Skip to contentSkip to conversation

dt-bindings: misc: add binding for Xilinx AXI-Stream FIFO

PATCH Superseded Aditya Chari 3 participants 10 messages Jun 21 – Jun 22

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 dt-bindings: misc: add binding for Xilinx AXI-Stream FIFO

The axis-fifo driver's compatible strings were undocumented, flagged by checkpatch.pl as UNDOCUMENTED_DT_STRING. Add a devicetree binding document describing the required properties
for drivers/staging/axis-fifo, derived from the properties the driver reads via of_property_read_u32()
in axis_fifo_parse_dt().

---
 .../bindings/misc/xlnx,axi-fifo-mm-s.yaml     | 95 +++++++++++++++++++
 1 file changed, 95 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml
Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml new file +95 -0
new file mode 100644
index 000000000..8ef0adde3
@@ -0,0 +1,95 @@
+# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/misc/xlnx,axi-fifo-mm-s.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Xilinx AXI-Stream FIFO (axis-fifo)
+
+maintainers:
+  - Jacob Feder <[email protected]>
+
+description:
+  The AXI-Stream FIFO (AXIS-FIFO) IP core provides a memory-mapped AXI4-Lite
+  interface for sending and receiving data over an AXI4-Stream interface
+  using FIFO buffers. See Xilinx PG080 for IP details.
+
+properties:
+  compatible:
+    enum:
+      - xlnx,axi-fifo-mm-s-4.1
+      - xlnx,axi-fifo-mm-s-4.2
+      - xlnx,axi-fifo-mm-s-4.3
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    maxItems: 1
+
+  xlnx,axi-str-rxd-tdata-width:
+    description:
+      Width in bits of the AXI4-Stream receive data interface. Only a
+      width of 32 is currently supported by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 32
+
+  xlnx,axi-str-txd-tdata-width:
+    description:
+      Width in bits of the AXI4-Stream transmit data interface. Only a
+      width of 32 is currently supported by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 32
+
+  xlnx,rx-fifo-depth:
+    description:
+      Depth in bytes of the receive FIFO, as configured in the IP core.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,tx-fifo-depth:
+    description:
+      Depth in bytes of the transmit FIFO, as configured in the IP core.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,use-rx-data:
+    description:
+      Indicates whether the receive data path is present, as configured
+      in the IP core. A value of 1 enables the receive path, 0 disables it.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    enum: [ 0, 1 ]
+
+  xlnx,use-tx-data:
+    description:
+      Indicates whether the transmit data path is present, as configured
+      in the IP core. A value of 1 enables the transmit path, 0 disables it.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    enum: [ 0, 1 ]
+
+required:
+  - compatible
+  - reg
+  - interrupts
+  - xlnx,axi-str-rxd-tdata-width
+  - xlnx,axi-str-txd-tdata-width
+  - xlnx,rx-fifo-depth
+  - xlnx,tx-fifo-depth
+  - xlnx,use-rx-data
+  - xlnx,use-tx-data
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/irq.h>
+
+    axi_fifo_mm_s_0: axi-fifo-mm-s@40000000 {
+        compatible = "xlnx,axi-fifo-mm-s-4.1";
+        reg = <0x40000000 0x10000>;
+        interrupts = <0 29 IRQ_TYPE_LEVEL_HIGH>;
+        xlnx,axi-str-rxd-tdata-width = <32>;
+        xlnx,axi-str-txd-tdata-width = <32>;
+        xlnx,rx-fifo-depth = <2048>;
+        xlnx,tx-fifo-depth = <2048>;
+        xlnx,use-rx-data = <1>;
+        xlnx,use-tx-data = <1>;
+    };
Signature · 1 line
2.53.0
#1 lore raw
Thank you for your contribution! Sashiko AI review found 5 potential issue(s) to consider:

Thank you for your contribution! Sashiko AI review found 5 potential issue(s) to consider:

New issues:
- [High] The YAML binding incorrectly documents the FIFO depth properties as being in 'bytes', while the driver and hardware expect them in 'words'.
- [Medium] The schema omits valid hardware properties while enforcing additionalProperties: false, falsely restricting the device tree to only the properties the current driver implementation reads and causing valid hardware device trees to fail schema validation.
- [Medium] Hardware data width properties are artificially constrained to const: 32 based on driver support rather than hardware capabilities.
- [Low] The previous text-based binding document was not removed.

Pre-existing issues:
- [Critical] A pre-existing use-after-free vulnerability occurs if the device is unbound while userspace holds an open file descriptor. --

commit 1499d3b34c3ed26799e82ed6630ba1f50a8aec04
Author: Aditya Chari <[email protected]>

dt-bindings: misc: add binding for Xilinx AXI-Stream FIFO

The axis-fifo driver's compatible strings were undocumented, flagged by checkpatch.pl as UNDOCUMENTED_DT_STRING. Add a devicetree binding document describing the required properties
for drivers/staging/axis-fifo, derived from the properties the driver reads via of_property_read_u32()
in axis_fifo_parse_dt().

Quoted 1 line

dt-bindings: misc: add binding for Xilinx AXI-Stream FIFO

[Severity: Low]
Should the previous text-based binding document be removed in this patch?

It looks like the legacy drivers/staging/axis-fifo/axis-fifo.txt binding was left in the tree, leaving two conflicting sources of truth for the same hardware compatible string.

Quoted 5 lines
Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml new file +0 -0
new file mode 100644
index 0000000000000..8ef0adde32ad8

[ ... ]

Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml line 42

Quoted 13 lines
Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml new file +13 -0
+  xlnx,axi-str-rxd-tdata-width:
+    description:
+      Width in bits of the AXI4-Stream receive data interface. Only a
+      width of 32 is currently supported by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 32
+
+  xlnx,axi-str-txd-tdata-width:
+    description:
+      Width in bits of the AXI4-Stream transmit data interface. Only a
+      width of 32 is currently supported by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 32

[Severity: Medium]
Should this schema describe the hardware capabilities rather than the software limitations?

Over-constraining the schema to const: 32 based on current driver support might cause validation errors for device trees that accurately represent hardware synthesized with different bus widths.

Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml line 52

Quoted 9 lines
Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml new file +9 -0
+  xlnx,rx-fifo-depth:
+    description:
+      Depth in bytes of the receive FIFO, as configured in the IP core.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,tx-fifo-depth:
+    description:
+      Depth in bytes of the transmit FIFO, as configured in the IP core.
+    $ref: /schemas/types.yaml#/definitions/uint32

[Severity: High]
Is it correct to document these properties as being in bytes?

Looking at the driver code, it evaluates the depth as word counts when checking for overflow:

drivers/staging/axis-fifo/axis-fifo.c:axis_fifo_write() {
    ...
    if (words_to_write > (fifo->tx_fifo_depth - 4)) {
        ...
    }
    ...
}

If a devicetree author specifies the depth in bytes per the new YAML documentation, could the driver load a value four times too large, allowing out-of-bounds writes that might trigger a hardware transmit overrun error?

[ ... ]

Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml line 79

Quoted 12 lines
Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml new file +12 -0
+required:
+  - compatible
+  - reg
+  - interrupts
+  - xlnx,axi-str-rxd-tdata-width
+  - xlnx,axi-str-txd-tdata-width
+  - xlnx,rx-fifo-depth
+  - xlnx,tx-fifo-depth
+  - xlnx,use-rx-data
+  - xlnx,use-tx-data
+
+additionalProperties: false

[Severity: Medium]
Does this schema artificially restrict the devicetree by omitting valid hardware properties?

The original text binding allowed properties such as xlnx,has-axis-tkeep and interrupt-names. Because the schema enforces additionalProperties: false, will running dt_binding_check on existing, valid device trees generated for this IP result in false-positive validation errors?

[Severity: Critical]
This is a pre-existing issue, but does the driver have a use-after-free vulnerability if the device is unbound while a userspace process holds an open file descriptor?

In the driver, the axis_fifo structure is allocated with devm_kzalloc() and I/O memory mapped with devm_platform_get_and_ioremap_resource():

drivers/staging/axis-fifo/axis-fifo.c:axis_fifo_probe() {
    ...
    fifo = devm_kzalloc(dev, sizeof(*fifo), GFP_KERNEL);
    ...
}

However, axis_fifo_remove() unregisters the character device without tracking or waiting for active file descriptors to close:

drivers/staging/axis-fifo/axis-fifo.c:axis_fifo_remove() {
    ...
    misc_deregister(&fifo->miscdev);
    ...
}

If the device is unbound via sysfs while a userspace process keeps the char device open, the driver core's devres cleanup automatically frees the fifo structure and unmaps base_addr. Would subsequent read(), write(), or poll() calls via the active file descriptor then access the freed fifo pointer and unmapped I/O memory?

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=1
#2 lore raw

PATCH v2 1/1 dt-bindings: misc: add binding for Xilinx AXI-Stream FIFO

In reply to Aditya Chari

The axis-fifo driver's compatible strings were undocumented, flagged by checkpatch.pl as UNDOCUMENTED_DT_STRING. Add a YAML devicetree binding document for drivers/staging/axis-fifo, converted from and replacing the existing free-form text binding (axis-fifo.txt), which this patch removes.

---

Changes since v1:
- Fixed xlnx,rx/tx-fifo-depth: depth is in 32-bit words, not bytes, matching the driver's overflow check in axis_fifo_write() and the wording of the original text binding.
- Restored the full set of hardware-generated properties (interrupt- names, AXI-Stream protocol/width properties, has-axis-t* feature flags, fifo threshold properties, etc.) so that additionalProperties:
false does not reject valid device trees generated for real hardware.
- Removed the now-superseded axis-fifo.txt text binding.

 .../bindings/misc/xlnx,axi-fifo-mm-s.yaml     | 221 ++++++++++++++++++
 drivers/staging/axis-fifo/axis-fifo.txt       |  96 --------
 2 files changed, 221 insertions(+), 96 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml
 delete mode 100644 drivers/staging/axis-fifo/axis-fifo.txt
Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml new file +221 -0
new file mode 100644
index 000000000..f4606b13c
@@ -0,0 +1,221 @@
+# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/misc/xlnx,axi-fifo-mm-s.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Xilinx AXI-Stream FIFO (axis-fifo)
+
+maintainers:
+  - Jacob Feder <[email protected]>
+
+description:
+  The AXI-Stream FIFO (AXIS-FIFO) IP core provides a memory-mapped AXI4-Lite
+  interface for sending and receiving data over an AXI4-Stream interface
+  using FIFO buffers. Currently supports only store-forward mode with a
+  32-bit AXI4-Lite interface; cut-through mode and full AXI4 are not
+  supported. See Xilinx PG080 for IP details.
+
+properties:
+  compatible:
+    enum:
+      - xlnx,axi-fifo-mm-s-4.1
+      - xlnx,axi-fifo-mm-s-4.2
+      - xlnx,axi-fifo-mm-s-4.3
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    maxItems: 1
+
+  interrupt-names:
+    items:
+      - const: interrupt
+
+  xlnx,axi-str-rxd-tdata-width:
+    description:
+      Width in bits of the AXI4-Stream receive data interface. Only a
+      width of 32 is currently supported by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 32
+
+  xlnx,axi-str-txd-tdata-width:
+    description:
+      Width in bits of the AXI4-Stream transmit data interface. Only a
+      width of 32 is currently supported by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 32
+
+  xlnx,axi-str-txc-tdata-width:
+    description:
+      Width in bits of the AXI4-Stream transmit control interface.
+      Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,axi-str-rxd-protocol:
+    description: AXI-Stream receive data protocol. Ignored by the driver.
+    enum: [ XIL_AXI_STREAM_ETH_DATA ]
+
+  xlnx,axi-str-txd-protocol:
+    description: AXI-Stream transmit data protocol. Ignored by the driver.
+    enum: [ XIL_AXI_STREAM_ETH_DATA ]
+
+  xlnx,axi-str-txc-protocol:
+    description: AXI-Stream transmit control protocol. Ignored by the driver.
+    enum: [ XIL_AXI_STREAM_ETH_CTRL ]
+
+  xlnx,axis-tdest-width:
+    description: AXI-Stream TDEST width. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,axis-tid-width:
+    description: AXI-Stream TID width. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,axis-tuser-width:
+    description: AXI-Stream TUSER width. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,data-interface-type:
+    description: Data interface type. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,has-axis-tdest:
+    description:
+      Whether the AXI-Stream interface has TDEST. This feature is not
+      supported by the driver and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,has-axis-tid:
+    description:
+      Whether the AXI-Stream interface has TID. This feature is not
+      supported by the driver and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,has-axis-tkeep:
+    description:
+      Whether the AXI-Stream interface has TKEEP. This feature is not
+      supported by the driver and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,has-axis-tstrb:
+    description:
+      Whether the AXI-Stream interface has TSTRB. This feature is not
+      supported by the driver and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,has-axis-tuser:
+    description:
+      Whether the AXI-Stream interface has TUSER. This feature is not
+      supported by the driver and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,rx-fifo-depth:
+    description:
+      Depth in 32-bit words of the receive FIFO, as configured in the
+      IP core.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,rx-fifo-pe-threshold:
+    description: RX programmable empty interrupt threshold. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,rx-fifo-pf-threshold:
+    description: RX programmable full interrupt threshold. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,s-axi-id-width:
+    description: AXI4-Lite ID width. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,s-axi4-data-width:
+    description: AXI4-Lite data width. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,select-xpm:
+    description: Whether XPM macros are used. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,tx-fifo-depth:
+    description:
+      Depth in 32-bit words of the transmit FIFO, as configured in the
+      IP core.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,tx-fifo-pe-threshold:
+    description: TX programmable empty interrupt threshold. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,tx-fifo-pf-threshold:
+    description: TX programmable full interrupt threshold. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,use-rx-cut-through:
+    description:
+      Whether RX cut-through mode is used. Not supported by the driver
+      and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,use-rx-data:
+    description:
+      Indicates whether the receive data path is present, as configured
+      in the IP core. A value of 1 enables the receive path, 0 disables it.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    enum: [ 0, 1 ]
+
+  xlnx,use-tx-ctrl:
+    description:
+      Whether the transmit control interface is used. Not supported by
+      the driver and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,use-tx-cut-through:
+    description:
+      Whether TX cut-through mode is used. Not supported by the driver
+      and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,use-tx-data:
+    description:
+      Indicates whether the transmit data path is present, as configured
+      in the IP core. A value of 1 enables the transmit path, 0 disables it.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    enum: [ 0, 1 ]
+
+required:
+  - compatible
+  - reg
+  - interrupts
+  - xlnx,axi-str-rxd-tdata-width
+  - xlnx,axi-str-txd-tdata-width
+  - xlnx,rx-fifo-depth
+  - xlnx,tx-fifo-depth
+  - xlnx,use-rx-data
+  - xlnx,use-tx-data
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/irq.h>
+
+    axi_fifo_mm_s_0: axi-fifo-mm-s@40000000 {
+        compatible = "xlnx,axi-fifo-mm-s-4.1";
+        reg = <0x40000000 0x10000>;
+        interrupts = <0 29 IRQ_TYPE_LEVEL_HIGH>;
+        xlnx,axi-str-rxd-tdata-width = <32>;
+        xlnx,axi-str-txd-tdata-width = <32>;
+        xlnx,rx-fifo-depth = <512>;
+        xlnx,tx-fifo-depth = <32768>;
+        xlnx,use-rx-data = <1>;
+        xlnx,use-tx-data = <1>;
+    };
drivers/staging/axis-fifo/axis-fifo.txt deleted +0 -96

96 lines not on this page · raw message

Signature · 1 line
2.53.0
#3 lore raw
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:

New issues:
- [Low] Missing $ref type definition for custom string properties in DT binding.

Pre-existing issues:
- [Critical] Use-After-Free on device unbind due to lack of synchronization with open file descriptors.
- [High] Integer underflow in bounds check allows bypassing transmit packet size restriction. --

commit 97e86b437f0cf66fc60eec47c4742a54db8053f8
Author: Aditya Chari <[email protected]>

dt-bindings: misc: add binding for Xilinx AXI-Stream FIFO

The axis-fifo driver's compatible strings were undocumented, flagged by checkpatch.pl as UNDOCUMENTED_DT_STRING. Add a YAML devicetree binding document for drivers/staging/axis-fifo, converted from and replacing the existing free-form text binding (axis-fifo.txt), which this patch removes.

Quoted 5 lines
Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml new file +0 -0
new file mode 100644
@@ -0,0 +1,221 @@

[ ... ]

Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml line 66

Quoted 11 lines
Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml new file +11 -0
+  xlnx,axi-str-rxd-protocol:
+    description: AXI-Stream receive data protocol. Ignored by the driver.
+    enum: [ XIL_AXI_STREAM_ETH_DATA ]
+
+  xlnx,axi-str-txd-protocol:
+    description: AXI-Stream transmit data protocol. Ignored by the driver.
+    enum: [ XIL_AXI_STREAM_ETH_DATA ]
+
+  xlnx,axi-str-txc-protocol:
+    description: AXI-Stream transmit control protocol. Ignored by the driver.
+    enum: [ XIL_AXI_STREAM_ETH_CTRL ]

[Severity: Low]
Do these custom string properties need a $ref to types.yaml?

Device tree schema validation (dt_binding_check) requires custom properties to explicitly specify their type. Providing an enum of strings might not be sufficient for dt-schema to infer the type, which could cause type missing or unknown schema validation errors.

Should these properties include a $ref: /schemas/types.yaml#/definitions/string constraint?

[ ... ]

Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml line 149

Quoted 5 lines
Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml new file +5 -0
+  xlnx,tx-fifo-depth:
+    description:
+      Depth in 32-bit words of the transmit FIFO, as configured in the
+      IP core.
+    $ref: /schemas/types.yaml#/definitions/uint32

[Severity: High]
This isn't a bug introduced by this patch, but should we add a minimum: 4 constraint here to protect against an integer underflow in the driver?

In drivers/staging/axis-fifo/axis-fifo.c:axis_fifo_write(), the bounds check assumes tx_fifo_depth is at least 4:

    if (!words_to_write || (len % sizeof(u32)) ||
        (words_to_write > (fifo->tx_fifo_depth - 4)))
        return -EINVAL;

If the device tree configures xlnx,tx-fifo-depth to a value less than 4, this subtraction underflows to a massive unsigned integer, bypassing the bounds check. This could allow an oversized packet to trigger a Transmit Packet Overrun Error in the hardware.

[ ... ]

[Severity: Critical]
This is a pre-existing issue in the driver documented by this binding, but is there a use-after-free vulnerability during device unbind?

The driver allocates its core state using devm_kzalloc():

drivers/staging/axis-fifo/axis-fifo.c:axis_fifo_probe() {
    ...
    fifo = devm_kzalloc(dev, sizeof(*fifo), GFP_KERNEL);
    ...
}

During removal, it deregisters the misc device but doesn't wait for active VFS file references to be dropped:

drivers/staging/axis-fifo/axis-fifo.c:axis_fifo_remove() {
    ...
    misc_deregister(&fifo->miscdev);
    ...
}

When axis_fifo_remove() returns, the devres layer automatically frees the axis_fifo structure. If a user process holds an open file descriptor and subsequently calls write() or read(), it will dereference the freed pointer:

drivers/staging/axis-fifo/axis-fifo.c:axis_fifo_write() {
    struct axis_fifo *fifo = f->private_data;
    ...
}

Should the driver use a different allocation and reference counting strategy to ensure the memory remains valid as long as the file descriptor is open?

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=1
#4 lore raw

PATCH v3 1/1 dt-bindings: misc: add binding for Xilinx AXI-Stream FIFO

In reply to Aditya Chari

The axis-fifo driver's compatible strings were undocumented, flagged by checkpatch.pl as UNDOCUMENTED_DT_STRING. Add a YAML devicetree binding document for drivers/staging/axis-fifo, converted from and replacing the existing free-form text binding (axis-fifo.txt), which this patch removes.

Constrain xlnx,tx-fifo-depth to a minimum of 4, since the driver subtracts 4 from this value in its transmit bounds check and a smaller value would underflow that check.

---

Changes since v2:
- Added $ref: /schemas/types.yaml#/definitions/string to the three AXI-Stream protocol enum properties (xlnx,axi-str-rxd-protocol, xlnx,axi-str-txd-protocol, xlnx,axi-str-txc-protocol) for explicit type consistency with the rest of the schema.
- Added minimum: 4 to xlnx,tx-fifo-depth, since the driver subtracts 4 from this value in its transmit bounds check (axis_fifo_write()) and a smaller configured value would underflow that unsigned check, bypassing the oversized-packet guard.

Changes since v1:
- Fixed xlnx,rx/tx-fifo-depth: depth is in 32-bit words, not bytes, matching the driver's overflow check in axis_fifo_write() and the wording of the original text binding.
- Restored the full set of hardware-generated properties (interrupt- names, AXI-Stream protocol/width properties, has-axis-t* feature flags, fifo threshold properties, etc.) so that additionalProperties:
false does not reject valid device trees generated for real hardware.
- Removed the now-superseded axis-fifo.txt text binding.

 .../bindings/misc/xlnx,axi-fifo-mm-s.yaml     | 227 ++++++++++++++++++
 drivers/staging/axis-fifo/axis-fifo.txt       |  96 --------
 2 files changed, 227 insertions(+), 96 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml
 delete mode 100644 drivers/staging/axis-fifo/axis-fifo.txt
Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml new file +227 -0
new file mode 100644
index 000000000..53bf52524
@@ -0,0 +1,227 @@
+# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/misc/xlnx,axi-fifo-mm-s.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Xilinx AXI-Stream FIFO (axis-fifo)
+
+maintainers:
+  - Jacob Feder <[email protected]>
+
+description:
+  The AXI-Stream FIFO (AXIS-FIFO) IP core provides a memory-mapped AXI4-Lite
+  interface for sending and receiving data over an AXI4-Stream interface
+  using FIFO buffers. Currently supports only store-forward mode with a
+  32-bit AXI4-Lite interface; cut-through mode and full AXI4 are not
+  supported. See Xilinx PG080 for IP details.
+
+properties:
+  compatible:
+    enum:
+      - xlnx,axi-fifo-mm-s-4.1
+      - xlnx,axi-fifo-mm-s-4.2
+      - xlnx,axi-fifo-mm-s-4.3
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    maxItems: 1
+
+  interrupt-names:
+    items:
+      - const: interrupt
+
+  xlnx,axi-str-rxd-tdata-width:
+    description:
+      Width in bits of the AXI4-Stream receive data interface. Only a
+      width of 32 is currently supported by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 32
+
+  xlnx,axi-str-txd-tdata-width:
+    description:
+      Width in bits of the AXI4-Stream transmit data interface. Only a
+      width of 32 is currently supported by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 32
+
+  xlnx,axi-str-txc-tdata-width:
+    description:
+      Width in bits of the AXI4-Stream transmit control interface.
+      Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,axi-str-rxd-protocol:
+    description: AXI-Stream receive data protocol. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/string
+    enum: [ XIL_AXI_STREAM_ETH_DATA ]
+
+  xlnx,axi-str-txd-protocol:
+    description: AXI-Stream transmit data protocol. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/string
+    enum: [ XIL_AXI_STREAM_ETH_DATA ]
+
+  xlnx,axi-str-txc-protocol:
+    description: AXI-Stream transmit control protocol. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/string
+    enum: [ XIL_AXI_STREAM_ETH_CTRL ]
+
+  xlnx,axis-tdest-width:
+    description: AXI-Stream TDEST width. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,axis-tid-width:
+    description: AXI-Stream TID width. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,axis-tuser-width:
+    description: AXI-Stream TUSER width. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,data-interface-type:
+    description: Data interface type. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,has-axis-tdest:
+    description:
+      Whether the AXI-Stream interface has TDEST. This feature is not
+      supported by the driver and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,has-axis-tid:
+    description:
+      Whether the AXI-Stream interface has TID. This feature is not
+      supported by the driver and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,has-axis-tkeep:
+    description:
+      Whether the AXI-Stream interface has TKEEP. This feature is not
+      supported by the driver and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,has-axis-tstrb:
+    description:
+      Whether the AXI-Stream interface has TSTRB. This feature is not
+      supported by the driver and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,has-axis-tuser:
+    description:
+      Whether the AXI-Stream interface has TUSER. This feature is not
+      supported by the driver and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,rx-fifo-depth:
+    description:
+      Depth in 32-bit words of the receive FIFO, as configured in the
+      IP core.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,rx-fifo-pe-threshold:
+    description: RX programmable empty interrupt threshold. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,rx-fifo-pf-threshold:
+    description: RX programmable full interrupt threshold. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,s-axi-id-width:
+    description: AXI4-Lite ID width. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,s-axi4-data-width:
+    description: AXI4-Lite data width. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,select-xpm:
+    description: Whether XPM macros are used. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,tx-fifo-depth:
+    description:
+      Depth in 32-bit words of the transmit FIFO, as configured in the
+      IP core. Must be at least 4; the driver subtracts 4 from this
+      value when bounds-checking transmit writes, and a smaller value
+      underflows that check.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    minimum: 4
+
+  xlnx,tx-fifo-pe-threshold:
+    description: TX programmable empty interrupt threshold. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,tx-fifo-pf-threshold:
+    description: TX programmable full interrupt threshold. Ignored by the driver.
+    $ref: /schemas/types.yaml#/definitions/uint32
+
+  xlnx,use-rx-cut-through:
+    description:
+      Whether RX cut-through mode is used. Not supported by the driver
+      and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,use-rx-data:
+    description:
+      Indicates whether the receive data path is present, as configured
+      in the IP core. A value of 1 enables the receive path, 0 disables it.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    enum: [ 0, 1 ]
+
+  xlnx,use-tx-ctrl:
+    description:
+      Whether the transmit control interface is used. Not supported by
+      the driver and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,use-tx-cut-through:
+    description:
+      Whether TX cut-through mode is used. Not supported by the driver
+      and must be 0.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    const: 0
+
+  xlnx,use-tx-data:
+    description:
+      Indicates whether the transmit data path is present, as configured
+      in the IP core. A value of 1 enables the transmit path, 0 disables it.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    enum: [ 0, 1 ]
+
+required:
+  - compatible
+  - reg
+  - interrupts
+  - xlnx,axi-str-rxd-tdata-width
+  - xlnx,axi-str-txd-tdata-width
+  - xlnx,rx-fifo-depth
+  - xlnx,tx-fifo-depth
+  - xlnx,use-rx-data
+  - xlnx,use-tx-data
+
+additionalProperties: false
+
+examples:
+  - |
+    #include <dt-bindings/interrupt-controller/irq.h>
+
+    axi_fifo_mm_s_0: axi-fifo-mm-s@40000000 {
+        compatible = "xlnx,axi-fifo-mm-s-4.1";
+        reg = <0x40000000 0x10000>;
+        interrupts = <0 29 IRQ_TYPE_LEVEL_HIGH>;
+        xlnx,axi-str-rxd-tdata-width = <32>;
+        xlnx,axi-str-txd-tdata-width = <32>;
+        xlnx,rx-fifo-depth = <512>;
+        xlnx,tx-fifo-depth = <32768>;
+        xlnx,use-rx-data = <1>;
+        xlnx,use-tx-data = <1>;
+    };
drivers/staging/axis-fifo/axis-fifo.txt deleted +0 -96

96 lines not on this page · raw message

Signature · 1 line
2.53.0
#5 lore raw
Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:

Thank you for your contribution! Sashiko AI review found 3 potential issue(s) to consider:

Pre-existing issues:
- [High] Missing runtime bounds check for tx_fifo_depth causes integer underflow and potential DoS
- [High] Unvalidated hardware read can cause massive busy loop and CPU soft lockup
- [High] Hardware interrupts enabled before IRQ handler registration
--

commit cd9dfa74e4b12151a534638f2622f83e05f40fee
Author: Aditya Chari <[email protected]>

dt-bindings: misc: add binding for Xilinx AXI-Stream FIFO

This patch converts the free-form text binding for the Xilinx AXI-Stream FIFO IP core into a YAML schema. It also enforces a minimum value of 4 for the xlnx,tx-fifo-depth property to match the driver's transmit bounds check.

Quoted 11 lines
Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml new file +5 -0
new file mode 100644
index 0000000000000..53bf52524e74d
@@ -0,0 +1,227 @@
+# SPDX-License-Identifier: (GPL-2.0 OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/misc/xlnx,axi-fifo-mm-s.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#

[ ... ]

Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml line 30

Quoted 12 lines
Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml new file +12 -0
+properties:
+  compatible:
+    enum:
+      - xlnx,axi-fifo-mm-s-4.1
+      - xlnx,axi-fifo-mm-s-4.2
+      - xlnx,axi-fifo-mm-s-4.3
+
+  reg:
+    maxItems: 1
+
+  interrupts:
+    maxItems: 1

[Severity: High]
This is a pre-existing issue, but does the driver enable interrupts before registering the IRQ handler?

Looking at drivers/staging/axis-fifo/axis-fifo.c:axis_fifo_probe():

    reset_ip_core(fifo);

    irq = platform_get_irq(pdev, 0);
    if (irq < 0)
        return irq;

    rc = devm_request_irq(fifo->dt_device, irq, &axis_fifo_irq, 0,
                          DRIVER_NAME, fifo);

And in reset_ip_core():

iowrite32(XLLF_INT_TC_MASK | XLLF_INT_RC_MASK | XLLF_INT_RPURE_MASK |
XLLF_INT_RPORE_MASK | XLLF_INT_RPUE_MASK | XLLF_INT_TPOE_MASK | XLLF_INT_TSE_MASK, fifo->base_addr + XLLF_IER_OFFSET);

If the hardware immediately asserts an interrupt, could this trigger a spurious interrupt before devm_request_irq() completes, permanently disabling the IRQ line?

[ ... ]

Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml line 155

Quoted 8 lines
Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml new file +8 -0
+  xlnx,tx-fifo-depth:
+    description:
+      Depth in 32-bit words of the transmit FIFO, as configured in the
+      IP core. Must be at least 4; the driver subtracts 4 from this
+      value when bounds-checking transmit writes, and a smaller value
+      underflows that check.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    minimum: 4

[Severity: High]
This isn't a bug introduced by this patch, but since the binding enforces a minimum of 4, what happens if a malformed device tree provides a smaller value?

In drivers/staging/axis-fifo/axis-fifo.c:axis_fifo_parse_dt():

    ret = of_property_read_u32(node, "xlnx,tx-fifo-depth",
                               &fifo->tx_fifo_depth);
    if (ret)
        return ret;

The value is parsed without enforcing the boundary constraint. Could this lead to an integer underflow later in axis_fifo_write()?

    if (!words_to_write || (len % sizeof(u32)) ||
        (words_to_write > (fifo->tx_fifo_depth - 4)))
        return -EINVAL;

If fifo->tx_fifo_depth is less than 4, this subtraction underflows, bypassing the maximum packet size check.

[ ... ]

Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml line 177

Quoted 6 lines
Documentation/devicetree/bindings/misc/xlnx,axi-fifo-mm-s.yaml new file +6 -0
+  xlnx,use-rx-data:
+    description:
+      Indicates whether the receive data path is present, as configured
+      in the IP core. A value of 1 enables the receive path, 0 disables it.
+    $ref: /schemas/types.yaml#/definitions/uint32
+    enum: [ 0, 1 ]

[Severity: High]
This is a pre-existing issue, but does the driver handle hardware read errors safely when processing received data?

In drivers/staging/axis-fifo/axis-fifo.c:axis_fifo_read():

    bytes_available = ioread32(fifo->base_addr + XLLF_RLR_OFFSET);
    words_available = bytes_available / sizeof(u32);
    ...
    if (bytes_available % sizeof(u32)) {
        /* this probably can't happen unless IP
         * registers were previously mishandled
         */
        dev_err(fifo->dt_device, "received a packet that isn't word-aligned\n");
        ret = -EIO;
        goto err_flush_rx;
    }

If the device is in an error state and ioread32() returns 0xFFFFFFFF, words_available becomes 0x3FFFFFFF. Would this cause the error recovery loop to stall the CPU?

    err_flush_rx:
        while (words_available--)
            ioread32(fifo->base_addr + XLLF_RDFD_OFFSET);

Executing over a billion memory-mapped reads sequentially could trigger the kernel's soft lockup watchdog.

Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=1
#6 lore raw
In reply to Aditya Chari
Quoted 23 lines · Aditya Chari

The axis-fifo driver's compatible strings were undocumented, flagged by checkpatch.pl as UNDOCUMENTED_DT_STRING. Add a YAML devicetree binding document for drivers/staging/axis-fifo, converted from and replacing the existing free-form text binding (axis-fifo.txt), which this patch removes.

Constrain xlnx,tx-fifo-depth to a minimum of 4, since the driver subtracts 4 from this value in its transmit bounds check and a smaller value would underflow that check.

---

Changes since v2:
- Added $ref: /schemas/types.yaml#/definitions/string to the three AXI-Stream protocol enum properties (xlnx,axi-str-rxd-protocol, xlnx,axi-str-txd-protocol, xlnx,axi-str-txc-protocol) for explicit type consistency with the rest of the schema.
- Added minimum: 4 to xlnx,tx-fifo-depth, since the driver subtracts 4 from this value in its transmit bounds check (axis_fifo_write()) and a smaller configured value would underflow that unsigned check, bypassing the oversized-packet guard.

Quoted 9 lines

Changes since v1:
- Fixed xlnx,rx/tx-fifo-depth: depth is in 32-bit words, not bytes, matching the driver's overflow check in axis_fifo_write() and the wording of the original text binding.
- Restored the full set of hardware-generated properties (interrupt- names, AXI-Stream protocol/width properties, has-axis-t* feature flags, fifo threshold properties, etc.) so that additionalProperties:
false does not reject valid device trees generated for real hardware.
- Removed the now-superseded axis-fifo.txt text binding.

Please slow down. Three versions within 1 hour! Why sending something and immediately sending fixes to it?

Quoted 3 lines
 .../bindings/misc/xlnx,axi-fifo-mm-s.yaml     | 227 ++++++++++++++++++
 drivers/staging/axis-fifo/axis-fifo.txt       |  96 --------

Why are you touching staging binding?

https://lore.kernel.org/all/?q=dfn%3Aaxis-fifo.txt

Best regards,
Krzysztof

#7 lore raw

Sorry about the pace - jumped from automated lint feedback straight to a new version without waiting for an actual reviewer. Won't happen again.

On the staging binding question - I found the driver-removal thread from June 2 and read through it. Michal Simek's reply makes clear this driver isn't going anywhere (their networking team has plans that depend on it), and that there's already an in-flight binding conversion from Lucas Faria Mendes that he's actively tracking. I wasn't aware of that series when I started this.

Given that, I'll withdraw mine rather than duplicate work that's already further along and already has the relevant maintainer's eyes on it.

Thanks for pointing me toward the history.

Regards,
Aditya

#8 lore raw
Quoted 11 lines · Aditya Chari S

Sorry about the pace - jumped from automated lint feedback straight to a new version without waiting for an actual reviewer. Won't happen again.

On the staging binding question - I found the driver-removal thread from June 2 and read through it. Michal Simek's reply makes clear this driver isn't going anywhere (their networking team has plans that depend on it), and that there's already an in-flight binding conversion from Lucas Faria Mendes that he's actively tracking. I wasn't aware of that series when I started this.

Quoted 4 lines

Given that, I'll withdraw mine rather than duplicate work that's already further along and already has the relevant maintainer's eyes on it.

The main point is that driver is in staging, so we don't take bindings for it.

Best regards,
Krzysztof

#9 lore raw
Quoted 2 lines · Krzysztof Kozlowski

The main point is that driver is in staging, so we don't take bindings for it.

Understood, thanks for clarifying.

Regards,
Aditya

Quoted 23 lines · Krzysztof Kozlowski
Quoted 11 lines · Aditya Chari S

Sorry about the pace - jumped from automated lint feedback straight to a new version without waiting for an actual reviewer. Won't happen again.

On the staging binding question - I found the driver-removal thread from June 2 and read through it. Michal Simek's reply makes clear this driver isn't going anywhere (their networking team has plans that depend on it), and that there's already an in-flight binding conversion from Lucas Faria Mendes that he's actively tracking. I wasn't aware of that series when I started this.

Quoted 4 lines

Given that, I'll withdraw mine rather than duplicate work that's already further along and already has the relevant maintainer's eyes on it.

The main point is that driver is in staging, so we don't take bindings for it.

Best regards,
Krzysztof