[libvirt] [Klaus@Ethgen.de: [Pkg-libvirt-maintainers] Bug#579208: USB Class/Number/... is base 16, not base 10 (Failed to convert 'ff' to unsigned int)]
12:11 p.m.
Hi,
this patch from Klau Ethgen looks right to me. O.k. to apply? I'll
remove the comment with this patch then.
Cheers,
-- Guido
----- Forwarded message from Klaus Ethgen <Klaus(a)Ethgen.de> -----
in node_device_udev.c is the base 10 used for conversion of USB values.
But they are hex values so that the conversion fail in the most of time.
Funny is that the upstream did know that but didn't fix it (comment in
line 551).
The patch will fix that. But it do not remove the comment.
--- a/src/node_device/node_device_udev.c
+++ b/src/node_device/node_device_udev.c
@@ -559,28 +559,28 @@ static int udevProcessUSBInterface(struc
if (udevGetUintSysfsAttr(device,
"bInterfaceNumber",
&data->usb_if.number,
- 10) == PROPERTY_ERROR) {
+ 16) == PROPERTY_ERROR) {
goto out;
}
if (udevGetUintSysfsAttr(device,
"bInterfaceClass",
&data->usb_if._class,
- 10) == PROPERTY_ERROR) {
+ 16) == PROPERTY_ERROR) {
goto out;
}
if (udevGetUintSysfsAttr(device,
"bInterfaceSubClass",
&data->usb_if.subclass,
- 10) == PROPERTY_ERROR) {
+ 16) == PROPERTY_ERROR) {
goto out;
}
if (udevGetUintSysfsAttr(device,
"bInterfaceProtocol",
&data->usb_if.protocol,
- 10) == PROPERTY_ERROR) {
+ 16) == PROPERTY_ERROR) {
goto out;
}
12:26 p.m.
On 04/26/2010 11:11 AM, Guido Günther wrote:
Hi,
this patch from Klau Ethgen looks right to me. O.k. to apply? I'll
remove the comment with this patch then.
Is there documentation somewhere that mentions what base the udev
strings are encoded in? At any rate, I agree with you that given the
XXX comment, that this patch looks right.
if (udevGetUintSysfsAttr(device,
"bInterfaceNumber",
&data->usb_if.number,
- 10) == PROPERTY_ERROR) {
+ 16) == PROPERTY_ERROR) {
ACK.
--
Eric Blake eblake(a)redhat.com +1-801-349-2682
Libvirt virtualization library http://libvirt.org
12:46 p.m.
On Mon, Apr 26, 2010 at 11:26:51AM -0600, Eric Blake wrote:
On 04/26/2010 11:11 AM, Guido Günther wrote:
> Hi,
> this patch from Klau Ethgen looks right to me. O.k. to apply? I'll
> remove the comment with this patch then.
Is there documentation somewhere that mentions what base the udev
strings are encoded in? At any rate, I agree with you that given the
XXX comment, that this patch looks right.
That's my comment in the code. I was uncertain at the time I wrote it
what the correct base was. If you're certain that the correct base is
16, then ACK, and please do remove the comment.
Dave
> if (udevGetUintSysfsAttr(device,
> "bInterfaceNumber",
> &data->usb_if.number,
> - 10) == PROPERTY_ERROR) {
> + 16) == PROPERTY_ERROR) {
ACK.
--
Eric Blake eblake(a)redhat.com +1-801-349-2682
Libvirt virtualization library http://libvirt.org
--
libvir-list mailing list
libvir-list(a)redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
6:08 p.m.
New subject: [libvirt] [Klaus@Ethgen.de: [Pkg-libvirt-maintainers] Bug#579208: USB Class/Number/... is base 16, not base 10 (Failed to convert 'ff' to unsigned int)]
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512
Hi,
Am Mo den 26. Apr 2010 um 18:26 schrieb Eric Blake:
Is there documentation somewhere that mentions what base the udev
strings are encoded in?
A short look to the code:
drivers/usb/core/sysfs.c:
usb_descriptor_attr(bDeviceClass, "%02x\n")
usb_descriptor_attr(bDeviceSubClass, "%02x\n")
usb_descriptor_attr(bDeviceProtocol, "%02x\n")
usb_descriptor_attr(bNumConfigurations, "%d\n")
usb_descriptor_attr(bMaxPacketSize0, "%d\n")
So, yes, it is documented. :-)
Regards
Klaus Ethgen
- --
Klaus Ethgen http://www.ethgen.de/
pub 2048R/D1A4EDE5 2000-02-26 Klaus Ethgen <Klaus(a)Ethgen.de>
Fingerprint: D7 67 71 C4 99 A6 D4 FE EA 40 30 57 3C 88 26 2B
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
iQEVAwUBS9Yc85+OKpjRpO3lAQr8kAf/WhUYb7Opu0inoSG39Fx1oXa6p2P4s4CD
FnHg5hZ0RyhE+gOkDKMC06GH/X19lEXEPpIAQIQKT9gmHTnyYBUvfouBrn945Oo5
CXA3hdbcquUkdK9VUPZy1rt97z4Y6Bhl5jpcq1qUmCWBX/5jjwBUbTGUczOuOrqd
fqa40E7zZGiIz+UaQ0n2IDJvj4dP2Zna+WQdX6JbklNLXVH3vbtxZ9PwDh6ZBtrS
3Ufvl4TRTJzh3XGc3Z+nOqOorAtoacMTDbCly0WvjpGKU7hkZMNrZDKaFa2ub+ae
KI1phuoHuDHPgbPBM2/uuhDuvZvRZ1knnF3VOLL7/fzDPIxi1FOJeA==
=4+C/
-----END PGP SIGNATURE-----
12:48 p.m.
On Mon, Apr 26, 2010 at 11:26:51AM -0600, Eric Blake wrote:
On 04/26/2010 11:11 AM, Guido Günther wrote:
> Hi,
> this patch from Klau Ethgen looks right to me. O.k. to apply? I'll
> remove the comment with this patch then.
Is there documentation somewhere that mentions what base the udev
strings are encoded in? At any rate, I agree with you that given the
XXX comment, that this patch looks right.
As Klaus pointed out the only docs seems
to be the kernel source.
> if (udevGetUintSysfsAttr(device,
> "bInterfaceNumber",
> &data->usb_if.number,
> - 10) == PROPERTY_ERROR) {
> + 16) == PROPERTY_ERROR) {
ACK.
Pushed now.
-- Guido
5323
days inactive
5324
days old
4 comments
4 participants
participants (4)
-
Dave Allan -
Eric Blake -
Guido Günther -
Klaus Ethgen