[XPU] llama add xpu support #8282
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
|
|
@@ -413,6 +413,13 @@ def forward(self, hidden_states): | |
| if self.config.use_fused_rms_norm: | ||
| if get_env_device() == "npu": | ||
| return core.eager._run_custom_op("rms_norm_npu", hidden_states, self.weight, self.variance_epsilon)[0] | ||
| elif get_env_device() == "xpu": | ||
| try: | ||
| import paddle_xpu_nn # noqa: F821 | ||
|
|
||
| return paddle_xpu_nn.xpu_rms_norm(hidden_states, self.weight, self.variance_epsilon)[0] | ||
| except ImportError: | ||
| pass | ||
|
dynamicheart marked this conversation as resolved.
Outdated
|
||
| return rms_norm_fused(hidden_states, self.weight, self.variance_epsilon) | ||
|
|
||
| if paddle.in_dynamic_mode(): | ||
|
|
@@ -582,12 +589,49 @@ def __init__(self, config): | |
|
|
||
| ColumnParallelLinear = MC2ColumnSeqParallelLinear | ||
| RowParallelLinear = MC2RowSeqParallelLinear | ||
| elif get_env_device() == "xpu": | ||
| try: | ||
| from paddle_xpu.layers.nn.sequence_parallel import ( # noqa: F401 | ||
| XPUColumnSequenceParallelLinear, | ||
| XPURowSequenceParallelLinear, | ||
| ) | ||
|
|
||
| ColumnParallelLinear = XPUColumnSequenceParallelLinear | ||
| RowParallelLinear = XPURowSequenceParallelLinear | ||
| except ImportError: | ||
| ColumnParallelLinear = ColumnSequenceParallelLinear | ||
| RowParallelLinear = RowSequenceParallelLinear | ||
|
dynamicheart marked this conversation as resolved.
Outdated
|
||
| else: | ||
| ColumnParallelLinear = ColumnSequenceParallelLinear | ||
| RowParallelLinear = RowSequenceParallelLinear | ||
| else: | ||
| if get_env_device() == "xpu": | ||
| try: | ||
| from paddle_xpu.layers.nn import ( # noqa: F401 | ||
| ColumnParallelLinear as XPUColumnParallelLinear, | ||
| ) | ||
| from paddle_xpu.layers.nn import ( # noqa: F401 | ||
| RowParallelLinear as XPURowParallelLinear, | ||
| ) | ||
|
|
||
| ColumnParallelLinear = XPUColumnParallelLinear | ||
| RowParallelLinear = XPURowParallelLinear | ||
| except ImportError: | ||
| ColumnParallelLinear = fleet.meta_parallel.ColumnParallelLinear | ||
|
dynamicheart marked this conversation as resolved.
Outdated
|
||
| RowParallelLinear = fleet.meta_parallel.RowParallelLinear | ||
| else: | ||
| ColumnParallelLinear = fleet.meta_parallel.ColumnParallelLinear | ||
| RowParallelLinear = fleet.meta_parallel.RowParallelLinear | ||
|
|
||
| if get_env_device() == "xpu": | ||
| try: | ||
| from paddle_xpu.layers.nn import Linear as XPULinear # noqa: F401 | ||
|
|
||
| Linear = XPULinear | ||
| except ImportError: | ||
| Linear = nn.Linear | ||
|
dynamicheart marked this conversation as resolved.
Outdated
|
||
| else: | ||
| Linear = nn.Linear | ||
|
|
||
| if config.tensor_parallel_degree > 1: | ||
| if config.fuse_attention_ffn: | ||
|
|
@@ -619,15 +663,29 @@ def __init__(self, config): | |
| ) | ||
| else: | ||
| if config.fuse_attention_ffn: | ||
| self.gate_up_fused_proj = Linear(self.hidden_size, self.intermediate_size * 2, bias_attr=False) | ||
| else: | ||
| self.gate_proj = Linear(self.hidden_size, self.intermediate_size, bias_attr=False) | ||
| self.up_proj = Linear(self.hidden_size, self.intermediate_size, bias_attr=False) | ||
|
|
||
| self.down_proj = Linear(self.intermediate_size, self.hidden_size, bias_attr=False) | ||
|
|
||
| def forward(self, x): | ||
| if self.fuse_attention_ffn: | ||
| # FIXME(yangjianbang): use paddle's native swiglu | ||
| if get_env_device() == "xpu": | ||
| try: | ||
| import paddle_xpu_nn # noqa: F821 | ||
|
|
||
| out = self.gate_up_fused_proj(x) | ||
| out = paddle_xpu_nn.xpu_swiglu(out, axis=-1, turn=True) | ||
| out = self.down_proj(out) | ||
| return out | ||
| except ImportError: | ||
| gate_out, up_out = paddle.chunk(self.gate_up_fused_proj(x), chunks=2, axis=-1) | ||
| out = self.down_proj(F.silu(gate_out) * up_out) | ||
| return out | ||
|
|
||
| x = swiglu(self.gate_up_fused_proj(x)) | ||
| else: | ||
| x = swiglu(self.gate_proj(x), self.up_proj(x)) | ||
|
|
@@ -688,7 +746,7 @@ def __init__(self, config: LlamaConfig, layerwise_recompute: bool = False): | |
| ) | ||
|
|
||
| self.use_fused_rope = config.use_fused_rope | ||
| if self.use_fused_rope and get_env_device() not in ["npu", "xpu"]: | ||
| if "gpu" not in paddle.device.get_device() or fused_rotary_position_embedding is None: | ||
| warnings.warn( | ||
| "Enable fuse rope in the config, but fuse rope is not available. " | ||
|
|
@@ -705,12 +763,49 @@ def __init__(self, config: LlamaConfig, layerwise_recompute: bool = False): | |
|
|
||
| ColumnParallelLinear = MC2ColumnSeqParallelLinear | ||
| RowParallelLinear = MC2RowSeqParallelLinear | ||
| elif get_env_device() == "xpu": | ||
| try: | ||
| from paddle_xpu.layers.nn.sequence_parallel import ( # noqa: F401 | ||
| XPUColumnSequenceParallelLinear, | ||
| XPURowSequenceParallelLinear, | ||
| ) | ||
|
|
||
| ColumnParallelLinear = XPUColumnSequenceParallelLinear | ||
| RowParallelLinear = XPURowSequenceParallelLinear | ||
| except ImportError: | ||
| ColumnParallelLinear = ColumnSequenceParallelLinear | ||
| RowParallelLinear = RowSequenceParallelLinear | ||
| else: | ||
| ColumnParallelLinear = ColumnSequenceParallelLinear | ||
| RowParallelLinear = RowSequenceParallelLinear | ||
| else: | ||
| if get_env_device() == "xpu": | ||
| try: | ||
| from paddle_xpu.layers.nn import ( # noqa: F401 | ||
| ColumnParallelLinear as XPUColumnParallelLinear, | ||
| ) | ||
| from paddle_xpu.layers.nn import ( # noqa: F401 | ||
| RowParallelLinear as XPURowParallelLinear, | ||
| ) | ||
|
|
||
| ColumnParallelLinear = XPUColumnParallelLinear | ||
| RowParallelLinear = XPURowParallelLinear | ||
| except ImportError: | ||
| ColumnParallelLinear = fleet.meta_parallel.ColumnParallelLinear | ||
| RowParallelLinear = fleet.meta_parallel.RowParallelLinear | ||
| else: | ||
| ColumnParallelLinear = fleet.meta_parallel.ColumnParallelLinear | ||
| RowParallelLinear = fleet.meta_parallel.RowParallelLinear | ||
|
|
||
| if get_env_device() == "xpu": | ||
| try: | ||
| from paddle_xpu.layers.nn import Linear as XPULinear # noqa: F401 | ||
|
|
||
| Linear = XPULinear | ||
| except: | ||
| Linear = nn.Linear | ||
| else: | ||
| Linear = nn.Linear | ||
|
|
||
| if config.tensor_parallel_degree > 1: | ||
| if self.fuse_attention_qkv: | ||
|
|
@@ -741,36 +836,36 @@ def __init__(self, config: LlamaConfig, layerwise_recompute: bool = False): | |
| gather_output=False, | ||
| ) | ||
| else: | ||
| self.k_proj = Linear( | ||
| self.hidden_size, | ||
| self.config.num_key_value_heads * self.head_dim, | ||
| bias_attr=False, | ||
| ) | ||
| self.v_proj = Linear( | ||
| self.hidden_size, | ||
| self.config.num_key_value_heads * self.head_dim, | ||
| bias_attr=False, | ||
| ) | ||
|
|
||
| else: | ||
| if self.fuse_attention_qkv: | ||
| self.qkv_proj = Linear( | ||
| self.hidden_size, | ||
| self.hidden_size + 2 * self.config.num_key_value_heads * self.head_dim, | ||
| bias_attr=False, | ||
| ) | ||
| else: | ||
| self.q_proj = Linear( | ||
| self.hidden_size, | ||
| self.hidden_size, | ||
| bias_attr=False, | ||
| ) | ||
| self.k_proj = Linear( | ||
| self.hidden_size, | ||
| self.config.num_key_value_heads * self.head_dim, | ||
| bias_attr=False, | ||
| ) | ||
| self.v_proj = Linear( | ||
| self.hidden_size, | ||
| self.config.num_key_value_heads * self.head_dim, | ||
| bias_attr=False, | ||
|
|
@@ -784,7 +879,7 @@ def __init__(self, config: LlamaConfig, layerwise_recompute: bool = False): | |
| input_is_parallel=True, | ||
| ) | ||
| else: | ||
| self.o_proj = Linear( | ||
| self.hidden_size, | ||
| self.hidden_size, | ||
| bias_attr=False, | ||
|
|
@@ -1428,6 +1523,11 @@ def _prepare_decoder_attention_mask(attention_mask, input_shape, past_key_values | |
| y = paddle.to_tensor(paddle.finfo(dtype).min, dtype="float16") | ||
| expanded_attn_mask = expanded_attn_mask.astype("float16") | ||
| expanded_attn_mask = paddle.where(expanded_attn_mask, x, y).astype(dtype) | ||
| elif get_env_device() == "xpu": | ||
| x = paddle.to_tensor(0.0, dtype=dtype) | ||
| y = paddle.to_tensor(paddle.finfo(dtype).min, dtype=dtype) | ||
| expanded_attn_mask = expanded_attn_mask.astype(dtype) | ||
| expanded_attn_mask = paddle.where(expanded_attn_mask, x, y).astype(dtype) | ||
|
Contributor
Author
There was a problem hiding this comment. 当传入的
Contributor
Author
There was a problem hiding this comment. 理论上是可以复用的，但是npu里面写死了dtype是
Contributor
Author
There was a problem hiding this comment. 根据 @wuhuachaocoding 意见，还是分成if elif两个单独的分支 |
||
| else: | ||
| expanded_attn_mask = paddle.where(expanded_attn_mask, 0.0, paddle.finfo(dtype).min).astype(dtype) | ||
| return expanded_attn_mask | ||
|
|
@@ -1708,6 +1808,15 @@ def __init__(self, config: LlamaConfig): | |
| self.weight.is_distributed = True if (vocab_size != config.vocab_size) else False | ||
| if self.weight.is_distributed: | ||
| self.weight.split_axis = 1 | ||
| if get_env_device() == "xpu": | ||
| try: | ||
| from paddle_xpu.layers.nn import ( # noqa: F401 | ||
| parallel_matmul as xpu_parallel_matmul, | ||
| ) | ||
|
|
||
| self.xpu_parallel_matmul = xpu_parallel_matmul() | ||
| except ImportError: | ||
| self.xpu_parallel_matmul = None | ||
|
|
||
| def forward(self, hidden_states, tensor_parallel_output=None): | ||
| if self.config.sequence_parallel: | ||
|
|
@@ -1721,7 +1830,12 @@ def forward(self, hidden_states, tensor_parallel_output=None): | |
| if tensor_parallel_output is None: | ||
| tensor_parallel_output = self.config.tensor_parallel_output | ||
|
|
||
| if get_env_device() == "xpu" and self.xpu_parallel_matmul is not None: | ||
| logits = self.xpu_parallel_matmul( | ||
| hidden_states, self.weight, tensor_parallel_output=tensor_parallel_output, training=self.training | ||
|
Comment on lines
+1742
to
+1743
Contributor
There was a problem hiding this comment. training 参数必须要吗？如果参数能一样的话，是不是 把 parallel_matmul 的实现在xpu下替换就好了？
Contributor
Author
There was a problem hiding this comment. 这里面有两个原因：
|
||
| ) | ||
| else: | ||
| logits = parallel_matmul(hidden_states, self.weight, tensor_parallel_output=tensor_parallel_output) | ||
| return logits | ||
|
|
||
|
|
||
|
|
||
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
这个是做什么的？
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
XPU针对accumulate_steps > 1的场景进行优化，配合下面的paddle_xpu里面的Linear层进行使用