Re: [PATCH 3/3] ufs: core: Optimize ufshcd_add_uic_command_trace()

[Bart Van Assche <bvanassche@acm.org>]

On 4/22/26 12:52 AM, Peter Wang (王信友) wrote:
> Yes, you are right! In general cases, this is true.
> But for some error cases, we need this trace debug log to check
> if the hardware is working as the software expects.
> Maybe the hardware is stuck or something like that.
> Therefore, we still need to read the register values from
> the hardware.

Hi Peter,

I have started testing the patch below. This patch is intended as a
replacement for patch 3/3 in this series:

diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
index a44ef7e97125..76416ee88b25 100644
--- a/drivers/ufs/core/ufshcd.c
+++ b/drivers/ufs/core/ufshcd.c
@@ -460,20 +460,19 @@ static void ufshcd_add_uic_command_trace(struct 
ufs_hba *hba,
  					 const struct uic_command *ucmd,
  					 enum ufs_trace_str_t str_t)
  {
-	u32 cmd;
-
  	if (!trace_ufshcd_uic_command_enabled())
  		return;

  	if (str_t == UFS_CMD_SEND)
-		cmd = ucmd->command;
+		trace_ufshcd_uic_command(hba, str_t, ucmd->command,
+					 ucmd->argument1, ucmd->argument2,
+					 ucmd->argument3);
  	else
-		cmd = ufshcd_readl(hba, REG_UIC_COMMAND);
-
-	trace_ufshcd_uic_command(hba, str_t, cmd,
-				 ufshcd_readl(hba, REG_UIC_COMMAND_ARG_1),
-				 ufshcd_readl(hba, REG_UIC_COMMAND_ARG_2),
-				 ufshcd_readl(hba, REG_UIC_COMMAND_ARG_3));
+		trace_ufshcd_uic_command(
+			hba, str_t, ufshcd_readl(hba, REG_UIC_COMMAND),
+			ufshcd_readl(hba, REG_UIC_COMMAND_ARG_1),
+			ufshcd_readl(hba, REG_UIC_COMMAND_ARG_2),
+			ufshcd_readl(hba, REG_UIC_COMMAND_ARG_3));
  }

  static void ufshcd_add_command_trace(struct ufs_hba *hba, struct 
scsi_cmnd *cmd,