu_supports_p2pdma() &&
- !host_bridge_whitelist(provider, client, verbose)) {
+ cpu_p2pdma = cpu_supports_p2pdma();
+ if (!cpu_p2pdma)
+ host_whitelisted = host_bridge_whitelist(provider, client,
+ verbose);
+
+ if (!cpu_p2pdma && !host_whitelisted) {
if (verbose)
pci_warn(client, "cannot be used for peer-to-peer DMA as the client and provider (%s) do not share an upstream bridge or whitelisted host bridge\n",
pci_name(provider));
@@ -1201,8 +1243,9 @@ enum pci_p2pdma_map_type pci_p2pdma_map_type(struct p2pdma_provider *provider,
{
enum pci_p2pdma_map_type type = PCI_P2PDMA_MAP_NOT_SUPPORTED;
struct pci_dev *pdev = to_pci_dev(provider->owner);
- struct pci_dev *client;
struct pci_p2pdma *p2pdma;
+ unsigned long cache_index;
+ struct pci_dev *client;
int dist;
if (!pdev->p2pdma)
@@ -1212,13 +1255,14 @@ enum pci_p2pdma_map_type pci_p2pdma_map_type(struct p2pdma_provider *provider,
return PCI_P2PDMA_MAP_NOT_SUPPORTED;
client = to_pci_dev(dev);
+ cache_index = map_types_idx(client);
rcu_read_lock();
p2pdma = rcu_dereference(pdev->p2pdma);
if (p2pdma)
type = xa_to_value(xa_load(&p2pdma->map_types,
- map_types_idx(client)));
+ cache_index));
rcu_read_unlock();
if (type == PCI_P2PDMA_MAP_UNKNOWN)
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* [PATCH v10 07/18] PCI/P2PDMA: Answer routing per TLP class
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (5 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 06/18] PCI/P2PDMA: Collect the path's ACS controls before deciding Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:18 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 08/18] PCI/P2PDMA: Route Relaxed Ordering Completions directly Leon Romanovsky
` (10 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
calc_map_type_and_dist() returns one mapping type per provider and client,
valid only for strictly ordered Requests carrying an Untranslated address.
Clients that use Relaxed Ordering or ATS cannot ask what the fabric would
do with their traffic.
Add enum pci_p2pdma_tlp_flags to name a class and pci_p2pdma_map_type_tlp()
to ask about one. The topology walk and the ACS Control reads do not depend
on the class, so decide all of them from the one walk and cache them
together, four bits each. Every class still answers alike; the controls
that tell them apart come next.
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
drivers/pci/p2pdma.c | 157 +++++++++++++++++++++++++++++++++++++++++----------
1 file changed, 126 insertions(+), 31 deletions(-)
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index ced177bf4c11..479bf58104ca 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -492,6 +492,31 @@ static struct pci_dev *find_parent_pci_dev(struct device *dev)
return NULL;
}
+/**
+ * enum pci_p2pdma_tlp_flags - Properties of the TLPs a client will issue
+ *
+ * These describe the traffic rather than the topology, and select which ACS
+ * controls apply along the peer-to-peer path. A value of 0 means strictly
+ * ordered Requests carrying an Untranslated address.
+ *
+ * @PCI_P2PDMA_TLP_TRANSLATED: Requests carry an ATS Translated address. PCIe
+ * r7.0 sec 6.12.3 routes those to the peer regardless of ACS P2P Request
+ * Redirect and ACS P2P Egress Control wherever ACS Direct Translated P2P
+ * is enabled.
+ * @PCI_P2PDMA_TLP_RELAXED_CPL: Non-Posted Requests carry the Relaxed
+ * Ordering attribute, and PCIe r7.0 sec 2.2.9.1 requires the Completer to
+ * return it in their Completions. Sec 6.12.1.1 never redirects those, so
+ * ACS P2P Completion Redirect does not gate the path.
+ */
+enum pci_p2pdma_tlp_flags {
+ PCI_P2PDMA_TLP_TRANSLATED = 1 << 0,
+ PCI_P2PDMA_TLP_RELAXED_CPL = 1 << 1,
+};
+
+/* Every combination of the flags above selects one routing class. */
+#define PCI_P2PDMA_TLP_CLASSES \
+ ((PCI_P2PDMA_TLP_TRANSLATED | PCI_P2PDMA_TLP_RELAXED_CPL) + 1)
+
enum pci_acs_p2pdma_state {
PCI_ACS_P2PDMA_NOT_SUPPORTED,
PCI_ACS_P2PDMA_DIRECT,
@@ -508,7 +533,7 @@ enum pci_acs_p2pdma_state {
* bus addressing be assumed.
*/
static enum pci_acs_p2pdma_state
-pci_acs_p2pdma_request(u16 ctrl)
+pci_acs_p2pdma_request(u16 ctrl, unsigned int tlp_flags)
{
return ctrl & (PCI_ACS_RR | PCI_ACS_EC) ?
PCI_ACS_P2PDMA_REDIRECT : PCI_ACS_P2PDMA_DIRECT;
@@ -520,7 +545,7 @@ pci_acs_p2pdma_request(u16 ctrl)
* affects a Completion.
*/
static enum pci_acs_p2pdma_state
-pci_acs_p2pdma_completion(u16 ctrl)
+pci_acs_p2pdma_completion(u16 ctrl, unsigned int tlp_flags)
{
return ctrl & PCI_ACS_CR ? PCI_ACS_P2PDMA_REDIRECT :
PCI_ACS_P2PDMA_DIRECT;
@@ -577,13 +602,16 @@ struct pci_p2pdma_acs_path {
* the peer's bus addresses.
*/
static enum pci_p2pdma_map_type
-pci_p2pdma_route(const struct pci_p2pdma_acs_path *path)
+pci_p2pdma_route(const struct pci_p2pdma_acs_path *path,
+ unsigned int tlp_flags)
{
if (path->unreadable)
return PCI_P2PDMA_MAP_NOT_SUPPORTED;
- if (pci_acs_p2pdma_request(path->req_ctrl) == PCI_ACS_P2PDMA_DIRECT &&
- pci_acs_p2pdma_completion(path->cpl_ctrl) == PCI_ACS_P2PDMA_DIRECT)
+ if (pci_acs_p2pdma_request(path->req_ctrl, tlp_flags) ==
+ PCI_ACS_P2PDMA_DIRECT &&
+ pci_acs_p2pdma_completion(path->cpl_ctrl, tlp_flags) ==
+ PCI_ACS_P2PDMA_DIRECT)
return PCI_P2PDMA_MAP_BUS_ADDR;
return PCI_P2PDMA_MAP_THRU_HOST_BRIDGE;
@@ -591,7 +619,9 @@ pci_p2pdma_route(const struct pci_p2pdma_acs_path *path)
/*
* Name the ports that keep this path off a direct route, so that the admin
- * can hand them to pci=disable_acs_redir=.
+ * can hand them to pci=disable_acs_redir=. That parameter clears the
+ * redirect controls, which decide the default class, so that is the class
+ * this reports on.
*/
static void pci_p2pdma_warn_path(struct pci_dev *client,
struct pci_dev *provider,
@@ -610,9 +640,10 @@ static void pci_p2pdma_warn_path(struct pci_dev *client,
}
seq_buf_init(&acs_list, buf, sizeof(buf));
- if (pci_acs_p2pdma_completion(path->cpl_ctrl) != PCI_ACS_P2PDMA_DIRECT)
+ if (pci_acs_p2pdma_completion(path->cpl_ctrl, 0) !=
+ PCI_ACS_P2PDMA_DIRECT)
seq_buf_print_bus_devfn(&acs_list, a_child);
- if (pci_acs_p2pdma_request(path->req_ctrl) != PCI_ACS_P2PDMA_DIRECT)
+ if (pci_acs_p2pdma_request(path->req_ctrl, 0) != PCI_ACS_P2PDMA_DIRECT)
seq_buf_print_bus_devfn(&acs_list, b_child);
/* Drop the final semicolon; the list is not empty here */
@@ -786,6 +817,31 @@ static unsigned long map_types_idx(struct pci_dev *client)
return (pci_domain_nr(client->bus) << 16) | pci_dev_id(client);
}
+/*
+ * One cache entry holds the routing of every TLP class, four bits each,
+ * indexed by the &enum pci_p2pdma_tlp_flags combination that selects it. An
+ * absent entry reads back as PCI_P2PDMA_MAP_UNKNOWN in every class.
+ */
+static_assert(PCI_P2PDMA_MAP_THRU_HOST_BRIDGE < 16);
+
+static unsigned long
+pci_p2pdma_map_types_pack(const enum pci_p2pdma_map_type *type)
+{
+ unsigned long val = 0;
+ unsigned int flags;
+
+ for (flags = 0; flags < PCI_P2PDMA_TLP_CLASSES; flags++)
+ val |= (unsigned long)type[flags] << (flags * 4);
+
+ return val;
+}
+
+static enum pci_p2pdma_map_type
+pci_p2pdma_map_types_unpack(unsigned long val, unsigned int tlp_flags)
+{
+ return (val >> (tlp_flags * 4)) & 0xf;
+}
+
/*
* Calculate the P2PDMA mapping type and distance between two PCI devices.
*
@@ -816,6 +872,10 @@ static unsigned long map_types_idx(struct pci_dev *client)
* port, and Completion Redirect for read Completions on the Completer-side
* port.
*
+ * Those controls apply to different TLPs, so every class named by &enum
+ * pci_p2pdma_tlp_flags is decided from the one walk and cached together;
+ * @tlp_flags selects which one is returned.
+ *
* If ACS redirects traffic at either divergence port, return
* PCI_P2PDMA_MAP_THRU_HOST_BRIDGE. If the ACS Control register cannot be
* read, return PCI_P2PDMA_MAP_NOT_SUPPORTED. Otherwise, return
@@ -829,14 +889,16 @@ static unsigned long map_types_idx(struct pci_dev *client)
*/
static enum pci_p2pdma_map_type
calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
- int *dist, bool verbose)
+ int *dist, unsigned int tlp_flags, bool verbose)
{
- enum pci_p2pdma_map_type map_type = PCI_P2PDMA_MAP_THRU_HOST_BRIDGE;
+ enum pci_p2pdma_map_type map_type[PCI_P2PDMA_TLP_CLASSES];
struct pci_dev *a = provider, *b = client, *bb;
struct pci_dev *a_child = NULL, *b_child = NULL;
struct pci_p2pdma_acs_path path = {};
struct pci_p2pdma *p2pdma;
bool cpu_p2pdma, host_whitelisted = false;
+ bool host_fallback = false;
+ unsigned int flags;
int dist_a = 0;
int dist_b = 0;
@@ -870,6 +932,8 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
* request can only get to the peer through the host bridge.
*/
*dist = dist_a + dist_b;
+ for (flags = 0; flags < PCI_P2PDMA_TLP_CLASSES; flags++)
+ map_type[flags] = PCI_P2PDMA_MAP_THRU_HOST_BRIDGE;
goto map_through_host_bridge;
check_paths_acs:
@@ -889,18 +953,24 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
}
/*
- * Below a shared upstream bridge, a path whose divergence ports do not
- * redirect routes the request directly.
+ * The walk and the config reads above serve every class; only the
+ * decision below depends on the kind of TLP being routed.
*/
- map_type = pci_p2pdma_route(&path);
- if (map_type == PCI_P2PDMA_MAP_BUS_ADDR)
- goto done;
+ for (flags = 0; flags < PCI_P2PDMA_TLP_CLASSES; flags++) {
+ map_type[flags] = pci_p2pdma_route(&path, flags);
+ if (map_type[flags] == PCI_P2PDMA_MAP_THRU_HOST_BRIDGE)
+ host_fallback = true;
+ }
- if (verbose)
- pci_p2pdma_warn_path(client, provider, &path, a_child, b_child);
+ if (verbose && map_type[0] != PCI_P2PDMA_MAP_BUS_ADDR)
+ pci_p2pdma_warn_path(client, provider, &path, a_child,
+ b_child);
- /* An unreadable control does not establish an upstream redirect. */
- if (path.unreadable)
+ /*
+ * Nothing needs the host bridge: the classes that did not get a direct
+ * route have no fallback that would use it.
+ */
+ if (!host_fallback)
goto done;
map_through_host_bridge:
@@ -913,16 +983,19 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
if (verbose)
pci_warn(client, "cannot be used for peer-to-peer DMA as the client and provider (%s) do not share an upstream bridge or whitelisted host bridge\n",
pci_name(provider));
- map_type = PCI_P2PDMA_MAP_NOT_SUPPORTED;
+ for (flags = 0; flags < PCI_P2PDMA_TLP_CLASSES; flags++)
+ if (map_type[flags] == PCI_P2PDMA_MAP_THRU_HOST_BRIDGE)
+ map_type[flags] = PCI_P2PDMA_MAP_NOT_SUPPORTED;
}
done:
rcu_read_lock();
p2pdma = rcu_dereference(provider->p2pdma);
if (p2pdma)
xa_store(&p2pdma->map_types, map_types_idx(client),
- xa_mk_value(map_type), GFP_ATOMIC);
+ xa_mk_value(pci_p2pdma_map_types_pack(map_type)),
+ GFP_ATOMIC);
rcu_read_unlock();
- return map_type;
+ return map_type[tlp_flags];
}
/**
@@ -963,7 +1036,7 @@ int pci_p2pdma_distance_many(struct pci_dev *provider, struct device **clients,
return -1;
}
- map = calc_map_type_and_dist(provider, pci_client, &distance,
+ map = calc_map_type_and_dist(provider, pci_client, &distance, 0,
verbose);
pci_dev_put(pci_client);
@@ -1229,22 +1302,28 @@ void pci_p2pmem_publish(struct pci_dev *pdev, bool publish)
EXPORT_SYMBOL_GPL(pci_p2pmem_publish);
/**
- * pci_p2pdma_map_type - Determine the mapping type for P2PDMA transfers
+ * pci_p2pdma_map_type_tlp - Determine the mapping type for P2PDMA transfers
* @provider: P2PDMA provider structure
* @dev: Client device that initiates the transfer
+ * @tlp_flags: &enum pci_p2pdma_tlp_flags describing the TLPs @dev will issue
*
* Determines how peer-to-peer DMA transfers should be mapped between
* the provider and the client device. The mapping type indicates whether
* the transfer can be done directly through PCI switches or must go
* through the host bridge.
+ *
+ * ACS routes a peer-to-peer transaction by the attributes its TLPs carry, so
+ * the answer depends on @tlp_flags. A caller that passes flags its traffic
+ * does not match gets a mapping the fabric will not deliver.
*/
-enum pci_p2pdma_map_type pci_p2pdma_map_type(struct p2pdma_provider *provider,
- struct device *dev)
+static enum pci_p2pdma_map_type
+pci_p2pdma_map_type_tlp(struct p2pdma_provider *provider, struct device *dev,
+ unsigned int tlp_flags)
{
- enum pci_p2pdma_map_type type = PCI_P2PDMA_MAP_NOT_SUPPORTED;
struct pci_dev *pdev = to_pci_dev(provider->owner);
+ unsigned long cache_index, cached = 0;
+ enum pci_p2pdma_map_type type;
struct pci_p2pdma *p2pdma;
- unsigned long cache_index;
struct pci_dev *client;
int dist;
@@ -1261,16 +1340,32 @@ enum pci_p2pdma_map_type pci_p2pdma_map_type(struct p2pdma_provider *provider,
p2pdma = rcu_dereference(pdev->p2pdma);
if (p2pdma)
- type = xa_to_value(xa_load(&p2pdma->map_types,
- cache_index));
+ cached = xa_to_value(xa_load(&p2pdma->map_types,
+ cache_index));
rcu_read_unlock();
+ type = pci_p2pdma_map_types_unpack(cached, tlp_flags);
if (type == PCI_P2PDMA_MAP_UNKNOWN)
- return calc_map_type_and_dist(pdev, client, &dist, true);
+ return calc_map_type_and_dist(pdev, client, &dist, tlp_flags,
+ true);
return type;
}
+/**
+ * pci_p2pdma_map_type - Determine the mapping type for P2PDMA transfers
+ * @provider: P2PDMA provider structure
+ * @dev: Client device that initiates the transfer
+ *
+ * Same as pci_p2pdma_map_type_tlp() for a client issuing strictly ordered
+ * Requests that carry an Untranslated address.
+ */
+enum pci_p2pdma_map_type pci_p2pdma_map_type(struct p2pdma_provider *provider,
+ struct device *dev)
+{
+ return pci_p2pdma_map_type_tlp(provider, dev, 0);
+}
+
void __pci_p2pdma_update_state(struct pci_p2pdma_map_state *state,
struct device *dev, struct page *page)
{
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* [PATCH v10 08/18] PCI/P2PDMA: Route Relaxed Ordering Completions directly
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (6 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 07/18] PCI/P2PDMA: Answer routing per TLP class Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:18 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 09/18] PCI/P2PDMA: Reject Translated Requests blocked by Translation Blocking Leon Romanovsky
` (9 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
ACS P2P Completion Redirect leaves Completions carrying the Relaxed
Ordering attribute alone. PCIe r7.0 sec 6.12.1.1 redirects only those "that
do not have the Relaxed Ordering Attribute bit set", and sec 7.7.12.5
describes the enable bit as "applicable only to Completions whose Relaxed
Ordering Attribute is clear". P2PDMA reports one answer for every kind of
TLP, so a client that sets Relaxed Ordering on its Non-Posted Requests,
and therefore receives such Completions (sec 2.2.9.1), is sent through
the host bridge for a redirect that never happens to it.
Route Completions directly for the classes that set
PCI_P2PDMA_TLP_RELAXED_CPL.
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
drivers/pci/p2pdma.c | 6 +++++-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index 479bf58104ca..e6482467e121 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -542,11 +542,15 @@ pci_acs_p2pdma_request(u16 ctrl, unsigned int tlp_flags)
/*
* Decide how a peer-to-peer Completion at an ACS-capable ingress port routes.
* PCIe r7.0 sec 6.12.1.1: no ACS control other than P2P Completion Redirect
- * affects a Completion.
+ * affects a Completion, and that one leaves Completions carrying the Relaxed
+ * Ordering attribute alone.
*/
static enum pci_acs_p2pdma_state
pci_acs_p2pdma_completion(u16 ctrl, unsigned int tlp_flags)
{
+ if (tlp_flags & PCI_P2PDMA_TLP_RELAXED_CPL)
+ return PCI_ACS_P2PDMA_DIRECT;
+
return ctrl & PCI_ACS_CR ? PCI_ACS_P2PDMA_REDIRECT :
PCI_ACS_P2PDMA_DIRECT;
}
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* [PATCH v10 09/18] PCI/P2PDMA: Reject Translated Requests blocked by Translation Blocking
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (7 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 08/18] PCI/P2PDMA: Route Relaxed Ordering Completions directly Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:20 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 10/18] PCI/P2PDMA: Route Translated Requests under Direct Translated P2P Leon Romanovsky
` (8 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
A Downstream Port with ACS Translation Blocking enabled treats every
Upstream Memory Request whose Address Type is not Untranslated as an ACS
Violation, ahead of "any applicable ACS P2P control mechanisms" per PCIe
r7.0 sec 6.12.1.1. P2PDMA never looks at that bit, so it reports a
bus-addressable path where the Requester's Translated Requests would be
rejected.
Add PCI_ACS_P2PDMA_BLOCKED, and because blocking is not a routing control,
scan the path for it rather than the divergence port alone. A blocked
Request has no host bridge fallback, since the Address Type is rejected
wherever the Request is addressed.
The two routes do not pass the same ports, so scan them separately. A
direct route turns around at the divergence and only passes the ports
below it, while the host bridge route keeps climbing and passes that port
and everything above it as well.
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
drivers/pci/p2pdma.c | 116 +++++++++++++++++++++++++++++++++++++++++++++++----
1 file changed, 109 insertions(+), 7 deletions(-)
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index e6482467e121..0f9bb96f2176 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -521,11 +521,12 @@ enum pci_acs_p2pdma_state {
PCI_ACS_P2PDMA_NOT_SUPPORTED,
PCI_ACS_P2PDMA_DIRECT,
PCI_ACS_P2PDMA_REDIRECT,
+ PCI_ACS_P2PDMA_BLOCKED,
};
/*
* Decide how a peer-to-peer Request at an ACS-capable ingress port routes,
- * from that port's ACS Control register.
+ * from that port's ACS Control register and the Request's Address Type.
*
* Linux does not read the Egress Control Vector, so Egress Control is treated
* conservatively as a redirect. Per PCIe r7.0 Table 6-11 the outcomes it
@@ -535,6 +536,18 @@ enum pci_acs_p2pdma_state {
static enum pci_acs_p2pdma_state
pci_acs_p2pdma_request(u16 ctrl, unsigned int tlp_flags)
{
+ if (tlp_flags & PCI_P2PDMA_TLP_TRANSLATED) {
+ /*
+ * PCIe r7.0 sec 6.12.1.1: Translation Blocking makes every
+ * Upstream Memory Request whose Address Type is not
+ * Untranslated an ACS Violation, taking precedence over the
+ * P2P controls. Sec 7.7.12.5: Direct Translated P2P "is
+ * ignored if ACS Translation Blocking Enable is 1b".
+ */
+ if (ctrl & PCI_ACS_TB)
+ return PCI_ACS_P2PDMA_BLOCKED;
+ }
+
return ctrl & (PCI_ACS_RR | PCI_ACS_EC) ?
PCI_ACS_P2PDMA_REDIRECT : PCI_ACS_P2PDMA_DIRECT;
}
@@ -574,6 +587,39 @@ static bool pci_acs_p2pdma_ctrl(struct pci_dev *pdev, u16 *ctrl)
return !pci_read_config_word(pdev, pos + PCI_ACS_CTRL, ctrl);
}
+/*
+ * Report whether any port between @client and @common rejects Translated
+ * addresses. If @common is NULL, every port up to the host bridge is walked.
+ * A port whose ACS Control cannot be read counts as blocking, which withdraws
+ * only the Translated classes because an Untranslated Request is routed by the
+ * redirect controls instead.
+ */
+static bool pci_p2pdma_path_blocks_translation(struct pci_dev *client,
+ struct pci_dev *common)
+{
+ struct pci_dev *pdev;
+ u16 ctrl;
+
+ /*
+ * @common is @client itself when the provider is the client or sits
+ * below it. The Request never travels upstream then, so no port sees
+ * it and none can reject its Address Type.
+ */
+ if (client == common)
+ return false;
+
+ for (pdev = pci_upstream_bridge(client); pdev && pdev != common;
+ pdev = pci_upstream_bridge(pdev)) {
+ if (!pci_acs_p2pdma_ctrl(pdev, &ctrl))
+ return true;
+
+ if (ctrl & PCI_ACS_TB)
+ return true;
+ }
+
+ return false;
+}
+
static void seq_buf_print_bus_devfn(struct seq_buf *buf, struct pci_dev *pdev)
{
if (!buf)
@@ -592,14 +638,42 @@ static void seq_buf_print_bus_devfn(struct seq_buf *buf, struct pci_dev *pdev)
* than upstream, so it is where the Request controls apply.
* @cpl_ctrl: ACS Control of the Completer-side divergence port, likewise for
* the Completions travelling back.
+ * @tb_on_path: A port below the divergence blocks Translated addresses.
+ * Every route out of @client passes those, so none carries them.
+ * @tb_above_divergence: A port at or above the divergence blocks Translated
+ * addresses. Only a Request continuing to the host bridge passes
+ * those, so a direct route is still open to them.
+ * @no_common_bridge: The two paths share no upstream bridge, so no direct
+ * route exists for ACS to gate.
* @unreadable: First port whose ACS Control could not be read, if any.
*/
struct pci_p2pdma_acs_path {
u16 req_ctrl;
u16 cpl_ctrl;
+ bool tb_on_path;
+ bool tb_above_divergence;
+ bool no_common_bridge;
struct pci_dev *unreadable;
};
+/*
+ * ACS Translation Blocking is not a routing control, so unlike the redirect
+ * controls it is not decided at the divergence alone. PCIe r7.0 sec 6.12.1.1
+ * has every Downstream Port check the Address Type of each Upstream Memory
+ * Request it receives, ahead of "any applicable ACS P2P control mechanisms".
+ * A port below the divergence cannot redirect the Request anywhere it was not
+ * already going, but it can still reject a Translated address.
+ */
+static enum pci_acs_p2pdma_state
+pci_p2pdma_request_state(const struct pci_p2pdma_acs_path *path,
+ unsigned int tlp_flags)
+{
+ if (tlp_flags & PCI_P2PDMA_TLP_TRANSLATED && path->tb_on_path)
+ return PCI_ACS_P2PDMA_BLOCKED;
+
+ return pci_acs_p2pdma_request(path->req_ctrl, tlp_flags);
+}
+
/*
* Combine both directions into a mapping type. Only a path that routes the
* Request and the Completions it generates directly can be programmed with
@@ -609,15 +683,35 @@ static enum pci_p2pdma_map_type
pci_p2pdma_route(const struct pci_p2pdma_acs_path *path,
unsigned int tlp_flags)
{
+ enum pci_acs_p2pdma_state req;
+
if (path->unreadable)
return PCI_P2PDMA_MAP_NOT_SUPPORTED;
- if (pci_acs_p2pdma_request(path->req_ctrl, tlp_flags) ==
- PCI_ACS_P2PDMA_DIRECT &&
+ req = pci_p2pdma_request_state(path, tlp_flags);
+
+ /*
+ * Translation Blocking rejects the Address Type rather than the
+ * target, so a blocked Request stays blocked however it is addressed.
+ * No host bridge fallback keeps a Translated address working; the
+ * caller has to issue a different kind of Request instead.
+ */
+ if (req == PCI_ACS_P2PDMA_BLOCKED)
+ return PCI_P2PDMA_MAP_NOT_SUPPORTED;
+
+ if (!path->no_common_bridge && req == PCI_ACS_P2PDMA_DIRECT &&
pci_acs_p2pdma_completion(path->cpl_ctrl, tlp_flags) ==
PCI_ACS_P2PDMA_DIRECT)
return PCI_P2PDMA_MAP_BUS_ADDR;
+ /*
+ * A Request that turns around at the divergence never reaches the
+ * ports above it, but one that keeps climbing to the host bridge
+ * does, so that route has to clear their Translation Blocking too.
+ */
+ if (tlp_flags & PCI_P2PDMA_TLP_TRANSLATED && path->tb_above_divergence)
+ return PCI_P2PDMA_MAP_NOT_SUPPORTED;
+
return PCI_P2PDMA_MAP_THRU_HOST_BRIDGE;
}
@@ -874,7 +968,8 @@ pci_p2pdma_map_types_unpack(unsigned long val, unsigned int tlp_flags)
* The client is the Requester and the provider the Completer. At the path
* divergence, check Request Redirect and Egress Control on the Requester-side
* port, and Completion Redirect for read Completions on the Completer-side
- * port.
+ * port. Translation Blocking is checked on every port the Request passes
+ * instead, because it rejects a Request rather than routing it.
*
* Those controls apply to different TLPs, so every class named by &enum
* pci_p2pdma_tlp_flags is decided from the one walk and cached together;
@@ -882,8 +977,8 @@ pci_p2pdma_map_types_unpack(unsigned long val, unsigned int tlp_flags)
*
* If ACS redirects traffic at either divergence port, return
* PCI_P2PDMA_MAP_THRU_HOST_BRIDGE. If the ACS Control register cannot be
- * read, return PCI_P2PDMA_MAP_NOT_SUPPORTED. Otherwise, return
- * PCI_P2PDMA_MAP_BUS_ADDR.
+ * read, or Translation Blocking rejects the class being asked about, return
+ * PCI_P2PDMA_MAP_NOT_SUPPORTED. Otherwise, return PCI_P2PDMA_MAP_BUS_ADDR.
*
* Any two devices that have a data path that goes through the host bridge
* will consult a whitelist. If the host bridge is in the whitelist, return
@@ -936,8 +1031,10 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
* request can only get to the peer through the host bridge.
*/
*dist = dist_a + dist_b;
+ path.no_common_bridge = true;
+ path.tb_on_path = pci_p2pdma_path_blocks_translation(client, NULL);
for (flags = 0; flags < PCI_P2PDMA_TLP_CLASSES; flags++)
- map_type[flags] = PCI_P2PDMA_MAP_THRU_HOST_BRIDGE;
+ map_type[flags] = pci_p2pdma_route(&path, flags);
goto map_through_host_bridge;
check_paths_acs:
@@ -956,6 +1053,11 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
path.unreadable = b_child;
}
+ path.tb_on_path = pci_p2pdma_path_blocks_translation(client, a);
+ if (b_child)
+ path.tb_above_divergence =
+ pci_p2pdma_path_blocks_translation(b_child, NULL);
+
/*
* The walk and the config reads above serve every class; only the
* decision below depends on the kind of TLP being routed.
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* Re: [PATCH v10 09/18] PCI/P2PDMA: Reject Translated Requests blocked by Translation Blocking
2026-10-12 13:07 ` [PATCH v10 09/18] PCI/P2PDMA: Reject Translated Requests blocked by Translation Blocking Leon Romanovsky
@ 2026-10-12 13:20 ` sashiko-bot
0 siblings, 0 replies; 38+ messages in thread
From: sashiko-bot @ 2026-10-12 13:20 UTC (permalink / raw)
To: Leon Romanovsky; +Cc: dri-devel, media-ci, linux-pci, kvm, linux-rdma
> From: Leon Romanovsky <leonro@nvidia.com>
>
> A Downstream Port with ACS Translation Blocking enabled treats every
> Upstream Memory Request whose Address Type is not Untranslated as an ACS
> Violation, ahead of "any applicable ACS P2P control mechanisms" per PCIe
> r7.0 sec 6.12.1.1. P2PDMA never looks at that bit, so it reports a
> bus-addressable path where the Requester's Translated Requests would be
> rejected.
>
> Add PCI_ACS_P2PDMA_BLOCKED, and because blocking is not a routing control,
> scan the path for it rather than the divergence port alone. A blocked
> Request has no host bridge fallback, since the Address Type is rejected
> wherever the Request is addressed.
>
> The two routes do not pass the same ports, so scan them separately. A
> direct route turns around at the divergence and only passes the ports
> below it, while the host bridge route keeps climbing and passes that port
> and everything above it as well.
>
> Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
> Tested-by: Tushar Dave <tdave@nvidia.com>
> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Sashiko has reviewed this patch and found no issues. It looks great!
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261012-fix-p2p-acs-v4-0-v10-0-6ffe7b7a4517@nvidia.com?part=9
^ permalink raw reply [flat|nested] 38+ messages in thread
* [PATCH v10 10/18] PCI/P2PDMA: Route Translated Requests under Direct Translated P2P
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (8 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 09/18] PCI/P2PDMA: Reject Translated Requests blocked by Translation Blocking Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:15 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 11/18] PCI/P2PDMA: Log detailed ACS routing diagnostics Leon Romanovsky
` (7 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
A Downstream Port with ACS Direct Translated P2P enabled routes a Request
whose Address Type is Translated "to the peer Egress Port without
redirection, regardless of ACS P2P Request Redirect and ACS P2P Egress
Control", per PCIe r7.0 sec 6.12.3. P2PDMA assumes every Request carries an
Untranslated address, so it sends an ATS client through the host bridge
even where the fabric would route it straight to the peer.
Consult Direct Translated P2P for the PCI_P2PDMA_TLP_TRANSLATED classes.
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
drivers/pci/p2pdma.c | 18 ++++++++++++++----
1 file changed, 14 insertions(+), 4 deletions(-)
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index 0f9bb96f2176..ebc34aae03a1 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -546,6 +546,15 @@ pci_acs_p2pdma_request(u16 ctrl, unsigned int tlp_flags)
*/
if (ctrl & PCI_ACS_TB)
return PCI_ACS_P2PDMA_BLOCKED;
+
+ /*
+ * PCIe r7.0 sec 6.12.3: ACS Direct Translated P2P routes a
+ * Request carrying a Translated address to the peer "without
+ * redirection, regardless of ACS P2P Request Redirect and ACS
+ * P2P Egress Control settings".
+ */
+ if (ctrl & PCI_ACS_DT)
+ return PCI_ACS_P2PDMA_DIRECT;
}
return ctrl & (PCI_ACS_RR | PCI_ACS_EC) ?
@@ -966,10 +975,11 @@ pci_p2pdma_map_types_unpack(unsigned long val, unsigned int tlp_flags)
* redirection setting of the ports along the path.
*
* The client is the Requester and the provider the Completer. At the path
- * divergence, check Request Redirect and Egress Control on the Requester-side
- * port, and Completion Redirect for read Completions on the Completer-side
- * port. Translation Blocking is checked on every port the Request passes
- * instead, because it rejects a Request rather than routing it.
+ * divergence, check Request Redirect, Egress Control and Direct Translated
+ * P2P on the Requester-side port, and Completion Redirect for read
+ * Completions on the Completer-side port. Translation Blocking is checked on
+ * every port the Request passes instead, because it rejects a Request rather
+ * than routing it.
*
* Those controls apply to different TLPs, so every class named by &enum
* pci_p2pdma_tlp_flags is decided from the one walk and cached together;
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* [PATCH v10 11/18] PCI/P2PDMA: Log detailed ACS routing diagnostics
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (9 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 10/18] PCI/P2PDMA: Route Translated Requests under Direct Translated P2P Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:18 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 12/18] PCI/P2PDMA: Add KUnit tests for the ACS routing decisions Leon Romanovsky
` (6 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
When P2PDMA rejects a mapping, existing warnings identify only the final
ACS or host-bridge result. They omit topology, live controls, divergence
ports, cache state, and intermediate routing decisions.
Emit debug-level messages for verbose calculations and cache lookups.
Report both paths, decoded ACS controls, directional decisions, host
fallback, and the final mapping. This keeps incidental unsupported probes
quiet while allowing the diagnostics to be enabled when needed.
Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
drivers/pci/p2pdma.c | 232 +++++++++++++++++++++++++++++++++++++++++++++++----
1 file changed, 215 insertions(+), 17 deletions(-)
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index ebc34aae03a1..46f7927968ff 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -577,23 +577,128 @@ pci_acs_p2pdma_completion(u16 ctrl, unsigned int tlp_flags)
PCI_ACS_P2PDMA_DIRECT;
}
+static const char *pci_acs_p2pdma_state_name(enum pci_acs_p2pdma_state state)
+{
+ switch (state) {
+ case PCI_ACS_P2PDMA_DIRECT:
+ return "direct";
+ case PCI_ACS_P2PDMA_REDIRECT:
+ return "redirect";
+ case PCI_ACS_P2PDMA_BLOCKED:
+ return "blocked";
+ case PCI_ACS_P2PDMA_NOT_SUPPORTED:
+ return "not-supported";
+ }
+
+ return "invalid";
+}
+
+static const char *pci_p2pdma_map_type_name(enum pci_p2pdma_map_type type)
+{
+ switch (type) {
+ case PCI_P2PDMA_MAP_UNKNOWN:
+ return "unknown";
+ case PCI_P2PDMA_MAP_NONE:
+ return "none";
+ case PCI_P2PDMA_MAP_NOT_SUPPORTED:
+ return "not-supported";
+ case PCI_P2PDMA_MAP_BUS_ADDR:
+ return "bus-address";
+ case PCI_P2PDMA_MAP_THRU_HOST_BRIDGE:
+ return "through-host-bridge";
+ }
+
+ return "invalid";
+}
+
/*
* Read @pdev's ACS Control register. A device without an ACS capability has
* no peer-to-peer controls at all, which routes the same as having them all
* clear. Returns false when the register is present but cannot be read; @ctrl
* is then meaningless.
*/
-static bool pci_acs_p2pdma_ctrl(struct pci_dev *pdev, u16 *ctrl)
+static bool pci_acs_p2pdma_ctrl(struct pci_dev *pdev, const char *what,
+ u16 *ctrl, bool verbose)
{
- int pos;
+ int pos, ret;
pos = pdev->acs_cap;
if (!pos) {
+ if (verbose)
+ pci_dbg(pdev,
+ "P2PDMA ACS: %s has no ACS capability\n", what);
*ctrl = 0;
return true;
}
- return !pci_read_config_word(pdev, pos + PCI_ACS_CTRL, ctrl);
+ ret = pci_read_config_word(pdev, pos + PCI_ACS_CTRL, ctrl);
+ if (ret) {
+ if (verbose)
+ pci_dbg(pdev,
+ "P2PDMA ACS: %s ACS Control read failed at %#x: %#x\n",
+ what, pos + PCI_ACS_CTRL, ret);
+ return false;
+ }
+
+ if (verbose) {
+ pci_dbg(pdev,
+ "P2PDMA ACS: %s cap=%#x caps=%#06x ctrl=%#06x\n",
+ what, pos, pdev->acs_capabilities, *ctrl);
+ pci_dbg(pdev,
+ "P2PDMA ACS: control bits SV=%u TB=%u RR=%u CR=%u UF=%u EC=%u DT=%u\n",
+ !!(*ctrl & PCI_ACS_SV), !!(*ctrl & PCI_ACS_TB),
+ !!(*ctrl & PCI_ACS_RR), !!(*ctrl & PCI_ACS_CR),
+ !!(*ctrl & PCI_ACS_UF), !!(*ctrl & PCI_ACS_EC),
+ !!(*ctrl & PCI_ACS_DT));
+ }
+
+ return true;
+}
+
+static void pci_p2pdma_log_path(const char *name, struct pci_dev *start,
+ struct pci_dev *common)
+{
+ struct pci_dev *pdev, *upstream;
+ int hop = 0, ret, type;
+ u16 ctrl;
+
+ for (pdev = start; pdev; pdev = upstream, hop++) {
+ upstream = pci_upstream_bridge(pdev);
+ type = pci_is_pcie(pdev) ? pci_pcie_type(pdev) : -1;
+ pci_dbg(pdev,
+ "P2PDMA ACS: %s path hop=%d common=%u pcie=%u type=%d class=%#08x vendor=%04x device=%04x upstream=%s\n",
+ name, hop, pdev == common, pci_is_pcie(pdev), type,
+ pdev->class, pdev->vendor, pdev->device,
+ upstream ? pci_name(upstream) : "<none>");
+
+ if (pdev->subordinate)
+ pci_dbg(pdev,
+ "P2PDMA ACS: bridge bus range=%02llx-%02llx\n",
+ (unsigned long long)pdev->subordinate->busn_res.start,
+ (unsigned long long)pdev->subordinate->busn_res.end);
+
+ if (!pdev->acs_cap) {
+ pci_dbg(pdev, "P2PDMA ACS: ACS capability absent\n");
+ continue;
+ }
+
+ ret = pci_read_config_word(pdev, pdev->acs_cap + PCI_ACS_CTRL,
+ &ctrl);
+ if (ret) {
+ pci_dbg(pdev,
+ "P2PDMA ACS: ACS cap=%#x caps=%#06x Control read failed: %#x\n",
+ pdev->acs_cap, pdev->acs_capabilities, ret);
+ continue;
+ }
+
+ pci_dbg(pdev,
+ "P2PDMA ACS: ACS cap=%#x caps=%#06x ctrl=%#06x SV=%u TB=%u RR=%u CR=%u UF=%u EC=%u DT=%u\n",
+ pdev->acs_cap, pdev->acs_capabilities, ctrl,
+ !!(ctrl & PCI_ACS_SV), !!(ctrl & PCI_ACS_TB),
+ !!(ctrl & PCI_ACS_RR), !!(ctrl & PCI_ACS_CR),
+ !!(ctrl & PCI_ACS_UF), !!(ctrl & PCI_ACS_EC),
+ !!(ctrl & PCI_ACS_DT));
+ }
}
/*
@@ -604,7 +709,8 @@ static bool pci_acs_p2pdma_ctrl(struct pci_dev *pdev, u16 *ctrl)
* redirect controls instead.
*/
static bool pci_p2pdma_path_blocks_translation(struct pci_dev *client,
- struct pci_dev *common)
+ struct pci_dev *common,
+ bool verbose)
{
struct pci_dev *pdev;
u16 ctrl;
@@ -619,11 +725,15 @@ static bool pci_p2pdma_path_blocks_translation(struct pci_dev *client,
for (pdev = pci_upstream_bridge(client); pdev && pdev != common;
pdev = pci_upstream_bridge(pdev)) {
- if (!pci_acs_p2pdma_ctrl(pdev, &ctrl))
+ if (!pci_acs_p2pdma_ctrl(pdev, "path hop", &ctrl, verbose))
return true;
- if (ctrl & PCI_ACS_TB)
+ if (ctrl & PCI_ACS_TB) {
+ if (verbose)
+ pci_dbg(pdev,
+ "P2PDMA ACS: Translation Blocking rejects Translated Requests on this path\n");
return true;
+ }
}
return false;
@@ -1003,14 +1113,22 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
enum pci_p2pdma_map_type map_type[PCI_P2PDMA_TLP_CLASSES];
struct pci_dev *a = provider, *b = client, *bb;
struct pci_dev *a_child = NULL, *b_child = NULL;
+ struct pci_host_bridge *provider_host, *client_host;
struct pci_p2pdma_acs_path path = {};
struct pci_p2pdma *p2pdma;
bool cpu_p2pdma, host_whitelisted = false;
+ bool cache_store = false;
bool host_fallback = false;
unsigned int flags;
int dist_a = 0;
int dist_b = 0;
+ if (verbose)
+ pci_dbg(client,
+ "P2PDMA ACS: begin provider=%s client=%s cache-index=%#lx\n",
+ pci_name(provider), pci_name(client),
+ map_types_idx(client));
+
/*
* Note, we don't need to take references to devices returned by
* pci_upstream_bridge() seeing we hold a reference to a child
@@ -1041,14 +1159,32 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
* request can only get to the peer through the host bridge.
*/
*dist = dist_a + dist_b;
+ if (verbose) {
+ pci_dbg(client,
+ "P2PDMA ACS: no common upstream bridge provider-distance=%d client-distance=%d total=%d\n",
+ dist_a, dist_b, *dist);
+ pci_p2pdma_log_path("provider", provider, NULL);
+ pci_p2pdma_log_path("client", client, NULL);
+ }
path.no_common_bridge = true;
- path.tb_on_path = pci_p2pdma_path_blocks_translation(client, NULL);
+ path.tb_on_path = pci_p2pdma_path_blocks_translation(client, NULL,
+ verbose);
for (flags = 0; flags < PCI_P2PDMA_TLP_CLASSES; flags++)
map_type[flags] = pci_p2pdma_route(&path, flags);
goto map_through_host_bridge;
check_paths_acs:
*dist = dist_a + dist_b;
+ if (verbose) {
+ pci_dbg(client,
+ "P2PDMA ACS: common=%s provider-divergence=%s client-divergence=%s provider-distance=%d client-distance=%d total=%d\n",
+ pci_name(a),
+ a_child ? pci_name(a_child) : "<none>",
+ b_child ? pci_name(b_child) : "<none>",
+ dist_a, dist_b, *dist);
+ pci_p2pdma_log_path("provider", provider, a);
+ pci_p2pdma_log_path("client", client, a);
+ }
/*
* ACS P2P routing controls apply where a TLP can route toward the peer
@@ -1056,17 +1192,33 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
* branch is upstream, so redirect controls do not affect the path.
*/
if (a_child && b_child) {
- if (!pci_acs_p2pdma_ctrl(a_child, &path.cpl_ctrl))
+ if (!pci_acs_p2pdma_ctrl(a_child, "completion", &path.cpl_ctrl,
+ verbose))
path.unreadable = a_child;
- if (!pci_acs_p2pdma_ctrl(b_child, &path.req_ctrl) &&
- !path.unreadable)
+ if (!pci_acs_p2pdma_ctrl(b_child, "request", &path.req_ctrl,
+ verbose) && !path.unreadable)
path.unreadable = b_child;
+ if (verbose && !path.unreadable)
+ pci_dbg(client,
+ "P2PDMA ACS: request=%s at %s completion=%s at %s\n",
+ pci_acs_p2pdma_state_name(
+ pci_p2pdma_request_state(&path, 0)),
+ pci_name(b_child),
+ pci_acs_p2pdma_state_name(
+ pci_acs_p2pdma_completion(path.cpl_ctrl,
+ 0)),
+ pci_name(a_child));
+ } else if (verbose) {
+ pci_dbg(client,
+ "P2PDMA ACS: peer divergence is incomplete; no ACS peer-routing controls evaluated\n");
}
- path.tb_on_path = pci_p2pdma_path_blocks_translation(client, a);
+ path.tb_on_path = pci_p2pdma_path_blocks_translation(client, a,
+ verbose);
if (b_child)
path.tb_above_divergence =
- pci_p2pdma_path_blocks_translation(b_child, NULL);
+ pci_p2pdma_path_blocks_translation(b_child, NULL,
+ verbose);
/*
* The walk and the config reads above serve every class; only the
@@ -1095,6 +1247,19 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
host_whitelisted = host_bridge_whitelist(provider, client,
verbose);
+ if (verbose) {
+ provider_host = pci_find_host_bridge(provider->bus);
+ client_host = pci_find_host_bridge(client->bus);
+ pci_dbg(client,
+ "P2PDMA ACS: host fallback cpu-support=%u whitelist=%s provider-host=%s client-host=%s same-host=%u\n",
+ cpu_p2pdma,
+ cpu_p2pdma ? "not-consulted" :
+ (host_whitelisted ? "yes" : "no"),
+ provider_host ? dev_name(&provider_host->dev) : "<none>",
+ client_host ? dev_name(&client_host->dev) : "<none>",
+ provider_host && provider_host == client_host);
+ }
+
if (!cpu_p2pdma && !host_whitelisted) {
if (verbose)
pci_warn(client, "cannot be used for peer-to-peer DMA as the client and provider (%s) do not share an upstream bridge or whitelisted host bridge\n",
@@ -1106,11 +1271,31 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
done:
rcu_read_lock();
p2pdma = rcu_dereference(provider->p2pdma);
- if (p2pdma)
+ if (p2pdma) {
xa_store(&p2pdma->map_types, map_types_idx(client),
- xa_mk_value(pci_p2pdma_map_types_pack(map_type)),
- GFP_ATOMIC);
+ xa_mk_value(pci_p2pdma_map_types_pack(map_type)), GFP_ATOMIC);
+ cache_store = true;
+ }
rcu_read_unlock();
+ if (verbose) {
+ pci_dbg(client,
+ "P2PDMA ACS: final provider=%s result=%s(%d) tlp-flags=%#x distance=%d unreadable=%s cache-store=%u index=%#lx\n",
+ pci_name(provider),
+ pci_p2pdma_map_type_name(map_type[tlp_flags]),
+ map_type[tlp_flags], tlp_flags, *dist,
+ path.unreadable ? pci_name(path.unreadable) : "<none>",
+ cache_store, map_types_idx(client));
+ pci_dbg(client,
+ "P2PDMA ACS: classes strict=%s relaxed=%s translated=%s translated+relaxed=%s\n",
+ pci_p2pdma_map_type_name(map_type[0]),
+ pci_p2pdma_map_type_name(
+ map_type[PCI_P2PDMA_TLP_RELAXED_CPL]),
+ pci_p2pdma_map_type_name(
+ map_type[PCI_P2PDMA_TLP_TRANSLATED]),
+ pci_p2pdma_map_type_name(
+ map_type[PCI_P2PDMA_TLP_TRANSLATED |
+ PCI_P2PDMA_TLP_RELAXED_CPL]));
+ }
return map_type[tlp_flags];
}
@@ -1441,13 +1626,21 @@ pci_p2pdma_map_type_tlp(struct p2pdma_provider *provider, struct device *dev,
enum pci_p2pdma_map_type type;
struct pci_p2pdma *p2pdma;
struct pci_dev *client;
+ bool provider_state;
int dist;
- if (!pdev->p2pdma)
+ if (!pdev->p2pdma) {
+ pci_dbg(pdev,
+ "P2PDMA ACS: map lookup rejected; provider state is absent\n");
return PCI_P2PDMA_MAP_NOT_SUPPORTED;
+ }
- if (!dev_is_pci(dev))
+ if (!dev_is_pci(dev)) {
+ dev_dbg(dev,
+ "P2PDMA ACS: provider=%s map lookup rejected; client is not PCI\n",
+ pci_name(pdev));
return PCI_P2PDMA_MAP_NOT_SUPPORTED;
+ }
client = to_pci_dev(dev);
cache_index = map_types_idx(client);
@@ -1458,8 +1651,13 @@ pci_p2pdma_map_type_tlp(struct p2pdma_provider *provider, struct device *dev,
if (p2pdma)
cached = xa_to_value(xa_load(&p2pdma->map_types,
cache_index));
+ provider_state = !!p2pdma;
rcu_read_unlock();
type = pci_p2pdma_map_types_unpack(cached, tlp_flags);
+ pci_dbg(client,
+ "P2PDMA ACS: map lookup provider=%s index=%#lx tlp-flags=%#x cached=%s(%d) provider-state=%u\n",
+ pci_name(pdev), cache_index, tlp_flags,
+ pci_p2pdma_map_type_name(type), type, provider_state);
if (type == PCI_P2PDMA_MAP_UNKNOWN)
return calc_map_type_and_dist(pdev, client, &dist, tlp_flags,
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* Re: [PATCH v10 11/18] PCI/P2PDMA: Log detailed ACS routing diagnostics
2026-10-12 13:07 ` [PATCH v10 11/18] PCI/P2PDMA: Log detailed ACS routing diagnostics Leon Romanovsky
@ 2026-10-12 13:18 ` sashiko-bot
0 siblings, 0 replies; 38+ messages in thread
From: sashiko-bot @ 2026-10-12 13:18 UTC (permalink / raw)
To: Leon Romanovsky; +Cc: linux-pci, linux-rdma, media-ci, kvm, dri-devel
> From: Leon Romanovsky <leonro@nvidia.com>
>
> When P2PDMA rejects a mapping, existing warnings identify only the final
> ACS or host-bridge result. They omit topology, live controls, divergence
> ports, cache state, and intermediate routing decisions.
>
> Emit debug-level messages for verbose calculations and cache lookups.
> Report both paths, decoded ACS controls, directional decisions, host
> fallback, and the final mapping. This keeps incidental unsupported probes
> quiet while allowing the diagnostics to be enabled when needed.
>
> Reviewed-by: Logan Gunthorpe <logang@deltatee.com>
> Tested-by: Tushar Dave <tdave@nvidia.com>
> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Sashiko has reviewed this patch and found no issues. It looks great!
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261012-fix-p2p-acs-v4-0-v10-0-6ffe7b7a4517@nvidia.com?part=11
^ permalink raw reply [flat|nested] 38+ messages in thread
* [PATCH v10 12/18] PCI/P2PDMA: Add KUnit tests for the ACS routing decisions
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (10 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 11/18] PCI/P2PDMA: Log detailed ACS routing diagnostics Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:19 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 13/18] PCI/P2PDMA: Test the ACS P2P routing walk Leon Romanovsky
` (5 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
pci_acs_p2pdma_request() and pci_acs_p2pdma_completion() turn an ACS
Control register and a TLP class into a routing decision. Which bits apply
to which direction and which class is easy to get wrong, and hardware that
exposes a given combination may not be at hand.
Drive both from a table of register values and classes, covering the
redirect controls per direction and Translation Blocking, Direct Translated
P2P and Relaxed Ordering. Direct Translated P2P gets a case with and
without a redirect to override, since it changes nothing without one.
Exposing the two helpers moves their state enum and the TLP flags into
pci.h.
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
drivers/pci/Kconfig | 15 ++++++
drivers/pci/Makefile | 1 +
drivers/pci/p2pdma.c | 38 ++------------
drivers/pci/pci.h | 41 +++++++++++++++
drivers/pci/pci_acs_test.c | 121 +++++++++++++++++++++++++++++++++++++++++++++
5 files changed, 182 insertions(+), 34 deletions(-)
diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig
index 0c7408509ba2..7a3eb5beb328 100644
--- a/drivers/pci/Kconfig
+++ b/drivers/pci/Kconfig
@@ -226,6 +226,21 @@ config PCI_P2PDMA
If unsure, say N.
+config PCI_ACS_KUNIT_TEST
+ tristate "KUnit tests for PCI ACS P2P routing" if !KUNIT_ALL_TESTS
+ depends on PCI_P2PDMA && KUNIT
+ default KUNIT_ALL_TESTS
+ help
+ Enable KUnit tests for the PCI ACS peer-to-peer routing decision
+ logic, including direction-specific Request and Completion
+ controls that cannot all be exercised on typical peer-to-peer
+ hardware.
+
+ For more information on KUnit and unit tests in general, refer to
+ the KUnit documentation in Documentation/dev-tools/kunit/.
+
+ If unsure, say N.
+
config PCI_LABEL
def_bool y if (DMI || ACPI)
select NLS
diff --git a/drivers/pci/Makefile b/drivers/pci/Makefile
index 41ebc3b9a518..6305d128d3df 100644
--- a/drivers/pci/Makefile
+++ b/drivers/pci/Makefile
@@ -31,6 +31,7 @@ obj-$(CONFIG_PCI_STUB) += pci-stub.o
obj-$(CONFIG_PCI_PF_STUB) += pci-pf-stub.o
obj-$(CONFIG_PCI_ECAM) += ecam.o
obj-$(CONFIG_PCI_P2PDMA) += p2pdma.o
+obj-$(CONFIG_PCI_ACS_KUNIT_TEST) += pci_acs_test.o
obj-$(CONFIG_XEN_PCIDEV_FRONTEND) += xen-pcifront.o
obj-$(CONFIG_VGA_ARB) += vgaarb.o
obj-$(CONFIG_PCI_DOE) += doe.o
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index 46f7927968ff..93735f110b14 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -492,38 +492,6 @@ static struct pci_dev *find_parent_pci_dev(struct device *dev)
return NULL;
}
-/**
- * enum pci_p2pdma_tlp_flags - Properties of the TLPs a client will issue
- *
- * These describe the traffic rather than the topology, and select which ACS
- * controls apply along the peer-to-peer path. A value of 0 means strictly
- * ordered Requests carrying an Untranslated address.
- *
- * @PCI_P2PDMA_TLP_TRANSLATED: Requests carry an ATS Translated address. PCIe
- * r7.0 sec 6.12.3 routes those to the peer regardless of ACS P2P Request
- * Redirect and ACS P2P Egress Control wherever ACS Direct Translated P2P
- * is enabled.
- * @PCI_P2PDMA_TLP_RELAXED_CPL: Non-Posted Requests carry the Relaxed
- * Ordering attribute, and PCIe r7.0 sec 2.2.9.1 requires the Completer to
- * return it in their Completions. Sec 6.12.1.1 never redirects those, so
- * ACS P2P Completion Redirect does not gate the path.
- */
-enum pci_p2pdma_tlp_flags {
- PCI_P2PDMA_TLP_TRANSLATED = 1 << 0,
- PCI_P2PDMA_TLP_RELAXED_CPL = 1 << 1,
-};
-
-/* Every combination of the flags above selects one routing class. */
-#define PCI_P2PDMA_TLP_CLASSES \
- ((PCI_P2PDMA_TLP_TRANSLATED | PCI_P2PDMA_TLP_RELAXED_CPL) + 1)
-
-enum pci_acs_p2pdma_state {
- PCI_ACS_P2PDMA_NOT_SUPPORTED,
- PCI_ACS_P2PDMA_DIRECT,
- PCI_ACS_P2PDMA_REDIRECT,
- PCI_ACS_P2PDMA_BLOCKED,
-};
-
/*
* Decide how a peer-to-peer Request at an ACS-capable ingress port routes,
* from that port's ACS Control register and the Request's Address Type.
@@ -533,7 +501,7 @@ enum pci_acs_p2pdma_state {
* selects are a direct route and an ACS Violation, and neither one lets peer
* bus addressing be assumed.
*/
-static enum pci_acs_p2pdma_state
+VISIBLE_IF_KUNIT enum pci_acs_p2pdma_state
pci_acs_p2pdma_request(u16 ctrl, unsigned int tlp_flags)
{
if (tlp_flags & PCI_P2PDMA_TLP_TRANSLATED) {
@@ -560,6 +528,7 @@ pci_acs_p2pdma_request(u16 ctrl, unsigned int tlp_flags)
return ctrl & (PCI_ACS_RR | PCI_ACS_EC) ?
PCI_ACS_P2PDMA_REDIRECT : PCI_ACS_P2PDMA_DIRECT;
}
+EXPORT_SYMBOL_IF_KUNIT(pci_acs_p2pdma_request);
/*
* Decide how a peer-to-peer Completion at an ACS-capable ingress port routes.
@@ -567,7 +536,7 @@ pci_acs_p2pdma_request(u16 ctrl, unsigned int tlp_flags)
* affects a Completion, and that one leaves Completions carrying the Relaxed
* Ordering attribute alone.
*/
-static enum pci_acs_p2pdma_state
+VISIBLE_IF_KUNIT enum pci_acs_p2pdma_state
pci_acs_p2pdma_completion(u16 ctrl, unsigned int tlp_flags)
{
if (tlp_flags & PCI_P2PDMA_TLP_RELAXED_CPL)
@@ -576,6 +545,7 @@ pci_acs_p2pdma_completion(u16 ctrl, unsigned int tlp_flags)
return ctrl & PCI_ACS_CR ? PCI_ACS_P2PDMA_REDIRECT :
PCI_ACS_P2PDMA_DIRECT;
}
+EXPORT_SYMBOL_IF_KUNIT(pci_acs_p2pdma_completion);
static const char *pci_acs_p2pdma_state_name(enum pci_acs_p2pdma_state state)
{
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index ba3c3fddddc2..302b79d3977f 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -2,6 +2,7 @@
#ifndef DRIVERS_PCI_H
#define DRIVERS_PCI_H
+#include <kunit/visibility.h>
#include <linux/bug.h>
#include <linux/align.h>
#include <linux/bitfield.h>
@@ -1093,6 +1094,46 @@ resource_size_t pci_min_window_alignment(struct pci_bus *bus,
void pci_acs_init(struct pci_dev *dev);
void pci_enable_acs(struct pci_dev *dev);
+
+/**
+ * enum pci_p2pdma_tlp_flags - Properties of the TLPs a client will issue
+ *
+ * These describe the traffic rather than the topology, and select which ACS
+ * controls apply along the peer-to-peer path. A value of 0 means strictly
+ * ordered Requests carrying an Untranslated address.
+ *
+ * @PCI_P2PDMA_TLP_TRANSLATED: Requests carry an ATS Translated address. PCIe
+ * r7.0 sec 6.12.3 routes those to the peer regardless of ACS P2P Request
+ * Redirect and ACS P2P Egress Control wherever ACS Direct Translated P2P
+ * is enabled.
+ * @PCI_P2PDMA_TLP_RELAXED_CPL: Non-Posted Requests carry the Relaxed
+ * Ordering attribute, and PCIe r7.0 sec 2.2.9.1 requires the Completer to
+ * return it in their Completions. Sec 6.12.1.1 never redirects those, so
+ * ACS P2P Completion Redirect does not gate the path.
+ */
+enum pci_p2pdma_tlp_flags {
+ PCI_P2PDMA_TLP_TRANSLATED = 1 << 0,
+ PCI_P2PDMA_TLP_RELAXED_CPL = 1 << 1,
+};
+
+/* Every combination of the flags above selects one routing class. */
+#define PCI_P2PDMA_TLP_CLASSES \
+ ((PCI_P2PDMA_TLP_TRANSLATED | PCI_P2PDMA_TLP_RELAXED_CPL) + 1)
+
+enum pci_acs_p2pdma_state {
+ PCI_ACS_P2PDMA_NOT_SUPPORTED,
+ PCI_ACS_P2PDMA_DIRECT,
+ PCI_ACS_P2PDMA_REDIRECT,
+ PCI_ACS_P2PDMA_BLOCKED,
+};
+
+#if IS_ENABLED(CONFIG_KUNIT)
+enum pci_acs_p2pdma_state pci_acs_p2pdma_request(u16 ctrl,
+ unsigned int tlp_flags);
+enum pci_acs_p2pdma_state pci_acs_p2pdma_completion(u16 ctrl,
+ unsigned int tlp_flags);
+#endif
+
#ifdef CONFIG_PCI_QUIRKS
int pci_dev_specific_acs_enabled(struct pci_dev *dev, u16 acs_flags);
int pci_dev_specific_enable_acs(struct pci_dev *dev);
diff --git a/drivers/pci/pci_acs_test.c b/drivers/pci/pci_acs_test.c
new file mode 100644
index 000000000000..ce6b9375da36
--- /dev/null
+++ b/drivers/pci/pci_acs_test.c
@@ -0,0 +1,121 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * KUnit tests for PCI ACS peer-to-peer routing decisions.
+ *
+ * These exercise Request and Completion routing independently of the ACS
+ * settings exposed by available PCIe hardware.
+ */
+#include <kunit/test.h>
+
+#include <linux/pci.h>
+#include <linux/pci-p2pdma.h>
+#include <linux/pci_regs.h>
+
+#include "pci.h"
+
+struct acs_decision_case {
+ const char *desc;
+ u16 ctrl;
+ unsigned int tlp_flags;
+ enum pci_acs_p2pdma_state expect;
+};
+
+/* Shorthands to keep the tables below readable. */
+#define ACS_DIRECT PCI_ACS_P2PDMA_DIRECT
+#define ACS_REDIR PCI_ACS_P2PDMA_REDIRECT
+#define ACS_RO PCI_P2PDMA_TLP_RELAXED_CPL
+#define ACS_AT PCI_P2PDMA_TLP_TRANSLATED
+#define ACS_BLOCK PCI_ACS_P2PDMA_BLOCKED
+
+/* Request routing ignores Completion Redirect. */
+static const struct acs_decision_case acs_request_cases[] = {
+ { "req/none", 0, 0, ACS_DIRECT },
+ { "req/rr", PCI_ACS_RR, 0, ACS_REDIR },
+ { "req/cr", PCI_ACS_CR, 0, ACS_DIRECT },
+ { "req/rr_cr", PCI_ACS_RR | PCI_ACS_CR, 0, ACS_REDIR },
+ { "req/ec", PCI_ACS_EC, 0, ACS_REDIR },
+ { "req/ec_cr", PCI_ACS_EC | PCI_ACS_CR, 0, ACS_REDIR },
+
+ /*
+ * Direct Translated P2P overrides the redirect controls, but only for
+ * a Request that actually carries a Translated address.
+ */
+ { "req/dt", PCI_ACS_DT, 0, ACS_DIRECT },
+ { "req/dt_rr", PCI_ACS_DT | PCI_ACS_RR, 0, ACS_REDIR },
+ { "req/at", 0, ACS_AT, ACS_DIRECT },
+ { "req/at_rr", PCI_ACS_RR, ACS_AT, ACS_REDIR },
+ { "req/at_dt_rr", PCI_ACS_DT | PCI_ACS_RR, ACS_AT, ACS_DIRECT },
+ { "req/at_dt_ec", PCI_ACS_DT | PCI_ACS_EC, ACS_AT, ACS_DIRECT },
+
+ /*
+ * Translation Blocking rejects a Translated address outright, and
+ * makes the port ignore Direct Translated P2P.
+ */
+ { "req/tb", PCI_ACS_TB, 0, ACS_DIRECT },
+ { "req/tb_rr", PCI_ACS_TB | PCI_ACS_RR, 0, ACS_REDIR },
+ { "req/at_tb", PCI_ACS_TB, ACS_AT, ACS_BLOCK },
+ { "req/at_tb_dt", PCI_ACS_TB | PCI_ACS_DT, ACS_AT, ACS_BLOCK },
+};
+
+/* Completion routing depends only on Completion Redirect. */
+static const struct acs_decision_case acs_completion_cases[] = {
+ { "cpl/none", 0, 0, ACS_DIRECT },
+ { "cpl/rr", PCI_ACS_RR, 0, ACS_DIRECT },
+ { "cpl/cr", PCI_ACS_CR, 0, ACS_REDIR },
+ { "cpl/rr_cr", PCI_ACS_RR | PCI_ACS_CR, 0, ACS_REDIR },
+ { "cpl/ec", PCI_ACS_EC, 0, ACS_DIRECT },
+ { "cpl/ec_cr", PCI_ACS_EC | PCI_ACS_CR, 0, ACS_REDIR },
+
+ /* Relaxed Ordering Completions are never redirected. */
+ { "cpl/ro", 0, ACS_RO, ACS_DIRECT },
+ { "cpl/ro_cr", PCI_ACS_CR, ACS_RO, ACS_DIRECT },
+ { "cpl/ro_rr_cr", PCI_ACS_RR | PCI_ACS_CR, ACS_RO, ACS_DIRECT },
+};
+
+#undef ACS_DIRECT
+#undef ACS_REDIR
+#undef ACS_RO
+#undef ACS_AT
+#undef ACS_BLOCK
+
+static void acs_decision_desc(const struct acs_decision_case *c, char *desc)
+{
+ strscpy(desc, c->desc, KUNIT_PARAM_DESC_SIZE);
+}
+
+KUNIT_ARRAY_PARAM(acs_request, acs_request_cases, acs_decision_desc);
+KUNIT_ARRAY_PARAM(acs_completion, acs_completion_cases, acs_decision_desc);
+
+static void pci_acs_p2pdma_request_test(struct kunit *test)
+{
+ const struct acs_decision_case *c = test->param_value;
+
+ KUNIT_EXPECT_EQ(test, pci_acs_p2pdma_request(c->ctrl, c->tlp_flags),
+ c->expect);
+}
+
+static void pci_acs_p2pdma_completion_test(struct kunit *test)
+{
+ const struct acs_decision_case *c = test->param_value;
+
+ KUNIT_EXPECT_EQ(test, pci_acs_p2pdma_completion(c->ctrl, c->tlp_flags),
+ c->expect);
+}
+
+static struct kunit_case pci_acs_test_cases[] = {
+ KUNIT_CASE_PARAM(pci_acs_p2pdma_request_test,
+ acs_request_gen_params),
+ KUNIT_CASE_PARAM(pci_acs_p2pdma_completion_test,
+ acs_completion_gen_params),
+ {}
+};
+
+static struct kunit_suite pci_acs_test_suite = {
+ .name = "pci_acs",
+ .test_cases = pci_acs_test_cases,
+};
+kunit_test_suite(pci_acs_test_suite);
+
+MODULE_IMPORT_NS("EXPORTED_FOR_KUNIT_TESTING");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("KUnit tests for PCI ACS peer-to-peer routing decisions");
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* [PATCH v10 13/18] PCI/P2PDMA: Test the ACS P2P routing walk
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (11 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 12/18] PCI/P2PDMA: Add KUnit tests for the ACS routing decisions Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:23 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 14/18] PCI: Add KUnit coverage for ACS isolation checks Leon Romanovsky
` (4 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
calc_map_type_and_dist() decides which ports along a path carry the
routing controls, and holds every class's answer in one cache entry.
Neither depends on a single register, so a table of them cannot reach the
walk itself.
Drive the walk over a fabricated fabric of two devices below a switch,
with fake config space supplying the ACS Control registers. Cover the
ports below the divergence, the three cases where two classes of one path
disagree, and the packed cache, which the fabric has no provider state to
exercise indirectly.
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
drivers/pci/p2pdma.c | 9 +-
drivers/pci/pci.h | 9 +
drivers/pci/pci_acs_test.c | 478 +++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 493 insertions(+), 3 deletions(-)
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index 93735f110b14..102930b114c1 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -1011,7 +1011,7 @@ static unsigned long map_types_idx(struct pci_dev *client)
*/
static_assert(PCI_P2PDMA_MAP_THRU_HOST_BRIDGE < 16);
-static unsigned long
+VISIBLE_IF_KUNIT unsigned long
pci_p2pdma_map_types_pack(const enum pci_p2pdma_map_type *type)
{
unsigned long val = 0;
@@ -1022,12 +1022,14 @@ pci_p2pdma_map_types_pack(const enum pci_p2pdma_map_type *type)
return val;
}
+EXPORT_SYMBOL_IF_KUNIT(pci_p2pdma_map_types_pack);
-static enum pci_p2pdma_map_type
+VISIBLE_IF_KUNIT enum pci_p2pdma_map_type
pci_p2pdma_map_types_unpack(unsigned long val, unsigned int tlp_flags)
{
return (val >> (tlp_flags * 4)) & 0xf;
}
+EXPORT_SYMBOL_IF_KUNIT(pci_p2pdma_map_types_unpack);
/*
* Calculate the P2PDMA mapping type and distance between two PCI devices.
@@ -1076,7 +1078,7 @@ pci_p2pdma_map_types_unpack(unsigned long val, unsigned int tlp_flags)
* ports per above. If the device is not in the whitelist, return
* PCI_P2PDMA_MAP_NOT_SUPPORTED.
*/
-static enum pci_p2pdma_map_type
+VISIBLE_IF_KUNIT enum pci_p2pdma_map_type
calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
int *dist, unsigned int tlp_flags, bool verbose)
{
@@ -1268,6 +1270,7 @@ calc_map_type_and_dist(struct pci_dev *provider, struct pci_dev *client,
}
return map_type[tlp_flags];
}
+EXPORT_SYMBOL_IF_KUNIT(calc_map_type_and_dist);
/**
* pci_p2pdma_distance_many - Determine the cumulative distance between
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index 302b79d3977f..b9b983ad6c34 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -7,6 +7,7 @@
#include <linux/align.h>
#include <linux/bitfield.h>
#include <linux/pci.h>
+#include <linux/pci-p2pdma.h>
#include <trace/events/pci.h>
struct pcie_tlp_log;
@@ -1132,6 +1133,14 @@ enum pci_acs_p2pdma_state pci_acs_p2pdma_request(u16 ctrl,
unsigned int tlp_flags);
enum pci_acs_p2pdma_state pci_acs_p2pdma_completion(u16 ctrl,
unsigned int tlp_flags);
+unsigned long pci_p2pdma_map_types_pack(const enum pci_p2pdma_map_type *type);
+enum pci_p2pdma_map_type pci_p2pdma_map_types_unpack(unsigned long val,
+ unsigned int tlp_flags);
+enum pci_p2pdma_map_type calc_map_type_and_dist(struct pci_dev *provider,
+ struct pci_dev *client,
+ int *dist,
+ unsigned int tlp_flags,
+ bool verbose);
#endif
#ifdef CONFIG_PCI_QUIRKS
diff --git a/drivers/pci/pci_acs_test.c b/drivers/pci/pci_acs_test.c
index ce6b9375da36..7f4f9cc04bfb 100644
--- a/drivers/pci/pci_acs_test.c
+++ b/drivers/pci/pci_acs_test.c
@@ -102,11 +102,489 @@ static void pci_acs_p2pdma_completion_test(struct kunit *test)
c->expect);
}
+/*
+ * Drive calc_map_type_and_dist() over a fabricated PCIe fabric matching the
+ * canonical topology of two devices below one switch:
+ *
+ * host bridge / root bus
+ * Root Port
+ * Switch Upstream Port
+ * Switch Downstream Port 0
+ * Nested Switch -- provider
+ * Switch Downstream Port 1
+ * Nested Switch -- client
+ *
+ * Fake config-space operations supply the ACS Control registers. This lets
+ * the cases vary both divergence ports and controls below the divergence
+ * without depending on real hardware.
+ */
+struct acs_port_cfg {
+ u16 ctrl;
+ bool fail_read;
+};
+
+struct acs_fabric {
+ struct pci_dev *rootport;
+ struct pci_dev *provider;
+ struct pci_dev *client;
+ struct pci_dev *dn0; /* Downstream Port 0 (provider side) */
+ struct pci_dev *dn1; /* Downstream Port 1 (client side) */
+ struct pci_dev *provider_leaf;
+ struct pci_dev *client_leaf;
+ struct acs_port_cfg dn0_cfg;
+ struct acs_port_cfg dn1_cfg;
+ struct acs_port_cfg provider_leaf_cfg;
+ struct acs_port_cfg client_leaf_cfg;
+ struct acs_port_cfg rootport_cfg;
+};
+
+static int acs_port_read(struct pci_dev *port, struct acs_port_cfg *cfg,
+ int where, int size, u32 *val)
+{
+ if (port->acs_cap && size == 2 &&
+ where == port->acs_cap + PCI_ACS_CTRL) {
+ if (cfg->fail_read)
+ return PCIBIOS_DEVICE_NOT_FOUND;
+ *val = cfg->ctrl;
+ }
+
+ return PCIBIOS_SUCCESSFUL;
+}
+
+static int acs_fabric_read(struct pci_bus *bus, unsigned int devfn,
+ int where, int size, u32 *val)
+{
+ struct acs_fabric *f = bus->sysdata;
+
+ *val = 0;
+ if (bus == f->rootport->bus && devfn == f->rootport->devfn)
+ return acs_port_read(f->rootport, &f->rootport_cfg,
+ where, size, val);
+ if (bus == f->dn0->bus && devfn == f->dn0->devfn)
+ return acs_port_read(f->dn0, &f->dn0_cfg, where, size, val);
+ if (bus == f->dn1->bus && devfn == f->dn1->devfn)
+ return acs_port_read(f->dn1, &f->dn1_cfg, where, size, val);
+ if (bus == f->provider_leaf->bus &&
+ devfn == f->provider_leaf->devfn)
+ return acs_port_read(f->provider_leaf, &f->provider_leaf_cfg,
+ where, size, val);
+ if (bus == f->client_leaf->bus && devfn == f->client_leaf->devfn)
+ return acs_port_read(f->client_leaf, &f->client_leaf_cfg,
+ where, size, val);
+
+ return PCIBIOS_SUCCESSFUL;
+}
+
+static int acs_fabric_write(struct pci_bus *bus, unsigned int devfn,
+ int where, int size, u32 val)
+{
+ return PCIBIOS_SUCCESSFUL;
+}
+
+static struct pci_ops acs_fabric_ops = {
+ .read = acs_fabric_read,
+ .write = acs_fabric_write,
+};
+
+static struct pci_bus *acs_add_bus(struct kunit *test, struct pci_bus *parent,
+ struct pci_dev *self, u8 nr, void *sysdata)
+{
+ struct pci_bus *bus = kunit_kzalloc(test, sizeof(*bus), GFP_KERNEL);
+
+ KUNIT_ASSERT_NOT_NULL(test, bus);
+ bus->parent = parent;
+ bus->self = self;
+ bus->number = nr;
+ bus->ops = &acs_fabric_ops;
+ bus->sysdata = sysdata;
+ INIT_LIST_HEAD(&bus->devices);
+ return bus;
+}
+
+static struct pci_dev *acs_add_dev(struct kunit *test, struct pci_bus *bus,
+ unsigned int devfn, int pcie_type)
+{
+ struct pci_dev *dev = kunit_kzalloc(test, sizeof(*dev), GFP_KERNEL);
+
+ KUNIT_ASSERT_NOT_NULL(test, dev);
+ dev->bus = bus;
+ dev->devfn = devfn;
+ dev->pcie_cap = 0x40;
+ dev->pcie_flags_reg = (pcie_type << 4) | 0x2;
+ list_add_tail(&dev->bus_list, &bus->devices);
+ return dev;
+}
+
+static void acs_build_fabric(struct kunit *test, struct acs_fabric *f)
+{
+ struct pci_bus *bus0, *bus1, *bus2, *bus3, *bus4, *bus5, *bus6;
+ struct pci_bus *bus7, *bus8;
+ struct pci_dev *swup, *provider_swup, *client_swup;
+ struct pci_dev *rootport;
+ struct pci_host_bridge *host;
+
+ host = kunit_kzalloc(test, sizeof(*host), GFP_KERNEL);
+ KUNIT_ASSERT_NOT_NULL(test, host);
+
+ bus0 = acs_add_bus(test, NULL, NULL, 0, f);
+ /* The Root Port doubles as the whitelisted host-bridge device. */
+ rootport = acs_add_dev(test, bus0, PCI_DEVFN(0, 0),
+ PCI_EXP_TYPE_ROOT_PORT);
+ f->rootport = rootport;
+ rootport->vendor = PCI_VENDOR_ID_GOOGLE;
+ rootport->device = 0x1234;
+ host->bus = bus0;
+ bus0->bridge = &host->dev;
+
+ bus1 = acs_add_bus(test, bus0, rootport, 1, f);
+ swup = acs_add_dev(test, bus1, PCI_DEVFN(0, 0),
+ PCI_EXP_TYPE_UPSTREAM);
+
+ bus2 = acs_add_bus(test, bus1, swup, 2, f);
+ f->dn0 = acs_add_dev(test, bus2, PCI_DEVFN(0, 0),
+ PCI_EXP_TYPE_DOWNSTREAM);
+ f->dn1 = acs_add_dev(test, bus2, PCI_DEVFN(1, 0),
+ PCI_EXP_TYPE_DOWNSTREAM);
+
+ bus3 = acs_add_bus(test, bus2, f->dn0, 3, f);
+ provider_swup = acs_add_dev(test, bus3, PCI_DEVFN(0, 0),
+ PCI_EXP_TYPE_UPSTREAM);
+ bus5 = acs_add_bus(test, bus3, provider_swup, 5, f);
+ f->provider_leaf = acs_add_dev(test, bus5, PCI_DEVFN(0, 0),
+ PCI_EXP_TYPE_DOWNSTREAM);
+ bus7 = acs_add_bus(test, bus5, f->provider_leaf, 7, f);
+ f->provider = acs_add_dev(test, bus7, PCI_DEVFN(0, 0),
+ PCI_EXP_TYPE_ENDPOINT);
+
+ bus4 = acs_add_bus(test, bus2, f->dn1, 4, f);
+ client_swup = acs_add_dev(test, bus4, PCI_DEVFN(0, 0),
+ PCI_EXP_TYPE_UPSTREAM);
+ bus6 = acs_add_bus(test, bus4, client_swup, 6, f);
+ f->client_leaf = acs_add_dev(test, bus6, PCI_DEVFN(0, 0),
+ PCI_EXP_TYPE_DOWNSTREAM);
+ bus8 = acs_add_bus(test, bus6, f->client_leaf, 8, f);
+ f->client = acs_add_dev(test, bus8, PCI_DEVFN(0, 0),
+ PCI_EXP_TYPE_ENDPOINT);
+}
+
+static enum pci_p2pdma_map_type acs_walk_map(struct acs_fabric *f,
+ unsigned int tlp_flags)
+{
+ int dist;
+
+ return calc_map_type_and_dist(f->provider, f->client, &dist, tlp_flags,
+ false);
+}
+
+static void acs_walk_bus_addr_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+
+ acs_build_fabric(test, &f);
+ KUNIT_EXPECT_EQ(test, acs_walk_map(&f, 0), PCI_P2PDMA_MAP_BUS_ADDR);
+}
+
+static void acs_walk_request_redirect_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+
+ acs_build_fabric(test, &f);
+ f.dn1->acs_cap = 0x100;
+ f.dn1->acs_capabilities = PCI_ACS_RR;
+ f.dn1_cfg.ctrl = PCI_ACS_RR;
+
+ KUNIT_EXPECT_EQ(test, acs_walk_map(&f, 0),
+ PCI_P2PDMA_MAP_THRU_HOST_BRIDGE);
+}
+
+static void acs_walk_completion_redirect_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+
+ acs_build_fabric(test, &f);
+ f.dn0->acs_cap = 0x100;
+ f.dn0->acs_capabilities = PCI_ACS_CR;
+ f.dn0_cfg.ctrl = PCI_ACS_CR;
+
+ KUNIT_EXPECT_EQ(test, acs_walk_map(&f, 0),
+ PCI_P2PDMA_MAP_THRU_HOST_BRIDGE);
+}
+
+static void acs_walk_egress_control_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+
+ acs_build_fabric(test, &f);
+ f.dn1->acs_cap = 0x100;
+ f.dn1->acs_capabilities = PCI_ACS_EC;
+ f.dn1_cfg.ctrl = PCI_ACS_EC;
+
+ KUNIT_EXPECT_EQ(test, acs_walk_map(&f, 0),
+ PCI_P2PDMA_MAP_THRU_HOST_BRIDGE);
+}
+
+static void acs_walk_asymmetric_direct_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+
+ acs_build_fabric(test, &f);
+ /* These controls affect only the reverse transaction directions. */
+ f.dn0->acs_cap = 0x100;
+ f.dn0->acs_capabilities = PCI_ACS_RR | PCI_ACS_EC;
+ f.dn0_cfg.ctrl = PCI_ACS_RR | PCI_ACS_EC;
+ f.dn1->acs_cap = 0x100;
+ f.dn1->acs_capabilities = PCI_ACS_CR;
+ f.dn1_cfg.ctrl = PCI_ACS_CR;
+
+ KUNIT_EXPECT_EQ(test, acs_walk_map(&f, 0), PCI_P2PDMA_MAP_BUS_ADDR);
+}
+
+static void acs_walk_nested_completion_redirect_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+
+ acs_build_fabric(test, &f);
+ f.provider_leaf->acs_cap = 0x100;
+ f.provider_leaf->acs_capabilities = PCI_ACS_CR;
+ f.provider_leaf_cfg.ctrl = PCI_ACS_CR;
+
+ KUNIT_EXPECT_EQ(test, acs_walk_map(&f, 0), PCI_P2PDMA_MAP_BUS_ADDR);
+}
+
+static void acs_walk_nested_request_redirect_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+
+ acs_build_fabric(test, &f);
+ f.client_leaf->acs_cap = 0x100;
+ f.client_leaf->acs_capabilities = PCI_ACS_RR;
+ f.client_leaf_cfg.ctrl = PCI_ACS_RR;
+
+ KUNIT_EXPECT_EQ(test, acs_walk_map(&f, 0), PCI_P2PDMA_MAP_BUS_ADDR);
+}
+
+static void acs_walk_translation_blocking_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+
+ acs_build_fabric(test, &f);
+ f.client_leaf->acs_cap = 0x100;
+ f.client_leaf->acs_capabilities = PCI_ACS_TB;
+ f.client_leaf_cfg.ctrl = PCI_ACS_TB;
+
+ /* Untranslated Requests are unaffected by Translation Blocking. */
+ KUNIT_EXPECT_EQ(test, acs_walk_map(&f, 0), PCI_P2PDMA_MAP_BUS_ADDR);
+ KUNIT_EXPECT_EQ(test, acs_walk_map(&f, PCI_P2PDMA_TLP_TRANSLATED),
+ PCI_P2PDMA_MAP_NOT_SUPPORTED);
+}
+
+static void acs_walk_relaxed_completion_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+
+ acs_build_fabric(test, &f);
+ f.dn0->acs_cap = 0x100;
+ f.dn0->acs_capabilities = PCI_ACS_CR;
+ f.dn0_cfg.ctrl = PCI_ACS_CR;
+
+ KUNIT_EXPECT_EQ(test, acs_walk_map(&f, 0),
+ PCI_P2PDMA_MAP_THRU_HOST_BRIDGE);
+ KUNIT_EXPECT_EQ(test, acs_walk_map(&f, PCI_P2PDMA_TLP_RELAXED_CPL),
+ PCI_P2PDMA_MAP_BUS_ADDR);
+}
+
+static void acs_walk_direct_translated_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+
+ acs_build_fabric(test, &f);
+ f.dn1->acs_cap = 0x100;
+ f.dn1->acs_capabilities = PCI_ACS_RR | PCI_ACS_DT;
+ f.dn1_cfg.ctrl = PCI_ACS_RR | PCI_ACS_DT;
+
+ KUNIT_EXPECT_EQ(test, acs_walk_map(&f, 0),
+ PCI_P2PDMA_MAP_THRU_HOST_BRIDGE);
+ KUNIT_EXPECT_EQ(test, acs_walk_map(&f, PCI_P2PDMA_TLP_TRANSLATED),
+ PCI_P2PDMA_MAP_BUS_ADDR);
+}
+
+/*
+ * The cache stores one packed value per client, so every class has to come
+ * back out under the flags that selected it.
+ */
+static void acs_map_types_pack_test(struct kunit *test)
+{
+ static const enum pci_p2pdma_map_type type[PCI_P2PDMA_TLP_CLASSES] = {
+ [0] = PCI_P2PDMA_MAP_THRU_HOST_BRIDGE,
+ [PCI_P2PDMA_TLP_TRANSLATED] = PCI_P2PDMA_MAP_NOT_SUPPORTED,
+ [PCI_P2PDMA_TLP_RELAXED_CPL] = PCI_P2PDMA_MAP_BUS_ADDR,
+ [PCI_P2PDMA_TLP_TRANSLATED | PCI_P2PDMA_TLP_RELAXED_CPL] =
+ PCI_P2PDMA_MAP_UNKNOWN,
+ };
+ unsigned long packed = pci_p2pdma_map_types_pack(type);
+ unsigned int flags;
+
+ for (flags = 0; flags < PCI_P2PDMA_TLP_CLASSES; flags++)
+ KUNIT_EXPECT_EQ(test,
+ pci_p2pdma_map_types_unpack(packed, flags),
+ type[flags]);
+
+ /* An absent cache entry reads back as unknown in every class. */
+ for (flags = 0; flags < PCI_P2PDMA_TLP_CLASSES; flags++)
+ KUNIT_EXPECT_EQ(test, pci_p2pdma_map_types_unpack(0, flags),
+ PCI_P2PDMA_MAP_UNKNOWN);
+}
+
+/*
+ * The provider can be an ancestor of the client, which leaves no divergence
+ * to evaluate. Translation Blocking still applies to every port the Request
+ * passes on its way up.
+ */
+static void acs_walk_ancestor_provider_tb_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+ int dist;
+
+ acs_build_fabric(test, &f);
+ f.client_leaf->acs_cap = 0x100;
+ f.client_leaf->acs_capabilities = PCI_ACS_TB;
+ f.client_leaf_cfg.ctrl = PCI_ACS_TB;
+
+ KUNIT_EXPECT_EQ(test,
+ calc_map_type_and_dist(f.dn1, f.client, &dist, 0,
+ false),
+ PCI_P2PDMA_MAP_BUS_ADDR);
+ KUNIT_EXPECT_EQ(test,
+ calc_map_type_and_dist(f.dn1, f.client, &dist,
+ PCI_P2PDMA_TLP_TRANSLATED,
+ false),
+ PCI_P2PDMA_MAP_NOT_SUPPORTED);
+}
+
+/*
+ * Without a common upstream bridge the Request still climbs towards the host
+ * bridge, so Translation Blocking on the way withdraws the Translated classes
+ * there too.
+ */
+static void acs_walk_no_common_bridge_tb_test(struct kunit *test)
+{
+ struct acs_fabric f = {}, g = {};
+ int dist;
+
+ acs_build_fabric(test, &f);
+ acs_build_fabric(test, &g);
+ f.client_leaf->acs_cap = 0x100;
+ f.client_leaf->acs_capabilities = PCI_ACS_TB;
+ f.client_leaf_cfg.ctrl = PCI_ACS_TB;
+
+ KUNIT_EXPECT_EQ(test,
+ calc_map_type_and_dist(g.provider, f.client, &dist, 0,
+ false),
+ PCI_P2PDMA_MAP_THRU_HOST_BRIDGE);
+ KUNIT_EXPECT_EQ(test,
+ calc_map_type_and_dist(g.provider, f.client, &dist,
+ PCI_P2PDMA_TLP_TRANSLATED,
+ false),
+ PCI_P2PDMA_MAP_NOT_SUPPORTED);
+}
+
+/*
+ * A Request between the client and itself never leaves the device, so no port
+ * is in a position to inspect its Address Type. Translation Blocking directly
+ * above the client must not withdraw the Translated classes.
+ */
+static void acs_walk_self_dma_tb_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+ int dist;
+
+ acs_build_fabric(test, &f);
+ f.client_leaf->acs_cap = 0x100;
+ f.client_leaf->acs_capabilities = PCI_ACS_TB;
+ f.client_leaf_cfg.ctrl = PCI_ACS_TB;
+
+ KUNIT_EXPECT_EQ(test,
+ calc_map_type_and_dist(f.client, f.client, &dist, 0,
+ false),
+ PCI_P2PDMA_MAP_BUS_ADDR);
+ KUNIT_EXPECT_EQ(test,
+ calc_map_type_and_dist(f.client, f.client, &dist,
+ PCI_P2PDMA_TLP_TRANSLATED,
+ false),
+ PCI_P2PDMA_MAP_BUS_ADDR);
+}
+
+/*
+ * A direct route turns around at the divergence, so Translation Blocking above
+ * it does not touch one. The host bridge route keeps climbing past that port,
+ * and the Translated classes have to go without it.
+ */
+static void acs_walk_tb_above_divergence_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+ int dist;
+
+ acs_build_fabric(test, &f);
+ f.rootport->acs_cap = 0x100;
+ f.rootport->acs_capabilities = PCI_ACS_TB;
+ f.rootport_cfg.ctrl = PCI_ACS_TB;
+
+ /* Nothing redirects yet, so the Request never reaches the Root Port. */
+ KUNIT_EXPECT_EQ(test,
+ calc_map_type_and_dist(f.provider, f.client, &dist,
+ PCI_P2PDMA_TLP_TRANSLATED,
+ false),
+ PCI_P2PDMA_MAP_BUS_ADDR);
+
+ /* Request Redirect sends it up past the Root Port instead. */
+ f.dn1->acs_cap = 0x100;
+ f.dn1->acs_capabilities = PCI_ACS_RR;
+ f.dn1_cfg.ctrl = PCI_ACS_RR;
+
+ KUNIT_EXPECT_EQ(test,
+ calc_map_type_and_dist(f.provider, f.client, &dist, 0,
+ false),
+ PCI_P2PDMA_MAP_THRU_HOST_BRIDGE);
+ KUNIT_EXPECT_EQ(test,
+ calc_map_type_and_dist(f.provider, f.client, &dist,
+ PCI_P2PDMA_TLP_TRANSLATED,
+ false),
+ PCI_P2PDMA_MAP_NOT_SUPPORTED);
+}
+
+static void acs_walk_unreadable_control_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+
+ acs_build_fabric(test, &f);
+ f.dn1->acs_cap = 0x100;
+ f.dn1_cfg.fail_read = true;
+
+ KUNIT_EXPECT_EQ(test, acs_walk_map(&f, 0),
+ PCI_P2PDMA_MAP_NOT_SUPPORTED);
+}
+
static struct kunit_case pci_acs_test_cases[] = {
KUNIT_CASE_PARAM(pci_acs_p2pdma_request_test,
acs_request_gen_params),
KUNIT_CASE_PARAM(pci_acs_p2pdma_completion_test,
acs_completion_gen_params),
+ KUNIT_CASE(acs_walk_bus_addr_test),
+ KUNIT_CASE(acs_walk_request_redirect_test),
+ KUNIT_CASE(acs_walk_completion_redirect_test),
+ KUNIT_CASE(acs_walk_egress_control_test),
+ KUNIT_CASE(acs_walk_asymmetric_direct_test),
+ KUNIT_CASE(acs_walk_nested_completion_redirect_test),
+ KUNIT_CASE(acs_walk_nested_request_redirect_test),
+ KUNIT_CASE(acs_walk_translation_blocking_test),
+ KUNIT_CASE(acs_walk_ancestor_provider_tb_test),
+ KUNIT_CASE(acs_walk_no_common_bridge_tb_test),
+ KUNIT_CASE(acs_walk_self_dma_tb_test),
+ KUNIT_CASE(acs_walk_relaxed_completion_test),
+ KUNIT_CASE(acs_walk_direct_translated_test),
+ KUNIT_CASE(acs_walk_tb_above_divergence_test),
+ KUNIT_CASE(acs_walk_unreadable_control_test),
+ KUNIT_CASE(acs_map_types_pack_test),
{}
};
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* Re: [PATCH v10 13/18] PCI/P2PDMA: Test the ACS P2P routing walk
2026-10-12 13:07 ` [PATCH v10 13/18] PCI/P2PDMA: Test the ACS P2P routing walk Leon Romanovsky
@ 2026-10-12 13:23 ` sashiko-bot
0 siblings, 0 replies; 38+ messages in thread
From: sashiko-bot @ 2026-10-12 13:23 UTC (permalink / raw)
To: Leon Romanovsky; +Cc: linux-rdma, linux-pci, media-ci, dri-devel, kvm
> From: Leon Romanovsky <leonro@nvidia.com>
>
> calc_map_type_and_dist() decides which ports along a path carry the
> routing controls, and holds every class's answer in one cache entry.
> Neither depends on a single register, so a table of them cannot reach the
> walk itself.
>
> Drive the walk over a fabricated fabric of two devices below a switch,
> with fake config space supplying the ACS Control registers. Cover the
> ports below the divergence, the three cases where two classes of one path
> disagree, and the packed cache, which the fabric has no provider state to
> exercise indirectly.
>
> Tested-by: Tushar Dave <tdave@nvidia.com>
> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Sashiko has reviewed this patch and found no issues. It looks great!
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261012-fix-p2p-acs-v4-0-v10-0-6ffe7b7a4517@nvidia.com?part=13
^ permalink raw reply [flat|nested] 38+ messages in thread
* [PATCH v10 14/18] PCI: Add KUnit coverage for ACS isolation checks
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (12 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 13/18] PCI/P2PDMA: Test the ACS P2P routing walk Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:20 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 15/18] PCI/P2PDMA: Document TLP-class routing Leon Romanovsky
` (3 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
Direct Translated P2P does not weaken IOMMU isolation because a Translated
Request carries an address supplied by the IOMMU. Config-space read
failures, however, leave ACS state unknown and must not report isolation.
Exercise both cases with fake config-space operations. Also cover missing
and unrequested controls and a missing ACS capability.
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
drivers/pci/pci.c | 4 +-
drivers/pci/pci.h | 1 +
drivers/pci/pci_acs_test.c | 138 +++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 142 insertions(+), 1 deletion(-)
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index f7d94ecf9157..4a9ab3882aac 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -3578,7 +3578,8 @@ void pci_configure_ari(struct pci_dev *dev)
}
}
-static bool pci_acs_flags_enabled(struct pci_dev *pdev, u16 acs_flags)
+VISIBLE_IF_KUNIT
+bool pci_acs_flags_enabled(struct pci_dev *pdev, u16 acs_flags)
{
int pos;
u16 ctrl;
@@ -3598,6 +3599,7 @@ static bool pci_acs_flags_enabled(struct pci_dev *pdev, u16 acs_flags)
return false;
return (ctrl & acs_flags) == acs_flags;
}
+EXPORT_SYMBOL_IF_KUNIT(pci_acs_flags_enabled);
/**
* pci_acs_enabled - test ACS against required flags for a given device
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index b9b983ad6c34..99852cf99732 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -1129,6 +1129,7 @@ enum pci_acs_p2pdma_state {
};
#if IS_ENABLED(CONFIG_KUNIT)
+bool pci_acs_flags_enabled(struct pci_dev *pdev, u16 acs_flags);
enum pci_acs_p2pdma_state pci_acs_p2pdma_request(u16 ctrl,
unsigned int tlp_flags);
enum pci_acs_p2pdma_state pci_acs_p2pdma_completion(u16 ctrl,
diff --git a/drivers/pci/pci_acs_test.c b/drivers/pci/pci_acs_test.c
index 7f4f9cc04bfb..6a2cd38d968f 100644
--- a/drivers/pci/pci_acs_test.c
+++ b/drivers/pci/pci_acs_test.c
@@ -102,6 +102,140 @@ static void pci_acs_p2pdma_completion_test(struct kunit *test)
c->expect);
}
+/* Flags an IOMMU asks for; see REQ_ACS_FLAGS in drivers/iommu/iommu.c. */
+#define ACS_REQ_FLAGS (PCI_ACS_SV | PCI_ACS_RR | PCI_ACS_CR | PCI_ACS_UF)
+#define ACS_ALL_CAPS (PCI_ACS_SV | PCI_ACS_TB | PCI_ACS_RR | PCI_ACS_CR | \
+ PCI_ACS_UF | PCI_ACS_DT)
+#define ACS_TEST_CAP 0x100
+
+struct acs_ctrl_cfg {
+ unsigned int devfn;
+ u16 cap; /* Offset where the ACS capability responds */
+ u16 ctrl;
+ bool fail_read;
+};
+
+static int acs_ctrl_read(struct pci_bus *bus, unsigned int devfn,
+ int where, int size, u32 *val)
+{
+ struct acs_ctrl_cfg *cfg = bus->sysdata;
+
+ *val = 0;
+ if (cfg->fail_read)
+ return PCIBIOS_DEVICE_NOT_FOUND;
+
+ if (devfn == cfg->devfn && size == 2 &&
+ where == cfg->cap + PCI_ACS_CTRL)
+ *val = cfg->ctrl;
+ return PCIBIOS_SUCCESSFUL;
+}
+
+static int acs_ctrl_write(struct pci_bus *bus, unsigned int devfn,
+ int where, int size, u32 val)
+{
+ return PCIBIOS_SUCCESSFUL;
+}
+
+static struct pci_ops acs_ctrl_ops = {
+ .read = acs_ctrl_read,
+ .write = acs_ctrl_write,
+};
+
+struct acs_isolation_case {
+ const char *desc;
+ u16 ctrl;
+ u16 req;
+ bool expect;
+};
+
+static const struct acs_isolation_case acs_isolation_cases[] = {
+ { "all_enabled", ACS_REQ_FLAGS, ACS_REQ_FLAGS, true },
+ /* Translated Requests remain isolated by their IOMMU translation. */
+ { "dt", ACS_REQ_FLAGS | PCI_ACS_DT, ACS_REQ_FLAGS, true },
+ { "rr_not_enabled", PCI_ACS_SV | PCI_ACS_CR | PCI_ACS_UF,
+ ACS_REQ_FLAGS, false },
+ { "rr_not_required", PCI_ACS_SV | PCI_ACS_CR | PCI_ACS_UF,
+ PCI_ACS_SV | PCI_ACS_CR | PCI_ACS_UF, true },
+};
+
+static void acs_isolation_desc(const struct acs_isolation_case *c, char *desc)
+{
+ strscpy(desc, c->desc, KUNIT_PARAM_DESC_SIZE);
+}
+
+KUNIT_ARRAY_PARAM(acs_isolation, acs_isolation_cases, acs_isolation_desc);
+
+static void pci_acs_flags_enabled_test(struct kunit *test)
+{
+ const struct acs_isolation_case *c = test->param_value;
+ struct acs_ctrl_cfg cfg = {
+ .devfn = PCI_DEVFN(0, 0),
+ .cap = ACS_TEST_CAP,
+ .ctrl = c->ctrl,
+ };
+ struct pci_bus *bus = kunit_kzalloc(test, sizeof(*bus), GFP_KERNEL);
+ struct pci_dev *pdev = kunit_kzalloc(test, sizeof(*pdev), GFP_KERNEL);
+
+ KUNIT_ASSERT_NOT_NULL(test, bus);
+ KUNIT_ASSERT_NOT_NULL(test, pdev);
+
+ bus->ops = &acs_ctrl_ops;
+ bus->sysdata = &cfg;
+
+ pdev->bus = bus;
+ pdev->devfn = cfg.devfn;
+ pdev->acs_cap = ACS_TEST_CAP;
+ pdev->acs_capabilities = ACS_ALL_CAPS;
+
+ KUNIT_EXPECT_EQ(test, pci_acs_flags_enabled(pdev, c->req), c->expect);
+}
+
+static bool acs_isolated(struct kunit *test, struct acs_ctrl_cfg *cfg,
+ u16 acs_cap, u16 acs_flags)
+{
+ struct pci_bus *bus = kunit_kzalloc(test, sizeof(*bus), GFP_KERNEL);
+ struct pci_dev *pdev = kunit_kzalloc(test, sizeof(*pdev), GFP_KERNEL);
+
+ KUNIT_ASSERT_NOT_NULL(test, bus);
+ KUNIT_ASSERT_NOT_NULL(test, pdev);
+
+ bus->ops = &acs_ctrl_ops;
+ bus->sysdata = cfg;
+
+ pdev->bus = bus;
+ pdev->devfn = cfg->devfn;
+ pdev->acs_cap = acs_cap;
+ pdev->acs_capabilities = ACS_ALL_CAPS;
+
+ return pci_acs_flags_enabled(pdev, acs_flags);
+}
+
+static void pci_acs_flags_no_cap_test(struct kunit *test)
+{
+ struct acs_ctrl_cfg cfg = {
+ .devfn = PCI_DEVFN(0, 0),
+ .cap = 0,
+ .ctrl = ACS_REQ_FLAGS,
+ };
+
+ KUNIT_EXPECT_FALSE(test, acs_isolated(test, &cfg, 0, ACS_REQ_FLAGS));
+}
+
+static void pci_acs_flags_read_fails_test(struct kunit *test)
+{
+ u16 no_rr = ACS_REQ_FLAGS & ~PCI_ACS_RR;
+ struct acs_ctrl_cfg cfg = {
+ .devfn = PCI_DEVFN(0, 0),
+ .cap = ACS_TEST_CAP,
+ .ctrl = ACS_REQ_FLAGS,
+ };
+
+ KUNIT_EXPECT_TRUE(test, acs_isolated(test, &cfg, ACS_TEST_CAP, no_rr));
+
+ cfg.fail_read = true;
+ KUNIT_EXPECT_FALSE(test, acs_isolated(test, &cfg, ACS_TEST_CAP, no_rr));
+}
+
/*
* Drive calc_map_type_and_dist() over a fabricated PCIe fabric matching the
* canonical topology of two devices below one switch:
@@ -569,6 +703,10 @@ static struct kunit_case pci_acs_test_cases[] = {
acs_request_gen_params),
KUNIT_CASE_PARAM(pci_acs_p2pdma_completion_test,
acs_completion_gen_params),
+ KUNIT_CASE_PARAM(pci_acs_flags_enabled_test,
+ acs_isolation_gen_params),
+ KUNIT_CASE(pci_acs_flags_no_cap_test),
+ KUNIT_CASE(pci_acs_flags_read_fails_test),
KUNIT_CASE(acs_walk_bus_addr_test),
KUNIT_CASE(acs_walk_request_redirect_test),
KUNIT_CASE(acs_walk_completion_redirect_test),
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* [PATCH v10 15/18] PCI/P2PDMA: Document TLP-class routing
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (13 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 14/18] PCI: Add KUnit coverage for ACS isolation checks Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:16 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 16/18] PCI/P2PDMA: Let a client declare that it selects ATS per mapping Leon Romanovsky
` (2 subsequent siblings)
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
The P2PDMA documentation stated that the mapping result is not defined for
Relaxed Ordering or ATS-translated Requests. It now is.
Replace that paragraph with what the three TLP-sensitive ACS controls do
and the table of outcomes per class.
Tested-by: Tushar Dave <tdave@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
Documentation/driver-api/pci/p2pdma.rst | 60 +++++++++++++++++++++++++++++----
1 file changed, 54 insertions(+), 6 deletions(-)
diff --git a/Documentation/driver-api/pci/p2pdma.rst b/Documentation/driver-api/pci/p2pdma.rst
index 40f5268ba9db..52725e58f656 100644
--- a/Documentation/driver-api/pci/p2pdma.rst
+++ b/Documentation/driver-api/pci/p2pdma.rst
@@ -33,12 +33,60 @@ through the host bridge when either applicable port redirects. If an ACS
Control register cannot be read, P2P DMA is rejected because the kernel cannot
establish a usable route.
-This evaluation assumes clients issue strictly ordered Requests carrying an
-Untranslated address. Its result is not defined when clients use Relaxed
-Ordering or issue ATS-translated Requests because those TLP attributes can
-select different routes through the fabric. Unless ACS Translation Blocking
-is enabled, a Port with ACS Direct Translated P2P enabled routes a
-Translated Request directly to the peer regardless of the redirect controls.
+Three of those controls act on TLP attributes that the Requester chooses rather
+than on the topology, so the same path routes differently for different
+traffic. ACS Translation Blocking rejects any Request whose Address Type is
+not Untranslated, and takes precedence over every other P2P control. ACS
+Direct Translated P2P routes a Translated Request to the peer regardless of
+Request Redirect and Egress Control. ACS Completion Redirect leaves alone
+Completions that carry the Relaxed Ordering attribute, which a Completion
+copies from the Request it answers.
+
+P2PDMA therefore decides each class of traffic separately, and
+``pci_p2pdma_map_type()`` answers for the one a DMA mapping is built for:
+strictly ordered Requests carrying an Untranslated address.
+
+The two directions are decided independently. Translation Blocking (TB),
+Direct Translated P2P (DT), Request Redirect (RR) and Egress Control (EC) on
+the Requester-side port decide the Request:
+
+===== ===== ======= ============ ==========
+TB DT RR/EC TLP class Request
+===== ===== ======= ============ ==========
+set x x translated blocked
+clear set x translated direct
+clear clear clear translated direct
+clear clear set translated redirected
+x x clear untranslated direct
+x x set untranslated redirected
+===== ===== ======= ============ ==========
+
+Completion Redirect (CR) on the Completer-side port decides the Completions:
+
+===== ========= ==========
+CR TLP class Completion
+===== ========= ==========
+x relaxed direct
+clear strict direct
+set strict redirected
+===== ========= ==========
+
+Combining the two tables gives the mapping type. A class is bus addressable
+when its Request and its Completions both route directly, and anything else
+goes through the host bridge. A blocked Request has neither route, because
+Translation Blocking rejects the Address Type itself rather than the
+destination. With no ACS control set anywhere, every class routes directly.
+
+Note that DT only matters where RR or EC would otherwise redirect: it
+overrides them for a Translated address rather than granting a direct route
+that was not already there.
+
+Translation Blocking is not a routing control, so it is evaluated on every
+port the Request passes rather than at the divergence alone. A direct route
+turns around at the divergence and only passes the ports below it, while the
+host bridge route keeps climbing and passes that port and everything above it
+as well. Neither route falls back to the other, because the Address Type is
+rejected wherever the Request is addressed.
However, if the P2P transaction reaches the host bridge then it might have to
hairpin back out the same root port, be routed inside the CPU SOC to another
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* [PATCH v10 16/18] PCI/P2PDMA: Let a client declare that it selects ATS per mapping
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (14 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 15/18] PCI/P2PDMA: Document TLP-class routing Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:26 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 17/18] PCI/P2PDMA: Evaluate the ATS path for clients with ATS enabled Leon Romanovsky
2026-10-12 13:07 ` [PATCH v10 18/18] PCI/P2PDMA: Test the routing of " Leon Romanovsky
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
The PCIe ATS Enable bit covers the whole device, so P2PDMA cannot tell
from it whether a client will translate a given address. Most devices
translate any address once ATS is enabled, but some choose ATS per DMA
mapping and can still use bus addresses for the rest.
Add pcim_p2pdma_set_ats_per_mapping() so the driver of such a device can
declare that before P2PDMA starts treating clients with ATS enabled as
translating everything. Keep it in the device's P2PDMA state, which
already ends with the driver binding.
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
drivers/pci/p2pdma.c | 27 +++++++++++++++++++++++++++
include/linux/pci-p2pdma.h | 4 ++++
2 files changed, 31 insertions(+)
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index 102930b114c1..e38081f79761 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -26,6 +26,7 @@
struct pci_p2pdma {
struct gen_pool *pool;
bool p2pmem_published;
+ bool ats_per_mapping;
struct xarray map_types;
struct p2pdma_provider mem[PCI_STD_NUM_BARS];
};
@@ -1575,6 +1576,32 @@ void pci_p2pmem_publish(struct pci_dev *pdev, bool publish)
}
EXPORT_SYMBOL_GPL(pci_p2pmem_publish);
+/**
+ * pcim_p2pdma_set_ats_per_mapping - Declare per-mapping ATS for a client
+ * @pdev: PCI device that initiates peer-to-peer DMA
+ *
+ * Declare that @pdev issues Translated Requests only for the DMA mappings its
+ * driver sets up to use ATS, rather than for any address once ATS is enabled.
+ * P2PDMA then routes this client's Requests with the Address Type each query
+ * asks about, rather than the one its ATS Enable bit implies.
+ */
+void pcim_p2pdma_set_ats_per_mapping(struct pci_dev *pdev)
+{
+ struct pci_p2pdma *p2p;
+
+ p2p = rcu_dereference_protected(pdev->p2pdma, 1);
+ if (!p2p)
+ /*
+ * ats_per_mapping is a performance optimization,
+ * if pcim_p2pdma_init() didn't set pdev->p2pdma pointer
+ * for some reason, let's simply use ATS global settings.
+ */
+ return;
+
+ p2p->ats_per_mapping = true;
+}
+EXPORT_SYMBOL_GPL(pcim_p2pdma_set_ats_per_mapping);
+
/**
* pci_p2pdma_map_type_tlp - Determine the mapping type for P2PDMA transfers
* @provider: P2PDMA provider structure
diff --git a/include/linux/pci-p2pdma.h b/include/linux/pci-p2pdma.h
index dd17501ba1b6..ce389e18d023 100644
--- a/include/linux/pci-p2pdma.h
+++ b/include/linux/pci-p2pdma.h
@@ -82,6 +82,7 @@ struct scatterlist *pci_p2pmem_alloc_sgl(struct pci_dev *pdev,
unsigned int *nents, u32 length);
void pci_p2pmem_free_sgl(struct pci_dev *pdev, struct scatterlist *sgl);
void pci_p2pmem_publish(struct pci_dev *pdev, bool publish);
+void pcim_p2pdma_set_ats_per_mapping(struct pci_dev *pdev);
int pci_p2pdma_enable_store(const char *page, struct pci_dev **p2p_dev,
bool *use_p2pdma);
ssize_t pci_p2pdma_enable_show(char *page, struct pci_dev *p2p_dev,
@@ -138,6 +139,9 @@ static inline void pci_p2pmem_free_sgl(struct pci_dev *pdev,
static inline void pci_p2pmem_publish(struct pci_dev *pdev, bool publish)
{
}
+static inline void pcim_p2pdma_set_ats_per_mapping(struct pci_dev *pdev)
+{
+}
static inline int pci_p2pdma_enable_store(const char *page,
struct pci_dev **p2p_dev, bool *use_p2pdma)
{
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* [PATCH v10 17/18] PCI/P2PDMA: Evaluate the ATS path for clients with ATS enabled
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (15 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 16/18] PCI/P2PDMA: Let a client declare that it selects ATS per mapping Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:27 ` sashiko-bot
2026-10-12 13:07 ` [PATCH v10 18/18] PCI/P2PDMA: Test the routing of " Leon Romanovsky
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
P2PDMA assumes every client issues Untranslated Requests. A device with
ATS enabled may translate any address it is handed, including a bus
address, and Translation Blocking can reject its Translated Requests on
a route the Untranslated answer called usable.
Unless the client declared per-mapping ATS, take the Address Type from
its ATS Enable bit. Never hand such a client bus addresses: only an IOVA
survives translation, and whatever it issues untranslated goes through
the host bridge, whose route must therefore be usable. Apply this after
the cache, which keeps holding answers that depend only on the topology.
Reviewed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
Documentation/driver-api/pci/p2pdma.rst | 8 +++++
drivers/pci/p2pdma.c | 53 +++++++++++++++++++++++++++++++--
2 files changed, 58 insertions(+), 3 deletions(-)
diff --git a/Documentation/driver-api/pci/p2pdma.rst b/Documentation/driver-api/pci/p2pdma.rst
index 52725e58f656..72ef2df3340a 100644
--- a/Documentation/driver-api/pci/p2pdma.rst
+++ b/Documentation/driver-api/pci/p2pdma.rst
@@ -46,6 +46,14 @@ P2PDMA therefore decides each class of traffic separately, and
``pci_p2pdma_map_type()`` answers for the one a DMA mapping is built for:
strictly ordered Requests carrying an Untranslated address.
+The PCIe ATS Enable bit covers the whole device, and most devices translate
+any address they are handed once it is set. Unless a driver has declared with
+``pcim_p2pdma_set_ats_per_mapping()`` that its device chooses ATS per mapping,
+that bit decides which Address Type P2PDMA assumes, and a client with ATS
+enabled is never handed a bus address, which it would translate as though it
+were an IOVA. Its Translated Requests may still route directly, but anything
+it issues untranslated reaches the host bridge, so that route has to work too.
+
The two directions are decided independently. Translation Blocking (TB),
Direct Translated P2P (DT), Request Redirect (RR) and Egress Control (EC) on
the Requester-side port decide the Request:
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index e38081f79761..a5fc622e3114 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -1602,6 +1602,41 @@ void pcim_p2pdma_set_ats_per_mapping(struct pci_dev *pdev)
}
EXPORT_SYMBOL_GPL(pcim_p2pdma_set_ats_per_mapping);
+static unsigned int pci_p2pdma_client_tlp_flags(struct pci_dev *client,
+ unsigned int tlp_flags)
+{
+ if (client->ats_enabled)
+ return tlp_flags | PCI_P2PDMA_TLP_TRANSLATED;
+
+ return tlp_flags & ~PCI_P2PDMA_TLP_TRANSLATED;
+}
+
+/*
+ * A client that translates every address it is handed cannot be handed a bus
+ * address, which it would translate as though it were an IOVA. Its Translated
+ * Requests may still route directly, but a Request it issues without a
+ * translation carries the IOVA to the host bridge, so that route has to work
+ * as well.
+ */
+static enum pci_p2pdma_map_type
+pci_p2pdma_client_map_type(struct pci_dev *provider, struct pci_dev *client,
+ bool per_mapping, enum pci_p2pdma_map_type type)
+{
+ if (per_mapping || !client->ats_enabled ||
+ type != PCI_P2PDMA_MAP_BUS_ADDR)
+ return type;
+
+ pci_dbg(client,
+ "P2PDMA ACS: provider=%s bus address withheld; ATS is enabled for the whole client\n",
+ pci_name(provider));
+
+ if (cpu_supports_p2pdma() ||
+ host_bridge_whitelist(provider, client, false))
+ return PCI_P2PDMA_MAP_THRU_HOST_BRIDGE;
+
+ return PCI_P2PDMA_MAP_NOT_SUPPORTED;
+}
+
/**
* pci_p2pdma_map_type_tlp - Determine the mapping type for P2PDMA transfers
* @provider: P2PDMA provider structure
@@ -1616,6 +1651,11 @@ EXPORT_SYMBOL_GPL(pcim_p2pdma_set_ats_per_mapping);
* ACS routes a peer-to-peer transaction by the attributes its TLPs carry, so
* the answer depends on @tlp_flags. A caller that passes flags its traffic
* does not match gets a mapping the fabric will not deliver.
+ *
+ * Only a client whose driver called pcim_p2pdma_set_ats_per_mapping() takes
+ * the Address Type from @tlp_flags. For any other client its ATS Enable bit
+ * decides, and a client that translates every address it is handed never gets
+ * %PCI_P2PDMA_MAP_BUS_ADDR.
*/
static enum pci_p2pdma_map_type
pci_p2pdma_map_type_tlp(struct p2pdma_provider *provider, struct device *dev,
@@ -1627,6 +1667,7 @@ pci_p2pdma_map_type_tlp(struct p2pdma_provider *provider, struct device *dev,
struct pci_p2pdma *p2pdma;
struct pci_dev *client;
bool provider_state;
+ bool per_mapping;
int dist;
if (!pdev->p2pdma) {
@@ -1646,6 +1687,9 @@ pci_p2pdma_map_type_tlp(struct p2pdma_provider *provider, struct device *dev,
cache_index = map_types_idx(client);
rcu_read_lock();
+ /* The declaration belongs to the client, the cache to the provider. */
+ p2pdma = rcu_dereference(client->p2pdma);
+ per_mapping = p2pdma && p2pdma->ats_per_mapping;
p2pdma = rcu_dereference(pdev->p2pdma);
if (p2pdma)
@@ -1653,6 +1697,8 @@ pci_p2pdma_map_type_tlp(struct p2pdma_provider *provider, struct device *dev,
cache_index));
provider_state = !!p2pdma;
rcu_read_unlock();
+ if (!per_mapping)
+ tlp_flags = pci_p2pdma_client_tlp_flags(client, tlp_flags);
type = pci_p2pdma_map_types_unpack(cached, tlp_flags);
pci_dbg(client,
"P2PDMA ACS: map lookup provider=%s index=%#lx tlp-flags=%#x cached=%s(%d) provider-state=%u\n",
@@ -1660,10 +1706,10 @@ pci_p2pdma_map_type_tlp(struct p2pdma_provider *provider, struct device *dev,
pci_p2pdma_map_type_name(type), type, provider_state);
if (type == PCI_P2PDMA_MAP_UNKNOWN)
- return calc_map_type_and_dist(pdev, client, &dist, tlp_flags,
+ type = calc_map_type_and_dist(pdev, client, &dist, tlp_flags,
true);
- return type;
+ return pci_p2pdma_client_map_type(pdev, client, per_mapping, type);
}
/**
@@ -1672,7 +1718,8 @@ pci_p2pdma_map_type_tlp(struct p2pdma_provider *provider, struct device *dev,
* @dev: Client device that initiates the transfer
*
* Same as pci_p2pdma_map_type_tlp() for a client issuing strictly ordered
- * Requests that carry an Untranslated address.
+ * Requests. Their Address Type is Untranslated unless the client enables ATS
+ * for the whole device.
*/
enum pci_p2pdma_map_type pci_p2pdma_map_type(struct p2pdma_provider *provider,
struct device *dev)
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread* Re: [PATCH v10 17/18] PCI/P2PDMA: Evaluate the ATS path for clients with ATS enabled
2026-10-12 13:07 ` [PATCH v10 17/18] PCI/P2PDMA: Evaluate the ATS path for clients with ATS enabled Leon Romanovsky
@ 2026-10-12 13:27 ` sashiko-bot
2026-10-12 18:08 ` Leon Romanovsky
0 siblings, 1 reply; 38+ messages in thread
From: sashiko-bot @ 2026-10-12 13:27 UTC (permalink / raw)
To: Leon Romanovsky; +Cc: media-ci, dri-devel, linux-rdma, linux-pci, kvm
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] pci: pci_p2pdma_distance_many() evaluates compatibility without applying client ATS constraints
--
commit 254b102fe424913edc3d5f94f8d41bc5d0de2b20
Author: Leon Romanovsky <leon@kernel.org>
PCI/P2PDMA: Evaluate the ATS path for clients with ATS enabled
P2PDMA assumes every client issues Untranslated Requests. A device with
ATS enabled may translate any address it is handed, including a bus
address, and Translation Blocking can reject its Translated Requests on
a route the Untranslated answer called usable.
> diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
> index e38081f797611..a5fc622e3114f 100644
> --- a/drivers/pci/p2pdma.c
> +++ b/drivers/pci/p2pdma.c
[ ... ]
> @@ -1660,10 +1706,10 @@ pci_p2pdma_map_type_tlp(struct p2pdma_provider *provider, struct device *dev,
> pci_p2pdma_map_type_name(type), type, provider_state);
>
> if (type == PCI_P2PDMA_MAP_UNKNOWN)
> - return calc_map_type_and_dist(pdev, client, &dist, tlp_flags,
> + type = calc_map_type_and_dist(pdev, client, &dist, tlp_flags,
> true);
>
> - return type;
> + return pci_p2pdma_client_map_type(pdev, client, per_mapping, type);
> }
[Severity: High]
Does pci_p2pdma_distance_many() also need to apply these new ATS constraints?
While this patch successfully updates pci_p2pdma_map_type_tlp() to evaluate ATS
routing constraints for P2PDMA clients, pci_p2pdma_distance_many() continues
to evaluate mapping compatibility without them:
drivers/pci/p2pdma.c:pci_p2pdma_distance_many() {
...
pci_client = find_parent_pci_dev(clients[i]);
if (!pci_client) {
...
map = calc_map_type_and_dist(provider, pci_client, &distance, 0,
verbose);
...
}
If a client has ATS enabled and its Untranslated route is usable but its
Translated route is blocked by Translation Blocking (or lacks a whitelisted
host bridge), pci_p2pdma_distance_many() evaluates the Untranslated mapping
with tlp_flags=0, and incorrectly returns success.
When the driver subsequently attempts to map memory, the ATS-aware mapping
logic will return PCI_P2PDMA_MAP_NOT_SUPPORTED and fail unexpectedly at
runtime, breaking the API contract of the distance validation.
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261012-fix-p2p-acs-v4-0-v10-0-6ffe7b7a4517@nvidia.com?part=17
^ permalink raw reply [flat|nested] 38+ messages in thread* Re: [PATCH v10 17/18] PCI/P2PDMA: Evaluate the ATS path for clients with ATS enabled
2026-10-12 13:27 ` sashiko-bot
@ 2026-10-12 18:08 ` Leon Romanovsky
0 siblings, 0 replies; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 18:08 UTC (permalink / raw)
To: sashiko-reviews; +Cc: media-ci, dri-devel, linux-rdma, linux-pci, kvm
On Mon, Oct 12, 2026 at 01:27:12PM +0000, sashiko-bot@kernel.org wrote:
> Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
> - [High] pci: pci_p2pdma_distance_many() evaluates compatibility without applying client ATS constraints
> --
>
> commit 254b102fe424913edc3d5f94f8d41bc5d0de2b20
> Author: Leon Romanovsky <leon@kernel.org>
>
> PCI/P2PDMA: Evaluate the ATS path for clients with ATS enabled
>
> P2PDMA assumes every client issues Untranslated Requests. A device with
> ATS enabled may translate any address it is handed, including a bus
> address, and Translation Blocking can reject its Translated Requests on
> a route the Untranslated answer called usable.
>
> > diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
> > index e38081f797611..a5fc622e3114f 100644
> > --- a/drivers/pci/p2pdma.c
> > +++ b/drivers/pci/p2pdma.c
>
> [ ... ]
>
> > @@ -1660,10 +1706,10 @@ pci_p2pdma_map_type_tlp(struct p2pdma_provider *provider, struct device *dev,
> > pci_p2pdma_map_type_name(type), type, provider_state);
> >
> > if (type == PCI_P2PDMA_MAP_UNKNOWN)
> > - return calc_map_type_and_dist(pdev, client, &dist, tlp_flags,
> > + type = calc_map_type_and_dist(pdev, client, &dist, tlp_flags,
> > true);
> >
> > - return type;
> > + return pci_p2pdma_client_map_type(pdev, client, per_mapping, type);
> > }
>
> [Severity: High]
> Does pci_p2pdma_distance_many() also need to apply these new ATS constraints?
>
> While this patch successfully updates pci_p2pdma_map_type_tlp() to evaluate ATS
> routing constraints for P2PDMA clients, pci_p2pdma_distance_many() continues
> to evaluate mapping compatibility without them:
>
> drivers/pci/p2pdma.c:pci_p2pdma_distance_many() {
> ...
> pci_client = find_parent_pci_dev(clients[i]);
> if (!pci_client) {
> ...
> map = calc_map_type_and_dist(provider, pci_client, &distance, 0,
> verbose);
> ...
> }
>
> If a client has ATS enabled and its Untranslated route is usable but its
> Translated route is blocked by Translation Blocking (or lacks a whitelisted
> host bridge), pci_p2pdma_distance_many() evaluates the Untranslated mapping
> with tlp_flags=0, and incorrectly returns success.
>
> When the driver subsequently attempts to map memory, the ATS-aware mapping
> logic will return PCI_P2PDMA_MAP_NOT_SUPPORTED and fail unexpectedly at
> runtime, breaking the API contract of the distance validation.
Yes, Sashiko is right, I'll send new version :(.
Thanks
>
> --
> Sashiko AI review · https://sashiko.dev/#/patchset/20261012-fix-p2p-acs-v4-0-v10-0-6ffe7b7a4517@nvidia.com?part=17
^ permalink raw reply [flat|nested] 38+ messages in thread
* [PATCH v10 18/18] PCI/P2PDMA: Test the routing of clients with ATS enabled
2026-10-12 13:07 [PATCH v10 00/18] PCI/P2PDMA: Route peer-to-peer DMA by TLP class Leon Romanovsky
` (16 preceding siblings ...)
2026-10-12 13:07 ` [PATCH v10 17/18] PCI/P2PDMA: Evaluate the ATS path for clients with ATS enabled Leon Romanovsky
@ 2026-10-12 13:07 ` Leon Romanovsky
2026-10-12 13:24 ` sashiko-bot
17 siblings, 1 reply; 38+ messages in thread
From: Leon Romanovsky @ 2026-10-12 13:07 UTC (permalink / raw)
To: Bjorn Helgaas, Logan Gunthorpe, Leon Romanovsky, Jason Gunthorpe,
Joerg Roedel (AMD), Will Deacon, Robin Murphy,
Christian König, Thomas Hellström
Cc: linux-pci, linux-kernel, linux-doc, iommu, Tushar Dave,
linux-media, dri-devel, linaro-mm-sig, linux-rdma, kvm,
Chaitanya Kulkarni, Greg Kroah-Hartman, Jens Axboe,
Alex Williamson, Ankit Agrawal, Jonathan Corbet, Shuah Khan,
Randy Dunlap, Sumit Semwal
From: Leon Romanovsky <leonro@nvidia.com>
A client that enables ATS for the whole device must never be handed a
bus address, and Translation Blocking must reject its peer-to-peer
traffic even when the caller asks for the default class. That has to
hold for answers taken from the cache as well.
Cover those on the fabricated fabric, together with a client that
declared per-mapping ATS keeping the per-class answers and a client
without ATS answering for the Untranslated Requests it will issue.
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
---
drivers/pci/p2pdma.c | 8 +--
drivers/pci/pci.h | 5 ++
drivers/pci/pci_acs_test.c | 123 +++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 133 insertions(+), 3 deletions(-)
diff --git a/drivers/pci/p2pdma.c b/drivers/pci/p2pdma.c
index a5fc622e3114..b0f859b20f35 100644
--- a/drivers/pci/p2pdma.c
+++ b/drivers/pci/p2pdma.c
@@ -1602,14 +1602,15 @@ void pcim_p2pdma_set_ats_per_mapping(struct pci_dev *pdev)
}
EXPORT_SYMBOL_GPL(pcim_p2pdma_set_ats_per_mapping);
-static unsigned int pci_p2pdma_client_tlp_flags(struct pci_dev *client,
- unsigned int tlp_flags)
+VISIBLE_IF_KUNIT unsigned int
+pci_p2pdma_client_tlp_flags(struct pci_dev *client, unsigned int tlp_flags)
{
if (client->ats_enabled)
return tlp_flags | PCI_P2PDMA_TLP_TRANSLATED;
return tlp_flags & ~PCI_P2PDMA_TLP_TRANSLATED;
}
+EXPORT_SYMBOL_IF_KUNIT(pci_p2pdma_client_tlp_flags);
/*
* A client that translates every address it is handed cannot be handed a bus
@@ -1618,7 +1619,7 @@ static unsigned int pci_p2pdma_client_tlp_flags(struct pci_dev *client,
* translation carries the IOVA to the host bridge, so that route has to work
* as well.
*/
-static enum pci_p2pdma_map_type
+VISIBLE_IF_KUNIT enum pci_p2pdma_map_type
pci_p2pdma_client_map_type(struct pci_dev *provider, struct pci_dev *client,
bool per_mapping, enum pci_p2pdma_map_type type)
{
@@ -1636,6 +1637,7 @@ pci_p2pdma_client_map_type(struct pci_dev *provider, struct pci_dev *client,
return PCI_P2PDMA_MAP_NOT_SUPPORTED;
}
+EXPORT_SYMBOL_IF_KUNIT(pci_p2pdma_client_map_type);
/**
* pci_p2pdma_map_type_tlp - Determine the mapping type for P2PDMA transfers
diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h
index 99852cf99732..9f9dbdbb66b7 100644
--- a/drivers/pci/pci.h
+++ b/drivers/pci/pci.h
@@ -1142,6 +1142,11 @@ enum pci_p2pdma_map_type calc_map_type_and_dist(struct pci_dev *provider,
int *dist,
unsigned int tlp_flags,
bool verbose);
+unsigned int pci_p2pdma_client_tlp_flags(struct pci_dev *client,
+ unsigned int tlp_flags);
+enum pci_p2pdma_map_type
+pci_p2pdma_client_map_type(struct pci_dev *provider, struct pci_dev *client,
+ bool per_mapping, enum pci_p2pdma_map_type type);
#endif
#ifdef CONFIG_PCI_QUIRKS
diff --git a/drivers/pci/pci_acs_test.c b/drivers/pci/pci_acs_test.c
index 6a2cd38d968f..a9a3b80ea0ac 100644
--- a/drivers/pci/pci_acs_test.c
+++ b/drivers/pci/pci_acs_test.c
@@ -698,6 +698,125 @@ static void acs_walk_unreadable_control_test(struct kunit *test)
PCI_P2PDMA_MAP_NOT_SUPPORTED);
}
+/*
+ * Route the way pci_p2pdma_map_type_tlp() does once it has a client: the
+ * client decides which flags apply, and the answer is then checked against
+ * what the client can be handed.
+ */
+static enum pci_p2pdma_map_type acs_client_map(struct acs_fabric *f,
+ bool per_mapping,
+ unsigned int tlp_flags)
+{
+ enum pci_p2pdma_map_type type;
+ int dist;
+
+ if (!per_mapping)
+ tlp_flags = pci_p2pdma_client_tlp_flags(f->client, tlp_flags);
+ type = calc_map_type_and_dist(f->provider, f->client, &dist, tlp_flags,
+ false);
+ return pci_p2pdma_client_map_type(f->provider, f->client, per_mapping,
+ type);
+}
+
+/*
+ * A client with ATS enabled for the whole device translates whatever it is
+ * handed, so it never gets bus addresses even on a direct route, and
+ * Translation Blocking rejects it even when its caller asks for the default
+ * class.
+ */
+static void acs_client_ats_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+
+ acs_build_fabric(test, &f);
+ f.client->ats_enabled = 1;
+
+ KUNIT_EXPECT_EQ(test, acs_client_map(&f, false, 0),
+ PCI_P2PDMA_MAP_THRU_HOST_BRIDGE);
+
+ f.client_leaf->acs_cap = 0x100;
+ f.client_leaf->acs_capabilities = PCI_ACS_TB;
+ f.client_leaf_cfg.ctrl = PCI_ACS_TB;
+
+ KUNIT_EXPECT_EQ(test, acs_client_map(&f, false, 0),
+ PCI_P2PDMA_MAP_NOT_SUPPORTED);
+}
+
+/*
+ * The cache holds what the topology allows, filled when the client first
+ * asked. A client whose ATS was enabled after that must still be kept off the
+ * bus addresses the cache recorded.
+ */
+static void acs_client_ats_cached_test(struct kunit *test)
+{
+ enum pci_p2pdma_map_type type[PCI_P2PDMA_TLP_CLASSES], hit;
+ struct acs_fabric f = {};
+ unsigned long cached;
+ unsigned int flags;
+ int dist;
+
+ acs_build_fabric(test, &f);
+ for (flags = 0; flags < PCI_P2PDMA_TLP_CLASSES; flags++)
+ type[flags] = calc_map_type_and_dist(f.provider, f.client,
+ &dist, flags, false);
+ cached = pci_p2pdma_map_types_pack(type);
+
+ f.client->ats_enabled = 1;
+ flags = pci_p2pdma_client_tlp_flags(f.client, 0);
+ hit = pci_p2pdma_map_types_unpack(cached, flags);
+
+ KUNIT_EXPECT_EQ(test,
+ pci_p2pdma_client_map_type(f.provider, f.client, false,
+ hit),
+ PCI_P2PDMA_MAP_THRU_HOST_BRIDGE);
+}
+
+/*
+ * A client that declared per-mapping ATS picks the Address Type per mapping,
+ * so it keeps the per-class answers even with ATS enabled.
+ */
+static void acs_client_ats_per_mapping_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+
+ acs_build_fabric(test, &f);
+ f.client->ats_enabled = 1;
+
+ KUNIT_EXPECT_EQ(test, acs_client_map(&f, true, 0),
+ PCI_P2PDMA_MAP_BUS_ADDR);
+ KUNIT_EXPECT_EQ(test,
+ acs_client_map(&f, true, PCI_P2PDMA_TLP_TRANSLATED),
+ PCI_P2PDMA_MAP_BUS_ADDR);
+
+ f.client_leaf->acs_cap = 0x100;
+ f.client_leaf->acs_capabilities = PCI_ACS_TB;
+ f.client_leaf_cfg.ctrl = PCI_ACS_TB;
+
+ KUNIT_EXPECT_EQ(test, acs_client_map(&f, true, 0),
+ PCI_P2PDMA_MAP_BUS_ADDR);
+ KUNIT_EXPECT_EQ(test,
+ acs_client_map(&f, true, PCI_P2PDMA_TLP_TRANSLATED),
+ PCI_P2PDMA_MAP_NOT_SUPPORTED);
+}
+
+/*
+ * A client with ATS disabled cannot issue Translated Requests, so a caller
+ * asking about them gets the answer for the Untranslated ones it will issue.
+ */
+static void acs_client_no_ats_test(struct kunit *test)
+{
+ struct acs_fabric f = {};
+
+ acs_build_fabric(test, &f);
+ f.dn1->acs_cap = 0x100;
+ f.dn1->acs_capabilities = PCI_ACS_RR | PCI_ACS_DT;
+ f.dn1_cfg.ctrl = PCI_ACS_RR | PCI_ACS_DT;
+
+ KUNIT_EXPECT_EQ(test,
+ acs_client_map(&f, false, PCI_P2PDMA_TLP_TRANSLATED),
+ PCI_P2PDMA_MAP_THRU_HOST_BRIDGE);
+}
+
static struct kunit_case pci_acs_test_cases[] = {
KUNIT_CASE_PARAM(pci_acs_p2pdma_request_test,
acs_request_gen_params),
@@ -723,6 +842,10 @@ static struct kunit_case pci_acs_test_cases[] = {
KUNIT_CASE(acs_walk_tb_above_divergence_test),
KUNIT_CASE(acs_walk_unreadable_control_test),
KUNIT_CASE(acs_map_types_pack_test),
+ KUNIT_CASE(acs_client_ats_test),
+ KUNIT_CASE(acs_client_ats_cached_test),
+ KUNIT_CASE(acs_client_ats_per_mapping_test),
+ KUNIT_CASE(acs_client_no_ats_test),
{}
};
--
2.55.0
^ permalink raw reply related [flat|nested] 38+ messages in thread