diff options
| author | Linus Torvalds <torvalds@linux-foundation.org> | 2025-07-31 09:45:28 -0700 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2025-07-31 09:45:28 -0700 |
| commit | 44a8c96edd0ee9320a1ad87afc7b10f38e55d5ec (patch) | |
| tree | 504034f60c5510ebeb2c0d1d93a68fba999f2896 | |
| parent | b4efd62564e96d1edb99eb00dd0ff620dbd1afab (diff) | |
| parent | bf24d64268544379d9a9b5b8efc2bb03967703b3 (diff) | |
Merge tag 'v6.17-p1' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
Pull crypto update from Herbert Xu:
"API:
- Allow hash drivers without fallbacks (e.g., hardware key)
Algorithms:
- Add hmac hardware key support (phmac) on s390
- Re-enable sha384 in FIPS mode
- Disable sha1 in FIPS mode
- Convert zstd to acomp
Drivers:
- Lower priority of qat skcipher and aead
- Convert aspeed to partial block API
- Add iMX8QXP support in caam
- Add rate limiting support for GEN6 devices in qat
- Enable telemetry for GEN6 devices in qat
- Implement full backlog mode for hisilicon/sec2"
* tag 'v6.17-p1' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: (116 commits)
crypto: keembay - Use min() to simplify ocs_create_linked_list_from_sg()
crypto: hisilicon/hpre - fix dma unmap sequence
crypto: qat - make adf_dev_autoreset() static
crypto: ccp - reduce stack usage in ccp_run_aes_gcm_cmd
crypto: qat - refactor ring-related debug functions
crypto: qat - fix seq_file position update in adf_ring_next()
crypto: qat - fix DMA direction for compression on GEN2 devices
crypto: jitter - replace ARRAY_SIZE definition with header include
crypto: engine - remove {prepare,unprepare}_crypt_hardware callbacks
crypto: engine - remove request batching support
crypto: qat - flush misc workqueue during device shutdown
crypto: qat - enable rate limiting feature for GEN6 devices
crypto: qat - add compression slice count for rate limiting
crypto: qat - add get_svc_slice_cnt() in device data structure
crypto: qat - add adf_rl_get_num_svc_aes() in rate limiting
crypto: qat - relocate service related functions
crypto: qat - consolidate service enums
crypto: qat - add decompression service for rate limiting
crypto: qat - validate service in rate limiting sysfs api
crypto: hisilicon/sec2 - implement full backlog mode for sec
...
150 files changed, 4088 insertions, 2044 deletions
diff --git a/Documentation/ABI/testing/debugfs-driver-qat b/Documentation/ABI/testing/debugfs-driver-qat index bd6793760f298..3f1efbbad6ca9 100644 --- a/Documentation/ABI/testing/debugfs-driver-qat +++ b/Documentation/ABI/testing/debugfs-driver-qat @@ -67,7 +67,7 @@ Contact: qat-linux@intel.com Description: (RO) Read returns power management information specific to the QAT device. - This attribute is only available for qat_4xxx devices. + This attribute is only available for qat_4xxx and qat_6xxx devices. What: /sys/kernel/debug/qat_<device>_<BDF>/cnv_errors Date: January 2024 diff --git a/Documentation/ABI/testing/debugfs-driver-qat_telemetry b/Documentation/ABI/testing/debugfs-driver-qat_telemetry index eacee20720882..0dfd8d97e1692 100644 --- a/Documentation/ABI/testing/debugfs-driver-qat_telemetry +++ b/Documentation/ABI/testing/debugfs-driver-qat_telemetry @@ -32,7 +32,7 @@ Description: (RW) Enables/disables the reporting of telemetry metrics. echo 0 > /sys/kernel/debug/qat_4xxx_0000:6b:00.0/telemetry/control - This attribute is only available for qat_4xxx devices. + This attribute is only available for qat_4xxx and qat_6xxx devices. What: /sys/kernel/debug/qat_<device>_<BDF>/telemetry/device_data Date: March 2024 @@ -67,6 +67,10 @@ Description: (RO) Reports device telemetry counters. exec_xlt<N> execution count of Translator slice N util_dcpr<N> utilization of Decompression slice N [%] exec_dcpr<N> execution count of Decompression slice N + util_cnv<N> utilization of Compression and verify slice N [%] + exec_cnv<N> execution count of Compression and verify slice N + util_dcprz<N> utilization of Decompression slice N [%] + exec_dcprz<N> execution count of Decompression slice N util_pke<N> utilization of PKE N [%] exec_pke<N> execution count of PKE N util_ucs<N> utilization of UCS slice N [%] @@ -100,7 +104,7 @@ Description: (RO) Reports device telemetry counters. If a device lacks of a specific accelerator, the corresponding attribute is not reported. - This attribute is only available for qat_4xxx devices. + This attribute is only available for qat_4xxx and qat_6xxx devices. What: /sys/kernel/debug/qat_<device>_<BDF>/telemetry/rp_<A/B/C/D>_data Date: March 2024 @@ -225,4 +229,4 @@ Description: (RW) Selects up to 4 Ring Pairs (RP) to monitor, one per file, ``rp2srv`` from sysfs. See Documentation/ABI/testing/sysfs-driver-qat for details. - This attribute is only available for qat_4xxx devices. + This attribute is only available for qat_4xxx and qat_6xxx devices. diff --git a/Documentation/ABI/testing/sysfs-driver-qat b/Documentation/ABI/testing/sysfs-driver-qat index f290e77cd590a..b0561b9fc4eb9 100644 --- a/Documentation/ABI/testing/sysfs-driver-qat +++ b/Documentation/ABI/testing/sysfs-driver-qat @@ -14,7 +14,7 @@ Description: (RW) Reports the current state of the QAT device. Write to It is possible to transition the device from up to down only if the device is up and vice versa. - This attribute is only available for qat_4xxx devices. + This attribute is available for qat_4xxx and qat_6xxx devices. What: /sys/bus/pci/devices/<BDF>/qat/cfg_services Date: June 2022 @@ -23,24 +23,28 @@ Contact: qat-linux@intel.com Description: (RW) Reports the current configuration of the QAT device. Write to the file to change the configured services. - The values are: - - * sym;asym: the device is configured for running crypto - services - * asym;sym: identical to sym;asym - * dc: the device is configured for running compression services - * dcc: identical to dc but enables the dc chaining feature, - hash then compression. If this is not required chose dc - * sym: the device is configured for running symmetric crypto - services - * asym: the device is configured for running asymmetric crypto - services - * asym;dc: the device is configured for running asymmetric - crypto services and compression services - * dc;asym: identical to asym;dc - * sym;dc: the device is configured for running symmetric crypto - services and compression services - * dc;sym: identical to sym;dc + One or more services can be enabled per device. + Certain configurations are restricted to specific device types; + where applicable this is explicitly indicated, for example + (qat_6xxx) denotes applicability exclusively to that device series. + + The available services include: + + * sym: Configures the device for symmetric cryptographic operations. + * asym: Configures the device for asymmetric cryptographic operations. + * dc: Configures the device for compression and decompression + operations. + * dcc: Similar to dc, but with the additional dc chaining feature + enabled, cipher then compress (qat_6xxx), hash then compression. + If this is not required choose dc. + * decomp: Configures the device for decompression operations (qat_6xxx). + + Service combinations are permitted for all services except dcc. + On QAT GEN4 devices (qat_4xxx driver) a maximum of two services can be + combined and on QAT GEN6 devices (qat_6xxx driver ) a maximum of three + services can be combined. + The order of services is not significant. For instance, sym;asym is + functionally equivalent to asym;sym. It is possible to set the configuration only if the device is in the `down` state (see /sys/bus/pci/devices/<BDF>/qat/state) @@ -59,7 +63,7 @@ Description: (RW) Reports the current configuration of the QAT device. # cat /sys/bus/pci/devices/<BDF>/qat/cfg_services dc - This attribute is only available for qat_4xxx devices. + This attribute is available for qat_4xxx and qat_6xxx devices. What: /sys/bus/pci/devices/<BDF>/qat/pm_idle_enabled Date: June 2023 @@ -94,7 +98,7 @@ Description: (RW) This configuration option provides a way to force the device i # cat /sys/bus/pci/devices/<BDF>/qat/pm_idle_enabled 0 - This attribute is only available for qat_4xxx devices. + This attribute is available for qat_4xxx and qat_6xxx devices. What: /sys/bus/pci/devices/<BDF>/qat/rp2srv Date: January 2024 @@ -126,7 +130,7 @@ Description: # cat /sys/bus/pci/devices/<BDF>/qat/rp2srv sym - This attribute is only available for qat_4xxx devices. + This attribute is available for qat_4xxx and qat_6xxx devices. What: /sys/bus/pci/devices/<BDF>/qat/num_rps Date: January 2024 @@ -140,7 +144,7 @@ Description: # cat /sys/bus/pci/devices/<BDF>/qat/num_rps 64 - This attribute is only available for qat_4xxx devices. + This attribute is available for qat_4xxx and qat_6xxx devices. What: /sys/bus/pci/devices/<BDF>/qat/auto_reset Date: May 2024 @@ -160,4 +164,4 @@ Description: (RW) Reports the current state of the autoreset feature * 0/Nn/off: auto reset disabled. If the device encounters an unrecoverable error, it will not be reset. - This attribute is only available for qat_4xxx devices. + This attribute is available for qat_4xxx and qat_6xxx devices. diff --git a/Documentation/ABI/testing/sysfs-driver-qat_rl b/Documentation/ABI/testing/sysfs-driver-qat_rl index 8c282ae3155dd..d534f89b49719 100644 --- a/Documentation/ABI/testing/sysfs-driver-qat_rl +++ b/Documentation/ABI/testing/sysfs-driver-qat_rl @@ -31,7 +31,7 @@ Description: * rm_all: Removes all the configured SLAs. * Inputs: None - This attribute is only available for qat_4xxx devices. + This attribute is only available for qat_4xxx and qat_6xxx devices. What: /sys/bus/pci/devices/<BDF>/qat_rl/rp Date: January 2024 @@ -68,7 +68,7 @@ Description: ## Write # echo 0x5 > /sys/bus/pci/devices/<BDF>/qat_rl/rp - This attribute is only available for qat_4xxx devices. + This attribute is only available for qat_4xxx and qat_6xxx devices. What: /sys/bus/pci/devices/<BDF>/qat_rl/id Date: January 2024 @@ -101,7 +101,7 @@ Description: # cat /sys/bus/pci/devices/<BDF>/qat_rl/rp 0x5 ## ring pair ID 0 and ring pair ID 2 - This attribute is only available for qat_4xxx devices. + This attribute is only available for qat_4xxx and qat_6xxx devices. What: /sys/bus/pci/devices/<BDF>/qat_rl/cir Date: January 2024 @@ -135,7 +135,7 @@ Description: # cat /sys/bus/pci/devices/<BDF>/qat_rl/cir 500 - This attribute is only available for qat_4xxx devices. + This attribute is only available for qat_4xxx and qat_6xxx devices. What: /sys/bus/pci/devices/<BDF>/qat_rl/pir Date: January 2024 @@ -169,7 +169,7 @@ Description: # cat /sys/bus/pci/devices/<BDF>/qat_rl/pir 750 - This attribute is only available for qat_4xxx devices. + This attribute is only available for qat_4xxx and qat_6xxx devices. What: /sys/bus/pci/devices/<BDF>/qat_rl/srv Date: January 2024 @@ -202,7 +202,7 @@ Description: # cat /sys/bus/pci/devices/<BDF>/qat_rl/srv dc - This attribute is only available for qat_4xxx devices. + This attribute is only available for qat_4xxx and qat_6xxx devices. What: /sys/bus/pci/devices/<BDF>/qat_rl/cap_rem Date: January 2024 @@ -223,4 +223,4 @@ Description: # cat /sys/bus/pci/devices/<BDF>/qat_rl/cap_rem 0 - This attribute is only available for qat_4xxx devices. + This attribute is only available for qat_4xxx and qat_6xxx devices. diff --git a/Documentation/crypto/crypto_engine.rst b/Documentation/crypto/crypto_engine.rst index d562ea17d9941..7ef850e28016a 100644 --- a/Documentation/crypto/crypto_engine.rst +++ b/Documentation/crypto/crypto_engine.rst @@ -36,12 +36,6 @@ engine using ``crypto_engine_stop()`` and destroy the engine with Before transferring any request, you have to fill the context enginectx by providing functions for the following: -* ``prepare_crypt_hardware``: Called once before any prepare functions are - called. - -* ``unprepare_crypt_hardware``: Called once after all unprepare functions have - been called. - * ``prepare_cipher_request``/``prepare_hash_request``: Called before each corresponding request is performed. If some processing or other preparatory work is required, do it here. diff --git a/Documentation/devicetree/bindings/crypto/atmel,at91sam9g46-aes.yaml b/Documentation/devicetree/bindings/crypto/atmel,at91sam9g46-aes.yaml index 7dc0748444fde..19010f90198a1 100644 --- a/Documentation/devicetree/bindings/crypto/atmel,at91sam9g46-aes.yaml +++ b/Documentation/devicetree/bindings/crypto/atmel,at91sam9g46-aes.yaml @@ -15,7 +15,9 @@ properties: oneOf: - const: atmel,at91sam9g46-aes - items: - - const: microchip,sam9x7-aes + - enum: + - microchip,sam9x7-aes + - microchip,sama7d65-aes - const: atmel,at91sam9g46-aes reg: diff --git a/Documentation/devicetree/bindings/crypto/atmel,at91sam9g46-sha.yaml b/Documentation/devicetree/bindings/crypto/atmel,at91sam9g46-sha.yaml index d378c53314dd0..39e076b275b39 100644 --- a/Documentation/devicetree/bindings/crypto/atmel,at91sam9g46-sha.yaml +++ b/Documentation/devicetree/bindings/crypto/atmel,at91sam9g46-sha.yaml @@ -15,7 +15,9 @@ properties: oneOf: - const: atmel,at91sam9g46-sha - items: - - const: microchip,sam9x7-sha + - enum: + - microchip,sam9x7-sha + - microchip,sama7d65-sha - const: atmel,at91sam9g46-sha reg: diff --git a/Documentation/devicetree/bindings/crypto/atmel,at91sam9g46-tdes.yaml b/Documentation/devicetree/bindings/crypto/atmel,at91sam9g46-tdes.yaml index 6a441f79efea5..6f16008c4251c 100644 --- a/Documentation/devicetree/bindings/crypto/atmel,at91sam9g46-tdes.yaml +++ b/Documentation/devicetree/bindings/crypto/atmel,at91sam9g46-tdes.yaml @@ -15,7 +15,9 @@ properties: oneOf: - const: atmel,at91sam9g46-tdes - items: - - const: microchip,sam9x7-tdes + - enum: + - microchip,sam9x7-tdes + - microchip,sama7d65-tdes - const: atmel,at91sam9g46-tdes reg: diff --git a/Documentation/devicetree/bindings/crypto/fsl,sec-v4.0.yaml b/Documentation/devicetree/bindings/crypto/fsl,sec-v4.0.yaml index 75afa441e019e..dcc755d2709a7 100644 --- a/Documentation/devicetree/bindings/crypto/fsl,sec-v4.0.yaml +++ b/Documentation/devicetree/bindings/crypto/fsl,sec-v4.0.yaml @@ -46,6 +46,8 @@ properties: - items: - enum: - fsl,imx6ul-caam + - fsl,imx8qm-caam + - fsl,imx8qxp-caam - fsl,sec-v5.0 - const: fsl,sec-v4.0 - const: fsl,sec-v4.0 @@ -77,6 +79,9 @@ properties: interrupts: maxItems: 1 + power-domains: + maxItems: 1 + fsl,sec-era: description: Defines the 'ERA' of the SEC device. $ref: /schemas/types.yaml#/definitions/uint32 @@ -106,7 +111,10 @@ patternProperties: - const: fsl,sec-v5.0-job-ring - const: fsl,sec-v4.0-job-ring - items: - - const: fsl,sec-v5.0-job-ring + - enum: + - fsl,imx8qm-job-ring + - fsl,imx8qxp-job-ring + - fsl,sec-v5.0-job-ring - const: fsl,sec-v4.0-job-ring - const: fsl,sec-v4.0-job-ring @@ -116,6 +124,9 @@ patternProperties: interrupts: maxItems: 1 + power-domains: + maxItems: 1 + fsl,liodn: description: Specifies the LIODN to be used in conjunction with the ppid-to-liodn @@ -125,6 +136,20 @@ patternProperties: $ref: /schemas/types.yaml#/definitions/uint32-array items: - maximum: 0xfff + allOf: + - if: + properties: + compatible: + contains: + enum: + - fsl,imx8qm-job-ring + - fsl,imx8qxp-job-ring + then: + required: + - power-domains + else: + properties: + power-domains: false '^rtic@[0-9a-f]+$': type: object @@ -212,6 +237,20 @@ required: - reg - ranges +if: + properties: + compatible: + contains: + enum: + - fsl,imx8qm-caam + - fsl,imx8qxp-caam +then: + required: + - power-domains +else: + properties: + power-domains: false + additionalProperties: false examples: diff --git a/Documentation/devicetree/bindings/crypto/omap-aes.txt b/Documentation/devicetree/bindings/crypto/omap-aes.txt deleted file mode 100644 index fd9717653cbb9..0000000000000 --- a/Documentation/devicetree/bindings/crypto/omap-aes.txt +++ /dev/null @@ -1,31 +0,0 @@ -OMAP SoC AES crypto Module - -Required properties: - -- compatible : Should contain entries for this and backward compatible - AES versions: - - "ti,omap2-aes" for OMAP2. - - "ti,omap3-aes" for OMAP3. - - "ti,omap4-aes" for OMAP4 and AM33XX. - Note that the OMAP2 and 3 versions are compatible (OMAP3 supports - more algorithms) but they are incompatible with OMAP4. -- ti,hwmods: Name of the hwmod associated with the AES module -- reg : Offset and length of the register set for the module -- interrupts : the interrupt-specifier for the AES module. - -Optional properties: -- dmas: DMA specifiers for tx and rx dma. See the DMA client binding, - Documentation/devicetree/bindings/dma/dma.txt -- dma-names: DMA request names should include "tx" and "rx" if present. - -Example: - /* AM335x */ - aes: aes@53500000 { - compatible = "ti,omap4-aes"; - ti,hwmods = "aes"; - reg = <0x53500000 0xa0>; - interrupts = <102>; - dmas = <&edma 6>, - <&edma 5>; - dma-names = "tx", "rx"; - }; diff --git a/Documentation/devicetree/bindings/crypto/omap-des.txt b/Documentation/devicetree/bindings/crypto/omap-des.txt deleted file mode 100644 index e8c63bf2e16db..0000000000000 --- a/Documentation/devicetree/bindings/crypto/omap-des.txt +++ /dev/null @@ -1,30 +0,0 @@ -OMAP SoC DES crypto Module - -Required properties: - -- compatible : Should contain "ti,omap4-des" -- ti,hwmods: Name of the hwmod associated with the DES module -- reg : Offset and length of the register set for the module -- interrupts : the interrupt-specifier for the DES module -- clocks : A phandle to the functional clock node of the DES module - corresponding to each entry in clock-names -- clock-names : Name of the functional clock, should be "fck" - -Optional properties: -- dmas: DMA specifiers for tx and rx dma. See the DMA client binding, - Documentation/devicetree/bindings/dma/dma.txt - Each entry corresponds to an entry in dma-names -- dma-names: DMA request names should include "tx" and "rx" if present - -Example: - /* DRA7xx SoC */ - des: des@480a5000 { - compatible = "ti,omap4-des"; - ti,hwmods = "des"; - reg = <0x480a5000 0xa0>; - interrupts = <GIC_SPI 82 IRQ_TYPE_LEVEL_HIGH>; - dmas = <&sdma 117>, <&sdma 116>; - dma-names = "tx", "rx"; - clocks = <&l3_iclk_div>; - clock-names = "fck"; - }; diff --git a/Documentation/devicetree/bindings/crypto/ti,omap2-aes.yaml b/Documentation/devicetree/bindings/crypto/ti,omap2-aes.yaml new file mode 100644 index 0000000000000..90e92050ad2eb --- /dev/null +++ b/Documentation/devicetree/bindings/crypto/ti,omap2-aes.yaml @@ -0,0 +1,58 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/ti,omap2-aes.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: OMAP SoC AES crypto Module + +maintainers: + - Aaro Koskinen <aaro.koskinen@iki.fi> + - Andreas Kemnade <andreas@kemnade.info> + - Kevin Hilman <khilman@baylibre.com> + - Roger Quadros <rogerq@kernel.org> + - Tony Lindgren <tony@atomide.com> + +properties: + compatible: + enum: + - ti,omap2-aes + - ti,omap3-aes + - ti,omap4-aes + + reg: + maxItems: 1 + + interrupts: + maxItems: 1 + + dmas: + maxItems: 2 + + dma-names: + items: + - const: tx + - const: rx + + ti,hwmods: + description: Name of the hwmod associated with the AES module + const: aes + deprecated: true + +required: + - compatible + - reg + - interrupts + +additionalProperties: false + +examples: + - | + aes@53500000 { + compatible = "ti,omap4-aes"; + reg = <0x53500000 0xa0>; + interrupts = <102>; + dmas = <&edma 6>, + <&edma 5>; + dma-names = "tx", "rx"; + }; diff --git a/Documentation/devicetree/bindings/crypto/ti,omap4-des.yaml b/Documentation/devicetree/bindings/crypto/ti,omap4-des.yaml new file mode 100644 index 0000000000000..f02f1e141218d --- /dev/null +++ b/Documentation/devicetree/bindings/crypto/ti,omap4-des.yaml @@ -0,0 +1,65 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/ti,omap4-des.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: OMAP4 DES crypto Module + +maintainers: + - Aaro Koskinen <aaro.koskinen@iki.fi> + - Andreas Kemnade <andreas@kemnade.info> + - Kevin Hilman <khilman@baylibre.com> + - Roger Quadros <rogerq@kernel.org> + - Tony Lindgren <tony@atomide.com> + +properties: + compatible: + const: ti,omap4-des + + reg: + maxItems: 1 + + interrupts: + maxItems: 1 + + dmas: + maxItems: 2 + + dma-names: + items: + - const: tx + - const: rx + + clocks: + maxItems: 1 + + clock-names: + items: + - const: fck + +dependencies: + dmas: [ dma-names ] + +required: + - compatible + - reg + - interrupts + - clocks + - clock-names + +additionalProperties: false + +examples: + - | + #include <dt-bindings/interrupt-controller/arm-gic.h> + + des@480a5000 { + compatible = "ti,omap4-des"; + reg = <0x480a5000 0xa0>; + interrupts = <GIC_SPI 82 IRQ_TYPE_LEVEL_HIGH>; + clocks = <&l3_iclk_div>; + clock-names = "fck"; + dmas = <&sdma 117>, <&sdma 116>; + dma-names = "tx", "rx"; + }; diff --git a/Documentation/devicetree/bindings/rng/atmel,at91-trng.yaml b/Documentation/devicetree/bindings/rng/atmel,at91-trng.yaml index b38f8252342ee..f78614100ea8b 100644 --- a/Documentation/devicetree/bindings/rng/atmel,at91-trng.yaml +++ b/Documentation/devicetree/bindings/rng/atmel,at91-trng.yaml @@ -24,6 +24,7 @@ properties: - items: - enum: - microchip,sam9x7-trng + - microchip,sama7d65-trng - const: microchip,sam9x60-trng clocks: diff --git a/arch/arm/crypto/aes-neonbs-glue.c b/arch/arm/crypto/aes-neonbs-glue.c index c60104dc15852..df5afe601e4a5 100644 --- a/arch/arm/crypto/aes-neonbs-glue.c +++ b/arch/arm/crypto/aes-neonbs-glue.c @@ -206,7 +206,7 @@ static int ctr_encrypt(struct skcipher_request *req) while (walk.nbytes > 0) { const u8 *src = walk.src.virt.addr; u8 *dst = walk.dst.virt.addr; - int bytes = walk.nbytes; + unsigned int bytes = walk.nbytes; if (unlikely(bytes < AES_BLOCK_SIZE)) src = dst = memcpy(buf + sizeof(buf) - bytes, diff --git a/arch/s390/configs/debug_defconfig b/arch/s390/configs/debug_defconfig index a8e84e80552d9..6b33429f1c4d4 100644 --- a/arch/s390/configs/debug_defconfig +++ b/arch/s390/configs/debug_defconfig @@ -816,6 +816,7 @@ CONFIG_PKEY_EP11=m CONFIG_PKEY_PCKMO=m CONFIG_PKEY_UV=m CONFIG_CRYPTO_PAES_S390=m +CONFIG_CRYPTO_PHMAC_S390=m CONFIG_CRYPTO_DEV_VIRTIO=m CONFIG_SYSTEM_BLACKLIST_KEYRING=y CONFIG_CRYPTO_KRB5=m diff --git a/arch/s390/configs/defconfig b/arch/s390/configs/defconfig index a209b6ae47914..b75eb27758507 100644 --- a/arch/s390/configs/defconfig +++ b/arch/s390/configs/defconfig @@ -803,6 +803,7 @@ CONFIG_PKEY_EP11=m CONFIG_PKEY_PCKMO=m CONFIG_PKEY_UV=m CONFIG_CRYPTO_PAES_S390=m +CONFIG_CRYPTO_PHMAC_S390=m CONFIG_CRYPTO_DEV_VIRTIO=m CONFIG_SYSTEM_BLACKLIST_KEYRING=y CONFIG_CRYPTO_KRB5=m diff --git a/arch/s390/crypto/Makefile b/arch/s390/crypto/Makefile index 1e5a1038d4914..998f4b656b18e 100644 --- a/arch/s390/crypto/Makefile +++ b/arch/s390/crypto/Makefile @@ -11,4 +11,5 @@ obj-$(CONFIG_CRYPTO_PAES_S390) += paes_s390.o obj-$(CONFIG_S390_PRNG) += prng.o obj-$(CONFIG_CRYPTO_GHASH_S390) += ghash_s390.o obj-$(CONFIG_CRYPTO_HMAC_S390) += hmac_s390.o +obj-$(CONFIG_CRYPTO_PHMAC_S390) += phmac_s390.o obj-y += arch_random.o diff --git a/arch/s390/crypto/hmac_s390.c b/arch/s390/crypto/hmac_s390.c index 93a1098d9f8d3..58444da9b004c 100644 --- a/arch/s390/crypto/hmac_s390.c +++ b/arch/s390/crypto/hmac_s390.c @@ -290,6 +290,7 @@ static int s390_hmac_export(struct shash_desc *desc, void *out) struct s390_kmac_sha2_ctx *ctx = shash_desc_ctx(desc); unsigned int bs = crypto_shash_blocksize(desc->tfm); unsigned int ds = bs / 2; + u64 lo = ctx->buflen[0]; union { u8 *u8; u64 *u64; @@ -301,9 +302,10 @@ static int s390_hmac_export(struct shash_desc *desc, void *out) else memcpy(p.u8, ctx->param, ds); p.u8 += ds; - put_unaligned(ctx->buflen[0], p.u64++); + lo += bs; + put_unaligned(lo, p.u64++); if (ds == SHA512_DIGEST_SIZE) - put_unaligned(ctx->buflen[1], p.u64); + put_unaligned(ctx->buflen[1] + (lo < bs), p.u64); return err; } @@ -316,14 +318,16 @@ static int s390_hmac_import(struct shash_desc *desc, const void *in) const u8 *u8; const u64 *u64; } p = { .u8 = in }; + u64 lo; int err; err = s390_hmac_sha2_init(desc); memcpy(ctx->param, p.u8, ds); p.u8 += ds; - ctx->buflen[0] = get_unaligned(p.u64++); + lo = get_unaligned(p.u64++); + ctx->buflen[0] = lo - bs; if (ds == SHA512_DIGEST_SIZE) - ctx->buflen[1] = get_unaligned(p.u64); + ctx->buflen[1] = get_unaligned(p.u64) - (lo < bs); if (ctx->buflen[0] | ctx->buflen[1]) ctx->gr0.ikp = 1; return err; diff --git a/arch/s390/crypto/paes_s390.c b/arch/s390/crypto/paes_s390.c index 8a340c16acb46..a624a43a2b540 100644 --- a/arch/s390/crypto/paes_s390.c +++ b/arch/s390/crypto/paes_s390.c @@ -1633,7 +1633,7 @@ static int __init paes_s390_init(void) /* with this pseudo devie alloc and start a crypto engine */ paes_crypto_engine = crypto_engine_alloc_init_and_set(paes_dev.this_device, - true, NULL, false, MAX_QLEN); + true, false, MAX_QLEN); if (!paes_crypto_engine) { rc = -ENOMEM; goto out_err; diff --git a/arch/s390/crypto/phmac_s390.c b/arch/s390/crypto/phmac_s390.c new file mode 100644 index 0000000000000..7ecfdc4fba2d0 --- /dev/null +++ b/arch/s390/crypto/phmac_s390.c @@ -0,0 +1,1048 @@ +// SPDX-License-Identifier: GPL-2.0+ +/* + * Copyright IBM Corp. 2025 + * + * s390 specific HMAC support for protected keys. + */ + +#define KMSG_COMPONENT "phmac_s390" +#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt + +#include <asm/cpacf.h> +#include <asm/pkey.h> +#include <crypto/engine.h> +#include <crypto/hash.h> +#include <crypto/internal/hash.h> +#include <crypto/sha2.h> +#include <linux/atomic.h> +#include <linux/cpufeature.h> +#include <linux/delay.h> +#include <linux/miscdevice.h> +#include <linux/module.h> +#include <linux/spinlock.h> + +static struct crypto_engine *phmac_crypto_engine; +#define MAX_QLEN 10 + +/* + * A simple hash walk helper + */ + +struct hash_walk_helper { + struct crypto_hash_walk walk; + const u8 *walkaddr; + int walkbytes; +}; + +/* + * Prepare hash walk helper. + * Set up the base hash walk, fill walkaddr and walkbytes. + * Returns 0 on success or negative value on error. + */ +static inline int hwh_prepare(struct ahash_request *req, + struct hash_walk_helper *hwh) +{ + hwh->walkbytes = crypto_hash_walk_first(req, &hwh->walk); + if (hwh->walkbytes < 0) + return hwh->walkbytes; + hwh->walkaddr = hwh->walk.data; + return 0; +} + +/* + * Advance hash walk helper by n bytes. + * Progress the walkbytes and walkaddr fields by n bytes. + * If walkbytes is then 0, pull next hunk from hash walk + * and update walkbytes and walkaddr. + * If n is negative, unmap hash walk and return error. + * Returns 0 on success or negative value on error. + */ +static inline int hwh_advance(struct hash_walk_helper *hwh, int n) +{ + if (n < 0) + return crypto_hash_walk_done(&hwh->walk, n); + + hwh->walkbytes -= n; + hwh->walkaddr += n; + if (hwh->walkbytes > 0) + return 0; + + hwh->walkbytes = crypto_hash_walk_done(&hwh->walk, 0); + if (hwh->walkbytes < 0) + return hwh->walkbytes; + + hwh->walkaddr = hwh->walk.data; + return 0; +} + +/* + * KMAC param block layout for sha2 function codes: + * The layout of the param block for the KMAC instruction depends on the + * blocksize of the used hashing sha2-algorithm function codes. The param block + * contains the hash chaining value (cv), the input message bit-length (imbl) + * and the hmac-secret (key). To prevent code duplication, the sizes of all + * these are calculated based on the blocksize. + * + * param-block: + * +-------+ + * | cv | + * +-------+ + * | imbl | + * +-------+ + * | key | + * +-------+ + * + * sizes: + * part | sh2-alg | calculation | size | type + * -----+---------+-------------+------+-------- + * cv | 224/256 | blocksize/2 | 32 | u64[8] + * | 384/512 | | 64 | u128[8] + * imbl | 224/256 | blocksize/8 | 8 | u64 + * | 384/512 | | 16 | u128 + * key | 224/256 | blocksize | 96 | u8[96] + * | 384/512 | | 160 | u8[160] + */ + +#define MAX_DIGEST_SIZE SHA512_DIGEST_SIZE +#define MAX_IMBL_SIZE sizeof(u128) +#define MAX_BLOCK_SIZE SHA512_BLOCK_SIZE + +#define SHA2_CV_SIZE(bs) ((bs) >> 1) +#define SHA2_IMBL_SIZE(bs) ((bs) >> 3) + +#define SHA2_IMBL_OFFSET(bs) (SHA2_CV_SIZE(bs)) +#define SHA2_KEY_OFFSET(bs) (SHA2_CV_SIZE(bs) + SHA2_IMBL_SIZE(bs)) + +#define PHMAC_MAX_KEYSIZE 256 +#define PHMAC_SHA256_PK_SIZE (SHA256_BLOCK_SIZE + 32) +#define PHMAC_SHA512_PK_SIZE (SHA512_BLOCK_SIZE + 32) +#define PHMAC_MAX_PK_SIZE PHMAC_SHA512_PK_SIZE + +/* phmac protected key struct */ +struct phmac_protkey { + u32 type; + u32 len; + u8 protkey[PHMAC_MAX_PK_SIZE]; +}; + +#define PK_STATE_NO_KEY 0 +#define PK_STATE_CONVERT_IN_PROGRESS 1 +#define PK_STATE_VALID 2 + +/* phmac tfm context */ +struct phmac_tfm_ctx { + /* source key material used to derive a protected key from */ + u8 keybuf[PHMAC_MAX_KEYSIZE]; + unsigned int keylen; + + /* cpacf function code to use with this protected key type */ + long fc; + + /* nr of requests enqueued via crypto engine which use this tfm ctx */ + atomic_t via_engine_ctr; + + /* spinlock to atomic read/update all the following fields */ + spinlock_t pk_lock; + + /* see PK_STATE* defines above, < 0 holds convert failure rc */ + int pk_state; + /* if state is valid, pk holds the protected key */ + struct phmac_protkey pk; +}; + +union kmac_gr0 { + unsigned long reg; + struct { + unsigned long : 48; + unsigned long ikp : 1; + unsigned long iimp : 1; + unsigned long ccup : 1; + unsigned long : 6; + unsigned long fc : 7; + }; +}; + +struct kmac_sha2_ctx { + u8 param[MAX_DIGEST_SIZE + MAX_IMBL_SIZE + PHMAC_MAX_PK_SIZE]; + union kmac_gr0 gr0; + u8 buf[MAX_BLOCK_SIZE]; + u64 buflen[2]; +}; + +/* phmac request context */ +struct phmac_req_ctx { + struct hash_walk_helper hwh; + struct kmac_sha2_ctx kmac_ctx; + bool final; +}; + +/* + * Pkey 'token' struct used to derive a protected key value from a clear key. + */ +struct hmac_clrkey_token { + u8 type; + u8 res0[3]; + u8 version; + u8 res1[3]; + u32 keytype; + u32 len; + u8 key[]; +} __packed; + +static int hash_key(const u8 *in, unsigned int inlen, + u8 *digest, unsigned int digestsize) +{ + unsigned long func; + union { + struct sha256_paramblock { + u32 h[8]; + u64 mbl; + } sha256; + struct sha512_paramblock { + u64 h[8]; + u128 mbl; + } sha512; + } __packed param; + +#define PARAM_INIT(x, y, z) \ + param.sha##x.h[0] = SHA##y ## _H0; \ + param.sha##x.h[1] = SHA##y ## _H1; \ + param.sha##x.h[2] = SHA##y ## _H2; \ + param.sha##x.h[3] = SHA##y ## _H3; \ + param.sha##x.h[4] = SHA##y ## _H4; \ + param.sha##x.h[5] = SHA##y ## _H5; \ + param.sha##x.h[6] = SHA##y ## _H6; \ + param.sha##x.h[7] = SHA##y ## _H7; \ + param.sha##x.mbl = (z) + + switch (digestsize) { + case SHA224_DIGEST_SIZE: + func = CPACF_KLMD_SHA_256; + PARAM_INIT(256, 224, inlen * 8); + break; + case SHA256_DIGEST_SIZE: + func = CPACF_KLMD_SHA_256; + PARAM_INIT(256, 256, inlen * 8); + break; + case SHA384_DIGEST_SIZE: + func = CPACF_KLMD_SHA_512; + PARAM_INIT(512, 384, inlen * 8); + break; + case SHA512_DIGEST_SIZE: + func = CPACF_KLMD_SHA_512; + PARAM_INIT(512, 512, inlen * 8); + break; + default: + return -EINVAL; + } + +#undef PARAM_INIT + + cpacf_klmd(func, ¶m, in, inlen); + + memcpy(digest, ¶m, digestsize); + + return 0; +} + +/* + * make_clrkey_token() - wrap the clear key into a pkey clearkey token. + */ +static inline int make_clrkey_token(const u8 *clrkey, size_t clrkeylen, + unsigned int digestsize, u8 *dest) +{ + struct hmac_clrkey_token *token = (struct hmac_clrkey_token *)dest; + unsigned int blocksize; + int rc; + + token->type = 0x00; + token->version = 0x02; + switch (digestsize) { + case SHA224_DIGEST_SIZE: + case SHA256_DIGEST_SIZE: + token->keytype = PKEY_KEYTYPE_HMAC_512; + blocksize = 64; + break; + case SHA384_DIGEST_SIZE: + case SHA512_DIGEST_SIZE: + token->keytype = PKEY_KEYTYPE_HMAC_1024; + blocksize = 128; + break; + default: + return -EINVAL; + } + token->len = blocksize; + + if (clrkeylen > blocksize) { + rc = hash_key(clrkey, clrkeylen, token->key, digestsize); + if (rc) + return rc; + } else { + memcpy(token->key, clrkey, clrkeylen); + } + + return 0; +} + +/* + * phmac_tfm_ctx_setkey() - Set key value into tfm context, maybe construct + * a clear key token digestible by pkey from a clear key value. + */ +static inline int phmac_tfm_ctx_setkey(struct phmac_tfm_ctx *tfm_ctx, + const u8 *key, unsigned int keylen) +{ + if (keylen > sizeof(tfm_ctx->keybuf)) + return -EINVAL; + + memcpy(tfm_ctx->keybuf, key, keylen); + tfm_ctx->keylen = keylen; + + return 0; +} + +/* + * Convert the raw key material into a protected key via PKEY api. + * This function may sleep - don't call in non-sleeping context. + */ +static inline int convert_key(const u8 *key, unsigned int keylen, + struct phmac_protkey *pk) +{ + int rc, i; + + pk->len = sizeof(pk->protkey); + + /* + * In case of a busy card retry with increasing delay + * of 200, 400, 800 and 1600 ms - in total 3 s. + */ + for (rc = -EIO, i = 0; rc && i < 5; i++) { + if (rc == -EBUSY && msleep_interruptible((1 << i) * 100)) { + rc = -EINTR; + goto out; + } + rc = pkey_key2protkey(key, keylen, + pk->protkey, &pk->len, &pk->type, + PKEY_XFLAG_NOMEMALLOC); + } + +out: + pr_debug("rc=%d\n", rc); + return rc; +} + +/* + * (Re-)Convert the raw key material from the tfm ctx into a protected + * key via convert_key() function. Update the pk_state, pk_type, pk_len + * and the protected key in the tfm context. + * Please note this function may be invoked concurrently with the very + * same tfm context. The pk_lock spinlock in the context ensures an + * atomic update of the pk and the pk state but does not guarantee any + * order of update. So a fresh converted valid protected key may get + * updated with an 'old' expired key value. As the cpacf instructions + * detect this, refuse to operate with an invalid key and the calling + * code triggers a (re-)conversion this does no harm. This may lead to + * unnecessary additional conversion but never to invalid data on the + * hash operation. + */ +static int phmac_convert_key(struct phmac_tfm_ctx *tfm_ctx) +{ + struct phmac_protkey pk; + int rc; + + spin_lock_bh(&tfm_ctx->pk_lock); + tfm_ctx->pk_state = PK_STATE_CONVERT_IN_PROGRESS; + spin_unlock_bh(&tfm_ctx->pk_lock); + + rc = convert_key(tfm_ctx->keybuf, tfm_ctx->keylen, &pk); + + /* update context */ + spin_lock_bh(&tfm_ctx->pk_lock); + if (rc) { + tfm_ctx->pk_state = rc; + } else { + tfm_ctx->pk_state = PK_STATE_VALID; + tfm_ctx->pk = pk; + } + spin_unlock_bh(&tfm_ctx->pk_lock); + + memzero_explicit(&pk, sizeof(pk)); + pr_debug("rc=%d\n", rc); + return rc; +} + +/* + * kmac_sha2_set_imbl - sets the input message bit-length based on the blocksize + */ +static inline void kmac_sha2_set_imbl(u8 *param, u64 buflen_lo, + u64 buflen_hi, unsigned int blocksize) +{ + u8 *imbl = param + SHA2_IMBL_OFFSET(blocksize); + + switch (blocksize) { + case SHA256_BLOCK_SIZE: + *(u64 *)imbl = buflen_lo * BITS_PER_BYTE; + break; + case SHA512_BLOCK_SIZE: + *(u128 *)imbl = (((u128)buflen_hi << 64) + buflen_lo) << 3; + break; + default: + break; + } +} + +static int phmac_kmac_update(struct ahash_request *req, bool maysleep) +{ + struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); + struct phmac_tfm_ctx *tfm_ctx = crypto_ahash_ctx(tfm); + struct phmac_req_ctx *req_ctx = ahash_request_ctx(req); + struct kmac_sha2_ctx *ctx = &req_ctx->kmac_ctx; + struct hash_walk_helper *hwh = &req_ctx->hwh; + unsigned int bs = crypto_ahash_blocksize(tfm); + unsigned int offset, k, n; + int rc = 0; + + /* + * The walk is always mapped when this function is called. + * Note that in case of partial processing or failure the walk + * is NOT unmapped here. So a follow up task may reuse the walk + * or in case of unrecoverable failure needs to unmap it. + */ + + while (hwh->walkbytes > 0) { + /* check sha2 context buffer */ + offset = ctx->buflen[0] % bs; + if (offset + hwh->walkbytes < bs) + goto store; + + if (offset) { + /* fill ctx buffer up to blocksize and process this block */ + n = bs - offset; + memcpy(ctx->buf + offset, hwh->walkaddr, n); + ctx->gr0.iimp = 1; + for (;;) { + k = _cpacf_kmac(&ctx->gr0.reg, ctx->param, ctx->buf, bs); + if (likely(k == bs)) + break; + if (unlikely(k > 0)) { + /* + * Can't deal with hunks smaller than blocksize. + * And kmac should always return the nr of + * processed bytes as 0 or a multiple of the + * blocksize. + */ + rc = -EIO; + goto out; + } + /* protected key is invalid and needs re-conversion */ + if (!maysleep) { + rc = -EKEYEXPIRED; + goto out; + } + rc = phmac_convert_key(tfm_ctx); + if (rc) + goto out; + spin_lock_bh(&tfm_ctx->pk_lock); + memcpy(ctx->param + SHA2_KEY_OFFSET(bs), + tfm_ctx->pk.protkey, tfm_ctx->pk.len); + spin_unlock_bh(&tfm_ctx->pk_lock); + } + ctx->buflen[0] += n; + if (ctx->buflen[0] < n) + ctx->buflen[1]++; + rc = hwh_advance(hwh, n); + if (unlikely(rc)) + goto out; + offset = 0; + } + + /* process as many blocks as possible from the walk */ + while (hwh->walkbytes >= bs) { + n = (hwh->walkbytes / bs) * bs; + ctx->gr0.iimp = 1; + k = _cpacf_kmac(&ctx->gr0.reg, ctx->param, hwh->walkaddr, n); + if (likely(k > 0)) { + ctx->buflen[0] += k; + if (ctx->buflen[0] < k) + ctx->buflen[1]++; + rc = hwh_advance(hwh, k); + if (unlikely(rc)) + goto out; + } + if (unlikely(k < n)) { + /* protected key is invalid and needs re-conversion */ + if (!maysleep) { + rc = -EKEYEXPIRED; + goto out; + } + rc = phmac_convert_key(tfm_ctx); + if (rc) + goto out; + spin_lock_bh(&tfm_ctx->pk_lock); + memcpy(ctx->param + SHA2_KEY_OFFSET(bs), + tfm_ctx->pk.protkey, tfm_ctx->pk.len); + spin_unlock_bh(&tfm_ctx->pk_lock); + } + } + +store: + /* store incomplete block in context buffer */ + if (hwh->walkbytes) { + memcpy(ctx->buf + offset, hwh->walkaddr, hwh->walkbytes); + ctx->buflen[0] += hwh->walkbytes; + if (ctx->buflen[0] < hwh->walkbytes) + ctx->buflen[1]++; + rc = hwh_advance(hwh, hwh->walkbytes); + if (unlikely(rc)) + goto out; + } + + } /* end of while (hwh->walkbytes > 0) */ + +out: + pr_debug("rc=%d\n", rc); + return rc; +} + +static int phmac_kmac_final(struct ahash_request *req, bool maysleep) +{ + struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); + struct phmac_tfm_ctx *tfm_ctx = crypto_ahash_ctx(tfm); + struct phmac_req_ctx *req_ctx = ahash_request_ctx(req); + struct kmac_sha2_ctx *ctx = &req_ctx->kmac_ctx; + unsigned int ds = crypto_ahash_digestsize(tfm); + unsigned int bs = crypto_ahash_blocksize(tfm); + unsigned int k, n; + int rc = 0; + + n = ctx->buflen[0] % bs; + ctx->gr0.iimp = 0; + kmac_sha2_set_imbl(ctx->param, ctx->buflen[0], ctx->buflen[1], bs); + for (;;) { + k = _cpacf_kmac(&ctx->gr0.reg, ctx->param, ctx->buf, n); + if (likely(k == n)) + break; + if (unlikely(k > 0)) { + /* Can't deal with hunks smaller than blocksize. */ + rc = -EIO; + goto out; + } + /* protected key is invalid and needs re-conversion */ + if (!maysleep) { + rc = -EKEYEXPIRED; + goto out; + } + rc = phmac_convert_key(tfm_ctx); + if (rc) + goto out; + spin_lock_bh(&tfm_ctx->pk_lock); + memcpy(ctx->param + SHA2_KEY_OFFSET(bs), + tfm_ctx->pk.protkey, tfm_ctx->pk.len); + spin_unlock_bh(&tfm_ctx->pk_lock); + } + + memcpy(req->result, ctx->param, ds); + +out: + pr_debug("rc=%d\n", rc); + return rc; +} + +static int phmac_init(struct ahash_request *req) +{ + struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); + struct phmac_tfm_ctx *tfm_ctx = crypto_ahash_ctx(tfm); + struct phmac_req_ctx *req_ctx = ahash_request_ctx(req); + struct kmac_sha2_ctx *kmac_ctx = &req_ctx->kmac_ctx; + unsigned int bs = crypto_ahash_blocksize(tfm); + int rc = 0; + + /* zero request context (includes the kmac sha2 context) */ + memset(req_ctx, 0, sizeof(*req_ctx)); + + /* + * setkey() should have set a valid fc into the tfm context. + * Copy this function code into the gr0 field of the kmac context. + */ + if (!tfm_ctx->fc) { + rc = -ENOKEY; + goto out; + } + kmac_ctx->gr0.fc = tfm_ctx->fc; + + /* + * Copy the pk from tfm ctx into kmac ctx. The protected key + * may be outdated but update() and final() will handle this. + */ + spin_lock_bh(&tfm_ctx->pk_lock); + memcpy(kmac_ctx->param + SHA2_KEY_OFFSET(bs), + tfm_ctx->pk.protkey, tfm_ctx->pk.len); + spin_unlock_bh(&tfm_ctx->pk_lock); + +out: + pr_debug("rc=%d\n", rc); + return rc; +} + +static int phmac_update(struct ahash_request *req) +{ + struct phmac_req_ctx *req_ctx = ahash_request_ctx(req); + struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); + struct phmac_tfm_ctx *tfm_ctx = crypto_ahash_ctx(tfm); + struct kmac_sha2_ctx *kmac_ctx = &req_ctx->kmac_ctx; + struct hash_walk_helper *hwh = &req_ctx->hwh; + int rc; + + /* prep the walk in the request context */ + rc = hwh_prepare(req, hwh); + if (rc) + goto out; + + /* Try synchronous operation if no active engine usage */ + if (!atomic_read(&tfm_ctx->via_engine_ctr)) { + rc = phmac_kmac_update(req, false); + if (rc == 0) + goto out; + } + + /* + * If sync operation failed or key expired or there are already + * requests enqueued via engine, fallback to async. Mark tfm as + * using engine to serialize requests. + */ + if (rc == 0 || rc == -EKEYEXPIRED) { + atomic_inc(&tfm_ctx->via_engine_ctr); + rc = crypto_transfer_hash_request_to_engine(phmac_crypto_engine, req); + if (rc != -EINPROGRESS) + atomic_dec(&tfm_ctx->via_engine_ctr); + } + + if (rc != -EINPROGRESS) { + hwh_advance(hwh, rc); + memzero_explicit(kmac_ctx, sizeof(*kmac_ctx)); + } + +out: + pr_debug("rc=%d\n", rc); + return rc; +} + +static int phmac_final(struct ahash_request *req) +{ + struct phmac_req_ctx *req_ctx = ahash_request_ctx(req); + struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); + struct phmac_tfm_ctx *tfm_ctx = crypto_ahash_ctx(tfm); + struct kmac_sha2_ctx *kmac_ctx = &req_ctx->kmac_ctx; + int rc = 0; + + /* Try synchronous operation if no active engine usage */ + if (!atomic_read(&tfm_ctx->via_engine_ctr)) { + rc = phmac_kmac_final(req, false); + if (rc == 0) + goto out; + } + + /* + * If sync operation failed or key expired or there are already + * requests enqueued via engine, fallback to async. Mark tfm as + * using engine to serialize requests. + */ + if (rc == 0 || rc == -EKEYEXPIRED) { + req->nbytes = 0; + req_ctx->final = true; + atomic_inc(&tfm_ctx->via_engine_ctr); + rc = crypto_transfer_hash_request_to_engine(phmac_crypto_engine, req); + if (rc != -EINPROGRESS) + atomic_dec(&tfm_ctx->via_engine_ctr); + } + +out: + if (rc != -EINPROGRESS) + memzero_explicit(kmac_ctx, sizeof(*kmac_ctx)); + pr_debug("rc=%d\n", rc); + return rc; +} + +static int phmac_finup(struct ahash_request *req) +{ + struct phmac_req_ctx *req_ctx = ahash_request_ctx(req); + struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); + struct phmac_tfm_ctx *tfm_ctx = crypto_ahash_ctx(tfm); + struct kmac_sha2_ctx *kmac_ctx = &req_ctx->kmac_ctx; + struct hash_walk_helper *hwh = &req_ctx->hwh; + int rc; + + /* prep the walk in the request context */ + rc = hwh_prepare(req, hwh); + if (rc) + goto out; + + /* Try synchronous operations if no active engine usage */ + if (!atomic_read(&tfm_ctx->via_engine_ctr)) { + rc = phmac_kmac_update(req, false); + if (rc == 0) + req->nbytes = 0; + } + if (!rc && !req->nbytes && !atomic_read(&tfm_ctx->via_engine_ctr)) { + rc = phmac_kmac_final(req, false); + if (rc == 0) + goto out; + } + + /* + * If sync operation failed or key expired or there are already + * requests enqueued via engine, fallback to async. Mark tfm as + * using engine to serialize requests. + */ + if (rc == 0 || rc == -EKEYEXPIRED) { + req_ctx->final = true; + atomic_inc(&tfm_ctx->via_engine_ctr); + rc = crypto_transfer_hash_request_to_engine(phmac_crypto_engine, req); + if (rc != -EINPROGRESS) + atomic_dec(&tfm_ctx->via_engine_ctr); + } + + if (rc != -EINPROGRESS) + hwh_advance(hwh, rc); + +out: + if (rc != -EINPROGRESS) + memzero_explicit(kmac_ctx, sizeof(*kmac_ctx)); + pr_debug("rc=%d\n", rc); + return rc; +} + +static int phmac_digest(struct ahash_request *req) +{ + int rc; + + rc = phmac_init(req); + if (rc) + goto out; + + rc = phmac_finup(req); + +out: + pr_debug("rc=%d\n", rc); + return rc; +} + +static int phmac_setkey(struct crypto_ahash *tfm, + const u8 *key, unsigned int keylen) +{ + struct phmac_tfm_ctx *tfm_ctx = crypto_ahash_ctx(tfm); + unsigned int ds = crypto_ahash_digestsize(tfm); + unsigned int bs = crypto_ahash_blocksize(tfm); + unsigned int tmpkeylen; + u8 *tmpkey = NULL; + int rc = 0; + + if (!crypto_ahash_tested(tfm)) { + /* + * selftest running: key is a raw hmac clear key and needs + * to get embedded into a 'clear key token' in order to have + * it correctly processed by the pkey module. + */ + tmpkeylen = sizeof(struct hmac_clrkey_token) + bs; + tmpkey = kzalloc(tmpkeylen, GFP_KERNEL); + if (!tmpkey) { + rc = -ENOMEM; + goto out; + } + rc = make_clrkey_token(key, keylen, ds, tmpkey); + if (rc) + goto out; + keylen = tmpkeylen; + key = tmpkey; + } + + /* copy raw key into tfm context */ + rc = phmac_tfm_ctx_setkey(tfm_ctx, key, keylen); + if (rc) + goto out; + + /* convert raw key into protected key */ + rc = phmac_convert_key(tfm_ctx); + if (rc) + goto out; + + /* set function code in tfm context, check for valid pk type */ + switch (ds) { + case SHA224_DIGEST_SIZE: + if (tfm_ctx->pk.type != PKEY_KEYTYPE_HMAC_512) + rc = -EINVAL; + else + tfm_ctx->fc = CPACF_KMAC_PHMAC_SHA_224; + break; + case SHA256_DIGEST_SIZE: + if (tfm_ctx->pk.type != PKEY_KEYTYPE_HMAC_512) + rc = -EINVAL; + else + tfm_ctx->fc = CPACF_KMAC_PHMAC_SHA_256; + break; + case SHA384_DIGEST_SIZE: + if (tfm_ctx->pk.type != PKEY_KEYTYPE_HMAC_1024) + rc = -EINVAL; + else + tfm_ctx->fc = CPACF_KMAC_PHMAC_SHA_384; + break; + case SHA512_DIGEST_SIZE: + if (tfm_ctx->pk.type != PKEY_KEYTYPE_HMAC_1024) + rc = -EINVAL; + else + tfm_ctx->fc = CPACF_KMAC_PHMAC_SHA_512; + break; + default: + tfm_ctx->fc = 0; + rc = -EINVAL; + } + +out: + kfree(tmpkey); + pr_debug("rc=%d\n", rc); + return rc; +} + +static int phmac_export(struct ahash_request *req, void *out) +{ + struct phmac_req_ctx *req_ctx = ahash_request_ctx(req); + struct kmac_sha2_ctx *ctx = &req_ctx->kmac_ctx; + + memcpy(out, ctx, sizeof(*ctx)); + + return 0; +} + +static int phmac_import(struct ahash_request *req, const void *in) +{ + struct phmac_req_ctx *req_ctx = ahash_request_ctx(req); + struct kmac_sha2_ctx *ctx = &req_ctx->kmac_ctx; + + memset(req_ctx, 0, sizeof(*req_ctx)); + memcpy(ctx, in, sizeof(*ctx)); + + return 0; +} + +static int phmac_init_tfm(struct crypto_ahash *tfm) +{ + struct phmac_tfm_ctx *tfm_ctx = crypto_ahash_ctx(tfm); + + memset(tfm_ctx, 0, sizeof(*tfm_ctx)); + spin_lock_init(&tfm_ctx->pk_lock); + + crypto_ahash_set_reqsize(tfm, sizeof(struct phmac_req_ctx)); + + return 0; +} + +static void phmac_exit_tfm(struct crypto_ahash *tfm) +{ + struct phmac_tfm_ctx *tfm_ctx = crypto_ahash_ctx(tfm); + + memzero_explicit(tfm_ctx->keybuf, sizeof(tfm_ctx->keybuf)); + memzero_explicit(&tfm_ctx->pk, sizeof(tfm_ctx->pk)); +} + +static int phmac_do_one_request(struct crypto_engine *engine, void *areq) +{ + struct ahash_request *req = ahash_request_cast(areq); + struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); + struct phmac_tfm_ctx *tfm_ctx = crypto_ahash_ctx(tfm); + struct phmac_req_ctx *req_ctx = ahash_request_ctx(req); + struct kmac_sha2_ctx *kmac_ctx = &req_ctx->kmac_ctx; + struct hash_walk_helper *hwh = &req_ctx->hwh; + int rc = -EINVAL; + + /* + * Three kinds of requests come in here: + * update when req->nbytes > 0 and req_ctx->final is false + * final when req->nbytes = 0 and req_ctx->final is true + * finup when req->nbytes > 0 and req_ctx->final is true + * For update and finup the hwh walk needs to be prepared and + * up to date but the actual nr of bytes in req->nbytes may be + * any non zero number. For final there is no hwh walk needed. + */ + + if (req->nbytes) { + rc = phmac_kmac_update(req, true); + if (rc == -EKEYEXPIRED) { + /* + * Protected key expired, conversion is in process. + * Trigger a re-schedule of this request by returning + * -ENOSPC ("hardware queue full") to the crypto engine. + * To avoid immediately re-invocation of this callback, + * tell scheduler to voluntarily give up the CPU here. + */ + pr_debug("rescheduling request\n"); + cond_resched(); + return -ENOSPC; + } else if (rc) { + hwh_advance(hwh, rc); + goto out; + } + req->nbytes = 0; + } + + if (req_ctx->final) { + rc = phmac_kmac_final(req, true); + if (rc == -EKEYEXPIRED) { + /* + * Protected key expired, conversion is in process. + * Trigger a re-schedule of this request by returning + * -ENOSPC ("hardware queue full") to the crypto engine. + * To avoid immediately re-invocation of this callback, + * tell scheduler to voluntarily give up the CPU here. + */ + pr_debug("rescheduling request\n"); + cond_resched(); + return -ENOSPC; + } + } + +out: + if (rc || req_ctx->final) + memzero_explicit(kmac_ctx, sizeof(*kmac_ctx)); + pr_debug("request complete with rc=%d\n", rc); + local_bh_disable(); + atomic_dec(&tfm_ctx->via_engine_ctr); + crypto_finalize_hash_request(engine, req, rc); + local_bh_enable(); + return rc; +} + +#define S390_ASYNC_PHMAC_ALG(x) \ +{ \ + .base = { \ + .init = phmac_init, \ + .update = phmac_update, \ + .final = phmac_final, \ + .finup = phmac_finup, \ + .digest = phmac_digest, \ + .setkey = phmac_setkey, \ + .import = phmac_import, \ + .export = phmac_export, \ + .init_tfm = phmac_init_tfm, \ + .exit_tfm = phmac_exit_tfm, \ + .halg = { \ + .digestsize = SHA##x##_DIGEST_SIZE, \ + .statesize = sizeof(struct kmac_sha2_ctx), \ + .base = { \ + .cra_name = "phmac(sha" #x ")", \ + .cra_driver_name = "phmac_s390_sha" #x, \ + .cra_blocksize = SHA##x##_BLOCK_SIZE, \ + .cra_priority = 400, \ + .cra_flags = CRYPTO_ALG_ASYNC | \ + CRYPTO_ALG_NO_FALLBACK, \ + .cra_ctxsize = sizeof(struct phmac_tfm_ctx), \ + .cra_module = THIS_MODULE, \ + }, \ + }, \ + }, \ + .op = { \ + .do_one_request = phmac_do_one_request, \ + }, \ +} + +static struct phmac_alg { + unsigned int fc; + struct ahash_engine_alg alg; + bool registered; +} phmac_algs[] = { + { + .fc = CPACF_KMAC_PHMAC_SHA_224, + .alg = S390_ASYNC_PHMAC_ALG(224), + }, { + .fc = CPACF_KMAC_PHMAC_SHA_256, + .alg = S390_ASYNC_PHMAC_ALG(256), + }, { + .fc = CPACF_KMAC_PHMAC_SHA_384, + .alg = S390_ASYNC_PHMAC_ALG(384), + }, { + .fc = CPACF_KMAC_PHMAC_SHA_512, + .alg = S390_ASYNC_PHMAC_ALG(512), + } +}; + +static struct miscdevice phmac_dev = { + .name = "phmac", + .minor = MISC_DYNAMIC_MINOR, +}; + +static void s390_phmac_exit(void) +{ + struct phmac_alg *phmac; + int i; + + if (phmac_crypto_engine) { + crypto_engine_stop(phmac_crypto_engine); + crypto_engine_exit(phmac_crypto_engine); + } + + for (i = ARRAY_SIZE(phmac_algs) - 1; i >= 0; i--) { + phmac = &phmac_algs[i]; + if (phmac->registered) + crypto_engine_unregister_ahash(&phmac->alg); + } + + misc_deregister(&phmac_dev); +} + +static int __init s390_phmac_init(void) +{ + struct phmac_alg *phmac; + int i, rc; + + /* for selftest cpacf klmd subfunction is needed */ + if (!cpacf_query_func(CPACF_KLMD, CPACF_KLMD_SHA_256)) + return -ENODEV; + if (!cpacf_query_func(CPACF_KLMD, CPACF_KLMD_SHA_512)) + return -ENODEV; + + /* register a simple phmac pseudo misc device */ + rc = misc_register(&phmac_dev); + if (rc) + return rc; + + /* with this pseudo device alloc and start a crypto engine */ + phmac_crypto_engine = + crypto_engine_alloc_init_and_set(phmac_dev.this_device, + true, false, MAX_QLEN); + if (!phmac_crypto_engine) { + rc = -ENOMEM; + goto out_err; + } + rc = crypto_engine_start(phmac_crypto_engine); + if (rc) { + crypto_engine_exit(phmac_crypto_engine); + phmac_crypto_engine = NULL; + goto out_err; + } + + for (i = 0; i < ARRAY_SIZE(phmac_algs); i++) { + phmac = &phmac_algs[i]; + if (!cpacf_query_func(CPACF_KMAC, phmac->fc)) + continue; + rc = crypto_engine_register_ahash(&phmac->alg); + if (rc) + goto out_err; + phmac->registered = true; + pr_debug("%s registered\n", phmac->alg.base.halg.base.cra_name); + } + + return 0; + +out_err: + s390_phmac_exit(); + return rc; +} + +module_init(s390_phmac_init); +module_exit(s390_phmac_exit); + +MODULE_ALIAS_CRYPTO("phmac(sha224)"); +MODULE_ALIAS_CRYPTO("phmac(sha256)"); +MODULE_ALIAS_CRYPTO("phmac(sha384)"); +MODULE_ALIAS_CRYPTO("phmac(sha512)"); + +MODULE_DESCRIPTION("S390 HMAC driver for protected keys"); +MODULE_LICENSE("GPL"); diff --git a/arch/s390/crypto/sha.h b/arch/s390/crypto/sha.h index d757ccbce2b4f..cadb4b13622ae 100644 --- a/arch/s390/crypto/sha.h +++ b/arch/s390/crypto/sha.h @@ -27,6 +27,9 @@ struct s390_sha_ctx { u64 state[SHA512_DIGEST_SIZE / sizeof(u64)]; u64 count_hi; } sha512; + struct { + __le64 state[SHA3_STATE_SIZE / sizeof(u64)]; + } sha3; }; int func; /* KIMD function to use */ bool first_message_part; diff --git a/arch/s390/crypto/sha3_256_s390.c b/arch/s390/crypto/sha3_256_s390.c index 4a7731ac6bcd6..03bb4f4bab701 100644 --- a/arch/s390/crypto/sha3_256_s390.c +++ b/arch/s390/crypto/sha3_256_s390.c @@ -35,23 +35,33 @@ static int sha3_256_init(struct shash_desc *desc) static int sha3_256_export(struct shash_desc *desc, void *out) { struct s390_sha_ctx *sctx = shash_desc_ctx(desc); - struct sha3_state *octx = out; + union { + u8 *u8; + u64 *u64; + } p = { .u8 = out }; + int i; if (sctx->first_message_part) { - memset(sctx->state, 0, sizeof(sctx->state)); - sctx->first_message_part = 0; + memset(out, 0, SHA3_STATE_SIZE); + return 0; } - memcpy(octx->st, sctx->state, sizeof(octx->st)); + for (i = 0; i < SHA3_STATE_SIZE / 8; i++) + put_unaligned(le64_to_cpu(sctx->sha3.state[i]), p.u64++); return 0; } static int sha3_256_import(struct shash_desc *desc, const void *in) { struct s390_sha_ctx *sctx = shash_desc_ctx(desc); - const struct sha3_state *ictx = in; - + union { + const u8 *u8; + const u64 *u64; + } p = { .u8 = in }; + int i; + + for (i = 0; i < SHA3_STATE_SIZE / 8; i++) + sctx->sha3.state[i] = cpu_to_le64(get_unaligned(p.u64++)); sctx->count = 0; - memcpy(sctx->state, ictx->st, sizeof(ictx->st)); sctx->first_message_part = 0; sctx->func = CPACF_KIMD_SHA3_256; diff --git a/arch/s390/crypto/sha3_512_s390.c b/arch/s390/crypto/sha3_512_s390.c index 018f02fff4446..a5c9690eecb19 100644 --- a/arch/s390/crypto/sha3_512_s390.c +++ b/arch/s390/crypto/sha3_512_s390.c @@ -34,24 +34,33 @@ static int sha3_512_init(struct shash_desc *desc) static int sha3_512_export(struct shash_desc *desc, void *out) { struct s390_sha_ctx *sctx = shash_desc_ctx(desc); - struct sha3_state *octx = out; - + union { + u8 *u8; + u64 *u64; + } p = { .u8 = out }; + int i; if (sctx->first_message_part) { - memset(sctx->state, 0, sizeof(sctx->state)); - sctx->first_message_part = 0; + memset(out, 0, SHA3_STATE_SIZE); + return 0; } - memcpy(octx->st, sctx->state, sizeof(octx->st)); + for (i = 0; i < SHA3_STATE_SIZE / 8; i++) + put_unaligned(le64_to_cpu(sctx->sha3.state[i]), p.u64++); return 0; } static int sha3_512_import(struct shash_desc *desc, const void *in) { struct s390_sha_ctx *sctx = shash_desc_ctx(desc); - const struct sha3_state *ictx = in; - + union { + const u8 *u8; + const u64 *u64; + } p = { .u8 = in }; + int i; + + for (i = 0; i < SHA3_STATE_SIZE / 8; i++) + sctx->sha3.state[i] = cpu_to_le64(get_unaligned(p.u64++)); sctx->count = 0; - memcpy(sctx->state, ictx->st, sizeof(ictx->st)); sctx->first_message_part = 0; sctx->func = CPACF_KIMD_SHA3_512; diff --git a/arch/s390/include/asm/cpacf.h b/arch/s390/include/asm/cpacf.h index 54cb97603ec05..4bc5317fbb123 100644 --- a/arch/s390/include/asm/cpacf.h +++ b/arch/s390/include/asm/cpacf.h @@ -129,6 +129,10 @@ #define CPACF_KMAC_HMAC_SHA_256 0x71 #define CPACF_KMAC_HMAC_SHA_384 0x72 #define CPACF_KMAC_HMAC_SHA_512 0x73 +#define CPACF_KMAC_PHMAC_SHA_224 0x78 +#define CPACF_KMAC_PHMAC_SHA_256 0x79 +#define CPACF_KMAC_PHMAC_SHA_384 0x7a +#define CPACF_KMAC_PHMAC_SHA_512 0x7b /* * Function codes for the PCKMO (PERFORM CRYPTOGRAPHIC KEY MANAGEMENT) diff --git a/arch/x86/crypto/aegis128-aesni-glue.c b/arch/x86/crypto/aegis128-aesni-glue.c index f1b6d40154e35..f1adfba1a76ea 100644 --- a/arch/x86/crypto/aegis128-aesni-glue.c +++ b/arch/x86/crypto/aegis128-aesni-glue.c @@ -104,10 +104,12 @@ static void crypto_aegis128_aesni_process_ad( } } -static __always_inline void +static __always_inline int crypto_aegis128_aesni_process_crypt(struct aegis_state *state, struct skcipher_walk *walk, bool enc) { + int err = 0; + while (walk->nbytes >= AEGIS128_BLOCK_SIZE) { if (enc) aegis128_aesni_enc(state, walk->src.virt.addr, @@ -119,7 +121,10 @@ crypto_aegis128_aesni_process_crypt(struct aegis_state *state, walk->dst.virt.addr, round_down(walk->nbytes, AEGIS128_BLOCK_SIZE)); - skcipher_walk_done(walk, walk->nbytes % AEGIS128_BLOCK_SIZE); + kernel_fpu_end(); + err = skcipher_walk_done(walk, + walk->nbytes % AEGIS128_BLOCK_SIZE); + kernel_fpu_begin(); } if (walk->nbytes) { @@ -131,8 +136,11 @@ crypto_aegis128_aesni_process_crypt(struct aegis_state *state, aegis128_aesni_dec_tail(state, walk->src.virt.addr, walk->dst.virt.addr, walk->nbytes); - skcipher_walk_done(walk, 0); + kernel_fpu_end(); + err = skcipher_walk_done(walk, 0); + kernel_fpu_begin(); } + return err; } static struct aegis_ctx *crypto_aegis128_aesni_ctx(struct crypto_aead *aead) @@ -165,7 +173,7 @@ static int crypto_aegis128_aesni_setauthsize(struct crypto_aead *tfm, return 0; } -static __always_inline void +static __always_inline int crypto_aegis128_aesni_crypt(struct aead_request *req, struct aegis_block *tag_xor, unsigned int cryptlen, bool enc) @@ -174,20 +182,24 @@ crypto_aegis128_aesni_crypt(struct aead_request *req, struct aegis_ctx *ctx = crypto_aegis128_aesni_ctx(tfm); struct skcipher_walk walk; struct aegis_state state; + int err; if (enc) - skcipher_walk_aead_encrypt(&walk, req, true); + err = skcipher_walk_aead_encrypt(&walk, req, false); else - skcipher_walk_aead_decrypt(&walk, req, true); + err = skcipher_walk_aead_decrypt(&walk, req, false); + if (err) + return err; kernel_fpu_begin(); aegis128_aesni_init(&state, &ctx->key, req->iv); crypto_aegis128_aesni_process_ad(&state, req->src, req->assoclen); - crypto_aegis128_aesni_process_crypt(&state, &walk, enc); - aegis128_aesni_final(&state, tag_xor, req->assoclen, cryptlen); - + err = crypto_aegis128_aesni_process_crypt(&state, &walk, enc); + if (err == 0) + aegis128_aesni_final(&state, tag_xor, req->assoclen, cryptlen); kernel_fpu_end(); + return err; } static int crypto_aegis128_aesni_encrypt(struct aead_request *req) @@ -196,8 +208,11 @@ static int crypto_aegis128_aesni_encrypt(struct aead_request *req) struct aegis_block tag = {}; unsigned int authsize = crypto_aead_authsize(tfm); unsigned int cryptlen = req->cryptlen; + int err; - crypto_aegis128_aesni_crypt(req, &tag, cryptlen, true); + err = crypto_aegis128_aesni_crypt(req, &tag, cryptlen, true); + if (err) + return err; scatterwalk_map_and_copy(tag.bytes, req->dst, req->assoclen + cryptlen, authsize, 1); @@ -212,11 +227,14 @@ static int crypto_aegis128_aesni_decrypt(struct aead_request *req) struct aegis_block tag; unsigned int authsize = crypto_aead_authsize(tfm); unsigned int cryptlen = req->cryptlen - authsize; + int err; scatterwalk_map_and_copy(tag.bytes, req->src, req->assoclen + cryptlen, authsize, 0); - crypto_aegis128_aesni_crypt(req, &tag, cryptlen, false); + err = crypto_aegis128_aesni_crypt(req, &tag, cryptlen, false); + if (err) + return err; return crypto_memneq(tag.bytes, zeros.bytes, authsize) ? -EBADMSG : 0; } diff --git a/arch/x86/crypto/aria_aesni_avx2_glue.c b/arch/x86/crypto/aria_aesni_avx2_glue.c index b4bddcd584577..007b250f774cd 100644 --- a/arch/x86/crypto/aria_aesni_avx2_glue.c +++ b/arch/x86/crypto/aria_aesni_avx2_glue.c @@ -9,6 +9,7 @@ #include <crypto/aria.h> #include <linux/crypto.h> #include <linux/err.h> +#include <linux/export.h> #include <linux/module.h> #include <linux/types.h> diff --git a/arch/x86/crypto/aria_aesni_avx_glue.c b/arch/x86/crypto/aria_aesni_avx_glue.c index ab9b38d05332a..4c88ef4eba827 100644 --- a/arch/x86/crypto/aria_aesni_avx_glue.c +++ b/arch/x86/crypto/aria_aesni_avx_glue.c @@ -9,6 +9,7 @@ #include <crypto/aria.h> #include <linux/crypto.h> #include <linux/err.h> +#include <linux/export.h> #include <linux/module.h> #include <linux/types.h> diff --git a/arch/x86/crypto/camellia_aesni_avx_glue.c b/arch/x86/crypto/camellia_aesni_avx_glue.c index a7d1623881424..5c321f255eb7b 100644 --- a/arch/x86/crypto/camellia_aesni_avx_glue.c +++ b/arch/x86/crypto/camellia_aesni_avx_glue.c @@ -8,6 +8,7 @@ #include <crypto/algapi.h> #include <linux/crypto.h> #include <linux/err.h> +#include <linux/export.h> #include <linux/module.h> #include <linux/types.h> diff --git a/arch/x86/crypto/camellia_glue.c b/arch/x86/crypto/camellia_glue.c index 3bd37d6641216..cbede120e5f29 100644 --- a/arch/x86/crypto/camellia_glue.c +++ b/arch/x86/crypto/camellia_glue.c @@ -10,6 +10,7 @@ #include <linux/unaligned.h> #include <linux/crypto.h> +#include <linux/export.h> #include <linux/init.h> #include <linux/module.h> #include <linux/types.h> diff --git a/arch/x86/crypto/curve25519-x86_64.c b/arch/x86/crypto/curve25519-x86_64.c index dcfc0de333dea..d587f05c3c8c3 100644 --- a/arch/x86/crypto/curve25519-x86_64.c +++ b/arch/x86/crypto/curve25519-x86_64.c @@ -7,6 +7,7 @@ #include <crypto/curve25519.h> #include <crypto/internal/kpp.h> +#include <linux/export.h> #include <linux/types.h> #include <linux/jump_label.h> #include <linux/kernel.h> diff --git a/arch/x86/crypto/serpent_avx_glue.c b/arch/x86/crypto/serpent_avx_glue.c index e640abc1cb8a7..9c8b3a335d5ca 100644 --- a/arch/x86/crypto/serpent_avx_glue.c +++ b/arch/x86/crypto/serpent_avx_glue.c @@ -12,6 +12,7 @@ #include <linux/types.h> #include <linux/crypto.h> #include <linux/err.h> +#include <linux/export.h> #include <crypto/algapi.h> #include <crypto/serpent.h> diff --git a/arch/x86/crypto/sm4_aesni_avx_glue.c b/arch/x86/crypto/sm4_aesni_avx_glue.c index 72867fc49ce8e..88caf418a06f1 100644 --- a/arch/x86/crypto/sm4_aesni_avx_glue.c +++ b/arch/x86/crypto/sm4_aesni_avx_glue.c @@ -11,6 +11,7 @@ #include <asm/fpu/api.h> #include <linux/module.h> #include <linux/crypto.h> +#include <linux/export.h> #include <linux/kernel.h> #include <crypto/internal/skcipher.h> #include <crypto/sm4.h> diff --git a/arch/x86/crypto/twofish_glue.c b/arch/x86/crypto/twofish_glue.c index 4c67184dc573e..8e9906d369021 100644 --- a/arch/x86/crypto/twofish_glue.c +++ b/arch/x86/crypto/twofish_glue.c @@ -40,6 +40,7 @@ #include <crypto/algapi.h> #include <crypto/twofish.h> +#include <linux/export.h> #include <linux/init.h> #include <linux/module.h> #include <linux/types.h> diff --git a/arch/x86/crypto/twofish_glue_3way.c b/arch/x86/crypto/twofish_glue_3way.c index 1a1ecfa7f72af..8ad77725bf60e 100644 --- a/arch/x86/crypto/twofish_glue_3way.c +++ b/arch/x86/crypto/twofish_glue_3way.c @@ -9,6 +9,7 @@ #include <crypto/algapi.h> #include <crypto/twofish.h> #include <linux/crypto.h> +#include <linux/export.h> #include <linux/init.h> #include <linux/module.h> #include <linux/types.h> diff --git a/crypto/ahash.c b/crypto/ahash.c index bc84a07c924c3..a227793d2c5b5 100644 --- a/crypto/ahash.c +++ b/crypto/ahash.c @@ -29,19 +29,6 @@ #define CRYPTO_ALG_TYPE_AHASH_MASK 0x0000000e -struct crypto_hash_walk { - const char *data; - - unsigned int offset; - unsigned int flags; - - struct page *pg; - unsigned int entrylen; - - unsigned int total; - struct scatterlist *sg; -}; - static int ahash_def_finup(struct ahash_request *req); static inline bool crypto_ahash_block_only(struct crypto_ahash *tfm) @@ -112,8 +99,8 @@ static int hash_walk_new_entry(struct crypto_hash_walk *walk) return hash_walk_next(walk); } -static int crypto_hash_walk_first(struct ahash_request *req, - struct crypto_hash_walk *walk) +int crypto_hash_walk_first(struct ahash_request *req, + struct crypto_hash_walk *walk) { walk->total = req->nbytes; walk->entrylen = 0; @@ -133,8 +120,9 @@ static int crypto_hash_walk_first(struct ahash_request *req, return hash_walk_new_entry(walk); } +EXPORT_SYMBOL_GPL(crypto_hash_walk_first); -static int crypto_hash_walk_done(struct crypto_hash_walk *walk, int err) +int crypto_hash_walk_done(struct crypto_hash_walk *walk, int err) { if ((walk->flags & CRYPTO_AHASH_REQ_VIRT)) return err; @@ -160,11 +148,7 @@ static int crypto_hash_walk_done(struct crypto_hash_walk *walk, int err) return hash_walk_new_entry(walk); } - -static inline int crypto_hash_walk_last(struct crypto_hash_walk *walk) -{ - return !(walk->entrylen | walk->total); -} +EXPORT_SYMBOL_GPL(crypto_hash_walk_done); /* * For an ahash tfm that is using an shash algorithm (instead of an ahash @@ -347,6 +331,12 @@ static int ahash_do_req_chain(struct ahash_request *req, if (crypto_ahash_statesize(tfm) > HASH_MAX_STATESIZE) return -ENOSYS; + if (!crypto_ahash_need_fallback(tfm)) + return -ENOSYS; + + if (crypto_hash_no_export_core(tfm)) + return -ENOSYS; + { u8 state[HASH_MAX_STATESIZE]; @@ -954,6 +944,10 @@ static int ahash_prepare_alg(struct ahash_alg *alg) base->cra_reqsize > MAX_SYNC_HASH_REQSIZE) return -EINVAL; + if (base->cra_flags & CRYPTO_ALG_NEED_FALLBACK && + base->cra_flags & CRYPTO_ALG_NO_FALLBACK) + return -EINVAL; + err = hash_prepare_alg(&alg->halg); if (err) return err; @@ -962,7 +956,8 @@ static int ahash_prepare_alg(struct ahash_alg *alg) base->cra_flags |= CRYPTO_ALG_TYPE_AHASH; if ((base->cra_flags ^ CRYPTO_ALG_REQ_VIRT) & - (CRYPTO_ALG_ASYNC | CRYPTO_ALG_REQ_VIRT)) + (CRYPTO_ALG_ASYNC | CRYPTO_ALG_REQ_VIRT) && + !(base->cra_flags & CRYPTO_ALG_NO_FALLBACK)) base->cra_flags |= CRYPTO_ALG_NEED_FALLBACK; if (!alg->setkey) diff --git a/crypto/cryptd.c b/crypto/cryptd.c index 5bb6f8d88cc2e..efff54e707cb5 100644 --- a/crypto/cryptd.c +++ b/crypto/cryptd.c @@ -34,6 +34,7 @@ MODULE_PARM_DESC(cryptd_max_cpu_qlen, "Set cryptd Max queue depth"); static struct workqueue_struct *cryptd_wq; struct cryptd_cpu_queue { + local_lock_t bh_lock; struct crypto_queue queue; struct work_struct work; }; @@ -110,6 +111,7 @@ static int cryptd_init_queue(struct cryptd_queue *queue, cpu_queue = per_cpu_ptr(queue->cpu_queue, cpu); crypto_init_queue(&cpu_queue->queue, max_cpu_qlen); INIT_WORK(&cpu_queue->work, cryptd_queue_worker); + local_lock_init(&cpu_queue->bh_lock); } pr_info("cryptd: max_cpu_qlen set to %d\n", max_cpu_qlen); return 0; @@ -135,6 +137,7 @@ static int cryptd_enqueue_request(struct cryptd_queue *queue, refcount_t *refcnt; local_bh_disable(); + local_lock_nested_bh(&queue->cpu_queue->bh_lock); cpu_queue = this_cpu_ptr(queue->cpu_queue); err = crypto_enqueue_request(&cpu_queue->queue, request); @@ -151,6 +154,7 @@ static int cryptd_enqueue_request(struct cryptd_queue *queue, refcount_inc(refcnt); out: + local_unlock_nested_bh(&queue->cpu_queue->bh_lock); local_bh_enable(); return err; @@ -169,8 +173,10 @@ static void cryptd_queue_worker(struct work_struct *work) * Only handle one request at a time to avoid hogging crypto workqueue. */ local_bh_disable(); + __local_lock_nested_bh(&cpu_queue->bh_lock); backlog = crypto_get_backlog(&cpu_queue->queue); req = crypto_dequeue_request(&cpu_queue->queue); + __local_unlock_nested_bh(&cpu_queue->bh_lock); local_bh_enable(); if (!req) diff --git a/crypto/crypto_engine.c b/crypto/crypto_engine.c index 445d3c113ee10..18e1689efe127 100644 --- a/crypto/crypto_engine.c +++ b/crypto/crypto_engine.c @@ -74,7 +74,6 @@ static void crypto_pump_requests(struct crypto_engine *engine, struct crypto_engine_alg *alg; struct crypto_engine_op *op; unsigned long flags; - bool was_busy = false; int ret; spin_lock_irqsave(&engine->queue_lock, flags); @@ -83,12 +82,6 @@ static void crypto_pump_requests(struct crypto_engine *engine, if (!engine->retry_support && engine->cur_req) goto out; - /* If another context is idling then defer */ - if (engine->idling) { - kthread_queue_work(engine->kworker, &engine->pump_requests); - goto out; - } - /* Check if the engine queue is idle */ if (!crypto_queue_len(&engine->queue) || !engine->running) { if (!engine->busy) @@ -102,15 +95,6 @@ static void crypto_pump_requests(struct crypto_engine *engine, } engine->busy = false; - engine->idling = true; - spin_unlock_irqrestore(&engine->queue_lock, flags); - - if (engine->unprepare_crypt_hardware && - engine->unprepare_crypt_hardware(engine)) - dev_err(engine->dev, "failed to unprepare crypt hardware\n"); - - spin_lock_irqsave(&engine->queue_lock, flags); - engine->idling = false; goto out; } @@ -129,22 +113,11 @@ start_request: if (!engine->retry_support) engine->cur_req = async_req; - if (engine->busy) - was_busy = true; - else + if (!engine->busy) engine->busy = true; spin_unlock_irqrestore(&engine->queue_lock, flags); - /* Until here we get the request need to be encrypted successfully */ - if (!was_busy && engine->prepare_crypt_hardware) { - ret = engine->prepare_crypt_hardware(engine); - if (ret) { - dev_err(engine->dev, "failed to prepare crypt hardware\n"); - goto req_err_1; - } - } - alg = container_of(async_req->tfm->__crt_alg, struct crypto_engine_alg, base); op = &alg->op; @@ -195,17 +168,6 @@ retry: out: spin_unlock_irqrestore(&engine->queue_lock, flags); - /* - * Batch requests is possible only if - * hardware can enqueue multiple requests - */ - if (engine->do_batch_requests) { - ret = engine->do_batch_requests(engine); - if (ret) - dev_err(engine->dev, "failed to do batch requests: %d\n", - ret); - } - return; } @@ -462,12 +424,6 @@ EXPORT_SYMBOL_GPL(crypto_engine_stop); * crypto-engine queue. * @dev: the device attached with one hardware engine * @retry_support: whether hardware has support for retry mechanism - * @cbk_do_batch: pointer to a callback function to be invoked when executing - * a batch of requests. - * This has the form: - * callback(struct crypto_engine *engine) - * where: - * engine: the crypto engine structure. * @rt: whether this queue is set to run as a realtime task * @qlen: maximum size of the crypto-engine queue * @@ -476,7 +432,6 @@ EXPORT_SYMBOL_GPL(crypto_engine_stop); */ struct crypto_engine *crypto_engine_alloc_init_and_set(struct device *dev, bool retry_support, - int (*cbk_do_batch)(struct crypto_engine *engine), bool rt, int qlen) { struct crypto_engine *engine; @@ -492,14 +447,8 @@ struct crypto_engine *crypto_engine_alloc_init_and_set(struct device *dev, engine->rt = rt; engine->running = false; engine->busy = false; - engine->idling = false; engine->retry_support = retry_support; engine->priv_data = dev; - /* - * Batch requests is possible only if - * hardware has support for retry mechanism. - */ - engine->do_batch_requests = retry_support ? cbk_do_batch : NULL; snprintf(engine->name, sizeof(engine->name), "%s-engine", dev_name(dev)); @@ -534,7 +483,7 @@ EXPORT_SYMBOL_GPL(crypto_engine_alloc_init_and_set); */ struct crypto_engine *crypto_engine_alloc_init(struct device *dev, bool rt) { - return crypto_engine_alloc_init_and_set(dev, false, NULL, rt, + return crypto_engine_alloc_init_and_set(dev, false, rt, CRYPTO_ENGINE_MAX_QLEN); } EXPORT_SYMBOL_GPL(crypto_engine_alloc_init); diff --git a/crypto/deflate.c b/crypto/deflate.c index fe8e4ad0fee10..21404515dc77e 100644 --- a/crypto/deflate.c +++ b/crypto/deflate.c @@ -48,9 +48,14 @@ static void *deflate_alloc_stream(void) return ctx; } +static void deflate_free_stream(void *ctx) +{ + kvfree(ctx); +} + static struct crypto_acomp_streams deflate_streams = { .alloc_ctx = deflate_alloc_stream, - .cfree_ctx = kvfree, + .free_ctx = deflate_free_stream, }; static int deflate_compress_one(struct acomp_req *req, diff --git a/crypto/jitterentropy-kcapi.c b/crypto/jitterentropy-kcapi.c index c24d4ff2b4a8b..1266eb790708b 100644 --- a/crypto/jitterentropy-kcapi.c +++ b/crypto/jitterentropy-kcapi.c @@ -144,7 +144,7 @@ int jent_hash_time(void *hash_state, __u64 time, u8 *addtl, * Inject the data from the previous loop into the pool. This data is * not considered to contain any entropy, but it stirs the pool a bit. */ - ret = crypto_shash_update(desc, intermediary, sizeof(intermediary)); + ret = crypto_shash_update(hash_state_desc, intermediary, sizeof(intermediary)); if (ret) goto err; @@ -157,11 +157,12 @@ int jent_hash_time(void *hash_state, __u64 time, u8 *addtl, * conditioning operation to have an identical amount of input data * according to section 3.1.5. */ - if (!stuck) { - ret = crypto_shash_update(hash_state_desc, (u8 *)&time, - sizeof(__u64)); + if (stuck) { + time = 0; } + ret = crypto_shash_update(hash_state_desc, (u8 *)&time, sizeof(__u64)); + err: shash_desc_zero(desc); memzero_explicit(intermediary, sizeof(intermediary)); diff --git a/crypto/jitterentropy.c b/crypto/jitterentropy.c index 3b390bd6c1196..3f93cdc9a7afb 100644 --- a/crypto/jitterentropy.c +++ b/crypto/jitterentropy.c @@ -145,6 +145,7 @@ struct rand_data { */ #define JENT_ENTROPY_SAFETY_FACTOR 64 +#include <linux/array_size.h> #include <linux/fips.h> #include <linux/minmax.h> #include "jitterentropy.h" @@ -178,7 +179,6 @@ static const unsigned int jent_apt_cutoff_lookup[15] = { static const unsigned int jent_apt_cutoff_permanent_lookup[15] = { 355, 447, 479, 494, 502, 507, 510, 512, 512, 512, 512, 512, 512, 512, 512 }; -#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0])) static void jent_apt_init(struct rand_data *ec, unsigned int osr) { diff --git a/crypto/krb5/selftest.c b/crypto/krb5/selftest.c index 2a81a6315a0d0..4519c572d37ef 100644 --- a/crypto/krb5/selftest.c +++ b/crypto/krb5/selftest.c @@ -152,6 +152,7 @@ static int krb5_test_one_prf(const struct krb5_prf_test *test) out: clear_buf(&result); + clear_buf(&prf); clear_buf(&octet); clear_buf(&key); return ret; diff --git a/crypto/pcrypt.c b/crypto/pcrypt.c index c33d29a523e02..c3a9d4f2995c7 100644 --- a/crypto/pcrypt.c +++ b/crypto/pcrypt.c @@ -178,7 +178,7 @@ static int pcrypt_aead_decrypt(struct aead_request *req) static int pcrypt_aead_init_tfm(struct crypto_aead *tfm) { - int cpu, cpu_index; + int cpu_index; struct aead_instance *inst = aead_alg_instance(tfm); struct pcrypt_instance_ctx *ictx = aead_instance_ctx(inst); struct pcrypt_aead_ctx *ctx = crypto_aead_ctx(tfm); @@ -187,10 +187,7 @@ static int pcrypt_aead_init_tfm(struct crypto_aead *tfm) cpu_index = (unsigned int)atomic_inc_return(&ictx->tfm_count) % cpumask_weight(cpu_online_mask); - ctx->cb_cpu = cpumask_first(cpu_online_mask); - for (cpu = 0; cpu < cpu_index; cpu++) - ctx->cb_cpu = cpumask_next(ctx->cb_cpu, cpu_online_mask); - + ctx->cb_cpu = cpumask_nth(cpu_index, cpu_online_mask); cipher = crypto_spawn_aead(&ictx->spawn); if (IS_ERR(cipher)) diff --git a/crypto/testmgr.c b/crypto/testmgr.c index d636e04b55d5c..ee33ba21ae2bc 100644 --- a/crypto/testmgr.c +++ b/crypto/testmgr.c @@ -4186,7 +4186,6 @@ static const struct alg_test_desc alg_test_descs[] = { .alg = "authenc(hmac(sha1),cbc(aes))", .generic_driver = "authenc(hmac-sha1-lib,cbc(aes-generic))", .test = alg_test_aead, - .fips_allowed = 1, .suite = { .aead = __VECS(hmac_sha1_aes_cbc_tv_temp) } @@ -4207,7 +4206,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha1),ctr(aes))", .test = alg_test_null, - .fips_allowed = 1, }, { .alg = "authenc(hmac(sha1),ecb(cipher_null))", .generic_driver = "authenc(hmac-sha1-lib,ecb-cipher_null)", @@ -4218,7 +4216,6 @@ static const struct alg_test_desc alg_test_descs[] = { }, { .alg = "authenc(hmac(sha1),rfc3686(ctr(aes)))", .test = alg_test_null, - .fips_allowed = 1, }, { .alg = "authenc(hmac(sha224),cbc(des))", .generic_driver = "authenc(hmac-sha224-lib,cbc(des-generic))", @@ -4712,6 +4709,7 @@ static const struct alg_test_desc alg_test_descs[] = { */ .alg = "drbg_nopr_hmac_sha384", .test = alg_test_null, + .fips_allowed = 1 }, { .alg = "drbg_nopr_hmac_sha512", .test = alg_test_drbg, @@ -4730,6 +4728,7 @@ static const struct alg_test_desc alg_test_descs[] = { /* covered by drbg_nopr_sha256 test */ .alg = "drbg_nopr_sha384", .test = alg_test_null, + .fips_allowed = 1 }, { .alg = "drbg_nopr_sha512", .fips_allowed = 1, @@ -4761,6 +4760,7 @@ static const struct alg_test_desc alg_test_descs[] = { /* covered by drbg_pr_hmac_sha256 test */ .alg = "drbg_pr_hmac_sha384", .test = alg_test_null, + .fips_allowed = 1 }, { .alg = "drbg_pr_hmac_sha512", .test = alg_test_null, @@ -4776,6 +4776,7 @@ static const struct alg_test_desc alg_test_descs[] = { /* covered by drbg_pr_sha256 test */ .alg = "drbg_pr_sha384", .test = alg_test_null, + .fips_allowed = 1 }, { .alg = "drbg_pr_sha512", .fips_allowed = 1, @@ -5077,7 +5078,6 @@ static const struct alg_test_desc alg_test_descs[] = { .alg = "hmac(sha1)", .generic_driver = "hmac-sha1-lib", .test = alg_test_hash, - .fips_allowed = 1, .suite = { .hash = __VECS(hmac_sha1_tv_template) } @@ -5291,6 +5291,36 @@ static const struct alg_test_desc alg_test_descs[] = { .cipher = __VECS(fcrypt_pcbc_tv_template) } }, { +#if IS_ENABLED(CONFIG_CRYPTO_PHMAC_S390) + .alg = "phmac(sha224)", + .test = alg_test_hash, + .fips_allowed = 1, + .suite = { + .hash = __VECS(hmac_sha224_tv_template) + } + }, { + .alg = "phmac(sha256)", + .test = alg_test_hash, + .fips_allowed = 1, + .suite = { + .hash = __VECS(hmac_sha256_tv_template) + } + }, { + .alg = "phmac(sha384)", + .test = alg_test_hash, + .fips_allowed = 1, + .suite = { + .hash = __VECS(hmac_sha384_tv_template) + } + }, { + .alg = "phmac(sha512)", + .test = alg_test_hash, + .fips_allowed = 1, + .suite = { + .hash = __VECS(hmac_sha512_tv_template) + } + }, { +#endif .alg = "pkcs1(rsa,none)", .test = alg_test_sig, .suite = { @@ -5418,7 +5448,6 @@ static const struct alg_test_desc alg_test_descs[] = { .alg = "sha1", .generic_driver = "sha1-lib", .test = alg_test_hash, - .fips_allowed = 1, .suite = { .hash = __VECS(sha1_tv_template) } diff --git a/crypto/zstd.c b/crypto/zstd.c index 7570e11b4ee60..c2a19cb0879d6 100644 --- a/crypto/zstd.c +++ b/crypto/zstd.c @@ -12,188 +12,304 @@ #include <linux/net.h> #include <linux/vmalloc.h> #include <linux/zstd.h> -#include <crypto/internal/scompress.h> +#include <crypto/internal/acompress.h> +#include <crypto/scatterwalk.h> -#define ZSTD_DEF_LEVEL 3 +#define ZSTD_DEF_LEVEL 3 +#define ZSTD_MAX_WINDOWLOG 18 +#define ZSTD_MAX_SIZE BIT(ZSTD_MAX_WINDOWLOG) struct zstd_ctx { zstd_cctx *cctx; zstd_dctx *dctx; - void *cwksp; - void *dwksp; + size_t wksp_size; + zstd_parameters params; + u8 wksp[] __aligned(8); }; -static zstd_parameters zstd_params(void) -{ - return zstd_get_params(ZSTD_DEF_LEVEL, 0); -} +static DEFINE_MUTEX(zstd_stream_lock); -static int zstd_comp_init(struct zstd_ctx *ctx) +static void *zstd_alloc_stream(void) { - int ret = 0; - const zstd_parameters params = zstd_params(); - const size_t wksp_size = zstd_cctx_workspace_bound(¶ms.cParams); + zstd_parameters params; + struct zstd_ctx *ctx; + size_t wksp_size; - ctx->cwksp = vzalloc(wksp_size); - if (!ctx->cwksp) { - ret = -ENOMEM; - goto out; - } + params = zstd_get_params(ZSTD_DEF_LEVEL, ZSTD_MAX_SIZE); - ctx->cctx = zstd_init_cctx(ctx->cwksp, wksp_size); - if (!ctx->cctx) { - ret = -EINVAL; - goto out_free; - } -out: - return ret; -out_free: - vfree(ctx->cwksp); - goto out; + wksp_size = max_t(size_t, + zstd_cstream_workspace_bound(¶ms.cParams), + zstd_dstream_workspace_bound(ZSTD_MAX_SIZE)); + if (!wksp_size) + return ERR_PTR(-EINVAL); + + ctx = kvmalloc(sizeof(*ctx) + wksp_size, GFP_KERNEL); + if (!ctx) + return ERR_PTR(-ENOMEM); + + ctx->params = params; + ctx->wksp_size = wksp_size; + + return ctx; } -static int zstd_decomp_init(struct zstd_ctx *ctx) +static void zstd_free_stream(void *ctx) +{ + kvfree(ctx); +} + +static struct crypto_acomp_streams zstd_streams = { + .alloc_ctx = zstd_alloc_stream, + .free_ctx = zstd_free_stream, +}; + +static int zstd_init(struct crypto_acomp *acomp_tfm) { int ret = 0; - const size_t wksp_size = zstd_dctx_workspace_bound(); - ctx->dwksp = vzalloc(wksp_size); - if (!ctx->dwksp) { - ret = -ENOMEM; - goto out; - } + mutex_lock(&zstd_stream_lock); + ret = crypto_acomp_alloc_streams(&zstd_streams); + mutex_unlock(&zstd_stream_lock); - ctx->dctx = zstd_init_dctx(ctx->dwksp, wksp_size); - if (!ctx->dctx) { - ret = -EINVAL; - goto out_free; - } -out: return ret; -out_free: - vfree(ctx->dwksp); - goto out; } -static void zstd_comp_exit(struct zstd_ctx *ctx) +static void zstd_exit(struct crypto_acomp *acomp_tfm) { - vfree(ctx->cwksp); - ctx->cwksp = NULL; - ctx->cctx = NULL; + crypto_acomp_free_streams(&zstd_streams); } -static void zstd_decomp_exit(struct zstd_ctx *ctx) +static int zstd_compress_one(struct acomp_req *req, struct zstd_ctx *ctx, + const void *src, void *dst, unsigned int *dlen) { - vfree(ctx->dwksp); - ctx->dwksp = NULL; - ctx->dctx = NULL; -} + unsigned int out_len; -static int __zstd_init(void *ctx) -{ - int ret; + ctx->cctx = zstd_init_cctx(ctx->wksp, ctx->wksp_size); + if (!ctx->cctx) + return -EINVAL; - ret = zstd_comp_init(ctx); - if (ret) - return ret; - ret = zstd_decomp_init(ctx); - if (ret) - zstd_comp_exit(ctx); - return ret; + out_len = zstd_compress_cctx(ctx->cctx, dst, req->dlen, src, req->slen, + &ctx->params); + if (zstd_is_error(out_len)) + return -EINVAL; + + *dlen = out_len; + + return 0; } -static void *zstd_alloc_ctx(void) +static int zstd_compress(struct acomp_req *req) { - int ret; + struct crypto_acomp_stream *s; + unsigned int pos, scur, dcur; + unsigned int total_out = 0; + bool data_available = true; + zstd_out_buffer outbuf; + struct acomp_walk walk; + zstd_in_buffer inbuf; struct zstd_ctx *ctx; + size_t pending_bytes; + size_t num_bytes; + int ret; - ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); - if (!ctx) - return ERR_PTR(-ENOMEM); + s = crypto_acomp_lock_stream_bh(&zstd_streams); + ctx = s->ctx; - ret = __zstd_init(ctx); - if (ret) { - kfree(ctx); - return ERR_PTR(ret); + ret = acomp_walk_virt(&walk, req, true); + if (ret) + goto out; + + ctx->cctx = zstd_init_cstream(&ctx->params, 0, ctx->wksp, ctx->wksp_size); + if (!ctx->cctx) { + ret = -EINVAL; + goto out; } - return ctx; -} + do { + dcur = acomp_walk_next_dst(&walk); + if (!dcur) { + ret = -ENOSPC; + goto out; + } -static void __zstd_exit(void *ctx) -{ - zstd_comp_exit(ctx); - zstd_decomp_exit(ctx); -} + outbuf.pos = 0; + outbuf.dst = (u8 *)walk.dst.virt.addr; + outbuf.size = dcur; -static void zstd_free_ctx(void *ctx) -{ - __zstd_exit(ctx); - kfree_sensitive(ctx); -} + do { + scur = acomp_walk_next_src(&walk); + if (dcur == req->dlen && scur == req->slen) { + ret = zstd_compress_one(req, ctx, walk.src.virt.addr, + walk.dst.virt.addr, &total_out); + acomp_walk_done_src(&walk, scur); + acomp_walk_done_dst(&walk, dcur); + goto out; + } -static int __zstd_compress(const u8 *src, unsigned int slen, - u8 *dst, unsigned int *dlen, void *ctx) -{ - size_t out_len; - struct zstd_ctx *zctx = ctx; - const zstd_parameters params = zstd_params(); + if (scur) { + inbuf.pos = 0; + inbuf.src = walk.src.virt.addr; + inbuf.size = scur; + } else { + data_available = false; + break; + } - out_len = zstd_compress_cctx(zctx->cctx, dst, *dlen, src, slen, ¶ms); - if (zstd_is_error(out_len)) - return -EINVAL; - *dlen = out_len; - return 0; -} + num_bytes = zstd_compress_stream(ctx->cctx, &outbuf, &inbuf); + if (ZSTD_isError(num_bytes)) { + ret = -EIO; + goto out; + } -static int zstd_scompress(struct crypto_scomp *tfm, const u8 *src, - unsigned int slen, u8 *dst, unsigned int *dlen, - void *ctx) -{ - return __zstd_compress(src, slen, dst, dlen, ctx); + pending_bytes = zstd_flush_stream(ctx->cctx, &outbuf); + if (ZSTD_isError(pending_bytes)) { + ret = -EIO; + goto out; + } + acomp_walk_done_src(&walk, inbuf.pos); + } while (dcur != outbuf.pos); + + total_out += outbuf.pos; + acomp_walk_done_dst(&walk, dcur); + } while (data_available); + + pos = outbuf.pos; + num_bytes = zstd_end_stream(ctx->cctx, &outbuf); + if (ZSTD_isError(num_bytes)) + ret = -EIO; + else + total_out += (outbuf.pos - pos); + +out: + if (ret) + req->dlen = 0; + else + req->dlen = total_out; + + crypto_acomp_unlock_stream_bh(s); + + return ret; } -static int __zstd_decompress(const u8 *src, unsigned int slen, - u8 *dst, unsigned int *dlen, void *ctx) +static int zstd_decompress_one(struct acomp_req *req, struct zstd_ctx *ctx, + const void *src, void *dst, unsigned int *dlen) { size_t out_len; - struct zstd_ctx *zctx = ctx; - out_len = zstd_decompress_dctx(zctx->dctx, dst, *dlen, src, slen); + ctx->dctx = zstd_init_dctx(ctx->wksp, ctx->wksp_size); + if (!ctx->dctx) + return -EINVAL; + + out_len = zstd_decompress_dctx(ctx->dctx, dst, req->dlen, src, req->slen); if (zstd_is_error(out_len)) return -EINVAL; + *dlen = out_len; + return 0; } -static int zstd_sdecompress(struct crypto_scomp *tfm, const u8 *src, - unsigned int slen, u8 *dst, unsigned int *dlen, - void *ctx) +static int zstd_decompress(struct acomp_req *req) { - return __zstd_decompress(src, slen, dst, dlen, ctx); -} + struct crypto_acomp_stream *s; + unsigned int total_out = 0; + unsigned int scur, dcur; + zstd_out_buffer outbuf; + struct acomp_walk walk; + zstd_in_buffer inbuf; + struct zstd_ctx *ctx; + size_t pending_bytes; + int ret; -static struct scomp_alg scomp = { - .alloc_ctx = zstd_alloc_ctx, - .free_ctx = zstd_free_ctx, - .compress = zstd_scompress, - .decompress = zstd_sdecompress, - .base = { - .cra_name = "zstd", - .cra_driver_name = "zstd-scomp", - .cra_module = THIS_MODULE, + s = crypto_acomp_lock_stream_bh(&zstd_streams); + ctx = s->ctx; + + ret = acomp_walk_virt(&walk, req, true); + if (ret) + goto out; + + ctx->dctx = zstd_init_dstream(ZSTD_MAX_SIZE, ctx->wksp, ctx->wksp_size); + if (!ctx->dctx) { + ret = -EINVAL; + goto out; } + + do { + scur = acomp_walk_next_src(&walk); + if (scur) { + inbuf.pos = 0; + inbuf.size = scur; + inbuf.src = walk.src.virt.addr; + } else { + break; + } + + do { + dcur = acomp_walk_next_dst(&walk); + if (dcur == req->dlen && scur == req->slen) { + ret = zstd_decompress_one(req, ctx, walk.src.virt.addr, + walk.dst.virt.addr, &total_out); + acomp_walk_done_dst(&walk, dcur); + acomp_walk_done_src(&walk, scur); + goto out; + } + + if (!dcur) { + ret = -ENOSPC; + goto out; + } + + outbuf.pos = 0; + outbuf.dst = (u8 *)walk.dst.virt.addr; + outbuf.size = dcur; + + pending_bytes = zstd_decompress_stream(ctx->dctx, &outbuf, &inbuf); + if (ZSTD_isError(pending_bytes)) { + ret = -EIO; + goto out; + } + + total_out += outbuf.pos; + + acomp_walk_done_dst(&walk, outbuf.pos); + } while (inbuf.pos != scur); + + acomp_walk_done_src(&walk, scur); + } while (ret == 0); + +out: + if (ret) + req->dlen = 0; + else + req->dlen = total_out; + + crypto_acomp_unlock_stream_bh(s); + + return ret; +} + +static struct acomp_alg zstd_acomp = { + .base = { + .cra_name = "zstd", + .cra_driver_name = "zstd-generic", + .cra_flags = CRYPTO_ALG_REQ_VIRT, + .cra_module = THIS_MODULE, + }, + .init = zstd_init, + .exit = zstd_exit, + .compress = zstd_compress, + .decompress = zstd_decompress, }; static int __init zstd_mod_init(void) { - return crypto_register_scomp(&scomp); + return crypto_register_acomp(&zstd_acomp); } static void __exit zstd_mod_fini(void) { - crypto_unregister_scomp(&scomp); + crypto_unregister_acomp(&zstd_acomp); } module_init(zstd_mod_init); diff --git a/drivers/char/hw_random/atmel-rng.c b/drivers/char/hw_random/atmel-rng.c index d2b00458761e1..6ed24be3481df 100644 --- a/drivers/char/hw_random/atmel-rng.c +++ b/drivers/char/hw_random/atmel-rng.c @@ -80,7 +80,6 @@ static int atmel_trng_read(struct hwrng *rng, void *buf, size_t max, ret = 4; out: - pm_runtime_mark_last_busy(trng->dev); pm_runtime_put_sync_autosuspend(trng->dev); return ret; } diff --git a/drivers/char/hw_random/cctrng.c b/drivers/char/hw_random/cctrng.c index 4db198849695d..a5be9258037f8 100644 --- a/drivers/char/hw_random/cctrng.c +++ b/drivers/char/hw_random/cctrng.c @@ -98,7 +98,6 @@ static void cc_trng_pm_put_suspend(struct device *dev) { int rc = 0; - pm_runtime_mark_last_busy(dev); rc = pm_runtime_put_autosuspend(dev); if (rc) dev_err(dev, "pm_runtime_put_autosuspend returned %x\n", rc); diff --git a/drivers/char/hw_random/mtk-rng.c b/drivers/char/hw_random/mtk-rng.c index b7fa1bc1122bd..5808d09d12c45 100644 --- a/drivers/char/hw_random/mtk-rng.c +++ b/drivers/char/hw_random/mtk-rng.c @@ -98,7 +98,6 @@ static int mtk_rng_read(struct hwrng *rng, void *buf, size_t max, bool wait) max -= sizeof(u32); } - pm_runtime_mark_last_busy(priv->dev); pm_runtime_put_sync_autosuspend(priv->dev); return retval || !wait ? retval : -EIO; @@ -143,7 +142,9 @@ static int mtk_rng_probe(struct platform_device *pdev) dev_set_drvdata(&pdev->dev, priv); pm_runtime_set_autosuspend_delay(&pdev->dev, RNG_AUTOSUSPEND_TIMEOUT); pm_runtime_use_autosuspend(&pdev->dev); - devm_pm_runtime_enable(&pdev->dev); + ret = devm_pm_runtime_enable(&pdev->dev); + if (ret) + return ret; dev_info(&pdev->dev, "registered RNG driver\n"); diff --git a/drivers/char/hw_random/npcm-rng.c b/drivers/char/hw_random/npcm-rng.c index 3e308c890bd25..40d6e29dea032 100644 --- a/drivers/char/hw_random/npcm-rng.c +++ b/drivers/char/hw_random/npcm-rng.c @@ -80,7 +80,6 @@ static int npcm_rng_read(struct hwrng *rng, void *buf, size_t max, bool wait) max--; } - pm_runtime_mark_last_busy(priv->dev); pm_runtime_put_sync_autosuspend(priv->dev); return retval || !wait ? retval : -EIO; diff --git a/drivers/char/hw_random/omap3-rom-rng.c b/drivers/char/hw_random/omap3-rom-rng.c index 8064c792caf0a..aa71f61c3dc9e 100644 --- a/drivers/char/hw_random/omap3-rom-rng.c +++ b/drivers/char/hw_random/omap3-rom-rng.c @@ -56,7 +56,6 @@ static int omap3_rom_rng_read(struct hwrng *rng, void *data, size_t max, bool w) else r = 4; - pm_runtime_mark_last_busy(ddata->dev); pm_runtime_put_autosuspend(ddata->dev); return r; diff --git a/drivers/char/hw_random/rockchip-rng.c b/drivers/char/hw_random/rockchip-rng.c index fb4a30b955075..6e3ed4b856051 100644 --- a/drivers/char/hw_random/rockchip-rng.c +++ b/drivers/char/hw_random/rockchip-rng.c @@ -223,7 +223,6 @@ static int rk3568_rng_read(struct hwrng *rng, void *buf, size_t max, bool wait) /* Read random data stored in the registers */ memcpy_fromio(buf, rk_rng->base + TRNG_RNG_DOUT, to_read); out: - pm_runtime_mark_last_busy(rk_rng->dev); pm_runtime_put_sync_autosuspend(rk_rng->dev); return (ret < 0) ? ret : to_read; @@ -263,7 +262,6 @@ static int rk3576_rng_read(struct hwrng *rng, void *buf, size_t max, bool wait) memcpy_fromio(buf, rk_rng->base + RKRNG_TRNG_DATA0, to_read); out: - pm_runtime_mark_last_busy(rk_rng->dev); pm_runtime_put_sync_autosuspend(rk_rng->dev); return (ret < 0) ? ret : to_read; @@ -355,7 +353,6 @@ out: /* close the TRNG */ rk_rng_writel(rk_rng, TRNG_V1_CTRL_NOP, TRNG_V1_CTRL); - pm_runtime_mark_last_busy(rk_rng->dev); pm_runtime_put_sync_autosuspend(rk_rng->dev); return (ret < 0) ? ret : to_read; diff --git a/drivers/char/hw_random/stm32-rng.c b/drivers/char/hw_random/stm32-rng.c index 98edbe796bc52..9a8c00586ab09 100644 --- a/drivers/char/hw_random/stm32-rng.c +++ b/drivers/char/hw_random/stm32-rng.c @@ -255,7 +255,6 @@ static int stm32_rng_read(struct hwrng *rng, void *data, size_t max, bool wait) } exit_rpm: - pm_runtime_mark_last_busy(priv->dev); pm_runtime_put_sync_autosuspend(priv->dev); return retval || !wait ? retval : -EIO; diff --git a/drivers/crypto/Kconfig b/drivers/crypto/Kconfig index 9f8a3a5bed7ec..04b4c43b6bae7 100644 --- a/drivers/crypto/Kconfig +++ b/drivers/crypto/Kconfig @@ -188,6 +188,19 @@ config CRYPTO_PAES_S390 Select this option if you want to use the paes cipher for example to use protected key encrypted devices. +config CRYPTO_PHMAC_S390 + tristate "PHMAC cipher algorithms" + depends on S390 + depends on PKEY + select CRYPTO_HASH + select CRYPTO_ENGINE + help + This is the s390 hardware accelerated implementation of the + protected key HMAC support for SHA224, SHA256, SHA384 and SHA512. + + Select this option if you want to use the phmac digests + for example to use dm-integrity with secure/protected keys. + config S390_PRNG tristate "Pseudo random number generator device driver" depends on S390 diff --git a/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c b/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c index f9cf00d690e2f..5663df49dd817 100644 --- a/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c +++ b/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-cipher.c @@ -206,15 +206,14 @@ static int sun8i_ce_cipher_prepare(struct crypto_engine *engine, void *async_req cet->t_key = desc_addr_val_le32(ce, rctx->addr_key); ivsize = crypto_skcipher_ivsize(tfm); - if (areq->iv && crypto_skcipher_ivsize(tfm) > 0) { - rctx->ivlen = ivsize; + if (areq->iv && ivsize > 0) { if (rctx->op_dir & CE_DECRYPTION) { offset = areq->cryptlen - ivsize; scatterwalk_map_and_copy(chan->backup_iv, areq->src, offset, ivsize, 0); } memcpy(chan->bounce_iv, areq->iv, ivsize); - rctx->addr_iv = dma_map_single(ce->dev, chan->bounce_iv, rctx->ivlen, + rctx->addr_iv = dma_map_single(ce->dev, chan->bounce_iv, ivsize, DMA_TO_DEVICE); if (dma_mapping_error(ce->dev, rctx->addr_iv)) { dev_err(ce->dev, "Cannot DMA MAP IV\n"); @@ -278,8 +277,8 @@ static int sun8i_ce_cipher_prepare(struct crypto_engine *engine, void *async_req } chan->timeout = areq->cryptlen; - rctx->nr_sgs = nr_sgs; - rctx->nr_sgd = nr_sgd; + rctx->nr_sgs = ns; + rctx->nr_sgd = nd; return 0; theend_sgs: @@ -296,7 +295,8 @@ theend_sgs: theend_iv: if (areq->iv && ivsize > 0) { if (!dma_mapping_error(ce->dev, rctx->addr_iv)) - dma_unmap_single(ce->dev, rctx->addr_iv, rctx->ivlen, DMA_TO_DEVICE); + dma_unmap_single(ce->dev, rctx->addr_iv, ivsize, + DMA_TO_DEVICE); offset = areq->cryptlen - ivsize; if (rctx->op_dir & CE_DECRYPTION) { @@ -345,7 +345,8 @@ static void sun8i_ce_cipher_unprepare(struct crypto_engine *engine, if (areq->iv && ivsize > 0) { if (cet->t_iv) - dma_unmap_single(ce->dev, rctx->addr_iv, rctx->ivlen, DMA_TO_DEVICE); + dma_unmap_single(ce->dev, rctx->addr_iv, ivsize, + DMA_TO_DEVICE); offset = areq->cryptlen - ivsize; if (rctx->op_dir & CE_DECRYPTION) { memcpy(areq->iv, chan->backup_iv, ivsize); diff --git a/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-hash.c b/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-hash.c index bef44f3501672..13bdfb8a2c627 100644 --- a/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-hash.c +++ b/drivers/crypto/allwinner/sun8i-ce/sun8i-ce-hash.c @@ -342,8 +342,8 @@ int sun8i_ce_hash_run(struct crypto_engine *engine, void *breq) algt = container_of(alg, struct sun8i_ce_alg_template, alg.hash.base); ce = algt->ce; - bs = algt->alg.hash.base.halg.base.cra_blocksize; - digestsize = algt->alg.hash.base.halg.digestsize; + bs = crypto_ahash_blocksize(tfm); + digestsize = crypto_ahash_digestsize(tfm); if (digestsize == SHA224_DIGEST_SIZE) digestsize = SHA256_DIGEST_SIZE; if (digestsize == SHA384_DIGEST_SIZE) @@ -455,7 +455,7 @@ int sun8i_ce_hash_run(struct crypto_engine *engine, void *breq) err_unmap_result: dma_unmap_single(ce->dev, addr_res, digestsize, DMA_FROM_DEVICE); if (!err) - memcpy(areq->result, result, algt->alg.hash.base.halg.digestsize); + memcpy(areq->result, result, crypto_ahash_digestsize(tfm)); err_unmap_src: dma_unmap_sg(ce->dev, areq->src, ns, DMA_TO_DEVICE); diff --git a/drivers/crypto/allwinner/sun8i-ce/sun8i-ce.h b/drivers/crypto/allwinner/sun8i-ce/sun8i-ce.h index 83df4d7190531..0f9a890670167 100644 --- a/drivers/crypto/allwinner/sun8i-ce/sun8i-ce.h +++ b/drivers/crypto/allwinner/sun8i-ce/sun8i-ce.h @@ -260,7 +260,6 @@ static inline __le32 desc_addr_val_le32(struct sun8i_ce_dev *dev, * struct sun8i_cipher_req_ctx - context for a skcipher request * @op_dir: direction (encrypt vs decrypt) for this request * @flow: the flow to use for this request - * @ivlen: size of bounce_iv * @nr_sgs: The number of source SG (as given by dma_map_sg()) * @nr_sgd: The number of destination SG (as given by dma_map_sg()) * @addr_iv: The IV addr returned by dma_map_single, need to unmap later @@ -270,7 +269,6 @@ static inline __le32 desc_addr_val_le32(struct sun8i_ce_dev *dev, struct sun8i_cipher_req_ctx { u32 op_dir; int flow; - unsigned int ivlen; int nr_sgs; int nr_sgd; dma_addr_t addr_iv; diff --git a/drivers/crypto/aspeed/aspeed-hace-hash.c b/drivers/crypto/aspeed/aspeed-hace-hash.c index 0b6e49c06eff4..f8f37c9d5f3c3 100644 --- a/drivers/crypto/aspeed/aspeed-hace-hash.c +++ b/drivers/crypto/aspeed/aspeed-hace-hash.c @@ -5,7 +5,6 @@ #include "aspeed-hace.h" #include <crypto/engine.h> -#include <crypto/hmac.h> #include <crypto/internal/hash.h> #include <crypto/scatterwalk.h> #include <crypto/sha1.h> @@ -14,6 +13,7 @@ #include <linux/err.h> #include <linux/io.h> #include <linux/kernel.h> +#include <linux/scatterlist.h> #include <linux/string.h> #ifdef CONFIG_CRYPTO_DEV_ASPEED_DEBUG @@ -59,6 +59,46 @@ static const __be64 sha512_iv[8] = { cpu_to_be64(SHA512_H6), cpu_to_be64(SHA512_H7) }; +static int aspeed_sham_init(struct ahash_request *req); +static int aspeed_ahash_req_update(struct aspeed_hace_dev *hace_dev); + +static int aspeed_sham_export(struct ahash_request *req, void *out) +{ + struct aspeed_sham_reqctx *rctx = ahash_request_ctx(req); + union { + u8 *u8; + u64 *u64; + } p = { .u8 = out }; + + memcpy(out, rctx->digest, rctx->ivsize); + p.u8 += rctx->ivsize; + put_unaligned(rctx->digcnt[0], p.u64++); + if (rctx->ivsize == 64) + put_unaligned(rctx->digcnt[1], p.u64); + return 0; +} + +static int aspeed_sham_import(struct ahash_request *req, const void *in) +{ + struct aspeed_sham_reqctx *rctx = ahash_request_ctx(req); + union { + const u8 *u8; + const u64 *u64; + } p = { .u8 = in }; + int err; + + err = aspeed_sham_init(req); + if (err) + return err; + + memcpy(rctx->digest, in, rctx->ivsize); + p.u8 += rctx->ivsize; + rctx->digcnt[0] = get_unaligned(p.u64++); + if (rctx->ivsize == 64) + rctx->digcnt[1] = get_unaligned(p.u64); + return 0; +} + /* The purpose of this padding is to ensure that the padded message is a * multiple of 512 bits (SHA1/SHA224/SHA256) or 1024 bits (SHA384/SHA512). * The bit "1" is appended at the end of the message followed by @@ -74,10 +114,10 @@ static const __be64 sha512_iv[8] = { * - if message length < 112 bytes then padlen = 112 - message length * - else padlen = 128 + 112 - message length */ -static void aspeed_ahash_fill_padding(struct aspeed_hace_dev *hace_dev, - struct aspeed_sham_reqctx *rctx) +static int aspeed_ahash_fill_padding(struct aspeed_hace_dev *hace_dev, + struct aspeed_sham_reqctx *rctx, u8 *buf) { - unsigned int index, padlen; + unsigned int index, padlen, bitslen; __be64 bits[2]; AHASH_DBG(hace_dev, "rctx flags:0x%x\n", (u32)rctx->flags); @@ -87,25 +127,32 @@ static void aspeed_ahash_fill_padding(struct aspeed_hace_dev *hace_dev, case SHA_FLAGS_SHA224: case SHA_FLAGS_SHA256: bits[0] = cpu_to_be64(rctx->digcnt[0] << 3); - index = rctx->bufcnt & 0x3f; + index = rctx->digcnt[0] & 0x3f; padlen = (index < 56) ? (56 - index) : ((64 + 56) - index); - *(rctx->buffer + rctx->bufcnt) = 0x80; - memset(rctx->buffer + rctx->bufcnt + 1, 0, padlen - 1); - memcpy(rctx->buffer + rctx->bufcnt + padlen, bits, 8); - rctx->bufcnt += padlen + 8; + bitslen = 8; break; default: bits[1] = cpu_to_be64(rctx->digcnt[0] << 3); bits[0] = cpu_to_be64(rctx->digcnt[1] << 3 | rctx->digcnt[0] >> 61); - index = rctx->bufcnt & 0x7f; + index = rctx->digcnt[0] & 0x7f; padlen = (index < 112) ? (112 - index) : ((128 + 112) - index); - *(rctx->buffer + rctx->bufcnt) = 0x80; - memset(rctx->buffer + rctx->bufcnt + 1, 0, padlen - 1); - memcpy(rctx->buffer + rctx->bufcnt + padlen, bits, 16); - rctx->bufcnt += padlen + 16; + bitslen = 16; break; } + buf[0] = 0x80; + memset(buf + 1, 0, padlen - 1); + memcpy(buf + padlen, bits, bitslen); + return padlen + bitslen; +} + +static void aspeed_ahash_update_counter(struct aspeed_sham_reqctx *rctx, + unsigned int len) +{ + rctx->offset += len; + rctx->digcnt[0] += len; + if (rctx->digcnt[0] < len) + rctx->digcnt[1]++; } /* @@ -117,31 +164,31 @@ static int aspeed_ahash_dma_prepare(struct aspeed_hace_dev *hace_dev) struct aspeed_engine_hash *hash_engine = &hace_dev->hash_engine; struct ahash_request *req = hash_engine->req; struct aspeed_sham_reqctx *rctx = ahash_request_ctx(req); - int length, remain; + unsigned int length, remain; + bool final = false; - length = rctx->total + rctx->bufcnt; - remain = length % rctx->block_size; + length = rctx->total - rctx->offset; + remain = length - round_down(length, rctx->block_size); AHASH_DBG(hace_dev, "length:0x%x, remain:0x%x\n", length, remain); - if (rctx->bufcnt) - memcpy(hash_engine->ahash_src_addr, rctx->buffer, rctx->bufcnt); - - if (rctx->total + rctx->bufcnt < ASPEED_CRYPTO_SRC_DMA_BUF_LEN) { - scatterwalk_map_and_copy(hash_engine->ahash_src_addr + - rctx->bufcnt, rctx->src_sg, - rctx->offset, rctx->total - remain, 0); - rctx->offset += rctx->total - remain; - - } else { - dev_warn(hace_dev->dev, "Hash data length is too large\n"); - return -EINVAL; - } - - scatterwalk_map_and_copy(rctx->buffer, rctx->src_sg, - rctx->offset, remain, 0); + if (length > ASPEED_HASH_SRC_DMA_BUF_LEN) + length = ASPEED_HASH_SRC_DMA_BUF_LEN; + else if (rctx->flags & SHA_FLAGS_FINUP) { + if (round_up(length, rctx->block_size) + rctx->block_size > + ASPEED_CRYPTO_SRC_DMA_BUF_LEN) + length = round_down(length - 1, rctx->block_size); + else + final = true; + } else + length -= remain; + scatterwalk_map_and_copy(hash_engine->ahash_src_addr, rctx->src_sg, + rctx->offset, length, 0); + aspeed_ahash_update_counter(rctx, length); + if (final) + length += aspeed_ahash_fill_padding( + hace_dev, rctx, hash_engine->ahash_src_addr + length); - rctx->bufcnt = remain; rctx->digest_dma_addr = dma_map_single(hace_dev->dev, rctx->digest, SHA512_DIGEST_SIZE, DMA_BIDIRECTIONAL); @@ -150,7 +197,7 @@ static int aspeed_ahash_dma_prepare(struct aspeed_hace_dev *hace_dev) return -ENOMEM; } - hash_engine->src_length = length - remain; + hash_engine->src_length = length; hash_engine->src_dma = hash_engine->ahash_src_dma_addr; hash_engine->digest_dma = rctx->digest_dma_addr; @@ -166,16 +213,20 @@ static int aspeed_ahash_dma_prepare_sg(struct aspeed_hace_dev *hace_dev) struct aspeed_engine_hash *hash_engine = &hace_dev->hash_engine; struct ahash_request *req = hash_engine->req; struct aspeed_sham_reqctx *rctx = ahash_request_ctx(req); + bool final = rctx->flags & SHA_FLAGS_FINUP; + int remain, sg_len, i, max_sg_nents; + unsigned int length, offset, total; struct aspeed_sg_list *src_list; struct scatterlist *s; - int length, remain, sg_len, i; int rc = 0; - remain = (rctx->total + rctx->bufcnt) % rctx->block_size; - length = rctx->total + rctx->bufcnt - remain; + offset = rctx->offset; + length = rctx->total - offset; + remain = final ? 0 : length - round_down(length, rctx->block_size); + length -= remain; - AHASH_DBG(hace_dev, "%s:0x%x, %s:%zu, %s:0x%x, %s:0x%x\n", - "rctx total", rctx->total, "bufcnt", rctx->bufcnt, + AHASH_DBG(hace_dev, "%s:0x%x, %s:0x%x, %s:0x%x\n", + "rctx total", rctx->total, "length", length, "remain", remain); sg_len = dma_map_sg(hace_dev->dev, rctx->src_sg, rctx->src_nents, @@ -186,6 +237,8 @@ static int aspeed_ahash_dma_prepare_sg(struct aspeed_hace_dev *hace_dev) goto end; } + max_sg_nents = ASPEED_HASH_SRC_DMA_BUF_LEN / sizeof(*src_list) - final; + sg_len = min(sg_len, max_sg_nents); src_list = (struct aspeed_sg_list *)hash_engine->ahash_src_addr; rctx->digest_dma_addr = dma_map_single(hace_dev->dev, rctx->digest, SHA512_DIGEST_SIZE, @@ -196,68 +249,66 @@ static int aspeed_ahash_dma_prepare_sg(struct aspeed_hace_dev *hace_dev) goto free_src_sg; } - if (rctx->bufcnt != 0) { - u32 phy_addr; - u32 len; + total = 0; + for_each_sg(rctx->src_sg, s, sg_len, i) { + u32 phy_addr = sg_dma_address(s); + u32 len = sg_dma_len(s); - rctx->buffer_dma_addr = dma_map_single(hace_dev->dev, - rctx->buffer, - rctx->block_size * 2, - DMA_TO_DEVICE); - if (dma_mapping_error(hace_dev->dev, rctx->buffer_dma_addr)) { - dev_warn(hace_dev->dev, "dma_map() rctx buffer error\n"); - rc = -ENOMEM; - goto free_rctx_digest; + if (len <= offset) { + offset -= len; + continue; } - phy_addr = rctx->buffer_dma_addr; - len = rctx->bufcnt; - length -= len; + len -= offset; + phy_addr += offset; + offset = 0; - /* Last sg list */ - if (length == 0) - len |= HASH_SG_LAST_LIST; + if (length > len) + length -= len; + else { + /* Last sg list */ + len = length; + length = 0; + } - src_list[0].phy_addr = cpu_to_le32(phy_addr); - src_list[0].len = cpu_to_le32(len); - src_list++; + total += len; + src_list[i].phy_addr = cpu_to_le32(phy_addr); + src_list[i].len = cpu_to_le32(len); } if (length != 0) { - for_each_sg(rctx->src_sg, s, sg_len, i) { - u32 phy_addr = sg_dma_address(s); - u32 len = sg_dma_len(s); - - if (length > len) - length -= len; - else { - /* Last sg list */ - len = length; - len |= HASH_SG_LAST_LIST; - length = 0; - } + total = round_down(total, rctx->block_size); + final = false; + } + + aspeed_ahash_update_counter(rctx, total); + if (final) { + int len = aspeed_ahash_fill_padding(hace_dev, rctx, + rctx->buffer); - src_list[i].phy_addr = cpu_to_le32(phy_addr); - src_list[i].len = cpu_to_le32(len); + total += len; + rctx->buffer_dma_addr = dma_map_single(hace_dev->dev, + rctx->buffer, + sizeof(rctx->buffer), + DMA_TO_DEVICE); + if (dma_mapping_error(hace_dev->dev, rctx->buffer_dma_addr)) { + dev_warn(hace_dev->dev, "dma_map() rctx buffer error\n"); + rc = -ENOMEM; + goto free_rctx_digest; } - } - if (length != 0) { - rc = -EINVAL; - goto free_rctx_buffer; + src_list[i].phy_addr = cpu_to_le32(rctx->buffer_dma_addr); + src_list[i].len = cpu_to_le32(len); + i++; } + src_list[i - 1].len |= cpu_to_le32(HASH_SG_LAST_LIST); - rctx->offset = rctx->total - remain; - hash_engine->src_length = rctx->total + rctx->bufcnt - remain; + hash_engine->src_length = total; hash_engine->src_dma = hash_engine->ahash_src_dma_addr; hash_engine->digest_dma = rctx->digest_dma_addr; return 0; -free_rctx_buffer: - if (rctx->bufcnt != 0) - dma_unmap_single(hace_dev->dev, rctx->buffer_dma_addr, - rctx->block_size * 2, DMA_TO_DEVICE); free_rctx_digest: dma_unmap_single(hace_dev->dev, rctx->digest_dma_addr, SHA512_DIGEST_SIZE, DMA_BIDIRECTIONAL); @@ -272,24 +323,6 @@ static int aspeed_ahash_complete(struct aspeed_hace_dev *hace_dev) { struct aspeed_engine_hash *hash_engine = &hace_dev->hash_engine; struct ahash_request *req = hash_engine->req; - - AHASH_DBG(hace_dev, "\n"); - - hash_engine->flags &= ~CRYPTO_FLAGS_BUSY; - - crypto_finalize_hash_request(hace_dev->crypt_engine_hash, req, 0); - - return 0; -} - -/* - * Copy digest to the corresponding request result. - * This function will be called at final() stage. - */ -static int aspeed_ahash_transfer(struct aspeed_hace_dev *hace_dev) -{ - struct aspeed_engine_hash *hash_engine = &hace_dev->hash_engine; - struct ahash_request *req = hash_engine->req; struct aspeed_sham_reqctx *rctx = ahash_request_ctx(req); AHASH_DBG(hace_dev, "\n"); @@ -297,12 +330,19 @@ static int aspeed_ahash_transfer(struct aspeed_hace_dev *hace_dev) dma_unmap_single(hace_dev->dev, rctx->digest_dma_addr, SHA512_DIGEST_SIZE, DMA_BIDIRECTIONAL); - dma_unmap_single(hace_dev->dev, rctx->buffer_dma_addr, - rctx->block_size * 2, DMA_TO_DEVICE); + if (rctx->total - rctx->offset >= rctx->block_size || + (rctx->total != rctx->offset && rctx->flags & SHA_FLAGS_FINUP)) + return aspeed_ahash_req_update(hace_dev); - memcpy(req->result, rctx->digest, rctx->digsize); + hash_engine->flags &= ~CRYPTO_FLAGS_BUSY; - return aspeed_ahash_complete(hace_dev); + if (rctx->flags & SHA_FLAGS_FINUP) + memcpy(req->result, rctx->digest, rctx->digsize); + + crypto_finalize_hash_request(hace_dev->crypt_engine_hash, req, + rctx->total - rctx->offset); + + return 0; } /* @@ -338,118 +378,6 @@ static int aspeed_hace_ahash_trigger(struct aspeed_hace_dev *hace_dev, return -EINPROGRESS; } -/* - * HMAC resume aims to do the second pass produces - * the final HMAC code derived from the inner hash - * result and the outer key. - */ -static int aspeed_ahash_hmac_resume(struct aspeed_hace_dev *hace_dev) -{ - struct aspeed_engine_hash *hash_engine = &hace_dev->hash_engine; - struct ahash_request *req = hash_engine->req; - struct aspeed_sham_reqctx *rctx = ahash_request_ctx(req); - struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); - struct aspeed_sham_ctx *tctx = crypto_ahash_ctx(tfm); - struct aspeed_sha_hmac_ctx *bctx = tctx->base; - int rc = 0; - - AHASH_DBG(hace_dev, "\n"); - - dma_unmap_single(hace_dev->dev, rctx->digest_dma_addr, - SHA512_DIGEST_SIZE, DMA_BIDIRECTIONAL); - - dma_unmap_single(hace_dev->dev, rctx->buffer_dma_addr, - rctx->block_size * 2, DMA_TO_DEVICE); - - /* o key pad + hash sum 1 */ - memcpy(rctx->buffer, bctx->opad, rctx->block_size); - memcpy(rctx->buffer + rctx->block_size, rctx->digest, rctx->digsize); - - rctx->bufcnt = rctx->block_size + rctx->digsize; - rctx->digcnt[0] = rctx->block_size + rctx->digsize; - - aspeed_ahash_fill_padding(hace_dev, rctx); - memcpy(rctx->digest, rctx->sha_iv, rctx->ivsize); - - rctx->digest_dma_addr = dma_map_single(hace_dev->dev, rctx->digest, - SHA512_DIGEST_SIZE, - DMA_BIDIRECTIONAL); - if (dma_mapping_error(hace_dev->dev, rctx->digest_dma_addr)) { - dev_warn(hace_dev->dev, "dma_map() rctx digest error\n"); - rc = -ENOMEM; - goto end; - } - - rctx->buffer_dma_addr = dma_map_single(hace_dev->dev, rctx->buffer, - rctx->block_size * 2, - DMA_TO_DEVICE); - if (dma_mapping_error(hace_dev->dev, rctx->buffer_dma_addr)) { - dev_warn(hace_dev->dev, "dma_map() rctx buffer error\n"); - rc = -ENOMEM; - goto free_rctx_digest; - } - - hash_engine->src_dma = rctx->buffer_dma_addr; - hash_engine->src_length = rctx->bufcnt; - hash_engine->digest_dma = rctx->digest_dma_addr; - - return aspeed_hace_ahash_trigger(hace_dev, aspeed_ahash_transfer); - -free_rctx_digest: - dma_unmap_single(hace_dev->dev, rctx->digest_dma_addr, - SHA512_DIGEST_SIZE, DMA_BIDIRECTIONAL); -end: - return rc; -} - -static int aspeed_ahash_req_final(struct aspeed_hace_dev *hace_dev) -{ - struct aspeed_engine_hash *hash_engine = &hace_dev->hash_engine; - struct ahash_request *req = hash_engine->req; - struct aspeed_sham_reqctx *rctx = ahash_request_ctx(req); - int rc = 0; - - AHASH_DBG(hace_dev, "\n"); - - aspeed_ahash_fill_padding(hace_dev, rctx); - - rctx->digest_dma_addr = dma_map_single(hace_dev->dev, - rctx->digest, - SHA512_DIGEST_SIZE, - DMA_BIDIRECTIONAL); - if (dma_mapping_error(hace_dev->dev, rctx->digest_dma_addr)) { - dev_warn(hace_dev->dev, "dma_map() rctx digest error\n"); - rc = -ENOMEM; - goto end; - } - - rctx->buffer_dma_addr = dma_map_single(hace_dev->dev, - rctx->buffer, - rctx->block_size * 2, - DMA_TO_DEVICE); - if (dma_mapping_error(hace_dev->dev, rctx->buffer_dma_addr)) { - dev_warn(hace_dev->dev, "dma_map() rctx buffer error\n"); - rc = -ENOMEM; - goto free_rctx_digest; - } - - hash_engine->src_dma = rctx->buffer_dma_addr; - hash_engine->src_length = rctx->bufcnt; - hash_engine->digest_dma = rctx->digest_dma_addr; - - if (rctx->flags & SHA_FLAGS_HMAC) - return aspeed_hace_ahash_trigger(hace_dev, - aspeed_ahash_hmac_resume); - - return aspeed_hace_ahash_trigger(hace_dev, aspeed_ahash_transfer); - -free_rctx_digest: - dma_unmap_single(hace_dev->dev, rctx->digest_dma_addr, - SHA512_DIGEST_SIZE, DMA_BIDIRECTIONAL); -end: - return rc; -} - static int aspeed_ahash_update_resume_sg(struct aspeed_hace_dev *hace_dev) { struct aspeed_engine_hash *hash_engine = &hace_dev->hash_engine; @@ -461,40 +389,12 @@ static int aspeed_ahash_update_resume_sg(struct aspeed_hace_dev *hace_dev) dma_unmap_sg(hace_dev->dev, rctx->src_sg, rctx->src_nents, DMA_TO_DEVICE); - if (rctx->bufcnt != 0) + if (rctx->flags & SHA_FLAGS_FINUP && rctx->total == rctx->offset) dma_unmap_single(hace_dev->dev, rctx->buffer_dma_addr, - rctx->block_size * 2, - DMA_TO_DEVICE); + sizeof(rctx->buffer), DMA_TO_DEVICE); - dma_unmap_single(hace_dev->dev, rctx->digest_dma_addr, - SHA512_DIGEST_SIZE, DMA_BIDIRECTIONAL); - - scatterwalk_map_and_copy(rctx->buffer, rctx->src_sg, rctx->offset, - rctx->total - rctx->offset, 0); - - rctx->bufcnt = rctx->total - rctx->offset; rctx->cmd &= ~HASH_CMD_HASH_SRC_SG_CTRL; - if (rctx->flags & SHA_FLAGS_FINUP) - return aspeed_ahash_req_final(hace_dev); - - return aspeed_ahash_complete(hace_dev); -} - -static int aspeed_ahash_update_resume(struct aspeed_hace_dev *hace_dev) -{ - struct aspeed_engine_hash *hash_engine = &hace_dev->hash_engine; - struct ahash_request *req = hash_engine->req; - struct aspeed_sham_reqctx *rctx = ahash_request_ctx(req); - - AHASH_DBG(hace_dev, "\n"); - - dma_unmap_single(hace_dev->dev, rctx->digest_dma_addr, - SHA512_DIGEST_SIZE, DMA_BIDIRECTIONAL); - - if (rctx->flags & SHA_FLAGS_FINUP) - return aspeed_ahash_req_final(hace_dev); - return aspeed_ahash_complete(hace_dev); } @@ -513,7 +413,7 @@ static int aspeed_ahash_req_update(struct aspeed_hace_dev *hace_dev) resume = aspeed_ahash_update_resume_sg; } else { - resume = aspeed_ahash_update_resume; + resume = aspeed_ahash_complete; } ret = hash_engine->dma_prepare(hace_dev); @@ -530,26 +430,47 @@ static int aspeed_hace_hash_handle_queue(struct aspeed_hace_dev *hace_dev, hace_dev->crypt_engine_hash, req); } +static noinline int aspeed_ahash_fallback(struct ahash_request *req) +{ + struct aspeed_sham_reqctx *rctx = ahash_request_ctx(req); + HASH_FBREQ_ON_STACK(fbreq, req); + u8 *state = rctx->buffer; + struct scatterlist sg[2]; + struct scatterlist *ssg; + int ret; + + ssg = scatterwalk_ffwd(sg, req->src, rctx->offset); + ahash_request_set_crypt(fbreq, ssg, req->result, + rctx->total - rctx->offset); + + ret = aspeed_sham_export(req, state) ?: + crypto_ahash_import_core(fbreq, state); + + if (rctx->flags & SHA_FLAGS_FINUP) + ret = ret ?: crypto_ahash_finup(fbreq); + else + ret = ret ?: crypto_ahash_update(fbreq) ?: + crypto_ahash_export_core(fbreq, state) ?: + aspeed_sham_import(req, state); + HASH_REQUEST_ZERO(fbreq); + return ret; +} + static int aspeed_ahash_do_request(struct crypto_engine *engine, void *areq) { struct ahash_request *req = ahash_request_cast(areq); - struct aspeed_sham_reqctx *rctx = ahash_request_ctx(req); struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); struct aspeed_sham_ctx *tctx = crypto_ahash_ctx(tfm); struct aspeed_hace_dev *hace_dev = tctx->hace_dev; struct aspeed_engine_hash *hash_engine; - int ret = 0; + int ret; hash_engine = &hace_dev->hash_engine; hash_engine->flags |= CRYPTO_FLAGS_BUSY; - if (rctx->op == SHA_OP_UPDATE) - ret = aspeed_ahash_req_update(hace_dev); - else if (rctx->op == SHA_OP_FINAL) - ret = aspeed_ahash_req_final(hace_dev); - + ret = aspeed_ahash_req_update(hace_dev); if (ret != -EINPROGRESS) - return ret; + return aspeed_ahash_fallback(req); return 0; } @@ -590,45 +511,7 @@ static int aspeed_sham_update(struct ahash_request *req) rctx->total = req->nbytes; rctx->src_sg = req->src; rctx->offset = 0; - rctx->src_nents = sg_nents(req->src); - rctx->op = SHA_OP_UPDATE; - - rctx->digcnt[0] += rctx->total; - if (rctx->digcnt[0] < rctx->total) - rctx->digcnt[1]++; - - if (rctx->bufcnt + rctx->total < rctx->block_size) { - scatterwalk_map_and_copy(rctx->buffer + rctx->bufcnt, - rctx->src_sg, rctx->offset, - rctx->total, 0); - rctx->bufcnt += rctx->total; - - return 0; - } - - return aspeed_hace_hash_handle_queue(hace_dev, req); -} - -static int aspeed_sham_shash_digest(struct crypto_shash *tfm, u32 flags, - const u8 *data, unsigned int len, u8 *out) -{ - SHASH_DESC_ON_STACK(shash, tfm); - - shash->tfm = tfm; - - return crypto_shash_digest(shash, data, len, out); -} - -static int aspeed_sham_final(struct ahash_request *req) -{ - struct aspeed_sham_reqctx *rctx = ahash_request_ctx(req); - struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); - struct aspeed_sham_ctx *tctx = crypto_ahash_ctx(tfm); - struct aspeed_hace_dev *hace_dev = tctx->hace_dev; - - AHASH_DBG(hace_dev, "req->nbytes:%d, rctx->total:%d\n", - req->nbytes, rctx->total); - rctx->op = SHA_OP_FINAL; + rctx->src_nents = sg_nents_for_len(req->src, req->nbytes); return aspeed_hace_hash_handle_queue(hace_dev, req); } @@ -639,23 +522,12 @@ static int aspeed_sham_finup(struct ahash_request *req) struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); struct aspeed_sham_ctx *tctx = crypto_ahash_ctx(tfm); struct aspeed_hace_dev *hace_dev = tctx->hace_dev; - int rc1, rc2; AHASH_DBG(hace_dev, "req->nbytes: %d\n", req->nbytes); rctx->flags |= SHA_FLAGS_FINUP; - rc1 = aspeed_sham_update(req); - if (rc1 == -EINPROGRESS || rc1 == -EBUSY) - return rc1; - - /* - * final() has to be always called to cleanup resources - * even if update() failed, except EINPROGRESS - */ - rc2 = aspeed_sham_final(req); - - return rc1 ? : rc2; + return aspeed_sham_update(req); } static int aspeed_sham_init(struct ahash_request *req) @@ -664,7 +536,6 @@ static int aspeed_sham_init(struct ahash_request *req) struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); struct aspeed_sham_ctx *tctx = crypto_ahash_ctx(tfm); struct aspeed_hace_dev *hace_dev = tctx->hace_dev; - struct aspeed_sha_hmac_ctx *bctx = tctx->base; AHASH_DBG(hace_dev, "%s: digest size:%d\n", crypto_tfm_alg_name(&tfm->base), @@ -679,7 +550,6 @@ static int aspeed_sham_init(struct ahash_request *req) rctx->flags |= SHA_FLAGS_SHA1; rctx->digsize = SHA1_DIGEST_SIZE; rctx->block_size = SHA1_BLOCK_SIZE; - rctx->sha_iv = sha1_iv; rctx->ivsize = 32; memcpy(rctx->digest, sha1_iv, rctx->ivsize); break; @@ -688,7 +558,6 @@ static int aspeed_sham_init(struct ahash_request *req) rctx->flags |= SHA_FLAGS_SHA224; rctx->digsize = SHA224_DIGEST_SIZE; rctx->block_size = SHA224_BLOCK_SIZE; - rctx->sha_iv = sha224_iv; rctx->ivsize = 32; memcpy(rctx->digest, sha224_iv, rctx->ivsize); break; @@ -697,7 +566,6 @@ static int aspeed_sham_init(struct ahash_request *req) rctx->flags |= SHA_FLAGS_SHA256; rctx->digsize = SHA256_DIGEST_SIZE; rctx->block_size = SHA256_BLOCK_SIZE; - rctx->sha_iv = sha256_iv; rctx->ivsize = 32; memcpy(rctx->digest, sha256_iv, rctx->ivsize); break; @@ -707,7 +575,6 @@ static int aspeed_sham_init(struct ahash_request *req) rctx->flags |= SHA_FLAGS_SHA384; rctx->digsize = SHA384_DIGEST_SIZE; rctx->block_size = SHA384_BLOCK_SIZE; - rctx->sha_iv = (const __be32 *)sha384_iv; rctx->ivsize = 64; memcpy(rctx->digest, sha384_iv, rctx->ivsize); break; @@ -717,7 +584,6 @@ static int aspeed_sham_init(struct ahash_request *req) rctx->flags |= SHA_FLAGS_SHA512; rctx->digsize = SHA512_DIGEST_SIZE; rctx->block_size = SHA512_BLOCK_SIZE; - rctx->sha_iv = (const __be32 *)sha512_iv; rctx->ivsize = 64; memcpy(rctx->digest, sha512_iv, rctx->ivsize); break; @@ -727,19 +593,10 @@ static int aspeed_sham_init(struct ahash_request *req) return -EINVAL; } - rctx->bufcnt = 0; rctx->total = 0; rctx->digcnt[0] = 0; rctx->digcnt[1] = 0; - /* HMAC init */ - if (tctx->flags & SHA_FLAGS_HMAC) { - rctx->digcnt[0] = rctx->block_size; - rctx->bufcnt = rctx->block_size; - memcpy(rctx->buffer, bctx->ipad, rctx->block_size); - rctx->flags |= SHA_FLAGS_HMAC; - } - return 0; } @@ -748,102 +605,14 @@ static int aspeed_sham_digest(struct ahash_request *req) return aspeed_sham_init(req) ? : aspeed_sham_finup(req); } -static int aspeed_sham_setkey(struct crypto_ahash *tfm, const u8 *key, - unsigned int keylen) +static int aspeed_sham_cra_init(struct crypto_ahash *tfm) { + struct ahash_alg *alg = crypto_ahash_alg(tfm); struct aspeed_sham_ctx *tctx = crypto_ahash_ctx(tfm); - struct aspeed_hace_dev *hace_dev = tctx->hace_dev; - struct aspeed_sha_hmac_ctx *bctx = tctx->base; - int ds = crypto_shash_digestsize(bctx->shash); - int bs = crypto_shash_blocksize(bctx->shash); - int err = 0; - int i; - - AHASH_DBG(hace_dev, "%s: keylen:%d\n", crypto_tfm_alg_name(&tfm->base), - keylen); - - if (keylen > bs) { - err = aspeed_sham_shash_digest(bctx->shash, - crypto_shash_get_flags(bctx->shash), - key, keylen, bctx->ipad); - if (err) - return err; - keylen = ds; - - } else { - memcpy(bctx->ipad, key, keylen); - } - - memset(bctx->ipad + keylen, 0, bs - keylen); - memcpy(bctx->opad, bctx->ipad, bs); - - for (i = 0; i < bs; i++) { - bctx->ipad[i] ^= HMAC_IPAD_VALUE; - bctx->opad[i] ^= HMAC_OPAD_VALUE; - } - - return err; -} - -static int aspeed_sham_cra_init(struct crypto_tfm *tfm) -{ - struct ahash_alg *alg = __crypto_ahash_alg(tfm->__crt_alg); - struct aspeed_sham_ctx *tctx = crypto_tfm_ctx(tfm); struct aspeed_hace_alg *ast_alg; ast_alg = container_of(alg, struct aspeed_hace_alg, alg.ahash.base); tctx->hace_dev = ast_alg->hace_dev; - tctx->flags = 0; - - crypto_ahash_set_reqsize(__crypto_ahash_cast(tfm), - sizeof(struct aspeed_sham_reqctx)); - - if (ast_alg->alg_base) { - /* hmac related */ - struct aspeed_sha_hmac_ctx *bctx = tctx->base; - - tctx->flags |= SHA_FLAGS_HMAC; - bctx->shash = crypto_alloc_shash(ast_alg->alg_base, 0, - CRYPTO_ALG_NEED_FALLBACK); - if (IS_ERR(bctx->shash)) { - dev_warn(ast_alg->hace_dev->dev, - "base driver '%s' could not be loaded.\n", - ast_alg->alg_base); - return PTR_ERR(bctx->shash); - } - } - - return 0; -} - -static void aspeed_sham_cra_exit(struct crypto_tfm *tfm) -{ - struct aspeed_sham_ctx *tctx = crypto_tfm_ctx(tfm); - struct aspeed_hace_dev *hace_dev = tctx->hace_dev; - - AHASH_DBG(hace_dev, "%s\n", crypto_tfm_alg_name(tfm)); - - if (tctx->flags & SHA_FLAGS_HMAC) { - struct aspeed_sha_hmac_ctx *bctx = tctx->base; - - crypto_free_shash(bctx->shash); - } -} - -static int aspeed_sham_export(struct ahash_request *req, void *out) -{ - struct aspeed_sham_reqctx *rctx = ahash_request_ctx(req); - - memcpy(out, rctx, sizeof(*rctx)); - - return 0; -} - -static int aspeed_sham_import(struct ahash_request *req, const void *in) -{ - struct aspeed_sham_reqctx *rctx = ahash_request_ctx(req); - - memcpy(rctx, in, sizeof(*rctx)); return 0; } @@ -853,11 +622,11 @@ static struct aspeed_hace_alg aspeed_ahash_algs[] = { .alg.ahash.base = { .init = aspeed_sham_init, .update = aspeed_sham_update, - .final = aspeed_sham_final, .finup = aspeed_sham_finup, .digest = aspeed_sham_digest, .export = aspeed_sham_export, .import = aspeed_sham_import, + .init_tfm = aspeed_sham_cra_init, .halg = { .digestsize = SHA1_DIGEST_SIZE, .statesize = sizeof(struct aspeed_sham_reqctx), @@ -867,13 +636,13 @@ static struct aspeed_hace_alg aspeed_ahash_algs[] = { .cra_priority = 300, .cra_flags = CRYPTO_ALG_TYPE_AHASH | CRYPTO_ALG_ASYNC | + CRYPTO_AHASH_ALG_BLOCK_ONLY | CRYPTO_ALG_KERN_DRIVER_ONLY, .cra_blocksize = SHA1_BLOCK_SIZE, .cra_ctxsize = sizeof(struct aspeed_sham_ctx), + .cra_reqsize = sizeof(struct aspeed_sham_reqctx), .cra_alignmask = 0, .cra_module = THIS_MODULE, - .cra_init = aspeed_sham_cra_init, - .cra_exit = aspeed_sham_cra_exit, } } }, @@ -885,11 +654,11 @@ static struct aspeed_hace_alg aspeed_ahash_algs[] = { .alg.ahash.base = { .init = aspeed_sham_init, .update = aspeed_sham_update, - .final = aspeed_sham_final, .finup = aspeed_sham_finup, .digest = aspeed_sham_digest, .export = aspeed_sham_export, .import = aspeed_sham_import, + .init_tfm = aspeed_sham_cra_init, .halg = { .digestsize = SHA256_DIGEST_SIZE, .statesize = sizeof(struct aspeed_sham_reqctx), @@ -899,13 +668,13 @@ static struct aspeed_hace_alg aspeed_ahash_algs[] = { .cra_priority = 300, .cra_flags = CRYPTO_ALG_TYPE_AHASH | CRYPTO_ALG_ASYNC | + CRYPTO_AHASH_ALG_BLOCK_ONLY | CRYPTO_ALG_KERN_DRIVER_ONLY, .cra_blocksize = SHA256_BLOCK_SIZE, .cra_ctxsize = sizeof(struct aspeed_sham_ctx), + .cra_reqsize = sizeof(struct aspeed_sham_reqctx), .cra_alignmask = 0, .cra_module = THIS_MODULE, - .cra_init = aspeed_sham_cra_init, - .cra_exit = aspeed_sham_cra_exit, } } }, @@ -917,11 +686,11 @@ static struct aspeed_hace_alg aspeed_ahash_algs[] = { .alg.ahash.base = { .init = aspeed_sham_init, .update = aspeed_sham_update, - .final = aspeed_sham_final, .finup = aspeed_sham_finup, .digest = aspeed_sham_digest, .export = aspeed_sham_export, .import = aspeed_sham_import, + .init_tfm = aspeed_sham_cra_init, .halg = { .digestsize = SHA224_DIGEST_SIZE, .statesize = sizeof(struct aspeed_sham_reqctx), @@ -931,118 +700,13 @@ static struct aspeed_hace_alg aspeed_ahash_algs[] = { .cra_priority = 300, .cra_flags = CRYPTO_ALG_TYPE_AHASH | CRYPTO_ALG_ASYNC | + CRYPTO_AHASH_ALG_BLOCK_ONLY | CRYPTO_ALG_KERN_DRIVER_ONLY, .cra_blocksize = SHA224_BLOCK_SIZE, .cra_ctxsize = sizeof(struct aspeed_sham_ctx), + .cra_reqsize = sizeof(struct aspeed_sham_reqctx), .cra_alignmask = 0, .cra_module = THIS_MODULE, - .cra_init = aspeed_sham_cra_init, - .cra_exit = aspeed_sham_cra_exit, - } - } - }, - .alg.ahash.op = { - .do_one_request = aspeed_ahash_do_one, - }, - }, - { - .alg_base = "sha1", - .alg.ahash.base = { - .init = aspeed_sham_init, - .update = aspeed_sham_update, - .final = aspeed_sham_final, - .finup = aspeed_sham_finup, - .digest = aspeed_sham_digest, - .setkey = aspeed_sham_setkey, - .export = aspeed_sham_export, - .import = aspeed_sham_import, - .halg = { - .digestsize = SHA1_DIGEST_SIZE, - .statesize = sizeof(struct aspeed_sham_reqctx), - .base = { - .cra_name = "hmac(sha1)", - .cra_driver_name = "aspeed-hmac-sha1", - .cra_priority = 300, - .cra_flags = CRYPTO_ALG_TYPE_AHASH | - CRYPTO_ALG_ASYNC | - CRYPTO_ALG_KERN_DRIVER_ONLY, - .cra_blocksize = SHA1_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct aspeed_sham_ctx) + - sizeof(struct aspeed_sha_hmac_ctx), - .cra_alignmask = 0, - .cra_module = THIS_MODULE, - .cra_init = aspeed_sham_cra_init, - .cra_exit = aspeed_sham_cra_exit, - } - } - }, - .alg.ahash.op = { - .do_one_request = aspeed_ahash_do_one, - }, - }, - { - .alg_base = "sha224", - .alg.ahash.base = { - .init = aspeed_sham_init, - .update = aspeed_sham_update, - .final = aspeed_sham_final, - .finup = aspeed_sham_finup, - .digest = aspeed_sham_digest, - .setkey = aspeed_sham_setkey, - .export = aspeed_sham_export, - .import = aspeed_sham_import, - .halg = { - .digestsize = SHA224_DIGEST_SIZE, - .statesize = sizeof(struct aspeed_sham_reqctx), - .base = { - .cra_name = "hmac(sha224)", - .cra_driver_name = "aspeed-hmac-sha224", - .cra_priority = 300, - .cra_flags = CRYPTO_ALG_TYPE_AHASH | - CRYPTO_ALG_ASYNC | - CRYPTO_ALG_KERN_DRIVER_ONLY, - .cra_blocksize = SHA224_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct aspeed_sham_ctx) + - sizeof(struct aspeed_sha_hmac_ctx), - .cra_alignmask = 0, - .cra_module = THIS_MODULE, - .cra_init = aspeed_sham_cra_init, - .cra_exit = aspeed_sham_cra_exit, - } - } - }, - .alg.ahash.op = { - .do_one_request = aspeed_ahash_do_one, - }, - }, - { - .alg_base = "sha256", - .alg.ahash.base = { - .init = aspeed_sham_init, - .update = aspeed_sham_update, - .final = aspeed_sham_final, - .finup = aspeed_sham_finup, - .digest = aspeed_sham_digest, - .setkey = aspeed_sham_setkey, - .export = aspeed_sham_export, - .import = aspeed_sham_import, - .halg = { - .digestsize = SHA256_DIGEST_SIZE, - .statesize = sizeof(struct aspeed_sham_reqctx), - .base = { - .cra_name = "hmac(sha256)", - .cra_driver_name = "aspeed-hmac-sha256", - .cra_priority = 300, - .cra_flags = CRYPTO_ALG_TYPE_AHASH | - CRYPTO_ALG_ASYNC | - CRYPTO_ALG_KERN_DRIVER_ONLY, - .cra_blocksize = SHA256_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct aspeed_sham_ctx) + - sizeof(struct aspeed_sha_hmac_ctx), - .cra_alignmask = 0, - .cra_module = THIS_MODULE, - .cra_init = aspeed_sham_cra_init, - .cra_exit = aspeed_sham_cra_exit, } } }, @@ -1057,11 +721,11 @@ static struct aspeed_hace_alg aspeed_ahash_algs_g6[] = { .alg.ahash.base = { .init = aspeed_sham_init, .update = aspeed_sham_update, - .final = aspeed_sham_final, .finup = aspeed_sham_finup, .digest = aspeed_sham_digest, .export = aspeed_sham_export, .import = aspeed_sham_import, + .init_tfm = aspeed_sham_cra_init, .halg = { .digestsize = SHA384_DIGEST_SIZE, .statesize = sizeof(struct aspeed_sham_reqctx), @@ -1071,13 +735,13 @@ static struct aspeed_hace_alg aspeed_ahash_algs_g6[] = { .cra_priority = 300, .cra_flags = CRYPTO_ALG_TYPE_AHASH | CRYPTO_ALG_ASYNC | + CRYPTO_AHASH_ALG_BLOCK_ONLY | CRYPTO_ALG_KERN_DRIVER_ONLY, .cra_blocksize = SHA384_BLOCK_SIZE, .cra_ctxsize = sizeof(struct aspeed_sham_ctx), + .cra_reqsize = sizeof(struct aspeed_sham_reqctx), .cra_alignmask = 0, .cra_module = THIS_MODULE, - .cra_init = aspeed_sham_cra_init, - .cra_exit = aspeed_sham_cra_exit, } } }, @@ -1089,11 +753,11 @@ static struct aspeed_hace_alg aspeed_ahash_algs_g6[] = { .alg.ahash.base = { .init = aspeed_sham_init, .update = aspeed_sham_update, - .final = aspeed_sham_final, .finup = aspeed_sham_finup, .digest = aspeed_sham_digest, .export = aspeed_sham_export, .import = aspeed_sham_import, + .init_tfm = aspeed_sham_cra_init, .halg = { .digestsize = SHA512_DIGEST_SIZE, .statesize = sizeof(struct aspeed_sham_reqctx), @@ -1103,83 +767,13 @@ static struct aspeed_hace_alg aspeed_ahash_algs_g6[] = { .cra_priority = 300, .cra_flags = CRYPTO_ALG_TYPE_AHASH | CRYPTO_ALG_ASYNC | + CRYPTO_AHASH_ALG_BLOCK_ONLY | CRYPTO_ALG_KERN_DRIVER_ONLY, .cra_blocksize = SHA512_BLOCK_SIZE, .cra_ctxsize = sizeof(struct aspeed_sham_ctx), + .cra_reqsize = sizeof(struct aspeed_sham_reqctx), .cra_alignmask = 0, .cra_module = THIS_MODULE, - .cra_init = aspeed_sham_cra_init, - .cra_exit = aspeed_sham_cra_exit, - } - } - }, - .alg.ahash.op = { - .do_one_request = aspeed_ahash_do_one, - }, - }, - { - .alg_base = "sha384", - .alg.ahash.base = { - .init = aspeed_sham_init, - .update = aspeed_sham_update, - .final = aspeed_sham_final, - .finup = aspeed_sham_finup, - .digest = aspeed_sham_digest, - .setkey = aspeed_sham_setkey, - .export = aspeed_sham_export, - .import = aspeed_sham_import, - .halg = { - .digestsize = SHA384_DIGEST_SIZE, - .statesize = sizeof(struct aspeed_sham_reqctx), - .base = { - .cra_name = "hmac(sha384)", - .cra_driver_name = "aspeed-hmac-sha384", - .cra_priority = 300, - .cra_flags = CRYPTO_ALG_TYPE_AHASH | - CRYPTO_ALG_ASYNC | - CRYPTO_ALG_KERN_DRIVER_ONLY, - .cra_blocksize = SHA384_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct aspeed_sham_ctx) + - sizeof(struct aspeed_sha_hmac_ctx), - .cra_alignmask = 0, - .cra_module = THIS_MODULE, - .cra_init = aspeed_sham_cra_init, - .cra_exit = aspeed_sham_cra_exit, - } - } - }, - .alg.ahash.op = { - .do_one_request = aspeed_ahash_do_one, - }, - }, - { - .alg_base = "sha512", - .alg.ahash.base = { - .init = aspeed_sham_init, - .update = aspeed_sham_update, - .final = aspeed_sham_final, - .finup = aspeed_sham_finup, - .digest = aspeed_sham_digest, - .setkey = aspeed_sham_setkey, - .export = aspeed_sham_export, - .import = aspeed_sham_import, - .halg = { - .digestsize = SHA512_DIGEST_SIZE, - .statesize = sizeof(struct aspeed_sham_reqctx), - .base = { - .cra_name = "hmac(sha512)", - .cra_driver_name = "aspeed-hmac-sha512", - .cra_priority = 300, - .cra_flags = CRYPTO_ALG_TYPE_AHASH | - CRYPTO_ALG_ASYNC | - CRYPTO_ALG_KERN_DRIVER_ONLY, - .cra_blocksize = SHA512_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct aspeed_sham_ctx) + - sizeof(struct aspeed_sha_hmac_ctx), - .cra_alignmask = 0, - .cra_module = THIS_MODULE, - .cra_init = aspeed_sham_cra_init, - .cra_exit = aspeed_sham_cra_exit, } } }, diff --git a/drivers/crypto/aspeed/aspeed-hace.h b/drivers/crypto/aspeed/aspeed-hace.h index 68f70e01fccbb..b1d07730d5436 100644 --- a/drivers/crypto/aspeed/aspeed-hace.h +++ b/drivers/crypto/aspeed/aspeed-hace.h @@ -119,7 +119,6 @@ #define SHA_FLAGS_SHA512 BIT(4) #define SHA_FLAGS_SHA512_224 BIT(5) #define SHA_FLAGS_SHA512_256 BIT(6) -#define SHA_FLAGS_HMAC BIT(8) #define SHA_FLAGS_FINUP BIT(9) #define SHA_FLAGS_MASK (0xff) @@ -161,22 +160,18 @@ struct aspeed_engine_hash { aspeed_hace_fn_t dma_prepare; }; -struct aspeed_sha_hmac_ctx { - struct crypto_shash *shash; - u8 ipad[SHA512_BLOCK_SIZE]; - u8 opad[SHA512_BLOCK_SIZE]; -}; - struct aspeed_sham_ctx { struct aspeed_hace_dev *hace_dev; - unsigned long flags; /* hmac flag */ - - struct aspeed_sha_hmac_ctx base[]; }; struct aspeed_sham_reqctx { + /* DMA buffer written by hardware */ + u8 digest[SHA512_DIGEST_SIZE] __aligned(64); + + /* Software state sorted by size. */ + u64 digcnt[2]; + unsigned long flags; /* final update flag should no use*/ - unsigned long op; /* final or update */ u32 cmd; /* trigger cmd */ /* walk state */ @@ -188,17 +183,12 @@ struct aspeed_sham_reqctx { size_t digsize; size_t block_size; size_t ivsize; - const __be32 *sha_iv; - /* remain data buffer */ - u8 buffer[SHA512_BLOCK_SIZE * 2]; dma_addr_t buffer_dma_addr; - size_t bufcnt; /* buffer counter */ - - /* output buffer */ - u8 digest[SHA512_DIGEST_SIZE] __aligned(64); dma_addr_t digest_dma_addr; - u64 digcnt[2]; + + /* This is DMA too but read-only for hardware. */ + u8 buffer[SHA512_BLOCK_SIZE + 16]; }; struct aspeed_engine_crypto { diff --git a/drivers/crypto/atmel-aes.c b/drivers/crypto/atmel-aes.c index 27c5d000b4b2b..3a2684208dda9 100644 --- a/drivers/crypto/atmel-aes.c +++ b/drivers/crypto/atmel-aes.c @@ -2297,6 +2297,7 @@ static void atmel_aes_get_cap(struct atmel_aes_dev *dd) /* keep only major version number */ switch (dd->hw_version & 0xff0) { + case 0x800: case 0x700: case 0x600: case 0x500: diff --git a/drivers/crypto/atmel-sha.c b/drivers/crypto/atmel-sha.c index 2cc36da163e84..3d7573c7bd1ca 100644 --- a/drivers/crypto/atmel-sha.c +++ b/drivers/crypto/atmel-sha.c @@ -2534,6 +2534,7 @@ static void atmel_sha_get_cap(struct atmel_sha_dev *dd) /* keep only major version number */ switch (dd->hw_version & 0xff0) { + case 0x800: case 0x700: case 0x600: case 0x510: diff --git a/drivers/crypto/caam/Makefile b/drivers/crypto/caam/Makefile index acf1b197eb84b..d2eaf5205b1ca 100644 --- a/drivers/crypto/caam/Makefile +++ b/drivers/crypto/caam/Makefile @@ -25,10 +25,6 @@ caam_jr-$(CONFIG_CRYPTO_DEV_FSL_CAAM_PKC_API) += caampkc.o pkc_desc.o caam_jr-$(CONFIG_CRYPTO_DEV_FSL_CAAM_BLOB_GEN) += blob_gen.o caam-$(CONFIG_CRYPTO_DEV_FSL_CAAM_CRYPTO_API_QI) += qi.o -ifneq ($(CONFIG_CRYPTO_DEV_FSL_CAAM_CRYPTO_API_QI),) - ccflags-y += -DCONFIG_CAAM_QI -endif - caam-$(CONFIG_DEBUG_FS) += debugfs.o obj-$(CONFIG_CRYPTO_DEV_FSL_DPAA2_CAAM) += dpaa2_caam.o diff --git a/drivers/crypto/caam/ctrl.c b/drivers/crypto/caam/ctrl.c index 38ff931059b49..a93be395c878c 100644 --- a/drivers/crypto/caam/ctrl.c +++ b/drivers/crypto/caam/ctrl.c @@ -24,7 +24,7 @@ bool caam_dpaa2; EXPORT_SYMBOL(caam_dpaa2); -#ifdef CONFIG_CAAM_QI +#ifdef CONFIG_CRYPTO_DEV_FSL_CAAM_CRYPTO_API_QI #include "qi.h" #endif @@ -573,7 +573,7 @@ static const struct soc_device_attribute caam_imx_soc_table[] = { { .soc_id = "i.MX7*", .data = &caam_imx7_data }, { .soc_id = "i.MX8M*", .data = &caam_imx7_data }, { .soc_id = "i.MX8ULP", .data = &caam_imx8ulp_data }, - { .soc_id = "i.MX8QM", .data = &caam_imx8ulp_data }, + { .soc_id = "i.MX8Q*", .data = &caam_imx8ulp_data }, { .soc_id = "VF*", .data = &caam_vf610_data }, { .family = "Freescale i.MX" }, { /* sentinel */ } @@ -831,7 +831,7 @@ static int caam_ctrl_suspend(struct device *dev) { const struct caam_drv_private *ctrlpriv = dev_get_drvdata(dev); - if (ctrlpriv->caam_off_during_pm && !ctrlpriv->optee_en) + if (ctrlpriv->caam_off_during_pm && !ctrlpriv->no_page0) caam_state_save(dev); return 0; @@ -842,7 +842,7 @@ static int caam_ctrl_resume(struct device *dev) struct caam_drv_private *ctrlpriv = dev_get_drvdata(dev); int ret = 0; - if (ctrlpriv->caam_off_during_pm && !ctrlpriv->optee_en) { + if (ctrlpriv->caam_off_during_pm && !ctrlpriv->no_page0) { caam_state_restore(dev); /* HW and rng will be reset so deinstantiation can be removed */ @@ -908,6 +908,7 @@ static int caam_probe(struct platform_device *pdev) imx_soc_data = imx_soc_match->data; reg_access = reg_access && imx_soc_data->page0_access; + ctrlpriv->no_page0 = !reg_access; /* * CAAM clocks cannot be controlled from kernel. */ @@ -967,7 +968,7 @@ iomap_ctrl: caam_dpaa2 = !!(comp_params & CTPR_MS_DPAA2); ctrlpriv->qi_present = !!(comp_params & CTPR_MS_QI_MASK); -#ifdef CONFIG_CAAM_QI +#ifdef CONFIG_CRYPTO_DEV_FSL_CAAM_CRYPTO_API_QI /* If (DPAA 1.x) QI present, check whether dependencies are available */ if (ctrlpriv->qi_present && !caam_dpaa2) { ret = qman_is_probed(); @@ -1098,7 +1099,7 @@ set_dma_mask: wr_reg32(&ctrlpriv->qi->qi_control_lo, QICTL_DQEN); /* If QMAN driver is present, init CAAM-QI backend */ -#ifdef CONFIG_CAAM_QI +#ifdef CONFIG_CRYPTO_DEV_FSL_CAAM_CRYPTO_API_QI ret = caam_qi_init(pdev); if (ret) dev_err(dev, "caam qi i/f init failed: %d\n", ret); diff --git a/drivers/crypto/caam/debugfs.c b/drivers/crypto/caam/debugfs.c index 6358d3cabf571..718352b7afb56 100644 --- a/drivers/crypto/caam/debugfs.c +++ b/drivers/crypto/caam/debugfs.c @@ -22,7 +22,7 @@ static int caam_debugfs_u32_get(void *data, u64 *val) DEFINE_DEBUGFS_ATTRIBUTE(caam_fops_u32_ro, caam_debugfs_u32_get, NULL, "%llu\n"); DEFINE_DEBUGFS_ATTRIBUTE(caam_fops_u64_ro, caam_debugfs_u64_get, NULL, "%llu\n"); -#ifdef CONFIG_CAAM_QI +#ifdef CONFIG_CRYPTO_DEV_FSL_CAAM_CRYPTO_API_QI /* * This is a counter for the number of times the congestion group (where all * the request and response queueus are) reached congestion. Incremented diff --git a/drivers/crypto/caam/debugfs.h b/drivers/crypto/caam/debugfs.h index 8b5d1acd21a7f..ef238c71f92a1 100644 --- a/drivers/crypto/caam/debugfs.h +++ b/drivers/crypto/caam/debugfs.h @@ -18,7 +18,7 @@ static inline void caam_debugfs_init(struct caam_drv_private *ctrlpriv, {} #endif -#if defined(CONFIG_DEBUG_FS) && defined(CONFIG_CAAM_QI) +#if defined(CONFIG_DEBUG_FS) && defined(CONFIG_CRYPTO_DEV_FSL_CAAM_CRYPTO_API_QI) void caam_debugfs_qi_congested(void); void caam_debugfs_qi_init(struct caam_drv_private *ctrlpriv); #else diff --git a/drivers/crypto/caam/intern.h b/drivers/crypto/caam/intern.h index e513201508720..a88da0d31b23b 100644 --- a/drivers/crypto/caam/intern.h +++ b/drivers/crypto/caam/intern.h @@ -115,6 +115,7 @@ struct caam_drv_private { u8 blob_present; /* Nonzero if BLOB support present in device */ u8 mc_en; /* Nonzero if MC f/w is active */ u8 optee_en; /* Nonzero if OP-TEE f/w is active */ + u8 no_page0; /* Nonzero if register page 0 is not controlled by Linux */ bool pr_support; /* RNG prediction resistance available */ int secvio_irq; /* Security violation interrupt number */ int virt_en; /* Virtualization enabled in CAAM */ @@ -226,7 +227,7 @@ static inline int caam_prng_register(struct device *dev) static inline void caam_prng_unregister(void *data) {} #endif /* CONFIG_CRYPTO_DEV_FSL_CAAM_PRNG_API */ -#ifdef CONFIG_CAAM_QI +#ifdef CONFIG_CRYPTO_DEV_FSL_CAAM_CRYPTO_API_QI int caam_qi_algapi_init(struct device *dev); void caam_qi_algapi_exit(void); @@ -242,7 +243,7 @@ static inline void caam_qi_algapi_exit(void) { } -#endif /* CONFIG_CAAM_QI */ +#endif /* CONFIG_CRYPTO_DEV_FSL_CAAM_CRYPTO_API_QI */ static inline u64 caam_get_dma_mask(struct device *dev) { diff --git a/drivers/crypto/caam/jr.c b/drivers/crypto/caam/jr.c index 9fcdb64084acc..0ef00df9730e5 100644 --- a/drivers/crypto/caam/jr.c +++ b/drivers/crypto/caam/jr.c @@ -629,8 +629,7 @@ static int caam_jr_probe(struct platform_device *pdev) } /* Initialize crypto engine */ - jrpriv->engine = crypto_engine_alloc_init_and_set(jrdev, true, NULL, - false, + jrpriv->engine = crypto_engine_alloc_init_and_set(jrdev, true, false, CRYPTO_ENGINE_MAX_QLEN); if (!jrpriv->engine) { dev_err(jrdev, "Could not init crypto-engine\n"); diff --git a/drivers/crypto/caam/qi.c b/drivers/crypto/caam/qi.c index b6e7c0b29d4e6..1e731ed8702b5 100644 --- a/drivers/crypto/caam/qi.c +++ b/drivers/crypto/caam/qi.c @@ -442,11 +442,8 @@ struct caam_drv_ctx *caam_drv_ctx_init(struct device *qidev, if (!cpumask_test_cpu(*cpu, cpus)) { int *pcpu = &get_cpu_var(last_cpu); - *pcpu = cpumask_next(*pcpu, cpus); - if (*pcpu >= nr_cpu_ids) - *pcpu = cpumask_first(cpus); + *pcpu = cpumask_next_wrap(*pcpu, cpus); *cpu = *pcpu; - put_cpu_var(last_cpu); } drv_ctx->cpu = *cpu; diff --git a/drivers/crypto/ccp/ccp-debugfs.c b/drivers/crypto/ccp/ccp-debugfs.c index a1055554b47a2..dc26bc22c91d1 100644 --- a/drivers/crypto/ccp/ccp-debugfs.c +++ b/drivers/crypto/ccp/ccp-debugfs.c @@ -319,5 +319,8 @@ void ccp5_debugfs_setup(struct ccp_device *ccp) void ccp5_debugfs_destroy(void) { + mutex_lock(&ccp_debugfs_lock); debugfs_remove_recursive(ccp_debugfs_dir); + ccp_debugfs_dir = NULL; + mutex_unlock(&ccp_debugfs_lock); } diff --git a/drivers/crypto/ccp/ccp-ops.c b/drivers/crypto/ccp/ccp-ops.c index 109b5aef40343..d78865d9d5f09 100644 --- a/drivers/crypto/ccp/ccp-ops.c +++ b/drivers/crypto/ccp/ccp-ops.c @@ -633,10 +633,16 @@ static noinline_for_stack int ccp_run_aes_gcm_cmd(struct ccp_cmd_queue *cmd_q, struct ccp_cmd *cmd) { struct ccp_aes_engine *aes = &cmd->u.aes; - struct ccp_dm_workarea key, ctx, final_wa, tag; - struct ccp_data src, dst; - struct ccp_data aad; - struct ccp_op op; + struct { + struct ccp_dm_workarea key; + struct ccp_dm_workarea ctx; + struct ccp_dm_workarea final; + struct ccp_dm_workarea tag; + struct ccp_data src; + struct ccp_data dst; + struct ccp_data aad; + struct ccp_op op; + } *wa __cleanup(kfree) = kzalloc(sizeof *wa, GFP_KERNEL); unsigned int dm_offset; unsigned int authsize; unsigned int jobid; @@ -650,6 +656,9 @@ ccp_run_aes_gcm_cmd(struct ccp_cmd_queue *cmd_q, struct ccp_cmd *cmd) struct scatterlist *p_outp, sg_outp[2]; struct scatterlist *p_aad; + if (!wa) + return -ENOMEM; + if (!aes->iv) return -EINVAL; @@ -696,26 +705,26 @@ ccp_run_aes_gcm_cmd(struct ccp_cmd_queue *cmd_q, struct ccp_cmd *cmd) jobid = CCP_NEW_JOBID(cmd_q->ccp); - memset(&op, 0, sizeof(op)); - op.cmd_q = cmd_q; - op.jobid = jobid; - op.sb_key = cmd_q->sb_key; /* Pre-allocated */ - op.sb_ctx = cmd_q->sb_ctx; /* Pre-allocated */ - op.init = 1; - op.u.aes.type = aes->type; + memset(&wa->op, 0, sizeof(wa->op)); + wa->op.cmd_q = cmd_q; + wa->op.jobid = jobid; + wa->op.sb_key = cmd_q->sb_key; /* Pre-allocated */ + wa->op.sb_ctx = cmd_q->sb_ctx; /* Pre-allocated */ + wa->op.init = 1; + wa->op.u.aes.type = aes->type; /* Copy the key to the LSB */ - ret = ccp_init_dm_workarea(&key, cmd_q, + ret = ccp_init_dm_workarea(&wa->key, cmd_q, CCP_AES_CTX_SB_COUNT * CCP_SB_BYTES, DMA_TO_DEVICE); if (ret) return ret; dm_offset = CCP_SB_BYTES - aes->key_len; - ret = ccp_set_dm_area(&key, dm_offset, aes->key, 0, aes->key_len); + ret = ccp_set_dm_area(&wa->key, dm_offset, aes->key, 0, aes->key_len); if (ret) goto e_key; - ret = ccp_copy_to_sb(cmd_q, &key, op.jobid, op.sb_key, + ret = ccp_copy_to_sb(cmd_q, &wa->key, wa->op.jobid, wa->op.sb_key, CCP_PASSTHRU_BYTESWAP_256BIT); if (ret) { cmd->engine_error = cmd_q->cmd_error; @@ -726,58 +735,58 @@ ccp_run_aes_gcm_cmd(struct ccp_cmd_queue *cmd_q, struct ccp_cmd *cmd) * There is an assumption here that the IV is 96 bits in length, plus * a nonce of 32 bits. If no IV is present, use a zeroed buffer. */ - ret = ccp_init_dm_workarea(&ctx, cmd_q, + ret = ccp_init_dm_workarea(&wa->ctx, cmd_q, CCP_AES_CTX_SB_COUNT * CCP_SB_BYTES, DMA_BIDIRECTIONAL); if (ret) goto e_key; dm_offset = CCP_AES_CTX_SB_COUNT * CCP_SB_BYTES - aes->iv_len; - ret = ccp_set_dm_area(&ctx, dm_offset, aes->iv, 0, aes->iv_len); + ret = ccp_set_dm_area(&wa->ctx, dm_offset, aes->iv, 0, aes->iv_len); if (ret) goto e_ctx; - ret = ccp_copy_to_sb(cmd_q, &ctx, op.jobid, op.sb_ctx, + ret = ccp_copy_to_sb(cmd_q, &wa->ctx, wa->op.jobid, wa->op.sb_ctx, CCP_PASSTHRU_BYTESWAP_256BIT); if (ret) { cmd->engine_error = cmd_q->cmd_error; goto e_ctx; } - op.init = 1; + wa->op.init = 1; if (aes->aad_len > 0) { /* Step 1: Run a GHASH over the Additional Authenticated Data */ - ret = ccp_init_data(&aad, cmd_q, p_aad, aes->aad_len, + ret = ccp_init_data(&wa->aad, cmd_q, p_aad, aes->aad_len, AES_BLOCK_SIZE, DMA_TO_DEVICE); if (ret) goto e_ctx; - op.u.aes.mode = CCP_AES_MODE_GHASH; - op.u.aes.action = CCP_AES_GHASHAAD; + wa->op.u.aes.mode = CCP_AES_MODE_GHASH; + wa->op.u.aes.action = CCP_AES_GHASHAAD; - while (aad.sg_wa.bytes_left) { - ccp_prepare_data(&aad, NULL, &op, AES_BLOCK_SIZE, true); + while (wa->aad.sg_wa.bytes_left) { + ccp_prepare_data(&wa->aad, NULL, &wa->op, AES_BLOCK_SIZE, true); - ret = cmd_q->ccp->vdata->perform->aes(&op); + ret = cmd_q->ccp->vdata->perform->aes(&wa->op); if (ret) { cmd->engine_error = cmd_q->cmd_error; goto e_aad; } - ccp_process_data(&aad, NULL, &op); - op.init = 0; + ccp_process_data(&wa->aad, NULL, &wa->op); + wa->op.init = 0; } } - op.u.aes.mode = CCP_AES_MODE_GCTR; - op.u.aes.action = aes->action; + wa->op.u.aes.mode = CCP_AES_MODE_GCTR; + wa->op.u.aes.action = aes->action; if (ilen > 0) { /* Step 2: Run a GCTR over the plaintext */ in_place = (sg_virt(p_inp) == sg_virt(p_outp)) ? true : false; - ret = ccp_init_data(&src, cmd_q, p_inp, ilen, + ret = ccp_init_data(&wa->src, cmd_q, p_inp, ilen, AES_BLOCK_SIZE, in_place ? DMA_BIDIRECTIONAL : DMA_TO_DEVICE); @@ -785,52 +794,52 @@ ccp_run_aes_gcm_cmd(struct ccp_cmd_queue *cmd_q, struct ccp_cmd *cmd) goto e_aad; if (in_place) { - dst = src; + wa->dst = wa->src; } else { - ret = ccp_init_data(&dst, cmd_q, p_outp, ilen, + ret = ccp_init_data(&wa->dst, cmd_q, p_outp, ilen, AES_BLOCK_SIZE, DMA_FROM_DEVICE); if (ret) goto e_src; } - op.soc = 0; - op.eom = 0; - op.init = 1; - while (src.sg_wa.bytes_left) { - ccp_prepare_data(&src, &dst, &op, AES_BLOCK_SIZE, true); - if (!src.sg_wa.bytes_left) { + wa->op.soc = 0; + wa->op.eom = 0; + wa->op.init = 1; + while (wa->src.sg_wa.bytes_left) { + ccp_prepare_data(&wa->src, &wa->dst, &wa->op, AES_BLOCK_SIZE, true); + if (!wa->src.sg_wa.bytes_left) { unsigned int nbytes = ilen % AES_BLOCK_SIZE; if (nbytes) { - op.eom = 1; - op.u.aes.size = (nbytes * 8) - 1; + wa->op.eom = 1; + wa->op.u.aes.size = (nbytes * 8) - 1; } } - ret = cmd_q->ccp->vdata->perform->aes(&op); + ret = cmd_q->ccp->vdata->perform->aes(&wa->op); if (ret) { cmd->engine_error = cmd_q->cmd_error; goto e_dst; } - ccp_process_data(&src, &dst, &op); - op.init = 0; + ccp_process_data(&wa->src, &wa->dst, &wa->op); + wa->op.init = 0; } } /* Step 3: Update the IV portion of the context with the original IV */ - ret = ccp_copy_from_sb(cmd_q, &ctx, op.jobid, op.sb_ctx, + ret = ccp_copy_from_sb(cmd_q, &wa->ctx, wa->op.jobid, wa->op.sb_ctx, CCP_PASSTHRU_BYTESWAP_256BIT); if (ret) { cmd->engine_error = cmd_q->cmd_error; goto e_dst; } - ret = ccp_set_dm_area(&ctx, dm_offset, aes->iv, 0, aes->iv_len); + ret = ccp_set_dm_area(&wa->ctx, dm_offset, aes->iv, 0, aes->iv_len); if (ret) goto e_dst; - ret = ccp_copy_to_sb(cmd_q, &ctx, op.jobid, op.sb_ctx, + ret = ccp_copy_to_sb(cmd_q, &wa->ctx, wa->op.jobid, wa->op.sb_ctx, CCP_PASSTHRU_BYTESWAP_256BIT); if (ret) { cmd->engine_error = cmd_q->cmd_error; @@ -840,75 +849,75 @@ ccp_run_aes_gcm_cmd(struct ccp_cmd_queue *cmd_q, struct ccp_cmd *cmd) /* Step 4: Concatenate the lengths of the AAD and source, and * hash that 16 byte buffer. */ - ret = ccp_init_dm_workarea(&final_wa, cmd_q, AES_BLOCK_SIZE, + ret = ccp_init_dm_workarea(&wa->final, cmd_q, AES_BLOCK_SIZE, DMA_BIDIRECTIONAL); if (ret) goto e_dst; - final = (__be64 *)final_wa.address; + final = (__be64 *)wa->final.address; final[0] = cpu_to_be64(aes->aad_len * 8); final[1] = cpu_to_be64(ilen * 8); - memset(&op, 0, sizeof(op)); - op.cmd_q = cmd_q; - op.jobid = jobid; - op.sb_key = cmd_q->sb_key; /* Pre-allocated */ - op.sb_ctx = cmd_q->sb_ctx; /* Pre-allocated */ - op.init = 1; - op.u.aes.type = aes->type; - op.u.aes.mode = CCP_AES_MODE_GHASH; - op.u.aes.action = CCP_AES_GHASHFINAL; - op.src.type = CCP_MEMTYPE_SYSTEM; - op.src.u.dma.address = final_wa.dma.address; - op.src.u.dma.length = AES_BLOCK_SIZE; - op.dst.type = CCP_MEMTYPE_SYSTEM; - op.dst.u.dma.address = final_wa.dma.address; - op.dst.u.dma.length = AES_BLOCK_SIZE; - op.eom = 1; - op.u.aes.size = 0; - ret = cmd_q->ccp->vdata->perform->aes(&op); + memset(&wa->op, 0, sizeof(wa->op)); + wa->op.cmd_q = cmd_q; + wa->op.jobid = jobid; + wa->op.sb_key = cmd_q->sb_key; /* Pre-allocated */ + wa->op.sb_ctx = cmd_q->sb_ctx; /* Pre-allocated */ + wa->op.init = 1; + wa->op.u.aes.type = aes->type; + wa->op.u.aes.mode = CCP_AES_MODE_GHASH; + wa->op.u.aes.action = CCP_AES_GHASHFINAL; + wa->op.src.type = CCP_MEMTYPE_SYSTEM; + wa->op.src.u.dma.address = wa->final.dma.address; + wa->op.src.u.dma.length = AES_BLOCK_SIZE; + wa->op.dst.type = CCP_MEMTYPE_SYSTEM; + wa->op.dst.u.dma.address = wa->final.dma.address; + wa->op.dst.u.dma.length = AES_BLOCK_SIZE; + wa->op.eom = 1; + wa->op.u.aes.size = 0; + ret = cmd_q->ccp->vdata->perform->aes(&wa->op); if (ret) goto e_final_wa; if (aes->action == CCP_AES_ACTION_ENCRYPT) { /* Put the ciphered tag after the ciphertext. */ - ccp_get_dm_area(&final_wa, 0, p_tag, 0, authsize); + ccp_get_dm_area(&wa->final, 0, p_tag, 0, authsize); } else { /* Does this ciphered tag match the input? */ - ret = ccp_init_dm_workarea(&tag, cmd_q, authsize, + ret = ccp_init_dm_workarea(&wa->tag, cmd_q, authsize, DMA_BIDIRECTIONAL); if (ret) goto e_final_wa; - ret = ccp_set_dm_area(&tag, 0, p_tag, 0, authsize); + ret = ccp_set_dm_area(&wa->tag, 0, p_tag, 0, authsize); if (ret) { - ccp_dm_free(&tag); + ccp_dm_free(&wa->tag); goto e_final_wa; } - ret = crypto_memneq(tag.address, final_wa.address, + ret = crypto_memneq(wa->tag.address, wa->final.address, authsize) ? -EBADMSG : 0; - ccp_dm_free(&tag); + ccp_dm_free(&wa->tag); } e_final_wa: - ccp_dm_free(&final_wa); + ccp_dm_free(&wa->final); e_dst: if (ilen > 0 && !in_place) - ccp_free_data(&dst, cmd_q); + ccp_free_data(&wa->dst, cmd_q); e_src: if (ilen > 0) - ccp_free_data(&src, cmd_q); + ccp_free_data(&wa->src, cmd_q); e_aad: if (aes->aad_len) - ccp_free_data(&aad, cmd_q); + ccp_free_data(&wa->aad, cmd_q); e_ctx: - ccp_dm_free(&ctx); + ccp_dm_free(&wa->ctx); e_key: - ccp_dm_free(&key); + ccp_dm_free(&wa->key); return ret; } diff --git a/drivers/crypto/ccp/sev-dev.c b/drivers/crypto/ccp/sev-dev.c index 3451bada884e1..e058ba0277929 100644 --- a/drivers/crypto/ccp/sev-dev.c +++ b/drivers/crypto/ccp/sev-dev.c @@ -434,7 +434,7 @@ cleanup: return rc; } -static struct page *__snp_alloc_firmware_pages(gfp_t gfp_mask, int order) +static struct page *__snp_alloc_firmware_pages(gfp_t gfp_mask, int order, bool locked) { unsigned long npages = 1ul << order, paddr; struct sev_device *sev; @@ -453,7 +453,7 @@ static struct page *__snp_alloc_firmware_pages(gfp_t gfp_mask, int order) return page; paddr = __pa((unsigned long)page_address(page)); - if (rmp_mark_pages_firmware(paddr, npages, false)) + if (rmp_mark_pages_firmware(paddr, npages, locked)) return NULL; return page; @@ -463,7 +463,7 @@ void *snp_alloc_firmware_page(gfp_t gfp_mask) { struct page *page; - page = __snp_alloc_firmware_pages(gfp_mask, 0); + page = __snp_alloc_firmware_pages(gfp_mask, 0, false); return page ? page_address(page) : NULL; } @@ -498,7 +498,7 @@ static void *sev_fw_alloc(unsigned long len) { struct page *page; - page = __snp_alloc_firmware_pages(GFP_KERNEL, get_order(len)); + page = __snp_alloc_firmware_pages(GFP_KERNEL, get_order(len), true); if (!page) return NULL; @@ -1276,9 +1276,11 @@ static int __sev_platform_init_handle_init_ex_path(struct sev_device *sev) static int __sev_platform_init_locked(int *error) { - int rc, psp_ret = SEV_RET_NO_FW_CALL; + int rc, psp_ret, dfflush_error; struct sev_device *sev; + psp_ret = dfflush_error = SEV_RET_NO_FW_CALL; + if (!psp_master || !psp_master->sev_data) return -ENODEV; @@ -1320,10 +1322,10 @@ static int __sev_platform_init_locked(int *error) /* Prepare for first SEV guest launch after INIT */ wbinvd_on_all_cpus(); - rc = __sev_do_cmd_locked(SEV_CMD_DF_FLUSH, NULL, error); + rc = __sev_do_cmd_locked(SEV_CMD_DF_FLUSH, NULL, &dfflush_error); if (rc) { dev_err(sev->dev, "SEV: DF_FLUSH failed %#x, rc %d\n", - *error, rc); + dfflush_error, rc); return rc; } @@ -1785,8 +1787,14 @@ static int __sev_snp_shutdown_locked(int *error, bool panic) sev->snp_initialized = false; dev_dbg(sev->dev, "SEV-SNP firmware shutdown\n"); - atomic_notifier_chain_unregister(&panic_notifier_list, - &snp_panic_notifier); + /* + * __sev_snp_shutdown_locked() deadlocks when it tries to unregister + * itself during panic as the panic notifier is called with RCU read + * lock held and notifier unregistration does RCU synchronization. + */ + if (!panic) + atomic_notifier_chain_unregister(&panic_notifier_list, + &snp_panic_notifier); /* Reset TMR size back to default */ sev_es_tmr_size = SEV_TMR_SIZE; diff --git a/drivers/crypto/ccp/sp-pci.c b/drivers/crypto/ccp/sp-pci.c index e1be2072d680e..e7bb803912a6d 100644 --- a/drivers/crypto/ccp/sp-pci.c +++ b/drivers/crypto/ccp/sp-pci.c @@ -453,6 +453,7 @@ static const struct psp_vdata pspv6 = { .cmdresp_reg = 0x10944, /* C2PMSG_17 */ .cmdbuff_addr_lo_reg = 0x10948, /* C2PMSG_18 */ .cmdbuff_addr_hi_reg = 0x1094c, /* C2PMSG_19 */ + .bootloader_info_reg = 0x109ec, /* C2PMSG_59 */ .feature_reg = 0x109fc, /* C2PMSG_63 */ .inten_reg = 0x10510, /* P2CMSG_INTEN */ .intsts_reg = 0x10514, /* P2CMSG_INTSTS */ diff --git a/drivers/crypto/ccree/cc_buffer_mgr.c b/drivers/crypto/ccree/cc_buffer_mgr.c index bcca55bff910e..3963bb91321fc 100644 --- a/drivers/crypto/ccree/cc_buffer_mgr.c +++ b/drivers/crypto/ccree/cc_buffer_mgr.c @@ -224,7 +224,7 @@ static int cc_generate_mlli(struct device *dev, struct buffer_array *sg_data, /* Set MLLI size for the bypass operation */ mlli_params->mlli_len = (total_nents * LLI_ENTRY_BYTE_SIZE); - dev_dbg(dev, "MLLI params: virt_addr=%pK dma_addr=%pad mlli_len=0x%X\n", + dev_dbg(dev, "MLLI params: virt_addr=%p dma_addr=%pad mlli_len=0x%X\n", mlli_params->mlli_virt_addr, &mlli_params->mlli_dma_addr, mlli_params->mlli_len); @@ -239,7 +239,7 @@ static void cc_add_sg_entry(struct device *dev, struct buffer_array *sgl_data, { unsigned int index = sgl_data->num_of_buffers; - dev_dbg(dev, "index=%u nents=%u sgl=%pK data_len=0x%08X is_last=%d\n", + dev_dbg(dev, "index=%u nents=%u sgl=%p data_len=0x%08X is_last=%d\n", index, nents, sgl, data_len, is_last_table); sgl_data->nents[index] = nents; sgl_data->entry[index].sgl = sgl; @@ -298,7 +298,7 @@ cc_set_aead_conf_buf(struct device *dev, struct aead_req_ctx *areq_ctx, dev_err(dev, "dma_map_sg() config buffer failed\n"); return -ENOMEM; } - dev_dbg(dev, "Mapped curr_buff: dma_address=%pad page=%p addr=%pK offset=%u length=%u\n", + dev_dbg(dev, "Mapped curr_buff: dma_address=%pad page=%p addr=%p offset=%u length=%u\n", &sg_dma_address(&areq_ctx->ccm_adata_sg), sg_page(&areq_ctx->ccm_adata_sg), sg_virt(&areq_ctx->ccm_adata_sg), @@ -323,7 +323,7 @@ static int cc_set_hash_buf(struct device *dev, struct ahash_req_ctx *areq_ctx, dev_err(dev, "dma_map_sg() src buffer failed\n"); return -ENOMEM; } - dev_dbg(dev, "Mapped curr_buff: dma_address=%pad page=%p addr=%pK offset=%u length=%u\n", + dev_dbg(dev, "Mapped curr_buff: dma_address=%pad page=%p addr=%p offset=%u length=%u\n", &sg_dma_address(areq_ctx->buff_sg), sg_page(areq_ctx->buff_sg), sg_virt(areq_ctx->buff_sg), areq_ctx->buff_sg->offset, areq_ctx->buff_sg->length); @@ -359,11 +359,11 @@ void cc_unmap_cipher_request(struct device *dev, void *ctx, if (src != dst) { dma_unmap_sg(dev, src, req_ctx->in_nents, DMA_TO_DEVICE); dma_unmap_sg(dev, dst, req_ctx->out_nents, DMA_FROM_DEVICE); - dev_dbg(dev, "Unmapped req->dst=%pK\n", sg_virt(dst)); - dev_dbg(dev, "Unmapped req->src=%pK\n", sg_virt(src)); + dev_dbg(dev, "Unmapped req->dst=%p\n", sg_virt(dst)); + dev_dbg(dev, "Unmapped req->src=%p\n", sg_virt(src)); } else { dma_unmap_sg(dev, src, req_ctx->in_nents, DMA_BIDIRECTIONAL); - dev_dbg(dev, "Unmapped req->src=%pK\n", sg_virt(src)); + dev_dbg(dev, "Unmapped req->src=%p\n", sg_virt(src)); } } @@ -391,11 +391,11 @@ int cc_map_cipher_request(struct cc_drvdata *drvdata, void *ctx, req_ctx->gen_ctx.iv_dma_addr = dma_map_single(dev, info, ivsize, DMA_BIDIRECTIONAL); if (dma_mapping_error(dev, req_ctx->gen_ctx.iv_dma_addr)) { - dev_err(dev, "Mapping iv %u B at va=%pK for DMA failed\n", + dev_err(dev, "Mapping iv %u B at va=%p for DMA failed\n", ivsize, info); return -ENOMEM; } - dev_dbg(dev, "Mapped iv %u B at va=%pK to dma=%pad\n", + dev_dbg(dev, "Mapped iv %u B at va=%p to dma=%pad\n", ivsize, info, &req_ctx->gen_ctx.iv_dma_addr); } else { req_ctx->gen_ctx.iv_dma_addr = 0; @@ -506,7 +506,7 @@ void cc_unmap_aead_request(struct device *dev, struct aead_request *req) if ((areq_ctx->assoc_buff_type == CC_DMA_BUF_MLLI || areq_ctx->data_buff_type == CC_DMA_BUF_MLLI) && (areq_ctx->mlli_params.mlli_virt_addr)) { - dev_dbg(dev, "free MLLI buffer: dma=%pad virt=%pK\n", + dev_dbg(dev, "free MLLI buffer: dma=%pad virt=%p\n", &areq_ctx->mlli_params.mlli_dma_addr, areq_ctx->mlli_params.mlli_virt_addr); dma_pool_free(areq_ctx->mlli_params.curr_pool, @@ -514,13 +514,13 @@ void cc_unmap_aead_request(struct device *dev, struct aead_request *req) areq_ctx->mlli_params.mlli_dma_addr); } - dev_dbg(dev, "Unmapping src sgl: req->src=%pK areq_ctx->src.nents=%u areq_ctx->assoc.nents=%u assoclen:%u cryptlen=%u\n", + dev_dbg(dev, "Unmapping src sgl: req->src=%p areq_ctx->src.nents=%u areq_ctx->assoc.nents=%u assoclen:%u cryptlen=%u\n", sg_virt(req->src), areq_ctx->src.nents, areq_ctx->assoc.nents, areq_ctx->assoclen, req->cryptlen); dma_unmap_sg(dev, req->src, areq_ctx->src.mapped_nents, src_direction); if (req->src != req->dst) { - dev_dbg(dev, "Unmapping dst sgl: req->dst=%pK\n", + dev_dbg(dev, "Unmapping dst sgl: req->dst=%p\n", sg_virt(req->dst)); dma_unmap_sg(dev, req->dst, areq_ctx->dst.mapped_nents, DMA_FROM_DEVICE); } @@ -566,7 +566,7 @@ static int cc_aead_chain_iv(struct cc_drvdata *drvdata, dma_map_single(dev, areq_ctx->gen_ctx.iv, hw_iv_size, DMA_BIDIRECTIONAL); if (dma_mapping_error(dev, areq_ctx->gen_ctx.iv_dma_addr)) { - dev_err(dev, "Mapping iv %u B at va=%pK for DMA failed\n", + dev_err(dev, "Mapping iv %u B at va=%p for DMA failed\n", hw_iv_size, req->iv); kfree_sensitive(areq_ctx->gen_ctx.iv); areq_ctx->gen_ctx.iv = NULL; @@ -574,7 +574,7 @@ static int cc_aead_chain_iv(struct cc_drvdata *drvdata, goto chain_iv_exit; } - dev_dbg(dev, "Mapped iv %u B at va=%pK to dma=%pad\n", + dev_dbg(dev, "Mapped iv %u B at va=%p to dma=%pad\n", hw_iv_size, req->iv, &areq_ctx->gen_ctx.iv_dma_addr); chain_iv_exit: @@ -977,7 +977,7 @@ int cc_map_aead_request(struct cc_drvdata *drvdata, struct aead_request *req) dma_addr = dma_map_single(dev, areq_ctx->mac_buf, MAX_MAC_SIZE, DMA_BIDIRECTIONAL); if (dma_mapping_error(dev, dma_addr)) { - dev_err(dev, "Mapping mac_buf %u B at va=%pK for DMA failed\n", + dev_err(dev, "Mapping mac_buf %u B at va=%p for DMA failed\n", MAX_MAC_SIZE, areq_ctx->mac_buf); rc = -ENOMEM; goto aead_map_failure; @@ -991,7 +991,7 @@ int cc_map_aead_request(struct cc_drvdata *drvdata, struct aead_request *req) DMA_TO_DEVICE); if (dma_mapping_error(dev, dma_addr)) { - dev_err(dev, "Mapping mac_buf %u B at va=%pK for DMA failed\n", + dev_err(dev, "Mapping mac_buf %u B at va=%p for DMA failed\n", AES_BLOCK_SIZE, addr); areq_ctx->ccm_iv0_dma_addr = 0; rc = -ENOMEM; @@ -1009,7 +1009,7 @@ int cc_map_aead_request(struct cc_drvdata *drvdata, struct aead_request *req) dma_addr = dma_map_single(dev, areq_ctx->hkey, AES_BLOCK_SIZE, DMA_BIDIRECTIONAL); if (dma_mapping_error(dev, dma_addr)) { - dev_err(dev, "Mapping hkey %u B at va=%pK for DMA failed\n", + dev_err(dev, "Mapping hkey %u B at va=%p for DMA failed\n", AES_BLOCK_SIZE, areq_ctx->hkey); rc = -ENOMEM; goto aead_map_failure; @@ -1019,7 +1019,7 @@ int cc_map_aead_request(struct cc_drvdata *drvdata, struct aead_request *req) dma_addr = dma_map_single(dev, &areq_ctx->gcm_len_block, AES_BLOCK_SIZE, DMA_TO_DEVICE); if (dma_mapping_error(dev, dma_addr)) { - dev_err(dev, "Mapping gcm_len_block %u B at va=%pK for DMA failed\n", + dev_err(dev, "Mapping gcm_len_block %u B at va=%p for DMA failed\n", AES_BLOCK_SIZE, &areq_ctx->gcm_len_block); rc = -ENOMEM; goto aead_map_failure; @@ -1030,7 +1030,7 @@ int cc_map_aead_request(struct cc_drvdata *drvdata, struct aead_request *req) AES_BLOCK_SIZE, DMA_TO_DEVICE); if (dma_mapping_error(dev, dma_addr)) { - dev_err(dev, "Mapping gcm_iv_inc1 %u B at va=%pK for DMA failed\n", + dev_err(dev, "Mapping gcm_iv_inc1 %u B at va=%p for DMA failed\n", AES_BLOCK_SIZE, (areq_ctx->gcm_iv_inc1)); areq_ctx->gcm_iv_inc1_dma_addr = 0; rc = -ENOMEM; @@ -1042,7 +1042,7 @@ int cc_map_aead_request(struct cc_drvdata *drvdata, struct aead_request *req) AES_BLOCK_SIZE, DMA_TO_DEVICE); if (dma_mapping_error(dev, dma_addr)) { - dev_err(dev, "Mapping gcm_iv_inc2 %u B at va=%pK for DMA failed\n", + dev_err(dev, "Mapping gcm_iv_inc2 %u B at va=%p for DMA failed\n", AES_BLOCK_SIZE, (areq_ctx->gcm_iv_inc2)); areq_ctx->gcm_iv_inc2_dma_addr = 0; rc = -ENOMEM; @@ -1152,7 +1152,7 @@ int cc_map_hash_request_final(struct cc_drvdata *drvdata, void *ctx, u32 dummy = 0; u32 mapped_nents = 0; - dev_dbg(dev, "final params : curr_buff=%pK curr_buff_cnt=0x%X nbytes = 0x%X src=%pK curr_index=%u\n", + dev_dbg(dev, "final params : curr_buff=%p curr_buff_cnt=0x%X nbytes = 0x%X src=%p curr_index=%u\n", curr_buff, *curr_buff_cnt, nbytes, src, areq_ctx->buff_index); /* Init the type of the dma buffer */ areq_ctx->data_dma_buf_type = CC_DMA_BUF_NULL; @@ -1236,7 +1236,7 @@ int cc_map_hash_request_update(struct cc_drvdata *drvdata, void *ctx, u32 dummy = 0; u32 mapped_nents = 0; - dev_dbg(dev, " update params : curr_buff=%pK curr_buff_cnt=0x%X nbytes=0x%X src=%pK curr_index=%u\n", + dev_dbg(dev, " update params : curr_buff=%p curr_buff_cnt=0x%X nbytes=0x%X src=%p curr_index=%u\n", curr_buff, *curr_buff_cnt, nbytes, src, areq_ctx->buff_index); /* Init the type of the dma buffer */ areq_ctx->data_dma_buf_type = CC_DMA_BUF_NULL; @@ -1246,7 +1246,7 @@ int cc_map_hash_request_update(struct cc_drvdata *drvdata, void *ctx, areq_ctx->in_nents = 0; if (total_in_len < block_size) { - dev_dbg(dev, " less than one block: curr_buff=%pK *curr_buff_cnt=0x%X copy_to=%pK\n", + dev_dbg(dev, " less than one block: curr_buff=%p *curr_buff_cnt=0x%X copy_to=%p\n", curr_buff, *curr_buff_cnt, &curr_buff[*curr_buff_cnt]); areq_ctx->in_nents = sg_nents_for_len(src, nbytes); sg_copy_to_buffer(src, areq_ctx->in_nents, @@ -1265,7 +1265,7 @@ int cc_map_hash_request_update(struct cc_drvdata *drvdata, void *ctx, /* Copy the new residue to next buffer */ if (*next_buff_cnt) { - dev_dbg(dev, " handle residue: next buff %pK skip data %u residue %u\n", + dev_dbg(dev, " handle residue: next buff %p skip data %u residue %u\n", next_buff, (update_data_len - *curr_buff_cnt), *next_buff_cnt); cc_copy_sg_portion(dev, next_buff, src, @@ -1338,7 +1338,7 @@ void cc_unmap_hash_request(struct device *dev, void *ctx, *allocated and should be released */ if (areq_ctx->mlli_params.curr_pool) { - dev_dbg(dev, "free MLLI buffer: dma=%pad virt=%pK\n", + dev_dbg(dev, "free MLLI buffer: dma=%pad virt=%p\n", &areq_ctx->mlli_params.mlli_dma_addr, areq_ctx->mlli_params.mlli_virt_addr); dma_pool_free(areq_ctx->mlli_params.curr_pool, @@ -1347,14 +1347,14 @@ void cc_unmap_hash_request(struct device *dev, void *ctx, } if (src && areq_ctx->in_nents) { - dev_dbg(dev, "Unmapped sg src: virt=%pK dma=%pad len=0x%X\n", + dev_dbg(dev, "Unmapped sg src: virt=%p dma=%pad len=0x%X\n", sg_virt(src), &sg_dma_address(src), sg_dma_len(src)); dma_unmap_sg(dev, src, areq_ctx->in_nents, DMA_TO_DEVICE); } if (*prev_len) { - dev_dbg(dev, "Unmapped buffer: areq_ctx->buff_sg=%pK dma=%pad len 0x%X\n", + dev_dbg(dev, "Unmapped buffer: areq_ctx->buff_sg=%p dma=%pad len 0x%X\n", sg_virt(areq_ctx->buff_sg), &sg_dma_address(areq_ctx->buff_sg), sg_dma_len(areq_ctx->buff_sg)); diff --git a/drivers/crypto/ccree/cc_cipher.c b/drivers/crypto/ccree/cc_cipher.c index d39c067672fd6..e2cbfdf7a0e46 100644 --- a/drivers/crypto/ccree/cc_cipher.c +++ b/drivers/crypto/ccree/cc_cipher.c @@ -211,11 +211,11 @@ static int cc_cipher_init(struct crypto_tfm *tfm) max_key_buf_size, DMA_TO_DEVICE); if (dma_mapping_error(dev, ctx_p->user.key_dma_addr)) { - dev_err(dev, "Mapping Key %u B at va=%pK for DMA failed\n", + dev_err(dev, "Mapping Key %u B at va=%p for DMA failed\n", max_key_buf_size, ctx_p->user.key); goto free_key; } - dev_dbg(dev, "Mapped key %u B at va=%pK to dma=%pad\n", + dev_dbg(dev, "Mapped key %u B at va=%p to dma=%pad\n", max_key_buf_size, ctx_p->user.key, &ctx_p->user.key_dma_addr); return 0; diff --git a/drivers/crypto/ccree/cc_hash.c b/drivers/crypto/ccree/cc_hash.c index d0612bec4d58d..c6d085c8ff797 100644 --- a/drivers/crypto/ccree/cc_hash.c +++ b/drivers/crypto/ccree/cc_hash.c @@ -125,7 +125,7 @@ static int cc_map_result(struct device *dev, struct ahash_req_ctx *state, digestsize); return -ENOMEM; } - dev_dbg(dev, "Mapped digest result buffer %u B at va=%pK to dma=%pad\n", + dev_dbg(dev, "Mapped digest result buffer %u B at va=%p to dma=%pad\n", digestsize, state->digest_result_buff, &state->digest_result_dma_addr); @@ -184,11 +184,11 @@ static int cc_map_req(struct device *dev, struct ahash_req_ctx *state, dma_map_single(dev, state->digest_buff, ctx->inter_digestsize, DMA_BIDIRECTIONAL); if (dma_mapping_error(dev, state->digest_buff_dma_addr)) { - dev_err(dev, "Mapping digest len %d B at va=%pK for DMA failed\n", + dev_err(dev, "Mapping digest len %d B at va=%p for DMA failed\n", ctx->inter_digestsize, state->digest_buff); return -EINVAL; } - dev_dbg(dev, "Mapped digest %d B at va=%pK to dma=%pad\n", + dev_dbg(dev, "Mapped digest %d B at va=%p to dma=%pad\n", ctx->inter_digestsize, state->digest_buff, &state->digest_buff_dma_addr); @@ -197,11 +197,11 @@ static int cc_map_req(struct device *dev, struct ahash_req_ctx *state, dma_map_single(dev, state->digest_bytes_len, HASH_MAX_LEN_SIZE, DMA_BIDIRECTIONAL); if (dma_mapping_error(dev, state->digest_bytes_len_dma_addr)) { - dev_err(dev, "Mapping digest len %u B at va=%pK for DMA failed\n", + dev_err(dev, "Mapping digest len %u B at va=%p for DMA failed\n", HASH_MAX_LEN_SIZE, state->digest_bytes_len); goto unmap_digest_buf; } - dev_dbg(dev, "Mapped digest len %u B at va=%pK to dma=%pad\n", + dev_dbg(dev, "Mapped digest len %u B at va=%p to dma=%pad\n", HASH_MAX_LEN_SIZE, state->digest_bytes_len, &state->digest_bytes_len_dma_addr); } @@ -212,12 +212,12 @@ static int cc_map_req(struct device *dev, struct ahash_req_ctx *state, ctx->inter_digestsize, DMA_BIDIRECTIONAL); if (dma_mapping_error(dev, state->opad_digest_dma_addr)) { - dev_err(dev, "Mapping opad digest %d B at va=%pK for DMA failed\n", + dev_err(dev, "Mapping opad digest %d B at va=%p for DMA failed\n", ctx->inter_digestsize, state->opad_digest_buff); goto unmap_digest_len; } - dev_dbg(dev, "Mapped opad digest %d B at va=%pK to dma=%pad\n", + dev_dbg(dev, "Mapped opad digest %d B at va=%p to dma=%pad\n", ctx->inter_digestsize, state->opad_digest_buff, &state->opad_digest_dma_addr); } @@ -272,7 +272,7 @@ static void cc_unmap_result(struct device *dev, struct ahash_req_ctx *state, if (state->digest_result_dma_addr) { dma_unmap_single(dev, state->digest_result_dma_addr, digestsize, DMA_BIDIRECTIONAL); - dev_dbg(dev, "unmpa digest result buffer va (%pK) pa (%pad) len %u\n", + dev_dbg(dev, "unmpa digest result buffer va (%p) pa (%pad) len %u\n", state->digest_result_buff, &state->digest_result_dma_addr, digestsize); memcpy(result, state->digest_result_buff, digestsize); @@ -287,7 +287,7 @@ static void cc_update_complete(struct device *dev, void *cc_req, int err) struct crypto_ahash *tfm = crypto_ahash_reqtfm(req); struct cc_hash_ctx *ctx = crypto_ahash_ctx_dma(tfm); - dev_dbg(dev, "req=%pK\n", req); + dev_dbg(dev, "req=%p\n", req); if (err != -EINPROGRESS) { /* Not a BACKLOG notification */ @@ -306,7 +306,7 @@ static void cc_digest_complete(struct device *dev, void *cc_req, int err) struct cc_hash_ctx *ctx = crypto_ahash_ctx_dma(tfm); u32 digestsize = crypto_ahash_digestsize(tfm); - dev_dbg(dev, "req=%pK\n", req); + dev_dbg(dev, "req=%p\n", req); if (err != -EINPROGRESS) { /* Not a BACKLOG notification */ @@ -326,7 +326,7 @@ static void cc_hash_complete(struct device *dev, void *cc_req, int err) struct cc_hash_ctx *ctx = crypto_ahash_ctx_dma(tfm); u32 digestsize = crypto_ahash_digestsize(tfm); - dev_dbg(dev, "req=%pK\n", req); + dev_dbg(dev, "req=%p\n", req); if (err != -EINPROGRESS) { /* Not a BACKLOG notification */ @@ -1077,11 +1077,11 @@ static int cc_alloc_ctx(struct cc_hash_ctx *ctx) dma_map_single(dev, ctx->digest_buff, sizeof(ctx->digest_buff), DMA_BIDIRECTIONAL); if (dma_mapping_error(dev, ctx->digest_buff_dma_addr)) { - dev_err(dev, "Mapping digest len %zu B at va=%pK for DMA failed\n", + dev_err(dev, "Mapping digest len %zu B at va=%p for DMA failed\n", sizeof(ctx->digest_buff), ctx->digest_buff); goto fail; } - dev_dbg(dev, "Mapped digest %zu B at va=%pK to dma=%pad\n", + dev_dbg(dev, "Mapped digest %zu B at va=%p to dma=%pad\n", sizeof(ctx->digest_buff), ctx->digest_buff, &ctx->digest_buff_dma_addr); @@ -1090,12 +1090,12 @@ static int cc_alloc_ctx(struct cc_hash_ctx *ctx) sizeof(ctx->opad_tmp_keys_buff), DMA_BIDIRECTIONAL); if (dma_mapping_error(dev, ctx->opad_tmp_keys_dma_addr)) { - dev_err(dev, "Mapping opad digest %zu B at va=%pK for DMA failed\n", + dev_err(dev, "Mapping opad digest %zu B at va=%p for DMA failed\n", sizeof(ctx->opad_tmp_keys_buff), ctx->opad_tmp_keys_buff); goto fail; } - dev_dbg(dev, "Mapped opad_tmp_keys %zu B at va=%pK to dma=%pad\n", + dev_dbg(dev, "Mapped opad_tmp_keys %zu B at va=%p to dma=%pad\n", sizeof(ctx->opad_tmp_keys_buff), ctx->opad_tmp_keys_buff, &ctx->opad_tmp_keys_dma_addr); diff --git a/drivers/crypto/ccree/cc_pm.c b/drivers/crypto/ccree/cc_pm.c index 6124fbbbed946..bbd118f8de0ea 100644 --- a/drivers/crypto/ccree/cc_pm.c +++ b/drivers/crypto/ccree/cc_pm.c @@ -77,6 +77,5 @@ int cc_pm_get(struct device *dev) void cc_pm_put_suspend(struct device *dev) { - pm_runtime_mark_last_busy(dev); pm_runtime_put_autosuspend(dev); } diff --git a/drivers/crypto/hisilicon/hpre/hpre_crypto.c b/drivers/crypto/hisilicon/hpre/hpre_crypto.c index 61b5e1c5d0192..1550c3818383a 100644 --- a/drivers/crypto/hisilicon/hpre/hpre_crypto.c +++ b/drivers/crypto/hisilicon/hpre/hpre_crypto.c @@ -1491,11 +1491,13 @@ static void hpre_ecdh_cb(struct hpre_ctx *ctx, void *resp) if (overtime_thrhld && hpre_is_bd_timeout(req, overtime_thrhld)) atomic64_inc(&dfx[HPRE_OVER_THRHLD_CNT].value); + /* Do unmap before data processing */ + hpre_ecdh_hw_data_clr_all(ctx, req, areq->dst, areq->src); + p = sg_virt(areq->dst); memmove(p, p + ctx->key_sz - curve_sz, curve_sz); memmove(p + curve_sz, p + areq->dst_len - curve_sz, curve_sz); - hpre_ecdh_hw_data_clr_all(ctx, req, areq->dst, areq->src); kpp_request_complete(areq, ret); atomic64_inc(&dfx[HPRE_RECV_CNT].value); @@ -1808,9 +1810,11 @@ static void hpre_curve25519_cb(struct hpre_ctx *ctx, void *resp) if (overtime_thrhld && hpre_is_bd_timeout(req, overtime_thrhld)) atomic64_inc(&dfx[HPRE_OVER_THRHLD_CNT].value); + /* Do unmap before data processing */ + hpre_curve25519_hw_data_clr_all(ctx, req, areq->dst, areq->src); + hpre_key_to_big_end(sg_virt(areq->dst), CURVE25519_KEY_SIZE); - hpre_curve25519_hw_data_clr_all(ctx, req, areq->dst, areq->src); kpp_request_complete(areq, ret); atomic64_inc(&dfx[HPRE_RECV_CNT].value); diff --git a/drivers/crypto/hisilicon/qm.c b/drivers/crypto/hisilicon/qm.c index 7c41f9593d039..2e4ee7ecfdfbb 100644 --- a/drivers/crypto/hisilicon/qm.c +++ b/drivers/crypto/hisilicon/qm.c @@ -912,7 +912,6 @@ static void qm_pm_put_sync(struct hisi_qm *qm) if (!test_bit(QM_SUPPORT_RPM, &qm->caps)) return; - pm_runtime_mark_last_busy(dev); pm_runtime_put_autosuspend(dev); } diff --git a/drivers/crypto/hisilicon/sec2/sec.h b/drivers/crypto/hisilicon/sec2/sec.h index 703920b49c7c0..81d0beda93b2b 100644 --- a/drivers/crypto/hisilicon/sec2/sec.h +++ b/drivers/crypto/hisilicon/sec2/sec.h @@ -7,6 +7,12 @@ #include <linux/hisi_acc_qm.h> #include "sec_crypto.h" +#define SEC_PBUF_SZ 512 +#define SEC_MAX_MAC_LEN 64 +#define SEC_IV_SIZE 24 +#define SEC_SGE_NR_NUM 4 +#define SEC_SGL_ALIGN_SIZE 64 + /* Algorithm resource per hardware SEC queue */ struct sec_alg_res { u8 *pbuf; @@ -20,6 +26,40 @@ struct sec_alg_res { u16 depth; }; +struct sec_hw_sge { + dma_addr_t buf; + void *page_ctrl; + __le32 len; + __le32 pad; + __le32 pad0; + __le32 pad1; +}; + +struct sec_hw_sgl { + dma_addr_t next_dma; + __le16 entry_sum_in_chain; + __le16 entry_sum_in_sgl; + __le16 entry_length_in_sgl; + __le16 pad0; + __le64 pad1[5]; + struct sec_hw_sgl *next; + struct sec_hw_sge sge_entries[SEC_SGE_NR_NUM]; +} __aligned(SEC_SGL_ALIGN_SIZE); + +struct sec_src_dst_buf { + struct sec_hw_sgl in; + struct sec_hw_sgl out; +}; + +struct sec_request_buf { + union { + struct sec_src_dst_buf data_buf; + __u8 pbuf[SEC_PBUF_SZ]; + }; + dma_addr_t in_dma; + dma_addr_t out_dma; +}; + /* Cipher request of SEC private */ struct sec_cipher_req { struct hisi_acc_hw_sgl *c_out; @@ -29,6 +69,7 @@ struct sec_cipher_req { struct skcipher_request *sk_req; u32 c_len; bool encrypt; + __u8 c_ivin_buf[SEC_IV_SIZE]; }; struct sec_aead_req { @@ -37,6 +78,13 @@ struct sec_aead_req { u8 *a_ivin; dma_addr_t a_ivin_dma; struct aead_request *aead_req; + __u8 a_ivin_buf[SEC_IV_SIZE]; + __u8 out_mac_buf[SEC_MAX_MAC_LEN]; +}; + +struct sec_instance_backlog { + struct list_head list; + spinlock_t lock; }; /* SEC request of Crypto */ @@ -55,15 +103,17 @@ struct sec_req { dma_addr_t in_dma; struct sec_cipher_req c_req; struct sec_aead_req aead_req; - struct list_head backlog_head; + struct crypto_async_request *base; int err_type; int req_id; u32 flag; - /* Status of the SEC request */ - bool fake_busy; bool use_pbuf; + + struct list_head list; + struct sec_instance_backlog *backlog; + struct sec_request_buf buf; }; /** @@ -119,9 +169,11 @@ struct sec_qp_ctx { struct sec_alg_res *res; struct sec_ctx *ctx; spinlock_t req_lock; - struct list_head backlog; + spinlock_t id_lock; struct hisi_acc_sgl_pool *c_in_pool; struct hisi_acc_sgl_pool *c_out_pool; + struct sec_instance_backlog backlog; + u16 send_head; }; enum sec_alg_type { @@ -139,9 +191,6 @@ struct sec_ctx { /* Half queues for encipher, and half for decipher */ u32 hlf_q_num; - /* Threshold for fake busy, trigger to return -EBUSY to user */ - u32 fake_req_limit; - /* Current cyclic index to select a queue for encipher */ atomic_t enc_qcyclic; diff --git a/drivers/crypto/hisilicon/sec2/sec_crypto.c b/drivers/crypto/hisilicon/sec2/sec_crypto.c index 8ea5305bc320f..d044ded0f2904 100644 --- a/drivers/crypto/hisilicon/sec2/sec_crypto.c +++ b/drivers/crypto/hisilicon/sec2/sec_crypto.c @@ -67,7 +67,6 @@ #define SEC_MAX_CCM_AAD_LEN 65279 #define SEC_TOTAL_MAC_SZ(depth) (SEC_MAX_MAC_LEN * (depth)) -#define SEC_PBUF_SZ 512 #define SEC_PBUF_IV_OFFSET SEC_PBUF_SZ #define SEC_PBUF_MAC_OFFSET (SEC_PBUF_SZ + SEC_IV_SIZE) #define SEC_PBUF_PKG (SEC_PBUF_SZ + SEC_IV_SIZE + \ @@ -102,6 +101,8 @@ #define IV_LAST_BYTE_MASK 0xFF #define IV_CTR_INIT 0x1 #define IV_BYTE_OFFSET 0x8 +#define SEC_GCM_MIN_AUTH_SZ 0x8 +#define SEC_RETRY_MAX_CNT 5U static DEFINE_MUTEX(sec_algs_lock); static unsigned int sec_available_devs; @@ -116,40 +117,19 @@ struct sec_aead { struct aead_alg alg; }; -/* Get an en/de-cipher queue cyclically to balance load over queues of TFM */ -static inline u32 sec_alloc_queue_id(struct sec_ctx *ctx, struct sec_req *req) -{ - if (req->c_req.encrypt) - return (u32)atomic_inc_return(&ctx->enc_qcyclic) % - ctx->hlf_q_num; - - return (u32)atomic_inc_return(&ctx->dec_qcyclic) % ctx->hlf_q_num + - ctx->hlf_q_num; -} - -static inline void sec_free_queue_id(struct sec_ctx *ctx, struct sec_req *req) -{ - if (req->c_req.encrypt) - atomic_dec(&ctx->enc_qcyclic); - else - atomic_dec(&ctx->dec_qcyclic); -} +static int sec_aead_soft_crypto(struct sec_ctx *ctx, + struct aead_request *aead_req, + bool encrypt); +static int sec_skcipher_soft_crypto(struct sec_ctx *ctx, + struct skcipher_request *sreq, bool encrypt); static int sec_alloc_req_id(struct sec_req *req, struct sec_qp_ctx *qp_ctx) { int req_id; - spin_lock_bh(&qp_ctx->req_lock); + spin_lock_bh(&qp_ctx->id_lock); req_id = idr_alloc_cyclic(&qp_ctx->req_idr, NULL, 0, qp_ctx->qp->sq_depth, GFP_ATOMIC); - spin_unlock_bh(&qp_ctx->req_lock); - if (unlikely(req_id < 0)) { - dev_err(req->ctx->dev, "alloc req id fail!\n"); - return req_id; - } - - req->qp_ctx = qp_ctx; - qp_ctx->req_list[req_id] = req; - + spin_unlock_bh(&qp_ctx->id_lock); return req_id; } @@ -163,12 +143,9 @@ static void sec_free_req_id(struct sec_req *req) return; } - qp_ctx->req_list[req_id] = NULL; - req->qp_ctx = NULL; - - spin_lock_bh(&qp_ctx->req_lock); + spin_lock_bh(&qp_ctx->id_lock); idr_remove(&qp_ctx->req_idr, req_id); - spin_unlock_bh(&qp_ctx->req_lock); + spin_unlock_bh(&qp_ctx->id_lock); } static u8 pre_parse_finished_bd(struct bd_status *status, void *resp) @@ -229,6 +206,90 @@ static int sec_cb_status_check(struct sec_req *req, return 0; } +static int qp_send_message(struct sec_req *req) +{ + struct sec_qp_ctx *qp_ctx = req->qp_ctx; + int ret; + + if (atomic_read(&qp_ctx->qp->qp_status.used) == qp_ctx->qp->sq_depth - 1) + return -EBUSY; + + spin_lock_bh(&qp_ctx->req_lock); + if (atomic_read(&qp_ctx->qp->qp_status.used) == qp_ctx->qp->sq_depth - 1) { + spin_unlock_bh(&qp_ctx->req_lock); + return -EBUSY; + } + + if (qp_ctx->ctx->type_supported == SEC_BD_TYPE2) { + req->sec_sqe.type2.tag = cpu_to_le16((u16)qp_ctx->send_head); + qp_ctx->req_list[qp_ctx->send_head] = req; + } + + ret = hisi_qp_send(qp_ctx->qp, &req->sec_sqe); + if (ret) { + spin_unlock_bh(&qp_ctx->req_lock); + return ret; + } + if (qp_ctx->ctx->type_supported == SEC_BD_TYPE2) + qp_ctx->send_head = (qp_ctx->send_head + 1) % qp_ctx->qp->sq_depth; + + spin_unlock_bh(&qp_ctx->req_lock); + + atomic64_inc(&req->ctx->sec->debug.dfx.send_cnt); + return -EINPROGRESS; +} + +static void sec_alg_send_backlog_soft(struct sec_ctx *ctx, struct sec_qp_ctx *qp_ctx) +{ + struct sec_req *req, *tmp; + int ret; + + list_for_each_entry_safe(req, tmp, &qp_ctx->backlog.list, list) { + list_del(&req->list); + ctx->req_op->buf_unmap(ctx, req); + if (req->req_id >= 0) + sec_free_req_id(req); + + if (ctx->alg_type == SEC_AEAD) + ret = sec_aead_soft_crypto(ctx, req->aead_req.aead_req, + req->c_req.encrypt); + else + ret = sec_skcipher_soft_crypto(ctx, req->c_req.sk_req, + req->c_req.encrypt); + + /* Wake up the busy thread first, then return the errno. */ + crypto_request_complete(req->base, -EINPROGRESS); + crypto_request_complete(req->base, ret); + } +} + +static void sec_alg_send_backlog(struct sec_ctx *ctx, struct sec_qp_ctx *qp_ctx) +{ + struct sec_req *req, *tmp; + int ret; + + spin_lock_bh(&qp_ctx->backlog.lock); + list_for_each_entry_safe(req, tmp, &qp_ctx->backlog.list, list) { + ret = qp_send_message(req); + switch (ret) { + case -EINPROGRESS: + list_del(&req->list); + crypto_request_complete(req->base, -EINPROGRESS); + break; + case -EBUSY: + /* Device is busy and stop send any request. */ + goto unlock; + default: + /* Release memory resources and send all requests through software. */ + sec_alg_send_backlog_soft(ctx, qp_ctx); + goto unlock; + } + } + +unlock: + spin_unlock_bh(&qp_ctx->backlog.lock); +} + static void sec_req_cb(struct hisi_qp *qp, void *resp) { struct sec_qp_ctx *qp_ctx = qp->qp_ctx; @@ -273,40 +334,54 @@ static void sec_req_cb(struct hisi_qp *qp, void *resp) ctx->req_op->callback(ctx, req, err); } -static int sec_bd_send(struct sec_ctx *ctx, struct sec_req *req) +static int sec_alg_send_message_retry(struct sec_req *req) { - struct sec_qp_ctx *qp_ctx = req->qp_ctx; + int ctr = 0; int ret; - if (ctx->fake_req_limit <= - atomic_read(&qp_ctx->qp->qp_status.used) && - !(req->flag & CRYPTO_TFM_REQ_MAY_BACKLOG)) - return -EBUSY; + do { + ret = qp_send_message(req); + } while (ret == -EBUSY && ctr++ < SEC_RETRY_MAX_CNT); - spin_lock_bh(&qp_ctx->req_lock); - ret = hisi_qp_send(qp_ctx->qp, &req->sec_sqe); - if (ctx->fake_req_limit <= - atomic_read(&qp_ctx->qp->qp_status.used) && !ret) { - list_add_tail(&req->backlog_head, &qp_ctx->backlog); - atomic64_inc(&ctx->sec->debug.dfx.send_cnt); - atomic64_inc(&ctx->sec->debug.dfx.send_busy_cnt); - spin_unlock_bh(&qp_ctx->req_lock); + return ret; +} + +static int sec_alg_try_enqueue(struct sec_req *req) +{ + /* Check if any request is already backlogged */ + if (!list_empty(&req->backlog->list)) return -EBUSY; - } - spin_unlock_bh(&qp_ctx->req_lock); - if (unlikely(ret == -EBUSY)) - return -ENOBUFS; + /* Try to enqueue to HW ring */ + return qp_send_message(req); +} - if (likely(!ret)) { - ret = -EINPROGRESS; - atomic64_inc(&ctx->sec->debug.dfx.send_cnt); - } + +static int sec_alg_send_message_maybacklog(struct sec_req *req) +{ + int ret; + + ret = sec_alg_try_enqueue(req); + if (ret != -EBUSY) + return ret; + + spin_lock_bh(&req->backlog->lock); + ret = sec_alg_try_enqueue(req); + if (ret == -EBUSY) + list_add_tail(&req->list, &req->backlog->list); + spin_unlock_bh(&req->backlog->lock); return ret; } -/* Get DMA memory resources */ +static int sec_bd_send(struct sec_ctx *ctx, struct sec_req *req) +{ + if (req->flag & CRYPTO_TFM_REQ_MAY_BACKLOG) + return sec_alg_send_message_maybacklog(req); + + return sec_alg_send_message_retry(req); +} + static int sec_alloc_civ_resource(struct device *dev, struct sec_alg_res *res) { u16 q_depth = res->depth; @@ -558,7 +633,10 @@ static int sec_create_qp_ctx(struct sec_ctx *ctx, int qp_ctx_id) spin_lock_init(&qp_ctx->req_lock); idr_init(&qp_ctx->req_idr); - INIT_LIST_HEAD(&qp_ctx->backlog); + spin_lock_init(&qp_ctx->backlog.lock); + spin_lock_init(&qp_ctx->id_lock); + INIT_LIST_HEAD(&qp_ctx->backlog.list); + qp_ctx->send_head = 0; ret = sec_alloc_qp_ctx_resource(ctx, qp_ctx); if (ret) @@ -602,9 +680,6 @@ static int sec_ctx_base_init(struct sec_ctx *ctx) ctx->hlf_q_num = sec->ctx_q_num >> 1; ctx->pbuf_supported = ctx->sec->iommu_used; - - /* Half of queue depth is taken as fake requests limit in the queue. */ - ctx->fake_req_limit = ctx->qps[0]->sq_depth >> 1; ctx->qp_ctx = kcalloc(sec->ctx_q_num, sizeof(struct sec_qp_ctx), GFP_KERNEL); if (!ctx->qp_ctx) { @@ -706,7 +781,7 @@ static int sec_skcipher_init(struct crypto_skcipher *tfm) int ret; ctx->alg_type = SEC_SKCIPHER; - crypto_skcipher_set_reqsize(tfm, sizeof(struct sec_req)); + crypto_skcipher_set_reqsize_dma(tfm, sizeof(struct sec_req)); ctx->c_ctx.ivsize = crypto_skcipher_ivsize(tfm); if (ctx->c_ctx.ivsize > SEC_IV_SIZE) { pr_err("get error skcipher iv size!\n"); @@ -883,24 +958,25 @@ GEN_SEC_SETKEY_FUNC(sm4_ctr, SEC_CALG_SM4, SEC_CMODE_CTR) static int sec_cipher_pbuf_map(struct sec_ctx *ctx, struct sec_req *req, struct scatterlist *src) { - struct sec_aead_req *a_req = &req->aead_req; - struct aead_request *aead_req = a_req->aead_req; + struct aead_request *aead_req = req->aead_req.aead_req; struct sec_cipher_req *c_req = &req->c_req; struct sec_qp_ctx *qp_ctx = req->qp_ctx; + struct sec_request_buf *buf = &req->buf; struct device *dev = ctx->dev; int copy_size, pbuf_length; int req_id = req->req_id; struct crypto_aead *tfm; + u8 *mac_offset, *pbuf; size_t authsize; - u8 *mac_offset; if (ctx->alg_type == SEC_AEAD) copy_size = aead_req->cryptlen + aead_req->assoclen; else copy_size = c_req->c_len; - pbuf_length = sg_copy_to_buffer(src, sg_nents(src), - qp_ctx->res[req_id].pbuf, copy_size); + + pbuf = req->req_id < 0 ? buf->pbuf : qp_ctx->res[req_id].pbuf; + pbuf_length = sg_copy_to_buffer(src, sg_nents(src), pbuf, copy_size); if (unlikely(pbuf_length != copy_size)) { dev_err(dev, "copy src data to pbuf error!\n"); return -EINVAL; @@ -908,8 +984,17 @@ static int sec_cipher_pbuf_map(struct sec_ctx *ctx, struct sec_req *req, if (!c_req->encrypt && ctx->alg_type == SEC_AEAD) { tfm = crypto_aead_reqtfm(aead_req); authsize = crypto_aead_authsize(tfm); - mac_offset = qp_ctx->res[req_id].pbuf + copy_size - authsize; - memcpy(a_req->out_mac, mac_offset, authsize); + mac_offset = pbuf + copy_size - authsize; + memcpy(req->aead_req.out_mac, mac_offset, authsize); + } + + if (req->req_id < 0) { + buf->in_dma = dma_map_single(dev, buf->pbuf, SEC_PBUF_SZ, DMA_BIDIRECTIONAL); + if (unlikely(dma_mapping_error(dev, buf->in_dma))) + return -ENOMEM; + + buf->out_dma = buf->in_dma; + return 0; } req->in_dma = qp_ctx->res[req_id].pbuf_dma; @@ -924,6 +1009,7 @@ static void sec_cipher_pbuf_unmap(struct sec_ctx *ctx, struct sec_req *req, struct aead_request *aead_req = req->aead_req.aead_req; struct sec_cipher_req *c_req = &req->c_req; struct sec_qp_ctx *qp_ctx = req->qp_ctx; + struct sec_request_buf *buf = &req->buf; int copy_size, pbuf_length; int req_id = req->req_id; @@ -932,10 +1018,16 @@ static void sec_cipher_pbuf_unmap(struct sec_ctx *ctx, struct sec_req *req, else copy_size = c_req->c_len; - pbuf_length = sg_copy_from_buffer(dst, sg_nents(dst), - qp_ctx->res[req_id].pbuf, copy_size); + if (req->req_id < 0) + pbuf_length = sg_copy_from_buffer(dst, sg_nents(dst), buf->pbuf, copy_size); + else + pbuf_length = sg_copy_from_buffer(dst, sg_nents(dst), qp_ctx->res[req_id].pbuf, + copy_size); if (unlikely(pbuf_length != copy_size)) dev_err(ctx->dev, "copy pbuf data to dst error!\n"); + + if (req->req_id < 0) + dma_unmap_single(ctx->dev, buf->in_dma, SEC_PBUF_SZ, DMA_BIDIRECTIONAL); } static int sec_aead_mac_init(struct sec_aead_req *req) @@ -957,14 +1049,95 @@ static int sec_aead_mac_init(struct sec_aead_req *req) return 0; } -static int sec_cipher_map(struct sec_ctx *ctx, struct sec_req *req, - struct scatterlist *src, struct scatterlist *dst) +static void fill_sg_to_hw_sge(struct scatterlist *sgl, struct sec_hw_sge *hw_sge) +{ + hw_sge->buf = sg_dma_address(sgl); + hw_sge->len = cpu_to_le32(sg_dma_len(sgl)); + hw_sge->page_ctrl = sg_virt(sgl); +} + +static int sec_cipher_to_hw_sgl(struct device *dev, struct scatterlist *src, + struct sec_hw_sgl *src_in, dma_addr_t *hw_sgl_dma, + int dma_dir) +{ + struct sec_hw_sge *curr_hw_sge = src_in->sge_entries; + u32 i, sg_n, sg_n_mapped; + struct scatterlist *sg; + u32 sge_var = 0; + + sg_n = sg_nents(src); + sg_n_mapped = dma_map_sg(dev, src, sg_n, dma_dir); + if (unlikely(!sg_n_mapped)) { + dev_err(dev, "dma mapping for SG error!\n"); + return -EINVAL; + } else if (unlikely(sg_n_mapped > SEC_SGE_NR_NUM)) { + dev_err(dev, "the number of entries in input scatterlist error!\n"); + dma_unmap_sg(dev, src, sg_n, dma_dir); + return -EINVAL; + } + + for_each_sg(src, sg, sg_n_mapped, i) { + fill_sg_to_hw_sge(sg, curr_hw_sge); + curr_hw_sge++; + sge_var++; + } + + src_in->entry_sum_in_sgl = cpu_to_le16(sge_var); + src_in->entry_sum_in_chain = cpu_to_le16(SEC_SGE_NR_NUM); + src_in->entry_length_in_sgl = cpu_to_le16(SEC_SGE_NR_NUM); + *hw_sgl_dma = dma_map_single(dev, src_in, sizeof(struct sec_hw_sgl), dma_dir); + if (unlikely(dma_mapping_error(dev, *hw_sgl_dma))) { + dma_unmap_sg(dev, src, sg_n, dma_dir); + return -ENOMEM; + } + + return 0; +} + +static void sec_cipher_put_hw_sgl(struct device *dev, struct scatterlist *src, + dma_addr_t src_in, int dma_dir) +{ + dma_unmap_single(dev, src_in, sizeof(struct sec_hw_sgl), dma_dir); + dma_unmap_sg(dev, src, sg_nents(src), dma_dir); +} + +static int sec_cipher_map_sgl(struct device *dev, struct sec_req *req, + struct scatterlist *src, struct scatterlist *dst) +{ + struct sec_hw_sgl *src_in = &req->buf.data_buf.in; + struct sec_hw_sgl *dst_out = &req->buf.data_buf.out; + int ret; + + if (dst == src) { + ret = sec_cipher_to_hw_sgl(dev, src, src_in, &req->buf.in_dma, + DMA_BIDIRECTIONAL); + req->buf.out_dma = req->buf.in_dma; + return ret; + } + + ret = sec_cipher_to_hw_sgl(dev, src, src_in, &req->buf.in_dma, DMA_TO_DEVICE); + if (unlikely(ret)) + return ret; + + ret = sec_cipher_to_hw_sgl(dev, dst, dst_out, &req->buf.out_dma, + DMA_FROM_DEVICE); + if (unlikely(ret)) { + sec_cipher_put_hw_sgl(dev, src, req->buf.in_dma, DMA_TO_DEVICE); + return ret; + } + + return 0; +} + +static int sec_cipher_map_inner(struct sec_ctx *ctx, struct sec_req *req, + struct scatterlist *src, struct scatterlist *dst) { struct sec_cipher_req *c_req = &req->c_req; struct sec_aead_req *a_req = &req->aead_req; struct sec_qp_ctx *qp_ctx = req->qp_ctx; struct sec_alg_res *res = &qp_ctx->res[req->req_id]; struct device *dev = ctx->dev; + enum dma_data_direction src_direction; int ret; if (req->use_pbuf) { @@ -977,10 +1150,9 @@ static int sec_cipher_map(struct sec_ctx *ctx, struct sec_req *req, a_req->out_mac_dma = res->pbuf_dma + SEC_PBUF_MAC_OFFSET; } - ret = sec_cipher_pbuf_map(ctx, req, src); - - return ret; + return sec_cipher_pbuf_map(ctx, req, src); } + c_req->c_ivin = res->c_ivin; c_req->c_ivin_dma = res->c_ivin_dma; if (ctx->alg_type == SEC_AEAD) { @@ -990,10 +1162,11 @@ static int sec_cipher_map(struct sec_ctx *ctx, struct sec_req *req, a_req->out_mac_dma = res->out_mac_dma; } + src_direction = dst == src ? DMA_BIDIRECTIONAL : DMA_TO_DEVICE; req->in = hisi_acc_sg_buf_map_to_hw_sgl(dev, src, qp_ctx->c_in_pool, req->req_id, - &req->in_dma); + &req->in_dma, src_direction); if (IS_ERR(req->in)) { dev_err(dev, "fail to dma map input sgl buffers!\n"); return PTR_ERR(req->in); @@ -1003,7 +1176,7 @@ static int sec_cipher_map(struct sec_ctx *ctx, struct sec_req *req, ret = sec_aead_mac_init(a_req); if (unlikely(ret)) { dev_err(dev, "fail to init mac data for ICV!\n"); - hisi_acc_sg_buf_unmap(dev, src, req->in); + hisi_acc_sg_buf_unmap(dev, src, req->in, src_direction); return ret; } } @@ -1015,11 +1188,12 @@ static int sec_cipher_map(struct sec_ctx *ctx, struct sec_req *req, c_req->c_out = hisi_acc_sg_buf_map_to_hw_sgl(dev, dst, qp_ctx->c_out_pool, req->req_id, - &c_req->c_out_dma); + &c_req->c_out_dma, + DMA_FROM_DEVICE); if (IS_ERR(c_req->c_out)) { dev_err(dev, "fail to dma map output sgl buffers!\n"); - hisi_acc_sg_buf_unmap(dev, src, req->in); + hisi_acc_sg_buf_unmap(dev, src, req->in, src_direction); return PTR_ERR(c_req->c_out); } } @@ -1027,19 +1201,108 @@ static int sec_cipher_map(struct sec_ctx *ctx, struct sec_req *req, return 0; } +static int sec_cipher_map(struct sec_ctx *ctx, struct sec_req *req, + struct scatterlist *src, struct scatterlist *dst) +{ + struct sec_aead_req *a_req = &req->aead_req; + struct sec_cipher_req *c_req = &req->c_req; + bool is_aead = (ctx->alg_type == SEC_AEAD); + struct device *dev = ctx->dev; + int ret = -ENOMEM; + + if (req->req_id >= 0) + return sec_cipher_map_inner(ctx, req, src, dst); + + c_req->c_ivin = c_req->c_ivin_buf; + c_req->c_ivin_dma = dma_map_single(dev, c_req->c_ivin, + SEC_IV_SIZE, DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(dev, c_req->c_ivin_dma))) + return -ENOMEM; + + if (is_aead) { + a_req->a_ivin = a_req->a_ivin_buf; + a_req->out_mac = a_req->out_mac_buf; + a_req->a_ivin_dma = dma_map_single(dev, a_req->a_ivin, + SEC_IV_SIZE, DMA_TO_DEVICE); + if (unlikely(dma_mapping_error(dev, a_req->a_ivin_dma))) + goto free_c_ivin_dma; + + a_req->out_mac_dma = dma_map_single(dev, a_req->out_mac, + SEC_MAX_MAC_LEN, DMA_BIDIRECTIONAL); + if (unlikely(dma_mapping_error(dev, a_req->out_mac_dma))) + goto free_a_ivin_dma; + } + if (req->use_pbuf) { + ret = sec_cipher_pbuf_map(ctx, req, src); + if (unlikely(ret)) + goto free_out_mac_dma; + + return 0; + } + + if (!c_req->encrypt && is_aead) { + ret = sec_aead_mac_init(a_req); + if (unlikely(ret)) { + dev_err(dev, "fail to init mac data for ICV!\n"); + goto free_out_mac_dma; + } + } + + ret = sec_cipher_map_sgl(dev, req, src, dst); + if (unlikely(ret)) { + dev_err(dev, "fail to dma map input sgl buffers!\n"); + goto free_out_mac_dma; + } + + return 0; + +free_out_mac_dma: + if (is_aead) + dma_unmap_single(dev, a_req->out_mac_dma, SEC_MAX_MAC_LEN, DMA_BIDIRECTIONAL); +free_a_ivin_dma: + if (is_aead) + dma_unmap_single(dev, a_req->a_ivin_dma, SEC_IV_SIZE, DMA_TO_DEVICE); +free_c_ivin_dma: + dma_unmap_single(dev, c_req->c_ivin_dma, SEC_IV_SIZE, DMA_TO_DEVICE); + return ret; +} + static void sec_cipher_unmap(struct sec_ctx *ctx, struct sec_req *req, struct scatterlist *src, struct scatterlist *dst) { + struct sec_aead_req *a_req = &req->aead_req; struct sec_cipher_req *c_req = &req->c_req; struct device *dev = ctx->dev; + if (req->req_id >= 0) { + if (req->use_pbuf) { + sec_cipher_pbuf_unmap(ctx, req, dst); + } else { + if (dst != src) { + hisi_acc_sg_buf_unmap(dev, dst, c_req->c_out, DMA_FROM_DEVICE); + hisi_acc_sg_buf_unmap(dev, src, req->in, DMA_TO_DEVICE); + } else { + hisi_acc_sg_buf_unmap(dev, src, req->in, DMA_BIDIRECTIONAL); + } + } + return; + } + if (req->use_pbuf) { sec_cipher_pbuf_unmap(ctx, req, dst); } else { - if (dst != src) - hisi_acc_sg_buf_unmap(dev, src, req->in); + if (dst != src) { + sec_cipher_put_hw_sgl(dev, dst, req->buf.out_dma, DMA_FROM_DEVICE); + sec_cipher_put_hw_sgl(dev, src, req->buf.in_dma, DMA_TO_DEVICE); + } else { + sec_cipher_put_hw_sgl(dev, src, req->buf.in_dma, DMA_BIDIRECTIONAL); + } + } - hisi_acc_sg_buf_unmap(dev, dst, c_req->c_out); + dma_unmap_single(dev, c_req->c_ivin_dma, SEC_IV_SIZE, DMA_TO_DEVICE); + if (ctx->alg_type == SEC_AEAD) { + dma_unmap_single(dev, a_req->a_ivin_dma, SEC_IV_SIZE, DMA_TO_DEVICE); + dma_unmap_single(dev, a_req->out_mac_dma, SEC_MAX_MAC_LEN, DMA_BIDIRECTIONAL); } } @@ -1257,8 +1520,15 @@ static int sec_skcipher_bd_fill(struct sec_ctx *ctx, struct sec_req *req) sec_sqe->type2.c_key_addr = cpu_to_le64(c_ctx->c_key_dma); sec_sqe->type2.c_ivin_addr = cpu_to_le64(c_req->c_ivin_dma); - sec_sqe->type2.data_src_addr = cpu_to_le64(req->in_dma); - sec_sqe->type2.data_dst_addr = cpu_to_le64(c_req->c_out_dma); + if (req->req_id < 0) { + sec_sqe->type2.data_src_addr = cpu_to_le64(req->buf.in_dma); + sec_sqe->type2.data_dst_addr = cpu_to_le64(req->buf.out_dma); + } else { + sec_sqe->type2.data_src_addr = cpu_to_le64(req->in_dma); + sec_sqe->type2.data_dst_addr = cpu_to_le64(c_req->c_out_dma); + } + if (sec_sqe->type2.data_src_addr != sec_sqe->type2.data_dst_addr) + de = 0x1 << SEC_DE_OFFSET; sec_sqe->type2.icvw_kmode |= cpu_to_le16(((u16)c_ctx->c_mode) << SEC_CMODE_OFFSET); @@ -1284,13 +1554,10 @@ static int sec_skcipher_bd_fill(struct sec_ctx *ctx, struct sec_req *req) sec_sqe->sdm_addr_type |= da_type; scene = SEC_COMM_SCENE << SEC_SCENE_OFFSET; - if (req->in_dma != c_req->c_out_dma) - de = 0x1 << SEC_DE_OFFSET; sec_sqe->sds_sa_type = (de | scene | sa_type); sec_sqe->type2.clen_ivhlen |= cpu_to_le32(c_req->c_len); - sec_sqe->type2.tag = cpu_to_le16((u16)req->req_id); return 0; } @@ -1307,8 +1574,15 @@ static int sec_skcipher_bd_fill_v3(struct sec_ctx *ctx, struct sec_req *req) sec_sqe3->c_key_addr = cpu_to_le64(c_ctx->c_key_dma); sec_sqe3->no_scene.c_ivin_addr = cpu_to_le64(c_req->c_ivin_dma); - sec_sqe3->data_src_addr = cpu_to_le64(req->in_dma); - sec_sqe3->data_dst_addr = cpu_to_le64(c_req->c_out_dma); + if (req->req_id < 0) { + sec_sqe3->data_src_addr = cpu_to_le64(req->buf.in_dma); + sec_sqe3->data_dst_addr = cpu_to_le64(req->buf.out_dma); + } else { + sec_sqe3->data_src_addr = cpu_to_le64(req->in_dma); + sec_sqe3->data_dst_addr = cpu_to_le64(c_req->c_out_dma); + } + if (sec_sqe3->data_src_addr != sec_sqe3->data_dst_addr) + bd_param |= 0x1 << SEC_DE_OFFSET_V3; sec_sqe3->c_mode_alg = ((u8)c_ctx->c_alg << SEC_CALG_OFFSET_V3) | c_ctx->c_mode; @@ -1334,8 +1608,6 @@ static int sec_skcipher_bd_fill_v3(struct sec_ctx *ctx, struct sec_req *req) } bd_param |= SEC_COMM_SCENE << SEC_SCENE_OFFSET_V3; - if (req->in_dma != c_req->c_out_dma) - bd_param |= 0x1 << SEC_DE_OFFSET_V3; bd_param |= SEC_BD_TYPE3; sec_sqe3->bd_param = cpu_to_le32(bd_param); @@ -1367,15 +1639,12 @@ static void sec_update_iv(struct sec_req *req, enum sec_alg_type alg_type) size_t sz; u8 *iv; - if (req->c_req.encrypt) - sgl = alg_type == SEC_SKCIPHER ? sk_req->dst : aead_req->dst; - else - sgl = alg_type == SEC_SKCIPHER ? sk_req->src : aead_req->src; - if (alg_type == SEC_SKCIPHER) { + sgl = req->c_req.encrypt ? sk_req->dst : sk_req->src; iv = sk_req->iv; cryptlen = sk_req->cryptlen; } else { + sgl = req->c_req.encrypt ? aead_req->dst : aead_req->src; iv = aead_req->iv; cryptlen = aead_req->cryptlen; } @@ -1386,57 +1655,26 @@ static void sec_update_iv(struct sec_req *req, enum sec_alg_type alg_type) if (unlikely(sz != iv_size)) dev_err(req->ctx->dev, "copy output iv error!\n"); } else { - sz = cryptlen / iv_size; - if (cryptlen % iv_size) - sz += 1; + sz = (cryptlen + iv_size - 1) / iv_size; ctr_iv_inc(iv, iv_size, sz); } } -static struct sec_req *sec_back_req_clear(struct sec_ctx *ctx, - struct sec_qp_ctx *qp_ctx) -{ - struct sec_req *backlog_req = NULL; - - spin_lock_bh(&qp_ctx->req_lock); - if (ctx->fake_req_limit >= - atomic_read(&qp_ctx->qp->qp_status.used) && - !list_empty(&qp_ctx->backlog)) { - backlog_req = list_first_entry(&qp_ctx->backlog, - typeof(*backlog_req), backlog_head); - list_del(&backlog_req->backlog_head); - } - spin_unlock_bh(&qp_ctx->req_lock); - - return backlog_req; -} - static void sec_skcipher_callback(struct sec_ctx *ctx, struct sec_req *req, int err) { - struct skcipher_request *sk_req = req->c_req.sk_req; struct sec_qp_ctx *qp_ctx = req->qp_ctx; - struct skcipher_request *backlog_sk_req; - struct sec_req *backlog_req; - sec_free_req_id(req); + if (req->req_id >= 0) + sec_free_req_id(req); /* IV output at encrypto of CBC/CTR mode */ if (!err && (ctx->c_ctx.c_mode == SEC_CMODE_CBC || ctx->c_ctx.c_mode == SEC_CMODE_CTR) && req->c_req.encrypt) sec_update_iv(req, SEC_SKCIPHER); - while (1) { - backlog_req = sec_back_req_clear(ctx, qp_ctx); - if (!backlog_req) - break; - - backlog_sk_req = backlog_req->c_req.sk_req; - skcipher_request_complete(backlog_sk_req, -EINPROGRESS); - atomic64_inc(&ctx->sec->debug.dfx.recv_busy_cnt); - } - - skcipher_request_complete(sk_req, err); + crypto_request_complete(req->base, err); + sec_alg_send_backlog(ctx, qp_ctx); } static void set_aead_auth_iv(struct sec_ctx *ctx, struct sec_req *req) @@ -1675,21 +1913,14 @@ static void sec_aead_callback(struct sec_ctx *c, struct sec_req *req, int err) struct aead_request *a_req = req->aead_req.aead_req; struct crypto_aead *tfm = crypto_aead_reqtfm(a_req); size_t authsize = crypto_aead_authsize(tfm); - struct sec_aead_req *aead_req = &req->aead_req; - struct sec_cipher_req *c_req = &req->c_req; struct sec_qp_ctx *qp_ctx = req->qp_ctx; - struct aead_request *backlog_aead_req; - struct sec_req *backlog_req; size_t sz; - if (!err && c->c_ctx.c_mode == SEC_CMODE_CBC && c_req->encrypt) - sec_update_iv(req, SEC_AEAD); - - /* Copy output mac */ - if (!err && c_req->encrypt) { - struct scatterlist *sgl = a_req->dst; + if (!err && req->c_req.encrypt) { + if (c->c_ctx.c_mode == SEC_CMODE_CBC) + sec_update_iv(req, SEC_AEAD); - sz = sg_pcopy_from_buffer(sgl, sg_nents(sgl), aead_req->out_mac, + sz = sg_pcopy_from_buffer(a_req->dst, sg_nents(a_req->dst), req->aead_req.out_mac, authsize, a_req->cryptlen + a_req->assoclen); if (unlikely(sz != authsize)) { dev_err(c->dev, "copy out mac err!\n"); @@ -1697,48 +1928,39 @@ static void sec_aead_callback(struct sec_ctx *c, struct sec_req *req, int err) } } - sec_free_req_id(req); + if (req->req_id >= 0) + sec_free_req_id(req); - while (1) { - backlog_req = sec_back_req_clear(c, qp_ctx); - if (!backlog_req) - break; - - backlog_aead_req = backlog_req->aead_req.aead_req; - aead_request_complete(backlog_aead_req, -EINPROGRESS); - atomic64_inc(&c->sec->debug.dfx.recv_busy_cnt); - } - - aead_request_complete(a_req, err); + crypto_request_complete(req->base, err); + sec_alg_send_backlog(c, qp_ctx); } -static void sec_request_uninit(struct sec_ctx *ctx, struct sec_req *req) +static void sec_request_uninit(struct sec_req *req) { - sec_free_req_id(req); - sec_free_queue_id(ctx, req); + if (req->req_id >= 0) + sec_free_req_id(req); } static int sec_request_init(struct sec_ctx *ctx, struct sec_req *req) { struct sec_qp_ctx *qp_ctx; - int queue_id; - - /* To load balance */ - queue_id = sec_alloc_queue_id(ctx, req); - qp_ctx = &ctx->qp_ctx[queue_id]; + int i; - req->req_id = sec_alloc_req_id(req, qp_ctx); - if (unlikely(req->req_id < 0)) { - sec_free_queue_id(ctx, req); - return req->req_id; + for (i = 0; i < ctx->sec->ctx_q_num; i++) { + qp_ctx = &ctx->qp_ctx[i]; + req->req_id = sec_alloc_req_id(req, qp_ctx); + if (req->req_id >= 0) + break; } + req->qp_ctx = qp_ctx; + req->backlog = &qp_ctx->backlog; + return 0; } static int sec_process(struct sec_ctx *ctx, struct sec_req *req) { - struct sec_cipher_req *c_req = &req->c_req; int ret; ret = sec_request_init(ctx, req); @@ -1755,8 +1977,7 @@ static int sec_process(struct sec_ctx *ctx, struct sec_req *req) sec_update_iv(req, ctx->alg_type); ret = ctx->req_op->bd_send(ctx, req); - if (unlikely((ret != -EBUSY && ret != -EINPROGRESS) || - (ret == -EBUSY && !(req->flag & CRYPTO_TFM_REQ_MAY_BACKLOG)))) { + if (unlikely((ret != -EBUSY && ret != -EINPROGRESS))) { dev_err_ratelimited(ctx->dev, "send sec request failed!\n"); goto err_send_req; } @@ -1767,16 +1988,23 @@ err_send_req: /* As failing, restore the IV from user */ if (ctx->c_ctx.c_mode == SEC_CMODE_CBC && !req->c_req.encrypt) { if (ctx->alg_type == SEC_SKCIPHER) - memcpy(req->c_req.sk_req->iv, c_req->c_ivin, + memcpy(req->c_req.sk_req->iv, req->c_req.c_ivin, ctx->c_ctx.ivsize); else - memcpy(req->aead_req.aead_req->iv, c_req->c_ivin, + memcpy(req->aead_req.aead_req->iv, req->c_req.c_ivin, ctx->c_ctx.ivsize); } sec_request_untransfer(ctx, req); + err_uninit_req: - sec_request_uninit(ctx, req); + sec_request_uninit(req); + if (ctx->alg_type == SEC_AEAD) + ret = sec_aead_soft_crypto(ctx, req->aead_req.aead_req, + req->c_req.encrypt); + else + ret = sec_skcipher_soft_crypto(ctx, req->c_req.sk_req, + req->c_req.encrypt); return ret; } @@ -1850,7 +2078,7 @@ static int sec_aead_init(struct crypto_aead *tfm) struct sec_ctx *ctx = crypto_aead_ctx(tfm); int ret; - crypto_aead_set_reqsize(tfm, sizeof(struct sec_req)); + crypto_aead_set_reqsize_dma(tfm, sizeof(struct sec_req)); ctx->alg_type = SEC_AEAD; ctx->c_ctx.ivsize = crypto_aead_ivsize(tfm); if (ctx->c_ctx.ivsize < SEC_AIV_SIZE || @@ -2087,7 +2315,7 @@ static int sec_skcipher_soft_crypto(struct sec_ctx *ctx, static int sec_skcipher_crypto(struct skcipher_request *sk_req, bool encrypt) { struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(sk_req); - struct sec_req *req = skcipher_request_ctx(sk_req); + struct sec_req *req = skcipher_request_ctx_dma(sk_req); struct sec_ctx *ctx = crypto_skcipher_ctx(tfm); bool need_fallback = false; int ret; @@ -2102,6 +2330,7 @@ static int sec_skcipher_crypto(struct skcipher_request *sk_req, bool encrypt) req->c_req.sk_req = sk_req; req->c_req.encrypt = encrypt; req->ctx = ctx; + req->base = &sk_req->base; ret = sec_skcipher_param_check(ctx, req, &need_fallback); if (unlikely(ret)) @@ -2236,6 +2465,9 @@ static int sec_aead_spec_check(struct sec_ctx *ctx, struct sec_req *sreq) return -EINVAL; if (unlikely(ctx->a_ctx.a_key_len & WORD_MASK)) return -EINVAL; + } else if (c_mode == SEC_CMODE_GCM) { + if (unlikely(sz < SEC_GCM_MIN_AUTH_SZ)) + return -EINVAL; } return 0; @@ -2309,7 +2541,7 @@ static int sec_aead_soft_crypto(struct sec_ctx *ctx, static int sec_aead_crypto(struct aead_request *a_req, bool encrypt) { struct crypto_aead *tfm = crypto_aead_reqtfm(a_req); - struct sec_req *req = aead_request_ctx(a_req); + struct sec_req *req = aead_request_ctx_dma(a_req); struct sec_ctx *ctx = crypto_aead_ctx(tfm); size_t sz = crypto_aead_authsize(tfm); bool need_fallback = false; @@ -2319,6 +2551,7 @@ static int sec_aead_crypto(struct aead_request *a_req, bool encrypt) req->aead_req.aead_req = a_req; req->c_req.encrypt = encrypt; req->ctx = ctx; + req->base = &a_req->base; req->c_req.c_len = a_req->cryptlen - (req->c_req.encrypt ? 0 : sz); ret = sec_aead_param_check(ctx, req, &need_fallback); diff --git a/drivers/crypto/hisilicon/sgl.c b/drivers/crypto/hisilicon/sgl.c index c974f95cd126f..7a9ef2a9972a2 100644 --- a/drivers/crypto/hisilicon/sgl.c +++ b/drivers/crypto/hisilicon/sgl.c @@ -210,15 +210,15 @@ static void clear_hw_sgl_sge(struct hisi_acc_hw_sgl *hw_sgl) * @pool: Pool which hw sgl memory will be allocated in. * @index: Index of hisi_acc_hw_sgl in pool. * @hw_sgl_dma: The dma address of allocated hw sgl. + * @dir: DMA direction. * * This function builds hw sgl according input sgl, user can use hw_sgl_dma * as src/dst in its BD. Only support single hw sgl currently. */ struct hisi_acc_hw_sgl * -hisi_acc_sg_buf_map_to_hw_sgl(struct device *dev, - struct scatterlist *sgl, - struct hisi_acc_sgl_pool *pool, - u32 index, dma_addr_t *hw_sgl_dma) +hisi_acc_sg_buf_map_to_hw_sgl(struct device *dev, struct scatterlist *sgl, + struct hisi_acc_sgl_pool *pool, u32 index, + dma_addr_t *hw_sgl_dma, enum dma_data_direction dir) { struct hisi_acc_hw_sgl *curr_hw_sgl; unsigned int i, sg_n_mapped; @@ -232,7 +232,7 @@ hisi_acc_sg_buf_map_to_hw_sgl(struct device *dev, sg_n = sg_nents(sgl); - sg_n_mapped = dma_map_sg(dev, sgl, sg_n, DMA_BIDIRECTIONAL); + sg_n_mapped = dma_map_sg(dev, sgl, sg_n, dir); if (!sg_n_mapped) { dev_err(dev, "DMA mapping for SG error!\n"); return ERR_PTR(-EINVAL); @@ -276,16 +276,17 @@ EXPORT_SYMBOL_GPL(hisi_acc_sg_buf_map_to_hw_sgl); * @dev: The device which hw sgl belongs to. * @sgl: Related scatterlist. * @hw_sgl: Virtual address of hw sgl. + * @dir: DMA direction. * * This function unmaps allocated hw sgl. */ void hisi_acc_sg_buf_unmap(struct device *dev, struct scatterlist *sgl, - struct hisi_acc_hw_sgl *hw_sgl) + struct hisi_acc_hw_sgl *hw_sgl, enum dma_data_direction dir) { if (!dev || !sgl || !hw_sgl) return; - dma_unmap_sg(dev, sgl, sg_nents(sgl), DMA_BIDIRECTIONAL); + dma_unmap_sg(dev, sgl, sg_nents(sgl), dir); clear_hw_sgl_sge(hw_sgl); hw_sgl->entry_sum_in_chain = 0; hw_sgl->entry_sum_in_sgl = 0; diff --git a/drivers/crypto/hisilicon/zip/zip_crypto.c b/drivers/crypto/hisilicon/zip/zip_crypto.c index 7327f8f29b013..b97513981a3b7 100644 --- a/drivers/crypto/hisilicon/zip/zip_crypto.c +++ b/drivers/crypto/hisilicon/zip/zip_crypto.c @@ -224,7 +224,8 @@ static int hisi_zip_do_work(struct hisi_zip_qp_ctx *qp_ctx, return -EINVAL; req->hw_src = hisi_acc_sg_buf_map_to_hw_sgl(dev, a_req->src, pool, - req->req_id << 1, &req->dma_src); + req->req_id << 1, &req->dma_src, + DMA_TO_DEVICE); if (IS_ERR(req->hw_src)) { dev_err(dev, "failed to map the src buffer to hw sgl (%ld)!\n", PTR_ERR(req->hw_src)); @@ -233,7 +234,7 @@ static int hisi_zip_do_work(struct hisi_zip_qp_ctx *qp_ctx, req->hw_dst = hisi_acc_sg_buf_map_to_hw_sgl(dev, a_req->dst, pool, (req->req_id << 1) + 1, - &req->dma_dst); + &req->dma_dst, DMA_FROM_DEVICE); if (IS_ERR(req->hw_dst)) { ret = PTR_ERR(req->hw_dst); dev_err(dev, "failed to map the dst buffer to hw slg (%d)!\n", @@ -258,9 +259,9 @@ static int hisi_zip_do_work(struct hisi_zip_qp_ctx *qp_ctx, return -EINPROGRESS; err_unmap_output: - hisi_acc_sg_buf_unmap(dev, a_req->dst, req->hw_dst); + hisi_acc_sg_buf_unmap(dev, a_req->dst, req->hw_dst, DMA_FROM_DEVICE); err_unmap_input: - hisi_acc_sg_buf_unmap(dev, a_req->src, req->hw_src); + hisi_acc_sg_buf_unmap(dev, a_req->src, req->hw_src, DMA_TO_DEVICE); return ret; } @@ -303,8 +304,8 @@ static void hisi_zip_acomp_cb(struct hisi_qp *qp, void *data) err = -EIO; } - hisi_acc_sg_buf_unmap(dev, acomp_req->src, req->hw_src); - hisi_acc_sg_buf_unmap(dev, acomp_req->dst, req->hw_dst); + hisi_acc_sg_buf_unmap(dev, acomp_req->dst, req->hw_dst, DMA_FROM_DEVICE); + hisi_acc_sg_buf_unmap(dev, acomp_req->src, req->hw_src, DMA_TO_DEVICE); acomp_req->dlen = ops->get_dstlen(sqe); diff --git a/drivers/crypto/img-hash.c b/drivers/crypto/img-hash.c index a8f735390f0dd..76b7ecb5624b1 100644 --- a/drivers/crypto/img-hash.c +++ b/drivers/crypto/img-hash.c @@ -436,7 +436,7 @@ static int img_hash_write_via_dma_stop(struct img_hash_dev *hdev) struct img_hash_request_ctx *ctx = ahash_request_ctx(hdev->req); if (ctx->flags & DRIVER_FLAGS_SG) - dma_unmap_sg(hdev->dev, ctx->sg, ctx->dma_ct, DMA_TO_DEVICE); + dma_unmap_sg(hdev->dev, ctx->sg, 1, DMA_TO_DEVICE); return 0; } diff --git a/drivers/crypto/inside-secure/safexcel_hash.c b/drivers/crypto/inside-secure/safexcel_hash.c index fd34dc8f5707d..ef0ba48329281 100644 --- a/drivers/crypto/inside-secure/safexcel_hash.c +++ b/drivers/crypto/inside-secure/safexcel_hash.c @@ -249,7 +249,9 @@ static int safexcel_handle_req_result(struct safexcel_crypto_priv *priv, safexcel_complete(priv, ring); if (sreq->nents) { - dma_unmap_sg(priv->dev, areq->src, sreq->nents, DMA_TO_DEVICE); + dma_unmap_sg(priv->dev, areq->src, + sg_nents_for_len(areq->src, areq->nbytes), + DMA_TO_DEVICE); sreq->nents = 0; } @@ -491,7 +493,9 @@ unmap_result: DMA_FROM_DEVICE); unmap_sg: if (req->nents) { - dma_unmap_sg(priv->dev, areq->src, req->nents, DMA_TO_DEVICE); + dma_unmap_sg(priv->dev, areq->src, + sg_nents_for_len(areq->src, areq->nbytes), + DMA_TO_DEVICE); req->nents = 0; } cdesc_rollback: diff --git a/drivers/crypto/intel/keembay/keembay-ocs-hcu-core.c b/drivers/crypto/intel/keembay/keembay-ocs-hcu-core.c index 95dc8979918d8..8f9e21ced0fe1 100644 --- a/drivers/crypto/intel/keembay/keembay-ocs-hcu-core.c +++ b/drivers/crypto/intel/keembay/keembay-ocs-hcu-core.c @@ -68,6 +68,7 @@ struct ocs_hcu_ctx { * @sg_data_total: Total data in the SG list at any time. * @sg_data_offset: Offset into the data of the current individual SG node. * @sg_dma_nents: Number of sg entries mapped in dma_list. + * @nents: Number of entries in the scatterlist. */ struct ocs_hcu_rctx { struct ocs_hcu_dev *hcu_dev; @@ -91,6 +92,7 @@ struct ocs_hcu_rctx { unsigned int sg_data_total; unsigned int sg_data_offset; unsigned int sg_dma_nents; + unsigned int nents; }; /** @@ -199,7 +201,7 @@ static void kmb_ocs_hcu_dma_cleanup(struct ahash_request *req, /* Unmap req->src (if mapped). */ if (rctx->sg_dma_nents) { - dma_unmap_sg(dev, req->src, rctx->sg_dma_nents, DMA_TO_DEVICE); + dma_unmap_sg(dev, req->src, rctx->nents, DMA_TO_DEVICE); rctx->sg_dma_nents = 0; } @@ -260,6 +262,10 @@ static int kmb_ocs_dma_prepare(struct ahash_request *req) rc = -ENOMEM; goto cleanup; } + + /* Save the value of nents to pass to dma_unmap_sg. */ + rctx->nents = nents; + /* * The value returned by dma_map_sg() can be < nents; so update * nents accordingly. diff --git a/drivers/crypto/intel/keembay/ocs-aes.c b/drivers/crypto/intel/keembay/ocs-aes.c index be9f32fc8f42a..bb6f33f6b4d39 100644 --- a/drivers/crypto/intel/keembay/ocs-aes.c +++ b/drivers/crypto/intel/keembay/ocs-aes.c @@ -7,6 +7,7 @@ #include <linux/dma-mapping.h> #include <linux/interrupt.h> +#include <linux/kernel.h> #include <linux/platform_device.h> #include <linux/slab.h> #include <linux/swab.h> @@ -1473,8 +1474,7 @@ int ocs_create_linked_list_from_sg(const struct ocs_aes_dev *aes_dev, ll = dll_desc->vaddr; for (i = 0; i < dma_nents; i++, sg = sg_next(sg)) { ll[i].src_addr = sg_dma_address(sg) + data_offset; - ll[i].src_len = (sg_dma_len(sg) - data_offset) < data_size ? - (sg_dma_len(sg) - data_offset) : data_size; + ll[i].src_len = min(sg_dma_len(sg) - data_offset, data_size); data_offset = 0; data_size -= ll[i].src_len; /* Current element points to the DMA address of the next one. */ diff --git a/drivers/crypto/intel/qat/qat_420xx/adf_420xx_hw_data.c b/drivers/crypto/intel/qat/qat_420xx/adf_420xx_hw_data.c index 7c3c0f561c956..53fa91d577ed0 100644 --- a/drivers/crypto/intel/qat/qat_420xx/adf_420xx_hw_data.c +++ b/drivers/crypto/intel/qat/qat_420xx/adf_420xx_hw_data.c @@ -191,7 +191,6 @@ static u32 get_accel_cap(struct adf_accel_dev *accel_dev) ICP_ACCEL_CAPABILITIES_SM4 | ICP_ACCEL_CAPABILITIES_AES_V2 | ICP_ACCEL_CAPABILITIES_ZUC | - ICP_ACCEL_CAPABILITIES_ZUC_256 | ICP_ACCEL_CAPABILITIES_WIRELESS_CRYPTO_EXT | ICP_ACCEL_CAPABILITIES_EXT_ALGCHAIN; @@ -223,17 +222,11 @@ static u32 get_accel_cap(struct adf_accel_dev *accel_dev) if (fusectl1 & ICP_ACCEL_GEN4_MASK_WCP_WAT_SLICE) { capabilities_sym &= ~ICP_ACCEL_CAPABILITIES_ZUC; - capabilities_sym &= ~ICP_ACCEL_CAPABILITIES_ZUC_256; capabilities_sym &= ~ICP_ACCEL_CAPABILITIES_WIRELESS_CRYPTO_EXT; } - if (fusectl1 & ICP_ACCEL_GEN4_MASK_EIA3_SLICE) { + if (fusectl1 & ICP_ACCEL_GEN4_MASK_EIA3_SLICE) capabilities_sym &= ~ICP_ACCEL_CAPABILITIES_ZUC; - capabilities_sym &= ~ICP_ACCEL_CAPABILITIES_ZUC_256; - } - - if (fusectl1 & ICP_ACCEL_GEN4_MASK_ZUC_256_SLICE) - capabilities_sym &= ~ICP_ACCEL_CAPABILITIES_ZUC_256; capabilities_asym = ICP_ACCEL_CAPABILITIES_CRYPTO_ASYMMETRIC | ICP_ACCEL_CAPABILITIES_SM2 | @@ -303,11 +296,13 @@ static void adf_init_rl_data(struct adf_rl_hw_data *rl_data) rl_data->pcie_scale_div = ADF_420XX_RL_PCIE_SCALE_FACTOR_DIV; rl_data->pcie_scale_mul = ADF_420XX_RL_PCIE_SCALE_FACTOR_MUL; rl_data->dcpr_correction = ADF_420XX_RL_DCPR_CORRECTION; - rl_data->max_tp[ADF_SVC_ASYM] = ADF_420XX_RL_MAX_TP_ASYM; - rl_data->max_tp[ADF_SVC_SYM] = ADF_420XX_RL_MAX_TP_SYM; - rl_data->max_tp[ADF_SVC_DC] = ADF_420XX_RL_MAX_TP_DC; + rl_data->max_tp[SVC_ASYM] = ADF_420XX_RL_MAX_TP_ASYM; + rl_data->max_tp[SVC_SYM] = ADF_420XX_RL_MAX_TP_SYM; + rl_data->max_tp[SVC_DC] = ADF_420XX_RL_MAX_TP_DC; rl_data->scan_interval = ADF_420XX_RL_SCANS_PER_SEC; rl_data->scale_ref = ADF_420XX_RL_SLICE_REF; + + adf_gen4_init_num_svc_aes(rl_data); } static int get_rp_group(struct adf_accel_dev *accel_dev, u32 ae_mask) @@ -473,6 +468,7 @@ void adf_init_hw_data_420xx(struct adf_hw_device_data *hw_data, u32 dev_id) hw_data->num_hb_ctrs = ADF_NUM_HB_CNT_PER_AE; hw_data->clock_frequency = ADF_420XX_AE_FREQ; hw_data->services_supported = adf_gen4_services_supported; + hw_data->get_svc_slice_cnt = adf_gen4_get_svc_slice_cnt; adf_gen4_set_err_mask(&hw_data->dev_err_mask); adf_gen4_init_hw_csr_ops(&hw_data->csr_ops); diff --git a/drivers/crypto/intel/qat/qat_4xxx/adf_4xxx_hw_data.c b/drivers/crypto/intel/qat/qat_4xxx/adf_4xxx_hw_data.c index bd0b1b1015c06..740f68a36ac51 100644 --- a/drivers/crypto/intel/qat/qat_4xxx/adf_4xxx_hw_data.c +++ b/drivers/crypto/intel/qat/qat_4xxx/adf_4xxx_hw_data.c @@ -3,6 +3,7 @@ #include <linux/iopoll.h> #include <adf_accel_devices.h> #include <adf_admin.h> +#include <adf_bank_state.h> #include <adf_cfg.h> #include <adf_cfg_services.h> #include <adf_clock.h> @@ -221,11 +222,13 @@ static void adf_init_rl_data(struct adf_rl_hw_data *rl_data) rl_data->pcie_scale_div = ADF_4XXX_RL_PCIE_SCALE_FACTOR_DIV; rl_data->pcie_scale_mul = ADF_4XXX_RL_PCIE_SCALE_FACTOR_MUL; rl_data->dcpr_correction = ADF_4XXX_RL_DCPR_CORRECTION; - rl_data->max_tp[ADF_SVC_ASYM] = ADF_4XXX_RL_MAX_TP_ASYM; - rl_data->max_tp[ADF_SVC_SYM] = ADF_4XXX_RL_MAX_TP_SYM; - rl_data->max_tp[ADF_SVC_DC] = ADF_4XXX_RL_MAX_TP_DC; + rl_data->max_tp[SVC_ASYM] = ADF_4XXX_RL_MAX_TP_ASYM; + rl_data->max_tp[SVC_SYM] = ADF_4XXX_RL_MAX_TP_SYM; + rl_data->max_tp[SVC_DC] = ADF_4XXX_RL_MAX_TP_DC; rl_data->scan_interval = ADF_4XXX_RL_SCANS_PER_SEC; rl_data->scale_ref = ADF_4XXX_RL_SLICE_REF; + + adf_gen4_init_num_svc_aes(rl_data); } static u32 uof_get_num_objs(struct adf_accel_dev *accel_dev) @@ -448,8 +451,8 @@ void adf_init_hw_data_4xxx(struct adf_hw_device_data *hw_data, u32 dev_id) hw_data->get_ring_to_svc_map = adf_gen4_get_ring_to_svc_map; hw_data->disable_iov = adf_disable_sriov; hw_data->ring_pair_reset = adf_gen4_ring_pair_reset; - hw_data->bank_state_save = adf_gen4_bank_state_save; - hw_data->bank_state_restore = adf_gen4_bank_state_restore; + hw_data->bank_state_save = adf_bank_state_save; + hw_data->bank_state_restore = adf_bank_state_restore; hw_data->enable_pm = adf_gen4_enable_pm; hw_data->handle_pm_interrupt = adf_gen4_handle_pm_interrupt; hw_data->dev_config = adf_gen4_dev_config; @@ -459,6 +462,7 @@ void adf_init_hw_data_4xxx(struct adf_hw_device_data *hw_data, u32 dev_id) hw_data->num_hb_ctrs = ADF_NUM_HB_CNT_PER_AE; hw_data->clock_frequency = ADF_4XXX_AE_FREQ; hw_data->services_supported = adf_gen4_services_supported; + hw_data->get_svc_slice_cnt = adf_gen4_get_svc_slice_cnt; adf_gen4_set_err_mask(&hw_data->dev_err_mask); adf_gen4_init_hw_csr_ops(&hw_data->csr_ops); diff --git a/drivers/crypto/intel/qat/qat_6xxx/adf_6xxx_hw_data.c b/drivers/crypto/intel/qat/qat_6xxx/adf_6xxx_hw_data.c index 359a6447ccb83..bed88d3ce8cac 100644 --- a/drivers/crypto/intel/qat/qat_6xxx/adf_6xxx_hw_data.c +++ b/drivers/crypto/intel/qat/qat_6xxx/adf_6xxx_hw_data.c @@ -10,6 +10,7 @@ #include <adf_accel_devices.h> #include <adf_admin.h> +#include <adf_bank_state.h> #include <adf_cfg.h> #include <adf_cfg_services.h> #include <adf_clock.h> @@ -18,6 +19,7 @@ #include <adf_gen6_pm.h> #include <adf_gen6_ras.h> #include <adf_gen6_shared.h> +#include <adf_gen6_tl.h> #include <adf_timer.h> #include "adf_6xxx_hw_data.h" #include "icp_qat_fw_comp.h" @@ -76,6 +78,10 @@ static const unsigned long thrd_mask_dcc[ADF_6XXX_MAX_ACCELENGINES] = { 0x00, 0x00, 0x00, 0x00, 0x07, 0x07, 0x03, 0x03, 0x00 }; +static const unsigned long thrd_mask_dcpr[ADF_6XXX_MAX_ACCELENGINES] = { + 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x00 +}; + static const char *const adf_6xxx_fw_objs[] = { [ADF_FW_CY_OBJ] = ADF_6XXX_CY_OBJ, [ADF_FW_DC_OBJ] = ADF_6XXX_DC_OBJ, @@ -97,7 +103,7 @@ static bool services_supported(unsigned long mask) { int num_svc; - if (mask >= BIT(SVC_BASE_COUNT)) + if (mask >= BIT(SVC_COUNT)) return false; num_svc = hweight_long(mask); @@ -126,10 +132,13 @@ static int get_service(unsigned long *mask) if (test_and_clear_bit(SVC_DCC, mask)) return SVC_DCC; + if (test_and_clear_bit(SVC_DECOMP, mask)) + return SVC_DECOMP; + return -EINVAL; } -static enum adf_cfg_service_type get_ring_type(enum adf_services service) +static enum adf_cfg_service_type get_ring_type(unsigned int service) { switch (service) { case SVC_SYM: @@ -139,12 +148,14 @@ static enum adf_cfg_service_type get_ring_type(enum adf_services service) case SVC_DC: case SVC_DCC: return COMP; + case SVC_DECOMP: + return DECOMP; default: return UNUSED; } } -static const unsigned long *get_thrd_mask(enum adf_services service) +static const unsigned long *get_thrd_mask(unsigned int service) { switch (service) { case SVC_SYM: @@ -155,6 +166,8 @@ static const unsigned long *get_thrd_mask(enum adf_services service) return thrd_mask_cpr; case SVC_DCC: return thrd_mask_dcc; + case SVC_DECOMP: + return thrd_mask_dcpr; default: return NULL; } @@ -511,6 +524,55 @@ static int adf_gen6_init_thd2arb_map(struct adf_accel_dev *accel_dev) return 0; } +static void init_num_svc_aes(struct adf_rl_hw_data *device_data) +{ + enum adf_fw_objs obj_type, obj_iter; + unsigned int svc, i, num_grp; + u32 ae_mask; + + for (svc = 0; svc < SVC_BASE_COUNT; svc++) { + switch (svc) { + case SVC_SYM: + case SVC_ASYM: + obj_type = ADF_FW_CY_OBJ; + break; + case SVC_DC: + case SVC_DECOMP: + obj_type = ADF_FW_DC_OBJ; + break; + } + + num_grp = ARRAY_SIZE(adf_default_fw_config); + for (i = 0; i < num_grp; i++) { + obj_iter = adf_default_fw_config[i].obj; + if (obj_iter == obj_type) { + ae_mask = adf_default_fw_config[i].ae_mask; + device_data->svc_ae_mask[svc] = hweight32(ae_mask); + break; + } + } + } +} + +static u32 adf_gen6_get_svc_slice_cnt(struct adf_accel_dev *accel_dev, + enum adf_base_services svc) +{ + struct adf_rl_hw_data *device_data = &accel_dev->hw_device->rl_data; + + switch (svc) { + case SVC_SYM: + return device_data->slices.cph_cnt; + case SVC_ASYM: + return device_data->slices.pke_cnt; + case SVC_DC: + return device_data->slices.cpr_cnt + device_data->slices.dcpr_cnt; + case SVC_DECOMP: + return device_data->slices.dcpr_cnt; + default: + return 0; + } +} + static void set_vc_csr_for_bank(void __iomem *csr, u32 bank_number) { u32 value; @@ -520,8 +582,8 @@ static void set_vc_csr_for_bank(void __iomem *csr, u32 bank_number) * driver must program the ringmodectl CSRs. */ value = ADF_CSR_RD(csr, ADF_GEN6_CSR_RINGMODECTL(bank_number)); - value |= FIELD_PREP(ADF_GEN6_RINGMODECTL_TC_MASK, ADF_GEN6_RINGMODECTL_TC_DEFAULT); - value |= FIELD_PREP(ADF_GEN6_RINGMODECTL_TC_EN_MASK, ADF_GEN6_RINGMODECTL_TC_EN_OP1); + FIELD_MODIFY(ADF_GEN6_RINGMODECTL_TC_MASK, &value, ADF_GEN6_RINGMODECTL_TC_DEFAULT); + FIELD_MODIFY(ADF_GEN6_RINGMODECTL_TC_EN_MASK, &value, ADF_GEN6_RINGMODECTL_TC_EN_OP1); ADF_CSR_WR(csr, ADF_GEN6_CSR_RINGMODECTL(bank_number), value); } @@ -537,7 +599,7 @@ static int set_vc_config(struct adf_accel_dev *accel_dev) * Read PVC0CTL then write the masked values. */ pci_read_config_dword(pdev, ADF_GEN6_PVC0CTL_OFFSET, &value); - value |= FIELD_PREP(ADF_GEN6_PVC0CTL_TCVCMAP_MASK, ADF_GEN6_PVC0CTL_TCVCMAP_DEFAULT); + FIELD_MODIFY(ADF_GEN6_PVC0CTL_TCVCMAP_MASK, &value, ADF_GEN6_PVC0CTL_TCVCMAP_DEFAULT); err = pci_write_config_dword(pdev, ADF_GEN6_PVC0CTL_OFFSET, value); if (err) { dev_err(&GET_DEV(accel_dev), "pci write to PVC0CTL failed\n"); @@ -546,8 +608,8 @@ static int set_vc_config(struct adf_accel_dev *accel_dev) /* Read PVC1CTL then write masked values */ pci_read_config_dword(pdev, ADF_GEN6_PVC1CTL_OFFSET, &value); - value |= FIELD_PREP(ADF_GEN6_PVC1CTL_TCVCMAP_MASK, ADF_GEN6_PVC1CTL_TCVCMAP_DEFAULT); - value |= FIELD_PREP(ADF_GEN6_PVC1CTL_VCEN_MASK, ADF_GEN6_PVC1CTL_VCEN_ON); + FIELD_MODIFY(ADF_GEN6_PVC1CTL_TCVCMAP_MASK, &value, ADF_GEN6_PVC1CTL_TCVCMAP_DEFAULT); + FIELD_MODIFY(ADF_GEN6_PVC1CTL_VCEN_MASK, &value, ADF_GEN6_PVC1CTL_VCEN_ON); err = pci_write_config_dword(pdev, ADF_GEN6_PVC1CTL_OFFSET, value); if (err) dev_err(&GET_DEV(accel_dev), "pci write to PVC1CTL failed\n"); @@ -618,7 +680,6 @@ static u32 get_accel_cap(struct adf_accel_dev *accel_dev) capabilities_sym &= ~ICP_ACCEL_CAPABILITIES_CHACHA_POLY; capabilities_sym &= ~ICP_ACCEL_CAPABILITIES_AESGCM_SPC; capabilities_sym &= ~ICP_ACCEL_CAPABILITIES_AES_V2; - capabilities_sym &= ~ICP_ACCEL_CAPABILITIES_CIPHER; } if (fusectl1 & ICP_ACCEL_GEN6_MASK_AUTH_SLICE) { capabilities_sym &= ~ICP_ACCEL_CAPABILITIES_AUTHENTICATION; @@ -627,7 +688,15 @@ static u32 get_accel_cap(struct adf_accel_dev *accel_dev) capabilities_sym &= ~ICP_ACCEL_CAPABILITIES_CIPHER; } - capabilities_asym = 0; + capabilities_asym = ICP_ACCEL_CAPABILITIES_CRYPTO_ASYMMETRIC | + ICP_ACCEL_CAPABILITIES_SM2 | + ICP_ACCEL_CAPABILITIES_ECEDMONT; + + if (fusectl1 & ICP_ACCEL_GEN6_MASK_PKE_SLICE) { + capabilities_asym &= ~ICP_ACCEL_CAPABILITIES_CRYPTO_ASYMMETRIC; + capabilities_asym &= ~ICP_ACCEL_CAPABILITIES_SM2; + capabilities_asym &= ~ICP_ACCEL_CAPABILITIES_ECEDMONT; + } capabilities_dc = ICP_ACCEL_CAPABILITIES_COMPRESSION | ICP_ACCEL_CAPABILITIES_LZ4_COMPRESSION | @@ -648,7 +717,7 @@ static u32 get_accel_cap(struct adf_accel_dev *accel_dev) caps |= capabilities_asym; if (test_bit(SVC_SYM, &mask)) caps |= capabilities_sym; - if (test_bit(SVC_DC, &mask)) + if (test_bit(SVC_DC, &mask) || test_bit(SVC_DECOMP, &mask)) caps |= capabilities_dc; if (test_bit(SVC_DCC, &mask)) { /* @@ -744,7 +813,16 @@ static int adf_init_device(struct adf_accel_dev *accel_dev) static int enable_pm(struct adf_accel_dev *accel_dev) { - return adf_init_admin_pm(accel_dev, ADF_GEN6_PM_DEFAULT_IDLE_FILTER); + int ret; + + ret = adf_init_admin_pm(accel_dev, ADF_GEN6_PM_DEFAULT_IDLE_FILTER); + if (ret) + return ret; + + /* Initialize PM internal data */ + adf_gen6_init_dev_pm_data(accel_dev); + + return 0; } static int dev_config(struct adf_accel_dev *accel_dev) @@ -776,6 +854,25 @@ static int dev_config(struct adf_accel_dev *accel_dev) return ret; } +static void adf_gen6_init_rl_data(struct adf_rl_hw_data *rl_data) +{ + rl_data->pciout_tb_offset = ADF_GEN6_RL_TOKEN_PCIEOUT_BUCKET_OFFSET; + rl_data->pciin_tb_offset = ADF_GEN6_RL_TOKEN_PCIEIN_BUCKET_OFFSET; + rl_data->r2l_offset = ADF_GEN6_RL_R2L_OFFSET; + rl_data->l2c_offset = ADF_GEN6_RL_L2C_OFFSET; + rl_data->c2s_offset = ADF_GEN6_RL_C2S_OFFSET; + rl_data->pcie_scale_div = ADF_6XXX_RL_PCIE_SCALE_FACTOR_DIV; + rl_data->pcie_scale_mul = ADF_6XXX_RL_PCIE_SCALE_FACTOR_MUL; + rl_data->max_tp[SVC_ASYM] = ADF_6XXX_RL_MAX_TP_ASYM; + rl_data->max_tp[SVC_SYM] = ADF_6XXX_RL_MAX_TP_SYM; + rl_data->max_tp[SVC_DC] = ADF_6XXX_RL_MAX_TP_DC; + rl_data->max_tp[SVC_DECOMP] = ADF_6XXX_RL_MAX_TP_DECOMP; + rl_data->scan_interval = ADF_6XXX_RL_SCANS_PER_SEC; + rl_data->scale_ref = ADF_6XXX_RL_SLICE_REF; + + init_num_svc_aes(rl_data); +} + void adf_init_hw_data_6xxx(struct adf_hw_device_data *hw_data) { hw_data->dev_class = &adf_6xxx_class; @@ -824,6 +921,8 @@ void adf_init_hw_data_6xxx(struct adf_hw_device_data *hw_data) hw_data->disable_iov = adf_disable_sriov; hw_data->ring_pair_reset = ring_pair_reset; hw_data->dev_config = dev_config; + hw_data->bank_state_save = adf_bank_state_save; + hw_data->bank_state_restore = adf_bank_state_restore; hw_data->get_hb_clock = get_heartbeat_clock; hw_data->num_hb_ctrs = ADF_NUM_HB_CNT_PER_AE; hw_data->start_timer = adf_timer_start; @@ -831,11 +930,17 @@ void adf_init_hw_data_6xxx(struct adf_hw_device_data *hw_data) hw_data->init_device = adf_init_device; hw_data->enable_pm = enable_pm; hw_data->services_supported = services_supported; + hw_data->num_rps = ADF_GEN6_ETR_MAX_BANKS; + hw_data->clock_frequency = ADF_6XXX_AE_FREQ; + hw_data->get_svc_slice_cnt = adf_gen6_get_svc_slice_cnt; adf_gen6_init_hw_csr_ops(&hw_data->csr_ops); adf_gen6_init_pf_pfvf_ops(&hw_data->pfvf_ops); adf_gen6_init_dc_ops(&hw_data->dc_ops); + adf_gen6_init_vf_mig_ops(&hw_data->vfmig_ops); adf_gen6_init_ras_ops(&hw_data->ras_ops); + adf_gen6_init_tl_data(&hw_data->tl_data); + adf_gen6_init_rl_data(&hw_data->rl_data); } void adf_clean_hw_data_6xxx(struct adf_hw_device_data *hw_data) diff --git a/drivers/crypto/intel/qat/qat_6xxx/adf_6xxx_hw_data.h b/drivers/crypto/intel/qat/qat_6xxx/adf_6xxx_hw_data.h index 78e2e2c5816e5..d822911fe68cc 100644 --- a/drivers/crypto/intel/qat/qat_6xxx/adf_6xxx_hw_data.h +++ b/drivers/crypto/intel/qat/qat_6xxx/adf_6xxx_hw_data.h @@ -99,7 +99,7 @@ #define ADF_GEN6_PVC0CTL_OFFSET 0x204 #define ADF_GEN6_PVC0CTL_TCVCMAP_OFFSET 1 #define ADF_GEN6_PVC0CTL_TCVCMAP_MASK GENMASK(7, 1) -#define ADF_GEN6_PVC0CTL_TCVCMAP_DEFAULT 0x7F +#define ADF_GEN6_PVC0CTL_TCVCMAP_DEFAULT 0x3F /* VC1 Resource Control Register */ #define ADF_GEN6_PVC1CTL_OFFSET 0x210 @@ -122,6 +122,13 @@ /* Number of heartbeat counter pairs */ #define ADF_NUM_HB_CNT_PER_AE ADF_NUM_THREADS_PER_AE +/* Rate Limiting */ +#define ADF_GEN6_RL_R2L_OFFSET 0x508000 +#define ADF_GEN6_RL_L2C_OFFSET 0x509000 +#define ADF_GEN6_RL_C2S_OFFSET 0x508818 +#define ADF_GEN6_RL_TOKEN_PCIEIN_BUCKET_OFFSET 0x508800 +#define ADF_GEN6_RL_TOKEN_PCIEOUT_BUCKET_OFFSET 0x508804 + /* Physical function fuses */ #define ADF_6XXX_ACCELENGINES_MASK GENMASK(8, 0) #define ADF_6XXX_ADMIN_AE_MASK GENMASK(8, 8) @@ -133,6 +140,19 @@ #define ADF_6XXX_DC_OBJ "qat_6xxx_dc.bin" #define ADF_6XXX_ADMIN_OBJ "qat_6xxx_admin.bin" +/* RL constants */ +#define ADF_6XXX_RL_PCIE_SCALE_FACTOR_DIV 100 +#define ADF_6XXX_RL_PCIE_SCALE_FACTOR_MUL 102 +#define ADF_6XXX_RL_SCANS_PER_SEC 954 +#define ADF_6XXX_RL_MAX_TP_ASYM 173750UL +#define ADF_6XXX_RL_MAX_TP_SYM 95000UL +#define ADF_6XXX_RL_MAX_TP_DC 40000UL +#define ADF_6XXX_RL_MAX_TP_DECOMP 40000UL +#define ADF_6XXX_RL_SLICE_REF 1000UL + +/* Clock frequency */ +#define ADF_6XXX_AE_FREQ (1000 * HZ_PER_MHZ) + enum icp_qat_gen6_slice_mask { ICP_ACCEL_GEN6_MASK_UCS_SLICE = BIT(0), ICP_ACCEL_GEN6_MASK_AUTH_SLICE = BIT(1), diff --git a/drivers/crypto/intel/qat/qat_common/Makefile b/drivers/crypto/intel/qat/qat_common/Makefile index 66bb295ace289..89845754841b8 100644 --- a/drivers/crypto/intel/qat/qat_common/Makefile +++ b/drivers/crypto/intel/qat/qat_common/Makefile @@ -4,6 +4,7 @@ ccflags-y += -DDEFAULT_SYMBOL_NAMESPACE='"CRYPTO_QAT"' intel_qat-y := adf_accel_engine.o \ adf_admin.o \ adf_aer.o \ + adf_bank_state.o \ adf_cfg.o \ adf_cfg_services.o \ adf_clock.o \ @@ -48,9 +49,12 @@ intel_qat-$(CONFIG_DEBUG_FS) += adf_cnv_dbgfs.o \ adf_fw_counters.o \ adf_gen4_pm_debugfs.o \ adf_gen4_tl.o \ + adf_gen6_pm_dbgfs.o \ + adf_gen6_tl.o \ adf_heartbeat_dbgfs.o \ adf_heartbeat.o \ adf_pm_dbgfs.o \ + adf_pm_dbgfs_utils.o \ adf_telemetry.o \ adf_tl_debugfs.o \ adf_transport_debug.o diff --git a/drivers/crypto/intel/qat/qat_common/adf_accel_devices.h b/drivers/crypto/intel/qat/qat_common/adf_accel_devices.h index 2ee5260632133..9fe3239f01143 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_accel_devices.h +++ b/drivers/crypto/intel/qat/qat_common/adf_accel_devices.h @@ -157,39 +157,7 @@ struct admin_info { u32 mailbox_offset; }; -struct ring_config { - u64 base; - u32 config; - u32 head; - u32 tail; - u32 reserved0; -}; - -struct bank_state { - u32 ringstat0; - u32 ringstat1; - u32 ringuostat; - u32 ringestat; - u32 ringnestat; - u32 ringnfstat; - u32 ringfstat; - u32 ringcstat0; - u32 ringcstat1; - u32 ringcstat2; - u32 ringcstat3; - u32 iaintflagen; - u32 iaintflagreg; - u32 iaintflagsrcsel0; - u32 iaintflagsrcsel1; - u32 iaintcolen; - u32 iaintcolctl; - u32 iaintflagandcolen; - u32 ringexpstat; - u32 ringexpintenable; - u32 ringsrvarben; - u32 reserved0; - struct ring_config rings[ADF_ETR_MAX_RINGS_PER_BANK]; -}; +struct adf_bank_state; struct adf_hw_csr_ops { u64 (*build_csr_ring_base_addr)(dma_addr_t addr, u32 size); @@ -338,9 +306,9 @@ struct adf_hw_device_data { void (*set_ssm_wdtimer)(struct adf_accel_dev *accel_dev); int (*ring_pair_reset)(struct adf_accel_dev *accel_dev, u32 bank_nr); int (*bank_state_save)(struct adf_accel_dev *accel_dev, u32 bank_number, - struct bank_state *state); + struct adf_bank_state *state); int (*bank_state_restore)(struct adf_accel_dev *accel_dev, - u32 bank_number, struct bank_state *state); + u32 bank_number, struct adf_bank_state *state); void (*reset_device)(struct adf_accel_dev *accel_dev); void (*set_msix_rttable)(struct adf_accel_dev *accel_dev); const char *(*uof_get_name)(struct adf_accel_dev *accel_dev, u32 obj_num); @@ -351,6 +319,8 @@ struct adf_hw_device_data { u32 (*get_ena_thd_mask)(struct adf_accel_dev *accel_dev, u32 obj_num); int (*dev_config)(struct adf_accel_dev *accel_dev); bool (*services_supported)(unsigned long mask); + u32 (*get_svc_slice_cnt)(struct adf_accel_dev *accel_dev, + enum adf_base_services svc); struct adf_pfvf_ops pfvf_ops; struct adf_hw_csr_ops csr_ops; struct adf_dc_ops dc_ops; diff --git a/drivers/crypto/intel/qat/qat_common/adf_aer.c b/drivers/crypto/intel/qat/qat_common/adf_aer.c index 4cb8bd83f5707..35679b21ff63b 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_aer.c +++ b/drivers/crypto/intel/qat/qat_common/adf_aer.c @@ -229,7 +229,7 @@ const struct pci_error_handlers adf_err_handler = { }; EXPORT_SYMBOL_GPL(adf_err_handler); -int adf_dev_autoreset(struct adf_accel_dev *accel_dev) +static int adf_dev_autoreset(struct adf_accel_dev *accel_dev) { if (accel_dev->autoreset_on_error) return adf_dev_aer_schedule_reset(accel_dev, ADF_DEV_RESET_ASYNC); diff --git a/drivers/crypto/intel/qat/qat_common/adf_bank_state.c b/drivers/crypto/intel/qat/qat_common/adf_bank_state.c new file mode 100644 index 0000000000000..225d55d56a4ba --- /dev/null +++ b/drivers/crypto/intel/qat/qat_common/adf_bank_state.c @@ -0,0 +1,238 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright(c) 2025 Intel Corporation */ + +#define pr_fmt(fmt) "QAT: " fmt + +#include <linux/bits.h> +#include <linux/dev_printk.h> +#include <linux/printk.h> +#include "adf_accel_devices.h" +#include "adf_bank_state.h" +#include "adf_common_drv.h" + +/* Ring interrupt masks */ +#define ADF_RP_INT_SRC_SEL_F_RISE_MASK GENMASK(1, 0) +#define ADF_RP_INT_SRC_SEL_F_FALL_MASK GENMASK(2, 0) +#define ADF_RP_INT_SRC_SEL_RANGE_WIDTH 4 + +static inline int check_stat(u32 (*op)(void __iomem *, u32), u32 expect_val, + const char *name, void __iomem *base, u32 bank) +{ + u32 actual_val = op(base, bank); + + if (expect_val == actual_val) + return 0; + + pr_err("Fail to restore %s register. Expected %#x, actual %#x\n", + name, expect_val, actual_val); + + return -EINVAL; +} + +static void bank_state_save(struct adf_hw_csr_ops *ops, void __iomem *base, + u32 bank, struct adf_bank_state *state, u32 num_rings) +{ + u32 i; + + state->ringstat0 = ops->read_csr_stat(base, bank); + state->ringuostat = ops->read_csr_uo_stat(base, bank); + state->ringestat = ops->read_csr_e_stat(base, bank); + state->ringnestat = ops->read_csr_ne_stat(base, bank); + state->ringnfstat = ops->read_csr_nf_stat(base, bank); + state->ringfstat = ops->read_csr_f_stat(base, bank); + state->ringcstat0 = ops->read_csr_c_stat(base, bank); + state->iaintflagen = ops->read_csr_int_en(base, bank); + state->iaintflagreg = ops->read_csr_int_flag(base, bank); + state->iaintflagsrcsel0 = ops->read_csr_int_srcsel(base, bank); + state->iaintcolen = ops->read_csr_int_col_en(base, bank); + state->iaintcolctl = ops->read_csr_int_col_ctl(base, bank); + state->iaintflagandcolen = ops->read_csr_int_flag_and_col(base, bank); + state->ringexpstat = ops->read_csr_exp_stat(base, bank); + state->ringexpintenable = ops->read_csr_exp_int_en(base, bank); + state->ringsrvarben = ops->read_csr_ring_srv_arb_en(base, bank); + + for (i = 0; i < num_rings; i++) { + state->rings[i].head = ops->read_csr_ring_head(base, bank, i); + state->rings[i].tail = ops->read_csr_ring_tail(base, bank, i); + state->rings[i].config = ops->read_csr_ring_config(base, bank, i); + state->rings[i].base = ops->read_csr_ring_base(base, bank, i); + } +} + +static int bank_state_restore(struct adf_hw_csr_ops *ops, void __iomem *base, + u32 bank, struct adf_bank_state *state, u32 num_rings, + int tx_rx_gap) +{ + u32 val, tmp_val, i; + int ret; + + for (i = 0; i < num_rings; i++) + ops->write_csr_ring_base(base, bank, i, state->rings[i].base); + + for (i = 0; i < num_rings; i++) + ops->write_csr_ring_config(base, bank, i, state->rings[i].config); + + for (i = 0; i < num_rings / 2; i++) { + int tx = i * (tx_rx_gap + 1); + int rx = tx + tx_rx_gap; + + ops->write_csr_ring_head(base, bank, tx, state->rings[tx].head); + ops->write_csr_ring_tail(base, bank, tx, state->rings[tx].tail); + + /* + * The TX ring head needs to be updated again to make sure that + * the HW will not consider the ring as full when it is empty + * and the correct state flags are set to match the recovered state. + */ + if (state->ringestat & BIT(tx)) { + val = ops->read_csr_int_srcsel(base, bank); + val |= ADF_RP_INT_SRC_SEL_F_RISE_MASK; + ops->write_csr_int_srcsel_w_val(base, bank, val); + ops->write_csr_ring_head(base, bank, tx, state->rings[tx].head); + } + + ops->write_csr_ring_tail(base, bank, rx, state->rings[rx].tail); + val = ops->read_csr_int_srcsel(base, bank); + val |= ADF_RP_INT_SRC_SEL_F_RISE_MASK << ADF_RP_INT_SRC_SEL_RANGE_WIDTH; + ops->write_csr_int_srcsel_w_val(base, bank, val); + + ops->write_csr_ring_head(base, bank, rx, state->rings[rx].head); + val = ops->read_csr_int_srcsel(base, bank); + val |= ADF_RP_INT_SRC_SEL_F_FALL_MASK << ADF_RP_INT_SRC_SEL_RANGE_WIDTH; + ops->write_csr_int_srcsel_w_val(base, bank, val); + + /* + * The RX ring tail needs to be updated again to make sure that + * the HW will not consider the ring as empty when it is full + * and the correct state flags are set to match the recovered state. + */ + if (state->ringfstat & BIT(rx)) + ops->write_csr_ring_tail(base, bank, rx, state->rings[rx].tail); + } + + ops->write_csr_int_flag_and_col(base, bank, state->iaintflagandcolen); + ops->write_csr_int_en(base, bank, state->iaintflagen); + ops->write_csr_int_col_en(base, bank, state->iaintcolen); + ops->write_csr_int_srcsel_w_val(base, bank, state->iaintflagsrcsel0); + ops->write_csr_exp_int_en(base, bank, state->ringexpintenable); + ops->write_csr_int_col_ctl(base, bank, state->iaintcolctl); + + /* + * Verify whether any exceptions were raised during the bank save process. + * If exceptions occurred, the status and exception registers cannot + * be directly restored. Consequently, further restoration is not + * feasible, and the current state of the ring should be maintained. + */ + val = state->ringexpstat; + if (val) { + pr_info("Bank %u state not fully restored due to exception in saved state (%#x)\n", + bank, val); + return 0; + } + + /* Ensure that the restoration process completed without exceptions */ + tmp_val = ops->read_csr_exp_stat(base, bank); + if (tmp_val) { + pr_err("Bank %u restored with exception: %#x\n", bank, tmp_val); + return -EFAULT; + } + + ops->write_csr_ring_srv_arb_en(base, bank, state->ringsrvarben); + + /* Check that all ring statuses match the saved state. */ + ret = check_stat(ops->read_csr_stat, state->ringstat0, "ringstat", + base, bank); + if (ret) + return ret; + + ret = check_stat(ops->read_csr_e_stat, state->ringestat, "ringestat", + base, bank); + if (ret) + return ret; + + ret = check_stat(ops->read_csr_ne_stat, state->ringnestat, "ringnestat", + base, bank); + if (ret) + return ret; + + ret = check_stat(ops->read_csr_nf_stat, state->ringnfstat, "ringnfstat", + base, bank); + if (ret) + return ret; + + ret = check_stat(ops->read_csr_f_stat, state->ringfstat, "ringfstat", + base, bank); + if (ret) + return ret; + + ret = check_stat(ops->read_csr_c_stat, state->ringcstat0, "ringcstat", + base, bank); + if (ret) + return ret; + + return 0; +} + +/** + * adf_bank_state_save() - save state of bank-related registers + * @accel_dev: Pointer to the device structure + * @bank_number: Bank number + * @state: Pointer to bank state structure + * + * This function saves the state of a bank by reading the bank CSRs and + * writing them in the @state structure. + * + * Returns 0 on success, error code otherwise + */ +int adf_bank_state_save(struct adf_accel_dev *accel_dev, u32 bank_number, + struct adf_bank_state *state) +{ + struct adf_hw_device_data *hw_data = GET_HW_DATA(accel_dev); + struct adf_hw_csr_ops *csr_ops = GET_CSR_OPS(accel_dev); + void __iomem *csr_base = adf_get_etr_base(accel_dev); + + if (bank_number >= hw_data->num_banks || !state) + return -EINVAL; + + dev_dbg(&GET_DEV(accel_dev), "Saving state of bank %d\n", bank_number); + + bank_state_save(csr_ops, csr_base, bank_number, state, + hw_data->num_rings_per_bank); + + return 0; +} +EXPORT_SYMBOL_GPL(adf_bank_state_save); + +/** + * adf_bank_state_restore() - restore state of bank-related registers + * @accel_dev: Pointer to the device structure + * @bank_number: Bank number + * @state: Pointer to bank state structure + * + * This function attempts to restore the state of a bank by writing the + * bank CSRs to the values in the state structure. + * + * Returns 0 on success, error code otherwise + */ +int adf_bank_state_restore(struct adf_accel_dev *accel_dev, u32 bank_number, + struct adf_bank_state *state) +{ + struct adf_hw_device_data *hw_data = GET_HW_DATA(accel_dev); + struct adf_hw_csr_ops *csr_ops = GET_CSR_OPS(accel_dev); + void __iomem *csr_base = adf_get_etr_base(accel_dev); + int ret; + + if (bank_number >= hw_data->num_banks || !state) + return -EINVAL; + + dev_dbg(&GET_DEV(accel_dev), "Restoring state of bank %d\n", bank_number); + + ret = bank_state_restore(csr_ops, csr_base, bank_number, state, + hw_data->num_rings_per_bank, hw_data->tx_rx_gap); + if (ret) + dev_err(&GET_DEV(accel_dev), + "Unable to restore state of bank %d\n", bank_number); + + return ret; +} +EXPORT_SYMBOL_GPL(adf_bank_state_restore); diff --git a/drivers/crypto/intel/qat/qat_common/adf_bank_state.h b/drivers/crypto/intel/qat/qat_common/adf_bank_state.h new file mode 100644 index 0000000000000..48b573d692dd7 --- /dev/null +++ b/drivers/crypto/intel/qat/qat_common/adf_bank_state.h @@ -0,0 +1,49 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright(c) 2025 Intel Corporation */ +#ifndef ADF_BANK_STATE_H_ +#define ADF_BANK_STATE_H_ + +#include <linux/types.h> + +struct adf_accel_dev; + +struct ring_config { + u64 base; + u32 config; + u32 head; + u32 tail; + u32 reserved0; +}; + +struct adf_bank_state { + u32 ringstat0; + u32 ringstat1; + u32 ringuostat; + u32 ringestat; + u32 ringnestat; + u32 ringnfstat; + u32 ringfstat; + u32 ringcstat0; + u32 ringcstat1; + u32 ringcstat2; + u32 ringcstat3; + u32 iaintflagen; + u32 iaintflagreg; + u32 iaintflagsrcsel0; + u32 iaintflagsrcsel1; + u32 iaintcolen; + u32 iaintcolctl; + u32 iaintflagandcolen; + u32 ringexpstat; + u32 ringexpintenable; + u32 ringsrvarben; + u32 reserved0; + struct ring_config rings[ADF_ETR_MAX_RINGS_PER_BANK]; +}; + +int adf_bank_state_restore(struct adf_accel_dev *accel_dev, u32 bank_number, + struct adf_bank_state *state); +int adf_bank_state_save(struct adf_accel_dev *accel_dev, u32 bank_number, + struct adf_bank_state *state); + +#endif diff --git a/drivers/crypto/intel/qat/qat_common/adf_cfg_common.h b/drivers/crypto/intel/qat/qat_common/adf_cfg_common.h index 15fdf9854b814..81e9e9d7eccd2 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_cfg_common.h +++ b/drivers/crypto/intel/qat/qat_common/adf_cfg_common.h @@ -29,6 +29,7 @@ enum adf_cfg_service_type { COMP, SYM, ASYM, + DECOMP, USED }; diff --git a/drivers/crypto/intel/qat/qat_common/adf_cfg_services.c b/drivers/crypto/intel/qat/qat_common/adf_cfg_services.c index c39871291da72..7d00bcb41ce70 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_cfg_services.c +++ b/drivers/crypto/intel/qat/qat_common/adf_cfg_services.c @@ -7,6 +7,7 @@ #include <linux/pci.h> #include <linux/string.h> #include "adf_cfg.h" +#include "adf_cfg_common.h" #include "adf_cfg_services.h" #include "adf_cfg_strings.h" @@ -15,13 +16,14 @@ static const char *const adf_cfg_services[] = { [SVC_SYM] = ADF_CFG_SYM, [SVC_DC] = ADF_CFG_DC, [SVC_DCC] = ADF_CFG_DCC, + [SVC_DECOMP] = ADF_CFG_DECOMP, }; /* * Ensure that the size of the array matches the number of services, - * SVC_BASE_COUNT, that is used to size the bitmap. + * SVC_COUNT, that is used to size the bitmap. */ -static_assert(ARRAY_SIZE(adf_cfg_services) == SVC_BASE_COUNT); +static_assert(ARRAY_SIZE(adf_cfg_services) == SVC_COUNT); /* * Ensure that the maximum number of concurrent services that can be @@ -34,7 +36,7 @@ static_assert(ARRAY_SIZE(adf_cfg_services) >= MAX_NUM_CONCURR_SVC); * Ensure that the number of services fit a single unsigned long, as each * service is represented by a bit in the mask. */ -static_assert(BITS_PER_LONG >= SVC_BASE_COUNT); +static_assert(BITS_PER_LONG >= SVC_COUNT); /* * Ensure that size of the concatenation of all service strings is smaller @@ -43,6 +45,7 @@ static_assert(BITS_PER_LONG >= SVC_BASE_COUNT); static_assert(sizeof(ADF_CFG_SYM ADF_SERVICES_DELIMITER ADF_CFG_ASYM ADF_SERVICES_DELIMITER ADF_CFG_DC ADF_SERVICES_DELIMITER + ADF_CFG_DECOMP ADF_SERVICES_DELIMITER ADF_CFG_DCC) < ADF_CFG_MAX_VAL_LEN_IN_BYTES); static int adf_service_string_to_mask(struct adf_accel_dev *accel_dev, const char *buf, @@ -88,7 +91,7 @@ static int adf_service_mask_to_string(unsigned long mask, char *buf, size_t len) if (len < ADF_CFG_MAX_VAL_LEN_IN_BYTES) return -ENOSPC; - for_each_set_bit(bit, &mask, SVC_BASE_COUNT) { + for_each_set_bit(bit, &mask, SVC_COUNT) { if (offset) offset += scnprintf(buf + offset, len - offset, ADF_SERVICES_DELIMITER); @@ -167,9 +170,43 @@ int adf_get_service_enabled(struct adf_accel_dev *accel_dev) if (test_bit(SVC_DC, &mask)) return SVC_DC; + if (test_bit(SVC_DECOMP, &mask)) + return SVC_DECOMP; + if (test_bit(SVC_DCC, &mask)) return SVC_DCC; return -EINVAL; } EXPORT_SYMBOL_GPL(adf_get_service_enabled); + +enum adf_cfg_service_type adf_srv_to_cfg_svc_type(enum adf_base_services svc) +{ + switch (svc) { + case SVC_ASYM: + return ASYM; + case SVC_SYM: + return SYM; + case SVC_DC: + return COMP; + case SVC_DECOMP: + return DECOMP; + default: + return UNUSED; + } +} + +bool adf_is_service_enabled(struct adf_accel_dev *accel_dev, enum adf_base_services svc) +{ + enum adf_cfg_service_type arb_srv = adf_srv_to_cfg_svc_type(svc); + struct adf_hw_device_data *hw_data = GET_HW_DATA(accel_dev); + u8 rps_per_bundle = hw_data->num_banks_per_vf; + int i; + + for (i = 0; i < rps_per_bundle; i++) { + if (GET_SRV_TYPE(accel_dev, i) == arb_srv) + return true; + } + + return false; +} diff --git a/drivers/crypto/intel/qat/qat_common/adf_cfg_services.h b/drivers/crypto/intel/qat/qat_common/adf_cfg_services.h index 3742c450878f5..913d717280afa 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_cfg_services.h +++ b/drivers/crypto/intel/qat/qat_common/adf_cfg_services.h @@ -7,16 +7,21 @@ struct adf_accel_dev; -enum adf_services { +enum adf_base_services { SVC_ASYM = 0, SVC_SYM, SVC_DC, - SVC_DCC, + SVC_DECOMP, SVC_BASE_COUNT }; +enum adf_extended_services { + SVC_DCC = SVC_BASE_COUNT, + SVC_COUNT +}; + enum adf_composed_services { - SVC_SYM_ASYM = SVC_BASE_COUNT, + SVC_SYM_ASYM = SVC_COUNT, SVC_SYM_DC, SVC_ASYM_DC, }; @@ -33,5 +38,7 @@ int adf_parse_service_string(struct adf_accel_dev *accel_dev, const char *in, size_t in_len, char *out, size_t out_len); int adf_get_service_enabled(struct adf_accel_dev *accel_dev); int adf_get_service_mask(struct adf_accel_dev *accel_dev, unsigned long *mask); +enum adf_cfg_service_type adf_srv_to_cfg_svc_type(enum adf_base_services svc); +bool adf_is_service_enabled(struct adf_accel_dev *accel_dev, enum adf_base_services svc); #endif diff --git a/drivers/crypto/intel/qat/qat_common/adf_cfg_strings.h b/drivers/crypto/intel/qat/qat_common/adf_cfg_strings.h index b79982c4a8565..30107a02ee7f4 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_cfg_strings.h +++ b/drivers/crypto/intel/qat/qat_common/adf_cfg_strings.h @@ -24,6 +24,7 @@ #define ADF_CY "Cy" #define ADF_DC "Dc" #define ADF_CFG_DC "dc" +#define ADF_CFG_DECOMP "decomp" #define ADF_CFG_CY "sym;asym" #define ADF_CFG_SYM "sym" #define ADF_CFG_ASYM "asym" diff --git a/drivers/crypto/intel/qat/qat_common/adf_common_drv.h b/drivers/crypto/intel/qat/qat_common/adf_common_drv.h index eaa6388a6678b..6cf3a95489e8a 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_common_drv.h +++ b/drivers/crypto/intel/qat/qat_common/adf_common_drv.h @@ -86,7 +86,6 @@ int adf_ae_stop(struct adf_accel_dev *accel_dev); extern const struct pci_error_handlers adf_err_handler; void adf_reset_sbr(struct adf_accel_dev *accel_dev); void adf_reset_flr(struct adf_accel_dev *accel_dev); -int adf_dev_autoreset(struct adf_accel_dev *accel_dev); void adf_dev_restore(struct adf_accel_dev *accel_dev); int adf_init_aer(void); void adf_exit_aer(void); @@ -189,6 +188,7 @@ void adf_exit_misc_wq(void); bool adf_misc_wq_queue_work(struct work_struct *work); bool adf_misc_wq_queue_delayed_work(struct delayed_work *work, unsigned long delay); +void adf_misc_wq_flush(void); #if defined(CONFIG_PCI_IOV) int adf_sriov_configure(struct pci_dev *pdev, int numvfs); void adf_disable_sriov(struct adf_accel_dev *accel_dev); diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.c b/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.c index 0406cb09c5bbb..349fdb323763c 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.c +++ b/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.c @@ -1,5 +1,8 @@ // SPDX-License-Identifier: (BSD-3-Clause OR GPL-2.0-only) /* Copyright(c) 2020 Intel Corporation */ + +#define pr_fmt(fmt) "QAT: " fmt + #include <linux/bitops.h> #include <linux/iopoll.h> #include <asm/div64.h> @@ -259,7 +262,10 @@ bool adf_gen4_services_supported(unsigned long mask) { unsigned long num_svc = hweight_long(mask); - if (mask >= BIT(SVC_BASE_COUNT)) + if (mask >= BIT(SVC_COUNT)) + return false; + + if (test_bit(SVC_DECOMP, &mask)) return false; switch (num_svc) { @@ -485,187 +491,6 @@ int adf_gen4_bank_drain_start(struct adf_accel_dev *accel_dev, return ret; } -static void bank_state_save(struct adf_hw_csr_ops *ops, void __iomem *base, - u32 bank, struct bank_state *state, u32 num_rings) -{ - u32 i; - - state->ringstat0 = ops->read_csr_stat(base, bank); - state->ringuostat = ops->read_csr_uo_stat(base, bank); - state->ringestat = ops->read_csr_e_stat(base, bank); - state->ringnestat = ops->read_csr_ne_stat(base, bank); - state->ringnfstat = ops->read_csr_nf_stat(base, bank); - state->ringfstat = ops->read_csr_f_stat(base, bank); - state->ringcstat0 = ops->read_csr_c_stat(base, bank); - state->iaintflagen = ops->read_csr_int_en(base, bank); - state->iaintflagreg = ops->read_csr_int_flag(base, bank); - state->iaintflagsrcsel0 = ops->read_csr_int_srcsel(base, bank); - state->iaintcolen = ops->read_csr_int_col_en(base, bank); - state->iaintcolctl = ops->read_csr_int_col_ctl(base, bank); - state->iaintflagandcolen = ops->read_csr_int_flag_and_col(base, bank); - state->ringexpstat = ops->read_csr_exp_stat(base, bank); - state->ringexpintenable = ops->read_csr_exp_int_en(base, bank); - state->ringsrvarben = ops->read_csr_ring_srv_arb_en(base, bank); - - for (i = 0; i < num_rings; i++) { - state->rings[i].head = ops->read_csr_ring_head(base, bank, i); - state->rings[i].tail = ops->read_csr_ring_tail(base, bank, i); - state->rings[i].config = ops->read_csr_ring_config(base, bank, i); - state->rings[i].base = ops->read_csr_ring_base(base, bank, i); - } -} - -#define CHECK_STAT(op, expect_val, name, args...) \ -({ \ - u32 __expect_val = (expect_val); \ - u32 actual_val = op(args); \ - (__expect_val == actual_val) ? 0 : \ - (pr_err("QAT: Fail to restore %s register. Expected 0x%x, actual 0x%x\n", \ - name, __expect_val, actual_val), -EINVAL); \ -}) - -static int bank_state_restore(struct adf_hw_csr_ops *ops, void __iomem *base, - u32 bank, struct bank_state *state, u32 num_rings, - int tx_rx_gap) -{ - u32 val, tmp_val, i; - int ret; - - for (i = 0; i < num_rings; i++) - ops->write_csr_ring_base(base, bank, i, state->rings[i].base); - - for (i = 0; i < num_rings; i++) - ops->write_csr_ring_config(base, bank, i, state->rings[i].config); - - for (i = 0; i < num_rings / 2; i++) { - int tx = i * (tx_rx_gap + 1); - int rx = tx + tx_rx_gap; - - ops->write_csr_ring_head(base, bank, tx, state->rings[tx].head); - ops->write_csr_ring_tail(base, bank, tx, state->rings[tx].tail); - - /* - * The TX ring head needs to be updated again to make sure that - * the HW will not consider the ring as full when it is empty - * and the correct state flags are set to match the recovered state. - */ - if (state->ringestat & BIT(tx)) { - val = ops->read_csr_int_srcsel(base, bank); - val |= ADF_RP_INT_SRC_SEL_F_RISE_MASK; - ops->write_csr_int_srcsel_w_val(base, bank, val); - ops->write_csr_ring_head(base, bank, tx, state->rings[tx].head); - } - - ops->write_csr_ring_tail(base, bank, rx, state->rings[rx].tail); - val = ops->read_csr_int_srcsel(base, bank); - val |= ADF_RP_INT_SRC_SEL_F_RISE_MASK << ADF_RP_INT_SRC_SEL_RANGE_WIDTH; - ops->write_csr_int_srcsel_w_val(base, bank, val); - - ops->write_csr_ring_head(base, bank, rx, state->rings[rx].head); - val = ops->read_csr_int_srcsel(base, bank); - val |= ADF_RP_INT_SRC_SEL_F_FALL_MASK << ADF_RP_INT_SRC_SEL_RANGE_WIDTH; - ops->write_csr_int_srcsel_w_val(base, bank, val); - - /* - * The RX ring tail needs to be updated again to make sure that - * the HW will not consider the ring as empty when it is full - * and the correct state flags are set to match the recovered state. - */ - if (state->ringfstat & BIT(rx)) - ops->write_csr_ring_tail(base, bank, rx, state->rings[rx].tail); - } - - ops->write_csr_int_flag_and_col(base, bank, state->iaintflagandcolen); - ops->write_csr_int_en(base, bank, state->iaintflagen); - ops->write_csr_int_col_en(base, bank, state->iaintcolen); - ops->write_csr_int_srcsel_w_val(base, bank, state->iaintflagsrcsel0); - ops->write_csr_exp_int_en(base, bank, state->ringexpintenable); - ops->write_csr_int_col_ctl(base, bank, state->iaintcolctl); - ops->write_csr_ring_srv_arb_en(base, bank, state->ringsrvarben); - - /* Check that all ring statuses match the saved state. */ - ret = CHECK_STAT(ops->read_csr_stat, state->ringstat0, "ringstat", - base, bank); - if (ret) - return ret; - - ret = CHECK_STAT(ops->read_csr_e_stat, state->ringestat, "ringestat", - base, bank); - if (ret) - return ret; - - ret = CHECK_STAT(ops->read_csr_ne_stat, state->ringnestat, "ringnestat", - base, bank); - if (ret) - return ret; - - ret = CHECK_STAT(ops->read_csr_nf_stat, state->ringnfstat, "ringnfstat", - base, bank); - if (ret) - return ret; - - ret = CHECK_STAT(ops->read_csr_f_stat, state->ringfstat, "ringfstat", - base, bank); - if (ret) - return ret; - - ret = CHECK_STAT(ops->read_csr_c_stat, state->ringcstat0, "ringcstat", - base, bank); - if (ret) - return ret; - - tmp_val = ops->read_csr_exp_stat(base, bank); - val = state->ringexpstat; - if (tmp_val && !val) { - pr_err("QAT: Bank was restored with exception: 0x%x\n", val); - return -EINVAL; - } - - return 0; -} - -int adf_gen4_bank_state_save(struct adf_accel_dev *accel_dev, u32 bank_number, - struct bank_state *state) -{ - struct adf_hw_device_data *hw_data = GET_HW_DATA(accel_dev); - struct adf_hw_csr_ops *csr_ops = GET_CSR_OPS(accel_dev); - void __iomem *csr_base = adf_get_etr_base(accel_dev); - - if (bank_number >= hw_data->num_banks || !state) - return -EINVAL; - - dev_dbg(&GET_DEV(accel_dev), "Saving state of bank %d\n", bank_number); - - bank_state_save(csr_ops, csr_base, bank_number, state, - hw_data->num_rings_per_bank); - - return 0; -} -EXPORT_SYMBOL_GPL(adf_gen4_bank_state_save); - -int adf_gen4_bank_state_restore(struct adf_accel_dev *accel_dev, u32 bank_number, - struct bank_state *state) -{ - struct adf_hw_device_data *hw_data = GET_HW_DATA(accel_dev); - struct adf_hw_csr_ops *csr_ops = GET_CSR_OPS(accel_dev); - void __iomem *csr_base = adf_get_etr_base(accel_dev); - int ret; - - if (bank_number >= hw_data->num_banks || !state) - return -EINVAL; - - dev_dbg(&GET_DEV(accel_dev), "Restoring state of bank %d\n", bank_number); - - ret = bank_state_restore(csr_ops, csr_base, bank_number, state, - hw_data->num_rings_per_bank, hw_data->tx_rx_gap); - if (ret) - dev_err(&GET_DEV(accel_dev), - "Unable to restore state of bank %d\n", bank_number); - - return ret; -} -EXPORT_SYMBOL_GPL(adf_gen4_bank_state_restore); - static int adf_gen4_build_comp_block(void *ctx, enum adf_dc_algo algo) { struct icp_qat_fw_comp_req *req_tmpl = ctx; @@ -733,3 +558,43 @@ void adf_gen4_init_dc_ops(struct adf_dc_ops *dc_ops) dc_ops->build_decomp_block = adf_gen4_build_decomp_block; } EXPORT_SYMBOL_GPL(adf_gen4_init_dc_ops); + +void adf_gen4_init_num_svc_aes(struct adf_rl_hw_data *device_data) +{ + struct adf_hw_device_data *hw_data; + unsigned int i; + u32 ae_cnt; + + hw_data = container_of(device_data, struct adf_hw_device_data, rl_data); + ae_cnt = hweight32(hw_data->get_ae_mask(hw_data)); + if (!ae_cnt) + return; + + for (i = 0; i < SVC_BASE_COUNT; i++) + device_data->svc_ae_mask[i] = ae_cnt - 1; + + /* + * The decompression service is not supported on QAT GEN4 devices. + * Therefore, set svc_ae_mask to 0. + */ + device_data->svc_ae_mask[SVC_DECOMP] = 0; +} +EXPORT_SYMBOL_GPL(adf_gen4_init_num_svc_aes); + +u32 adf_gen4_get_svc_slice_cnt(struct adf_accel_dev *accel_dev, + enum adf_base_services svc) +{ + struct adf_rl_hw_data *device_data = &accel_dev->hw_device->rl_data; + + switch (svc) { + case SVC_SYM: + return device_data->slices.cph_cnt; + case SVC_ASYM: + return device_data->slices.pke_cnt; + case SVC_DC: + return device_data->slices.dcpr_cnt; + default: + return 0; + } +} +EXPORT_SYMBOL_GPL(adf_gen4_get_svc_slice_cnt); diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.h b/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.h index e4f4d5fa616de..cd26b6724c430 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.h +++ b/drivers/crypto/intel/qat/qat_common/adf_gen4_hw_data.h @@ -84,9 +84,6 @@ #define ADF_WQM_CSR_RPRESETSTS(bank) (ADF_WQM_CSR_RPRESETCTL(bank) + 4) /* Ring interrupt */ -#define ADF_RP_INT_SRC_SEL_F_RISE_MASK GENMASK(1, 0) -#define ADF_RP_INT_SRC_SEL_F_FALL_MASK GENMASK(2, 0) -#define ADF_RP_INT_SRC_SEL_RANGE_WIDTH 4 #define ADF_COALESCED_POLL_TIMEOUT_US (1 * USEC_PER_SEC) #define ADF_COALESCED_POLL_DELAY_US 1000 #define ADF_WQM_CSR_RPINTSOU(bank) (0x200000 + ((bank) << 12)) @@ -176,11 +173,10 @@ int adf_gen4_bank_drain_start(struct adf_accel_dev *accel_dev, u32 bank_number, int timeout_us); void adf_gen4_bank_drain_finish(struct adf_accel_dev *accel_dev, u32 bank_number); -int adf_gen4_bank_state_save(struct adf_accel_dev *accel_dev, u32 bank_number, - struct bank_state *state); -int adf_gen4_bank_state_restore(struct adf_accel_dev *accel_dev, - u32 bank_number, struct bank_state *state); bool adf_gen4_services_supported(unsigned long service_mask); void adf_gen4_init_dc_ops(struct adf_dc_ops *dc_ops); +void adf_gen4_init_num_svc_aes(struct adf_rl_hw_data *device_data); +u32 adf_gen4_get_svc_slice_cnt(struct adf_accel_dev *accel_dev, + enum adf_base_services svc); #endif diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen4_pm_debugfs.c b/drivers/crypto/intel/qat/qat_common/adf_gen4_pm_debugfs.c index 2e4095c4c12c9..b7e38842a46dc 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_gen4_pm_debugfs.c +++ b/drivers/crypto/intel/qat/qat_common/adf_gen4_pm_debugfs.c @@ -1,47 +1,18 @@ // SPDX-License-Identifier: GPL-2.0-only /* Copyright(c) 2023 Intel Corporation */ #include <linux/dma-mapping.h> -#include <linux/kernel.h> #include <linux/string_helpers.h> -#include <linux/stringify.h> #include "adf_accel_devices.h" #include "adf_admin.h" #include "adf_common_drv.h" #include "adf_gen4_pm.h" +#include "adf_pm_dbgfs_utils.h" #include "icp_qat_fw_init_admin.h" -/* - * This is needed because a variable is used to index the mask at - * pm_scnprint_table(), making it not compile time constant, so the compile - * asserts from FIELD_GET() or u32_get_bits() won't be fulfilled. - */ -#define field_get(_mask, _reg) (((_reg) & (_mask)) >> (ffs(_mask) - 1)) - -#define PM_INFO_MEMBER_OFF(member) \ - (offsetof(struct icp_qat_fw_init_admin_pm_info, member) / sizeof(u32)) - -#define PM_INFO_REGSET_ENTRY_MASK(_reg_, _field_, _mask_) \ -{ \ - .reg_offset = PM_INFO_MEMBER_OFF(_reg_), \ - .key = __stringify(_field_), \ - .field_mask = _mask_, \ -} - -#define PM_INFO_REGSET_ENTRY32(_reg_, _field_) \ - PM_INFO_REGSET_ENTRY_MASK(_reg_, _field_, GENMASK(31, 0)) - #define PM_INFO_REGSET_ENTRY(_reg_, _field_) \ PM_INFO_REGSET_ENTRY_MASK(_reg_, _field_, ADF_GEN4_PM_##_field_##_MASK) -#define PM_INFO_MAX_KEY_LEN 21 - -struct pm_status_row { - int reg_offset; - u32 field_mask; - const char *key; -}; - static const struct pm_status_row pm_fuse_rows[] = { PM_INFO_REGSET_ENTRY(fusectl0, ENABLE_PM), PM_INFO_REGSET_ENTRY(fusectl0, ENABLE_PM_IDLE), @@ -109,44 +80,6 @@ static const struct pm_status_row pm_csrs_rows[] = { PM_INFO_REGSET_ENTRY32(pm.pwrreq, CPM_PM_PWRREQ), }; -static int pm_scnprint_table(char *buff, const struct pm_status_row *table, - u32 *pm_info_regs, size_t buff_size, int table_len, - bool lowercase) -{ - char key[PM_INFO_MAX_KEY_LEN]; - int wr = 0; - int i; - - for (i = 0; i < table_len; i++) { - if (lowercase) - string_lower(key, table[i].key); - else - string_upper(key, table[i].key); - - wr += scnprintf(&buff[wr], buff_size - wr, "%s: %#x\n", key, - field_get(table[i].field_mask, - pm_info_regs[table[i].reg_offset])); - } - - return wr; -} - -static int pm_scnprint_table_upper_keys(char *buff, const struct pm_status_row *table, - u32 *pm_info_regs, size_t buff_size, - int table_len) -{ - return pm_scnprint_table(buff, table, pm_info_regs, buff_size, - table_len, false); -} - -static int pm_scnprint_table_lower_keys(char *buff, const struct pm_status_row *table, - u32 *pm_info_regs, size_t buff_size, - int table_len) -{ - return pm_scnprint_table(buff, table, pm_info_regs, buff_size, - table_len, true); -} - static_assert(sizeof(struct icp_qat_fw_init_admin_pm_info) < PAGE_SIZE); static ssize_t adf_gen4_print_pm_status(struct adf_accel_dev *accel_dev, @@ -191,9 +124,9 @@ static ssize_t adf_gen4_print_pm_status(struct adf_accel_dev *accel_dev, /* Fusectl related */ len += scnprintf(&pm_kv[len], PAGE_SIZE - len, "----------- PM Fuse info ---------\n"); - len += pm_scnprint_table_lower_keys(&pm_kv[len], pm_fuse_rows, - pm_info_regs, PAGE_SIZE - len, - ARRAY_SIZE(pm_fuse_rows)); + len += adf_pm_scnprint_table_lower_keys(&pm_kv[len], pm_fuse_rows, + pm_info_regs, PAGE_SIZE - len, + ARRAY_SIZE(pm_fuse_rows)); len += scnprintf(&pm_kv[len], PAGE_SIZE - len, "max_pwrreq: %#x\n", pm_info->max_pwrreq); len += scnprintf(&pm_kv[len], PAGE_SIZE - len, "min_pwrreq: %#x\n", @@ -204,28 +137,28 @@ static ssize_t adf_gen4_print_pm_status(struct adf_accel_dev *accel_dev, "------------ PM Info ------------\n"); len += scnprintf(&pm_kv[len], PAGE_SIZE - len, "power_level: %s\n", pm_info->pwr_state == PM_SET_MIN ? "min" : "max"); - len += pm_scnprint_table_lower_keys(&pm_kv[len], pm_info_rows, - pm_info_regs, PAGE_SIZE - len, - ARRAY_SIZE(pm_info_rows)); + len += adf_pm_scnprint_table_lower_keys(&pm_kv[len], pm_info_rows, + pm_info_regs, PAGE_SIZE - len, + ARRAY_SIZE(pm_info_rows)); len += scnprintf(&pm_kv[len], PAGE_SIZE - len, "pm_mode: STATIC\n"); /* SSM related */ len += scnprintf(&pm_kv[len], PAGE_SIZE - len, "----------- SSM_PM Info ----------\n"); - len += pm_scnprint_table_lower_keys(&pm_kv[len], pm_ssm_rows, - pm_info_regs, PAGE_SIZE - len, - ARRAY_SIZE(pm_ssm_rows)); + len += adf_pm_scnprint_table_lower_keys(&pm_kv[len], pm_ssm_rows, + pm_info_regs, PAGE_SIZE - len, + ARRAY_SIZE(pm_ssm_rows)); /* Log related */ len += scnprintf(&pm_kv[len], PAGE_SIZE - len, "------------- PM Log -------------\n"); - len += pm_scnprint_table_lower_keys(&pm_kv[len], pm_log_rows, - pm_info_regs, PAGE_SIZE - len, - ARRAY_SIZE(pm_log_rows)); + len += adf_pm_scnprint_table_lower_keys(&pm_kv[len], pm_log_rows, + pm_info_regs, PAGE_SIZE - len, + ARRAY_SIZE(pm_log_rows)); - len += pm_scnprint_table_lower_keys(&pm_kv[len], pm_event_rows, - pm_info_regs, PAGE_SIZE - len, - ARRAY_SIZE(pm_event_rows)); + len += adf_pm_scnprint_table_lower_keys(&pm_kv[len], pm_event_rows, + pm_info_regs, PAGE_SIZE - len, + ARRAY_SIZE(pm_event_rows)); len += scnprintf(&pm_kv[len], PAGE_SIZE - len, "idle_irq_count: %#x\n", pm->idle_irq_counters); @@ -241,9 +174,9 @@ static ssize_t adf_gen4_print_pm_status(struct adf_accel_dev *accel_dev, /* CSRs content */ len += scnprintf(&pm_kv[len], PAGE_SIZE - len, "----------- HW PM CSRs -----------\n"); - len += pm_scnprint_table_upper_keys(&pm_kv[len], pm_csrs_rows, - pm_info_regs, PAGE_SIZE - len, - ARRAY_SIZE(pm_csrs_rows)); + len += adf_pm_scnprint_table_upper_keys(&pm_kv[len], pm_csrs_rows, + pm_info_regs, PAGE_SIZE - len, + ARRAY_SIZE(pm_csrs_rows)); val = ADF_CSR_RD(pmisc, ADF_GEN4_PM_HOST_MSG); len += scnprintf(&pm_kv[len], PAGE_SIZE - len, diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen4_vf_mig.c b/drivers/crypto/intel/qat/qat_common/adf_gen4_vf_mig.c index a62eb5e8dbe6a..adb21656a3ba8 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_gen4_vf_mig.c +++ b/drivers/crypto/intel/qat/qat_common/adf_gen4_vf_mig.c @@ -9,6 +9,7 @@ #include <asm/errno.h> #include "adf_accel_devices.h" +#include "adf_bank_state.h" #include "adf_common_drv.h" #include "adf_gen4_hw_data.h" #include "adf_gen4_pfvf.h" @@ -358,7 +359,7 @@ static int adf_gen4_vfmig_load_etr_regs(struct adf_mstate_mgr *sub_mgr, pf_bank_nr = vf_bank_info->bank_nr + vf_bank_info->vf_nr * hw_data->num_banks_per_vf; ret = hw_data->bank_state_restore(accel_dev, pf_bank_nr, - (struct bank_state *)state); + (struct adf_bank_state *)state); if (ret) { dev_err(&GET_DEV(accel_dev), "Failed to load regs for vf%d bank%d\n", @@ -585,7 +586,7 @@ static int adf_gen4_vfmig_save_etr_regs(struct adf_mstate_mgr *subs, u8 *state, pf_bank_nr += vf_bank_info->vf_nr * hw_data->num_banks_per_vf; ret = hw_data->bank_state_save(accel_dev, pf_bank_nr, - (struct bank_state *)state); + (struct adf_bank_state *)state); if (ret) { dev_err(&GET_DEV(accel_dev), "Failed to save regs for vf%d bank%d\n", @@ -593,7 +594,7 @@ static int adf_gen4_vfmig_save_etr_regs(struct adf_mstate_mgr *subs, u8 *state, return ret; } - return sizeof(struct bank_state); + return sizeof(struct adf_bank_state); } static int adf_gen4_vfmig_save_etr_bank(struct adf_accel_dev *accel_dev, diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen6_pm.h b/drivers/crypto/intel/qat/qat_common/adf_gen6_pm.h index 9a5b995f7ada2..4c0d576e8c218 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_gen6_pm.h +++ b/drivers/crypto/intel/qat/qat_common/adf_gen6_pm.h @@ -24,5 +24,29 @@ struct adf_accel_dev; /* cpm_pm_status bitfields */ #define ADF_GEN6_PM_INIT_STATE BIT(21) +#define ADF_GEN6_PM_CPM_PM_STATE_MASK GENMASK(22, 20) + +/* fusectl0 bitfields */ +#define ADF_GEN6_PM_ENABLE_PM_MASK BIT(21) +#define ADF_GEN6_PM_ENABLE_PM_IDLE_MASK BIT(22) +#define ADF_GEN6_PM_ENABLE_DEEP_PM_IDLE_MASK BIT(23) + +/* cpm_pm_fw_init bitfields */ +#define ADF_GEN6_PM_IDLE_FILTER_MASK GENMASK(5, 3) +#define ADF_GEN6_PM_IDLE_ENABLE_MASK BIT(2) + +/* ssm_pm_enable bitfield */ +#define ADF_GEN6_PM_SSM_PM_ENABLE_MASK BIT(0) + +/* ssm_pm_domain_status bitfield */ +#define ADF_GEN6_PM_DOMAIN_POWERED_UP_MASK BIT(0) + +#ifdef CONFIG_DEBUG_FS +void adf_gen6_init_dev_pm_data(struct adf_accel_dev *accel_dev); +#else +static inline void adf_gen6_init_dev_pm_data(struct adf_accel_dev *accel_dev) +{ +} +#endif /* CONFIG_DEBUG_FS */ #endif /* ADF_GEN6_PM_H */ diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen6_pm_dbgfs.c b/drivers/crypto/intel/qat/qat_common/adf_gen6_pm_dbgfs.c new file mode 100644 index 0000000000000..603aefba0fdba --- /dev/null +++ b/drivers/crypto/intel/qat/qat_common/adf_gen6_pm_dbgfs.c @@ -0,0 +1,124 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright(c) 2025 Intel Corporation */ +#include <linux/dma-mapping.h> +#include <linux/export.h> +#include <linux/string_helpers.h> + +#include "adf_admin.h" +#include "adf_common_drv.h" +#include "adf_gen6_pm.h" +#include "adf_pm_dbgfs_utils.h" +#include "icp_qat_fw_init_admin.h" + +#define PM_INFO_REGSET_ENTRY(_reg_, _field_) \ + PM_INFO_REGSET_ENTRY_MASK(_reg_, _field_, ADF_GEN6_PM_##_field_##_MASK) + +static struct pm_status_row pm_fuse_rows[] = { + PM_INFO_REGSET_ENTRY(fusectl0, ENABLE_PM), + PM_INFO_REGSET_ENTRY(fusectl0, ENABLE_PM_IDLE), + PM_INFO_REGSET_ENTRY(fusectl0, ENABLE_DEEP_PM_IDLE), +}; + +static struct pm_status_row pm_info_rows[] = { + PM_INFO_REGSET_ENTRY(pm.status, CPM_PM_STATE), + PM_INFO_REGSET_ENTRY(pm.fw_init, IDLE_ENABLE), + PM_INFO_REGSET_ENTRY(pm.fw_init, IDLE_FILTER), +}; + +static struct pm_status_row pm_ssm_rows[] = { + PM_INFO_REGSET_ENTRY(ssm.pm_enable, SSM_PM_ENABLE), + PM_INFO_REGSET_ENTRY(ssm.pm_domain_status, DOMAIN_POWERED_UP), +}; + +static struct pm_status_row pm_csrs_rows[] = { + PM_INFO_REGSET_ENTRY32(pm.fw_init, CPM_PM_FW_INIT), + PM_INFO_REGSET_ENTRY32(pm.status, CPM_PM_STATUS), +}; + +static_assert(sizeof(struct icp_qat_fw_init_admin_pm_info) < PAGE_SIZE); + +static ssize_t adf_gen6_print_pm_status(struct adf_accel_dev *accel_dev, + char __user *buf, size_t count, + loff_t *pos) +{ + void __iomem *pmisc = adf_get_pmisc_base(accel_dev); + struct icp_qat_fw_init_admin_pm_info *pm_info; + dma_addr_t p_state_addr; + u32 *pm_info_regs; + size_t len = 0; + char *pm_kv; + u32 val; + int ret; + + pm_info = kzalloc(PAGE_SIZE, GFP_KERNEL); + if (!pm_info) + return -ENOMEM; + + pm_kv = kzalloc(PAGE_SIZE, GFP_KERNEL); + if (!pm_kv) { + kfree(pm_info); + return -ENOMEM; + } + + p_state_addr = dma_map_single(&GET_DEV(accel_dev), pm_info, PAGE_SIZE, + DMA_FROM_DEVICE); + ret = dma_mapping_error(&GET_DEV(accel_dev), p_state_addr); + if (ret) + goto out_free; + + /* Query power management information from QAT FW */ + ret = adf_get_pm_info(accel_dev, p_state_addr, PAGE_SIZE); + dma_unmap_single(&GET_DEV(accel_dev), p_state_addr, PAGE_SIZE, + DMA_FROM_DEVICE); + if (ret) + goto out_free; + + pm_info_regs = (u32 *)pm_info; + + /* Fuse control register */ + len += scnprintf(&pm_kv[len], PAGE_SIZE - len, + "----------- PM Fuse info ---------\n"); + len += adf_pm_scnprint_table_lower_keys(&pm_kv[len], pm_fuse_rows, + pm_info_regs, PAGE_SIZE - len, + ARRAY_SIZE(pm_fuse_rows)); + + /* Power management */ + len += scnprintf(&pm_kv[len], PAGE_SIZE - len, + "----------- PM Info --------------\n"); + + len += adf_pm_scnprint_table_lower_keys(&pm_kv[len], pm_info_rows, + pm_info_regs, PAGE_SIZE - len, + ARRAY_SIZE(pm_info_rows)); + len += scnprintf(&pm_kv[len], PAGE_SIZE - len, "pm_mode: ACTIVE\n"); + + /* Shared Slice Module */ + len += scnprintf(&pm_kv[len], PAGE_SIZE - len, + "----------- SSM_PM Info ----------\n"); + len += adf_pm_scnprint_table_lower_keys(&pm_kv[len], pm_ssm_rows, + pm_info_regs, PAGE_SIZE - len, + ARRAY_SIZE(pm_ssm_rows)); + + /* Control status register content */ + len += scnprintf(&pm_kv[len], PAGE_SIZE - len, + "----------- HW PM CSRs -----------\n"); + len += adf_pm_scnprint_table_upper_keys(&pm_kv[len], pm_csrs_rows, + pm_info_regs, PAGE_SIZE - len, + ARRAY_SIZE(pm_csrs_rows)); + + val = ADF_CSR_RD(pmisc, ADF_GEN6_PM_INTERRUPT); + len += scnprintf(&pm_kv[len], PAGE_SIZE - len, "CPM_PM_INTERRUPT: %#x\n", val); + ret = simple_read_from_buffer(buf, count, pos, pm_kv, len); + +out_free: + kfree(pm_info); + kfree(pm_kv); + + return ret; +} + +void adf_gen6_init_dev_pm_data(struct adf_accel_dev *accel_dev) +{ + accel_dev->power_management.print_pm_status = adf_gen6_print_pm_status; + accel_dev->power_management.present = true; +} +EXPORT_SYMBOL_GPL(adf_gen6_init_dev_pm_data); diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen6_shared.c b/drivers/crypto/intel/qat/qat_common/adf_gen6_shared.c index 58a072e2f9362..c9b151006dcae 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_gen6_shared.c +++ b/drivers/crypto/intel/qat/qat_common/adf_gen6_shared.c @@ -5,6 +5,7 @@ #include "adf_gen4_config.h" #include "adf_gen4_hw_csr_data.h" #include "adf_gen4_pfvf.h" +#include "adf_gen4_vf_mig.h" #include "adf_gen6_shared.h" struct adf_accel_dev; @@ -47,3 +48,9 @@ int adf_gen6_no_dev_config(struct adf_accel_dev *accel_dev) return adf_no_dev_config(accel_dev); } EXPORT_SYMBOL_GPL(adf_gen6_no_dev_config); + +void adf_gen6_init_vf_mig_ops(struct qat_migdev_ops *vfmig_ops) +{ + adf_gen4_init_vf_mig_ops(vfmig_ops); +} +EXPORT_SYMBOL_GPL(adf_gen6_init_vf_mig_ops); diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen6_shared.h b/drivers/crypto/intel/qat/qat_common/adf_gen6_shared.h index bc8e71e984fc7..fc6fad029a70d 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_gen6_shared.h +++ b/drivers/crypto/intel/qat/qat_common/adf_gen6_shared.h @@ -4,6 +4,7 @@ #define ADF_GEN6_SHARED_H_ struct adf_hw_csr_ops; +struct qat_migdev_ops; struct adf_accel_dev; struct adf_pfvf_ops; @@ -12,4 +13,5 @@ void adf_gen6_init_hw_csr_ops(struct adf_hw_csr_ops *csr_ops); int adf_gen6_cfg_dev_init(struct adf_accel_dev *accel_dev); int adf_gen6_comp_dev_config(struct adf_accel_dev *accel_dev); int adf_gen6_no_dev_config(struct adf_accel_dev *accel_dev); +void adf_gen6_init_vf_mig_ops(struct qat_migdev_ops *vfmig_ops); #endif/* ADF_GEN6_SHARED_H_ */ diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen6_tl.c b/drivers/crypto/intel/qat/qat_common/adf_gen6_tl.c new file mode 100644 index 0000000000000..cf804f95838a3 --- /dev/null +++ b/drivers/crypto/intel/qat/qat_common/adf_gen6_tl.c @@ -0,0 +1,146 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright (c) 2025 Intel Corporation. */ +#include <linux/export.h> + +#include "adf_gen6_tl.h" +#include "adf_telemetry.h" +#include "adf_tl_debugfs.h" +#include "icp_qat_fw_init_admin.h" + +#define ADF_GEN6_TL_DEV_REG_OFF(reg) ADF_TL_DEV_REG_OFF(reg, gen6) + +#define ADF_GEN6_TL_RP_REG_OFF(reg) ADF_TL_RP_REG_OFF(reg, gen6) + +#define ADF_GEN6_TL_SL_UTIL_COUNTER(_name) \ + ADF_TL_COUNTER("util_" #_name, ADF_TL_SIMPLE_COUNT, \ + ADF_TL_SLICE_REG_OFF(_name, reg_tm_slice_util, gen6)) + +#define ADF_GEN6_TL_SL_EXEC_COUNTER(_name) \ + ADF_TL_COUNTER("exec_" #_name, ADF_TL_SIMPLE_COUNT, \ + ADF_TL_SLICE_REG_OFF(_name, reg_tm_slice_exec_cnt, gen6)) + +#define SLICE_IDX(sl) offsetof(struct icp_qat_fw_init_admin_slice_cnt, sl##_cnt) + +/* Device level counters. */ +static const struct adf_tl_dbg_counter dev_counters[] = { + /* PCIe partial transactions. */ + ADF_TL_COUNTER(PCI_TRANS_CNT_NAME, ADF_TL_SIMPLE_COUNT, + ADF_GEN6_TL_DEV_REG_OFF(reg_tl_prt_trans_cnt)), + /* Max read latency[ns]. */ + ADF_TL_COUNTER(MAX_RD_LAT_NAME, ADF_TL_COUNTER_NS, + ADF_GEN6_TL_DEV_REG_OFF(reg_tl_rd_lat_max)), + /* Read latency average[ns]. */ + ADF_TL_COUNTER_LATENCY(RD_LAT_ACC_NAME, ADF_TL_COUNTER_NS_AVG, + ADF_GEN6_TL_DEV_REG_OFF(reg_tl_rd_lat_acc), + ADF_GEN6_TL_DEV_REG_OFF(reg_tl_rd_cmpl_cnt)), + /* Max "get to put" latency[ns]. */ + ADF_TL_COUNTER(MAX_LAT_NAME, ADF_TL_COUNTER_NS, + ADF_GEN6_TL_DEV_REG_OFF(reg_tl_gp_lat_max)), + /* "Get to put" latency average[ns]. */ + ADF_TL_COUNTER_LATENCY(LAT_ACC_NAME, ADF_TL_COUNTER_NS_AVG, + ADF_GEN6_TL_DEV_REG_OFF(reg_tl_gp_lat_acc), + ADF_GEN6_TL_DEV_REG_OFF(reg_tl_ae_put_cnt)), + /* PCIe write bandwidth[Mbps]. */ + ADF_TL_COUNTER(BW_IN_NAME, ADF_TL_COUNTER_MBPS, + ADF_GEN6_TL_DEV_REG_OFF(reg_tl_bw_in)), + /* PCIe read bandwidth[Mbps]. */ + ADF_TL_COUNTER(BW_OUT_NAME, ADF_TL_COUNTER_MBPS, + ADF_GEN6_TL_DEV_REG_OFF(reg_tl_bw_out)), + /* Page request latency average[ns]. */ + ADF_TL_COUNTER_LATENCY(PAGE_REQ_LAT_NAME, ADF_TL_COUNTER_NS_AVG, + ADF_GEN6_TL_DEV_REG_OFF(reg_tl_at_page_req_lat_acc), + ADF_GEN6_TL_DEV_REG_OFF(reg_tl_at_page_req_cnt)), + /* Page translation latency average[ns]. */ + ADF_TL_COUNTER_LATENCY(AT_TRANS_LAT_NAME, ADF_TL_COUNTER_NS_AVG, + ADF_GEN6_TL_DEV_REG_OFF(reg_tl_at_trans_lat_acc), + ADF_GEN6_TL_DEV_REG_OFF(reg_tl_at_trans_lat_cnt)), + /* Maximum uTLB used. */ + ADF_TL_COUNTER(AT_MAX_UTLB_USED_NAME, ADF_TL_SIMPLE_COUNT, + ADF_GEN6_TL_DEV_REG_OFF(reg_tl_at_max_utlb_used)), +}; + +/* Accelerator utilization counters */ +static const struct adf_tl_dbg_counter sl_util_counters[ADF_TL_SL_CNT_COUNT] = { + /* Compression accelerator utilization. */ + [SLICE_IDX(cpr)] = ADF_GEN6_TL_SL_UTIL_COUNTER(cnv), + /* Decompression accelerator utilization. */ + [SLICE_IDX(dcpr)] = ADF_GEN6_TL_SL_UTIL_COUNTER(dcprz), + /* PKE accelerator utilization. */ + [SLICE_IDX(pke)] = ADF_GEN6_TL_SL_UTIL_COUNTER(pke), + /* Wireless Authentication accelerator utilization. */ + [SLICE_IDX(wat)] = ADF_GEN6_TL_SL_UTIL_COUNTER(wat), + /* Wireless Cipher accelerator utilization. */ + [SLICE_IDX(wcp)] = ADF_GEN6_TL_SL_UTIL_COUNTER(wcp), + /* UCS accelerator utilization. */ + [SLICE_IDX(ucs)] = ADF_GEN6_TL_SL_UTIL_COUNTER(ucs), + /* Authentication accelerator utilization. */ + [SLICE_IDX(ath)] = ADF_GEN6_TL_SL_UTIL_COUNTER(ath), +}; + +/* Accelerator execution counters */ +static const struct adf_tl_dbg_counter sl_exec_counters[ADF_TL_SL_CNT_COUNT] = { + /* Compression accelerator execution count. */ + [SLICE_IDX(cpr)] = ADF_GEN6_TL_SL_EXEC_COUNTER(cnv), + /* Decompression accelerator execution count. */ + [SLICE_IDX(dcpr)] = ADF_GEN6_TL_SL_EXEC_COUNTER(dcprz), + /* PKE execution count. */ + [SLICE_IDX(pke)] = ADF_GEN6_TL_SL_EXEC_COUNTER(pke), + /* Wireless Authentication accelerator execution count. */ + [SLICE_IDX(wat)] = ADF_GEN6_TL_SL_EXEC_COUNTER(wat), + /* Wireless Cipher accelerator execution count. */ + [SLICE_IDX(wcp)] = ADF_GEN6_TL_SL_EXEC_COUNTER(wcp), + /* UCS accelerator execution count. */ + [SLICE_IDX(ucs)] = ADF_GEN6_TL_SL_EXEC_COUNTER(ucs), + /* Authentication accelerator execution count. */ + [SLICE_IDX(ath)] = ADF_GEN6_TL_SL_EXEC_COUNTER(ath), +}; + +/* Ring pair counters. */ +static const struct adf_tl_dbg_counter rp_counters[] = { + /* PCIe partial transactions. */ + ADF_TL_COUNTER(PCI_TRANS_CNT_NAME, ADF_TL_SIMPLE_COUNT, + ADF_GEN6_TL_RP_REG_OFF(reg_tl_prt_trans_cnt)), + /* "Get to put" latency average[ns]. */ + ADF_TL_COUNTER_LATENCY(LAT_ACC_NAME, ADF_TL_COUNTER_NS_AVG, + ADF_GEN6_TL_RP_REG_OFF(reg_tl_gp_lat_acc), + ADF_GEN6_TL_RP_REG_OFF(reg_tl_ae_put_cnt)), + /* PCIe write bandwidth[Mbps]. */ + ADF_TL_COUNTER(BW_IN_NAME, ADF_TL_COUNTER_MBPS, + ADF_GEN6_TL_RP_REG_OFF(reg_tl_bw_in)), + /* PCIe read bandwidth[Mbps]. */ + ADF_TL_COUNTER(BW_OUT_NAME, ADF_TL_COUNTER_MBPS, + ADF_GEN6_TL_RP_REG_OFF(reg_tl_bw_out)), + /* Message descriptor DevTLB hit rate. */ + ADF_TL_COUNTER(AT_GLOB_DTLB_HIT_NAME, ADF_TL_SIMPLE_COUNT, + ADF_GEN6_TL_RP_REG_OFF(reg_tl_at_glob_devtlb_hit)), + /* Message descriptor DevTLB miss rate. */ + ADF_TL_COUNTER(AT_GLOB_DTLB_MISS_NAME, ADF_TL_SIMPLE_COUNT, + ADF_GEN6_TL_RP_REG_OFF(reg_tl_at_glob_devtlb_miss)), + /* Payload DevTLB hit rate. */ + ADF_TL_COUNTER(AT_PAYLD_DTLB_HIT_NAME, ADF_TL_SIMPLE_COUNT, + ADF_GEN6_TL_RP_REG_OFF(reg_tl_at_payld_devtlb_hit)), + /* Payload DevTLB miss rate. */ + ADF_TL_COUNTER(AT_PAYLD_DTLB_MISS_NAME, ADF_TL_SIMPLE_COUNT, + ADF_GEN6_TL_RP_REG_OFF(reg_tl_at_payld_devtlb_miss)), +}; + +void adf_gen6_init_tl_data(struct adf_tl_hw_data *tl_data) +{ + tl_data->layout_sz = ADF_GEN6_TL_LAYOUT_SZ; + tl_data->slice_reg_sz = ADF_GEN6_TL_SLICE_REG_SZ; + tl_data->rp_reg_sz = ADF_GEN6_TL_RP_REG_SZ; + tl_data->num_hbuff = ADF_GEN6_TL_NUM_HIST_BUFFS; + tl_data->max_rp = ADF_GEN6_TL_MAX_RP_NUM; + tl_data->msg_cnt_off = ADF_GEN6_TL_MSG_CNT_OFF; + tl_data->cpp_ns_per_cycle = ADF_GEN6_CPP_NS_PER_CYCLE; + tl_data->bw_units_to_bytes = ADF_GEN6_TL_BW_HW_UNITS_TO_BYTES; + + tl_data->dev_counters = dev_counters; + tl_data->num_dev_counters = ARRAY_SIZE(dev_counters); + tl_data->sl_util_counters = sl_util_counters; + tl_data->sl_exec_counters = sl_exec_counters; + tl_data->rp_counters = rp_counters; + tl_data->num_rp_counters = ARRAY_SIZE(rp_counters); + tl_data->max_sl_cnt = ADF_GEN6_TL_MAX_SLICES_PER_TYPE; +} +EXPORT_SYMBOL_GPL(adf_gen6_init_tl_data); diff --git a/drivers/crypto/intel/qat/qat_common/adf_gen6_tl.h b/drivers/crypto/intel/qat/qat_common/adf_gen6_tl.h new file mode 100644 index 0000000000000..49db660b8eb96 --- /dev/null +++ b/drivers/crypto/intel/qat/qat_common/adf_gen6_tl.h @@ -0,0 +1,198 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright (c) 2025 Intel Corporation. */ +#ifndef ADF_GEN6_TL_H +#define ADF_GEN6_TL_H + +#include <linux/types.h> + +struct adf_tl_hw_data; + +/* Computation constants. */ +#define ADF_GEN6_CPP_NS_PER_CYCLE 2 +#define ADF_GEN6_TL_BW_HW_UNITS_TO_BYTES 64 + +/* Maximum aggregation time. Value is in milliseconds. */ +#define ADF_GEN6_TL_MAX_AGGR_TIME_MS 4000 +/* Number of buffers to store historic values. */ +#define ADF_GEN6_TL_NUM_HIST_BUFFS \ + (ADF_GEN6_TL_MAX_AGGR_TIME_MS / ADF_TL_DATA_WR_INTERVAL_MS) + +/* Max number of HW resources of one type */ +#define ADF_GEN6_TL_MAX_SLICES_PER_TYPE 32 +#define MAX_ATH_SL_COUNT 7 +#define MAX_CNV_SL_COUNT 2 +#define MAX_DCPRZ_SL_COUNT 2 +#define MAX_PKE_SL_COUNT 32 +#define MAX_UCS_SL_COUNT 4 +#define MAX_WAT_SL_COUNT 5 +#define MAX_WCP_SL_COUNT 5 + +#define MAX_ATH_CMDQ_COUNT 14 +#define MAX_CNV_CMDQ_COUNT 6 +#define MAX_DCPRZ_CMDQ_COUNT 6 +#define MAX_PKE_CMDQ_COUNT 32 +#define MAX_UCS_CMDQ_COUNT 12 +#define MAX_WAT_CMDQ_COUNT 35 +#define MAX_WCP_CMDQ_COUNT 35 + +/* Max number of simultaneously monitored ring pairs. */ +#define ADF_GEN6_TL_MAX_RP_NUM 4 + +/** + * struct adf_gen6_tl_slice_data_regs - HW slice data as populated by FW. + * @reg_tm_slice_exec_cnt: Slice execution count. + * @reg_tm_slice_util: Slice utilization. + */ +struct adf_gen6_tl_slice_data_regs { + __u32 reg_tm_slice_exec_cnt; + __u32 reg_tm_slice_util; +}; + +#define ADF_GEN6_TL_SLICE_REG_SZ sizeof(struct adf_gen6_tl_slice_data_regs) + +/** + * struct adf_gen6_tl_cmdq_data_regs - HW CMDQ data as populated by FW. + * @reg_tm_cmdq_wait_cnt: CMDQ wait count. + * @reg_tm_cmdq_exec_cnt: CMDQ execution count. + * @reg_tm_cmdq_drain_cnt: CMDQ drain count. + */ +struct adf_gen6_tl_cmdq_data_regs { + __u32 reg_tm_cmdq_wait_cnt; + __u32 reg_tm_cmdq_exec_cnt; + __u32 reg_tm_cmdq_drain_cnt; + __u32 reserved; +}; + +#define ADF_GEN6_TL_CMDQ_REG_SZ sizeof(struct adf_gen6_tl_cmdq_data_regs) + +/** + * struct adf_gen6_tl_device_data_regs - This structure stores device telemetry + * counter values as are being populated periodically by device. + * @reg_tl_rd_lat_acc: read latency accumulator + * @reg_tl_gp_lat_acc: "get to put" latency accumulator + * @reg_tl_at_page_req_lat_acc: AT/DevTLB page request latency accumulator + * @reg_tl_at_trans_lat_acc: DevTLB transaction latency accumulator + * @reg_tl_re_acc: accumulated ring empty time + * @reg_tl_prt_trans_cnt: PCIe partial transactions + * @reg_tl_rd_lat_max: maximum logged read latency + * @reg_tl_rd_cmpl_cnt: read requests completed count + * @reg_tl_gp_lat_max: maximum logged get to put latency + * @reg_tl_ae_put_cnt: Accelerator Engine put counts across all rings + * @reg_tl_bw_in: PCIe write bandwidth + * @reg_tl_bw_out: PCIe read bandwidth + * @reg_tl_at_page_req_cnt: DevTLB page requests count + * @reg_tl_at_trans_lat_cnt: DevTLB transaction latency samples count + * @reg_tl_at_max_utlb_used: maximum uTLB used + * @reg_tl_re_cnt: ring empty time samples count + * @reserved: reserved + * @ath_slices: array of Authentication slices utilization registers + * @cnv_slices: array of Compression slices utilization registers + * @dcprz_slices: array of Decompression slices utilization registers + * @pke_slices: array of PKE slices utilization registers + * @ucs_slices: array of UCS slices utilization registers + * @wat_slices: array of Wireless Authentication slices utilization registers + * @wcp_slices: array of Wireless Cipher slices utilization registers + * @ath_cmdq: array of Authentication cmdq telemetry registers + * @cnv_cmdq: array of Compression cmdq telemetry registers + * @dcprz_cmdq: array of Decomopression cmdq telemetry registers + * @pke_cmdq: array of PKE cmdq telemetry registers + * @ucs_cmdq: array of UCS cmdq telemetry registers + * @wat_cmdq: array of Wireless Authentication cmdq telemetry registers + * @wcp_cmdq: array of Wireless Cipher cmdq telemetry registers + */ +struct adf_gen6_tl_device_data_regs { + __u64 reg_tl_rd_lat_acc; + __u64 reg_tl_gp_lat_acc; + __u64 reg_tl_at_page_req_lat_acc; + __u64 reg_tl_at_trans_lat_acc; + __u64 reg_tl_re_acc; + __u32 reg_tl_prt_trans_cnt; + __u32 reg_tl_rd_lat_max; + __u32 reg_tl_rd_cmpl_cnt; + __u32 reg_tl_gp_lat_max; + __u32 reg_tl_ae_put_cnt; + __u32 reg_tl_bw_in; + __u32 reg_tl_bw_out; + __u32 reg_tl_at_page_req_cnt; + __u32 reg_tl_at_trans_lat_cnt; + __u32 reg_tl_at_max_utlb_used; + __u32 reg_tl_re_cnt; + __u32 reserved; + struct adf_gen6_tl_slice_data_regs ath_slices[MAX_ATH_SL_COUNT]; + struct adf_gen6_tl_slice_data_regs cnv_slices[MAX_CNV_SL_COUNT]; + struct adf_gen6_tl_slice_data_regs dcprz_slices[MAX_DCPRZ_SL_COUNT]; + struct adf_gen6_tl_slice_data_regs pke_slices[MAX_PKE_SL_COUNT]; + struct adf_gen6_tl_slice_data_regs ucs_slices[MAX_UCS_SL_COUNT]; + struct adf_gen6_tl_slice_data_regs wat_slices[MAX_WAT_SL_COUNT]; + struct adf_gen6_tl_slice_data_regs wcp_slices[MAX_WCP_SL_COUNT]; + struct adf_gen6_tl_cmdq_data_regs ath_cmdq[MAX_ATH_CMDQ_COUNT]; + struct adf_gen6_tl_cmdq_data_regs cnv_cmdq[MAX_CNV_CMDQ_COUNT]; + struct adf_gen6_tl_cmdq_data_regs dcprz_cmdq[MAX_DCPRZ_CMDQ_COUNT]; + struct adf_gen6_tl_cmdq_data_regs pke_cmdq[MAX_PKE_CMDQ_COUNT]; + struct adf_gen6_tl_cmdq_data_regs ucs_cmdq[MAX_UCS_CMDQ_COUNT]; + struct adf_gen6_tl_cmdq_data_regs wat_cmdq[MAX_WAT_CMDQ_COUNT]; + struct adf_gen6_tl_cmdq_data_regs wcp_cmdq[MAX_WCP_CMDQ_COUNT]; +}; + +/** + * struct adf_gen6_tl_ring_pair_data_regs - This structure stores ring pair + * telemetry counter values as they are being populated periodically by device. + * @reg_tl_gp_lat_acc: get-put latency accumulator + * @reg_tl_re_acc: accumulated ring empty time + * @reg_tl_pci_trans_cnt: PCIe partial transactions + * @reg_tl_ae_put_cnt: Accelerator Engine put counts across all rings + * @reg_tl_bw_in: PCIe write bandwidth + * @reg_tl_bw_out: PCIe read bandwidth + * @reg_tl_at_glob_devtlb_hit: Message descriptor DevTLB hit rate + * @reg_tl_at_glob_devtlb_miss: Message descriptor DevTLB miss rate + * @reg_tl_at_payld_devtlb_hit: Payload DevTLB hit rate + * @reg_tl_at_payld_devtlb_miss: Payload DevTLB miss rate + * @reg_tl_re_cnt: ring empty time samples count + * @reserved1: reserved + */ +struct adf_gen6_tl_ring_pair_data_regs { + __u64 reg_tl_gp_lat_acc; + __u64 reg_tl_re_acc; + __u32 reg_tl_prt_trans_cnt; + __u32 reg_tl_ae_put_cnt; + __u32 reg_tl_bw_in; + __u32 reg_tl_bw_out; + __u32 reg_tl_at_glob_devtlb_hit; + __u32 reg_tl_at_glob_devtlb_miss; + __u32 reg_tl_at_payld_devtlb_hit; + __u32 reg_tl_at_payld_devtlb_miss; + __u32 reg_tl_re_cnt; + __u32 reserved1; +}; + +#define ADF_GEN6_TL_RP_REG_SZ sizeof(struct adf_gen6_tl_ring_pair_data_regs) + +/** + * struct adf_gen6_tl_layout - This structure represents the entire telemetry + * counters data: Device + 4 Ring Pairs as they are being populated periodically + * by device. + * @tl_device_data_regs: structure of device telemetry registers + * @tl_ring_pairs_data_regs: array of ring pairs telemetry registers + * @reg_tl_msg_cnt: telemetry message counter + * @reserved: reserved + */ +struct adf_gen6_tl_layout { + struct adf_gen6_tl_device_data_regs tl_device_data_regs; + struct adf_gen6_tl_ring_pair_data_regs + tl_ring_pairs_data_regs[ADF_GEN6_TL_MAX_RP_NUM]; + __u32 reg_tl_msg_cnt; + __u32 reserved; +}; + +#define ADF_GEN6_TL_LAYOUT_SZ sizeof(struct adf_gen6_tl_layout) +#define ADF_GEN6_TL_MSG_CNT_OFF \ + offsetof(struct adf_gen6_tl_layout, reg_tl_msg_cnt) + +#ifdef CONFIG_DEBUG_FS +void adf_gen6_init_tl_data(struct adf_tl_hw_data *tl_data); +#else +static inline void adf_gen6_init_tl_data(struct adf_tl_hw_data *tl_data) +{ +} +#endif /* CONFIG_DEBUG_FS */ +#endif /* ADF_GEN6_TL_H */ diff --git a/drivers/crypto/intel/qat/qat_common/adf_init.c b/drivers/crypto/intel/qat/qat_common/adf_init.c index f189cce7d1535..46491048e0bb4 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_init.c +++ b/drivers/crypto/intel/qat/qat_common/adf_init.c @@ -404,6 +404,7 @@ static void adf_dev_shutdown(struct adf_accel_dev *accel_dev) hw_data->exit_admin_comms(accel_dev); adf_cleanup_etr_data(accel_dev); + adf_misc_wq_flush(); adf_dev_restore(accel_dev); } diff --git a/drivers/crypto/intel/qat/qat_common/adf_isr.c b/drivers/crypto/intel/qat/qat_common/adf_isr.c index cae1aee5479af..12e5656136610 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_isr.c +++ b/drivers/crypto/intel/qat/qat_common/adf_isr.c @@ -407,3 +407,8 @@ bool adf_misc_wq_queue_delayed_work(struct delayed_work *work, { return queue_delayed_work(adf_misc_wq, work, delay); } + +void adf_misc_wq_flush(void) +{ + flush_workqueue(adf_misc_wq); +} diff --git a/drivers/crypto/intel/qat/qat_common/adf_pm_dbgfs_utils.c b/drivers/crypto/intel/qat/qat_common/adf_pm_dbgfs_utils.c new file mode 100644 index 0000000000000..69295a9ddf0ac --- /dev/null +++ b/drivers/crypto/intel/qat/qat_common/adf_pm_dbgfs_utils.c @@ -0,0 +1,52 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* Copyright(c) 2025 Intel Corporation */ +#include <linux/bitops.h> +#include <linux/sprintf.h> +#include <linux/string_helpers.h> + +#include "adf_pm_dbgfs_utils.h" + +/* + * This is needed because a variable is used to index the mask at + * pm_scnprint_table(), making it not compile time constant, so the compile + * asserts from FIELD_GET() or u32_get_bits() won't be fulfilled. + */ +#define field_get(_mask, _reg) (((_reg) & (_mask)) >> (ffs(_mask) - 1)) + +#define PM_INFO_MAX_KEY_LEN 21 + +static int pm_scnprint_table(char *buff, const struct pm_status_row *table, + u32 *pm_info_regs, size_t buff_size, int table_len, + bool lowercase) +{ + char key[PM_INFO_MAX_KEY_LEN]; + int wr = 0; + int i; + + for (i = 0; i < table_len; i++) { + if (lowercase) + string_lower(key, table[i].key); + else + string_upper(key, table[i].key); + + wr += scnprintf(&buff[wr], buff_size - wr, "%s: %#x\n", key, + field_get(table[i].field_mask, + pm_info_regs[table[i].reg_offset])); + } + + return wr; +} + +int adf_pm_scnprint_table_upper_keys(char *buff, const struct pm_status_row *table, + u32 *pm_info_regs, size_t buff_size, int table_len) +{ + return pm_scnprint_table(buff, table, pm_info_regs, buff_size, + table_len, false); +} + +int adf_pm_scnprint_table_lower_keys(char *buff, const struct pm_status_row *table, + u32 *pm_info_regs, size_t buff_size, int table_len) +{ + return pm_scnprint_table(buff, table, pm_info_regs, buff_size, + table_len, true); +} diff --git a/drivers/crypto/intel/qat/qat_common/adf_pm_dbgfs_utils.h b/drivers/crypto/intel/qat/qat_common/adf_pm_dbgfs_utils.h new file mode 100644 index 0000000000000..854f058b35edf --- /dev/null +++ b/drivers/crypto/intel/qat/qat_common/adf_pm_dbgfs_utils.h @@ -0,0 +1,36 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* Copyright(c) 2025 Intel Corporation */ +#ifndef ADF_PM_DBGFS_UTILS_H_ +#define ADF_PM_DBGFS_UTILS_H_ + +#include <linux/stddef.h> +#include <linux/stringify.h> +#include <linux/types.h> +#include "icp_qat_fw_init_admin.h" + +#define PM_INFO_MEMBER_OFF(member) \ + (offsetof(struct icp_qat_fw_init_admin_pm_info, member) / sizeof(u32)) + +#define PM_INFO_REGSET_ENTRY_MASK(_reg_, _field_, _mask_) \ +{ \ + .reg_offset = PM_INFO_MEMBER_OFF(_reg_), \ + .key = __stringify(_field_), \ + .field_mask = _mask_, \ +} + +#define PM_INFO_REGSET_ENTRY32(_reg_, _field_) \ + PM_INFO_REGSET_ENTRY_MASK(_reg_, _field_, GENMASK(31, 0)) + +struct pm_status_row { + int reg_offset; + u32 field_mask; + const char *key; +}; + +int adf_pm_scnprint_table_upper_keys(char *buff, const struct pm_status_row *table, + u32 *pm_info_regs, size_t buff_size, int table_len); + +int adf_pm_scnprint_table_lower_keys(char *buff, const struct pm_status_row *table, + u32 *pm_info_regs, size_t buff_size, int table_len); + +#endif /* ADF_PM_DBGFS_UTILS_H_ */ diff --git a/drivers/crypto/intel/qat/qat_common/adf_rl.c b/drivers/crypto/intel/qat/qat_common/adf_rl.c index e782c23fc1bfc..c6a54e4659316 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_rl.c +++ b/drivers/crypto/intel/qat/qat_common/adf_rl.c @@ -13,6 +13,7 @@ #include <linux/units.h> #include "adf_accel_devices.h" +#include "adf_cfg_services.h" #include "adf_common_drv.h" #include "adf_rl_admin.h" #include "adf_rl.h" @@ -55,7 +56,7 @@ static int validate_user_input(struct adf_accel_dev *accel_dev, } } - if (sla_in->srv >= ADF_SVC_NONE) { + if (sla_in->srv >= SVC_BASE_COUNT) { dev_notice(&GET_DEV(accel_dev), "Wrong service type\n"); return -EINVAL; @@ -168,20 +169,6 @@ static struct rl_sla *find_parent(struct adf_rl *rl_data, return NULL; } -static enum adf_cfg_service_type srv_to_cfg_svc_type(enum adf_base_services rl_srv) -{ - switch (rl_srv) { - case ADF_SVC_ASYM: - return ASYM; - case ADF_SVC_SYM: - return SYM; - case ADF_SVC_DC: - return COMP; - default: - return UNUSED; - } -} - /** * adf_rl_get_sla_arr_of_type() - Returns a pointer to SLA type specific array * @rl_data: pointer to ratelimiting data @@ -209,22 +196,6 @@ u32 adf_rl_get_sla_arr_of_type(struct adf_rl *rl_data, enum rl_node_type type, } } -static bool is_service_enabled(struct adf_accel_dev *accel_dev, - enum adf_base_services rl_srv) -{ - enum adf_cfg_service_type arb_srv = srv_to_cfg_svc_type(rl_srv); - struct adf_hw_device_data *hw_data = GET_HW_DATA(accel_dev); - u8 rps_per_bundle = hw_data->num_banks_per_vf; - int i; - - for (i = 0; i < rps_per_bundle; i++) { - if (GET_SRV_TYPE(accel_dev, i) == arb_srv) - return true; - } - - return false; -} - /** * prepare_rp_ids() - Creates an array of ring pair IDs from bitmask * @accel_dev: pointer to acceleration device structure @@ -243,7 +214,7 @@ static bool is_service_enabled(struct adf_accel_dev *accel_dev, static int prepare_rp_ids(struct adf_accel_dev *accel_dev, struct rl_sla *sla, const unsigned long rp_mask) { - enum adf_cfg_service_type arb_srv = srv_to_cfg_svc_type(sla->srv); + enum adf_cfg_service_type arb_srv = adf_srv_to_cfg_svc_type(sla->srv); u16 rps_per_bundle = GET_HW_DATA(accel_dev)->num_banks_per_vf; bool *rp_in_use = accel_dev->rate_limiting->rp_in_use; size_t rp_cnt_max = ARRAY_SIZE(sla->ring_pairs_ids); @@ -558,21 +529,9 @@ u32 adf_rl_calculate_slice_tokens(struct adf_accel_dev *accel_dev, u32 sla_val, if (!sla_val) return 0; + /* Handle generation specific slice count adjustment */ avail_slice_cycles = hw_data->clock_frequency; - - switch (svc_type) { - case ADF_SVC_ASYM: - avail_slice_cycles *= device_data->slices.pke_cnt; - break; - case ADF_SVC_SYM: - avail_slice_cycles *= device_data->slices.cph_cnt; - break; - case ADF_SVC_DC: - avail_slice_cycles *= device_data->slices.dcpr_cnt; - break; - default: - break; - } + avail_slice_cycles *= hw_data->get_svc_slice_cnt(accel_dev, svc_type); do_div(avail_slice_cycles, device_data->scan_interval); allocated_tokens = avail_slice_cycles * sla_val; @@ -581,6 +540,17 @@ u32 adf_rl_calculate_slice_tokens(struct adf_accel_dev *accel_dev, u32 sla_val, return allocated_tokens; } +static u32 adf_rl_get_num_svc_aes(struct adf_accel_dev *accel_dev, + enum adf_base_services svc) +{ + struct adf_rl_hw_data *device_data = &accel_dev->hw_device->rl_data; + + if (svc >= SVC_BASE_COUNT) + return 0; + + return device_data->svc_ae_mask[svc]; +} + u32 adf_rl_calculate_ae_cycles(struct adf_accel_dev *accel_dev, u32 sla_val, enum adf_base_services svc_type) { @@ -592,7 +562,7 @@ u32 adf_rl_calculate_ae_cycles(struct adf_accel_dev *accel_dev, u32 sla_val, return 0; avail_ae_cycles = hw_data->clock_frequency; - avail_ae_cycles *= hw_data->get_num_aes(hw_data) - 1; + avail_ae_cycles *= adf_rl_get_num_svc_aes(accel_dev, svc_type); do_div(avail_ae_cycles, device_data->scan_interval); sla_val *= device_data->max_tp[svc_type]; @@ -617,9 +587,8 @@ u32 adf_rl_calculate_pci_bw(struct adf_accel_dev *accel_dev, u32 sla_val, sla_to_bytes *= device_data->max_tp[svc_type]; do_div(sla_to_bytes, device_data->scale_ref); - sla_to_bytes *= (svc_type == ADF_SVC_ASYM) ? RL_TOKEN_ASYM_SIZE : - BYTES_PER_MBIT; - if (svc_type == ADF_SVC_DC && is_bw_out) + sla_to_bytes *= (svc_type == SVC_ASYM) ? RL_TOKEN_ASYM_SIZE : BYTES_PER_MBIT; + if (svc_type == SVC_DC && is_bw_out) sla_to_bytes *= device_data->slices.dcpr_cnt - device_data->dcpr_correction; @@ -660,7 +629,7 @@ static int add_new_sla_entry(struct adf_accel_dev *accel_dev, } *sla_out = sla; - if (!is_service_enabled(accel_dev, sla_in->srv)) { + if (!adf_is_service_enabled(accel_dev, sla_in->srv)) { dev_notice(&GET_DEV(accel_dev), "Provided service is not enabled\n"); ret = -EINVAL; @@ -730,8 +699,8 @@ static int initialize_default_nodes(struct adf_accel_dev *accel_dev) sla_in.type = RL_ROOT; sla_in.parent_id = RL_PARENT_DEFAULT_ID; - for (i = 0; i < ADF_SVC_NONE; i++) { - if (!is_service_enabled(accel_dev, i)) + for (i = 0; i < SVC_BASE_COUNT; i++) { + if (!adf_is_service_enabled(accel_dev, i)) continue; sla_in.cir = device_data->scale_ref; @@ -745,10 +714,9 @@ static int initialize_default_nodes(struct adf_accel_dev *accel_dev) /* Init default cluster for each root */ sla_in.type = RL_CLUSTER; - for (i = 0; i < ADF_SVC_NONE; i++) { + for (i = 0; i < SVC_BASE_COUNT; i++) { if (!rl_data->root[i]) continue; - sla_in.cir = rl_data->root[i]->cir; sla_in.pir = sla_in.cir; sla_in.srv = rl_data->root[i]->srv; @@ -987,7 +955,7 @@ int adf_rl_get_capability_remaining(struct adf_accel_dev *accel_dev, struct rl_sla *sla = NULL; int i; - if (srv >= ADF_SVC_NONE) + if (srv >= SVC_BASE_COUNT) return -EINVAL; if (sla_id > RL_SLA_EMPTY_ID && !validate_sla_id(accel_dev, sla_id)) { @@ -1086,9 +1054,9 @@ int adf_rl_init(struct adf_accel_dev *accel_dev) int ret = 0; /* Validate device parameters */ - if (RL_VALIDATE_NON_ZERO(rl_hw_data->max_tp[ADF_SVC_ASYM]) || - RL_VALIDATE_NON_ZERO(rl_hw_data->max_tp[ADF_SVC_SYM]) || - RL_VALIDATE_NON_ZERO(rl_hw_data->max_tp[ADF_SVC_DC]) || + if (RL_VALIDATE_NON_ZERO(rl_hw_data->max_tp[SVC_ASYM]) || + RL_VALIDATE_NON_ZERO(rl_hw_data->max_tp[SVC_SYM]) || + RL_VALIDATE_NON_ZERO(rl_hw_data->max_tp[SVC_DC]) || RL_VALIDATE_NON_ZERO(rl_hw_data->scan_interval) || RL_VALIDATE_NON_ZERO(rl_hw_data->pcie_scale_div) || RL_VALIDATE_NON_ZERO(rl_hw_data->pcie_scale_mul) || diff --git a/drivers/crypto/intel/qat/qat_common/adf_rl.h b/drivers/crypto/intel/qat/qat_common/adf_rl.h index bfe750ea0e839..c1f3f9a511958 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_rl.h +++ b/drivers/crypto/intel/qat/qat_common/adf_rl.h @@ -7,6 +7,8 @@ #include <linux/mutex.h> #include <linux/types.h> +#include "adf_cfg_services.h" + struct adf_accel_dev; #define RL_ROOT_MAX 4 @@ -24,13 +26,6 @@ enum rl_node_type { RL_LEAF, }; -enum adf_base_services { - ADF_SVC_ASYM = 0, - ADF_SVC_SYM, - ADF_SVC_DC, - ADF_SVC_NONE, -}; - /** * struct adf_rl_sla_input_data - ratelimiting user input data structure * @rp_mask: 64 bit bitmask of ring pair IDs which will be assigned to SLA. @@ -73,6 +68,7 @@ struct rl_slice_cnt { u8 dcpr_cnt; u8 pke_cnt; u8 cph_cnt; + u8 cpr_cnt; }; struct adf_rl_interface_data { @@ -94,6 +90,7 @@ struct adf_rl_hw_data { u32 pcie_scale_div; u32 dcpr_correction; u32 max_tp[RL_ROOT_MAX]; + u32 svc_ae_mask[SVC_BASE_COUNT]; struct rl_slice_cnt slices; }; diff --git a/drivers/crypto/intel/qat/qat_common/adf_rl_admin.c b/drivers/crypto/intel/qat/qat_common/adf_rl_admin.c index 698a14f4ce66a..4a3e0591fdba9 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_rl_admin.c +++ b/drivers/crypto/intel/qat/qat_common/adf_rl_admin.c @@ -63,6 +63,7 @@ int adf_rl_send_admin_init_msg(struct adf_accel_dev *accel_dev, slices_int->pke_cnt = slices_resp.pke_cnt; /* For symmetric crypto, slice tokens are relative to the UCS slice */ slices_int->cph_cnt = slices_resp.ucs_cnt; + slices_int->cpr_cnt = slices_resp.cpr_cnt; return 0; } diff --git a/drivers/crypto/intel/qat/qat_common/adf_sriov.c b/drivers/crypto/intel/qat/qat_common/adf_sriov.c index c75d0b6cb0ada..31d1ef0cb1f52 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_sriov.c +++ b/drivers/crypto/intel/qat/qat_common/adf_sriov.c @@ -155,7 +155,6 @@ static int adf_do_enable_sriov(struct adf_accel_dev *accel_dev) if (!device_iommu_mapped(&GET_DEV(accel_dev))) { dev_warn(&GET_DEV(accel_dev), "IOMMU should be enabled for SR-IOV to work correctly\n"); - return -EINVAL; } if (adf_dev_started(accel_dev)) { diff --git a/drivers/crypto/intel/qat/qat_common/adf_sysfs.c b/drivers/crypto/intel/qat/qat_common/adf_sysfs.c index 6c39194647f03..79c63dfa8ff30 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_sysfs.c +++ b/drivers/crypto/intel/qat/qat_common/adf_sysfs.c @@ -269,6 +269,8 @@ static ssize_t rp2srv_show(struct device *dev, struct device_attribute *attr, return sysfs_emit(buf, "%s\n", ADF_CFG_SYM); case ASYM: return sysfs_emit(buf, "%s\n", ADF_CFG_ASYM); + case DECOMP: + return sysfs_emit(buf, "%s\n", ADF_CFG_DECOMP); default: break; } diff --git a/drivers/crypto/intel/qat/qat_common/adf_sysfs_rl.c b/drivers/crypto/intel/qat/qat_common/adf_sysfs_rl.c index bedb514d4e304..f31556beed8b6 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_sysfs_rl.c +++ b/drivers/crypto/intel/qat/qat_common/adf_sysfs_rl.c @@ -32,9 +32,10 @@ enum rl_params { }; static const char *const rl_services[] = { - [ADF_SVC_ASYM] = "asym", - [ADF_SVC_SYM] = "sym", - [ADF_SVC_DC] = "dc", + [SVC_ASYM] = "asym", + [SVC_SYM] = "sym", + [SVC_DC] = "dc", + [SVC_DECOMP] = "decomp", }; static const char *const rl_operations[] = { @@ -282,7 +283,7 @@ static ssize_t srv_show(struct device *dev, struct device_attribute *attr, if (ret) return ret; - if (get == ADF_SVC_NONE) + if (get == SVC_BASE_COUNT) return -EINVAL; return sysfs_emit(buf, "%s\n", rl_services[get]); @@ -291,14 +292,22 @@ static ssize_t srv_show(struct device *dev, struct device_attribute *attr, static ssize_t srv_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) { + struct adf_accel_dev *accel_dev; unsigned int val; int ret; + accel_dev = adf_devmgr_pci_to_accel_dev(to_pci_dev(dev)); + if (!accel_dev) + return -EINVAL; + ret = sysfs_match_string(rl_services, buf); if (ret < 0) return ret; val = ret; + if (!adf_is_service_enabled(accel_dev, val)) + return -EINVAL; + ret = set_param_u(dev, SRV, val); if (ret) return ret; @@ -439,8 +448,8 @@ int adf_sysfs_rl_add(struct adf_accel_dev *accel_dev) dev_err(&GET_DEV(accel_dev), "Failed to create qat_rl attribute group\n"); - data->cap_rem_srv = ADF_SVC_NONE; - data->input.srv = ADF_SVC_NONE; + data->cap_rem_srv = SVC_BASE_COUNT; + data->input.srv = SVC_BASE_COUNT; data->sysfs_added = true; return ret; diff --git a/drivers/crypto/intel/qat/qat_common/adf_tl_debugfs.c b/drivers/crypto/intel/qat/qat_common/adf_tl_debugfs.c index f20ae7e35a0dc..a32db273842a5 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_tl_debugfs.c +++ b/drivers/crypto/intel/qat/qat_common/adf_tl_debugfs.c @@ -538,6 +538,9 @@ static void tl_print_rp_srv(struct adf_accel_dev *accel_dev, struct seq_file *s, case ASYM: seq_printf(s, "%*s\n", TL_VALUE_MIN_PADDING, ADF_CFG_ASYM); break; + case DECOMP: + seq_printf(s, "%*s\n", TL_VALUE_MIN_PADDING, ADF_CFG_DECOMP); + break; default: seq_printf(s, "%*s\n", TL_VALUE_MIN_PADDING, TL_RP_SRV_UNKNOWN); break; diff --git a/drivers/crypto/intel/qat/qat_common/adf_transport_debug.c b/drivers/crypto/intel/qat/qat_common/adf_transport_debug.c index e2dd568b87b51..6c22bc9b28e47 100644 --- a/drivers/crypto/intel/qat/qat_common/adf_transport_debug.c +++ b/drivers/crypto/intel/qat/qat_common/adf_transport_debug.c @@ -10,16 +10,21 @@ static DEFINE_MUTEX(ring_read_lock); static DEFINE_MUTEX(bank_read_lock); +#define ADF_RING_NUM_MSGS(ring) \ + (ADF_SIZE_TO_RING_SIZE_IN_BYTES(ring->ring_size) / \ + ADF_MSG_SIZE_TO_BYTES(ring->msg_size)) + static void *adf_ring_start(struct seq_file *sfile, loff_t *pos) { struct adf_etr_ring_data *ring = sfile->private; + unsigned int num_msg = ADF_RING_NUM_MSGS(ring); + loff_t val = *pos; mutex_lock(&ring_read_lock); - if (*pos == 0) + if (val == 0) return SEQ_START_TOKEN; - if (*pos >= (ADF_SIZE_TO_RING_SIZE_IN_BYTES(ring->ring_size) / - ADF_MSG_SIZE_TO_BYTES(ring->msg_size))) + if (val >= num_msg) return NULL; return ring->base_addr + @@ -29,13 +34,15 @@ static void *adf_ring_start(struct seq_file *sfile, loff_t *pos) static void *adf_ring_next(struct seq_file *sfile, void *v, loff_t *pos) { struct adf_etr_ring_data *ring = sfile->private; + unsigned int num_msg = ADF_RING_NUM_MSGS(ring); + loff_t val = *pos; + + (*pos)++; - if (*pos >= (ADF_SIZE_TO_RING_SIZE_IN_BYTES(ring->ring_size) / - ADF_MSG_SIZE_TO_BYTES(ring->msg_size))) + if (val >= num_msg) return NULL; - return ring->base_addr + - (ADF_MSG_SIZE_TO_BYTES(ring->msg_size) * (*pos)++); + return ring->base_addr + (ADF_MSG_SIZE_TO_BYTES(ring->msg_size) * val); } static int adf_ring_show(struct seq_file *sfile, void *v) diff --git a/drivers/crypto/intel/qat/qat_common/qat_algs.c b/drivers/crypto/intel/qat/qat_common/qat_algs.c index c03a698511142..43e6dd9b77b7d 100644 --- a/drivers/crypto/intel/qat/qat_common/qat_algs.c +++ b/drivers/crypto/intel/qat/qat_common/qat_algs.c @@ -1277,7 +1277,7 @@ static struct aead_alg qat_aeads[] = { { .base = { .cra_name = "authenc(hmac(sha1),cbc(aes))", .cra_driver_name = "qat_aes_cbc_hmac_sha1", - .cra_priority = 4001, + .cra_priority = 100, .cra_flags = CRYPTO_ALG_ASYNC | CRYPTO_ALG_ALLOCATES_MEMORY, .cra_blocksize = AES_BLOCK_SIZE, .cra_ctxsize = sizeof(struct qat_alg_aead_ctx), @@ -1294,7 +1294,7 @@ static struct aead_alg qat_aeads[] = { { .base = { .cra_name = "authenc(hmac(sha256),cbc(aes))", .cra_driver_name = "qat_aes_cbc_hmac_sha256", - .cra_priority = 4001, + .cra_priority = 100, .cra_flags = CRYPTO_ALG_ASYNC | CRYPTO_ALG_ALLOCATES_MEMORY, .cra_blocksize = AES_BLOCK_SIZE, .cra_ctxsize = sizeof(struct qat_alg_aead_ctx), @@ -1311,7 +1311,7 @@ static struct aead_alg qat_aeads[] = { { .base = { .cra_name = "authenc(hmac(sha512),cbc(aes))", .cra_driver_name = "qat_aes_cbc_hmac_sha512", - .cra_priority = 4001, + .cra_priority = 100, .cra_flags = CRYPTO_ALG_ASYNC | CRYPTO_ALG_ALLOCATES_MEMORY, .cra_blocksize = AES_BLOCK_SIZE, .cra_ctxsize = sizeof(struct qat_alg_aead_ctx), @@ -1329,7 +1329,7 @@ static struct aead_alg qat_aeads[] = { { static struct skcipher_alg qat_skciphers[] = { { .base.cra_name = "cbc(aes)", .base.cra_driver_name = "qat_aes_cbc", - .base.cra_priority = 4001, + .base.cra_priority = 100, .base.cra_flags = CRYPTO_ALG_ASYNC | CRYPTO_ALG_ALLOCATES_MEMORY, .base.cra_blocksize = AES_BLOCK_SIZE, .base.cra_ctxsize = sizeof(struct qat_alg_skcipher_ctx), @@ -1347,7 +1347,7 @@ static struct skcipher_alg qat_skciphers[] = { { }, { .base.cra_name = "ctr(aes)", .base.cra_driver_name = "qat_aes_ctr", - .base.cra_priority = 4001, + .base.cra_priority = 100, .base.cra_flags = CRYPTO_ALG_ASYNC | CRYPTO_ALG_ALLOCATES_MEMORY, .base.cra_blocksize = 1, .base.cra_ctxsize = sizeof(struct qat_alg_skcipher_ctx), @@ -1365,7 +1365,7 @@ static struct skcipher_alg qat_skciphers[] = { { }, { .base.cra_name = "xts(aes)", .base.cra_driver_name = "qat_aes_xts", - .base.cra_priority = 4001, + .base.cra_priority = 100, .base.cra_flags = CRYPTO_ALG_ASYNC | CRYPTO_ALG_NEED_FALLBACK | CRYPTO_ALG_ALLOCATES_MEMORY, .base.cra_blocksize = AES_BLOCK_SIZE, diff --git a/drivers/crypto/intel/qat/qat_common/qat_bl.c b/drivers/crypto/intel/qat/qat_common/qat_bl.c index 5e4dad4693caa..9b2338f58d97c 100644 --- a/drivers/crypto/intel/qat/qat_common/qat_bl.c +++ b/drivers/crypto/intel/qat/qat_common/qat_bl.c @@ -38,7 +38,7 @@ void qat_bl_free_bufl(struct adf_accel_dev *accel_dev, for (i = 0; i < blout->num_mapped_bufs; i++) { dma_unmap_single(dev, blout->buffers[i].addr, blout->buffers[i].len, - DMA_FROM_DEVICE); + DMA_BIDIRECTIONAL); } dma_unmap_single(dev, blpout, sz_out, DMA_TO_DEVICE); @@ -162,7 +162,7 @@ static int __qat_bl_sgl_to_bufl(struct adf_accel_dev *accel_dev, } buffers[y].addr = dma_map_single(dev, sg_virt(sg) + left, sg->length - left, - DMA_FROM_DEVICE); + DMA_BIDIRECTIONAL); if (unlikely(dma_mapping_error(dev, buffers[y].addr))) goto err_out; buffers[y].len = sg->length; @@ -204,7 +204,7 @@ err_out: if (!dma_mapping_error(dev, buflout->buffers[i].addr)) dma_unmap_single(dev, buflout->buffers[i].addr, buflout->buffers[i].len, - DMA_FROM_DEVICE); + DMA_BIDIRECTIONAL); } if (!buf->sgl_dst_valid) diff --git a/drivers/crypto/intel/qat/qat_common/qat_compression.c b/drivers/crypto/intel/qat/qat_common/qat_compression.c index c285b45b8679d..53a4db5507ec2 100644 --- a/drivers/crypto/intel/qat/qat_common/qat_compression.c +++ b/drivers/crypto/intel/qat/qat_common/qat_compression.c @@ -196,7 +196,7 @@ static int qat_compression_alloc_dc_data(struct adf_accel_dev *accel_dev) struct adf_dc_data *dc_data = NULL; u8 *obuff = NULL; - dc_data = devm_kzalloc(dev, sizeof(*dc_data), GFP_KERNEL); + dc_data = kzalloc_node(sizeof(*dc_data), GFP_KERNEL, dev_to_node(dev)); if (!dc_data) goto err; @@ -204,7 +204,7 @@ static int qat_compression_alloc_dc_data(struct adf_accel_dev *accel_dev) if (!obuff) goto err; - obuff_p = dma_map_single(dev, obuff, ovf_buff_sz, DMA_FROM_DEVICE); + obuff_p = dma_map_single(dev, obuff, ovf_buff_sz, DMA_BIDIRECTIONAL); if (unlikely(dma_mapping_error(dev, obuff_p))) goto err; @@ -232,9 +232,9 @@ static void qat_free_dc_data(struct adf_accel_dev *accel_dev) return; dma_unmap_single(dev, dc_data->ovf_buff_p, dc_data->ovf_buff_sz, - DMA_FROM_DEVICE); + DMA_BIDIRECTIONAL); kfree_sensitive(dc_data->ovf_buff); - devm_kfree(dev, dc_data); + kfree(dc_data); accel_dev->dc_data = NULL; } diff --git a/drivers/crypto/marvell/cesa/cipher.c b/drivers/crypto/marvell/cesa/cipher.c index 48c5c8ea8c43e..3fe0fd9226cf7 100644 --- a/drivers/crypto/marvell/cesa/cipher.c +++ b/drivers/crypto/marvell/cesa/cipher.c @@ -75,9 +75,12 @@ mv_cesa_skcipher_dma_cleanup(struct skcipher_request *req) static inline void mv_cesa_skcipher_cleanup(struct skcipher_request *req) { struct mv_cesa_skcipher_req *creq = skcipher_request_ctx(req); + struct mv_cesa_engine *engine = creq->base.engine; if (mv_cesa_req_get_type(&creq->base) == CESA_DMA_REQ) mv_cesa_skcipher_dma_cleanup(req); + + atomic_sub(req->cryptlen, &engine->load); } static void mv_cesa_skcipher_std_step(struct skcipher_request *req) @@ -212,7 +215,6 @@ mv_cesa_skcipher_complete(struct crypto_async_request *req) struct mv_cesa_engine *engine = creq->base.engine; unsigned int ivsize; - atomic_sub(skreq->cryptlen, &engine->load); ivsize = crypto_skcipher_ivsize(crypto_skcipher_reqtfm(skreq)); if (mv_cesa_req_get_type(&creq->base) == CESA_DMA_REQ) { diff --git a/drivers/crypto/marvell/cesa/hash.c b/drivers/crypto/marvell/cesa/hash.c index 6815eddc90681..5103d36cdfdb1 100644 --- a/drivers/crypto/marvell/cesa/hash.c +++ b/drivers/crypto/marvell/cesa/hash.c @@ -110,9 +110,12 @@ static inline void mv_cesa_ahash_dma_cleanup(struct ahash_request *req) static inline void mv_cesa_ahash_cleanup(struct ahash_request *req) { struct mv_cesa_ahash_req *creq = ahash_request_ctx(req); + struct mv_cesa_engine *engine = creq->base.engine; if (mv_cesa_req_get_type(&creq->base) == CESA_DMA_REQ) mv_cesa_ahash_dma_cleanup(req); + + atomic_sub(req->nbytes, &engine->load); } static void mv_cesa_ahash_last_cleanup(struct ahash_request *req) @@ -362,16 +365,13 @@ static void mv_cesa_ahash_complete(struct crypto_async_request *req) if (mv_cesa_req_get_type(&creq->base) == CESA_DMA_REQ && (creq->base.chain.last->flags & CESA_TDMA_TYPE_MSK) == CESA_TDMA_RESULT) { - __le32 *data = NULL; + const void *data; /* * Result is already in the correct endianness when the SA is * used */ data = creq->base.chain.last->op->ctx.hash.hash; - for (i = 0; i < digsize / 4; i++) - creq->state[i] = le32_to_cpu(data[i]); - memcpy(ahashreq->result, data, digsize); } else { for (i = 0; i < digsize / 4; i++) @@ -395,8 +395,6 @@ static void mv_cesa_ahash_complete(struct crypto_async_request *req) } } } - - atomic_sub(ahashreq->nbytes, &engine->load); } static void mv_cesa_ahash_prepare(struct crypto_async_request *req, diff --git a/drivers/crypto/marvell/octeontx2/otx2_cpt_reqmgr.h b/drivers/crypto/marvell/octeontx2/otx2_cpt_reqmgr.h index e27e849b01dfc..e64ca30335dea 100644 --- a/drivers/crypto/marvell/octeontx2/otx2_cpt_reqmgr.h +++ b/drivers/crypto/marvell/octeontx2/otx2_cpt_reqmgr.h @@ -34,6 +34,9 @@ #define SG_COMP_2 2 #define SG_COMP_1 1 +#define OTX2_CPT_DPTR_RPTR_ALIGN 8 +#define OTX2_CPT_RES_ADDR_ALIGN 32 + union otx2_cpt_opcode { u16 flags; struct { @@ -347,22 +350,48 @@ static inline struct otx2_cpt_inst_info * cn10k_sgv2_info_create(struct pci_dev *pdev, struct otx2_cpt_req_info *req, gfp_t gfp) { - u32 dlen = 0, g_len, sg_len, info_len; - int align = OTX2_CPT_DMA_MINALIGN; + u32 dlen = 0, g_len, s_len, sg_len, info_len; struct otx2_cpt_inst_info *info; - u16 g_sz_bytes, s_sz_bytes; u32 total_mem_len; int i; - g_sz_bytes = ((req->in_cnt + 2) / 3) * - sizeof(struct cn10kb_cpt_sglist_component); - s_sz_bytes = ((req->out_cnt + 2) / 3) * - sizeof(struct cn10kb_cpt_sglist_component); + /* Allocate memory to meet below alignment requirement: + * ------------------------------------ + * | struct otx2_cpt_inst_info | + * | (No alignment required) | + * | --------------------------------| + * | | padding for ARCH_DMA_MINALIGN | + * | | alignment | + * |------------------------------------| + * | SG List Gather/Input memory | + * | Length = multiple of 32Bytes | + * | Alignment = 8Byte | + * |---------------------------------- | + * | SG List Scatter/Output memory | + * | Length = multiple of 32Bytes | + * | Alignment = 8Byte | + * | -------------------------------| + * | | padding for 32B alignment | + * |------------------------------------| + * | Result response memory | + * | Alignment = 32Byte | + * ------------------------------------ + */ + + info_len = sizeof(*info); + + g_len = ((req->in_cnt + 2) / 3) * + sizeof(struct cn10kb_cpt_sglist_component); + s_len = ((req->out_cnt + 2) / 3) * + sizeof(struct cn10kb_cpt_sglist_component); + sg_len = g_len + s_len; - g_len = ALIGN(g_sz_bytes, align); - sg_len = ALIGN(g_len + s_sz_bytes, align); - info_len = ALIGN(sizeof(*info), align); - total_mem_len = sg_len + info_len + sizeof(union otx2_cpt_res_s); + /* Allocate extra memory for SG and response address alignment */ + total_mem_len = ALIGN(info_len, OTX2_CPT_DPTR_RPTR_ALIGN); + total_mem_len += (ARCH_DMA_MINALIGN - 1) & + ~(OTX2_CPT_DPTR_RPTR_ALIGN - 1); + total_mem_len += ALIGN(sg_len, OTX2_CPT_RES_ADDR_ALIGN); + total_mem_len += sizeof(union otx2_cpt_res_s); info = kzalloc(total_mem_len, gfp); if (unlikely(!info)) @@ -372,7 +401,8 @@ cn10k_sgv2_info_create(struct pci_dev *pdev, struct otx2_cpt_req_info *req, dlen += req->in[i].size; info->dlen = dlen; - info->in_buffer = (u8 *)info + info_len; + info->in_buffer = PTR_ALIGN((u8 *)info + info_len, ARCH_DMA_MINALIGN); + info->out_buffer = info->in_buffer + g_len; info->gthr_sz = req->in_cnt; info->sctr_sz = req->out_cnt; @@ -384,7 +414,7 @@ cn10k_sgv2_info_create(struct pci_dev *pdev, struct otx2_cpt_req_info *req, } if (sgv2io_components_setup(pdev, req->out, req->out_cnt, - &info->in_buffer[g_len])) { + info->out_buffer)) { dev_err(&pdev->dev, "Failed to setup scatter list\n"); goto destroy_info; } @@ -401,8 +431,10 @@ cn10k_sgv2_info_create(struct pci_dev *pdev, struct otx2_cpt_req_info *req, * Get buffer for union otx2_cpt_res_s response * structure and its physical address */ - info->completion_addr = info->in_buffer + sg_len; - info->comp_baddr = info->dptr_baddr + sg_len; + info->completion_addr = PTR_ALIGN((info->in_buffer + sg_len), + OTX2_CPT_RES_ADDR_ALIGN); + info->comp_baddr = ALIGN((info->dptr_baddr + sg_len), + OTX2_CPT_RES_ADDR_ALIGN); return info; @@ -417,10 +449,9 @@ static inline struct otx2_cpt_inst_info * otx2_sg_info_create(struct pci_dev *pdev, struct otx2_cpt_req_info *req, gfp_t gfp) { - int align = OTX2_CPT_DMA_MINALIGN; struct otx2_cpt_inst_info *info; - u32 dlen, align_dlen, info_len; - u16 g_sz_bytes, s_sz_bytes; + u32 dlen, info_len; + u16 g_len, s_len; u32 total_mem_len; if (unlikely(req->in_cnt > OTX2_CPT_MAX_SG_IN_CNT || @@ -429,22 +460,54 @@ otx2_sg_info_create(struct pci_dev *pdev, struct otx2_cpt_req_info *req, return NULL; } - g_sz_bytes = ((req->in_cnt + 3) / 4) * - sizeof(struct otx2_cpt_sglist_component); - s_sz_bytes = ((req->out_cnt + 3) / 4) * - sizeof(struct otx2_cpt_sglist_component); + /* Allocate memory to meet below alignment requirement: + * ------------------------------------ + * | struct otx2_cpt_inst_info | + * | (No alignment required) | + * | --------------------------------| + * | | padding for ARCH_DMA_MINALIGN | + * | | alignment | + * |------------------------------------| + * | SG List Header of 8 Byte | + * |------------------------------------| + * | SG List Gather/Input memory | + * | Length = multiple of 32Bytes | + * | Alignment = 8Byte | + * |---------------------------------- | + * | SG List Scatter/Output memory | + * | Length = multiple of 32Bytes | + * | Alignment = 8Byte | + * | -------------------------------| + * | | padding for 32B alignment | + * |------------------------------------| + * | Result response memory | + * | Alignment = 32Byte | + * ------------------------------------ + */ + + info_len = sizeof(*info); + + g_len = ((req->in_cnt + 3) / 4) * + sizeof(struct otx2_cpt_sglist_component); + s_len = ((req->out_cnt + 3) / 4) * + sizeof(struct otx2_cpt_sglist_component); + + dlen = g_len + s_len + SG_LIST_HDR_SIZE; - dlen = g_sz_bytes + s_sz_bytes + SG_LIST_HDR_SIZE; - align_dlen = ALIGN(dlen, align); - info_len = ALIGN(sizeof(*info), align); - total_mem_len = align_dlen + info_len + sizeof(union otx2_cpt_res_s); + /* Allocate extra memory for SG and response address alignment */ + total_mem_len = ALIGN(info_len, OTX2_CPT_DPTR_RPTR_ALIGN); + total_mem_len += (ARCH_DMA_MINALIGN - 1) & + ~(OTX2_CPT_DPTR_RPTR_ALIGN - 1); + total_mem_len += ALIGN(dlen, OTX2_CPT_RES_ADDR_ALIGN); + total_mem_len += sizeof(union otx2_cpt_res_s); info = kzalloc(total_mem_len, gfp); if (unlikely(!info)) return NULL; info->dlen = dlen; - info->in_buffer = (u8 *)info + info_len; + info->in_buffer = PTR_ALIGN((u8 *)info + info_len, ARCH_DMA_MINALIGN); + info->out_buffer = info->in_buffer + SG_LIST_HDR_SIZE + g_len; ((u16 *)info->in_buffer)[0] = req->out_cnt; ((u16 *)info->in_buffer)[1] = req->in_cnt; @@ -460,7 +523,7 @@ otx2_sg_info_create(struct pci_dev *pdev, struct otx2_cpt_req_info *req, } if (setup_sgio_components(pdev, req->out, req->out_cnt, - &info->in_buffer[8 + g_sz_bytes])) { + info->out_buffer)) { dev_err(&pdev->dev, "Failed to setup scatter list\n"); goto destroy_info; } @@ -476,8 +539,10 @@ otx2_sg_info_create(struct pci_dev *pdev, struct otx2_cpt_req_info *req, * Get buffer for union otx2_cpt_res_s response * structure and its physical address */ - info->completion_addr = info->in_buffer + align_dlen; - info->comp_baddr = info->dptr_baddr + align_dlen; + info->completion_addr = PTR_ALIGN((info->in_buffer + dlen), + OTX2_CPT_RES_ADDR_ALIGN); + info->comp_baddr = ALIGN((info->dptr_baddr + dlen), + OTX2_CPT_RES_ADDR_ALIGN); return info; @@ -490,6 +555,7 @@ struct otx2_cptlf_wqe; int otx2_cpt_do_request(struct pci_dev *pdev, struct otx2_cpt_req_info *req, int cpu_num); void otx2_cpt_post_process(struct otx2_cptlf_wqe *wqe); -int otx2_cpt_get_kcrypto_eng_grp_num(struct pci_dev *pdev); +int otx2_cpt_get_eng_grp_num(struct pci_dev *pdev, + enum otx2_cpt_eng_type); #endif /* __OTX2_CPT_REQMGR_H */ diff --git a/drivers/crypto/marvell/octeontx2/otx2_cptlf.h b/drivers/crypto/marvell/octeontx2/otx2_cptlf.h index 6e004a5568d88..1b9f75214d18a 100644 --- a/drivers/crypto/marvell/octeontx2/otx2_cptlf.h +++ b/drivers/crypto/marvell/octeontx2/otx2_cptlf.h @@ -124,7 +124,8 @@ struct otx2_cptlfs_info { struct cpt_hw_ops *ops; u8 are_lfs_attached; /* Whether CPT LFs are attached */ u8 lfs_num; /* Number of CPT LFs */ - u8 kcrypto_eng_grp_num; /* Kernel crypto engine group number */ + u8 kcrypto_se_eng_grp_num; /* Crypto symmetric engine group number */ + u8 kcrypto_ae_eng_grp_num; /* Crypto asymmetric engine group number */ u8 kvf_limits; /* Kernel crypto limits */ atomic_t state; /* LF's state. started/reset */ int blkaddr; /* CPT blkaddr: BLKADDR_CPT0/BLKADDR_CPT1 */ diff --git a/drivers/crypto/marvell/octeontx2/otx2_cptpf_ucode.c b/drivers/crypto/marvell/octeontx2/otx2_cptpf_ucode.c index 7180944ece500..cc47e361089a0 100644 --- a/drivers/crypto/marvell/octeontx2/otx2_cptpf_ucode.c +++ b/drivers/crypto/marvell/octeontx2/otx2_cptpf_ucode.c @@ -1493,11 +1493,13 @@ int otx2_cpt_discover_eng_capabilities(struct otx2_cptpf_dev *cptpf) union otx2_cpt_opcode opcode; union otx2_cpt_res_s *result; union otx2_cpt_inst_s inst; + dma_addr_t result_baddr; dma_addr_t rptr_baddr; struct pci_dev *pdev; - u32 len, compl_rlen; + int timeout = 10000; + void *base, *rptr; int ret, etype; - void *rptr; + u32 len; /* * We don't get capabilities if it was already done @@ -1520,22 +1522,28 @@ int otx2_cpt_discover_eng_capabilities(struct otx2_cptpf_dev *cptpf) if (ret) goto delete_grps; - compl_rlen = ALIGN(sizeof(union otx2_cpt_res_s), OTX2_CPT_DMA_MINALIGN); - len = compl_rlen + LOADFVC_RLEN; + /* Allocate extra memory for "rptr" and "result" pointer alignment */ + len = LOADFVC_RLEN + ARCH_DMA_MINALIGN + + sizeof(union otx2_cpt_res_s) + OTX2_CPT_RES_ADDR_ALIGN; - result = kzalloc(len, GFP_KERNEL); - if (!result) { + base = kzalloc(len, GFP_KERNEL); + if (!base) { ret = -ENOMEM; goto lf_cleanup; } - rptr_baddr = dma_map_single(&pdev->dev, (void *)result, len, - DMA_BIDIRECTIONAL); + + rptr = PTR_ALIGN(base, ARCH_DMA_MINALIGN); + rptr_baddr = dma_map_single(&pdev->dev, rptr, len, DMA_BIDIRECTIONAL); if (dma_mapping_error(&pdev->dev, rptr_baddr)) { dev_err(&pdev->dev, "DMA mapping failed\n"); ret = -EFAULT; - goto free_result; + goto free_rptr; } - rptr = (u8 *)result + compl_rlen; + + result = (union otx2_cpt_res_s *)PTR_ALIGN(rptr + LOADFVC_RLEN, + OTX2_CPT_RES_ADDR_ALIGN); + result_baddr = ALIGN(rptr_baddr + LOADFVC_RLEN, + OTX2_CPT_RES_ADDR_ALIGN); /* Fill in the command */ opcode.s.major = LOADFVC_MAJOR_OP; @@ -1547,27 +1555,38 @@ int otx2_cpt_discover_eng_capabilities(struct otx2_cptpf_dev *cptpf) /* 64-bit swap for microcode data reads, not needed for addresses */ cpu_to_be64s(&iq_cmd.cmd.u); iq_cmd.dptr = 0; - iq_cmd.rptr = rptr_baddr + compl_rlen; + iq_cmd.rptr = rptr_baddr; iq_cmd.cptr.u = 0; for (etype = 1; etype < OTX2_CPT_MAX_ENG_TYPES; etype++) { result->s.compcode = OTX2_CPT_COMPLETION_CODE_INIT; iq_cmd.cptr.s.grp = otx2_cpt_get_eng_grp(&cptpf->eng_grps, etype); - otx2_cpt_fill_inst(&inst, &iq_cmd, rptr_baddr); + otx2_cpt_fill_inst(&inst, &iq_cmd, result_baddr); lfs->ops->send_cmd(&inst, 1, &cptpf->lfs.lf[0]); + timeout = 10000; while (lfs->ops->cpt_get_compcode(result) == - OTX2_CPT_COMPLETION_CODE_INIT) + OTX2_CPT_COMPLETION_CODE_INIT) { cpu_relax(); + udelay(1); + timeout--; + if (!timeout) { + ret = -ENODEV; + cptpf->is_eng_caps_discovered = false; + dev_warn(&pdev->dev, "Timeout on CPT load_fvc completion poll\n"); + goto error_no_response; + } + } cptpf->eng_caps[etype].u = be64_to_cpup(rptr); } - dma_unmap_single(&pdev->dev, rptr_baddr, len, DMA_BIDIRECTIONAL); cptpf->is_eng_caps_discovered = true; -free_result: - kfree(result); +error_no_response: + dma_unmap_single(&pdev->dev, rptr_baddr, len, DMA_BIDIRECTIONAL); +free_rptr: + kfree(base); lf_cleanup: otx2_cptlf_shutdown(lfs); delete_grps: diff --git a/drivers/crypto/marvell/octeontx2/otx2_cptvf_algs.c b/drivers/crypto/marvell/octeontx2/otx2_cptvf_algs.c index 7eb0bc13994d9..8d9f394d6b50e 100644 --- a/drivers/crypto/marvell/octeontx2/otx2_cptvf_algs.c +++ b/drivers/crypto/marvell/octeontx2/otx2_cptvf_algs.c @@ -384,7 +384,8 @@ static inline int cpt_enc_dec(struct skcipher_request *req, u32 enc) req_info->req_type = OTX2_CPT_ENC_DEC_REQ; req_info->is_enc = enc; req_info->is_trunc_hmac = false; - req_info->ctrl.s.grp = otx2_cpt_get_kcrypto_eng_grp_num(pdev); + req_info->ctrl.s.grp = otx2_cpt_get_eng_grp_num(pdev, + OTX2_CPT_SE_TYPES); req_info->req.cptr = ctx->er_ctx.hw_ctx; req_info->req.cptr_dma = ctx->er_ctx.cptr_dma; @@ -1288,7 +1289,8 @@ static int cpt_aead_enc_dec(struct aead_request *req, u8 reg_type, u8 enc) if (status) return status; - req_info->ctrl.s.grp = otx2_cpt_get_kcrypto_eng_grp_num(pdev); + req_info->ctrl.s.grp = otx2_cpt_get_eng_grp_num(pdev, + OTX2_CPT_SE_TYPES); /* * We perform an asynchronous send and once diff --git a/drivers/crypto/marvell/octeontx2/otx2_cptvf_main.c b/drivers/crypto/marvell/octeontx2/otx2_cptvf_main.c index 56904bdfd6e82..c1c44a7b89fa8 100644 --- a/drivers/crypto/marvell/octeontx2/otx2_cptvf_main.c +++ b/drivers/crypto/marvell/octeontx2/otx2_cptvf_main.c @@ -265,17 +265,33 @@ static int cptvf_lf_init(struct otx2_cptvf_dev *cptvf) u8 eng_grp_msk; /* Get engine group number for symmetric crypto */ - cptvf->lfs.kcrypto_eng_grp_num = OTX2_CPT_INVALID_CRYPTO_ENG_GRP; + cptvf->lfs.kcrypto_se_eng_grp_num = OTX2_CPT_INVALID_CRYPTO_ENG_GRP; ret = otx2_cptvf_send_eng_grp_num_msg(cptvf, OTX2_CPT_SE_TYPES); if (ret) return ret; - if (cptvf->lfs.kcrypto_eng_grp_num == OTX2_CPT_INVALID_CRYPTO_ENG_GRP) { - dev_err(dev, "Engine group for kernel crypto not available\n"); - ret = -ENOENT; + if (cptvf->lfs.kcrypto_se_eng_grp_num == + OTX2_CPT_INVALID_CRYPTO_ENG_GRP) { + dev_err(dev, + "Symmetric Engine group for crypto not available\n"); + return -ENOENT; + } + + /* Get engine group number for asymmetric crypto */ + cptvf->lfs.kcrypto_ae_eng_grp_num = OTX2_CPT_INVALID_CRYPTO_ENG_GRP; + ret = otx2_cptvf_send_eng_grp_num_msg(cptvf, OTX2_CPT_AE_TYPES); + if (ret) return ret; + + if (cptvf->lfs.kcrypto_ae_eng_grp_num == + OTX2_CPT_INVALID_CRYPTO_ENG_GRP) { + dev_err(dev, + "Asymmetric Engine group for crypto not available\n"); + return -ENOENT; } - eng_grp_msk = 1 << cptvf->lfs.kcrypto_eng_grp_num; + + eng_grp_msk = BIT(cptvf->lfs.kcrypto_se_eng_grp_num) | + BIT(cptvf->lfs.kcrypto_ae_eng_grp_num); ret = otx2_cptvf_send_kvf_limits_msg(cptvf); if (ret) diff --git a/drivers/crypto/marvell/octeontx2/otx2_cptvf_mbox.c b/drivers/crypto/marvell/octeontx2/otx2_cptvf_mbox.c index 92e49babd79af..5277bcfa275e0 100644 --- a/drivers/crypto/marvell/octeontx2/otx2_cptvf_mbox.c +++ b/drivers/crypto/marvell/octeontx2/otx2_cptvf_mbox.c @@ -75,6 +75,7 @@ static void process_pfvf_mbox_mbox_msg(struct otx2_cptvf_dev *cptvf, struct otx2_cpt_caps_rsp *eng_caps; struct cpt_rd_wr_reg_msg *rsp_reg; struct msix_offset_rsp *rsp_msix; + u8 grp_num; int i; if (msg->id >= MBOX_MSG_MAX) { @@ -122,7 +123,11 @@ static void process_pfvf_mbox_mbox_msg(struct otx2_cptvf_dev *cptvf, break; case MBOX_MSG_GET_ENG_GRP_NUM: rsp_grp = (struct otx2_cpt_egrp_num_rsp *) msg; - cptvf->lfs.kcrypto_eng_grp_num = rsp_grp->eng_grp_num; + grp_num = rsp_grp->eng_grp_num; + if (rsp_grp->eng_type == OTX2_CPT_SE_TYPES) + cptvf->lfs.kcrypto_se_eng_grp_num = grp_num; + else if (rsp_grp->eng_type == OTX2_CPT_AE_TYPES) + cptvf->lfs.kcrypto_ae_eng_grp_num = grp_num; break; case MBOX_MSG_GET_KVF_LIMITS: rsp_limits = (struct otx2_cpt_kvf_limits_rsp *) msg; diff --git a/drivers/crypto/marvell/octeontx2/otx2_cptvf_reqmgr.c b/drivers/crypto/marvell/octeontx2/otx2_cptvf_reqmgr.c index 4262441070372..e71494486c641 100644 --- a/drivers/crypto/marvell/octeontx2/otx2_cptvf_reqmgr.c +++ b/drivers/crypto/marvell/octeontx2/otx2_cptvf_reqmgr.c @@ -391,9 +391,19 @@ void otx2_cpt_post_process(struct otx2_cptlf_wqe *wqe) &wqe->lfs->lf[wqe->lf_num].pqueue); } -int otx2_cpt_get_kcrypto_eng_grp_num(struct pci_dev *pdev) +int otx2_cpt_get_eng_grp_num(struct pci_dev *pdev, + enum otx2_cpt_eng_type eng_type) { struct otx2_cptvf_dev *cptvf = pci_get_drvdata(pdev); - return cptvf->lfs.kcrypto_eng_grp_num; + switch (eng_type) { + case OTX2_CPT_SE_TYPES: + return cptvf->lfs.kcrypto_se_eng_grp_num; + case OTX2_CPT_AE_TYPES: + return cptvf->lfs.kcrypto_ae_eng_grp_num; + default: + dev_err(&cptvf->pdev->dev, "Unsupported engine type"); + break; + } + return -ENXIO; } diff --git a/drivers/crypto/omap-aes-gcm.c b/drivers/crypto/omap-aes-gcm.c index c498950402e89..1f4586509ca41 100644 --- a/drivers/crypto/omap-aes-gcm.c +++ b/drivers/crypto/omap-aes-gcm.c @@ -38,7 +38,6 @@ static void omap_aes_gcm_finish_req(struct omap_aes_dev *dd, int ret) crypto_finalize_aead_request(dd->engine, req, ret); - pm_runtime_mark_last_busy(dd->dev); pm_runtime_put_autosuspend(dd->dev); } diff --git a/drivers/crypto/omap-aes.c b/drivers/crypto/omap-aes.c index 1ecf5f6ac04e8..244e24e52987e 100644 --- a/drivers/crypto/omap-aes.c +++ b/drivers/crypto/omap-aes.c @@ -400,7 +400,6 @@ static void omap_aes_finish_req(struct omap_aes_dev *dd, int err) crypto_finalize_skcipher_request(dd->engine, req, err); - pm_runtime_mark_last_busy(dd->dev); pm_runtime_put_autosuspend(dd->dev); } diff --git a/drivers/crypto/omap-des.c b/drivers/crypto/omap-des.c index a099460d5f215..9c5538ae17db2 100644 --- a/drivers/crypto/omap-des.c +++ b/drivers/crypto/omap-des.c @@ -489,7 +489,6 @@ static void omap_des_finish_req(struct omap_des_dev *dd, int err) crypto_finalize_skcipher_request(dd->engine, req, err); - pm_runtime_mark_last_busy(dd->dev); pm_runtime_put_autosuspend(dd->dev); } diff --git a/drivers/crypto/omap-sham.c b/drivers/crypto/omap-sham.c index 56f192cb976dc..6328e8026b91e 100644 --- a/drivers/crypto/omap-sham.c +++ b/drivers/crypto/omap-sham.c @@ -1167,7 +1167,6 @@ static void omap_sham_finish_req(struct ahash_request *req, int err) dd->flags &= ~(BIT(FLAGS_FINAL) | BIT(FLAGS_CPU) | BIT(FLAGS_DMA_READY) | BIT(FLAGS_OUTPUT_READY)); - pm_runtime_mark_last_busy(dd->dev); pm_runtime_put_autosuspend(dd->dev); ctx->offset = 0; diff --git a/drivers/crypto/stm32/stm32-cryp.c b/drivers/crypto/stm32/stm32-cryp.c index 5ce88e7a8f657..a89b4c5d62a09 100644 --- a/drivers/crypto/stm32/stm32-cryp.c +++ b/drivers/crypto/stm32/stm32-cryp.c @@ -851,7 +851,6 @@ static void stm32_cryp_finish_req(struct stm32_cryp *cryp, int err) if (!err && (!(is_gcm(cryp) || is_ccm(cryp) || is_ecb(cryp)))) stm32_cryp_get_iv(cryp); - pm_runtime_mark_last_busy(cryp->dev); pm_runtime_put_autosuspend(cryp->dev); if (is_gcm(cryp) || is_ccm(cryp)) diff --git a/drivers/crypto/stm32/stm32-hash.c b/drivers/crypto/stm32/stm32-hash.c index 768b27de47378..a4436728b0dba 100644 --- a/drivers/crypto/stm32/stm32-hash.c +++ b/drivers/crypto/stm32/stm32-hash.c @@ -1373,7 +1373,6 @@ static void stm32_hash_unprepare_request(struct ahash_request *req) *preg++ = stm32_hash_read(hdev, HASH_CSR(i)); pm_runtime: - pm_runtime_mark_last_busy(hdev->dev); pm_runtime_put_autosuspend(hdev->dev); } diff --git a/drivers/crypto/virtio/virtio_crypto_common.h b/drivers/crypto/virtio/virtio_crypto_common.h index 7059bbe5a2eba..19c934af3df6b 100644 --- a/drivers/crypto/virtio/virtio_crypto_common.h +++ b/drivers/crypto/virtio/virtio_crypto_common.h @@ -113,8 +113,6 @@ struct virtio_crypto_request { int virtcrypto_devmgr_add_dev(struct virtio_crypto *vcrypto_dev); struct list_head *virtcrypto_devmgr_get_head(void); void virtcrypto_devmgr_rm_dev(struct virtio_crypto *vcrypto_dev); -struct virtio_crypto *virtcrypto_devmgr_get_first(void); -int virtcrypto_dev_in_use(struct virtio_crypto *vcrypto_dev); int virtcrypto_dev_get(struct virtio_crypto *vcrypto_dev); void virtcrypto_dev_put(struct virtio_crypto *vcrypto_dev); int virtcrypto_dev_started(struct virtio_crypto *vcrypto_dev); diff --git a/drivers/crypto/virtio/virtio_crypto_core.c b/drivers/crypto/virtio/virtio_crypto_core.c index 0d522049f595b..3d241446099cc 100644 --- a/drivers/crypto/virtio/virtio_crypto_core.c +++ b/drivers/crypto/virtio/virtio_crypto_core.c @@ -139,7 +139,7 @@ static int virtcrypto_find_vqs(struct virtio_crypto *vi) spin_lock_init(&vi->data_vq[i].lock); vi->data_vq[i].vq = vqs[i]; /* Initialize crypto engine */ - vi->data_vq[i].engine = crypto_engine_alloc_init_and_set(dev, true, NULL, true, + vi->data_vq[i].engine = crypto_engine_alloc_init_and_set(dev, true, true, virtqueue_get_vring_size(vqs[i])); if (!vi->data_vq[i].engine) { ret = -ENOMEM; diff --git a/drivers/crypto/virtio/virtio_crypto_mgr.c b/drivers/crypto/virtio/virtio_crypto_mgr.c index bddbd8ebfebe6..06c74fa132cdd 100644 --- a/drivers/crypto/virtio/virtio_crypto_mgr.c +++ b/drivers/crypto/virtio/virtio_crypto_mgr.c @@ -82,42 +82,6 @@ void virtcrypto_devmgr_rm_dev(struct virtio_crypto *vcrypto_dev) } /* - * virtcrypto_devmgr_get_first() - * - * Function returns the first virtio crypto device from the acceleration - * framework. - * - * To be used by virtio crypto device specific drivers. - * - * Return: pointer to vcrypto_dev or NULL if not found. - */ -struct virtio_crypto *virtcrypto_devmgr_get_first(void) -{ - struct virtio_crypto *dev = NULL; - - mutex_lock(&table_lock); - if (!list_empty(&virtio_crypto_table)) - dev = list_first_entry(&virtio_crypto_table, - struct virtio_crypto, - list); - mutex_unlock(&table_lock); - return dev; -} - -/* - * virtcrypto_dev_in_use() - Check whether vcrypto_dev is currently in use - * @vcrypto_dev: Pointer to virtio crypto device. - * - * To be used by virtio crypto device specific drivers. - * - * Return: 1 when device is in use, 0 otherwise. - */ -int virtcrypto_dev_in_use(struct virtio_crypto *vcrypto_dev) -{ - return atomic_read(&vcrypto_dev->ref_count) != 0; -} - -/* * virtcrypto_dev_get() - Increment vcrypto_dev reference count * @vcrypto_dev: Pointer to virtio crypto device. * diff --git a/include/crypto/engine.h b/include/crypto/engine.h index 545dbefe3e13c..2e60344437daf 100644 --- a/include/crypto/engine.h +++ b/include/crypto/engine.h @@ -76,7 +76,6 @@ int crypto_engine_stop(struct crypto_engine *engine); struct crypto_engine *crypto_engine_alloc_init(struct device *dev, bool rt); struct crypto_engine *crypto_engine_alloc_init_and_set(struct device *dev, bool retry_support, - int (*cbk_do_batch)(struct crypto_engine *engine), bool rt, int qlen); void crypto_engine_exit(struct crypto_engine *engine); diff --git a/include/crypto/internal/acompress.h b/include/crypto/internal/acompress.h index ffffd88bbbad3..2d97440028ffd 100644 --- a/include/crypto/internal/acompress.h +++ b/include/crypto/internal/acompress.h @@ -63,10 +63,7 @@ struct crypto_acomp_stream { struct crypto_acomp_streams { /* These must come first because of struct scomp_alg. */ void *(*alloc_ctx)(void); - union { - void (*free_ctx)(void *); - void (*cfree_ctx)(const void *); - }; + void (*free_ctx)(void *); struct crypto_acomp_stream __percpu *streams; struct work_struct stream_work; diff --git a/include/crypto/internal/engine.h b/include/crypto/internal/engine.h index b6a4ea2240fca..f19ef376833f3 100644 --- a/include/crypto/internal/engine.h +++ b/include/crypto/internal/engine.h @@ -21,7 +21,6 @@ struct device; /* * struct crypto_engine - crypto hardware engine * @name: the engine name - * @idling: the engine is entering idle state * @busy: request pump is busy * @running: the engine is on working * @retry_support: indication that the hardware allows re-execution @@ -31,14 +30,6 @@ struct device; * @list: link with the global crypto engine list * @queue_lock: spinlock to synchronise access to request queue * @queue: the crypto queue of the engine - * @prepare_crypt_hardware: a request will soon arrive from the queue - * so the subsystem requests the driver to prepare the hardware - * by issuing this call - * @unprepare_crypt_hardware: there are currently no more requests on the - * queue so the subsystem notifies the driver that it may relax the - * hardware by issuing this call - * @do_batch_requests: execute a batch of requests. Depends on multiple - * requests support. * @kworker: kthread worker struct for request pump * @pump_requests: work struct for scheduling work to the request pump * @priv_data: the engine private data @@ -46,7 +37,6 @@ struct device; */ struct crypto_engine { char name[ENGINE_NAME_LEN]; - bool idling; bool busy; bool running; @@ -58,11 +48,6 @@ struct crypto_engine { struct crypto_queue queue; struct device *dev; - int (*prepare_crypt_hardware)(struct crypto_engine *engine); - int (*unprepare_crypt_hardware)(struct crypto_engine *engine); - int (*do_batch_requests)(struct crypto_engine *engine); - - struct kthread_worker *kworker; struct kthread_work pump_requests; diff --git a/include/crypto/internal/hash.h b/include/crypto/internal/hash.h index 0f85c543f80bf..6ec5f2f37ccbe 100644 --- a/include/crypto/internal/hash.h +++ b/include/crypto/internal/hash.h @@ -30,6 +30,20 @@ __##name##_req, (req)) struct ahash_request; +struct scatterlist; + +struct crypto_hash_walk { + const char *data; + + unsigned int offset; + unsigned int flags; + + struct page *pg; + unsigned int entrylen; + + unsigned int total; + struct scatterlist *sg; +}; struct ahash_instance { void (*free)(struct ahash_instance *inst); @@ -61,6 +75,15 @@ struct crypto_shash_spawn { struct crypto_spawn base; }; +int crypto_hash_walk_done(struct crypto_hash_walk *walk, int err); +int crypto_hash_walk_first(struct ahash_request *req, + struct crypto_hash_walk *walk); + +static inline int crypto_hash_walk_last(struct crypto_hash_walk *walk) +{ + return !(walk->entrylen | walk->total); +} + int crypto_register_ahash(struct ahash_alg *alg); void crypto_unregister_ahash(struct ahash_alg *alg); int crypto_register_ahashes(struct ahash_alg *algs, int count); @@ -91,6 +114,12 @@ static inline bool crypto_hash_alg_needs_key(struct hash_alg_common *alg) !(alg->base.cra_flags & CRYPTO_ALG_OPTIONAL_KEY); } +static inline bool crypto_hash_no_export_core(struct crypto_ahash *tfm) +{ + return crypto_hash_alg_common(tfm)->base.cra_flags & + CRYPTO_AHASH_ALG_NO_EXPORT_CORE; +} + int crypto_grab_ahash(struct crypto_ahash_spawn *spawn, struct crypto_instance *inst, const char *name, u32 type, u32 mask); @@ -167,6 +196,13 @@ static inline void crypto_ahash_set_reqsize(struct crypto_ahash *tfm, tfm->reqsize = reqsize; } +static inline bool crypto_ahash_tested(struct crypto_ahash *tfm) +{ + struct crypto_tfm *tfm_base = crypto_ahash_tfm(tfm); + + return tfm_base->__crt_alg->cra_flags & CRYPTO_ALG_TESTED; +} + static inline void crypto_ahash_set_reqsize_dma(struct crypto_ahash *ahash, unsigned int reqsize) { diff --git a/include/linux/crypto.h b/include/linux/crypto.h index b50f1954d1bba..a2137e19be7d8 100644 --- a/include/linux/crypto.h +++ b/include/linux/crypto.h @@ -136,6 +136,9 @@ /* Set if the algorithm supports virtual addresses. */ #define CRYPTO_ALG_REQ_VIRT 0x00040000 +/* Set if the algorithm cannot have a fallback (e.g., phmac). */ +#define CRYPTO_ALG_NO_FALLBACK 0x00080000 + /* The high bits 0xff000000 are reserved for type-specific flags. */ /* diff --git a/include/linux/hisi_acc_qm.h b/include/linux/hisi_acc_qm.h index 99fcf65d575fa..0c4c84b8c3be9 100644 --- a/include/linux/hisi_acc_qm.h +++ b/include/linux/hisi_acc_qm.h @@ -556,9 +556,9 @@ int hisi_qm_mb(struct hisi_qm *qm, u8 cmd, dma_addr_t dma_addr, u16 queue, struct hisi_acc_sgl_pool; struct hisi_acc_hw_sgl *hisi_acc_sg_buf_map_to_hw_sgl(struct device *dev, struct scatterlist *sgl, struct hisi_acc_sgl_pool *pool, - u32 index, dma_addr_t *hw_sgl_dma); + u32 index, dma_addr_t *hw_sgl_dma, enum dma_data_direction dir); void hisi_acc_sg_buf_unmap(struct device *dev, struct scatterlist *sgl, - struct hisi_acc_hw_sgl *hw_sgl); + struct hisi_acc_hw_sgl *hw_sgl, enum dma_data_direction dir); struct hisi_acc_sgl_pool *hisi_acc_create_sgl_pool(struct device *dev, u32 count, u32 sge_nr); void hisi_acc_free_sgl_pool(struct device *dev, diff --git a/include/linux/padata.h b/include/linux/padata.h index 0146daf344306..765f2778e264a 100644 --- a/include/linux/padata.h +++ b/include/linux/padata.h @@ -90,8 +90,6 @@ struct padata_cpumask { * @processed: Number of already processed objects. * @cpu: Next CPU to be processed. * @cpumask: The cpumasks in use for parallel and serial workers. - * @reorder_work: work struct for reordering. - * @lock: Reorder lock. */ struct parallel_data { struct padata_shell *ps; @@ -102,8 +100,6 @@ struct parallel_data { unsigned int processed; int cpu; struct padata_cpumask cpumask; - struct work_struct reorder_work; - spinlock_t ____cacheline_aligned lock; }; /** diff --git a/kernel/padata.c b/kernel/padata.c index 7eee94166357a..f85f8bd788d0d 100644 --- a/kernel/padata.c +++ b/kernel/padata.c @@ -63,17 +63,6 @@ static inline void padata_put_pd(struct parallel_data *pd) padata_put_pd_cnt(pd, 1); } -static int padata_index_to_cpu(struct parallel_data *pd, int cpu_index) -{ - int cpu, target_cpu; - - target_cpu = cpumask_first(pd->cpumask.pcpu); - for (cpu = 0; cpu < cpu_index; cpu++) - target_cpu = cpumask_next(target_cpu, pd->cpumask.pcpu); - - return target_cpu; -} - static int padata_cpu_hash(struct parallel_data *pd, unsigned int seq_nr) { /* @@ -82,7 +71,7 @@ static int padata_cpu_hash(struct parallel_data *pd, unsigned int seq_nr) */ int cpu_index = seq_nr % cpumask_weight(pd->cpumask.pcpu); - return padata_index_to_cpu(pd, cpu_index); + return cpumask_nth(cpu_index, pd->cpumask.pcpu); } static struct padata_work *padata_work_alloc(void) @@ -192,9 +181,9 @@ int padata_do_parallel(struct padata_shell *ps, struct padata_priv *padata, int *cb_cpu) { struct padata_instance *pinst = ps->pinst; - int i, cpu, cpu_index, err; struct parallel_data *pd; struct padata_work *pw; + int cpu_index, err; rcu_read_lock_bh(); @@ -210,12 +199,7 @@ int padata_do_parallel(struct padata_shell *ps, /* Select an alternate fallback CPU and notify the caller. */ cpu_index = *cb_cpu % cpumask_weight(pd->cpumask.cbcpu); - - cpu = cpumask_first(pd->cpumask.cbcpu); - for (i = 0; i < cpu_index; i++) - cpu = cpumask_next(cpu, pd->cpumask.cbcpu); - - *cb_cpu = cpu; + *cb_cpu = cpumask_nth(cpu_index, pd->cpumask.cbcpu); } err = -EBUSY; @@ -261,20 +245,17 @@ EXPORT_SYMBOL(padata_do_parallel); * be parallel processed by another cpu and is not yet present in * the cpu's reorder queue. */ -static struct padata_priv *padata_find_next(struct parallel_data *pd, - bool remove_object) +static struct padata_priv *padata_find_next(struct parallel_data *pd, int cpu, + unsigned int processed) { struct padata_priv *padata; struct padata_list *reorder; - int cpu = pd->cpu; reorder = per_cpu_ptr(pd->reorder_list, cpu); spin_lock(&reorder->lock); - if (list_empty(&reorder->list)) { - spin_unlock(&reorder->lock); - return NULL; - } + if (list_empty(&reorder->list)) + goto notfound; padata = list_entry(reorder->list.next, struct padata_priv, list); @@ -282,97 +263,52 @@ static struct padata_priv *padata_find_next(struct parallel_data *pd, * Checks the rare case where two or more parallel jobs have hashed to * the same CPU and one of the later ones finishes first. */ - if (padata->seq_nr != pd->processed) { - spin_unlock(&reorder->lock); - return NULL; - } - - if (remove_object) { - list_del_init(&padata->list); - ++pd->processed; - pd->cpu = cpumask_next_wrap(cpu, pd->cpumask.pcpu); - } + if (padata->seq_nr != processed) + goto notfound; + list_del_init(&padata->list); spin_unlock(&reorder->lock); return padata; + +notfound: + pd->processed = processed; + pd->cpu = cpu; + spin_unlock(&reorder->lock); + return NULL; } -static void padata_reorder(struct parallel_data *pd) +static void padata_reorder(struct padata_priv *padata) { + struct parallel_data *pd = padata->pd; struct padata_instance *pinst = pd->ps->pinst; - int cb_cpu; - struct padata_priv *padata; - struct padata_serial_queue *squeue; - struct padata_list *reorder; + unsigned int processed; + int cpu; - /* - * We need to ensure that only one cpu can work on dequeueing of - * the reorder queue the time. Calculating in which percpu reorder - * queue the next object will arrive takes some time. A spinlock - * would be highly contended. Also it is not clear in which order - * the objects arrive to the reorder queues. So a cpu could wait to - * get the lock just to notice that there is nothing to do at the - * moment. Therefore we use a trylock and let the holder of the lock - * care for all the objects enqueued during the holdtime of the lock. - */ - if (!spin_trylock_bh(&pd->lock)) - return; + processed = pd->processed; + cpu = pd->cpu; - while (1) { - padata = padata_find_next(pd, true); + do { + struct padata_serial_queue *squeue; + int cb_cpu; - /* - * If the next object that needs serialization is parallel - * processed by another cpu and is still on it's way to the - * cpu's reorder queue, nothing to do for now. - */ - if (!padata) - break; + cpu = cpumask_next_wrap(cpu, pd->cpumask.pcpu); + processed++; cb_cpu = padata->cb_cpu; squeue = per_cpu_ptr(pd->squeue, cb_cpu); spin_lock(&squeue->serial.lock); list_add_tail(&padata->list, &squeue->serial.list); - spin_unlock(&squeue->serial.lock); - queue_work_on(cb_cpu, pinst->serial_wq, &squeue->work); - } - - spin_unlock_bh(&pd->lock); - - /* - * The next object that needs serialization might have arrived to - * the reorder queues in the meantime. - * - * Ensure reorder queue is read after pd->lock is dropped so we see - * new objects from another task in padata_do_serial. Pairs with - * smp_mb in padata_do_serial. - */ - smp_mb(); - reorder = per_cpu_ptr(pd->reorder_list, pd->cpu); - if (!list_empty(&reorder->list) && padata_find_next(pd, false)) { /* - * Other context(eg. the padata_serial_worker) can finish the request. - * To avoid UAF issue, add pd ref here, and put pd ref after reorder_work finish. + * If the next object that needs serialization is parallel + * processed by another cpu and is still on it's way to the + * cpu's reorder queue, end the loop. */ - padata_get_pd(pd); - if (!queue_work(pinst->serial_wq, &pd->reorder_work)) - padata_put_pd(pd); - } -} - -static void invoke_padata_reorder(struct work_struct *work) -{ - struct parallel_data *pd; - - local_bh_disable(); - pd = container_of(work, struct parallel_data, reorder_work); - padata_reorder(pd); - local_bh_enable(); - /* Pairs with putting the reorder_work in the serial_wq */ - padata_put_pd(pd); + padata = padata_find_next(pd, cpu, processed); + spin_unlock(&squeue->serial.lock); + } while (padata); } static void padata_serial_worker(struct work_struct *serial_work) @@ -423,6 +359,7 @@ void padata_do_serial(struct padata_priv *padata) struct padata_list *reorder = per_cpu_ptr(pd->reorder_list, hashed_cpu); struct padata_priv *cur; struct list_head *pos; + bool gotit = true; spin_lock(&reorder->lock); /* Sort in ascending order of sequence number. */ @@ -432,17 +369,14 @@ void padata_do_serial(struct padata_priv *padata) if ((signed int)(cur->seq_nr - padata->seq_nr) < 0) break; } - list_add(&padata->list, pos); + if (padata->seq_nr != pd->processed) { + gotit = false; + list_add(&padata->list, pos); + } spin_unlock(&reorder->lock); - /* - * Ensure the addition to the reorder list is ordered correctly - * with the trylock of pd->lock in padata_reorder. Pairs with smp_mb - * in padata_reorder. - */ - smp_mb(); - - padata_reorder(pd); + if (gotit) + padata_reorder(padata); } EXPORT_SYMBOL(padata_do_serial); @@ -632,9 +566,7 @@ static struct parallel_data *padata_alloc_pd(struct padata_shell *ps) padata_init_squeues(pd); pd->seq_nr = -1; refcount_set(&pd->refcnt, 1); - spin_lock_init(&pd->lock); pd->cpu = cpumask_first(pd->cpumask.pcpu); - INIT_WORK(&pd->reorder_work, invoke_padata_reorder); return pd; @@ -1144,12 +1076,6 @@ void padata_free_shell(struct padata_shell *ps) if (!ps) return; - /* - * Wait for all _do_serial calls to finish to avoid touching - * freed pd's and ps's. - */ - synchronize_rcu(); - mutex_lock(&ps->pinst->lock); list_del(&ps->list); pd = rcu_dereference_protected(ps->pd, 1); |