From: Taotao Chen <chentao...@didiglobal.com> Refactors shmem_pwrite() to replace the ->write_begin/end logic with a write_iter-based implementation using kiocb and iov_iter.
While kernel_write() was considered, it caused about 50% performance regression. vfs_write() is not exported for kernel use. Therefore, file->f_op->write_iter() is called directly with a synchronously initialized kiocb to preserve performance and remove write_begin usage. Performance results use gem_pwrite on Intel CPU i7-10700 (average of 10 runs): - ./gem_pwrite --run-subtest bench -s 16384 Before: 0.205s, After: 0.214s - ./gem_pwrite --run-subtest bench -s 524288 Before: 6.1021s, After: 4.8047s Part of a series refactoring address_space_operations write_begin and write_end callbacks to use struct kiocb for passing write context and flags. Signed-off-by: Taotao Chen <chentao...@didiglobal.com> --- drivers/gpu/drm/i915/gem/i915_gem_shmem.c | 82 ++++++----------------- 1 file changed, 22 insertions(+), 60 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c index 1e8f66ac48ca..9cbb0f68a5bb 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_shmem.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_shmem.c @@ -6,6 +6,7 @@ #include <linux/pagevec.h> #include <linux/shmem_fs.h> #include <linux/swap.h> +#include <linux/uio.h> #include <drm/drm_cache.h> @@ -400,12 +401,12 @@ static int shmem_pwrite(struct drm_i915_gem_object *obj, const struct drm_i915_gem_pwrite *arg) { - struct address_space *mapping = obj->base.filp->f_mapping; - const struct address_space_operations *aops = mapping->a_ops; char __user *user_data = u64_to_user_ptr(arg->data_ptr); - u64 remain; - loff_t pos; - unsigned int pg; + struct file *file = obj->base.filp; + struct kiocb kiocb; + struct iov_iter iter; + ssize_t written; + u64 size = arg->size; /* Caller already validated user args */ GEM_BUG_ON(!access_ok(user_data, arg->size)); @@ -428,63 +429,24 @@ shmem_pwrite(struct drm_i915_gem_object *obj, if (obj->mm.madv != I915_MADV_WILLNEED) return -EFAULT; - /* - * Before the pages are instantiated the object is treated as being - * in the CPU domain. The pages will be clflushed as required before - * use, and we can freely write into the pages directly. If userspace - * races pwrite with any other operation; corruption will ensue - - * that is userspace's prerogative! - */ + if (size > MAX_RW_COUNT) + return -EFBIG; - remain = arg->size; - pos = arg->offset; - pg = offset_in_page(pos); + if (!file->f_op->write_iter) + return -EINVAL; - do { - unsigned int len, unwritten; - struct folio *folio; - void *data, *vaddr; - int err; - char __maybe_unused c; - - len = PAGE_SIZE - pg; - if (len > remain) - len = remain; - - /* Prefault the user page to reduce potential recursion */ - err = __get_user(c, user_data); - if (err) - return err; - - err = __get_user(c, user_data + len - 1); - if (err) - return err; - - err = aops->write_begin(obj->base.filp, mapping, pos, len, - &folio, &data); - if (err < 0) - return err; - - vaddr = kmap_local_folio(folio, offset_in_folio(folio, pos)); - pagefault_disable(); - unwritten = __copy_from_user_inatomic(vaddr, user_data, len); - pagefault_enable(); - kunmap_local(vaddr); - - err = aops->write_end(obj->base.filp, mapping, pos, len, - len - unwritten, folio, data); - if (err < 0) - return err; - - /* We don't handle -EFAULT, leave it to the caller to check */ - if (unwritten) - return -ENODEV; - - remain -= len; - user_data += len; - pos += len; - pg = 0; - } while (remain); + init_sync_kiocb(&kiocb, file); + kiocb.ki_pos = arg->offset; + iov_iter_ubuf(&iter, ITER_SOURCE, (void __user *)user_data, size); + + written = file->f_op->write_iter(&kiocb, &iter); + BUG_ON(written == -EIOCBQUEUED); + + if (written != size) + return -EIO; + + if (written < 0) + return written; return 0; } -- 2.34.1