Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

gh-99108: Release the GIL around hashlib built-in computation #104675

Merged
merged 7 commits into from
May 23, 2023
Merged

gh-99108: Release the GIL around hashlib built-in computation #104675

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
73094a1
gh-99108: Release the GIL around hashlib built-in hash updates.
gpshead May 19, 2023
621f9a6
Do the same for sha2, sha1, & md5.
gpshead May 20, 2023
4cf2a19
reword NEWS.
gpshead May 20, 2023
88f190f
Better describe the ENTER/LEAVE macros.
gpshead May 22, 2023
c8e0b01
Explain the locking strategy in comments.
gpshead May 22, 2023
2830028
Merge branch 'main' into hashlib_hacl_dropGIL
gpshead May 22, 2023
0e60cec
Merge branch 'main' into hashlib_hacl_dropGIL
gpshead May 22, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 3 additions & 0 deletions Misc/NEWS.d/next/Library/2023-05-19-19-46-22.gh-issue-99108.wqCg0t.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
We now release the GIL around built-in :mod:`hashlib` computations of
reasonable size for the SHA families and MD5 hash functions, matching
what our OpenSSL backed hash computations already does.
2 changes: 1 addition & 1 deletion Modules/hashlib.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,7 +53,7 @@
PyThread_release_lock((obj)->lock); \
}

/* TODO(gps): We should probably make this a module or EVPobject attribute
/* TODO(gpshead): We should make this a module or class attribute
* to allow the user to optimize based on the platform they're using. */
#define HASHLIB_GIL_MINSIZE 2048

33 changes: 30 additions & 3 deletions Modules/md5module.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -49,7 +49,7 @@ typedef long long MD5_INT64; /* 64-bit integer */

typedef struct {
PyObject_HEAD

PyThread_type_lock lock;
Hacl_Streaming_MD5_state *hash_state;
} MD5object;

Expand All @@ -72,6 +72,7 @@ static MD5object *
newMD5object(MD5State * st)
{
MD5object *md5 = (MD5object *)PyObject_GC_New(MD5object, st->md5_type);
md5->lock = NULL;
gpshead marked this conversation as resolved.
Show resolved Hide resolved
PyObject_GC_Track(md5);
return md5;
}
Expand All @@ -88,6 +89,9 @@ static void
MD5_dealloc(MD5object *ptr)
{
Hacl_Streaming_MD5_legacy_free(ptr->hash_state);
if (ptr->lock != NULL) {
PyThread_free_lock(ptr->lock);
}
PyTypeObject *tp = Py_TYPE(ptr);
PyObject_GC_UnTrack(ptr);
PyObject_GC_Del(ptr);
Expand Down Expand Up @@ -115,7 +119,9 @@ MD5Type_copy_impl(MD5object *self, PyTypeObject *cls)
if ((newobj = newMD5object(st))==NULL)
return NULL;

ENTER_HASHLIB(self);
newobj->hash_state = Hacl_Streaming_MD5_legacy_copy(self->hash_state);
LEAVE_HASHLIB(self);
return (PyObject *)newobj;
}

Expand All @@ -130,7 +136,9 @@ MD5Type_digest_impl(MD5object *self)
/*[clinic end generated code: output=eb691dc4190a07ec input=bc0c4397c2994be6]*/
{
unsigned char digest[MD5_DIGESTSIZE];
ENTER_HASHLIB(self);
Hacl_Streaming_MD5_legacy_finish(self->hash_state, digest);
LEAVE_HASHLIB(self);
return PyBytes_FromStringAndSize((const char *)digest, MD5_DIGESTSIZE);
}

Expand All @@ -145,7 +153,9 @@ MD5Type_hexdigest_impl(MD5object *self)
/*[clinic end generated code: output=17badced1f3ac932 input=b60b19de644798dd]*/
{
unsigned char digest[MD5_DIGESTSIZE];
ENTER_HASHLIB(self);
Hacl_Streaming_MD5_legacy_finish(self->hash_state, digest);
LEAVE_HASHLIB(self);
return _Py_strhex((const char*)digest, MD5_DIGESTSIZE);
}

Expand Down Expand Up @@ -177,7 +187,18 @@ MD5Type_update(MD5object *self, PyObject *obj)

GET_BUFFER_VIEW_OR_ERROUT(obj, &buf);

update(self->hash_state, buf.buf, buf.len);
if (self->lock == NULL && buf.len >= HASHLIB_GIL_MINSIZE) {
self->lock = PyThread_allocate_lock();
}
if (self->lock != NULL) {
Py_BEGIN_ALLOW_THREADS
PyThread_acquire_lock(self->lock, 1);
gpshead marked this conversation as resolved.
Show resolved Hide resolved
update(self->hash_state, buf.buf, buf.len);
PyThread_release_lock(self->lock);
Py_END_ALLOW_THREADS
} else {
update(self->hash_state, buf.buf, buf.len);
}

PyBuffer_Release(&buf);
Py_RETURN_NONE;
Expand Down Expand Up @@ -279,7 +300,13 @@ _md5_md5_impl(PyObject *module, PyObject *string, int usedforsecurity)
return NULL;
}
if (string) {
update(new->hash_state, buf.buf, buf.len);
if (buf.len >= HASHLIB_GIL_MINSIZE) {
gpshead marked this conversation as resolved.
Show resolved Hide resolved
Py_BEGIN_ALLOW_THREADS
update(new->hash_state, buf.buf, buf.len);
Py_END_ALLOW_THREADS
} else {
update(new->hash_state, buf.buf, buf.len);
}
PyBuffer_Release(&buf);
}

Expand Down
33 changes: 30 additions & 3 deletions Modules/sha1module.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -48,7 +48,7 @@ typedef long long SHA1_INT64; /* 64-bit integer */

typedef struct {
PyObject_HEAD

PyThread_type_lock lock;
Hacl_Streaming_SHA1_state *hash_state;
} SHA1object;

Expand All @@ -71,6 +71,7 @@ static SHA1object *
newSHA1object(SHA1State *st)
{
SHA1object *sha = (SHA1object *)PyObject_GC_New(SHA1object, st->sha1_type);
sha->lock = NULL;
PyObject_GC_Track(sha);
return sha;
}
Expand All @@ -88,6 +89,9 @@ static void
SHA1_dealloc(SHA1object *ptr)
{
Hacl_Streaming_SHA1_legacy_free(ptr->hash_state);
if (ptr->lock != NULL) {
PyThread_free_lock(ptr->lock);
}
PyTypeObject *tp = Py_TYPE(ptr);
PyObject_GC_UnTrack(ptr);
PyObject_GC_Del(ptr);
Expand Down Expand Up @@ -115,7 +119,9 @@ SHA1Type_copy_impl(SHA1object *self, PyTypeObject *cls)
if ((newobj = newSHA1object(st)) == NULL)
return NULL;

ENTER_HASHLIB(self);
newobj->hash_state = Hacl_Streaming_SHA1_legacy_copy(self->hash_state);
LEAVE_HASHLIB(self);
return (PyObject *)newobj;
}

Expand All @@ -130,7 +136,9 @@ SHA1Type_digest_impl(SHA1object *self)
/*[clinic end generated code: output=2f05302a7aa2b5cb input=13824b35407444bd]*/
{
unsigned char digest[SHA1_DIGESTSIZE];
ENTER_HASHLIB(self);
Hacl_Streaming_SHA1_legacy_finish(self->hash_state, digest);
LEAVE_HASHLIB(self);
return PyBytes_FromStringAndSize((const char *)digest, SHA1_DIGESTSIZE);
}

Expand All @@ -145,7 +153,9 @@ SHA1Type_hexdigest_impl(SHA1object *self)
/*[clinic end generated code: output=4161fd71e68c6659 input=97691055c0c74ab0]*/
{
unsigned char digest[SHA1_DIGESTSIZE];
ENTER_HASHLIB(self);
Hacl_Streaming_SHA1_legacy_finish(self->hash_state, digest);
LEAVE_HASHLIB(self);
return _Py_strhex((const char *)digest, SHA1_DIGESTSIZE);
}

Expand Down Expand Up @@ -177,7 +187,18 @@ SHA1Type_update(SHA1object *self, PyObject *obj)

GET_BUFFER_VIEW_OR_ERROUT(obj, &buf);

update(self->hash_state, buf.buf, buf.len);
if (self->lock == NULL && buf.len >= HASHLIB_GIL_MINSIZE) {
self->lock = PyThread_allocate_lock();
}
if (self->lock != NULL) {
Py_BEGIN_ALLOW_THREADS
PyThread_acquire_lock(self->lock, 1);
update(self->hash_state, buf.buf, buf.len);
PyThread_release_lock(self->lock);
Py_END_ALLOW_THREADS
} else {
update(self->hash_state, buf.buf, buf.len);
}

PyBuffer_Release(&buf);
Py_RETURN_NONE;
Expand Down Expand Up @@ -279,7 +300,13 @@ _sha1_sha1_impl(PyObject *module, PyObject *string, int usedforsecurity)
return NULL;
}
if (string) {
update(new->hash_state, buf.buf, buf.len);
if (buf.len >= HASHLIB_GIL_MINSIZE) {
Py_BEGIN_ALLOW_THREADS
update(new->hash_state, buf.buf, buf.len);
Py_END_ALLOW_THREADS
} else {
update(new->hash_state, buf.buf, buf.len);
}
PyBuffer_Release(&buf);
}

Expand Down
Loading