Skip to contentSkip to conversation

usb: remove redundant conditions

PATCH In review Geyslan G. Bem 4 participants 11 messages Dec 10, 2015

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 usb: remove redundant conditions

This patch removes redundant conditions.

- (!A || (A && B)) is the same as (!A || B).
- (length && length > 5) can be reduced to a single evaluation.

Caught by: cppcheck

---
 drivers/usb/gadget/udc/s3c-hsudc.c |  2 +-
 drivers/usb/host/fhci-sched.c      |  2 +-
 drivers/usb/musb/musb_gadget.c     |  5 ++---
 drivers/usb/serial/io_edgeport.c   | 35 ++++++++++++++---------------------
 drivers/usb/serial/mos7840.c       |  2 +-
 5 files changed, 19 insertions(+), 27 deletions(-)
drivers/usb/gadget/udc/s3c-hsudc.c +1 -1
index e9def42..82a9e2a 100644
@@ -569,7 +569,7 @@ static int s3c_hsudc_handle_reqfeat(struct s3c_hsudc *hsudc,
 		hsep = &hsudc->ep[ep_num];
 		switch (le16_to_cpu(ctrl->wValue)) {
 		case USB_ENDPOINT_HALT:
-			if (set || (!set && !hsep->wedge))
+			if (set || !hsep->wedge)
 				s3c_hsudc_set_halt(&hsep->ep, set);
 			return 0;
 		}
drivers/usb/host/fhci-sched.c +1 -1
index 95ca598..23ecac5 100644
@@ -288,7 +288,7 @@ static int scan_ed_list(struct fhci_usb *usb,
 	list_for_each_entry(ed, list, node) {
 		td = ed->td_head;
 
-		if (!td || (td && td->status == USB_TD_INPROGRESS))
+		if (!td || (td->status == USB_TD_INPROGRESS))
 			continue;
 
 		if (ed->state != FHCI_ED_OPER) {
drivers/usb/musb/musb_gadget.c +2 -3
index 67ad630..87bd578 100644
@@ -353,9 +353,8 @@ static void txstate(struct musb *musb, struct musb_request *req)
 					 *	1	>0	Yes(FS bulk)
 					 */
 					if (!musb_ep->hb_mult ||
-						(musb_ep->hb_mult &&
-						 can_bulk_split(musb,
-						    musb_ep->type)))
+					    can_bulk_split(musb,
+							   musb_ep->type))
 						csr |= MUSB_TXCSR_AUTOSET;
 				}
 				csr &= ~MUSB_TXCSR_P_UNDERRUN;
drivers/usb/serial/io_edgeport.c +14 -21
index c086697..f49327d 100644
@@ -1046,9 +1046,8 @@ static void edge_close(struct usb_serial_port *port)
 
 	edge_port->closePending = true;
 
-	if ((!edge_serial->is_epic) ||
-	    ((edge_serial->is_epic) &&
-	     (edge_serial->epic_descriptor.Supports.IOSPChase))) {
+	if (!edge_serial->is_epic ||
+	    edge_serial->epic_descriptor.Supports.IOSPChase) {
 		/* flush and chase */
 		edge_port->chaseResponsePending = true;
 
@@ -1061,9 +1060,8 @@ static void edge_close(struct usb_serial_port *port)
 			edge_port->chaseResponsePending = false;
 	}
 
-	if ((!edge_serial->is_epic) ||
-	    ((edge_serial->is_epic) &&
-	     (edge_serial->epic_descriptor.Supports.IOSPClose))) {
+	if (!edge_serial->is_epic ||
+	    edge_serial->epic_descriptor.Supports.IOSPClose) {
 	       /* close the port */
 		dev_dbg(&port->dev, "%s - Sending IOSP_CMD_CLOSE_PORT\n", __func__);
 		send_iosp_ext_cmd(edge_port, IOSP_CMD_CLOSE_PORT, 0);
@@ -1612,9 +1610,8 @@ static void edge_break(struct tty_struct *tty, int break_state)
 	struct edgeport_serial *edge_serial = usb_get_serial_data(port->serial);
 	int status;
 
-	if ((!edge_serial->is_epic) ||
-	    ((edge_serial->is_epic) &&
-	     (edge_serial->epic_descriptor.Supports.IOSPChase))) {
+	if (!edge_serial->is_epic ||
+	    edge_serial->epic_descriptor.Supports.IOSPChase) {
 		/* flush and chase */
 		edge_port->chaseResponsePending = true;
 
@@ -1628,9 +1625,8 @@ static void edge_break(struct tty_struct *tty, int break_state)
 		}
 	}
 
-	if ((!edge_serial->is_epic) ||
-	    ((edge_serial->is_epic) &&
-	     (edge_serial->epic_descriptor.Supports.IOSPSetClrBreak))) {
+	if (!edge_serial->is_epic ||
+	    edge_serial->epic_descriptor.Supports.IOSPSetClrBreak) {
 		if (break_state == -1) {
 			dev_dbg(&port->dev, "%s - Sending IOSP_CMD_SET_BREAK\n", __func__);
 			status = send_iosp_ext_cmd(edge_port,
@@ -2465,9 +2461,8 @@ static void change_port_settings(struct tty_struct *tty,
 		unsigned char stop_char  = STOP_CHAR(tty);
 		unsigned char start_char = START_CHAR(tty);
 
-		if ((!edge_serial->is_epic) ||
-		    ((edge_serial->is_epic) &&
-		     (edge_serial->epic_descriptor.Supports.IOSPSetXChar))) {
+		if (!edge_serial->is_epic ||
+		    edge_serial->epic_descriptor.Supports.IOSPSetXChar) {
 			send_iosp_ext_cmd(edge_port,
 					IOSP_CMD_SET_XON_CHAR, start_char);
 			send_iosp_ext_cmd(edge_port,
@@ -2494,13 +2489,11 @@ static void change_port_settings(struct tty_struct *tty,
 	}
 
 	/* Set flow control to the configured value */
-	if ((!edge_serial->is_epic) ||
-	    ((edge_serial->is_epic) &&
-	     (edge_serial->epic_descriptor.Supports.IOSPSetRxFlow)))
+	if (!edge_serial->is_epic ||
+	    edge_serial->epic_descriptor.Supports.IOSPSetRxFlow)
 		send_iosp_ext_cmd(edge_port, IOSP_CMD_SET_RX_FLOW, rxFlow);
-	if ((!edge_serial->is_epic) ||
-	    ((edge_serial->is_epic) &&
-	     (edge_serial->epic_descriptor.Supports.IOSPSetTxFlow)))
+	if (!edge_serial->is_epic ||
+	    edge_serial->epic_descriptor.Supports.IOSPSetTxFlow)
 		send_iosp_ext_cmd(edge_port, IOSP_CMD_SET_TX_FLOW, txFlow);
 
 
drivers/usb/serial/mos7840.c +1 -1
index 8ac9b55..2c69bfc 100644
@@ -635,7 +635,7 @@ static void mos7840_interrupt_callback(struct urb *urb)
 	 * Byte 4 IIR Port 4 (port.number is 3)
 	 * Byte 5 FIFO status for both */
 
-	if (length && length > 5) {
+	if (length > 5) {
 		dev_dbg(&urb->dev->dev, "%s", "Wrong data !!!\n");
 		return;
 	}
Signature · 1 line
2.6.3
maintainer
#1 lore raw
Quoted 7 lines · Geyslan G. Bem

This patch removes redundant conditions.

- (!A || (A && B)) is the same as (!A || B).
- (length && length > 5) can be reduced to a single evaluation.

Caught by: cppcheck

Quoted 8 lines
---
 drivers/usb/gadget/udc/s3c-hsudc.c |  2 +-
 drivers/usb/host/fhci-sched.c      |  2 +-
 drivers/usb/musb/musb_gadget.c     |  5 ++---
 drivers/usb/serial/io_edgeport.c   | 35 ++++++++++++++---------------------
 drivers/usb/serial/mos7840.c       |  2 +-
 5 files changed, 19 insertions(+), 27 deletions(-)

Please split out (at least) the usb-serial changes into two patches.

Johan

#2 lore raw

2015-12-10 7:44 GMT-03:00 Johan Hovold <[email protected]>:

Quoted 17 lines · Geyslan G. Bem
Quoted 7 lines

This patch removes redundant conditions.

- (!A || (A && B)) is the same as (!A || B).
- (length && length > 5) can be reduced to a single evaluation.

Caught by: cppcheck

Quoted 8 lines
---
 drivers/usb/gadget/udc/s3c-hsudc.c |  2 +-
 drivers/usb/host/fhci-sched.c      |  2 +-
 drivers/usb/musb/musb_gadget.c     |  5 ++---
 drivers/usb/serial/io_edgeport.c   | 35 ++++++++++++++---------------------
 drivers/usb/serial/mos7840.c       |  2 +-
 5 files changed, 19 insertions(+), 27 deletions(-)
Quoted 1 line

Please split out (at least) the usb-serial changes into two patches.

Ok, I'll send v2 later.

Quoted 2 lines

Johan

Signature · 4 lines
Regards,

Geyslan G. Bem
hackingbits.com
#3 lore raw
In reply to Geyslan G. Bem

Hi,

Quoted 8 lines · Geyslan G. Bem

This patch removes redundant conditions.

- (!A || (A && B)) is the same as (!A || B).
- (length && length > 5) can be reduced to a single evaluation.

Caught by: cppcheck

Can you split this into one patch per driver, that way it's easier to apply by different maintainers.

Signature · 1 line
balbi
#4 lore raw

2015-12-10 11:50 GMT-03:00 Felipe Balbi <[email protected]>:

Quoted 9 lines

Hi,

Quoted 5 lines · Geyslan G. Bem

This patch removes redundant conditions.

- (!A || (A && B)) is the same as (!A || B).
- (length && length > 5) can be reduced to a single evaluation.

Quoted 6 lines
Quoted 3 lines

Caught by: cppcheck

Can you split this into one patch per driver, that way it's easier to apply by different maintainers.

Balbi, sure, I can.

Tks.

Quoted 3 lines

--
balbi

Signature · 4 lines
Regards,

Geyslan G. Bem
hackingbits.com
#5 lore raw
In reply to Geyslan G. Bem
Quoted 9 lines · Geyslan G. Bem

This patch removes redundant conditions.

- (!A || (A && B)) is the same as (!A || B).
- (length && length > 5) can be reduced to a single evaluation.

Caught by: cppcheck

---

I guess you didn't get previous comment in time; let's split this per driver so different maintainers can pick their parts.

Signature · 1 line
balbi
#6 lore raw

2015-12-10 12:13 GMT-03:00 Felipe Balbi <[email protected]>:

Quoted 7 lines
Quoted 5 lines · Geyslan G. Bem

This patch removes redundant conditions.

- (!A || (A && B)) is the same as (!A || B).
- (length && length > 5) can be reduced to a single evaluation.

Quoted 7 lines
Quoted 4 lines

Caught by: cppcheck

---

I guess you didn't get previous comment in time; let's split this per driver so different maintainers can pick their parts.

Okeydokey. :)

Quoted 3 lines

--
balbi

Signature · 4 lines
Regards,

Geyslan G. Bem
hackingbits.com
#7 lore raw

Hi,

Quoted 10 lines · Geyslan G. Bem

2015-12-10 12:13 GMT-03:00 Felipe Balbi <[email protected]>:

Quoted 9 lines
Quoted 7 lines · Geyslan G. Bem

This patch removes redundant conditions.

- (!A || (A && B)) is the same as (!A || B).
- (length && length > 5) can be reduced to a single evaluation.

Caught by: cppcheck

Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

---

I guess you didn't get previous comment in time; let's split this per driver so different maintainers can pick their parts.

Okeydokey. :)

thanks :-)

Signature · 1 line
balbi
#8 lore raw
In reply to Felipe Balbi

From: Felipe Balbi

Quoted 8 lines

Sent: 10 December 2015 15:14

Quoted 5 lines · Geyslan G. Bem

This patch removes redundant conditions.

- (!A || (A && B)) is the same as (!A || B).
- (length && length > 5) can be reduced to a single evaluation.

Quoted 7 lines
Quoted 4 lines

Caught by: cppcheck

---

I guess you didn't get previous comment in time; let's split this per driver so different maintainers can pick their parts.

I also suspect that gcc will optimise out the redundant checks as well.

David

#9 lore raw

2015-12-10 12:17 GMT-03:00 David Laight <[email protected]>:

Quoted 11 lines

From: Felipe Balbi

Quoted 10 lines

Sent: 10 December 2015 15:14

Quoted 7 lines · Geyslan G. Bem

This patch removes redundant conditions.

- (!A || (A && B)) is the same as (!A || B).
- (length && length > 5) can be reduced to a single evaluation.

Caught by: cppcheck

Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

---

I guess you didn't get previous comment in time; let's split this per driver so different maintainers can pick their parts.

I also suspect that gcc will optimise out the redundant checks as well.

Yes, David. it will.

Let's see.

void f(int f, int s)

{
        if (!f || (f && s))
                printf("branch\n");
}

Generates without optimization three comparisons:

        cmpl    $0, -4(%rbp)
        je      .L2
        cmpl    $0, -4(%rbp)
        je      .L4
        cmpl    $0, -8(%rbp)
        je      .L4

But with -O2 it generates only two:

        testl   %edi, %edi
        je      .L2
        testl   %esi, %esi
        je      .L1

Despite that, I think that the patches are welcome since they silence checkpatch and make code clearer. Don't you think?

Quoted 3 lines

David

Signature · 4 lines
Regards,

Geyslan G. Bem
hackingbits.com
#10 lore raw

2015-12-10 12:29 GMT-03:00 Geyslan G. Bem <[email protected]>:

Quoted 40 lines

2015-12-10 12:17 GMT-03:00 David Laight <[email protected]>:

Quoted 11 lines

From: Felipe Balbi

Quoted 10 lines

Sent: 10 December 2015 15:14

Quoted 7 lines · Geyslan G. Bem

This patch removes redundant conditions.

- (!A || (A && B)) is the same as (!A || B).
- (length && length > 5) can be reduced to a single evaluation.

Caught by: cppcheck

Quoted 7 lines
Quoted 5 lines
Quoted 2 lines

---

I guess you didn't get previous comment in time; let's split this per driver so different maintainers can pick their parts.

I also suspect that gcc will optimise out the redundant checks as well.

Yes, David. it will.

Let's see.

void f(int f, int s)

{
        if (!f || (f && s))
                printf("branch\n");
}

Generates without optimization three comparisons:

        cmpl    $0, -4(%rbp)
        je      .L2
        cmpl    $0, -4(%rbp)
        je      .L4
        cmpl    $0, -8(%rbp)
        je      .L4

But with -O2 it generates only two:

Quoted 7 lines
        testl   %edi, %edi
        je      .L2
        testl   %esi, %esi
        je      .L1

Despite that, I think that the patches are welcome since they silence checkpatch and make code clearer. Don't you think?

Sorry, silence "cppcheck". I'm mistaking the tool name.

Quoted 12 lines
Quoted 4 lines

David

--
Regards,

Geyslan G. Bem
hackingbits.com

Signature · 4 lines
Regards,

Geyslan G. Bem
hackingbits.com