Hello All,
This is a simple patch that changes several "return EBLAH"'s in drivers/*
for "return -EBLAH". I have done my best to check the call stack to ensure
that the change in sign of the return values wont break anything.
The patch supplied is agaist linux-2.4.19-pre6
Yours Tony.
/*
* "The significant problems we face cannot be solved at the
* same level of thinking we were at when we created them."
* --Albert Einstein
*/
--------------------------------------------------------------------------------
diff -urN -X /usr/src/dontdiff linux-2.4.19-pre6.clean/drivers/acorn/net/etherh.c linux-2.4.19-pre6/drivers/acorn/net/etherh.c
--- linux-2.4.19-pre6.clean/drivers/acorn/net/etherh.c Mon Oct 22 16:04:29 2001
+++ linux-2.4.19-pre6/drivers/acorn/net/etherh.c Wed Apr 17 10:42:41 2002
@@ -475,7 +475,7 @@
if (i == 6)
return 0;
}
- return ENODEV;
+ return -ENODEV;
}
/*
@@ -486,7 +486,7 @@
unsigned int serial;
if (system_serial_low == 0 && system_serial_high == 0)
- return ENODEV;
+ return -ENODEV;
serial = system_serial_low | system_serial_high;
diff -urN -X /usr/src/dontdiff linux-2.4.19-pre6.clean/drivers/block/ps2esdi.c linux-2.4.19-pre6/drivers/block/ps2esdi.c
--- linux-2.4.19-pre6.clean/drivers/block/ps2esdi.c Thu Dec 13 11:00:43 2001
+++ linux-2.4.19-pre6/drivers/block/ps2esdi.c Wed Apr 17 11:01:26 2002
@@ -630,7 +630,7 @@
/* if device is still busy - then just time out */
if (inb(ESDI_STATUS) & STATUS_BUSY) {
printk("%s: ps2esdi_out_cmd timed out (1)\n", DEVICE_NAME);
- return ERROR;
+ return -ERROR;
} /* timeout ??? */
/* Set up the attention register in the controller */
outb(((*cmd_blk) & 0xE0) | 1, ESDI_ATTN);
@@ -653,7 +653,7 @@
} else {
printk("%s: ps2esdi_out_cmd timed out while sending command (status=%02X)\n",
DEVICE_NAME, status);
- return ERROR;
+ return -ERROR;
}
} /* send all words out */
return OK;
diff -urN -X /usr/src/dontdiff linux-2.4.19-pre6.clean/drivers/mtd/mtdblock_ro.c linux-2.4.19-pre6/drivers/mtd/mtdblock_ro.c
--- linux-2.4.19-pre6.clean/drivers/mtd/mtdblock_ro.c Tue Apr 16 16:33:28 2002
+++ linux-2.4.19-pre6/drivers/mtd/mtdblock_ro.c Wed Apr 17 10:41:57 2002
@@ -268,7 +268,7 @@
if (register_blkdev(MAJOR_NR,DEVICE_NAME,&mtd_fops)) {
printk(KERN_NOTICE "Can't allocate major number %d for Memory Technology Devices.\n",
MTD_BLOCK_MAJOR);
- return EAGAIN;
+ return -EAGAIN;
}
/* We fill it in at open() time. */
diff -urN -X /usr/src/dontdiff linux-2.4.19-pre6.clean/drivers/net/plip.c linux-2.4.19-pre6/drivers/net/plip.c
--- linux-2.4.19-pre6.clean/drivers/net/plip.c Mon Oct 22 16:05:31 2001
+++ linux-2.4.19-pre6/drivers/net/plip.c Wed Apr 17 11:08:35 2002
@@ -685,13 +685,13 @@
if (rcv->length.h > dev->mtu + dev->hard_header_len
|| rcv->length.h < 8) {
printk(KERN_WARNING "%s: bogus packet size %d.\n", dev->name, rcv->length.h);
- return ERROR;
+ return -ERROR;
}
/* Malloc up new buffer. */
rcv->skb = dev_alloc_skb(rcv->length.h + 2);
if (rcv->skb == NULL) {
printk(KERN_ERR "%s: Memory squeeze.\n", dev->name);
- return ERROR;
+ return -ERROR;
}
skb_reserve(rcv->skb, 2); /* Align IP on 16 byte boundaries */
skb_put(rcv->skb,rcv->length.h);
@@ -720,7 +720,7 @@
nl->enet_stats.rx_crc_errors++;
if (net_debug)
printk(KERN_DEBUG "%s: checksum error\n", dev->name);
- return ERROR;
+ return -ERROR;
}
rcv->state = PLIP_PK_DONE;
@@ -816,7 +816,7 @@
printk(KERN_DEBUG "%s: send skb lost\n", dev->name);
snd->state = PLIP_PK_DONE;
snd->skb = NULL;
- return ERROR;
+ return -ERROR;
}
switch (snd->state) {
diff -urN -X /usr/src/dontdiff linux-2.4.19-pre6.clean/drivers/scsi/sym53c8xx_2/sym_hipd.c linux-2.4.19-pre6/drivers/scsi/sym53c8xx_2/sym_hipd.c
--- linux-2.4.19-pre6.clean/drivers/scsi/sym53c8xx_2/sym_hipd.c Fri Feb 1 17:42:40 2002
+++ linux-2.4.19-pre6/drivers/scsi/sym53c8xx_2/sym_hipd.c Wed Apr 17 10:50:04 2002
@@ -5947,7 +5947,7 @@
*/
attach_failed:
sym_hcb_free(np);
- return ENXIO;
+ return -ENXIO;
}
/*
diff -urN -X /usr/src/dontdiff linux-2.4.19-pre6.clean/drivers/sound/cs46xx.c linux-2.4.19-pre6/drivers/sound/cs46xx.c
--- linux-2.4.19-pre6.clean/drivers/sound/cs46xx.c Tue Apr 16 16:33:36 2002
+++ linux-2.4.19-pre6/drivers/sound/cs46xx.c Wed Apr 17 10:43:09 2002
@@ -2212,7 +2212,7 @@
if(!state)
return -ENODEV;
if (!access_ok(VERIFY_READ, buffer, count))
- return EFAULT;
+ return -EFAULT;
dmabuf = &state->dmabuf;
if (ppos != &file->f_pos)
diff -urN -X /usr/src/dontdiff linux-2.4.19-pre6.clean/drivers/usb/se401.c linux-2.4.19-pre6/drivers/usb/se401.c
--- linux-2.4.19-pre6.clean/drivers/usb/se401.c Mon Oct 22 16:04:55 2001
+++ linux-2.4.19-pre6/drivers/usb/se401.c Wed Apr 17 10:42:17 2002
@@ -646,7 +646,7 @@
for (i=0; i<SE401_NUMSBUF; i++) {
urb=usb_alloc_urb(0);
if(!urb)
- return ENOMEM;
+ return -ENOMEM;
FILL_BULK_URB(urb, se401->dev,
usb_rcvbulkpipe(se401->dev, SE401_VIDEO_ENDPOINT),
diff -urN -X /usr/src/dontdiff linux-2.4.19-pre6.clean/drivers/usb/stv680.c linux-2.4.19-pre6/drivers/usb/stv680.c
--- linux-2.4.19-pre6.clean/drivers/usb/stv680.c Tue Apr 16 16:33:40 2002
+++ linux-2.4.19-pre6/drivers/usb/stv680.c Wed Apr 17 10:42:30 2002
@@ -806,7 +806,7 @@
for (i = 0; i < STV680_NUMSBUF; i++) {
urb = usb_alloc_urb (0);
if (!urb)
- return ENOMEM;
+ return -ENOMEM;
/* sbuf is urb->transfer_buffer, later gets memcpyed to scratch */
usb_fill_bulk_urb (urb, stv680->udev,
On Wednesday 17 April 2002 07:42 am, Anthony J. Breeds-Taurima wrote:
> --- linux-2.4.19-pre6.clean/drivers/net/plip.c Mon Oct 22 16:05:31 2001
> +++ linux-2.4.19-pre6/drivers/net/plip.c Wed Apr 17 11:08:35 2002
...
> - return ERROR;
> + return -ERROR;
At least the changes to plip.c are wrong.
ERROR is not an errno value:
#define ERROR 2
and the return from this function is propgated to a test
if (error != ERROR) { /* Timeout */
Which will go wrong with the suggested patch.
-- Itai
On Wed, Apr 17, 2002 at 12:42:23PM +0800, Anthony J. Breeds-Taurima wrote:
> Hello All,
> This is a simple patch that changes several "return EBLAH"'s in drivers/*
> for "return -EBLAH". I have done my best to check the call stack to ensure
> that the change in sign of the return values wont break anything.
I'll add the usb driver changes to my tree, thanks.
Now if only those function return values were actually checked in the
code... :)
greg k-h