summaryrefslogtreecommitdiff
path: root/arch/x86/crypto/sha-mb
diff options
context:
space:
mode:
Diffstat (limited to 'arch/x86/crypto/sha-mb')
-rw-r--r--arch/x86/crypto/sha-mb/sha1_mb.c35
-rw-r--r--arch/x86/crypto/sha-mb/sha1_mb_mgr_flush_avx2.S35
-rw-r--r--arch/x86/crypto/sha-mb/sha1_mb_mgr_submit_avx2.S38
-rw-r--r--arch/x86/crypto/sha-mb/sha1_x8_avx2.S13
4 files changed, 61 insertions, 60 deletions
diff --git a/arch/x86/crypto/sha-mb/sha1_mb.c b/arch/x86/crypto/sha-mb/sha1_mb.c
index 8381c09d2..081255cea 100644
--- a/arch/x86/crypto/sha-mb/sha1_mb.c
+++ b/arch/x86/crypto/sha-mb/sha1_mb.c
@@ -762,6 +762,38 @@ static int sha1_mb_async_digest(struct ahash_request *req)
return crypto_ahash_digest(mcryptd_req);
}
+static int sha1_mb_async_export(struct ahash_request *req, void *out)
+{
+ struct ahash_request *mcryptd_req = ahash_request_ctx(req);
+ struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+ struct sha1_mb_ctx *ctx = crypto_ahash_ctx(tfm);
+ struct mcryptd_ahash *mcryptd_tfm = ctx->mcryptd_tfm;
+
+ memcpy(mcryptd_req, req, sizeof(*req));
+ ahash_request_set_tfm(mcryptd_req, &mcryptd_tfm->base);
+ return crypto_ahash_export(mcryptd_req, out);
+}
+
+static int sha1_mb_async_import(struct ahash_request *req, const void *in)
+{
+ struct ahash_request *mcryptd_req = ahash_request_ctx(req);
+ struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+ struct sha1_mb_ctx *ctx = crypto_ahash_ctx(tfm);
+ struct mcryptd_ahash *mcryptd_tfm = ctx->mcryptd_tfm;
+ struct crypto_shash *child = mcryptd_ahash_child(mcryptd_tfm);
+ struct mcryptd_hash_request_ctx *rctx;
+ struct shash_desc *desc;
+
+ memcpy(mcryptd_req, req, sizeof(*req));
+ ahash_request_set_tfm(mcryptd_req, &mcryptd_tfm->base);
+ rctx = ahash_request_ctx(mcryptd_req);
+ desc = &rctx->desc;
+ desc->tfm = child;
+ desc->flags = CRYPTO_TFM_REQ_MAY_SLEEP;
+
+ return crypto_ahash_import(mcryptd_req, in);
+}
+
static int sha1_mb_async_init_tfm(struct crypto_tfm *tfm)
{
struct mcryptd_ahash *mcryptd_tfm;
@@ -796,8 +828,11 @@ static struct ahash_alg sha1_mb_async_alg = {
.final = sha1_mb_async_final,
.finup = sha1_mb_async_finup,
.digest = sha1_mb_async_digest,
+ .export = sha1_mb_async_export,
+ .import = sha1_mb_async_import,
.halg = {
.digestsize = SHA1_DIGEST_SIZE,
+ .statesize = sizeof(struct sha1_hash_ctx),
.base = {
.cra_name = "sha1",
.cra_driver_name = "sha1_mb",
diff --git a/arch/x86/crypto/sha-mb/sha1_mb_mgr_flush_avx2.S b/arch/x86/crypto/sha-mb/sha1_mb_mgr_flush_avx2.S
index 85c4e1cf7..96df6a39d 100644
--- a/arch/x86/crypto/sha-mb/sha1_mb_mgr_flush_avx2.S
+++ b/arch/x86/crypto/sha-mb/sha1_mb_mgr_flush_avx2.S
@@ -52,6 +52,7 @@
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <linux/linkage.h>
+#include <asm/frame.h>
#include "sha1_mb_mgr_datastruct.S"
@@ -86,16 +87,6 @@
#define extra_blocks %arg2
#define p %arg2
-
-# STACK_SPACE needs to be an odd multiple of 8
-_XMM_SAVE_SIZE = 10*16
-_GPR_SAVE_SIZE = 8*8
-_ALIGN_SIZE = 8
-
-_XMM_SAVE = 0
-_GPR_SAVE = _XMM_SAVE + _XMM_SAVE_SIZE
-STACK_SPACE = _GPR_SAVE + _GPR_SAVE_SIZE + _ALIGN_SIZE
-
.macro LABEL prefix n
\prefix\n\():
.endm
@@ -113,16 +104,8 @@ offset = \_offset
# JOB* sha1_mb_mgr_flush_avx2(MB_MGR *state)
# arg 1 : rcx : state
ENTRY(sha1_mb_mgr_flush_avx2)
- mov %rsp, %r10
- sub $STACK_SPACE, %rsp
- and $~31, %rsp
- mov %rbx, _GPR_SAVE(%rsp)
- mov %r10, _GPR_SAVE+8*1(%rsp) #save rsp
- mov %rbp, _GPR_SAVE+8*3(%rsp)
- mov %r12, _GPR_SAVE+8*4(%rsp)
- mov %r13, _GPR_SAVE+8*5(%rsp)
- mov %r14, _GPR_SAVE+8*6(%rsp)
- mov %r15, _GPR_SAVE+8*7(%rsp)
+ FRAME_BEGIN
+ push %rbx
# If bit (32+3) is set, then all lanes are empty
mov _unused_lanes(state), unused_lanes
@@ -230,16 +213,8 @@ len_is_0:
mov tmp2_w, offset(job_rax)
return:
-
- mov _GPR_SAVE(%rsp), %rbx
- mov _GPR_SAVE+8*1(%rsp), %r10 #saved rsp
- mov _GPR_SAVE+8*3(%rsp), %rbp
- mov _GPR_SAVE+8*4(%rsp), %r12
- mov _GPR_SAVE+8*5(%rsp), %r13
- mov _GPR_SAVE+8*6(%rsp), %r14
- mov _GPR_SAVE+8*7(%rsp), %r15
- mov %r10, %rsp
-
+ pop %rbx
+ FRAME_END
ret
return_null:
diff --git a/arch/x86/crypto/sha-mb/sha1_mb_mgr_submit_avx2.S b/arch/x86/crypto/sha-mb/sha1_mb_mgr_submit_avx2.S
index 2ab9560b5..63a0d9c8e 100644
--- a/arch/x86/crypto/sha-mb/sha1_mb_mgr_submit_avx2.S
+++ b/arch/x86/crypto/sha-mb/sha1_mb_mgr_submit_avx2.S
@@ -53,6 +53,7 @@
*/
#include <linux/linkage.h>
+#include <asm/frame.h>
#include "sha1_mb_mgr_datastruct.S"
@@ -86,33 +87,21 @@ job_rax = %rax
len = %rax
DWORD_len = %eax
-lane = %rbp
-tmp3 = %rbp
+lane = %r12
+tmp3 = %r12
tmp = %r9
DWORD_tmp = %r9d
lane_data = %r10
-# STACK_SPACE needs to be an odd multiple of 8
-STACK_SPACE = 8*8 + 16*10 + 8
-
# JOB* submit_mb_mgr_submit_avx2(MB_MGR *state, job_sha1 *job)
# arg 1 : rcx : state
# arg 2 : rdx : job
ENTRY(sha1_mb_mgr_submit_avx2)
-
- mov %rsp, %r10
- sub $STACK_SPACE, %rsp
- and $~31, %rsp
-
- mov %rbx, (%rsp)
- mov %r10, 8*2(%rsp) #save old rsp
- mov %rbp, 8*3(%rsp)
- mov %r12, 8*4(%rsp)
- mov %r13, 8*5(%rsp)
- mov %r14, 8*6(%rsp)
- mov %r15, 8*7(%rsp)
+ FRAME_BEGIN
+ push %rbx
+ push %r12
mov _unused_lanes(state), unused_lanes
mov unused_lanes, lane
@@ -197,22 +186,15 @@ len_is_0:
vpinsrd $1, _args_digest+1*32(state , idx, 4), %xmm0, %xmm0
vpinsrd $2, _args_digest+2*32(state , idx, 4), %xmm0, %xmm0
vpinsrd $3, _args_digest+3*32(state , idx, 4), %xmm0, %xmm0
- movl 4*32(state, idx, 4), DWORD_tmp
+ movl _args_digest+4*32(state, idx, 4), DWORD_tmp
vmovdqu %xmm0, _result_digest(job_rax)
movl DWORD_tmp, _result_digest+1*16(job_rax)
return:
-
- mov (%rsp), %rbx
- mov 8*2(%rsp), %r10 #save old rsp
- mov 8*3(%rsp), %rbp
- mov 8*4(%rsp), %r12
- mov 8*5(%rsp), %r13
- mov 8*6(%rsp), %r14
- mov 8*7(%rsp), %r15
- mov %r10, %rsp
-
+ pop %r12
+ pop %rbx
+ FRAME_END
ret
return_null:
diff --git a/arch/x86/crypto/sha-mb/sha1_x8_avx2.S b/arch/x86/crypto/sha-mb/sha1_x8_avx2.S
index 8e1b47792..c9dae1cd2 100644
--- a/arch/x86/crypto/sha-mb/sha1_x8_avx2.S
+++ b/arch/x86/crypto/sha-mb/sha1_x8_avx2.S
@@ -296,7 +296,11 @@ W14 = TMP_
#
ENTRY(sha1_x8_avx2)
- push RSP_SAVE
+ # save callee-saved clobbered registers to comply with C function ABI
+ push %r12
+ push %r13
+ push %r14
+ push %r15
#save rsp
mov %rsp, RSP_SAVE
@@ -446,7 +450,12 @@ lloop:
## Postamble
mov RSP_SAVE, %rsp
- pop RSP_SAVE
+
+ # restore callee-saved clobbered registers
+ pop %r15
+ pop %r14
+ pop %r13
+ pop %r12
ret
ENDPROC(sha1_x8_avx2)
generated by cgit v1.2.3-54-g00ecf (git 2.45.2) at 2024-07-04 03:50:20 +0000