Skip to contentSkip to conversation

remove implicit sign bit in msp3400.h

PATCH Applied Darren Jenkins\ 4 participants 8 messages Feb 8, 2006 – Feb 19, 2006

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

PATCH 1/1 remove implicit sign bit in msp3400.h

G'day

the two fields here

	int restart:1;
	int watch_stereo:1;

they both seem to be used as single bit flags. So this patch just makes them unsigned.

Note : Not sure if something should be done in msp3400-driver.c as "int msp_sleep(struct msp_state *state, int timeout)"
is declared to return an int then returns
"return state->restart;" which is now "unsigned:1;"
This seems ok but I don't know if gcc will issue a warning without a cast.
Note: if anyone can tell me if it's possible/how to compile msp3400-driver.c by its self that would be handy.

drivers/media/video/msp3400.h renamed from drivers/media/video/msp3400.h.orig +2 -2
@@ -86,8 +86,8 @@ struct msp_state {
 	/* thread */
 	struct task_struct   *kthread;
 	wait_queue_head_t    wq;
-	int                  restart:1;
-	int                  watch_stereo:1;
+	unsigned restart:1;
+	unsigned watch_stereo:1;
 };
 
 /* msp3400-driver.c */

_______________________________________________
Kernel-janitors mailing list
[email protected]
https://lists.osdl.org/mailman/listinfo/kernel-janitors

#1 lore raw

PATCH 1/1 remove implicit sign bit in l64781.c

G'day

Below is a patch which removes the implicit sign bit on a single bit field. The field 'first' is used as a flag to indicate a first attempt at init. So the patch just converts it to unsigned.

drivers/media/dvb/frontends/l64781.c renamed from drivers/media/dvb/frontends/l64781.c.orig +1 -1
@@ -37,7 +37,7 @@ struct l64781_state {
 	struct dvb_frontend frontend;
 
 	/* private demodulator data */
-	int first:1;
+	unsigned first:1;
 };
 
 #define dprintk(args...) \

_______________________________________________
Kernel-janitors mailing list
[email protected]
https://lists.osdl.org/mailman/listinfo/kernel-janitors

#2 lore raw

PATCH 1/1 remove implicit sign bit in isdnhdlc.h

In reply to Darren Jenkins\

G'day list.

This patch removes the implicit sign bit in a couple of one bit struct fields.
The fields here are used as one bit flags. So the patch just makes them unsigned. As a bonus it converts spaces to a tab before the last field.

drivers/isdn/hisax/isdnhdlc.h renamed from drivers/isdn/hisax/isdnhdlc.h.orig +4 -4
@@ -41,10 +41,10 @@ struct isdnhdlc_vars {
 	unsigned char shift_reg;
 	unsigned char ffvalue;
 
-	int data_received:1; 	// set if transferring data
-	int dchannel:1; 	// set if D channel (send idle instead of flags)
-	int do_adapt56:1; 	// set if 56K adaptation
-        int do_closing:1; 	// set if in closing phase (need to send CRC + flag
+	unsigned data_received:1; 	// set if transferring data
+	unsigned dchannel:1; 	// set if D channel (send idle instead of flags)
+	unsigned do_adapt56:1; 	// set if 56K adaptation
+	unsigned do_closing:1; 	// set if in closing phase (need to send CRC + flag
 };
 
 

_______________________________________________
Kernel-janitors mailing list
[email protected]
https://lists.osdl.org/mailman/listinfo/kernel-janitors

#3 lore raw
In reply to Darren Jenkins\
Quoted 24 lines · Darren Jenkins\

G'day

the two fields here

	int restart:1;
	int watch_stereo:1;

they both seem to be used as single bit flags. So this patch just makes them unsigned.

Note : Not sure if something should be done in msp3400-driver.c as "int msp_sleep(struct msp_state *state, int timeout)"
is declared to return an int then returns
"return state->restart;" which is now "unsigned:1;"
This seems ok but I don't know if gcc will issue a warning without a cast.
Note: if anyone can tell me if it's possible/how to compile msp3400-driver.c by its self that would be handy.

drivers/media/video/msp3400.h struct msp_state { · line 91

Quoted 8 lines
drivers/media/video/msp3400.h renamed from drivers/media/video/msp3400.h.orig +2 -2
@@ -86,8 +86,8 @@ struct msp_state {
 	/* thread */
 	struct task_struct   *kthread;
 	wait_queue_head_t    wq;
-	int                  restart:1;
-	int                  watch_stereo:1;
+	unsigned restart:1;
+	unsigned watch_stereo:1;
 };

Please keep the same spacing/tab/alignment.
And usually "unsigned int" is preferred in kernel source code.

Signature · 1 line
~Randy

_______________________________________________
Kernel-janitors mailing list
[email protected]
https://lists.osdl.org/mailman/listinfo/kernel-janitors

#4 lore raw
In reply to Darren Jenkins\
Quoted 1 line · Randy.Dunlap

Please keep the same spacing/tab/alignment.

Fair enough

Quoted 1 line

And usually "unsigned int" is preferred in kernel source code.

After doing a grep on the kernel I have to agree, most of them are "unsigned int foo :1;"
Weird if you ask me. I'm pretty sure the C standard says they are promoted to an int for calculations. But I suppose as it makes no difference it is better to stick with convention.

So here is a re-send of the patch.

drivers/media/video/msp3400.h renamed from drivers/media/video/msp3400.h.orig +2 -2
@@ -86,8 +86,8 @@ struct msp_state {
 	/* thread */
 	struct task_struct   *kthread;
 	wait_queue_head_t    wq;
-	int                  restart:1;
-	int                  watch_stereo:1;
+	unsigned int         restart:1;
+	unsigned int         watch_stereo:1;
 };
 
 /* msp3400-driver.c */

_______________________________________________
Kernel-janitors mailing list
[email protected]
https://lists.osdl.org/mailman/listinfo/kernel-janitors

#5 lore raw
In reply to Darren Jenkins\
Quoted 2 lines · Darren Jenkins

Weird if you ask me. I'm pretty sure the C standard says they are promoted to an int for calculations.

It's very logical.

	int a:1;

means 1 bit for sign (it's int) + >=1 bit for absolute value (it's int). So, >=2 bits you should somehow fit into 1 bit.

int | unsigned int |
-------------------+-------------------+---
doesn't make sense | OK |:1

	OK	   |	     OK	       |:2
	OK	   |	     OK	       |:3
			...

I guess C standard says "undefined behaviour" or "implementation-defined".

drivers/media/video/msp3400.h struct msp_state { · line 90

Quoted 13 lines

But I suppose as it makes no
difference it is better to stick with convention.

drivers/media/video/msp3400.h renamed from drivers/media/video/msp3400.h.orig +2 -2
@@ -86,8 +86,8 @@ struct msp_state {
 	/* thread */
 	struct task_struct   *kthread;
 	wait_queue_head_t    wq;
-	int                  restart:1;
-	int                  watch_stereo:1;
+	unsigned int         restart:1;
+	unsigned int         watch_stereo:1;

Applied.

BTW, someone could do

	int a:1;
	if (a < 0)

so every time type is flipped, submitter should check every usage of "a".

_______________________________________________
Kernel-janitors mailing list
[email protected]
https://lists.osdl.org/mailman/listinfo/kernel-janitors

#6 lore raw
In reply to Darren Jenkins\
Quoted 11 lines · Alexey Dobriyan
Quoted 2 lines · Darren Jenkins

Weird if you ask me. I'm pretty sure the C standard says they are promoted to an int for calculations.

It's very logical.

	int a:1;

means 1 bit for sign (it's int) + >=1 bit for absolute value (it's int). So, >=2 bits you should somehow fit into 1 bit.

Quoted 8 lines

int | unsigned int |
-------------------+-------------------+---
doesn't make sense | OK |:1

	OK	   |	     OK	       |:2
	OK	   |	     OK	       |:3
			...

I guess C standard says "undefined behaviour" or "implementation-defined".

Yep, that's how I understand it.

What I thought was weird was that the common coding style in the kernel was to include the int, when it is implicit. Where the rest of the style in the kernel seems to be to remove anything that is implicit. Just an observation.

Quoted 7 lines

BTW, someone could do

	int a:1;
	if (a < 0)

so every time type is flipped, submitter should check every usage of "a".

I did not think of that kind of usage. Will check it in future.

Darren J.

_______________________________________________
Kernel-janitors mailing list
[email protected]
https://lists.osdl.org/mailman/listinfo/kernel-janitors

#7 lore raw
In reply to Darren Jenkins\
Quoted 6 lines · Alexey Dobriyan

It's very logical.

	int a:1;

means 1 bit for sign (it's int) + >=1 bit for absolute value (it's int). So, >=2 bits you should somehow fit into 1 bit.

Standard 2's-complement representation. The MSB is negative. So instead of the bits meaning 1, 2, 4, 8, 16, 32, 64, 128, the top bit represents -128. So the one bit case represents -1.

ie the two values it can have are 0 and -1.

Quoted 1 line

I guess C standard says "undefined behaviour" or "implementation-defined".

It does ... sort of:

6.7.2:
for bit-fields, it is implementation-defined whether the specifier int designates the same type as signed int or the same type as unsigned int.

so relying on int to be signed was already broken.

Quoted 5 lines

BTW, someone could do

	int a:1;
	if (a < 0)
Quoted 7 lines

so every time type is flipped, submitter should check every usage of "a".

_______________________________________________
Kernel-janitors mailing list
[email protected]
https://lists.osdl.org/mailman/listinfo/kernel-janitors

_______________________________________________
Kernel-janitors mailing list
[email protected]
https://lists.osdl.org/mailman/listinfo/kernel-janitors