[2.6 patch] kill drive_info

[2.6 patch] kill drive_info

[Adrian Bunk]

The only user of drive_info in 2.6 seems to be drivers/ide/legacy/hd.c .

What about the patch below to kill drive_info?

Please double-check this patch, all I checked was a test of compilation
on i386.

diffstat output:
 arch/cris/kernel/setup.c         |    1 -
 arch/i386/kernel/i386_ksyms.c    |    5 -----
 arch/i386/kernel/setup.c         |    2 --
 arch/ia64/dig/setup.c            |   10 ----------
 arch/ia64/sn/kernel/setup.c      |   14 --------------
 arch/ppc/platforms/pmac_setup.c  |    2 --
 arch/sh64/kernel/sh_ksyms.c      |    8 --------
 arch/x86_64/kernel/setup.c       |    2 --
 arch/x86_64/kernel/x8664_ksyms.c |    5 -----
 drivers/ide/legacy/hd.c          |    7 +++++--
 10 files changed, 5 insertions(+), 51 deletions(-)


Signed-off-by: Adrian Bunk <bunk@fs.tum.de>

--- linux-2.6.7-mm7-full/arch/x86_64/kernel/setup.c.old	2004-07-14 01:18:28.000000000 +0200
+++ linux-2.6.7-mm7-full/arch/x86_64/kernel/setup.c	2004-07-14 01:19:45.000000000 +0200
@@ -85,7 +85,6 @@
 /*
  * Setup options
  */
-struct drive_info_struct { char dummy[32]; } drive_info;
 struct screen_info screen_info;
 struct sys_desc_table_struct {
 	unsigned short length;
@@ -429,7 +428,6 @@
 	unsigned long kernel_end;
 
  	ROOT_DEV = old_decode_dev(ORIG_ROOT_DEV);
- 	drive_info = DRIVE_INFO;
  	screen_info = SCREEN_INFO;
 	edid_info = EDID_INFO;
 	aux_device_present = AUX_DEVICE_INFO;
--- linux-2.6.7-mm7-full/arch/x86_64/kernel/x8664_ksyms.c.old	2004-07-14 01:20:15.000000000 +0200
+++ linux-2.6.7-mm7-full/arch/x86_64/kernel/x8664_ksyms.c	2004-07-14 01:20:25.000000000 +0200
@@ -41,11 +41,6 @@
 extern void __read_lock_failed(rwlock_t *rw);
 #endif
 
-#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_HD) || defined(CONFIG_BLK_DEV_IDE_MODULE) || defined(CONFIG_BLK_DEV_HD_MODULE)
-extern struct drive_info_struct drive_info;
-EXPORT_SYMBOL(drive_info);
-#endif
-
 extern unsigned long get_cmos_time(void);
 
 /* platform dependent support */
--- linux-2.6.7-mm7-full/arch/i386/kernel/i386_ksyms.c.old	2004-07-14 01:22:27.000000000 +0200
+++ linux-2.6.7-mm7-full/arch/i386/kernel/i386_ksyms.c	2004-07-14 01:22:39.000000000 +0200
@@ -51,11 +51,6 @@
 extern void FASTCALL( __read_lock_failed(rwlock_t *rw));
 #endif
 
-#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_HD) || defined(CONFIG_BLK_DEV_IDE_MODULE) || defined(CONFIG_BLK_DEV_HD_MODULE)
-extern struct drive_info_struct drive_info;
-EXPORT_SYMBOL(drive_info);
-#endif
-
 extern unsigned long cpu_khz;
 extern unsigned long get_cmos_time(void);
 
--- linux-2.6.7-mm7-full/arch/ia64/sn/kernel/setup.c.old	2004-07-14 01:23:27.000000000 +0200
+++ linux-2.6.7-mm7-full/arch/ia64/sn/kernel/setup.c	2004-07-14 01:24:19.000000000 +0200
@@ -109,20 +109,6 @@
 };
 
 /*
- * This is here so we can use the CMOS detection in ide-probe.c to
- * determine what drives are present.  In theory, we don't need this
- * as the auto-detection could be done via ide-probe.c:do_probe() but
- * in practice that would be much slower, which is painful when
- * running in the simulator.  Note that passing zeroes in DRIVE_INFO
- * is sufficient (the IDE driver will autodetect the drive geometry).
- */
-#ifdef CONFIG_IA64_GENERIC
-extern char drive_info[4*16];
-#else
-char drive_info[4*16];
-#endif
-
-/*
  * This routine can only be used during init, since
  * smp_boot_data is an init data structure.
  * We have to use smp_boot_data.cpu_phys_id to find
--- linux-2.6.7-mm7-full/arch/ia64/dig/setup.c.old	2004-07-14 01:24:45.000000000 +0200
+++ linux-2.6.7-mm7-full/arch/ia64/dig/setup.c	2004-07-14 01:25:02.000000000 +0200
@@ -25,16 +25,6 @@
 #include <asm/machvec.h>
 #include <asm/system.h>
 
-/*
- * This is here so we can use the CMOS detection in ide-probe.c to
- * determine what drives are present.  In theory, we don't need this
- * as the auto-detection could be done via ide-probe.c:do_probe() but
- * in practice that would be much slower, which is painful when
- * running in the simulator.  Note that passing zeroes in DRIVE_INFO
- * is sufficient (the IDE driver will autodetect the drive geometry).
- */
-char drive_info[4*16];
-
 void __init
 dig_setup (char **cmdline_p)
 {
--- linux-2.6.7-mm7-full/arch/ppc/platforms/pmac_setup.c.old	2004-07-14 01:25:51.000000000 +0200
+++ linux-2.6.7-mm7-full/arch/ppc/platforms/pmac_setup.c	2004-07-14 01:25:57.000000000 +0200
@@ -95,8 +95,6 @@
 
 struct device_node *memory_node;
 
-unsigned char drive_info;
-
 int ppc_override_l2cr = 0;
 int ppc_override_l2cr_value;
 int has_l2cache = 0;
--- linux-2.6.7-mm7-full/arch/cris/kernel/setup.c.old	2004-07-14 01:26:26.000000000 +0200
+++ linux-2.6.7-mm7-full/arch/cris/kernel/setup.c	2004-07-14 01:26:42.000000000 +0200
@@ -23,7 +23,6 @@
 /*
  * Setup options
  */
-struct drive_info_struct { char dummy[32]; } drive_info;
 struct screen_info screen_info;
 
 unsigned char aux_device_present;
--- linux-2.6.7-mm7-full/arch/sh64/kernel/sh_ksyms.c.old	2004-07-14 01:27:07.000000000 +0200
+++ linux-2.6.7-mm7-full/arch/sh64/kernel/sh_ksyms.c	2004-07-14 01:27:16.000000000 +0200
@@ -32,14 +32,6 @@
 extern void dump_thread(struct pt_regs *, struct user *);
 extern int dump_fpu(struct pt_regs *, elf_fpregset_t *);
 
-#if 0
-/* Not yet - there's no declaration of drive_info anywhere. */
-#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_HD) || defined(CONFIG_BLK_DEV_IDE_MODULE) || defined(CONFIG_BLK_DEV_HD_MODULE)
-extern struct drive_info_struct drive_info;
-EXPORT_SYMBOL(drive_info);
-#endif
-#endif
-
 /* platform dependent support */
 EXPORT_SYMBOL(dump_thread);
 EXPORT_SYMBOL(dump_fpu);
--- linux-2.6.7-mm7-full/drivers/ide/legacy/hd.c.old	2004-07-14 01:29:42.000000000 +0200
+++ linux-2.6.7-mm7-full/drivers/ide/legacy/hd.c	2004-07-14 01:53:21.000000000 +0200
@@ -47,6 +47,10 @@
 #include <asm/io.h>
 #include <asm/uaccess.h>
 
+#ifdef __i386__
+#include <asm/setup.h>
+#endif
+
 #ifdef __arm__
 #undef  HD_IRQ
 #endif
@@ -726,8 +730,7 @@
 
 #ifdef __i386__
 	if (!NR_HD) {
-		extern struct drive_info drive_info;
-		unsigned char *BIOS = (unsigned char *) &drive_info;
+		unsigned char *BIOS = &boot_params[0x80];
 		unsigned long flags;
 		int cmos_disks;
 
--- linux-2.6.7-mm7-full/arch/i386/kernel/setup.c.old	2004-07-14 01:28:56.000000000 +0200
+++ linux-2.6.7-mm7-full/arch/i386/kernel/setup.c	2004-07-14 01:59:25.000000000 +0200
@@ -102,7 +102,6 @@
 /*
  * Setup options
  */
-struct drive_info_struct { char dummy[32]; } drive_info;
 struct screen_info screen_info;
 struct apm_info apm_info;
 struct sys_desc_table_struct {
@@ -1288,7 +1287,6 @@
 #endif
 
  	ROOT_DEV = old_decode_dev(ORIG_ROOT_DEV);
- 	drive_info = DRIVE_INFO;
  	screen_info = SCREEN_INFO;
 	edid_info = EDID_INFO;
 	apm_info.bios = APM_BIOS_INFO;

Re: [2.6 patch] kill drive_info

[Andries Brouwer]

On Wed, Jul 14, 2004 at 02:08:10AM +0200, Adrian Bunk wrote:

> What about the patch below to kill drive_info?
> 
> Please double-check this patch, all I checked was a test of compilation
> on i386.

> -		unsigned char *BIOS = (unsigned char *) &drive_info;
> +		unsigned char *BIOS = &boot_params[0x80];

Ugly. There is already a define DRIVE_INFO for this right hand side:

  setup.h:#define PARAM (boot_params)
  setup.h:#define DRIVE_INFO (*(struct drive_info_struct *) (PARAM+0x80))


> - 	drive_info = DRIVE_INFO;

Hmm. setup.c copies this info from where it was left after booting
to some safe place. You seem to think that this saving is not required.
Is it not?

Andries

Re: [2.6 patch] kill drive_info

[Adrian Bunk]

On Wed, Jul 14, 2004 at 11:01:59AM +0200, Andries Brouwer wrote:
>...
> > - 	drive_info = DRIVE_INFO;
> 
> Hmm. setup.c copies this info from where it was left after booting
> to some safe place. You seem to think that this saving is not required.
> Is it not?

- boot_params is __initdata
- hd_init in legacy/hd.c is __init
- legacy/hd.c can't be built modular

Did I miss something?

The situation is different in 2.4 where the new (possibly modular) IDE 
driver also uses drive_info .

> Andries

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

cannot ping from a usb-serial device

[pietro canevarolo]

Hi all,
I am developing a portable device that I must connect to Internet via usb.
I connect to my linux box with pppd on ttyUSB0 and the link layer is 
working ok. I can ping my device from all the machines connected, but 
when the device pings some address known reacheable my linux box sees 
the packets ("ifconfig ppp0" returns the correct number of packets 
exchanged and "netstat -i" reports the correct number of bytes on the 
interface), but no echo replies come back to my device.
I understand that my question may be silly for this forum, but I am 
really out of resources.
Can someone out there point me to some info to understand why this is 
happening?

many tnx

pietro canevarolo

Re: [2.6 patch] kill drive_info

[Andries Brouwer]

On Thu, Jul 15, 2004 at 04:59:48AM +0200, Adrian Bunk wrote:
> On Wed, Jul 14, 2004 at 11:01:59AM +0200, Andries Brouwer wrote:
> >...
> > > - 	drive_info = DRIVE_INFO;
> > 
> > Hmm. setup.c copies this info from where it was left after booting
> > to some safe place. You seem to think that this saving is not required.
> > Is it not?
> 
> - boot_params is __initdata
> - hd_init in legacy/hd.c is __init
> - legacy/hd.c can't be built modular
> 
> The situation is different in 2.4 where the new (possibly modular) IDE 
> driver also uses drive_info .

Yes, I see that this area (which used to be empty_zero_page)
is independent since 2.6.5.

So, maybe your patch is correct.

[Still, the patch brings knowledge about boot_params that earlier was
concentrated in head.S, setup.c, setup.h to elsewhere in the kernel. Ach.]

The doc file is still called zero-page.txt. One might add

0x228   4 bytes         command line buffer

and possibly rename it.

Andries