diff --git a/src/amd/common/sid.h b/src/amd/common/sid.h
index fe3351be5a5..62177e853e9 100644
--- a/src/amd/common/sid.h
+++ b/src/amd/common/sid.h
@@ -248,6 +248,7 @@
 #define PKT3_INCREMENT_CE_COUNTER                  0x84
 #define PKT3_INCREMENT_DE_COUNTER                  0x85
 #define PKT3_WAIT_ON_CE_COUNTER                    0x86
+#define PKT3_HDP_FLUSH                             0x95
 #define PKT3_SET_SH_REG_INDEX                      0x9B
 #define PKT3_LOAD_CONTEXT_REG_INDEX                0x9F /* GFX8+ */
 #define PKT3_DISPATCH_DIRECT_INTERLEAVED           0xA7 /* GFX12+ */
diff --git a/src/gallium/winsys/amdgpu/drm/amdgpu_cs.cpp b/src/gallium/winsys/amdgpu/drm/amdgpu_cs.cpp
index 8bb9d848ed6..ebe6201be87 100644
--- a/src/gallium/winsys/amdgpu/drm/amdgpu_cs.cpp
+++ b/src/gallium/winsys/amdgpu/drm/amdgpu_cs.cpp
@@ -1397,6 +1397,9 @@ static void amdgpu_cs_add_userq_packets(struct amdgpu_userq *userq,
          }
       }
 
+      amdgpu_pkt_add_dw(PKT3(PKT3_HDP_FLUSH, 0, 0));
+      amdgpu_pkt_add_dw(0x0);
+
       amdgpu_pkt_add_dw(PKT3(PKT3_INDIRECT_BUFFER, 2, 0));
       amdgpu_pkt_add_dw(cs->chunk_ib[IB_MAIN].va_start);
       amdgpu_pkt_add_dw(cs->chunk_ib[IB_MAIN].va_start >> 32);