Bug#619932: evtest does not decode KEY_NUMERIC_* key events
Package: evtest
Version: 20051019-9
Owner: Stephen Kitt
evtest does no show the fancy name for KEY_NUMERIC_* key events.
The patch attached below adds support for decoding these events.
--
--- a/utils/evtest.c2011-03-04 12:13:15.931651003 +0100
+++ b/utils/evtest.c2011-03-04 12:14:33.011651003 +0100
@@ -225,6 +225,11 @@
[KEY_TWEN] = "TWEN",[KEY_DEL_EOL] = "Delete EOL",
[KEY_DEL_EOS] = "Delete EOS", [KEY_INS_LINE] = "Insert line",
[KEY_DEL_LINE] = "Delete line",
+ [KEY_NUMERIC_0] = "0", [KEY_NUMERIC_1] = "1",
+ [KEY_NUMERIC_2] = "2", [KEY_NUMERIC_3] = "3",
+ [KEY_NUMERIC_4] = "4", [KEY_NUMERIC_5] = "5",
+ [KEY_NUMERIC_6] = "6", [KEY_NUMERIC_7] = "7",
+ [KEY_NUMERIC_8] = "8", [KEY_NUMERIC_9] = "9",
};
char *absval[5] = { "Value", "Min ", "Max ", "Fuzz ", "Flat " };
--
Florian
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#584326: debian-installer fails formatting / as ext3 on a LVM volume group
Package: debian-installer Version: 20101202 Severity: important Using the debian installer on an AMD64 with the following steps: - Using guided partitionning - Guided - user entire disk and set up LVM - All files in one partition When the / partition is to be formatted as ext3, it fails, and VT4 contains the following message: partman: mkfs.ext3 /lib/libblkid.so.1: version `BLKID_2.17' not found (required by mkfs.ext3) Using the Lenny installer on the same operations works fine. I did not try changing the filesystem, but I assume using ext2 would work just fine, since the /boot was correctly formatted previously. Thanks. -- Florian -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#581740: evtest does not flush stdout between events reports
Package: evtest
Version: 20051019-9
Owner: Stephen Kitt
evtest does no stdout flushing at all between consecutive events reports, which
can prevent its standard output redirection to a file from working:
evtest /dev/input/event6 >/tmp/evtest.log
(some mouse moving)
cat /tmp/evtest.log
(nothing)
This happened on a custom i686 eglibc build, most likely because the stdio
buffer lenght is different from a glibc one.
The patch attached adds the required flushing to fix this behaviour.
--- a/utils/evtest.c2010-05-14 18:25:11.0 +0200
+++ b/utils/evtest.c2010-05-14 18:23:20.0 +0200
@@ -364,6 +364,7 @@
perror("\nevtest: error reading");
return 1;
}
+ fflush(stdout);
for (i = 0; i < rd / sizeof(struct input_event); i++)
--
Florian
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#433321: Debconf for postfix should not accept multiple relay hosts
Package: postfix Version: 2.3.8-2+b1 Subject : Debconf for postfix should not accept multiple relay hosts If we specify multiple relay hosts while configuring postfix with debconfi, postfix will fail instaling. Debconf should warn us about that and refuse to set multiple relay hosts. Thank you. -- signature.asc Description: This is a digitally signed message part.
Bug#431977: Realtek r8169 driver is not included by default for sparc64/smp kernels
Package: linux-image-2.6.18-4-sparc64-smp Version: 2.6.18+6 Description : The realtek r8169 driver is not included by default in the linux-image-2.6-sparc64-smp kernel package. Since this driver causes kernel unaligned accesses, you will need to apply a fix, as proposed in Bug#430934 for it to be working. Once fixed, please include this driver into the next linux-image package release. Thank you very much in advance. Best regards, Florian signature.asc Description: This is a digitally signed message part.
Bug#430934: (no subject)
Looking at similar things that were done for the Tulip driver, and also for
Via-rhine, here comes a patch that fixes the bug. This patch will be sent to
the netdev mailing-list as well.
--
--
Cordialement, Florian Fainelli
-
--- drivers/net/r8169.c.old 2007-07-06 12:30:10.0 +0200
+++ drivers/net/r8169.c 2007-07-06 12:28:58.0 +0200
@@ -68,6 +68,7 @@
#include
#include
+#include
#ifdef CONFIG_R8169_NAPI
#define NAPI_SUFFIX"-NAPI"
@@ -223,7 +224,9 @@
MODULE_DEVICE_TABLE(pci, rtl8169_pci_tbl);
+#if !defined(__sparc__)
static int rx_copybreak = 200;
+#endif
static int use_dac;
static struct {
u32 msg_enable;
@@ -463,8 +466,10 @@
MODULE_DESCRIPTION("RealTek RTL-8169 Gigabit Ethernet driver");
module_param_array(media, int, &num_media, 0);
MODULE_PARM_DESC(media, "force phy operation. Deprecated by ethtool (8).");
+#if !defined(__sparc__)
module_param(rx_copybreak, int, 0);
MODULE_PARM_DESC(rx_copybreak, "Copy breakpoint for copy-only-tiny-frames");
+#endif
module_param(use_dac, int, 0);
MODULE_PARM_DESC(use_dac, "Enable PCI DAC. Unsafe on 32 bit PCI slot.");
module_param_named(debug, debug.msg_enable, int, 0);
@@ -2472,10 +2477,24 @@
{
int ret = -1;
- if (pkt_size < rx_copybreak) {
+#if defined(__sparc__)
+ if (pkt_size) {
struct sk_buff *skb;
+ skb = dev_alloc_skb(pkt_size + 4);
+ int i;
+ /* align the data to the ip header - should be faster than
copying the entire packet */
+ for (i = pkt_size - (pkt_size % 4); i >= 0; i -= 4) {
+ put_unaligned(*((u32 *) (skb->data + i)), (u32 *)
(skb->data + i + 2));
+ }
+ skb->data += 2;
+ skb->tail += 2;
+#else
+ if (pkt_size < rx_copybreak) {
+struct sk_buff *skb;
+
skb = dev_alloc_skb(pkt_size + align);
+#endif
if (skb) {
skb_reserve(skb, (align - 1) & (u32)skb->data);
eth_copy_and_sum(skb, sk_buff[0]->data, pkt_size, 0);
signature.asc
Description: This is a digitally signed message part.
Bug#430934: Kernel unaligned accesses with ip_fast_csum and ip_rcv with r8169 driver
Package: linux-image-2.6-sparc64-smp Version: 2.6.18+6 Description : I use the r8169 driver from the kernel-source-2.6.18 package on a Sun Entreprise 450 (sparc64-smp) and got a bunch of aligned accesses on ip_rcv followed by 4 unaligned access on ip_fast_csum, like that : Kernel unaligned access at TPC[5fd878] ip_rcv+0xf0/0x608 Kernel unaligned access at TPC[52cac4] ip_fast_csum+0xc/0x80 Kernel unaligned access at TPC[52cac8] ip_fast_csum+0x10/0x80 Kernel unaligned access at TPC[52cacc] ip_fast_csum+0x14/0x80 Kernel unaligned access at TPC[52cad4] ip_fast_csum+0x1c/0x80 I tried a similar fix as the one suggested in bug #409313, which points to this patch : http://article.gmane.org/gmane.linux.ports.sparc/7178 Unfortunately, increasing the rx_copybreak for sparc does not help. If you need further informations, please let me know. I know the driver is not included by default in the linux-image-2.6-sparc64-smp package, and as soon as this driver gets fixed, I could open a request for it to be included. Distribution is Debian etch freshly installed from yesterday. -- Best regards, Florian signature.asc Description: This is a digitally signed message part.
