Skip to contentSkip to conversation

drivers/tty: delete break after return or goto

PATCH In review Bernard Zhao 3 participants 6 messages Nov 7, 2020 – Nov 9, 2020

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

cover drivers/tty: delete break after return or goto

Hi, Greg:

This patch sereies delete code which never run:
{
case XXX:

	return XXX;
	break; //The break is meanless, so just delete it.
case YYY:
	goto YYY;
	break; //The break is meanless, so just delete it.

......
}

Bernard Zhao (2):
tty/serial: delete break after return
drivers/tty: delete break after goto/return

 drivers/tty/nozomi.c     | 4 ----
 drivers/tty/serial/imx.c | 5 -----
 2 files changed, 9 deletions(-)
Signature · 7 lines
2.29.0


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
#1 lore raw

PATCH 1/2 tty/serial: delete break after return

Delete break after return, which will never run.

---
 drivers/tty/serial/imx.c | 5 -----
 1 file changed, 5 deletions(-)
drivers/tty/serial/imx.c +0 -5
index 1731d9728865..09703079db7b 100644
@@ -320,7 +320,6 @@ static u32 imx_uart_readl(struct imx_port *sport, u32 offset)
 	switch (offset) {
 	case UCR1:
 		return sport->ucr1;
-		break;
 	case UCR2:
 		/*
 		 * UCR2_SRST is the only bit in the cached registers that might
@@ -331,16 +330,12 @@ static u32 imx_uart_readl(struct imx_port *sport, u32 offset)
 		if (!(sport->ucr2 & UCR2_SRST))
 			sport->ucr2 = readl(sport->port.membase + offset);
 		return sport->ucr2;
-		break;
 	case UCR3:
 		return sport->ucr3;
-		break;
 	case UCR4:
 		return sport->ucr4;
-		break;
 	case UFCR:
 		return sport->ufcr;
-		break;
 	default:
 		return readl(sport->port.membase + offset);
 	}
Signature · 7 lines
2.29.0


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
#2 lore raw

Re: [PATCH 1/2] tty/serial: delete break after return

Hello,

the Subject is wrong, it should use a prefix similar to "serial: imx:". It's a good idea to check previous patches to the same file to pick a suitable prefix. (E.g. git log --oneline drivers/tty/serial/imx.c)

Quoted 23 lines · Bernard Zhao

Delete break after return, which will never run.

---
 drivers/tty/serial/imx.c | 5 -----
 1 file changed, 5 deletions(-)
drivers/tty/serial/imx.c +0 -1
index 1731d9728865..09703079db7b 100644
@@ -320,7 +320,6 @@ static u32 imx_uart_readl(struct imx_port *sport, u32 offset)
 	switch (offset) {
 	case UCR1:
 		return sport->ucr1;
-		break;
 	case UCR2:
 		/*
 		 * UCR2_SRST is the only bit in the cached registers that might
@@ -331,16 +330,12 @@ static u32 imx_uart_readl(struct imx_port *sport, u32 offset)
 		if (!(sport->ucr2 & UCR2_SRST))
 			sport->ucr2 = readl(sport->port.membase + offset);
 		return sport->ucr2;

drivers/tty/serial/imx.c static u32 imx_uart_readl(struct imx_port *sport, u32 offset) · line 343

Quoted 10 lines
drivers/tty/serial/imx.c +0 -4
-		break;
 	case UCR3:
 		return sport->ucr3;
-		break;
 	case UCR4:
 		return sport->ucr4;
-		break;
 	case UFCR:
 		return sport->ufcr;
-		break;

you're the third to send this patch since October 20:

	https://lore.kernel.org/r/[email protected]
	https://lore.kernel.org/r/[email protected]

My comment for these was:

Quoted 3 lines

this might be subjective, but I like the break being there for clearity. So I object to make a patch to remove them. In case I'm outvoted I'd at least want empty lines instead.

Zhang Qilong wrote he found the patch opportunity by manual code inspection, I would have expected that there is a tool that identifies a break after a return. If you had tool support, please mention the tool in the commit log (if you really want to keep following the patch's idea).

Best regards
Uwe

Signature · 2 lines
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | https://www.pengutronix.de/ |
author
#3 lore raw

Re:Re: [PATCH 1/2] tty/serial: delete break after return

From: "Uwe Kleine-König" <[email protected]>
Date: 2020-11-07 22:01:29
To: Bernard Zhao <[email protected]>
Cc: Greg Kroah-Hartman <[email protected]>,Jiri Slaby <[email protected]>,Shawn Guo <[email protected]>,Sascha Hauer <[email protected]>,Pengutronix Kernel Team <[email protected]>,Fabio Estevam <[email protected]>,NXP Linux Team <[email protected]>,[email protected],[email protected],[email protected],[email protected]
Subject: Re: [PATCH 1/2] tty/serial: delete break after return>Hello,

Quoted 4 lines

the Subject is wrong, it should use a prefix similar to "serial: imx:". It's a good idea to check previous patches to the same file to pick a suitable prefix. (E.g. git log --oneline drivers/tty/serial/imx.c)

Hi, Uwe:

Thank you for your suggestion, I will make a more accurate subject in my future patches.

Quoted 46 lines · Bernard Zhao
Quoted 23 lines

Delete break after return, which will never run.

---
 drivers/tty/serial/imx.c | 5 -----
 1 file changed, 5 deletions(-)
drivers/tty/serial/imx.c +0 -1
index 1731d9728865..09703079db7b 100644
@@ -320,7 +320,6 @@ static u32 imx_uart_readl(struct imx_port *sport, u32 offset)
 	switch (offset) {
 	case UCR1:
 		return sport->ucr1;
-		break;
 	case UCR2:
 		/*
 		 * UCR2_SRST is the only bit in the cached registers that might
@@ -331,16 +330,12 @@ static u32 imx_uart_readl(struct imx_port *sport, u32 offset)
 		if (!(sport->ucr2 & UCR2_SRST))
 			sport->ucr2 = readl(sport->port.membase + offset);
 		return sport->ucr2;

drivers/tty/serial/imx.c static u32 imx_uart_readl(struct imx_port *sport, u32 offset) · line 343

Quoted 10 lines
drivers/tty/serial/imx.c +0 -4
-		break;
 	case UCR3:
 		return sport->ucr3;
-		break;
 	case UCR4:
 		return sport->ucr4;
-		break;
 	case UFCR:
 		return sport->ufcr;
-		break;

you're the third to send this patch since October 20:

	https://lore.kernel.org/r/[email protected]
	https://lore.kernel.org/r/[email protected]

My comment for these was:

Quoted 3 lines

this might be subjective, but I like the break being there for clearity. So I object to make a patch to remove them. In case I'm outvoted I'd at least want empty lines instead.

Quoted 5 lines

Zhang Qilong wrote he found the patch opportunity by manual code inspection, I would have expected that there is a tool that identifies a break after a return. If you had tool support, please mention the tool in the commit log (if you really want to keep following the patch's idea).

For this part:
I wrote a python script to check if there is a break after a return or goto.
The script currently has some issues in its handling of special characters, the search result is not precise enough. It is still under debugging. These are the only places which the script checks out in path: driver/tty.
I also checked other path(like: driver/vme/, driver/video, driver/usb), these paths also have a certain number of similar issues. And I will try to submit these patch later. Thanks!

BR//Bernard

Quoted 6 lines

Best regards
Uwe

Signature · 2 lines
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | https://www.pengutronix.de/ |
#4 lore raw

PATCH 2/2 drivers/tty: delete break after goto/return

In reply to Bernard Zhao

Delete break after goto/return, which will never run.

---
 drivers/tty/nozomi.c | 4 ----
 1 file changed, 4 deletions(-)
drivers/tty/nozomi.c +0 -4
index d42b854cb7df..946cc16827aa 100644
@@ -414,11 +414,9 @@ static void read_mem32(u32 *buf, const void __iomem *mem_addr_start,
 		buf16 = (u16 *) buf;
 		*buf16 = __le16_to_cpu(readw(ptr));
 		goto out;
-		break;
 	case 4:	/* 4 bytes */
 		*(buf) = __le32_to_cpu(readl(ptr));
 		goto out;
-		break;
 	}
 
 	while (i < size_bytes) {
@@ -460,7 +458,6 @@ static u32 write_mem32(void __iomem *mem_addr_start, const u32 *buf,
 		buf16 = (const u16 *)buf;
 		writew(__cpu_to_le16(*buf16), ptr);
 		return 2;
-		break;
 	case 1: /*
 		 * also needs to write 4 bytes in this case
 		 * so falling through..
@@ -468,7 +465,6 @@ static u32 write_mem32(void __iomem *mem_addr_start, const u32 *buf,
 	case 4: /* 4 bytes */
 		writel(__cpu_to_le32(*buf), ptr);
 		return 4;
-		break;
 	}
 
 	while (i < size_bytes) {
Signature · 7 lines
2.29.0


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
#5 lore raw

Re: [PATCH 2/2] drivers/tty: delete break after goto/return

Quoted 6 lines · Bernard Zhao

Delete break after goto/return, which will never run.

---
 drivers/tty/nozomi.c | 4 ----
 1 file changed, 4 deletions(-)

If you look at the commits for this file:

	$ git log --oneline drivers/tty/nozomi.c | head -n 5
	1a460c36078e tty: nozomi: remove unneeded break
	caa47cc63947 tty: nozomi: Use scnprintf() for avoiding potential buffer overflow
	e2c2e7987106 tty: nozomi: fix spelling mistake "reserverd" -> "reserved"
	18b1345e60ae tty: nozomi: Use dev_get_drvdata
	c392ed464205 tty/nozomi: use pci_iomap instead of ioremap_nocache

You will notice that you should probably put the driver name in the subject line. Otherwise this patch really looks like you are doing this action on all of drivers/tty/ right?

Same for patch 1/2 as was pointed out by others.

thanks,

greg k-h