typo in sbc_mediagx.c

typo in sbc_mediagx.c

[Amy Fong]

afong@ifurita:~/mtd/2.2.18-test/mtd/kernel$ diff -uNr sbc_mediagx.c.orig
sbc_mediagx.c      
--- sbc_mediagx.c.orig  Mon Oct 16 14:57:44 2000
+++ sbc_mediagx.c       Mon Oct 16 14:52:52 2000
@@ -67,7 +67,7 @@
     { name: "SBC-MediaGX flash boot partition", 
       offset: 0, 
       size: 640*1024 },
-    { mame: "SBC-MediaGX flash partition 1", 
+    { name: "SBC-MediaGX flash partition 1", 
       offset: 640*1024, 
       size: 2*1024*1024-640*1024 },
     { name: "SBC-MediaGX flash partition 2", 


To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org

Re: typo in sbc_mediagx.c

[Nicolas Pitre]

Fixed.

On 16 Oct 2000, Amy Fong wrote:

> afong@ifurita:~/mtd/2.2.18-test/mtd/kernel$ diff -uNr sbc_mediagx.c.orig
> sbc_mediagx.c      
> --- sbc_mediagx.c.orig  Mon Oct 16 14:57:44 2000
> +++ sbc_mediagx.c       Mon Oct 16 14:52:52 2000
> @@ -67,7 +67,7 @@
>      { name: "SBC-MediaGX flash boot partition", 
>        offset: 0, 
>        size: 640*1024 },
> -    { mame: "SBC-MediaGX flash partition 1", 
> +    { name: "SBC-MediaGX flash partition 1", 
>        offset: 640*1024, 
>        size: 2*1024*1024-640*1024 },
>      { name: "SBC-MediaGX flash partition 2", 
> 
> 



To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org

Re: typo in sbc_mediagx.c

[David Woodhouse]

nico@cam.org said:
> Fixed. 

Thanks. 

Note that if you're using mtdblock.c you'll need to use v1.23 - 1.24 is my 
first attempt at using a kernel thread and it doesn't actually appear to 
work. It'll probably be a few days before I get a chance to look at it 
again - volunteers are welcome if you feel so inclined :)

--
dwmw2




To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org

Re: typo in sbc_mediagx.c

[Amy Fong]

> 
> nico@cam.org said:
> > Fixed. 
> 
> Thanks. 
> 
> Note that if you're using mtdblock.c you'll need to use v1.23 - 1.24 is my 
> first attempt at using a kernel thread and it doesn't actually appear to 
> work. It'll probably be a few days before I get a chance to look at it 
> again - volunteers are welcome if you feel so inclined :)
> 
> --
> dwmw2

Uh yeah. You're using the QUEUE_EMPTY macro which doesn't really exist
in 2.2 or at least exists as 

include/lists.h:

#define QUEUE_EMPTY(head, listnam) \
        ((QUEUE_FIRST(head, listnam) == QUEUE_LAST(head, listnam)) && \
	((u_long)QUEUE_FIRST(head, listnam) == (u_long)head))

as opposed to 

2.4's include/blk.h
#define QUEUE_EMPTY list_empty(&blk_dev[MAJOR_NR].request_queue.queue_head)

<twitch> 

;)




To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org

Re: typo in sbc_mediagx.c

[David Woodhouse]

afong@furryterror.org said:
>  Uh yeah. You're using the QUEUE_EMPTY macro which doesn't really
> exist in 2.2 or at least exists as 

Hmmm. It doesn't work in 2.4 either. But I've illustrated how I think it 
can be done with a kernel thread. If anyone disagrees with me sufficiently, 
I'm quite happy to be proved wrong, as long as the proof is under GPL.

Of course, if you agree with my method in spite of my implementation, 
you're welcome to fix that too :)


--
dwmw2




To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org

Re: typo in sbc_mediagx.c

[Amy Fong]

> 
> afong@furryterror.org said:
> >  Uh yeah. You're using the QUEUE_EMPTY macro which doesn't really
> > exist in 2.2 or at least exists as 
> 
> Hmmm. It doesn't work in 2.4 either. But I've illustrated how I think it 
> can be done with a kernel thread. If anyone disagrees with me sufficiently, 
> I'm quite happy to be proved wrong, as long as the proof is under GPL.
> 
> Of course, if you agree with my method in spite of my implementation, 
> you're welcome to fix that too :)
> 
> 
> --
> dwmw2

This seems to work so far... (I hope I got the kernel version correct)
(with 2.2.18pre11)

--- mtdblock.c  Tue Oct 17 14:29:10 2000
+++ mtdblock.c.tmp      Tue Oct 17 14:27:28 2000
@@ -298,7 +298,11 @@
        again:
                set_current_state(TASK_UNINTERRUPTIBLE);
                spin_lock_irq(&io_request_lock);
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,3,46)
                if (QUEUE_EMPTY) {
+#else
+               if (!CURRENT) {
+#endif
                        spin_unlock_irq(&io_request_lock);
                        schedule();
                        if (leaving)

Oh... while I'm at it, can we add this into the Config.in? (for
2.2.16 and the 2.4's)

--- /home/afong/mtd/mtd/patches/mtd-patch-2.2.16        Mon Jun 26 04:22:56 2000
+++ mtd-patch-2.2.16    Tue Oct 17 14:44:20 2000
@@ -212,6 +212,10 @@
      define_bool CONFIG_SGI_PARTITION y
    fi
 +  tristate 'Journalling Flash filesystem (JFFS) support
(experimental)' CONFIG_JFFS_FS
++ if [ "$CONFIG_JFFS_FS" = "y" -o "$CONFIG_JFFS_FS" = "m" ]; then
++     int 'JFFS verbose level 0 (quiet), 1, 2, or 3'
CONFIG_JFFS_FS_VERBOSE 0
++ fi
+
  fi
  
  if [ "$CONFIG_NET" = "y" ]; then



To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org

Re: typo in sbc_mediagx.c

[David Woodhouse]

Committed, thanks. I didn't bother with the version check - !CURRENT ought 
to still work in 2.4.

I've amended the 2.2.16 patch, and just deleted the 2.4 patches - there's 
no point in them now that MTD is in Linus' tree.

--
dwmw2




To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org