Closed
[Bug-Report|缺陷反馈]: aclnnRmsNormQuant x/y 前置维不一致时报 561103 而非 161002(仅比对末维) #7400
liuhanwen8Add on  26 days agoClosed at  17 days ago
liuhanwen8
liuhanwen8
26 days ago Create

现象

x 与 y 的前置维不一致(x=[1,3,5]、y=[3,5])时第一段返回 561103,资料要求该形态返回 161002。对照:末维不一致(x=[3,5]、y=[3,4])能被正确拦下并返回 161002。

场景 x shape y shape 第一段
合法基线 [3,5] [3,5] 0
缺陷(前置维不等) [1,3,5] [3,5] 561103
阴性对照(末维不等) [3,5] [3,4] 161002

期望

资料 norm/rms_norm_quant/docs/aclnnRmsNormQuant.md「返回值」表:「输入/输出的 shape 关系不符合预期」→ 161002。

环境

  • 硬件:Atlas A2 训练系列产品(Ascend 910B2,单卡复现)
  • CANN:9.2.0(Ascend-cann-toolkit_9.2.0_linux-aarch64)
  • 仓库:cann/ops-nn @ 711a4f5dda0d(norm/rms_norm_quant),2026-10-26

复现步骤

  1. 把下面「复现脚本」保存为 repro.cpp;
  2. 编译:
g++ -O2 -std=c++17 repro.cpp -o repro -I$ASCEND_HOME_PATH/include \
    -L$ASCEND_HOME_PATH/lib64 -lascendcl -lnnopbase -lopapi
  1. 运行对照:./repro ctrl 0、./repro xyrank3 0;
  2. 读两行返回码(第一段 GetWorkspaceSize 与第二段 RmsNormQuant)。

复现脚本

#include <cstdio>
#include <cstdlib>
#include <string>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_rms_norm_quant.h"

static void* MkDev(size_t bytes) { void* p = nullptr; aclrtMalloc(&p, bytes, ACL_MEM_MALLOC_HUGE_FIRST); return p; }

static aclTensor* MkTen(const std::vector<int64_t>& sh, aclDataType dt, void* dev, size_t esz) {
  int64_t n = 1; for (auto v : sh) n *= v;
  std::vector<int64_t> st(sh.size(), 1);
  for (int i = (int)sh.size() - 2; i >= 0; --i) st[i] = sh[i + 1] * st[i + 1];
  return aclCreateTensor(sh.data(), sh.size(), dt, st.data(), 0, ACL_FORMAT_ND, sh.data(), sh.size(), dev);
}

int main(int argc, char** argv) {
  std::string sc = argc > 1 ? argv[1] : "ctrl";
  std::vector<int64_t> xs{3,5}, gs{1,5}, bs{1,5}, ss{1}, os{1}, ys{3,5};
  aclDataType ydt = ACL_INT8;
  if (sc == "gamma2d") gs = {3,5};
  else if (sc == "scale2d") { ss = {1,1}; os = {1,1}; }
  else if (sc == "scalelen3") { ss = {3}; os = {3}; }
  else if (sc == "empty") { xs = {0,5}; ys = {0,5}; }
  else if (sc == "yint4") { xs = {3,16}; gs = {1,16}; bs = {1,16}; ys = {3,16}; ydt = ACL_INT4; }
  else if (sc == "yint32") { xs = {3,40}; gs = {1,40}; bs = {1,40}; ys = {3,5}; ydt = ACL_INT32; }
  else if (sc == "xyrank3") { xs = {1,3,5}; ys = {3,5}; }
  if (aclInit(nullptr) != ACL_SUCCESS) return 1;
  if (aclrtSetDevice(atoi(argc > 2 ? argv[2] : "0")) != ACL_SUCCESS) return 1;
  aclrtStream stm = nullptr; aclrtCreateStream(&stm);
  auto num = [](const std::vector<int64_t>& s) { int64_t n = 1; for (auto v : s) n *= v; return (int)(n > 0 ? n : 1); };
  int nx = num(xs), ng = num(gs), nb = num(bs), ns = num(ss), no = num(os), ny = num(ys);
  std::vector<uint16_t> xh(nx), gh(ng), bh(nb), sh(ns);
  for (auto& v : xh) v = aclFloatToFloat16(-1.5f);
  for (auto& v : gh) v = aclFloatToFloat16(1.0f);
  for (auto& v : bh) v = aclFloatToFloat16(0.25f);
  for (int i = 0; i < ns; ++i) sh[i] = aclFloatToFloat16(sc == "scalelen3" ? std::vector<float>{0.01f,0.02f,0.05f}[i] : 0.01f);
  std::vector<int8_t> oh(no, 2), yh(ny * (ydt == ACL_INT4 ? 1 : 4), 0);
  void *dx = MkDev(nx*2), *dg = MkDev(ng*2), *db = MkDev(nb*2), *ds = MkDev(ns*2), *dof = MkDev(no), *dy = MkDev(yh.size());
  aclrtMemcpy(dx, nx*2, xh.data(), nx*2, ACL_MEMCPY_HOST_TO_DEVICE);
  aclrtMemcpy(dg, ng*2, gh.data(), ng*2, ACL_MEMCPY_HOST_TO_DEVICE);
  aclrtMemcpy(db, nb*2, bh.data(), nb*2, ACL_MEMCPY_HOST_TO_DEVICE);
  aclrtMemcpy(ds, ns*2, sh.data(), ns*2, ACL_MEMCPY_HOST_TO_DEVICE);
  aclrtMemcpy(dof, no, oh.data(), no, ACL_MEMCPY_HOST_TO_DEVICE);
  aclTensor* x = MkTen(xs, ACL_FLOAT16, dx, 2); aclTensor* g = MkTen(gs, ACL_FLOAT16, dg, 2);
  aclTensor* b = MkTen(bs, ACL_FLOAT16, db, 2); aclTensor* s = MkTen(ss, ACL_FLOAT16, ds, 2);
  aclTensor* o = MkTen(os, ACL_INT8, dof, 1);  aclTensor* y = MkTen(ys, ydt, dy, 1);
  uint64_t ws = 0; aclOpExecutor* ex = nullptr;
  auto r1 = aclnnRmsNormQuantGetWorkspaceSize(x, g, b, s, o, 1e-6, y, &ws, &ex);
  printf("[%s] GetWorkspaceSize = %d\n", sc.c_str(), (int)r1);
  if (r1 == 0) {
    void* p = ws > 0 ? MkDev((size_t)ws) : nullptr;
    auto r2 = aclnnRmsNormQuant(p, ws, ex, stm);
    printf("[%s] RmsNormQuant = %d\n", sc.c_str(), (int)r2);
    if (r2 == 0) {
      aclrtSynchronizeStream(stm);
      std::vector<int8_t> out(ny, 0);
      aclrtMemcpy(out.data(), ny, dy, ny, ACL_MEMCPY_DEVICE_TO_HOST);
      printf("  y[0..7]="); for (int i = 0; i < ny && i < 8; ++i) printf("%d ", out[i]); printf("\n");
    }
    if (p) aclrtFree(p);
  }
  aclFinalize();
  return 0;
}

实测输出:

[xyrank3] GetWorkspaceSize = 561103
(同名程序另测 `./repro yint32`/对照场景时,末维不等形态为 161002)

影响

同 B1:非法 shape 未在契约面被拦下,错误码语义类别与资料不符;调用方无法据返回码定位是自己传错。规避方式:调用方自行保证 y.shape == x.shape。

建议

CheckShapeDim 的非 regbase 分支只做 xLastDim == outLastDim,前置维比对(CheckShapeDimWithFrontN)只在 IsRegbase() 分支执行。建议在非 regbase 分支同样比对前 n-1 维,命中时返回 161002。

likedislike
liuhanwen8liuhanwen8
26 days ago Added label: bug-report
yuning_chen
yuning_chenCollaborators
26 days ago Comments:

初步分析

感谢反馈,我们已开始核对这个问题。

现象理解:缺陷反馈;aclnnRmsNormQuant x/y 前置维不一致时报 561103 而非 161002(仅比对末维)。
当前判断:从描述看,这是缺陷反馈「aclnnRmsNormQuant x/y 前置维不一致时报 561103 而非 161002(仅比对末维)」(正文涉及返回码 561103 / 161002)。正文已给出对照用例与期望/实际差异,更像参数校验、错误码语义或静默错值类问题。以上为初步判断,待复现确认,不能当成已确认根因。
涉及范围:aclnnRmsNormQuant、aclnnop、aclnn_rms_norm_quant
下一步:我们会对照资料契约与实现路径继续定位,并按需补充回归用例。

likedislike
yuning_chenyuning_chenCollaborators
25 days ago Assign liujie12345678 as responsible person
yuning_chen
yuning_chenCollaborators
25 days ago Comments:

您好,问题已收到,当前 @liujie12345678 正在处理中。

likedislike
liujie12345678liujie12345678Collaborators
25 days ago Assign liuyun_nj as responsible person
liujie12345678liujie12345678Collaborators
25 days ago Removed responsible person liujie12345678
liuyun_njliuyun_njCollaborators
17 days ago changed custom state from 进行中 to 已完成
liuyun_njliuyun_njCollaborators
17 days ago Close issue
chenjiayi_
chenjiayi_Collaborators
11 days ago Comments:

补充说明本问题的修复结果:

结论:x / y 前置维一致性校验已覆盖 regbase 与 A2/A3 路径,前置维存在差异时第一段返回 161002,错误码语义与资料一致。

修复内容(文件/函数):

  • norm/rms_norm_quant/op_host/op_api/aclnn_rms_norm_quant.cpp 的 CheckShapeDim():
    • DT_INT32 输出分支:CheckShapeDimWithFrontN(x, y, xDimNum - 1) 外层增加所有平台的统一校验,前 n-1 维在 regbase 与 A2/A3 上均参与比对。
    • 其余输出分支:新增 CheckShapeDimWithFrontN(x, y, xDimNum),按全部维度比对,存在差异时报 x shape should be the same as y shape。
    • 两个分支的 OP_LOGE 改为 OP_LOGE_FOR_INVALID_SHAPES_WITH_REASON 等结构化错误宏,返回码为 ACLNN_ERR_PARAM_INVALID(161002)。
  • A2/A3 路径在 CheckParams 通过后遇空 Tensor 直接返回 workspaceSize=0 与 ACLNN_SUCCESS。

验证:

  • 修复后 x=[1,3,5]、y=[3,5] 在第一段被拒绝并返回 161002;
  • 合法基线 x=[3,5]、y=[3,5] 两段均返回 0;
  • MR 中 UT 已更新,覆盖 x/y 前置维存在差异的新场景。

该 issue 已随修复合入关闭。MR 链接:PR#13356

如果还有后续需要,请重新提 issue。

likedislike
chenjiayi_chenjiayi_Collaborators
9 days ago Assign chenjiayi_ as responsible person
chenjiayi_chenjiayi_Collaborators
1 day ago Removed responsible person chenjiayi_