Hi Dan, Yes, you are right. It should be left shift to check the AE ctx enable. Thanks for your defect, I will send a patch to fix this issue. Best Regards, Pingchao yang -----Original Message----- From: Dan Carpenter [mailto:dan.carpenter@xxxxxxxxxx] Sent: Friday, December 11, 2015 6:03 AM To: Yang, Pingchao Cc: qat-linux; linux-crypto@xxxxxxxxxxxxxxx Subject: re: crypto: qat - add support for new devices to FW loader Hello Pingchao Yang, The patch b0272276d903: "crypto: qat - add support for new devices to FW loader" from Dec 4, 2015, leads to the following static checker warning: drivers/crypto/qat/qat_common/qat_hal.c:421 qat_hal_check_ae_active() warn: bitwise AND condition is false here drivers/crypto/qat/qat_common/qat_hal.c 414 int qat_hal_check_ae_active(struct icp_qat_fw_loader_handle *handle, 415 unsigned int ae) 416 { 417 unsigned int enable = 0, active = 0; 418 419 qat_hal_rd_ae_csr(handle, ae, CTX_ENABLES, &enable); 420 qat_hal_rd_ae_csr(handle, ae, ACTIVE_CTX_STATUS, &active); 421 if ((enable & (0xff >> CE_ENABLE_BITPOS)) || ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Never true. CE_ENABLE_BITPOS is 8. Perhaps the intention was to left shift like we do with ACS_ABO_BITPOS on the next line? 422 (active & (1 << ACS_ABO_BITPOS))) 423 return 1; 424 else 425 return 0; 426 } regards, dan carpenter -- To unsubscribe from this list: send the line "unsubscribe linux-crypto" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
