http://blog.gmane.org/gmane.linux.debian.devel.kernel hourly 1 1901-01-01T00:00+00:00 http://gmane.org/img/gmane-25t.png http://gmane.org http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43323 reassign 404503 kbd clone 404503 -1 reassign -1 console-tools thanks Pavel Vávra, le Fri 17 Oct 2008 23:18:14 +0200, a écrit : And kbd doesn't yet include the patch anyway... http://bugzilla.kernel.org/attachment.cgi?id=12359 Attached is a console-tools port of the patch, could you please check that it works? You should just need to $ apt-get source console-tools $ sudo apt-get build-dep console-tools $ cd console-tools-version $ ./debian/rules stampdir/patchapply $ (cd build-tree/console-tools-version ; patch -p1 < ~/patch-console-tools) $ fakeroot ./debian/rules binary-arch and install console-tools_version_arch.deb Samuel diff -ur console-tools-0.2.3-orig/include/lct/ksyms.h console-tools-0.2.3/include/lct/ksyms.h --- console-tools-0.2.3-orig/include/lct/ksyms.h2008-12-01 23:48:27.000000000 +0100 +++ console-tools-0.2.3/include/lct/ksyms.h2008-12-01 23:54:50.000000000 +0100 < at >< at > -21,6 +21,7 < at >< at > extern const int charsets_size; extern int set_charset(const char *name); +extern const char *chosen_charset; extern int add_number(int code); extern int add_capslock(int code); extern const char *codetoksym(int code); diff -ur console-tools-0.2.3-orig/kbdtools/loadkeys.y console-tools-0.2.3/kbdtools/loadkeys.y --- console-tools-0.2.3-orig/kbdtools/loadkeys.y2008-12-01 23:48:27.000000000 +0100 +++ console-tools-0.2.3/kbdtools/loadkeys.y2008-12-01 23:56:36.000000000 +0100 < at >< at > -70,6 +70,7 < at >< at > #include <linux/keyboard.h> #include <sys/ioctl.h> #include <ctype.h> +#include <iconv.h> #include <sysexits.h> #include <signal.h> < at >< at > -111,7 +112,11 < at >< at > /* the kernel structures we want to set or print */ u_short *key_map[MAX_NR_KEYMAPS]; char *func_table[MAX_NR_FUNC]; +#ifdef KDSKBDIACRUC +struct kbdiacruc accent_table[MAX_DIACR]; +#else struct kbdiacr accent_table[MAX_DIACR]; +#endif unsigned int accent_table_size = 0; char key_is_constant[512]; // 512 == Max. value of NR_KEYS across kernels <= 2.6.1 < at >< at > -692,7 +697,7 < at >< at > static void compose(int diacr, int base, int res) { - struct kbdiacr *p; + typeof(*accent_table) *p; if (accent_table_size == MAX_DIACR) { fprintf(stderr, _("compose table overflow\n")); < at >< at > -860,11 +865,80 < at >< at > return ct; } +#ifdef KDSKBDIACRUC +static wchar_t convert_char(iconv_t conv, unsigned char c) { +wchar_t wc; +char *i = (void*)&c; +size_t is = sizeof(c); +char *o = (void*)&wc; +size_t os = sizeof(wc); +if (iconv(conv, &i, &is, &o, &os) == (size_t) -1) { + perror("iconv"); + fprintf(stderr, _("Couldn't convert character %u from charset %s to unicode\n"), c, chosen_charset); + wc = 0; +} +return wc; +} +static int +defdiacsuc(int fd){ + struct kbdiacrsuc kd; +int i; +iconv_t conv; + +if (!chosen_charset) + return 0; + +kd.kb_cnt = accent_table_size; +if (kd.kb_cnt > MAX_DIACR) { + kd.kb_cnt = MAX_DIACR; + fprintf(stderr, _("too many compose definitions\n")); +} +conv = iconv_open("WCHAR_T", chosen_charset); +if (conv == (iconv_t)(-1)) { + perror("iconv_open"); + fprintf(stderr, _("Can't convert from charset %s to unicode\n"), chosen_charset); + return 0; +} +for (i = 0; i < kd.kb_cnt; i++) { + wchar_t wc; + wc = convert_char(conv, accent_table[i].diacr); + if (!wc) +goto err; + kd.kbdiacruc[i].diacr = wc; + wc = convert_char(conv, accent_table[i].base); + if (!wc) +goto err; + kd.kbdiacruc[i].base = wc; + wc = convert_char(conv, accent_table[i].result); + if (!wc) +goto err; + kd.kbdiacruc[i].result = wc; +} + +if(ioctl(fd, KDSKBDIACRUC, (unsigned long) &kd)) { + if (errno == EINVAL) +return 0; + perror("KDSKBDIACR"); + exit(1); +} +return kd.kb_cnt; +err: +iconv_close(conv); +return 0; +} + +#endif static int defdiacs(int fd) { struct kbdiacrs kd; unsigned i; +#ifdef KDSKBDIACRUC + int cnt = defdiacsuc(fd); + if (cnt) + return cnt; +#endif + kd.kb_cnt = accent_table_size; if (kd.kb_cnt > MAX_DIACR) { < at >< at > -872,7 +946,11 < at >< at > fprintf(stderr, _("too many compose definitions\n")); } for (i = 0; i < kd.kb_cnt; i++) - kd.kbdiacr[i] = accent_table[i]; + { + kd.kbdiacr[i].diacr = accent_table[i].diacr; + kd.kbdiacr[i].base = accent_table[i].base; + kd.kbdiacr[i].result = accent_table[i].result; + } if(ioctl(fd, KDSKBDIACR, (unsigned long) &kd)) { diff -ur console-tools-0.2.3-orig/lib/ksyms.c console-tools-0.2.3/lib/ksyms.c --- console-tools-0.2.3-orig/lib/ksyms.c2008-12-01 23:48:27.000000000 +0100 +++ console-tools-0.2.3/lib/ksyms.c2008-12-01 23:53:43.000000000 +0100 < at >< at > -1647,6 +1647,7 < at >< at > /* Functions for both dumpkeys and loadkeys. */ static int prefer_unicode = 0; +const char *chosen_charset = NULL; void list_charsets(FILE *f) { < at >< at > -1700,6 +1701,7 < at >< at > if(p->name[0]) syms[0].table[i] = p->name; } +chosen_charset = strdup(charset); return 0; } } Samuel Thibault 2008-12-01T23:01:28 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43322 Processing commands for control< at >bugs.debian.org: Bug#404503: unfixed capitalization of accented letters by caps lock Bug reassigned from package `linux-2.6' to `kbd'. Bug#404503: unfixed capitalization of accented letters by caps lock Bug 404503 cloned as bug 507513. Bug#507513: unfixed capitalization of accented letters by caps lock Bug reassigned from package `kbd' to `console-tools'. Stopping processing here. Please contact me if you need assistance. Debian bug tracking system administrator (administrator, Debian Bugs database) Debian Bug Tracking System 2008-12-01T23:06:14 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43321 Yes it works fine with -10. You can close this report. Cheers, Bill Allombert 2008-12-01T21:00:16 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43320 I've turned off AppleTalk on the problem system because of this bug, so I'm not sure if it will still happen with 2.6.18-6-686, and even though the oops didn't affect other aspects of the system's operation, I'd rather not try it on my production file server. I'll try it on a test system now and get back to you in a few weeks (it may take a while before it oopses). I can't guarantee that failure to reproduce it on 2.6.18-6-686 on my test system means it won't happen on 2.6.18-6-686 on the machine that was having problems, but if it runs cleanly on the test system, I may risk it on the production system over the holiday break. Will Aoki 2008-12-01T17:16:27 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43319 I'll try ... ... if I can get this linux-kernel compiled and installed within our Debian (etchnhalf) installation. In our Debian repository, the newest kernel I saw is a 2.6.26 version. Thus I cannot predict, if and how quick I'll get a working build of version 2.6.28-rc6, but I'll try. Best regards, Ruediger Oberhage Ruediger Oberhage 2008-12-01T16:02:58 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43318 El Dissabte 29 Novembre 2008 18:07:11 Moritz Muehlenhoff va escriure: I'm installing 2.6.26-10 right now and I'll give it a try the following days. I have a busy week so maybe I won't be able to run and watch it until Thursday or Friday. Eloi Notario 2008-12-01T14:31:31 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43317 Hi! On Sunday 30 November 2008 02:30:51 Adeodato Simó wrote: I am sorry we are kind of forcing this now, this was absolutely not intended. When I was looking what to cherrypick for the old version in Debian, I should have backported more than 3/4 of the commits, all being bugfixes relevant for normal operation, among them security issues (CVE-2008-4579, CVE-2008-4580 and others found via code audit) and grave functionality bugs like not working noatime support (what you really want on a shared filesystem, to avoid exclusive write locks on every file access). I should have asked on this list first, before uploading. GFS still has problems on high load, not fixed yet upstream, caused by DLM changes in linux-2.6 2.6.26 (2.6.25 works), which probably need to be addressed there. Best regards Frederik Schüler Frederik Schüler 2008-12-01T12:07:09 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43316 Package: linux-image-2.6.26-1-amd64 Version: 2.6.26-10 Severity: minor when I boot my Samsumg Q45 (laptop), I get the following messages at the _bottom_ of the screen. They remain there until the normal boot messages get to the bottom of the screen and overwrite them. Kernel alive kernel direct mapping tables up to 100000000 < at > 8000-d000 Sam Morris 2008-12-01T10:35:47 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43315 I too had the same problem then I gave acpi=off then the system goes to halt stage instead of shutdown when I shutdown my pc. So I gave the kernel parameter as "pci=nommconf iommu=calgary" then its working fine. Why is it happening like that... On Sun, Nov 30, 2008 at 9:55 PM, Gerrit Jan Baarda <gjbaarda< at >xs4all.nl>wrote: boss ganesh 2008-12-01T07:28:44 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43314 is there anything like /etc/modprobe.d/alsa* you can add the modules under /etc/modules - it will fetch module during boot itself On Sun, Nov 30, 2008 at 11:14 PM, Phred Xmidt <fsmithred< at >gmail.com> wrote: boss ganesh 2008-12-01T07:22:52 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43313 On Sat, 29 Nov 2008 23:27:12 +0100 Moritz Muehlenhoff <jmm< at >inutil.org> wrote: Hi, I am afraid I can no longer verify, as the card in question stopped working completely (even in other OSes) some time ago and I threw it away. (Which is too bad because it is one of the few cards I have found that accepts Master mode.....) Clayton Clayton 2008-12-01T05:05:57 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43312 Just compiled a new kernel 2.6.26 (from backports.org) and lshw now seems to talk correctly to the scsi system and the panic doesn't occur. The Eclectic One 2008-12-01T02:49:47 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43311 I have recently found this problem on an old socket 7 machine with an SiS900 network card. The lspci output for the card is 00:0b.0 Ethernet controller: Silicon Integrated Systems [SiS] SiS900 PCI Fast Ethernet (rev 02) The motherboard is a PC Chips (Hsin Tech) M537DMA33: http://www.users.zetnet.co.uk/thorfin/motherboards/mb01/motherboard_1.htm Kernel is 2.6.26 from the Lenny RC1 installer. I'm afraid I don't know how to troubleshoot this bug. For example, earlier messages mention setting the card to half-duplex mode - should I try that, and if so how is it done? I don't seem to get as many of these errors as the previous reporter, but enough to notice in the logs. Grateful for any help, Many thanks, Chris Chris Carr 2008-11-30T17:51:09 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43310 Added option to /etc/modprobe.d/sound: model=3stack Phred Xmidt 2008-11-30T17:44:40 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43309 Hi Moritz, With the d-i RC1 it's even worse on the console: ----------------- client boot log--------------------------- 0 > boot net:192.168.22.51,vmlinuz-chrp.initrd,192.168.22.58 BOOTP S = 1 FILE: vmlinuz-chrp.initrd Load Addr=0x4000 Max Size=0xbfc000 FINAL Packet Count = 13774 FINAL File Size = 7052236 bytes. zImage starting: loaded at 0x00400000 (sp: 0x0121fea0) Allocating 0x431e28 bytes for kernel ... OF version = 'IBM,SPH00221' gunzipping (0x01400000 <- 0x00407000:0x005c14cf)...done 0x3db000 bytes Attached initrd image at 0x005c2000-0x00aa8b3d initrd head: 0x1f8b0808 Linux/PowerPC load: Finalizing device tree... using OF tree (promptr=00c1b320) OF stdout device is: /pci< at >fef00000/isa< at >b/serial< at >i3f8 command line: console=ttyS0,9600 console=tty0 memory layout at init: alloc_bottom : 01836000 alloc_top : 30000000 alloc_top_hi : 40000000 rmo_top : 30000000 ram_top : 40000000 Looking for displays found display : /pci< at >fee00000/display< at >b, opening ... done instantiating rtas at 0x2ffb5000 ... done copying OF device tree ... Building dt strings... Building dt structure... Device tree strings 0x01837000 -> 0x01837f15 Device tree struct 0x01838000 -> 0x0183d000 Calling quiesce ... returning from prom_init ----------------- client boot log--------------------------- That's it. The command line seem to be valid (ttyS0), but it's opening a display? Sounds strange to me. Unfortunately, I do not have a matching monitor available anymore, so I can't see if something is displayed there. I'll try the daily build next, If you like, CU, Uli Ulrich Teichert 2008-11-30T17:11:15 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43308 Processing commands for control< at >bugs.debian.org: Bug#422675: sis900: tons of corrupted packets are reported Bug reopened, originator set to Chris Carr <rantingman< at >gmail.com>. Stopping processing here. Please contact me if you need assistance. Debian bug tracking system administrator (administrator, Debian Bugs database) Debian Bug Tracking System 2008-11-30T16:57:03 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43307 Package: linux-image-2.6.26-1-686 Version: 2.6.26-11 Followup-For: Bug #497692 I had exactly the same behaviour when upgrading to 2.6.26. I managed to make my system bootable by adding pci=nomsi to the kernel startup line. However I see the following messages in dmesg: [ 797.309745] +------ PCI-Express Device Error ------+ [ 797.309745] Error Severity : Uncorrected (Non-Fatal) [ 797.309745] PCIE Bus Error type : Transaction Layer [ 797.309745] Flow Control Protocol : First [ 797.309745] Receiver ID : 0010 [ 797.309745] VendorID=1106h, DeviceID=a327h, Bus=00h, Device=02h, Function=00h [ 797.309745] Broadcast error_detected message [ 797.309745] Broadcast mmio_enabled message [ 797.309745] Broadcast resume message [ 797.309745] AER driver successfully recovered It looks like this error is also related to the console. For example, an entry is inserted almost everytime a debconf dialog is opened. And if I run less <somefile> in a console, and then scroll beyond the last line I get this message most every 'beep'. During my struggles to get the system bootable again I noticed that the current (lenny) installer apparently has similar problems. I tried to run the console rescue mode, but that results in a blank screen. However running the graphical rescue mode works just fine. Please feel free to ask me for more information Gerrit Jan Baarda 2008-11-30T16:25:58 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43306 I meant to provide additional information for bug #497692 Please excuse me and ignore my report! Gerrit Jan Baarda 2008-11-30T16:16:47 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43305 Package: linux-image-2.6.26-1-686 Version: 2.6.26-11 Followup-For: Bug #494007 I had exactly the same behaviour when upgrading to 2.6.26. I managed to make my system bootable by adding pci=nomsi to the kernel startup line. However I see the following messages in dmesg: [ 797.309745] +------ PCI-Express Device Error ------+ [ 797.309745] Error Severity : Uncorrected (Non-Fatal) [ 797.309745] PCIE Bus Error type : Transaction Layer [ 797.309745] Flow Control Protocol : First [ 797.309745] Receiver ID : 0010 [ 797.309745] VendorID=1106h, DeviceID=a327h, Bus=00h, Device=02h, Function=00h [ 797.309745] Broadcast error_detected message [ 797.309745] Broadcast mmio_enabled message [ 797.309745] Broadcast resume message [ 797.309745] AER driver successfully recovered It looks like this error is also related to the console. For example, an entry is inserted almost everytime a debconf dialog is opened. And if I run less <somefile> in a console, and then scroll beyond the last line I get this message most every 'beep'. During my struggles to get the system bootable again I noticed that the current (lenny) installer apparently has similar problems. I tried to run the console rescue mode, but that results in a blank screen. However running the graphical rescue mode works just fine. Please feel free to ask me for more information. Gerrit Jan Baarda 2008-11-30T15:49:23 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43304 Package: linux-image-2.6.26-1-openvz-amd64 Version: 2.6.26-10 Followup-For: Bug #500645 Apparently this has been fixed in upstream already [1], could you please integrate the patch? best regards, Michael [1] http://bugzilla.openvz.org/show_bug.cgi?id=1034#c6 Michael Renner 2008-11-30T14:14:48 http://permalink.gmane.org/gmane.linux.debian.devel.kernel/43303 I figured it out: linux-image-2.6.26-1-686 has "CONFIG_EDD=m" set and I don't set that in my kernel. Hugo hugo vanwoerkom 2008-11-30T13:24:40 Search the mailing list at Gmane query http://search.gmane.org/?group=$group=gmane.linux.debian.devel.kernel