summaryrefslogtreecommitdiff
path: root/arch/s390
diff options
context:
space:
mode:
Diffstat (limited to 'arch/s390')
-rw-r--r--arch/s390/Kconfig2
-rw-r--r--arch/s390/configs/debug_defconfig2
-rw-r--r--arch/s390/configs/defconfig2
-rw-r--r--arch/s390/crypto/Makefile1
-rw-r--r--arch/s390/crypto/hmac_s390.c12
-rw-r--r--arch/s390/crypto/paes_s390.c2
-rw-r--r--arch/s390/crypto/phmac_s390.c1048
-rw-r--r--arch/s390/crypto/sha.h3
-rw-r--r--arch/s390/crypto/sha3_256_s390.c24
-rw-r--r--arch/s390/crypto/sha3_512_s390.c25
-rw-r--r--arch/s390/include/asm/cpacf.h4
-rw-r--r--arch/s390/include/asm/entry-common.h10
-rw-r--r--arch/s390/include/asm/kvm_host.h3
-rw-r--r--arch/s390/include/asm/percpu.h5
-rw-r--r--arch/s390/kvm/kvm-s390.c51
-rw-r--r--arch/s390/kvm/vsie.c17
-rw-r--r--arch/s390/mm/dump_pagetables.c2
-rw-r--r--arch/s390/net/bpf_jit.h55
-rw-r--r--arch/s390/net/bpf_jit_comp.c113
19 files changed, 1230 insertions, 151 deletions
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index 25a773e6596e..f0c0469e553d 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -131,6 +131,7 @@ config S390
select ARCH_INLINE_WRITE_UNLOCK_IRQ
select ARCH_INLINE_WRITE_UNLOCK_IRQRESTORE
select ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE
+ select ARCH_MODULE_NEEDS_WEAK_PER_CPU
select ARCH_STACKWALK
select ARCH_SUPPORTS_ATOMIC_RMW
select ARCH_SUPPORTS_DEBUG_PAGEALLOC
@@ -198,7 +199,6 @@ config S390
select HAVE_GUP_FAST
select HAVE_FENTRY
select HAVE_FTRACE_GRAPH_FUNC
- select HAVE_FTRACE_MCOUNT_RECORD
select HAVE_FUNCTION_ARG_ACCESS_API
select HAVE_FUNCTION_ERROR_INJECTION
select HAVE_FUNCTION_GRAPH_FREGS
diff --git a/arch/s390/configs/debug_defconfig b/arch/s390/configs/debug_defconfig
index a7db7ed28720..6b33429f1c4d 100644
--- a/arch/s390/configs/debug_defconfig
+++ b/arch/s390/configs/debug_defconfig
@@ -248,7 +248,6 @@ CONFIG_NETFILTER_XT_MATCH_CONNLIMIT=m
CONFIG_NETFILTER_XT_MATCH_CONNMARK=m
CONFIG_NETFILTER_XT_MATCH_CONNTRACK=m
CONFIG_NETFILTER_XT_MATCH_CPU=m
-CONFIG_NETFILTER_XT_MATCH_DCCP=m
CONFIG_NETFILTER_XT_MATCH_DEVGROUP=m
CONFIG_NETFILTER_XT_MATCH_DSCP=m
CONFIG_NETFILTER_XT_MATCH_ESP=m
@@ -817,6 +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 0217ed5616bf..b75eb2775850 100644
--- a/arch/s390/configs/defconfig
+++ b/arch/s390/configs/defconfig
@@ -239,7 +239,6 @@ CONFIG_NETFILTER_XT_MATCH_CONNLIMIT=m
CONFIG_NETFILTER_XT_MATCH_CONNMARK=m
CONFIG_NETFILTER_XT_MATCH_CONNTRACK=m
CONFIG_NETFILTER_XT_MATCH_CPU=m
-CONFIG_NETFILTER_XT_MATCH_DCCP=m
CONFIG_NETFILTER_XT_MATCH_DEVGROUP=m
CONFIG_NETFILTER_XT_MATCH_DSCP=m
CONFIG_NETFILTER_XT_MATCH_ESP=m
@@ -804,6 +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 1e5a1038d491..998f4b656b18 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 93a1098d9f8d..58444da9b004 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 8a340c16acb4..a624a43a2b54 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 000000000000..7ecfdc4fba2d
--- /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, &param, in, inlen);
+
+ memcpy(digest, &param, 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 d757ccbce2b4..cadb4b13622a 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 4a7731ac6bcd..03bb4f4bab70 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 018f02fff444..a5c9690eecb1 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 54cb97603ec0..4bc5317fbb12 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/s390/include/asm/entry-common.h b/arch/s390/include/asm/entry-common.h
index 35555c944630..979af986a8fe 100644
--- a/arch/s390/include/asm/entry-common.h
+++ b/arch/s390/include/asm/entry-common.h
@@ -59,4 +59,14 @@ static inline void arch_exit_to_user_mode_prepare(struct pt_regs *regs,
#define arch_exit_to_user_mode_prepare arch_exit_to_user_mode_prepare
+static __always_inline bool arch_in_rcu_eqs(void)
+{
+ if (IS_ENABLED(CONFIG_KVM))
+ return current->flags & PF_VCPU;
+
+ return false;
+}
+
+#define arch_in_rcu_eqs arch_in_rcu_eqs
+
#endif
diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index cb89e54ada25..f870d09515cc 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -716,6 +716,9 @@ extern char sie_exit;
bool kvm_s390_pv_is_protected(struct kvm *kvm);
bool kvm_s390_pv_cpu_is_protected(struct kvm_vcpu *vcpu);
+extern int kvm_s390_enter_exit_sie(struct kvm_s390_sie_block *scb,
+ u64 *gprs, unsigned long gasce);
+
extern int kvm_s390_gisc_register(struct kvm *kvm, u32 gisc);
extern int kvm_s390_gisc_unregister(struct kvm *kvm, u32 gisc);
diff --git a/arch/s390/include/asm/percpu.h b/arch/s390/include/asm/percpu.h
index 84f6b8357b45..96af7d964014 100644
--- a/arch/s390/include/asm/percpu.h
+++ b/arch/s390/include/asm/percpu.h
@@ -16,10 +16,9 @@
* For 64 bit module code, the module may be more than 4G above the
* per cpu area, use weak definitions to force the compiler to
* generate external references.
+ * Therefore, we have enabled CONFIG_ARCH_MODULE_NEEDS_WEAK_PER_CPU
+ * in the Kconfig.
*/
-#if defined(MODULE)
-#define ARCH_NEEDS_WEAK_PER_CPU
-#endif
/*
* We use a compare-and-swap loop since that uses less cpu cycles than
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 78c9a310efa5..bf6fa8b9ca73 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -5063,6 +5063,30 @@ static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason)
return vcpu_post_run_handle_fault(vcpu);
}
+int noinstr kvm_s390_enter_exit_sie(struct kvm_s390_sie_block *scb,
+ u64 *gprs, unsigned long gasce)
+{
+ int ret;
+
+ guest_state_enter_irqoff();
+
+ /*
+ * The guest_state_{enter,exit}_irqoff() functions inform lockdep and
+ * tracing that entry to the guest will enable host IRQs, and exit from
+ * the guest will disable host IRQs.
+ *
+ * We must not use lockdep/tracing/RCU in this critical section, so we
+ * use the low-level arch_local_irq_*() helpers to enable/disable IRQs.
+ */
+ arch_local_irq_enable();
+ ret = sie64a(scb, gprs, gasce);
+ arch_local_irq_disable();
+
+ guest_state_exit_irqoff();
+
+ return ret;
+}
+
#define PSW_INT_MASK (PSW_MASK_EXT | PSW_MASK_IO | PSW_MASK_MCHECK)
static int __vcpu_run(struct kvm_vcpu *vcpu)
{
@@ -5083,20 +5107,27 @@ static int __vcpu_run(struct kvm_vcpu *vcpu)
kvm_vcpu_srcu_read_unlock(vcpu);
/*
* As PF_VCPU will be used in fault handler, between
- * guest_enter and guest_exit should be no uaccess.
+ * guest_timing_enter_irqoff and guest_timing_exit_irqoff
+ * should be no uaccess.
*/
- local_irq_disable();
- guest_enter_irqoff();
- __disable_cpu_timer_accounting(vcpu);
- local_irq_enable();
if (kvm_s390_pv_cpu_is_protected(vcpu)) {
memcpy(sie_page->pv_grregs,
vcpu->run->s.regs.gprs,
sizeof(sie_page->pv_grregs));
}
- exit_reason = sie64a(vcpu->arch.sie_block,
- vcpu->run->s.regs.gprs,
- vcpu->arch.gmap->asce);
+
+ local_irq_disable();
+ guest_timing_enter_irqoff();
+ __disable_cpu_timer_accounting(vcpu);
+
+ exit_reason = kvm_s390_enter_exit_sie(vcpu->arch.sie_block,
+ vcpu->run->s.regs.gprs,
+ vcpu->arch.gmap->asce);
+
+ __enable_cpu_timer_accounting(vcpu);
+ guest_timing_exit_irqoff();
+ local_irq_enable();
+
if (kvm_s390_pv_cpu_is_protected(vcpu)) {
memcpy(vcpu->run->s.regs.gprs,
sie_page->pv_grregs,
@@ -5112,10 +5143,6 @@ static int __vcpu_run(struct kvm_vcpu *vcpu)
vcpu->arch.sie_block->gpsw.mask &= ~PSW_INT_MASK;
}
}
- local_irq_disable();
- __enable_cpu_timer_accounting(vcpu);
- guest_exit_irqoff();
- local_irq_enable();
kvm_vcpu_srcu_read_lock(vcpu);
rc = vcpu_post_run(vcpu, exit_reason);
diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c
index 13a9661d2b28..347268f89f2f 100644
--- a/arch/s390/kvm/vsie.c
+++ b/arch/s390/kvm/vsie.c
@@ -1170,10 +1170,6 @@ static int do_vsie_run(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page)
vcpu->arch.sie_block->fpf & FPF_BPBC)
set_thread_flag(TIF_ISOLATE_BP_GUEST);
- local_irq_disable();
- guest_enter_irqoff();
- local_irq_enable();
-
/*
* Simulate a SIE entry of the VCPU (see sie64a), so VCPU blocking
* and VCPU requests also hinder the vSIE from running and lead
@@ -1183,15 +1179,16 @@ static int do_vsie_run(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page)
vcpu->arch.sie_block->prog0c |= PROG_IN_SIE;
current->thread.gmap_int_code = 0;
barrier();
- if (!kvm_s390_vcpu_sie_inhibited(vcpu))
- rc = sie64a(scb_s, vcpu->run->s.regs.gprs, vsie_page->gmap->asce);
+ if (!kvm_s390_vcpu_sie_inhibited(vcpu)) {
+ local_irq_disable();
+ guest_timing_enter_irqoff();
+ rc = kvm_s390_enter_exit_sie(scb_s, vcpu->run->s.regs.gprs, vsie_page->gmap->asce);
+ guest_timing_exit_irqoff();
+ local_irq_enable();
+ }
barrier();
vcpu->arch.sie_block->prog0c &= ~PROG_IN_SIE;
- local_irq_disable();
- guest_exit_irqoff();
- local_irq_enable();
-
/* restore guest state for bp isolation override */
if (!guest_bp_isolation)
clear_thread_flag(TIF_ISOLATE_BP_GUEST);
diff --git a/arch/s390/mm/dump_pagetables.c b/arch/s390/mm/dump_pagetables.c
index ac604b176660..9af2aae0a515 100644
--- a/arch/s390/mm/dump_pagetables.c
+++ b/arch/s390/mm/dump_pagetables.c
@@ -247,11 +247,9 @@ static int ptdump_show(struct seq_file *m, void *v)
.marker = markers,
};
- get_online_mems();
mutex_lock(&cpa_mutex);
ptdump_walk_pgd(&st.ptdump, &init_mm, NULL);
mutex_unlock(&cpa_mutex);
- put_online_mems();
return 0;
}
DEFINE_SHOW_ATTRIBUTE(ptdump);
diff --git a/arch/s390/net/bpf_jit.h b/arch/s390/net/bpf_jit.h
deleted file mode 100644
index 615e6da71374..000000000000
--- a/arch/s390/net/bpf_jit.h
+++ /dev/null
@@ -1,55 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/*
- * BPF Jit compiler defines
- *
- * Copyright IBM Corp. 2012,2015
- *
- * Author(s): Martin Schwidefsky <schwidefsky@de.ibm.com>
- * Michael Holzheu <holzheu@linux.vnet.ibm.com>
- */
-
-#ifndef __ARCH_S390_NET_BPF_JIT_H
-#define __ARCH_S390_NET_BPF_JIT_H
-
-#ifndef __ASSEMBLER__
-
-#include <linux/filter.h>
-#include <linux/types.h>
-
-#endif /* __ASSEMBLER__ */
-
-/*
- * Stackframe layout (packed stack):
- *
- * ^ high
- * +---------------+ |
- * | old backchain | |
- * +---------------+ |
- * | r15 - r6 | |
- * +---------------+ |
- * | 4 byte align | |
- * | tail_call_cnt | |
- * BFP -> +===============+ |
- * | | |
- * | BPF stack | |
- * | | |
- * R15+160 -> +---------------+ |
- * | new backchain | |
- * R15+152 -> +---------------+ |
- * | + 152 byte SA | |
- * R15 -> +---------------+ + low
- *
- * We get 160 bytes stack space from calling function, but only use
- * 12 * 8 byte for old backchain, r15..r6, and tail_call_cnt.
- *
- * The stack size used by the BPF program ("BPF stack" above) is passed
- * via "aux->stack_depth".
- */
-#define STK_SPACE_ADD (160)
-#define STK_160_UNUSED (160 - 12 * 8)
-#define STK_OFF (STK_SPACE_ADD - STK_160_UNUSED)
-
-#define STK_OFF_R6 (160 - 11 * 8) /* Offset of r6 on stack */
-#define STK_OFF_TCCNT (160 - 12 * 8) /* Offset of tail_call_cnt on stack */
-
-#endif /* __ARCH_S390_NET_BPF_JIT_H */
diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index 0c9a35782c83..bb17efe29d65 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -32,7 +32,6 @@
#include <asm/set_memory.h>
#include <asm/text-patching.h>
#include <asm/unwind.h>
-#include "bpf_jit.h"
struct bpf_jit {
u32 seen; /* Flags to remember seen eBPF instructions */
@@ -54,6 +53,7 @@ struct bpf_jit {
int prologue_plt; /* Start of prologue hotpatch PLT */
int kern_arena; /* Pool offset of kernel arena address */
u64 user_arena; /* User arena address */
+ u32 frame_off; /* Offset of struct bpf_prog from %r15 */
};
#define SEEN_MEM BIT(0) /* use mem[] for temporary storage */
@@ -426,11 +426,25 @@ static void jit_fill_hole(void *area, unsigned int size)
}
/*
+ * Caller-allocated part of the frame.
+ * Thanks to packed stack, its otherwise unused initial part can be used for
+ * the BPF stack and for the next frame.
+ */
+struct prog_frame {
+ u64 unused[8];
+ /* BPF stack starts here and grows towards 0 */
+ u32 tail_call_cnt;
+ u32 pad;
+ u64 r6[10]; /* r6 - r15 */
+ u64 backchain;
+} __packed;
+
+/*
* Save registers from "rs" (register start) to "re" (register end) on stack
*/
static void save_regs(struct bpf_jit *jit, u32 rs, u32 re)
{
- u32 off = STK_OFF_R6 + (rs - 6) * 8;
+ u32 off = offsetof(struct prog_frame, r6) + (rs - 6) * 8;
if (rs == re)
/* stg %rs,off(%r15) */
@@ -443,12 +457,9 @@ static void save_regs(struct bpf_jit *jit, u32 rs, u32 re)
/*
* Restore registers from "rs" (register start) to "re" (register end) on stack
*/
-static void restore_regs(struct bpf_jit *jit, u32 rs, u32 re, u32 stack_depth)
+static void restore_regs(struct bpf_jit *jit, u32 rs, u32 re)
{
- u32 off = STK_OFF_R6 + (rs - 6) * 8;
-
- if (jit->seen & SEEN_STACK)
- off += STK_OFF + stack_depth;
+ u32 off = jit->frame_off + offsetof(struct prog_frame, r6) + (rs - 6) * 8;
if (rs == re)
/* lg %rs,off(%r15) */
@@ -492,8 +503,7 @@ static int get_end(u16 seen_regs, int start)
* Save and restore clobbered registers (6-15) on stack.
* We save/restore registers in chunks with gap >= 2 registers.
*/
-static void save_restore_regs(struct bpf_jit *jit, int op, u32 stack_depth,
- u16 extra_regs)
+static void save_restore_regs(struct bpf_jit *jit, int op, u16 extra_regs)
{
u16 seen_regs = jit->seen_regs | extra_regs;
const int last = 15, save_restore_size = 6;
@@ -516,7 +526,7 @@ static void save_restore_regs(struct bpf_jit *jit, int op, u32 stack_depth,
if (op == REGS_SAVE)
save_regs(jit, rs, re);
else
- restore_regs(jit, rs, re, stack_depth);
+ restore_regs(jit, rs, re);
re++;
} while (re <= last);
}
@@ -581,11 +591,12 @@ static void bpf_jit_plt(struct bpf_plt *plt, void *ret, void *target)
* Emit function prologue
*
* Save registers and create stack frame if necessary.
- * See stack frame layout description in "bpf_jit.h"!
+ * Stack frame layout is described by struct prog_frame.
*/
-static void bpf_jit_prologue(struct bpf_jit *jit, struct bpf_prog *fp,
- u32 stack_depth)
+static void bpf_jit_prologue(struct bpf_jit *jit, struct bpf_prog *fp)
{
+ BUILD_BUG_ON(sizeof(struct prog_frame) != STACK_FRAME_OVERHEAD);
+
/* No-op for hotpatching */
/* brcl 0,prologue_plt */
EMIT6_PCREL_RILC(0xc0040000, 0, jit->prologue_plt);
@@ -593,8 +604,9 @@ static void bpf_jit_prologue(struct bpf_jit *jit, struct bpf_prog *fp,
if (!bpf_is_subprog(fp)) {
/* Initialize the tail call counter in the main program. */
- /* xc STK_OFF_TCCNT(4,%r15),STK_OFF_TCCNT(%r15) */
- _EMIT6(0xd703f000 | STK_OFF_TCCNT, 0xf000 | STK_OFF_TCCNT);
+ /* xc tail_call_cnt(4,%r15),tail_call_cnt(%r15) */
+ _EMIT6(0xd703f000 | offsetof(struct prog_frame, tail_call_cnt),
+ 0xf000 | offsetof(struct prog_frame, tail_call_cnt));
} else {
/*
* Skip the tail call counter initialization in subprograms.
@@ -617,7 +629,7 @@ static void bpf_jit_prologue(struct bpf_jit *jit, struct bpf_prog *fp,
jit->seen_regs |= NVREGS;
} else {
/* Save registers */
- save_restore_regs(jit, REGS_SAVE, stack_depth,
+ save_restore_regs(jit, REGS_SAVE,
fp->aux->exception_boundary ? NVREGS : 0);
}
/* Setup literal pool */
@@ -637,13 +649,15 @@ static void bpf_jit_prologue(struct bpf_jit *jit, struct bpf_prog *fp,
if (is_first_pass(jit) || (jit->seen & SEEN_STACK)) {
/* lgr %w1,%r15 (backchain) */
EMIT4(0xb9040000, REG_W1, REG_15);
- /* la %bfp,STK_160_UNUSED(%r15) (BPF frame pointer) */
- EMIT4_DISP(0x41000000, BPF_REG_FP, REG_15, STK_160_UNUSED);
- /* aghi %r15,-STK_OFF */
- EMIT4_IMM(0xa70b0000, REG_15, -(STK_OFF + stack_depth));
- /* stg %w1,152(%r15) (backchain) */
+ /* la %bfp,unused_end(%r15) (BPF frame pointer) */
+ EMIT4_DISP(0x41000000, BPF_REG_FP, REG_15,
+ offsetofend(struct prog_frame, unused));
+ /* aghi %r15,-frame_off */
+ EMIT4_IMM(0xa70b0000, REG_15, -jit->frame_off);
+ /* stg %w1,backchain(%r15) */
EMIT6_DISP_LH(0xe3000000, 0x0024, REG_W1, REG_0,
- REG_15, 152);
+ REG_15,
+ offsetof(struct prog_frame, backchain));
}
}
@@ -677,13 +691,13 @@ static void call_r1(struct bpf_jit *jit)
/*
* Function epilogue
*/
-static void bpf_jit_epilogue(struct bpf_jit *jit, u32 stack_depth)
+static void bpf_jit_epilogue(struct bpf_jit *jit)
{
jit->exit_ip = jit->prg;
/* Load exit code: lgr %r2,%b0 */
EMIT4(0xb9040000, REG_2, BPF_REG_0);
/* Restore registers */
- save_restore_regs(jit, REGS_RESTORE, stack_depth, 0);
+ save_restore_regs(jit, REGS_RESTORE, 0);
EMIT_JUMP_REG(14);
jit->prg = ALIGN(jit->prg, 8);
@@ -865,7 +879,7 @@ static int sign_extend(struct bpf_jit *jit, int r, u8 size, u8 flags)
* stack space for the large switch statement.
*/
static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
- int i, bool extra_pass, u32 stack_depth)
+ int i, bool extra_pass)
{
struct bpf_insn *insn = &fp->insnsi[i];
s32 branch_oc_off = insn->off;
@@ -1786,9 +1800,10 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
* Note 2: We assume that the verifier does not let us call the
* main program, which clears the tail call counter on entry.
*/
- /* mvc STK_OFF_TCCNT(4,%r15),N(%r15) */
- _EMIT6(0xd203f000 | STK_OFF_TCCNT,
- 0xf000 | (STK_OFF_TCCNT + STK_OFF + stack_depth));
+ /* mvc tail_call_cnt(4,%r15),frame_off+tail_call_cnt(%r15) */
+ _EMIT6(0xd203f000 | offsetof(struct prog_frame, tail_call_cnt),
+ 0xf000 | (jit->frame_off +
+ offsetof(struct prog_frame, tail_call_cnt)));
/* Sign-extend the kfunc arguments. */
if (insn->src_reg == BPF_PSEUDO_KFUNC_CALL) {
@@ -1839,10 +1854,8 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
* goto out;
*/
- if (jit->seen & SEEN_STACK)
- off = STK_OFF_TCCNT + STK_OFF + stack_depth;
- else
- off = STK_OFF_TCCNT;
+ off = jit->frame_off +
+ offsetof(struct prog_frame, tail_call_cnt);
/* lhi %w0,1 */
EMIT4_IMM(0xa7080000, REG_W0, 1);
/* laal %w1,%w0,off(%r15) */
@@ -1872,7 +1885,7 @@ static noinline int bpf_jit_insn(struct bpf_jit *jit, struct bpf_prog *fp,
/*
* Restore registers before calling function
*/
- save_restore_regs(jit, REGS_RESTORE, stack_depth, 0);
+ save_restore_regs(jit, REGS_RESTORE, 0);
/*
* goto *(prog->bpf_func + tail_call_start);
@@ -2165,7 +2178,7 @@ static int bpf_set_addr(struct bpf_jit *jit, int i)
* Compile eBPF program into s390x code
*/
static int bpf_jit_prog(struct bpf_jit *jit, struct bpf_prog *fp,
- bool extra_pass, u32 stack_depth)
+ bool extra_pass)
{
int i, insn_count, lit32_size, lit64_size;
u64 kern_arena;
@@ -2174,24 +2187,30 @@ static int bpf_jit_prog(struct bpf_jit *jit, struct bpf_prog *fp,
jit->lit64 = jit->lit64_start;
jit->prg = 0;
jit->excnt = 0;
+ if (is_first_pass(jit) || (jit->seen & SEEN_STACK))
+ jit->frame_off = sizeof(struct prog_frame) -
+ offsetofend(struct prog_frame, unused) +
+ round_up(fp->aux->stack_depth, 8);
+ else
+ jit->frame_off = 0;
kern_arena = bpf_arena_get_kern_vm_start(fp->aux->arena);
if (kern_arena)
jit->kern_arena = _EMIT_CONST_U64(kern_arena);
jit->user_arena = bpf_arena_get_user_vm_start(fp->aux->arena);
- bpf_jit_prologue(jit, fp, stack_depth);
+ bpf_jit_prologue(jit, fp);
if (bpf_set_addr(jit, 0) < 0)
return -1;
for (i = 0; i < fp->len; i += insn_count) {
- insn_count = bpf_jit_insn(jit, fp, i, extra_pass, stack_depth);
+ insn_count = bpf_jit_insn(jit, fp, i, extra_pass);
if (insn_count < 0)
return -1;
/* Next instruction address */
if (bpf_set_addr(jit, i + insn_count) < 0)
return -1;
}
- bpf_jit_epilogue(jit, stack_depth);
+ bpf_jit_epilogue(jit);
lit32_size = jit->lit32 - jit->lit32_start;
lit64_size = jit->lit64 - jit->lit64_start;
@@ -2267,7 +2286,6 @@ static struct bpf_binary_header *bpf_jit_alloc(struct bpf_jit *jit,
*/
struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
{
- u32 stack_depth = round_up(fp->aux->stack_depth, 8);
struct bpf_prog *tmp, *orig_fp = fp;
struct bpf_binary_header *header;
struct s390_jit_data *jit_data;
@@ -2320,7 +2338,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
* - 3: Calculate program size and addrs array
*/
for (pass = 1; pass <= 3; pass++) {
- if (bpf_jit_prog(&jit, fp, extra_pass, stack_depth)) {
+ if (bpf_jit_prog(&jit, fp, extra_pass)) {
fp = orig_fp;
goto free_addrs;
}
@@ -2334,7 +2352,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
goto free_addrs;
}
skip_init_ctx:
- if (bpf_jit_prog(&jit, fp, extra_pass, stack_depth)) {
+ if (bpf_jit_prog(&jit, fp, extra_pass)) {
bpf_jit_binary_free(header);
fp = orig_fp;
goto free_addrs;
@@ -2654,9 +2672,10 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
/* stg %r1,backchain_off(%r15) */
EMIT6_DISP_LH(0xe3000000, 0x0024, REG_1, REG_0, REG_15,
tjit->backchain_off);
- /* mvc tccnt_off(4,%r15),stack_size+STK_OFF_TCCNT(%r15) */
+ /* mvc tccnt_off(4,%r15),stack_size+tail_call_cnt(%r15) */
_EMIT6(0xd203f000 | tjit->tccnt_off,
- 0xf000 | (tjit->stack_size + STK_OFF_TCCNT));
+ 0xf000 | (tjit->stack_size +
+ offsetof(struct prog_frame, tail_call_cnt)));
/* stmg %r2,%rN,fwd_reg_args_off(%r15) */
if (nr_reg_args)
EMIT6_DISP_LH(0xeb000000, 0x0024, REG_2,
@@ -2793,8 +2812,9 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
(nr_stack_args * sizeof(u64) - 1) << 16 |
tjit->stack_args_off,
0xf000 | tjit->orig_stack_args_off);
- /* mvc STK_OFF_TCCNT(4,%r15),tccnt_off(%r15) */
- _EMIT6(0xd203f000 | STK_OFF_TCCNT, 0xf000 | tjit->tccnt_off);
+ /* mvc tail_call_cnt(4,%r15),tccnt_off(%r15) */
+ _EMIT6(0xd203f000 | offsetof(struct prog_frame, tail_call_cnt),
+ 0xf000 | tjit->tccnt_off);
/* lgr %r1,%r8 */
EMIT4(0xb9040000, REG_1, REG_8);
/* %r1() */
@@ -2851,8 +2871,9 @@ static int __arch_prepare_bpf_trampoline(struct bpf_tramp_image *im,
if (flags & (BPF_TRAMP_F_CALL_ORIG | BPF_TRAMP_F_RET_FENTRY_RET))
EMIT6_DISP_LH(0xe3000000, 0x0004, REG_2, REG_0, REG_15,
tjit->retval_off);
- /* mvc stack_size+STK_OFF_TCCNT(4,%r15),tccnt_off(%r15) */
- _EMIT6(0xd203f000 | (tjit->stack_size + STK_OFF_TCCNT),
+ /* mvc stack_size+tail_call_cnt(4,%r15),tccnt_off(%r15) */
+ _EMIT6(0xd203f000 | (tjit->stack_size +
+ offsetof(struct prog_frame, tail_call_cnt)),
0xf000 | tjit->tccnt_off);
/* aghi %r15,stack_size */
EMIT4_IMM(0xa70b0000, REG_15, tjit->stack_size);
generated by cgit (git 2.34.1) at 2025-08-01 15:16:52 +0000