[RFA] sh-tdep.c: Fix erroneus register skipping in sh_print_registers_info

[RFA] sh-tdep.c: Fix erroneus register skipping in sh_print_registers_info

[Corinna Vinschen]

Hi,

the below patch fixes a long standing bug in sh_print_registers_info.
When the print loop encounters a float type register and fpregs is
not set, then the loop counter (regnum) is not just incremented by
one, but instead it's incremented by FP_LAST_REGNUM - FP0_REGNUM.

The problem with this is, that FPUL is also a float type register.
FPUL is two register numbers below FP0_REGNUM.  So when the loop
arrives at FPUL, it skips the next 16 registers.  The next evaluated
register then (fr14) is still a float type register, so the loop
skips again 16 registers.  As a result, the register set is never
printed completely when calling `info registers'.

The below patch fixes that and shortens the code slightly by using
a "for" loop instead of "while".


Corinna


ChangeLog:

	* sh-tdep.c (sh_print_registers_info): Use for loop.
	Don't skip multiple registers when a float register is encountered.

Index: sh-tdep.c
===================================================================
RCS file: /cvs/src/src/gdb/sh-tdep.c,v
retrieving revision 1.156
diff -u -p -r1.156 sh-tdep.c
--- sh-tdep.c	26 Jan 2004 20:52:12 -0000	1.156
+++ sh-tdep.c	18 Feb 2004 13:41:21 -0000
@@ -1799,35 +1799,23 @@ sh_print_registers_info (struct gdbarch 
   else
     /* do all (or most) registers */
     {
-      regnum = 0;
-      while (regnum < NUM_REGS)
+      for (regnum = 0; regnum < NUM_REGS; ++regnum)
 	{
 	  /* If the register name is empty, it is undefined for this
 	     processor, so don't display anything.  */
 	  if (REGISTER_NAME (regnum) == NULL
 	      || *(REGISTER_NAME (regnum)) == '\0')
-	    {
-	      regnum++;
-	      continue;
-	    }
+	    continue;
 
 	  if (TYPE_CODE (gdbarch_register_type (gdbarch, regnum)) ==
 	      TYPE_CODE_FLT)
 	    {
+	      /* true for "INFO ALL-REGISTERS" command */
 	      if (fpregs)
-		{
-		  /* true for "INFO ALL-REGISTERS" command */
-		  sh_do_fp_register (gdbarch, file, regnum);	/* FP regs */
-		  regnum++;
-		}
-	      else
-		regnum += (FP_LAST_REGNUM - FP0_REGNUM);	/* skip FP regs */
+		sh_do_fp_register (gdbarch, file, regnum);	/* FP regs */
 	    }
 	  else
-	    {
-	      sh_do_register (gdbarch, file, regnum);	/* All other regs */
-	      regnum++;
-	    }
+	    sh_do_register (gdbarch, file, regnum);	/* All other regs */
 	}
 
       if (fpregs)

-- 
Corinna Vinschen
Cygwin Developer
Red Hat, Inc.

Re: [RFA] sh-tdep.c: Fix erroneus register skipping in sh_print_registers_info

[Corinna Vinschen]

Ping?

Corinna

On Feb 18 14:59, Corinna Vinschen wrote:
> Hi,
> 
> the below patch fixes a long standing bug in sh_print_registers_info.
> When the print loop encounters a float type register and fpregs is
> not set, then the loop counter (regnum) is not just incremented by
> one, but instead it's incremented by FP_LAST_REGNUM - FP0_REGNUM.
> 
> The problem with this is, that FPUL is also a float type register.
> FPUL is two register numbers below FP0_REGNUM.  So when the loop
> arrives at FPUL, it skips the next 16 registers.  The next evaluated
> register then (fr14) is still a float type register, so the loop
> skips again 16 registers.  As a result, the register set is never
> printed completely when calling `info registers'.
> 
> The below patch fixes that and shortens the code slightly by using
> a "for" loop instead of "while".
> 
> 
> Corinna
> 
> 
> ChangeLog:
> 
> 	* sh-tdep.c (sh_print_registers_info): Use for loop.
> 	Don't skip multiple registers when a float register is encountered.
> 
> Index: sh-tdep.c
> ===================================================================
> RCS file: /cvs/src/src/gdb/sh-tdep.c,v
> retrieving revision 1.156
> diff -u -p -r1.156 sh-tdep.c
> --- sh-tdep.c	26 Jan 2004 20:52:12 -0000	1.156
> +++ sh-tdep.c	18 Feb 2004 13:41:21 -0000
> @@ -1799,35 +1799,23 @@ sh_print_registers_info (struct gdbarch 
>    else
>      /* do all (or most) registers */
>      {
> -      regnum = 0;
> -      while (regnum < NUM_REGS)
> +      for (regnum = 0; regnum < NUM_REGS; ++regnum)
>  	{
>  	  /* If the register name is empty, it is undefined for this
>  	     processor, so don't display anything.  */
>  	  if (REGISTER_NAME (regnum) == NULL
>  	      || *(REGISTER_NAME (regnum)) == '\0')
> -	    {
> -	      regnum++;
> -	      continue;
> -	    }
> +	    continue;
>  
>  	  if (TYPE_CODE (gdbarch_register_type (gdbarch, regnum)) ==
>  	      TYPE_CODE_FLT)
>  	    {
> +	      /* true for "INFO ALL-REGISTERS" command */
>  	      if (fpregs)
> -		{
> -		  /* true for "INFO ALL-REGISTERS" command */
> -		  sh_do_fp_register (gdbarch, file, regnum);	/* FP regs */
> -		  regnum++;
> -		}
> -	      else
> -		regnum += (FP_LAST_REGNUM - FP0_REGNUM);	/* skip FP regs */
> +		sh_do_fp_register (gdbarch, file, regnum);	/* FP regs */
>  	    }
>  	  else
> -	    {
> -	      sh_do_register (gdbarch, file, regnum);	/* All other regs */
> -	      regnum++;
> -	    }
> +	    sh_do_register (gdbarch, file, regnum);	/* All other regs */
>  	}
>  
>        if (fpregs)
> 
> -- 
> Corinna Vinschen
> Cygwin Developer
> Red Hat, Inc.

-- 
Corinna Vinschen
Cygwin Developer
Red Hat, Inc.

Re: [RFA] sh-tdep.c: Fix erroneus register skipping in sh_print_registers_info

[Elena Zannoni]

Corinna Vinschen writes:
 > Hi,
 > 
 > the below patch fixes a long standing bug in sh_print_registers_info.
 > When the print loop encounters a float type register and fpregs is
 > not set, then the loop counter (regnum) is not just incremented by
 > one, but instead it's incremented by FP_LAST_REGNUM - FP0_REGNUM.
 > 
 > The problem with this is, that FPUL is also a float type register.
 > FPUL is two register numbers below FP0_REGNUM.  So when the loop
 > arrives at FPUL, it skips the next 16 registers.  The next evaluated
 > register then (fr14) is still a float type register, so the loop
 > skips again 16 registers.  As a result, the register set is never
 > printed completely when calling `info registers'.
 > 
 > The below patch fixes that and shortens the code slightly by using
 > a "for" loop instead of "while".
 > 
 > 
 > Corinna
 > 
 > 
 > ChangeLog:
 > 
 > 	* sh-tdep.c (sh_print_registers_info): Use for loop.
 > 	Don't skip multiple registers when a float register is encountered.
 > 

ok

Re: [RFA] sh-tdep.c: Fix erroneus register skipping in sh_print_registers_info

[Elena Zannoni]

Corinna Vinschen writes:
 > Hi,
 > 
 > the below patch fixes a long standing bug in sh_print_registers_info.
 > When the print loop encounters a float type register and fpregs is
 > not set, then the loop counter (regnum) is not just incremented by
 > one, but instead it's incremented by FP_LAST_REGNUM - FP0_REGNUM.
 > 
 > The problem with this is, that FPUL is also a float type register.
 > FPUL is two register numbers below FP0_REGNUM.  So when the loop
 > arrives at FPUL, it skips the next 16 registers.  The next evaluated
 > register then (fr14) is still a float type register, so the loop
 > skips again 16 registers.  As a result, the register set is never
 > printed completely when calling `info registers'.
 > 
 > The below patch fixes that and shortens the code slightly by using
 > a "for" loop instead of "while".
 > 
 > 
 > Corinna
 > 
 > 
 > ChangeLog:
 > 
 > 	* sh-tdep.c (sh_print_registers_info): Use for loop.
 > 	Don't skip multiple registers when a float register is encountered.
 > 

ok

Re: [RFA] sh-tdep.c: Fix erroneus register skipping in sh_print_registers_info

[Corinna Vinschen]

On Mar  5 17:31, Elena Zannoni wrote:
> Corinna Vinschen writes:
>  > Hi,
>  > 
>  > the below patch fixes a long standing bug in sh_print_registers_info.
>  > When the print loop encounters a float type register and fpregs is
>  > not set, then the loop counter (regnum) is not just incremented by
>  > one, but instead it's incremented by FP_LAST_REGNUM - FP0_REGNUM.
>  > 
>  > The problem with this is, that FPUL is also a float type register.
>  > FPUL is two register numbers below FP0_REGNUM.  So when the loop
>  > arrives at FPUL, it skips the next 16 registers.  The next evaluated
>  > register then (fr14) is still a float type register, so the loop
>  > skips again 16 registers.  As a result, the register set is never
>  > printed completely when calling `info registers'.
>  > 
>  > The below patch fixes that and shortens the code slightly by using
>  > a "for" loop instead of "while".
>  > 
>  > 
>  > Corinna
>  > 
>  > 
>  > ChangeLog:
>  > 
>  > 	* sh-tdep.c (sh_print_registers_info): Use for loop.
>  > 	Don't skip multiple registers when a float register is encountered.
>  > 
> 
> ok

Thanks, applied.

Corinna

-- 
Corinna Vinschen
Cygwin Developer
Red Hat, Inc.

Re: [RFA] sh-tdep.c: Fix erroneus register skipping in sh_print_registers_info

[Corinna Vinschen]

On Mar  5 17:31, Elena Zannoni wrote:
> Corinna Vinschen writes:
>  > Hi,
>  > 
>  > the below patch fixes a long standing bug in sh_print_registers_info.
>  > When the print loop encounters a float type register and fpregs is
>  > not set, then the loop counter (regnum) is not just incremented by
>  > one, but instead it's incremented by FP_LAST_REGNUM - FP0_REGNUM.
>  > 
>  > The problem with this is, that FPUL is also a float type register.
>  > FPUL is two register numbers below FP0_REGNUM.  So when the loop
>  > arrives at FPUL, it skips the next 16 registers.  The next evaluated
>  > register then (fr14) is still a float type register, so the loop
>  > skips again 16 registers.  As a result, the register set is never
>  > printed completely when calling `info registers'.
>  > 
>  > The below patch fixes that and shortens the code slightly by using
>  > a "for" loop instead of "while".
>  > 
>  > 
>  > Corinna
>  > 
>  > 
>  > ChangeLog:
>  > 
>  > 	* sh-tdep.c (sh_print_registers_info): Use for loop.
>  > 	Don't skip multiple registers when a float register is encountered.
>  > 
> 
> ok

Thanks, applied.

Corinna

-- 
Corinna Vinschen
Cygwin Developer
Red Hat, Inc.