diff options
Diffstat (limited to 'arch/powerpc')
33 files changed, 635 insertions, 470 deletions
diff --git a/arch/powerpc/crypto/aes-spe-glue.c b/arch/powerpc/crypto/aes-spe-glue.c index 3a4ca7d32477..1fad5d4c658d 100644 --- a/arch/powerpc/crypto/aes-spe-glue.c +++ b/arch/powerpc/crypto/aes-spe-glue.c @@ -17,7 +17,10 @@ #include <asm/byteorder.h> #include <asm/switch_to.h> #include <crypto/algapi.h> +#include <crypto/internal/skcipher.h> #include <crypto/xts.h> +#include <crypto/gf128mul.h> +#include <crypto/scatterwalk.h> /* * MAX_BYTES defines the number of bytes that are allowed to be processed @@ -118,13 +121,19 @@ static int ppc_aes_setkey(struct crypto_tfm *tfm, const u8 *in_key, return 0; } -static int ppc_xts_setkey(struct crypto_tfm *tfm, const u8 *in_key, +static int ppc_aes_setkey_skcipher(struct crypto_skcipher *tfm, + const u8 *in_key, unsigned int key_len) +{ + return ppc_aes_setkey(crypto_skcipher_tfm(tfm), in_key, key_len); +} + +static int ppc_xts_setkey(struct crypto_skcipher *tfm, const u8 *in_key, unsigned int key_len) { - struct ppc_xts_ctx *ctx = crypto_tfm_ctx(tfm); + struct ppc_xts_ctx *ctx = crypto_skcipher_ctx(tfm); int err; - err = xts_check_key(tfm, in_key, key_len); + err = xts_verify_key(tfm, in_key, key_len); if (err) return err; @@ -133,7 +142,7 @@ static int ppc_xts_setkey(struct crypto_tfm *tfm, const u8 *in_key, if (key_len != AES_KEYSIZE_128 && key_len != AES_KEYSIZE_192 && key_len != AES_KEYSIZE_256) { - tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; + crypto_skcipher_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN); return -EINVAL; } @@ -178,208 +187,229 @@ static void ppc_aes_decrypt(struct crypto_tfm *tfm, u8 *out, const u8 *in) spe_end(); } -static int ppc_ecb_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) +static int ppc_ecb_crypt(struct skcipher_request *req, bool enc) { - struct ppc_aes_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - struct blkcipher_walk walk; - unsigned int ubytes; + struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req); + struct ppc_aes_ctx *ctx = crypto_skcipher_ctx(tfm); + struct skcipher_walk walk; + unsigned int nbytes; int err; - desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); + err = skcipher_walk_virt(&walk, req, false); - while ((nbytes = walk.nbytes)) { - ubytes = nbytes > MAX_BYTES ? - nbytes - MAX_BYTES : nbytes & (AES_BLOCK_SIZE - 1); - nbytes -= ubytes; + while ((nbytes = walk.nbytes) != 0) { + nbytes = min_t(unsigned int, nbytes, MAX_BYTES); + nbytes = round_down(nbytes, AES_BLOCK_SIZE); spe_begin(); - ppc_encrypt_ecb(walk.dst.virt.addr, walk.src.virt.addr, - ctx->key_enc, ctx->rounds, nbytes); + if (enc) + ppc_encrypt_ecb(walk.dst.virt.addr, walk.src.virt.addr, + ctx->key_enc, ctx->rounds, nbytes); + else + ppc_decrypt_ecb(walk.dst.virt.addr, walk.src.virt.addr, + ctx->key_dec, ctx->rounds, nbytes); spe_end(); - err = blkcipher_walk_done(desc, &walk, ubytes); + err = skcipher_walk_done(&walk, walk.nbytes - nbytes); } return err; } -static int ppc_ecb_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) +static int ppc_ecb_encrypt(struct skcipher_request *req) { - struct ppc_aes_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - struct blkcipher_walk walk; - unsigned int ubytes; - int err; - - desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); - - while ((nbytes = walk.nbytes)) { - ubytes = nbytes > MAX_BYTES ? - nbytes - MAX_BYTES : nbytes & (AES_BLOCK_SIZE - 1); - nbytes -= ubytes; - - spe_begin(); - ppc_decrypt_ecb(walk.dst.virt.addr, walk.src.virt.addr, - ctx->key_dec, ctx->rounds, nbytes); - spe_end(); - - err = blkcipher_walk_done(desc, &walk, ubytes); - } + return ppc_ecb_crypt(req, true); +} - return err; +static int ppc_ecb_decrypt(struct skcipher_request *req) +{ + return ppc_ecb_crypt(req, false); } -static int ppc_cbc_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) +static int ppc_cbc_crypt(struct skcipher_request *req, bool enc) { - struct ppc_aes_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - struct blkcipher_walk walk; - unsigned int ubytes; + struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req); + struct ppc_aes_ctx *ctx = crypto_skcipher_ctx(tfm); + struct skcipher_walk walk; + unsigned int nbytes; int err; - desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); + err = skcipher_walk_virt(&walk, req, false); - while ((nbytes = walk.nbytes)) { - ubytes = nbytes > MAX_BYTES ? - nbytes - MAX_BYTES : nbytes & (AES_BLOCK_SIZE - 1); - nbytes -= ubytes; + while ((nbytes = walk.nbytes) != 0) { + nbytes = min_t(unsigned int, nbytes, MAX_BYTES); + nbytes = round_down(nbytes, AES_BLOCK_SIZE); spe_begin(); - ppc_encrypt_cbc(walk.dst.virt.addr, walk.src.virt.addr, - ctx->key_enc, ctx->rounds, nbytes, walk.iv); + if (enc) + ppc_encrypt_cbc(walk.dst.virt.addr, walk.src.virt.addr, + ctx->key_enc, ctx->rounds, nbytes, + walk.iv); + else + ppc_decrypt_cbc(walk.dst.virt.addr, walk.src.virt.addr, + ctx->key_dec, ctx->rounds, nbytes, + walk.iv); spe_end(); - err = blkcipher_walk_done(desc, &walk, ubytes); + err = skcipher_walk_done(&walk, walk.nbytes - nbytes); } return err; } -static int ppc_cbc_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) +static int ppc_cbc_encrypt(struct skcipher_request *req) { - struct ppc_aes_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - struct blkcipher_walk walk; - unsigned int ubytes; - int err; - - desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); - - while ((nbytes = walk.nbytes)) { - ubytes = nbytes > MAX_BYTES ? - nbytes - MAX_BYTES : nbytes & (AES_BLOCK_SIZE - 1); - nbytes -= ubytes; - - spe_begin(); - ppc_decrypt_cbc(walk.dst.virt.addr, walk.src.virt.addr, - ctx->key_dec, ctx->rounds, nbytes, walk.iv); - spe_end(); - - err = blkcipher_walk_done(desc, &walk, ubytes); - } + return ppc_cbc_crypt(req, true); +} - return err; +static int ppc_cbc_decrypt(struct skcipher_request *req) +{ + return ppc_cbc_crypt(req, false); } -static int ppc_ctr_crypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) +static int ppc_ctr_crypt(struct skcipher_request *req) { - struct ppc_aes_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - struct blkcipher_walk walk; - unsigned int pbytes, ubytes; + struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req); + struct ppc_aes_ctx *ctx = crypto_skcipher_ctx(tfm); + struct skcipher_walk walk; + unsigned int nbytes; int err; - desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt_block(desc, &walk, AES_BLOCK_SIZE); + err = skcipher_walk_virt(&walk, req, false); - while ((pbytes = walk.nbytes)) { - pbytes = pbytes > MAX_BYTES ? MAX_BYTES : pbytes; - pbytes = pbytes == nbytes ? - nbytes : pbytes & ~(AES_BLOCK_SIZE - 1); - ubytes = walk.nbytes - pbytes; + while ((nbytes = walk.nbytes) != 0) { + nbytes = min_t(unsigned int, nbytes, MAX_BYTES); + if (nbytes < walk.total) + nbytes = round_down(nbytes, AES_BLOCK_SIZE); spe_begin(); ppc_crypt_ctr(walk.dst.virt.addr, walk.src.virt.addr, - ctx->key_enc, ctx->rounds, pbytes , walk.iv); + ctx->key_enc, ctx->rounds, nbytes, walk.iv); spe_end(); - nbytes -= pbytes; - err = blkcipher_walk_done(desc, &walk, ubytes); + err = skcipher_walk_done(&walk, walk.nbytes - nbytes); } return err; } -static int ppc_xts_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) +static int ppc_xts_crypt(struct skcipher_request *req, bool enc) { - struct ppc_xts_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - struct blkcipher_walk walk; - unsigned int ubytes; + struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req); + struct ppc_xts_ctx *ctx = crypto_skcipher_ctx(tfm); + struct skcipher_walk walk; + unsigned int nbytes; int err; u32 *twk; - desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); + err = skcipher_walk_virt(&walk, req, false); twk = ctx->key_twk; - while ((nbytes = walk.nbytes)) { - ubytes = nbytes > MAX_BYTES ? - nbytes - MAX_BYTES : nbytes & (AES_BLOCK_SIZE - 1); - nbytes -= ubytes; + while ((nbytes = walk.nbytes) != 0) { + nbytes = min_t(unsigned int, nbytes, MAX_BYTES); + nbytes = round_down(nbytes, AES_BLOCK_SIZE); spe_begin(); - ppc_encrypt_xts(walk.dst.virt.addr, walk.src.virt.addr, - ctx->key_enc, ctx->rounds, nbytes, walk.iv, twk); + if (enc) + ppc_encrypt_xts(walk.dst.virt.addr, walk.src.virt.addr, + ctx->key_enc, ctx->rounds, nbytes, + walk.iv, twk); + else + ppc_decrypt_xts(walk.dst.virt.addr, walk.src.virt.addr, + ctx->key_dec, ctx->rounds, nbytes, + walk.iv, twk); spe_end(); twk = NULL; - err = blkcipher_walk_done(desc, &walk, ubytes); + err = skcipher_walk_done(&walk, walk.nbytes - nbytes); } return err; } -static int ppc_xts_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, - struct scatterlist *src, unsigned int nbytes) +static int ppc_xts_encrypt(struct skcipher_request *req) { - struct ppc_xts_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); - struct blkcipher_walk walk; - unsigned int ubytes; + struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req); + struct ppc_xts_ctx *ctx = crypto_skcipher_ctx(tfm); + int tail = req->cryptlen % AES_BLOCK_SIZE; + int offset = req->cryptlen - tail - AES_BLOCK_SIZE; + struct skcipher_request subreq; + u8 b[2][AES_BLOCK_SIZE]; int err; - u32 *twk; - desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; - blkcipher_walk_init(&walk, dst, src, nbytes); - err = blkcipher_walk_virt(desc, &walk); - twk = ctx->key_twk; + if (req->cryptlen < AES_BLOCK_SIZE) + return -EINVAL; + + if (tail) { + subreq = *req; + skcipher_request_set_crypt(&subreq, req->src, req->dst, + req->cryptlen - tail, req->iv); + req = &subreq; + } - while ((nbytes = walk.nbytes)) { - ubytes = nbytes > MAX_BYTES ? - nbytes - MAX_BYTES : nbytes & (AES_BLOCK_SIZE - 1); - nbytes -= ubytes; + err = ppc_xts_crypt(req, true); + if (err || !tail) + return err; - spe_begin(); - ppc_decrypt_xts(walk.dst.virt.addr, walk.src.virt.addr, - ctx->key_dec, ctx->rounds, nbytes, walk.iv, twk); - spe_end(); + scatterwalk_map_and_copy(b[0], req->dst, offset, AES_BLOCK_SIZE, 0); + memcpy(b[1], b[0], tail); + scatterwalk_map_and_copy(b[0], req->src, offset + AES_BLOCK_SIZE, tail, 0); - twk = NULL; - err = blkcipher_walk_done(desc, &walk, ubytes); + spe_begin(); + ppc_encrypt_xts(b[0], b[0], ctx->key_enc, ctx->rounds, AES_BLOCK_SIZE, + req->iv, NULL); + spe_end(); + + scatterwalk_map_and_copy(b[0], req->dst, offset, AES_BLOCK_SIZE + tail, 1); + + return 0; +} + +static int ppc_xts_decrypt(struct skcipher_request *req) +{ + struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(req); + struct ppc_xts_ctx *ctx = crypto_skcipher_ctx(tfm); + int tail = req->cryptlen % AES_BLOCK_SIZE; + int offset = req->cryptlen - tail - AES_BLOCK_SIZE; + struct skcipher_request subreq; + u8 b[3][AES_BLOCK_SIZE]; + le128 twk; + int err; + + if (req->cryptlen < AES_BLOCK_SIZE) + return -EINVAL; + + if (tail) { + subreq = *req; + skcipher_request_set_crypt(&subreq, req->src, req->dst, + offset, req->iv); + req = &subreq; } - return err; + err = ppc_xts_crypt(req, false); + if (err || !tail) + return err; + + scatterwalk_map_and_copy(b[1], req->src, offset, AES_BLOCK_SIZE + tail, 0); + + spe_begin(); + if (!offset) + ppc_encrypt_ecb(req->iv, req->iv, ctx->key_twk, ctx->rounds, + AES_BLOCK_SIZE); + + gf128mul_x_ble(&twk, (le128 *)req->iv); + + ppc_decrypt_xts(b[1], b[1], ctx->key_dec, ctx->rounds, AES_BLOCK_SIZE, + (u8 *)&twk, NULL); + memcpy(b[0], b[2], tail); + memcpy(b[0] + tail, b[1] + tail, AES_BLOCK_SIZE - tail); + ppc_decrypt_xts(b[0], b[0], ctx->key_dec, ctx->rounds, AES_BLOCK_SIZE, + req->iv, NULL); + spe_end(); + + scatterwalk_map_and_copy(b[0], req->dst, offset, AES_BLOCK_SIZE + tail, 1); + + return 0; } /* @@ -388,9 +418,9 @@ static int ppc_xts_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, * This improves IPsec thoughput by another few percent. Additionally we assume * that AES context is always aligned to at least 8 bytes because it is created * with kmalloc() in the crypto infrastructure - * */ -static struct crypto_alg aes_algs[] = { { + +static struct crypto_alg aes_cipher_alg = { .cra_name = "aes", .cra_driver_name = "aes-ppc-spe", .cra_priority = 300, @@ -408,96 +438,84 @@ static struct crypto_alg aes_algs[] = { { .cia_decrypt = ppc_aes_decrypt } } -}, { - .cra_name = "ecb(aes)", - .cra_driver_name = "ecb-ppc-spe", - .cra_priority = 300, - .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, - .cra_blocksize = AES_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct ppc_aes_ctx), - .cra_alignmask = 0, - .cra_type = &crypto_blkcipher_type, - .cra_module = THIS_MODULE, - .cra_u = { - .blkcipher = { - .min_keysize = AES_MIN_KEY_SIZE, - .max_keysize = AES_MAX_KEY_SIZE, - .ivsize = AES_BLOCK_SIZE, - .setkey = ppc_aes_setkey, - .encrypt = ppc_ecb_encrypt, - .decrypt = ppc_ecb_decrypt, - } - } -}, { - .cra_name = "cbc(aes)", - .cra_driver_name = "cbc-ppc-spe", - .cra_priority = 300, - .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, - .cra_blocksize = AES_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct ppc_aes_ctx), - .cra_alignmask = 0, - .cra_type = &crypto_blkcipher_type, - .cra_module = THIS_MODULE, - .cra_u = { - .blkcipher = { - .min_keysize = AES_MIN_KEY_SIZE, - .max_keysize = AES_MAX_KEY_SIZE, - .ivsize = AES_BLOCK_SIZE, - .setkey = ppc_aes_setkey, - .encrypt = ppc_cbc_encrypt, - .decrypt = ppc_cbc_decrypt, - } - } -}, { - .cra_name = "ctr(aes)", - .cra_driver_name = "ctr-ppc-spe", - .cra_priority = 300, - .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, - .cra_blocksize = 1, - .cra_ctxsize = sizeof(struct ppc_aes_ctx), - .cra_alignmask = 0, - .cra_type = &crypto_blkcipher_type, - .cra_module = THIS_MODULE, - .cra_u = { - .blkcipher = { - .min_keysize = AES_MIN_KEY_SIZE, - .max_keysize = AES_MAX_KEY_SIZE, - .ivsize = AES_BLOCK_SIZE, - .setkey = ppc_aes_setkey, - .encrypt = ppc_ctr_crypt, - .decrypt = ppc_ctr_crypt, - } - } -}, { - .cra_name = "xts(aes)", - .cra_driver_name = "xts-ppc-spe", - .cra_priority = 300, - .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, - .cra_blocksize = AES_BLOCK_SIZE, - .cra_ctxsize = sizeof(struct ppc_xts_ctx), - .cra_alignmask = 0, - .cra_type = &crypto_blkcipher_type, - .cra_module = THIS_MODULE, - .cra_u = { - .blkcipher = { - .min_keysize = AES_MIN_KEY_SIZE * 2, - .max_keysize = AES_MAX_KEY_SIZE * 2, - .ivsize = AES_BLOCK_SIZE, - .setkey = ppc_xts_setkey, - .encrypt = ppc_xts_encrypt, - .decrypt = ppc_xts_decrypt, - } +}; + +static struct skcipher_alg aes_skcipher_algs[] = { + { + .base.cra_name = "ecb(aes)", + .base.cra_driver_name = "ecb-ppc-spe", + .base.cra_priority = 300, + .base.cra_blocksize = AES_BLOCK_SIZE, + .base.cra_ctxsize = sizeof(struct ppc_aes_ctx), + .base.cra_module = THIS_MODULE, + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .setkey = ppc_aes_setkey_skcipher, + .encrypt = ppc_ecb_encrypt, + .decrypt = ppc_ecb_decrypt, + }, { + .base.cra_name = "cbc(aes)", + .base.cra_driver_name = "cbc-ppc-spe", + .base.cra_priority = 300, + .base.cra_blocksize = AES_BLOCK_SIZE, + .base.cra_ctxsize = sizeof(struct ppc_aes_ctx), + .base.cra_module = THIS_MODULE, + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + .setkey = ppc_aes_setkey_skcipher, + .encrypt = ppc_cbc_encrypt, + .decrypt = ppc_cbc_decrypt, + }, { + .base.cra_name = "ctr(aes)", + .base.cra_driver_name = "ctr-ppc-spe", + .base.cra_priority = 300, + .base.cra_blocksize = 1, + .base.cra_ctxsize = sizeof(struct ppc_aes_ctx), + .base.cra_module = THIS_MODULE, + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + .setkey = ppc_aes_setkey_skcipher, + .encrypt = ppc_ctr_crypt, + .decrypt = ppc_ctr_crypt, + .chunksize = AES_BLOCK_SIZE, + }, { + .base.cra_name = "xts(aes)", + .base.cra_driver_name = "xts-ppc-spe", + .base.cra_priority = 300, + .base.cra_blocksize = AES_BLOCK_SIZE, + .base.cra_ctxsize = sizeof(struct ppc_xts_ctx), + .base.cra_module = THIS_MODULE, + .min_keysize = AES_MIN_KEY_SIZE * 2, + .max_keysize = AES_MAX_KEY_SIZE * 2, + .ivsize = AES_BLOCK_SIZE, + .setkey = ppc_xts_setkey, + .encrypt = ppc_xts_encrypt, + .decrypt = ppc_xts_decrypt, } -} }; +}; static int __init ppc_aes_mod_init(void) { - return crypto_register_algs(aes_algs, ARRAY_SIZE(aes_algs)); + int err; + + err = crypto_register_alg(&aes_cipher_alg); + if (err) + return err; + + err = crypto_register_skciphers(aes_skcipher_algs, + ARRAY_SIZE(aes_skcipher_algs)); + if (err) + crypto_unregister_alg(&aes_cipher_alg); + return err; } static void __exit ppc_aes_mod_fini(void) { - crypto_unregister_algs(aes_algs, ARRAY_SIZE(aes_algs)); + crypto_unregister_alg(&aes_cipher_alg); + crypto_unregister_skciphers(aes_skcipher_algs, + ARRAY_SIZE(aes_skcipher_algs)); } module_init(ppc_aes_mod_init); diff --git a/arch/powerpc/include/asm/book3s/32/kup.h b/arch/powerpc/include/asm/book3s/32/kup.h index 677e9babef80..f9dc597b0b86 100644 --- a/arch/powerpc/include/asm/book3s/32/kup.h +++ b/arch/powerpc/include/asm/book3s/32/kup.h @@ -91,6 +91,7 @@ static inline void kuap_update_sr(u32 sr, u32 addr, u32 end) { + addr &= 0xf0000000; /* align addr to start of segment */ barrier(); /* make sure thread.kuap is updated before playing with SRs */ while (addr < end) { mtsrin(sr, addr); diff --git a/arch/powerpc/include/asm/elf.h b/arch/powerpc/include/asm/elf.h index 409c9bfb43d9..57c229a86f08 100644 --- a/arch/powerpc/include/asm/elf.h +++ b/arch/powerpc/include/asm/elf.h @@ -175,4 +175,7 @@ do { \ ARCH_DLINFO_CACHE_GEOMETRY; \ } while (0) +/* Relocate the kernel image to @final_address */ +void relocate(unsigned long final_address); + #endif /* _ASM_POWERPC_ELF_H */ diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h index 6fe6ad64cba5..4273e799203d 100644 --- a/arch/powerpc/include/asm/kvm_host.h +++ b/arch/powerpc/include/asm/kvm_host.h @@ -401,7 +401,6 @@ struct kvmppc_mmu { u32 (*mfsrin)(struct kvm_vcpu *vcpu, u32 srnum); int (*xlate)(struct kvm_vcpu *vcpu, gva_t eaddr, struct kvmppc_pte *pte, bool data, bool iswrite); - void (*reset_msr)(struct kvm_vcpu *vcpu); void (*tlbie)(struct kvm_vcpu *vcpu, ulong addr, bool large); int (*esid_to_vsid)(struct kvm_vcpu *vcpu, ulong esid, u64 *vsid); u64 (*ea_to_vp)(struct kvm_vcpu *vcpu, gva_t eaddr, bool data); diff --git a/arch/powerpc/include/asm/kvm_ppc.h b/arch/powerpc/include/asm/kvm_ppc.h index ee62776e5433..d63f649fe713 100644 --- a/arch/powerpc/include/asm/kvm_ppc.h +++ b/arch/powerpc/include/asm/kvm_ppc.h @@ -271,6 +271,7 @@ struct kvmppc_ops { union kvmppc_one_reg *val); void (*vcpu_load)(struct kvm_vcpu *vcpu, int cpu); void (*vcpu_put)(struct kvm_vcpu *vcpu); + void (*inject_interrupt)(struct kvm_vcpu *vcpu, int vec, u64 srr1_flags); void (*set_msr)(struct kvm_vcpu *vcpu, u64 msr); int (*vcpu_run)(struct kvm_run *run, struct kvm_vcpu *vcpu); struct kvm_vcpu *(*vcpu_create)(struct kvm *kvm, unsigned int id); diff --git a/arch/powerpc/include/asm/local.h b/arch/powerpc/include/asm/local.h index fdd00939270b..bc4bd19b7fc2 100644 --- a/arch/powerpc/include/asm/local.h +++ b/arch/powerpc/include/asm/local.h @@ -17,7 +17,7 @@ typedef struct #define LOCAL_INIT(i) { (i) } -static __inline__ long local_read(local_t *l) +static __inline__ long local_read(const local_t *l) { return READ_ONCE(l->v); } diff --git a/arch/powerpc/include/asm/page.h b/arch/powerpc/include/asm/page.h index c8bb14ff4713..f6c562acc3f8 100644 --- a/arch/powerpc/include/asm/page.h +++ b/arch/powerpc/include/asm/page.h @@ -329,13 +329,4 @@ struct vm_area_struct; #endif /* __ASSEMBLY__ */ #include <asm/slice.h> -/* - * Allow 30-bit DMA for very limited Broadcom wifi chips on many powerbooks. - */ -#ifdef CONFIG_PPC32 -#define ARCH_ZONE_DMA_BITS 30 -#else -#define ARCH_ZONE_DMA_BITS 31 -#endif - #endif /* _ASM_POWERPC_PAGE_H */ diff --git a/arch/powerpc/include/asm/reg.h b/arch/powerpc/include/asm/reg.h index b3cbb1136bce..75c7e95a321b 100644 --- a/arch/powerpc/include/asm/reg.h +++ b/arch/powerpc/include/asm/reg.h @@ -748,6 +748,18 @@ #define SPRN_USPRG7 0x107 /* SPRG7 userspace read */ #define SPRN_SRR0 0x01A /* Save/Restore Register 0 */ #define SPRN_SRR1 0x01B /* Save/Restore Register 1 */ + +#ifdef CONFIG_PPC_BOOK3S +/* + * Bits loaded from MSR upon interrupt. + * PPC (64-bit) bits 33-36,42-47 are interrupt dependent, the others are + * loaded from MSR. The exception is that SRESET and MCE do not always load + * bit 62 (RI) from MSR. Don't use PPC_BITMASK for this because 32-bit uses + * it. + */ +#define SRR1_MSR_BITS (~0x783f0000UL) +#endif + #define SRR1_ISI_NOPT 0x40000000 /* ISI: Not found in hash */ #define SRR1_ISI_N_OR_G 0x10000000 /* ISI: Access is no-exec or G */ #define SRR1_ISI_PROT 0x08000000 /* ISI: Other protection fault */ diff --git a/arch/powerpc/include/uapi/asm/kvm.h b/arch/powerpc/include/uapi/asm/kvm.h index b0f72dea8b11..264e266a85bf 100644 --- a/arch/powerpc/include/uapi/asm/kvm.h +++ b/arch/powerpc/include/uapi/asm/kvm.h @@ -667,6 +667,8 @@ struct kvm_ppc_cpu_char { /* PPC64 eXternal Interrupt Controller Specification */ #define KVM_DEV_XICS_GRP_SOURCES 1 /* 64-bit source attributes */ +#define KVM_DEV_XICS_GRP_CTRL 2 +#define KVM_DEV_XICS_NR_SERVERS 1 /* Layout of 64-bit source attribute values */ #define KVM_XICS_DESTINATION_SHIFT 0 @@ -683,6 +685,7 @@ struct kvm_ppc_cpu_char { #define KVM_DEV_XIVE_GRP_CTRL 1 #define KVM_DEV_XIVE_RESET 1 #define KVM_DEV_XIVE_EQ_SYNC 2 +#define KVM_DEV_XIVE_NR_SERVERS 3 #define KVM_DEV_XIVE_GRP_SOURCE 2 /* 64-bit source identifier */ #define KVM_DEV_XIVE_GRP_SOURCE_CONFIG 3 /* 64-bit source identifier */ #define KVM_DEV_XIVE_GRP_EQ_CONFIG 4 /* 64-bit EQ identifier */ diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c index a4e7762dd286..100f1b57ec2f 100644 --- a/arch/powerpc/kernel/prom_init.c +++ b/arch/powerpc/kernel/prom_init.c @@ -3249,7 +3249,20 @@ static void setup_secure_guest(unsigned long kbase, unsigned long fdt) /* Switch to secure mode. */ prom_printf("Switching to secure mode.\n"); + /* + * The ultravisor will do an integrity check of the kernel image but we + * relocated it so the check will fail. Restore the original image by + * relocating it back to the kernel virtual base address. + */ + if (IS_ENABLED(CONFIG_RELOCATABLE)) + relocate(KERNELBASE); + ret = enter_secure_mode(kbase, fdt); + + /* Relocate the kernel again. */ + if (IS_ENABLED(CONFIG_RELOCATABLE)) + relocate(kbase); + if (ret != U_SUCCESS) { prom_printf("Returned %d from switching to secure mode.\n", ret); prom_rtas_os_term("Switch to secure mode failed.\n"); diff --git a/arch/powerpc/kernel/prom_init_check.sh b/arch/powerpc/kernel/prom_init_check.sh index 78bab17b1396..b183ab9c5107 100644 --- a/arch/powerpc/kernel/prom_init_check.sh +++ b/arch/powerpc/kernel/prom_init_check.sh @@ -26,7 +26,8 @@ _end enter_prom $MEM_FUNCS reloc_offset __secondary_hold __secondary_hold_acknowledge __secondary_hold_spinloop __start logo_linux_clut224 btext_prepare_BAT reloc_got2 kernstart_addr memstart_addr linux_banner _stext -__prom_init_toc_start __prom_init_toc_end btext_setup_display TOC." +__prom_init_toc_start __prom_init_toc_end btext_setup_display TOC. +relocate" NM="$1" OBJ="$2" diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index 694522308cd5..84827da01d45 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -338,7 +338,7 @@ static unsigned long vtime_delta(struct task_struct *tsk, return stime; } -void vtime_account_system(struct task_struct *tsk) +void vtime_account_kernel(struct task_struct *tsk) { unsigned long stime, stime_scaled, steal_time; struct cpu_accounting_data *acct = get_accounting(tsk); @@ -366,7 +366,7 @@ void vtime_account_system(struct task_struct *tsk) #endif } } -EXPORT_SYMBOL_GPL(vtime_account_system); +EXPORT_SYMBOL_GPL(vtime_account_kernel); void vtime_account_idle(struct task_struct *tsk) { @@ -395,7 +395,7 @@ static void vtime_flush_scaled(struct task_struct *tsk, /* * Account the whole cputime accumulated in the paca * Must be called with interrupts disabled. - * Assumes that vtime_account_system/idle() has been called + * Assumes that vtime_account_kernel/idle() has been called * recently (i.e. since the last entry from usermode) so that * get_paca()->user_time_scaled is up to date. */ diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S index 060a1acd7c6d..8834220036a5 100644 --- a/arch/powerpc/kernel/vmlinux.lds.S +++ b/arch/powerpc/kernel/vmlinux.lds.S @@ -6,6 +6,8 @@ #endif #define BSS_FIRST_SECTIONS *(.bss.prominit) +#define EMITS_PT_NOTE +#define RO_EXCEPTION_TABLE_ALIGN 0 #include <asm/page.h> #include <asm-generic/vmlinux.lds.h> @@ -18,22 +20,8 @@ ENTRY(_stext) PHDRS { - kernel PT_LOAD FLAGS(7); /* RWX */ - notes PT_NOTE FLAGS(0); - dummy PT_NOTE FLAGS(0); - - /* binutils < 2.18 has a bug that makes it misbehave when taking an - ELF file with all segments at load address 0 as input. This - happens when running "strip" on vmlinux, because of the AT() magic - in this linker script. People using GCC >= 4.2 won't run into - this problem, because the "build-id" support will put some data - into the "notes" segment (at a non-zero load address). - - To work around this, we force some data into both the "dummy" - segment and the kernel segment, so the dummy segment will get a - non-zero load address. It's not enough to always create the - "notes" segment, since if nothing gets assigned to it, its load - address will be zero. */ + text PT_LOAD FLAGS(7); /* RWX */ + note PT_NOTE FLAGS(0); } #ifdef CONFIG_PPC64 @@ -77,7 +65,7 @@ SECTIONS #else /* !CONFIG_PPC64 */ HEAD_TEXT #endif - } :kernel + } :text __head_end = .; @@ -126,7 +114,7 @@ SECTIONS __got2_end = .; #endif /* CONFIG_PPC32 */ - } :kernel + } :text . = ALIGN(ETEXT_ALIGN_SIZE); _etext = .; @@ -175,17 +163,6 @@ SECTIONS __stop__btb_flush_fixup = .; } #endif - EXCEPTION_TABLE(0) - - NOTES :kernel :notes - - /* The dummy segment contents for the bug workaround mentioned above - near PHDRS. */ - .dummy : AT(ADDR(.dummy) - LOAD_OFFSET) { - LONG(0) - LONG(0) - LONG(0) - } :kernel :dummy /* * Init sections discarded at runtime @@ -200,7 +177,7 @@ SECTIONS #ifdef CONFIG_PPC64 *(.tramp.ftrace.init); #endif - } :kernel + } :text /* .exit.text is discarded at runtime, not link time, * to deal with references from __bug_table diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c index ec2547cc5ecb..58a59ee998e2 100644 --- a/arch/powerpc/kvm/book3s.c +++ b/arch/powerpc/kvm/book3s.c @@ -74,27 +74,6 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { NULL } }; -void kvmppc_unfixup_split_real(struct kvm_vcpu *vcpu) -{ - if (vcpu->arch.hflags & BOOK3S_HFLAG_SPLIT_HACK) { - ulong pc = kvmppc_get_pc(vcpu); - ulong lr = kvmppc_get_lr(vcpu); - if ((pc & SPLIT_HACK_MASK) == SPLIT_HACK_OFFS) - kvmppc_set_pc(vcpu, pc & ~SPLIT_HACK_MASK); - if ((lr & SPLIT_HACK_MASK) == SPLIT_HACK_OFFS) - kvmppc_set_lr(vcpu, lr & ~SPLIT_HACK_MASK); - vcpu->arch.hflags &= ~BOOK3S_HFLAG_SPLIT_HACK; - } -} -EXPORT_SYMBOL_GPL(kvmppc_unfixup_split_real); - -static inline unsigned long kvmppc_interrupt_offset(struct kvm_vcpu *vcpu) -{ - if (!is_kvmppc_hv_enabled(vcpu->kvm)) - return to_book3s(vcpu)->hior; - return 0; -} - static inline void kvmppc_update_int_pending(struct kvm_vcpu *vcpu, unsigned long pending_now, unsigned long old_pending) { @@ -134,11 +113,7 @@ static inline bool kvmppc_critical_section(struct kvm_vcpu *vcpu) void kvmppc_inject_interrupt(struct kvm_vcpu *vcpu, int vec, u64 flags) { - kvmppc_unfixup_split_real(vcpu); - kvmppc_set_srr0(vcpu, kvmppc_get_pc(vcpu)); - kvmppc_set_srr1(vcpu, (kvmppc_get_msr(vcpu) & ~0x783f0000ul) | flags); - kvmppc_set_pc(vcpu, kvmppc_interrupt_offset(vcpu) + vec); - vcpu->arch.mmu.reset_msr(vcpu); + vcpu->kvm->arch.kvm_ops->inject_interrupt(vcpu, vec, flags); } static int kvmppc_book3s_vec2irqprio(unsigned int vec) diff --git a/arch/powerpc/kvm/book3s.h b/arch/powerpc/kvm/book3s.h index 2ef1311a2a13..3a4613985949 100644 --- a/arch/powerpc/kvm/book3s.h +++ b/arch/powerpc/kvm/book3s.h @@ -32,4 +32,7 @@ extern void kvmppc_emulate_tabort(struct kvm_vcpu *vcpu, int ra_val); static inline void kvmppc_emulate_tabort(struct kvm_vcpu *vcpu, int ra_val) {} #endif +extern void kvmppc_set_msr_hv(struct kvm_vcpu *vcpu, u64 msr); +extern void kvmppc_inject_interrupt_hv(struct kvm_vcpu *vcpu, int vec, u64 srr1_flags); + #endif diff --git a/arch/powerpc/kvm/book3s_32_mmu.c b/arch/powerpc/kvm/book3s_32_mmu.c index 18f244aad7aa..f21e73492ce3 100644 --- a/arch/powerpc/kvm/book3s_32_mmu.c +++ b/arch/powerpc/kvm/book3s_32_mmu.c @@ -90,11 +90,6 @@ static u64 kvmppc_mmu_book3s_32_ea_to_vp(struct kvm_vcpu *vcpu, gva_t eaddr, return (((u64)eaddr >> 12) & 0xffff) | (vsid << 16); } -static void kvmppc_mmu_book3s_32_reset_msr(struct kvm_vcpu *vcpu) -{ - kvmppc_set_msr(vcpu, 0); -} - static hva_t kvmppc_mmu_book3s_32_get_pteg(struct kvm_vcpu *vcpu, u32 sre, gva_t eaddr, bool primary) @@ -406,7 +401,6 @@ void kvmppc_mmu_book3s_32_init(struct kvm_vcpu *vcpu) mmu->mtsrin = kvmppc_mmu_book3s_32_mtsrin; mmu->mfsrin = kvmppc_mmu_book3s_32_mfsrin; mmu->xlate = kvmppc_mmu_book3s_32_xlate; - mmu->reset_msr = kvmppc_mmu_book3s_32_reset_msr; mmu->tlbie = kvmppc_mmu_book3s_32_tlbie; mmu->esid_to_vsid = kvmppc_mmu_book3s_32_esid_to_vsid; mmu->ea_to_vp = kvmppc_mmu_book3s_32_ea_to_vp; diff --git a/arch/powerpc/kvm/book3s_64_mmu.c b/arch/powerpc/kvm/book3s_64_mmu.c index 5f63a5f7f24f..599133256a95 100644 --- a/arch/powerpc/kvm/book3s_64_mmu.c +++ b/arch/powerpc/kvm/book3s_64_mmu.c @@ -24,20 +24,6 @@ #define dprintk(X...) do { } while(0) #endif -static void kvmppc_mmu_book3s_64_reset_msr(struct kvm_vcpu *vcpu) -{ - unsigned long msr = vcpu->arch.intr_msr; - unsigned long cur_msr = kvmppc_get_msr(vcpu); - - /* If transactional, change to suspend mode on IRQ delivery */ - if (MSR_TM_TRANSACTIONAL(cur_msr)) - msr |= MSR_TS_S; - else - msr |= cur_msr & MSR_TS_MASK; - - kvmppc_set_msr(vcpu, msr); -} - static struct kvmppc_slb *kvmppc_mmu_book3s_64_find_slbe( struct kvm_vcpu *vcpu, gva_t eaddr) @@ -676,7 +662,6 @@ void kvmppc_mmu_book3s_64_init(struct kvm_vcpu *vcpu) mmu->slbie = kvmppc_mmu_book3s_64_slbie; mmu->slbia = kvmppc_mmu_book3s_64_slbia; mmu->xlate = kvmppc_mmu_book3s_64_xlate; - mmu->reset_msr = kvmppc_mmu_book3s_64_reset_msr; mmu->tlbie = kvmppc_mmu_book3s_64_tlbie; mmu->esid_to_vsid = kvmppc_mmu_book3s_64_esid_to_vsid; mmu->ea_to_vp = kvmppc_mmu_book3s_64_ea_to_vp; diff --git a/arch/powerpc/kvm/book3s_64_mmu_hv.c b/arch/powerpc/kvm/book3s_64_mmu_hv.c index 9a75f0e1933b..d381526c5c9b 100644 --- a/arch/powerpc/kvm/book3s_64_mmu_hv.c +++ b/arch/powerpc/kvm/book3s_64_mmu_hv.c @@ -275,18 +275,6 @@ int kvmppc_mmu_hv_init(void) return 0; } -static void kvmppc_mmu_book3s_64_hv_reset_msr(struct kvm_vcpu *vcpu) -{ - unsigned long msr = vcpu->arch.intr_msr; - - /* If transactional, change to suspend mode on IRQ delivery */ - if (MSR_TM_TRANSACTIONAL(vcpu->arch.shregs.msr)) - msr |= MSR_TS_S; - else - msr |= vcpu->arch.shregs.msr & MSR_TS_MASK; - kvmppc_set_msr(vcpu, msr); -} - static long kvmppc_virtmode_do_h_enter(struct kvm *kvm, unsigned long flags, long pte_index, unsigned long pteh, unsigned long ptel, unsigned long *pte_idx_ret) @@ -508,6 +496,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu, struct vm_area_struct *vma; unsigned long rcbits; long mmio_update; + struct mm_struct *mm; if (kvm_is_radix(kvm)) return kvmppc_book3s_radix_page_fault(run, vcpu, ea, dsisr); @@ -584,6 +573,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu, is_ci = false; pfn = 0; page = NULL; + mm = current->mm; pte_size = PAGE_SIZE; writing = (dsisr & DSISR_ISSTORE) != 0; /* If writing != 0, then the HPTE must allow writing, if we get here */ @@ -592,8 +582,8 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu, npages = get_user_pages_fast(hva, 1, writing ? FOLL_WRITE : 0, pages); if (npages < 1) { /* Check if it's an I/O mapping */ - down_read(¤t->mm->mmap_sem); - vma = find_vma(current->mm, hva); + down_read(&mm->mmap_sem); + vma = find_vma(mm, hva); if (vma && vma->vm_start <= hva && hva + psize <= vma->vm_end && (vma->vm_flags & VM_PFNMAP)) { pfn = vma->vm_pgoff + @@ -602,7 +592,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu, is_ci = pte_ci(__pte((pgprot_val(vma->vm_page_prot)))); write_ok = vma->vm_flags & VM_WRITE; } - up_read(¤t->mm->mmap_sem); + up_read(&mm->mmap_sem); if (!pfn) goto out_put; } else { @@ -621,8 +611,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu, * hugepage split and collapse. */ local_irq_save(flags); - ptep = find_current_mm_pte(current->mm->pgd, - hva, NULL, NULL); + ptep = find_current_mm_pte(mm->pgd, hva, NULL, NULL); if (ptep) { pte = kvmppc_read_update_linux_pte(ptep, 1); if (__pte_write(pte)) @@ -2000,7 +1989,7 @@ int kvm_vm_ioctl_get_htab_fd(struct kvm *kvm, struct kvm_get_htab_fd *ghf) ret = anon_inode_getfd("kvm-htab", &kvm_htab_fops, ctx, rwflag | O_CLOEXEC); if (ret < 0) { kfree(ctx); - kvm_put_kvm(kvm); + kvm_put_kvm_no_destroy(kvm); return ret; } @@ -2161,7 +2150,6 @@ void kvmppc_mmu_book3s_hv_init(struct kvm_vcpu *vcpu) vcpu->arch.slb_nr = 32; /* POWER7/POWER8 */ mmu->xlate = kvmppc_mmu_book3s_64_hv_xlate; - mmu->reset_msr = kvmppc_mmu_book3s_64_hv_reset_msr; vcpu->arch.hflags |= BOOK3S_HFLAG_SLB; } diff --git a/arch/powerpc/kvm/book3s_64_vio.c b/arch/powerpc/kvm/book3s_64_vio.c index 5834db0a54c6..883a66e76638 100644 --- a/arch/powerpc/kvm/book3s_64_vio.c +++ b/arch/powerpc/kvm/book3s_64_vio.c @@ -317,7 +317,7 @@ long kvm_vm_ioctl_create_spapr_tce(struct kvm *kvm, if (ret >= 0) list_add_rcu(&stt->list, &kvm->arch.spapr_tce_tables); else - kvm_put_kvm(kvm); + kvm_put_kvm_no_destroy(kvm); mutex_unlock(&kvm->lock); diff --git a/arch/powerpc/kvm/book3s_hv.c b/arch/powerpc/kvm/book3s_hv.c index 709cf1fd4cf4..ec5c0379296a 100644 --- a/arch/powerpc/kvm/book3s_hv.c +++ b/arch/powerpc/kvm/book3s_hv.c @@ -133,7 +133,6 @@ static inline bool nesting_enabled(struct kvm *kvm) /* If set, the threads on each CPU core have to be in the same MMU mode */ static bool no_mixing_hpt_and_radix; -static void kvmppc_end_cede(struct kvm_vcpu *vcpu); static int kvmppc_hv_setup_htab_rma(struct kvm_vcpu *vcpu); /* @@ -338,18 +337,6 @@ static void kvmppc_core_vcpu_put_hv(struct kvm_vcpu *vcpu) spin_unlock_irqrestore(&vcpu->arch.tbacct_lock, flags); } -static void kvmppc_set_msr_hv(struct kvm_vcpu *vcpu, u64 msr) -{ - /* - * Check for illegal transactional state bit combination - * and if we find it, force the TS field to a safe state. - */ - if ((msr & MSR_TS_MASK) == MSR_TS_MASK) - msr &= ~MSR_TS_MASK; - vcpu->arch.shregs.msr = msr; - kvmppc_end_cede(vcpu); -} - static void kvmppc_set_pvr_hv(struct kvm_vcpu *vcpu, u32 pvr) { vcpu->arch.pvr = pvr; @@ -792,6 +779,11 @@ static int kvmppc_h_set_mode(struct kvm_vcpu *vcpu, unsigned long mflags, vcpu->arch.dawr = value1; vcpu->arch.dawrx = value2; return H_SUCCESS; + case H_SET_MODE_RESOURCE_ADDR_TRANS_MODE: + /* KVM does not support mflags=2 (AIL=2) */ + if (mflags != 0 && mflags != 3) + return H_UNSUPPORTED_FLAG_START; + return H_TOO_HARD; default: return H_TOO_HARD; } @@ -2454,15 +2446,6 @@ static void kvmppc_set_timer(struct kvm_vcpu *vcpu) vcpu->arch.timer_running = 1; } -static void kvmppc_end_cede(struct kvm_vcpu *vcpu) -{ - vcpu->arch.ceded = 0; - if (vcpu->arch.timer_running) { - hrtimer_try_to_cancel(&vcpu->arch.dec_timer); - vcpu->arch.timer_running = 0; - } -} - extern int __kvmppc_vcore_entry(void); static void kvmppc_remove_runnable(struct kvmppc_vcore *vc, @@ -5401,6 +5384,7 @@ static struct kvmppc_ops kvm_ops_hv = { .set_one_reg = kvmppc_set_one_reg_hv, .vcpu_load = kvmppc_core_vcpu_load_hv, .vcpu_put = kvmppc_core_vcpu_put_hv, + .inject_interrupt = kvmppc_inject_interrupt_hv, .set_msr = kvmppc_set_msr_hv, .vcpu_run = kvmppc_vcpu_run_hv, .vcpu_create = kvmppc_core_vcpu_create_hv, diff --git a/arch/powerpc/kvm/book3s_hv_builtin.c b/arch/powerpc/kvm/book3s_hv_builtin.c index 7c1909657b55..7cd3cf3d366b 100644 --- a/arch/powerpc/kvm/book3s_hv_builtin.c +++ b/arch/powerpc/kvm/book3s_hv_builtin.c @@ -755,6 +755,71 @@ void kvmhv_p9_restore_lpcr(struct kvm_split_mode *sip) local_paca->kvm_hstate.kvm_split_mode = NULL; } +static void kvmppc_end_cede(struct kvm_vcpu *vcpu) +{ + vcpu->arch.ceded = 0; + if (vcpu->arch.timer_running) { + hrtimer_try_to_cancel(&vcpu->arch.dec_timer); + vcpu->arch.timer_running = 0; + } +} + +void kvmppc_set_msr_hv(struct kvm_vcpu *vcpu, u64 msr) +{ + /* + * Check for illegal transactional state bit combination + * and if we find it, force the TS field to a safe state. + */ + if ((msr & MSR_TS_MASK) == MSR_TS_MASK) + msr &= ~MSR_TS_MASK; + vcpu->arch.shregs.msr = msr; + kvmppc_end_cede(vcpu); +} +EXPORT_SYMBOL_GPL(kvmppc_set_msr_hv); + +static void inject_interrupt(struct kvm_vcpu *vcpu, int vec, u64 srr1_flags) +{ + unsigned long msr, pc, new_msr, new_pc; + + msr = kvmppc_get_msr(vcpu); + pc = kvmppc_get_pc(vcpu); + new_msr = vcpu->arch.intr_msr; + new_pc = vec; + + /* If transactional, change to suspend mode on IRQ delivery */ + if (MSR_TM_TRANSACTIONAL(msr)) + new_msr |= MSR_TS_S; + else + new_msr |= msr & MSR_TS_MASK; + + /* + * Perform MSR and PC adjustment for LPCR[AIL]=3 if it is set and + * applicable. AIL=2 is not supported. + * + * AIL does not apply to SRESET, MCE, or HMI (which is never + * delivered to the guest), and does not apply if IR=0 or DR=0. + */ + if (vec != BOOK3S_INTERRUPT_SYSTEM_RESET && + vec != BOOK3S_INTERRUPT_MACHINE_CHECK && + (vcpu->arch.vcore->lpcr & LPCR_AIL) == LPCR_AIL_3 && + (msr & (MSR_IR|MSR_DR)) == (MSR_IR|MSR_DR) ) { + new_msr |= MSR_IR | MSR_DR; + new_pc += 0xC000000000004000ULL; + } + + kvmppc_set_srr0(vcpu, pc); + kvmppc_set_srr1(vcpu, (msr & SRR1_MSR_BITS) | srr1_flags); + kvmppc_set_pc(vcpu, new_pc); + vcpu->arch.shregs.msr = new_msr; +} + +void kvmppc_inject_interrupt_hv(struct kvm_vcpu *vcpu, int vec, u64 srr1_flags) +{ + inject_interrupt(vcpu, vec, srr1_flags); + kvmppc_end_cede(vcpu); +} +EXPORT_SYMBOL_GPL(kvmppc_inject_interrupt_hv); + /* * Is there a PRIV_DOORBELL pending for the guest (on POWER9)? * Can we inject a Decrementer or a External interrupt? @@ -762,7 +827,6 @@ void kvmhv_p9_restore_lpcr(struct kvm_split_mode *sip) void kvmppc_guest_entry_inject_int(struct kvm_vcpu *vcpu) { int ext; - unsigned long vec = 0; unsigned long lpcr; /* Insert EXTERNAL bit into LPCR at the MER bit position */ @@ -774,26 +838,16 @@ void kvmppc_guest_entry_inject_int(struct kvm_vcpu *vcpu) if (vcpu->arch.shregs.msr & MSR_EE) { if (ext) { - vec = BOOK3S_INTERRUPT_EXTERNAL; + inject_interrupt(vcpu, BOOK3S_INTERRUPT_EXTERNAL, 0); } else { long int dec = mfspr(SPRN_DEC); if (!(lpcr & LPCR_LD)) dec = (int) dec; if (dec < 0) - vec = BOOK3S_INTERRUPT_DECREMENTER; + inject_interrupt(vcpu, + BOOK3S_INTERRUPT_DECREMENTER, 0); } } - if (vec) { - unsigned long msr, old_msr = vcpu->arch.shregs.msr; - - kvmppc_set_srr0(vcpu, kvmppc_get_pc(vcpu)); - kvmppc_set_srr1(vcpu, old_msr); - kvmppc_set_pc(vcpu, vec); - msr = vcpu->arch.intr_msr; - if (MSR_TM_ACTIVE(old_msr)) - msr |= MSR_TS_S; - vcpu->arch.shregs.msr = msr; - } if (vcpu->arch.doorbell_request) { mtspr(SPRN_DPDES, 1); diff --git a/arch/powerpc/kvm/book3s_hv_nested.c b/arch/powerpc/kvm/book3s_hv_nested.c index cdf30c6eaf54..dc97e5be76f6 100644 --- a/arch/powerpc/kvm/book3s_hv_nested.c +++ b/arch/powerpc/kvm/book3s_hv_nested.c @@ -1186,7 +1186,7 @@ static int kvmhv_translate_addr_nested(struct kvm_vcpu *vcpu, forward_to_l1: vcpu->arch.fault_dsisr = flags; if (vcpu->arch.trap == BOOK3S_INTERRUPT_H_INST_STORAGE) { - vcpu->arch.shregs.msr &= ~0x783f0000ul; + vcpu->arch.shregs.msr &= SRR1_MSR_BITS; vcpu->arch.shregs.msr |= flags; } return RESUME_HOST; diff --git a/arch/powerpc/kvm/book3s_pr.c b/arch/powerpc/kvm/book3s_pr.c index cc65af8fe6f7..ce4fcf76e53e 100644 --- a/arch/powerpc/kvm/book3s_pr.c +++ b/arch/powerpc/kvm/book3s_pr.c @@ -90,7 +90,43 @@ static void kvmppc_fixup_split_real(struct kvm_vcpu *vcpu) kvmppc_set_pc(vcpu, pc | SPLIT_HACK_OFFS); } -void kvmppc_unfixup_split_real(struct kvm_vcpu *vcpu); +static void kvmppc_unfixup_split_real(struct kvm_vcpu *vcpu) +{ + if (vcpu->arch.hflags & BOOK3S_HFLAG_SPLIT_HACK) { + ulong pc = kvmppc_get_pc(vcpu); + ulong lr = kvmppc_get_lr(vcpu); + if ((pc & SPLIT_HACK_MASK) == SPLIT_HACK_OFFS) + kvmppc_set_pc(vcpu, pc & ~SPLIT_HACK_MASK); + if ((lr & SPLIT_HACK_MASK) == SPLIT_HACK_OFFS) + kvmppc_set_lr(vcpu, lr & ~SPLIT_HACK_MASK); + vcpu->arch.hflags &= ~BOOK3S_HFLAG_SPLIT_HACK; + } +} + +static void kvmppc_inject_interrupt_pr(struct kvm_vcpu *vcpu, int vec, u64 srr1_flags) +{ + unsigned long msr, pc, new_msr, new_pc; + + kvmppc_unfixup_split_real(vcpu); + + msr = kvmppc_get_msr(vcpu); + pc = kvmppc_get_pc(vcpu); + new_msr = vcpu->arch.intr_msr; + new_pc = to_book3s(vcpu)->hior + vec; + +#ifdef CONFIG_PPC_BOOK3S_64 + /* If transactional, change to suspend mode on IRQ delivery */ + if (MSR_TM_TRANSACTIONAL(msr)) + new_msr |= MSR_TS_S; + else + new_msr |= msr & MSR_TS_MASK; +#endif + + kvmppc_set_srr0(vcpu, pc); + kvmppc_set_srr1(vcpu, (msr & SRR1_MSR_BITS) | srr1_flags); + kvmppc_set_pc(vcpu, new_pc); + kvmppc_set_msr(vcpu, new_msr); +} static void kvmppc_core_vcpu_load_pr(struct kvm_vcpu *vcpu, int cpu) { @@ -1761,6 +1797,7 @@ static struct kvm_vcpu *kvmppc_core_vcpu_create_pr(struct kvm *kvm, #else /* default to book3s_32 (750) */ vcpu->arch.pvr = 0x84202; + vcpu->arch.intr_msr = 0; #endif kvmppc_set_pvr_pr(vcpu, vcpu->arch.pvr); vcpu->arch.slb_nr = 64; @@ -2058,6 +2095,7 @@ static struct kvmppc_ops kvm_ops_pr = { .set_one_reg = kvmppc_set_one_reg_pr, .vcpu_load = kvmppc_core_vcpu_load_pr, .vcpu_put = kvmppc_core_vcpu_put_pr, + .inject_interrupt = kvmppc_inject_interrupt_pr, .set_msr = kvmppc_set_msr_pr, .vcpu_run = kvmppc_vcpu_run_pr, .vcpu_create = kvmppc_core_vcpu_create_pr, diff --git a/arch/powerpc/kvm/book3s_xive.c b/arch/powerpc/kvm/book3s_xive.c index a3f9c665bb5b..66858b7d3c6b 100644 --- a/arch/powerpc/kvm/book3s_xive.c +++ b/arch/powerpc/kvm/book3s_xive.c @@ -1211,6 +1211,45 @@ void kvmppc_xive_cleanup_vcpu(struct kvm_vcpu *vcpu) vcpu->arch.xive_vcpu = NULL; } +static bool kvmppc_xive_vcpu_id_valid(struct kvmppc_xive *xive, u32 cpu) +{ + /* We have a block of xive->nr_servers VPs. We just need to check + * raw vCPU ids are below the expected limit for this guest's + * core stride ; kvmppc_pack_vcpu_id() will pack them down to an + * index that can be safely used to compute a VP id that belongs + * to the VP block. + */ + return cpu < xive->nr_servers * xive->kvm->arch.emul_smt_mode; +} + +int kvmppc_xive_compute_vp_id(struct kvmppc_xive *xive, u32 cpu, u32 *vp) +{ + u32 vp_id; + + if (!kvmppc_xive_vcpu_id_valid(xive, cpu)) { + pr_devel("Out of bounds !\n"); + return -EINVAL; + } + + if (xive->vp_base == XIVE_INVALID_VP) { + xive->vp_base = xive_native_alloc_vp_block(xive->nr_servers); + pr_devel("VP_Base=%x nr_servers=%d\n", xive->vp_base, xive->nr_servers); + + if (xive->vp_base == XIVE_INVALID_VP) + return -ENOSPC; + } + + vp_id = kvmppc_xive_vp(xive, cpu); + if (kvmppc_xive_vp_in_use(xive->kvm, vp_id)) { + pr_devel("Duplicate !\n"); + return -EEXIST; + } + + *vp = vp_id; + + return 0; +} + int kvmppc_xive_connect_vcpu(struct kvm_device *dev, struct kvm_vcpu *vcpu, u32 cpu) { @@ -1229,20 +1268,13 @@ int kvmppc_xive_connect_vcpu(struct kvm_device *dev, return -EPERM; if (vcpu->arch.irq_type != KVMPPC_IRQ_DEFAULT) return -EBUSY; - if (cpu >= (KVM_MAX_VCPUS * vcpu->kvm->arch.emul_smt_mode)) { - pr_devel("Out of bounds !\n"); - return -EINVAL; - } /* We need to synchronize with queue provisioning */ mutex_lock(&xive->lock); - vp_id = kvmppc_xive_vp(xive, cpu); - if (kvmppc_xive_vp_in_use(xive->kvm, vp_id)) { - pr_devel("Duplicate !\n"); - r = -EEXIST; + r = kvmppc_xive_compute_vp_id(xive, cpu, &vp_id); + if (r) goto bail; - } xc = kzalloc(sizeof(*xc), GFP_KERNEL); if (!xc) { @@ -1834,6 +1866,43 @@ int kvmppc_xive_set_irq(struct kvm *kvm, int irq_source_id, u32 irq, int level, return 0; } +int kvmppc_xive_set_nr_servers(struct kvmppc_xive *xive, u64 addr) +{ + u32 __user *ubufp = (u32 __user *) addr; + u32 nr_servers; + int rc = 0; + + if (get_user(nr_servers, ubufp)) + return -EFAULT; + + pr_devel("%s nr_servers=%u\n", __func__, nr_servers); + + if (!nr_servers || nr_servers > KVM_MAX_VCPU_ID) + return -EINVAL; + + mutex_lock(&xive->lock); + if (xive->vp_base != XIVE_INVALID_VP) + /* The VP block is allocated once and freed when the device + * is released. Better not allow to change its size since its + * used by connect_vcpu to validate vCPU ids are valid (eg, + * setting it back to a higher value could allow connect_vcpu + * to come up with a VP id that goes beyond the VP block, which + * is likely to cause a crash in OPAL). + */ + rc = -EBUSY; + else if (nr_servers > KVM_MAX_VCPUS) + /* We don't need more servers. Higher vCPU ids get packed + * down below KVM_MAX_VCPUS by kvmppc_pack_vcpu_id(). + */ + xive->nr_servers = KVM_MAX_VCPUS; + else + xive->nr_servers = nr_servers; + + mutex_unlock(&xive->lock); + + return rc; +} + static int xive_set_attr(struct kvm_device *dev, struct kvm_device_attr *attr) { struct kvmppc_xive *xive = dev->private; @@ -1842,6 +1911,11 @@ static int xive_set_attr(struct kvm_device *dev, struct kvm_device_attr *attr) switch (attr->group) { case KVM_DEV_XICS_GRP_SOURCES: return xive_set_source(xive, attr->attr, attr->addr); + case KVM_DEV_XICS_GRP_CTRL: + switch (attr->attr) { + case KVM_DEV_XICS_NR_SERVERS: + return kvmppc_xive_set_nr_servers(xive, attr->addr); + } } return -ENXIO; } @@ -1867,6 +1941,11 @@ static int xive_has_attr(struct kvm_device *dev, struct kvm_device_attr *attr) attr->attr < KVMPPC_XICS_NR_IRQS) return 0; break; + case KVM_DEV_XICS_GRP_CTRL: + switch (attr->attr) { + case KVM_DEV_XICS_NR_SERVERS: + return 0; + } } return -ENXIO; } @@ -2001,10 +2080,13 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type) { struct kvmppc_xive *xive; struct kvm *kvm = dev->kvm; - int ret = 0; pr_devel("Creating xive for partition\n"); + /* Already there ? */ + if (kvm->arch.xive) + return -EEXIST; + xive = kvmppc_xive_get_device(kvm, type); if (!xive) return -ENOMEM; @@ -2014,12 +2096,6 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type) xive->kvm = kvm; mutex_init(&xive->lock); - /* Already there ? */ - if (kvm->arch.xive) - ret = -EEXIST; - else - kvm->arch.xive = xive; - /* We use the default queue size set by the host */ xive->q_order = xive_native_default_eq_shift(); if (xive->q_order < PAGE_SHIFT) @@ -2027,18 +2103,16 @@ static int kvmppc_xive_create(struct kvm_device *dev, u32 type) else xive->q_page_order = xive->q_order - PAGE_SHIFT; - /* Allocate a bunch of VPs */ - xive->vp_base = xive_native_alloc_vp_block(KVM_MAX_VCPUS); - pr_devel("VP_Base=%x\n", xive->vp_base); - - if (xive->vp_base == XIVE_INVALID_VP) - ret = -ENOMEM; + /* VP allocation is delayed to the first call to connect_vcpu */ + xive->vp_base = XIVE_INVALID_VP; + /* KVM_MAX_VCPUS limits the number of VMs to roughly 64 per sockets + * on a POWER9 system. + */ + xive->nr_servers = KVM_MAX_VCPUS; xive->single_escalation = xive_native_has_single_escalation(); - if (ret) - return ret; - + kvm->arch.xive = xive; return 0; } @@ -2108,9 +2182,9 @@ static int xive_debug_show(struct seq_file *m, void *private) if (!xc) continue; - seq_printf(m, "cpu server %#x CPPR:%#x HWCPPR:%#x" + seq_printf(m, "cpu server %#x VP:%#x CPPR:%#x HWCPPR:%#x" " MFRR:%#x PEND:%#x h_xirr: R=%lld V=%lld\n", - xc->server_num, xc->cppr, xc->hw_cppr, + xc->server_num, xc->vp_id, xc->cppr, xc->hw_cppr, xc->mfrr, xc->pending, xc->stat_rm_h_xirr, xc->stat_vm_h_xirr); diff --git a/arch/powerpc/kvm/book3s_xive.h b/arch/powerpc/kvm/book3s_xive.h index fe3ed50e0818..382e3a56e789 100644 --- a/arch/powerpc/kvm/book3s_xive.h +++ b/arch/powerpc/kvm/book3s_xive.h @@ -135,6 +135,9 @@ struct kvmppc_xive { /* Flags */ u8 single_escalation; + /* Number of entries in the VP block */ + u32 nr_servers; + struct kvmppc_xive_ops *ops; struct address_space *mapping; struct mutex mapping_lock; @@ -296,6 +299,8 @@ int kvmppc_xive_attach_escalation(struct kvm_vcpu *vcpu, u8 prio, struct kvmppc_xive *kvmppc_xive_get_device(struct kvm *kvm, u32 type); void xive_cleanup_single_escalation(struct kvm_vcpu *vcpu, struct kvmppc_xive_vcpu *xc, int irq); +int kvmppc_xive_compute_vp_id(struct kvmppc_xive *xive, u32 cpu, u32 *vp); +int kvmppc_xive_set_nr_servers(struct kvmppc_xive *xive, u64 addr); #endif /* CONFIG_KVM_XICS */ #endif /* _KVM_PPC_BOOK3S_XICS_H */ diff --git a/arch/powerpc/kvm/book3s_xive_native.c b/arch/powerpc/kvm/book3s_xive_native.c index 78b906ffa0d2..d83adb1e1490 100644 --- a/arch/powerpc/kvm/book3s_xive_native.c +++ b/arch/powerpc/kvm/book3s_xive_native.c @@ -50,6 +50,24 @@ static void kvmppc_xive_native_cleanup_queue(struct kvm_vcpu *vcpu, int prio) } } +static int kvmppc_xive_native_configure_queue(u32 vp_id, struct xive_q *q, + u8 prio, __be32 *qpage, + u32 order, bool can_escalate) +{ + int rc; + __be32 *qpage_prev = q->qpage; + + rc = xive_native_configure_queue(vp_id, q, prio, qpage, order, + can_escalate); + if (rc) + return rc; + + if (qpage_prev) + put_page(virt_to_page(qpage_prev)); + + return rc; +} + void kvmppc_xive_native_cleanup_vcpu(struct kvm_vcpu *vcpu) { struct kvmppc_xive_vcpu *xc = vcpu->arch.xive_vcpu; @@ -118,19 +136,12 @@ int kvmppc_xive_native_connect_vcpu(struct kvm_device *dev, return -EPERM; if (vcpu->arch.irq_type != KVMPPC_IRQ_DEFAULT) return -EBUSY; - if (server_num >= (KVM_MAX_VCPUS * vcpu->kvm->arch.emul_smt_mode)) { - pr_devel("Out of bounds !\n"); - return -EINVAL; - } mutex_lock(&xive->lock); - vp_id = kvmppc_xive_vp(xive, server_num); - if (kvmppc_xive_vp_in_use(xive->kvm, vp_id)) { - pr_devel("Duplicate !\n"); - rc = -EEXIST; + rc = kvmppc_xive_compute_vp_id(xive, server_num, &vp_id); + if (rc) goto bail; - } xc = kzalloc(sizeof(*xc), GFP_KERNEL); if (!xc) { @@ -582,19 +593,14 @@ static int kvmppc_xive_native_set_queue_config(struct kvmppc_xive *xive, q->guest_qaddr = 0; q->guest_qshift = 0; - rc = xive_native_configure_queue(xc->vp_id, q, priority, - NULL, 0, true); + rc = kvmppc_xive_native_configure_queue(xc->vp_id, q, priority, + NULL, 0, true); if (rc) { pr_err("Failed to reset queue %d for VCPU %d: %d\n", priority, xc->server_num, rc); return rc; } - if (q->qpage) { - put_page(virt_to_page(q->qpage)); - q->qpage = NULL; - } - return 0; } @@ -624,12 +630,6 @@ static int kvmppc_xive_native_set_queue_config(struct kvmppc_xive *xive, srcu_idx = srcu_read_lock(&kvm->srcu); gfn = gpa_to_gfn(kvm_eq.qaddr); - page = gfn_to_page(kvm, gfn); - if (is_error_page(page)) { - srcu_read_unlock(&kvm->srcu, srcu_idx); - pr_err("Couldn't get queue page %llx!\n", kvm_eq.qaddr); - return -EINVAL; - } page_size = kvm_host_page_size(kvm, gfn); if (1ull << kvm_eq.qshift > page_size) { @@ -638,6 +638,13 @@ static int kvmppc_xive_native_set_queue_config(struct kvmppc_xive *xive, return -EINVAL; } + page = gfn_to_page(kvm, gfn); + if (is_error_page(page)) { + srcu_read_unlock(&kvm->srcu, srcu_idx); + pr_err("Couldn't get queue page %llx!\n", kvm_eq.qaddr); + return -EINVAL; + } + qaddr = page_to_virt(page) + (kvm_eq.qaddr & ~PAGE_MASK); srcu_read_unlock(&kvm->srcu, srcu_idx); @@ -653,8 +660,8 @@ static int kvmppc_xive_native_set_queue_config(struct kvmppc_xive *xive, * OPAL level because the use of END ESBs is not supported by * Linux. */ - rc = xive_native_configure_queue(xc->vp_id, q, priority, - (__be32 *) qaddr, kvm_eq.qshift, true); + rc = kvmppc_xive_native_configure_queue(xc->vp_id, q, priority, + (__be32 *) qaddr, kvm_eq.qshift, true); if (rc) { pr_err("Failed to configure queue %d for VCPU %d: %d\n", priority, xc->server_num, rc); @@ -928,6 +935,8 @@ static int kvmppc_xive_native_set_attr(struct kvm_device *dev, return kvmppc_xive_reset(xive); case KVM_DEV_XIVE_EQ_SYNC: return kvmppc_xive_native_eq_sync(xive); + case KVM_DEV_XIVE_NR_SERVERS: + return kvmppc_xive_set_nr_servers(xive, attr->addr); } break; case KVM_DEV_XIVE_GRP_SOURCE: @@ -967,6 +976,7 @@ static int kvmppc_xive_native_has_attr(struct kvm_device *dev, switch (attr->attr) { case KVM_DEV_XIVE_RESET: case KVM_DEV_XIVE_EQ_SYNC: + case KVM_DEV_XIVE_NR_SERVERS: return 0; } break; @@ -1067,7 +1077,6 @@ static int kvmppc_xive_native_create(struct kvm_device *dev, u32 type) { struct kvmppc_xive *xive; struct kvm *kvm = dev->kvm; - int ret = 0; pr_devel("Creating xive native device\n"); @@ -1081,27 +1090,20 @@ static int kvmppc_xive_native_create(struct kvm_device *dev, u32 type) dev->private = xive; xive->dev = dev; xive->kvm = kvm; - kvm->arch.xive = xive; mutex_init(&xive->mapping_lock); mutex_init(&xive->lock); - /* - * Allocate a bunch of VPs. KVM_MAX_VCPUS is a large value for - * a default. Getting the max number of CPUs the VM was - * configured with would improve our usage of the XIVE VP space. + /* VP allocation is delayed to the first call to connect_vcpu */ + xive->vp_base = XIVE_INVALID_VP; + /* KVM_MAX_VCPUS limits the number of VMs to roughly 64 per sockets + * on a POWER9 system. */ - xive->vp_base = xive_native_alloc_vp_block(KVM_MAX_VCPUS); - pr_devel("VP_Base=%x\n", xive->vp_base); - - if (xive->vp_base == XIVE_INVALID_VP) - ret = -ENXIO; + xive->nr_servers = KVM_MAX_VCPUS; xive->single_escalation = xive_native_has_single_escalation(); xive->ops = &kvmppc_xive_native_ops; - if (ret) - return ret; - + kvm->arch.xive = xive; return 0; } @@ -1204,8 +1206,8 @@ static int xive_native_debug_show(struct seq_file *m, void *private) if (!xc) continue; - seq_printf(m, "cpu server %#x NSR=%02x CPPR=%02x IBP=%02x PIPR=%02x w01=%016llx w2=%08x\n", - xc->server_num, + seq_printf(m, "cpu server %#x VP=%#x NSR=%02x CPPR=%02x IBP=%02x PIPR=%02x w01=%016llx w2=%08x\n", + xc->server_num, xc->vp_id, vcpu->arch.xive_saved_state.nsr, vcpu->arch.xive_saved_state.cppr, vcpu->arch.xive_saved_state.ipb, diff --git a/arch/powerpc/kvm/e500_mmu_host.c b/arch/powerpc/kvm/e500_mmu_host.c index 321db0fdb9db..425d13806645 100644 --- a/arch/powerpc/kvm/e500_mmu_host.c +++ b/arch/powerpc/kvm/e500_mmu_host.c @@ -355,9 +355,9 @@ static inline int kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500, if (tlbsel == 1) { struct vm_area_struct *vma; - down_read(¤t->mm->mmap_sem); + down_read(&kvm->mm->mmap_sem); - vma = find_vma(current->mm, hva); + vma = find_vma(kvm->mm, hva); if (vma && hva >= vma->vm_start && (vma->vm_flags & VM_PFNMAP)) { /* @@ -441,7 +441,7 @@ static inline int kvmppc_e500_shadow_map(struct kvmppc_vcpu_e500 *vcpu_e500, tsize = max(BOOK3E_PAGESZ_4K, tsize & ~1); } - up_read(¤t->mm->mmap_sem); + up_read(&kvm->mm->mmap_sem); } if (likely(!pfnmap)) { diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index 3a77bb643452..9e085e931d74 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -522,6 +522,8 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) case KVM_CAP_IMMEDIATE_EXIT: r = 1; break; + case KVM_CAP_PPC_GUEST_DEBUG_SSTEP: + /* fall through */ case KVM_CAP_PPC_PAIRED_SINGLES: case KVM_CAP_PPC_OSI: case KVM_CAP_PPC_GET_PVINFO: diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c index be941d382c8d..c95b7fe9f298 100644 --- a/arch/powerpc/mm/mem.c +++ b/arch/powerpc/mm/mem.c @@ -31,6 +31,7 @@ #include <linux/slab.h> #include <linux/vmalloc.h> #include <linux/memremap.h> +#include <linux/dma-direct.h> #include <asm/pgalloc.h> #include <asm/prom.h> @@ -201,10 +202,10 @@ static int __init mark_nonram_nosave(void) * everything else. GFP_DMA32 page allocations automatically fall back to * ZONE_DMA. * - * By using 31-bit unconditionally, we can exploit ARCH_ZONE_DMA_BITS to - * inform the generic DMA mapping code. 32-bit only devices (if not handled - * by an IOMMU anyway) will take a first dip into ZONE_NORMAL and get - * otherwise served by ZONE_DMA. + * By using 31-bit unconditionally, we can exploit zone_dma_bits to inform the + * generic DMA mapping code. 32-bit only devices (if not handled by an IOMMU + * anyway) will take a first dip into ZONE_NORMAL and get otherwise served by + * ZONE_DMA. */ static unsigned long max_zone_pfns[MAX_NR_ZONES]; @@ -237,9 +238,18 @@ void __init paging_init(void) printk(KERN_DEBUG "Memory hole size: %ldMB\n", (long int)((top_of_ram - total_ram) >> 20)); + /* + * Allow 30-bit DMA for very limited Broadcom wifi chips on many + * powerbooks. + */ + if (IS_ENABLED(CONFIG_PPC32)) + zone_dma_bits = 30; + else + zone_dma_bits = 31; + #ifdef CONFIG_ZONE_DMA max_zone_pfns[ZONE_DMA] = min(max_low_pfn, - 1UL << (ARCH_ZONE_DMA_BITS - PAGE_SHIFT)); + 1UL << (zone_dma_bits - PAGE_SHIFT)); #endif max_zone_pfns[ZONE_NORMAL] = max_low_pfn; #ifdef CONFIG_HIGHMEM diff --git a/arch/powerpc/net/bpf_jit_comp64.c b/arch/powerpc/net/bpf_jit_comp64.c index 02a59946a78a..be3517ef0574 100644 --- a/arch/powerpc/net/bpf_jit_comp64.c +++ b/arch/powerpc/net/bpf_jit_comp64.c @@ -1142,6 +1142,19 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp) } /* + * If we have seen a tail call, we need a second pass. + * This is because bpf_jit_emit_common_epilogue() is called + * from bpf_jit_emit_tail_call() with a not yet stable ctx->seen. + */ + if (cgctx.seen & SEEN_TAILCALL) { + cgctx.idx = 0; + if (bpf_jit_build_body(fp, 0, &cgctx, addrs, false)) { + fp = org_fp; + goto out_addrs; + } + } + + /* * Pretend to build prologue, given the features we've seen. This will * update ctgtx.idx as it pretends to output instructions, then we can * calculate total size from idx. diff --git a/arch/powerpc/perf/core-book3s.c b/arch/powerpc/perf/core-book3s.c index ca92e01d0bd1..48604625ab31 100644 --- a/arch/powerpc/perf/core-book3s.c +++ b/arch/powerpc/perf/core-book3s.c @@ -96,7 +96,7 @@ static inline unsigned long perf_ip_adjust(struct pt_regs *regs) { return 0; } -static inline void perf_get_data_addr(struct pt_regs *regs, u64 *addrp) { } +static inline void perf_get_data_addr(struct perf_event *event, struct pt_regs *regs, u64 *addrp) { } static inline u32 perf_get_misc_flags(struct pt_regs *regs) { return 0; @@ -127,7 +127,7 @@ static unsigned long ebb_switch_in(bool ebb, struct cpu_hw_events *cpuhw) static inline void power_pmu_bhrb_enable(struct perf_event *event) {} static inline void power_pmu_bhrb_disable(struct perf_event *event) {} static void power_pmu_sched_task(struct perf_event_context *ctx, bool sched_in) {} -static inline void power_pmu_bhrb_read(struct cpu_hw_events *cpuhw) {} +static inline void power_pmu_bhrb_read(struct perf_event *event, struct cpu_hw_events *cpuhw) {} static void pmao_restore_workaround(bool ebb) { } #endif /* CONFIG_PPC32 */ @@ -179,7 +179,7 @@ static inline unsigned long perf_ip_adjust(struct pt_regs *regs) * pointed to by SIAR; this is indicated by the [POWER6_]MMCRA_SDSYNC, the * [POWER7P_]MMCRA_SDAR_VALID bit in MMCRA, or the SDAR_VALID bit in SIER. */ -static inline void perf_get_data_addr(struct pt_regs *regs, u64 *addrp) +static inline void perf_get_data_addr(struct perf_event *event, struct pt_regs *regs, u64 *addrp) { unsigned long mmcra = regs->dsisr; bool sdar_valid; @@ -204,8 +204,7 @@ static inline void perf_get_data_addr(struct pt_regs *regs, u64 *addrp) if (!(mmcra & MMCRA_SAMPLE_ENABLE) || sdar_valid) *addrp = mfspr(SPRN_SDAR); - if (perf_paranoid_kernel() && !capable(CAP_SYS_ADMIN) && - is_kernel_addr(mfspr(SPRN_SDAR))) + if (is_kernel_addr(mfspr(SPRN_SDAR)) && perf_allow_kernel(&event->attr) != 0) *addrp = 0; } @@ -444,7 +443,7 @@ static __u64 power_pmu_bhrb_to(u64 addr) } /* Processing BHRB entries */ -static void power_pmu_bhrb_read(struct cpu_hw_events *cpuhw) +static void power_pmu_bhrb_read(struct perf_event *event, struct cpu_hw_events *cpuhw) { u64 val; u64 addr; @@ -472,8 +471,7 @@ static void power_pmu_bhrb_read(struct cpu_hw_events *cpuhw) * exporting it to userspace (avoid exposure of regions * where we could have speculative execution) */ - if (perf_paranoid_kernel() && !capable(CAP_SYS_ADMIN) && - is_kernel_addr(addr)) + if (is_kernel_addr(addr) && perf_allow_kernel(&event->attr) != 0) continue; /* Branches are read most recent first (ie. mfbhrb 0 is @@ -2087,12 +2085,12 @@ static void record_and_restart(struct perf_event *event, unsigned long val, if (event->attr.sample_type & (PERF_SAMPLE_ADDR | PERF_SAMPLE_PHYS_ADDR)) - perf_get_data_addr(regs, &data.addr); + perf_get_data_addr(event, regs, &data.addr); if (event->attr.sample_type & PERF_SAMPLE_BRANCH_STACK) { struct cpu_hw_events *cpuhw; cpuhw = this_cpu_ptr(&cpu_hw_events); - power_pmu_bhrb_read(cpuhw); + power_pmu_bhrb_read(event, cpuhw); data.br_stack = &cpuhw->bhrb_stack; } diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c index 6bc24a47e9ef..6f300ab7f0e9 100644 --- a/arch/powerpc/platforms/powernv/eeh-powernv.c +++ b/arch/powerpc/platforms/powernv/eeh-powernv.c @@ -42,7 +42,7 @@ void pnv_pcibios_bus_add_device(struct pci_dev *pdev) { struct pci_dn *pdn = pci_get_pdn(pdev); - if (eeh_has_flag(EEH_FORCE_DISABLED)) + if (!pdn || eeh_has_flag(EEH_FORCE_DISABLED)) return; dev_dbg(&pdev->dev, "EEH: Setting up device\n"); diff --git a/arch/powerpc/platforms/powernv/smp.c b/arch/powerpc/platforms/powernv/smp.c index fbd6e6b7bbf2..13e251699346 100644 --- a/arch/powerpc/platforms/powernv/smp.c +++ b/arch/powerpc/platforms/powernv/smp.c @@ -146,20 +146,25 @@ static int pnv_smp_cpu_disable(void) return 0; } +static void pnv_flush_interrupts(void) +{ + if (cpu_has_feature(CPU_FTR_ARCH_300)) { + if (xive_enabled()) + xive_flush_interrupt(); + else + icp_opal_flush_interrupt(); + } else { + icp_native_flush_interrupt(); + } +} + static void pnv_smp_cpu_kill_self(void) { + unsigned long srr1, unexpected_mask, wmask; unsigned int cpu; - unsigned long srr1, wmask; u64 lpcr_val; /* Standard hot unplug procedure */ - /* - * This hard disables local interurpts, ensuring we have no lazy - * irqs pending. - */ - WARN_ON(irqs_disabled()); - hard_irq_disable(); - WARN_ON(lazy_irq_pending()); idle_task_exit(); current->active_mm = NULL; /* for sanity */ @@ -173,6 +178,27 @@ static void pnv_smp_cpu_kill_self(void) wmask = SRR1_WAKEMASK_P8; /* + * This turns the irq soft-disabled state we're called with, into a + * hard-disabled state with pending irq_happened interrupts cleared. + * + * PACA_IRQ_DEC - Decrementer should be ignored. + * PACA_IRQ_HMI - Can be ignored, processing is done in real mode. + * PACA_IRQ_DBELL, EE, PMI - Unexpected. + */ + hard_irq_disable(); + if (generic_check_cpu_restart(cpu)) + goto out; + + unexpected_mask = ~(PACA_IRQ_DEC | PACA_IRQ_HMI | PACA_IRQ_HARD_DIS); + if (local_paca->irq_happened & unexpected_mask) { + if (local_paca->irq_happened & PACA_IRQ_EE) + pnv_flush_interrupts(); + DBG("CPU%d Unexpected exit while offline irq_happened=%lx!\n", + cpu, local_paca->irq_happened); + } + local_paca->irq_happened = PACA_IRQ_HARD_DIS; + + /* * We don't want to take decrementer interrupts while we are * offline, so clear LPCR:PECE1. We keep PECE2 (and * LPCR_PECE_HVEE on P9) enabled so as to let IPIs in. @@ -197,6 +223,7 @@ static void pnv_smp_cpu_kill_self(void) srr1 = pnv_cpu_offline(cpu); + WARN_ON_ONCE(!irqs_disabled()); WARN_ON(lazy_irq_pending()); /* @@ -212,13 +239,7 @@ static void pnv_smp_cpu_kill_self(void) */ if (((srr1 & wmask) == SRR1_WAKEEE) || ((srr1 & wmask) == SRR1_WAKEHVI)) { - if (cpu_has_feature(CPU_FTR_ARCH_300)) { - if (xive_enabled()) - xive_flush_interrupt(); - else - icp_opal_flush_interrupt(); - } else - icp_native_flush_interrupt(); + pnv_flush_interrupts(); } else if ((srr1 & wmask) == SRR1_WAKEHDBELL) { unsigned long msg = PPC_DBELL_TYPE(PPC_DBELL_SERVER); asm volatile(PPC_MSGCLR(%0) : : "r" (msg)); @@ -266,7 +287,7 @@ static void pnv_smp_cpu_kill_self(void) */ lpcr_val = mfspr(SPRN_LPCR) | (u64)LPCR_PECE1; pnv_program_cpu_hotplug_lpcr(cpu, lpcr_val); - +out: DBG("CPU%d coming online...\n", cpu); } |