[ET-VK] Using TmpTensor for width packed versions of q_linear op shader to reduce memory usage. - #7929

[ET-VK] Using TmpTensor for width packed versions of q_linear op shader to reduce memory usage.#7929

facebook-github-bot merged 8 commits into

gh/trivedivivek/53/basepytorch/executorch:gh/trivedivivek/53/basefrom

gh/trivedivivek/53/headpytorch/executorch:gh/trivedivivek/53/headCopy head branch name to clipboard