Author christian.heimes
Recipients christian.heimes, petr.viktorin, serhiy.storchaka, vstinner
Date 2021-10-13.11:49:14
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1634125754.21.0.178579880616.issue45459@roundup.psfhosted.org>
In-reply-to
Content
IIRC shape, strides, and suboffsets are all arrays of ndims length.

We could optimize allocation if we would require users to specify the value of ndims and don't allow them to change the value afterwards. PyBuffer_New(int ndims) then would allocate view of size sizeof(Py_buffer) + (3 * ndims * sizeof(Py_ssize_t *)). This would give us sufficient space to memcpy() shape, strides, and suboffsets arguments into memory after the Py_buffer struct.
History
Date User Action Args
2021-10-13 11:49:14christian.heimessetrecipients: + christian.heimes, vstinner, petr.viktorin, serhiy.storchaka
2021-10-13 11:49:14christian.heimessetmessageid: <1634125754.21.0.178579880616.issue45459@roundup.psfhosted.org>
2021-10-13 11:49:14christian.heimeslinkissue45459 messages
2021-10-13 11:49:14christian.heimescreate