| |||
| Home > Programmer’s Model > VFP11 system registers > Media and VFP Feature Registers | |||
The purpose of the Media and VFP Feature Registers is to provide information about the features that the VFP unit contains.
The Media and VFP Feature Registers:
are two 32-bit read-only registers
when the VFP is enabled by the EN bit in the Floating Point Exception Register, are accessible in any mode
when the VFP is disabled by the EN bit, are accessible only in Privileged mode.
See Floating-Point Exception Register, FPEXC for more information about the EN bit.
Figure 3.8 shows the Media and VFP Feature Register 0 bit fields.
This register is first implemented in the rev1 (r1p0) release of the ARM1136JF-S processor.
The values in the Media and VFP Feature Register 0 are implementation defined. Table 3.9 describes the Media and VFP Feature Register 0 bit fields for the VFP11 coprocessor.
Table 3.9. Media and VFP Feature Register 0 bit fields
| Bit range | Field name | Function |
|---|---|---|
| [31:28] | - | Indicates the VFP hardware support level when user traps are disabled.
|
| [27:24] | - | Indicates support for short vectors.
|
| [23:20] | - | Indicates support for hardware square root.
|
| [19:16] | - | Indicates support for hardware divide.
|
| [15:12] | - | Indicates support for user traps.
|
| [11:8] | - | Indicates support for double precision VFP.
|
| [7:4] | - | Indicates support for single precision VFP.
|
| [3:0] | - | Indicates support for the media register bank.
|
Figure 3.9 shows the Media and VFP Feature Register 1 bit fields.
This register is first implemented in the rev1 (r1p0) release of the ARM1136JF-S processor.
The values in the Media and VFP Feature Register 1 are implementation defined. Table 3.10 describes the Media and VFP Feature Register 0 bit fields for the VFP11 coprocessor.
Table 3.10. Media and VFP Feature Register 1 bit fields
| Bit range | Field name | Function |
|---|---|---|
| [31:12] | - | Reserved. Read as zero. |
| [11:8] | - | Indicates support for media extension, single precision floating point instructions.
|
| [7:4] | - | Indicates support for media extension, integer instructions.
|
| [3:0] | - | Indicates support for media extension, load/store instructions.
|