- 07 Nov, 2004 40 commits
-
-
Antonino Daplas authored
The following fields in struct fb_info is specific only to fbcon: - struct timer_ cursor_timer - int currcon Remove them from struct fb_info and place them in fbcon_par so they become invisible from the drivers. Signed-off-by:
Antonino Daplas <adaplas@pol.net> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
Antonino Daplas authored
Fix copy system-io copy bug. Signed-off-by:
-
Antonino Daplas authored
From: David Lazar <david@panzer.utcluj.ro> I've tried to use the 6x11 (mac console) font on the framebuffer console (on the x86 architecture), and I've noticed that the font didn't have the correct box drawing glyphs. As a result I've used as a model the 8x16 font and edited the glyphs for all 256 US ASCII characters, so that they display correctly. Signed-off-by:
-
Antonino Daplas authored
Fix IO access in rivafb: - change NV_RD*/WR* macros to use read*/write* family - use NV_RD*/WR* to access IO memory - add __iomem annotations Signed-off-by:
-
Antonino Daplas authored
Fix access to ROM in aty128fb. Found by sparse. Signed-off-by:
-
Antonino Daplas authored
The frambuffer console is still writing to the graphics hardware even if the hardware is not owned anymore by the console. This can lead to crashes with some hardware. A trace of the source indicates that it comes from almost anywhere, anytime. The fix is to return immediately if vc_mode != KD_TEXT. The test is placed in the following functions if not already present. fbcon_cursor() fbcon_putcs() fbcon_clear() fbcon_bmove() fbcon_blank() Signed-off-by:
-
Arjan van de Ven authored
With the demise of intermezzo, the journal callback stuff in jbd is entirely unused (neither ext3 nor ocfs2 use it), and thus will only bitrot and bloat the kernel with code and datastructure growth. If intermezzo ever gets resurrected this will be the least of the problems they have to face (both with generic kernel as jbd). Signed-off-by:
Arjan van de Ven <arjan@infradead.org> Signed-off-by:
-
Stelian Pop authored
The following patchlet to videodev2.h brings in the "__user" definition from linux/compiler.h, making it suitable for inclusion in both kernel or user code. Signed-off-by:
Stelian Pop <stelian@popies.net> Signed-off-by:
-
Stelian Pop authored
In the latest round of updates for the sonypi driver, I fergot to send the attached patch which fixes the driver documentation. The patch contains mostly whitespace fixes and some additional information regarding the input layer usage. Signed-off-by:
-
Stelian Pop authored
Retrieving the current settings from the camera does not work very well, we need to cache the values in the driver. Signed-off-by:
-
Stelian Pop authored
Signed-off-by:
-
Stelian Pop authored
Signed-off-by:
-
Stelian Pop authored
Signed-off-by:
-
Stelian Pop authored
Signed-off-by:
-
Stelian Pop authored
Signed-off-by:
-
Stelian Pop authored
Signed-off-by:
-
Stelian Pop authored
Signed-off-by:
-
Stelian Pop authored
Signed-off-by:
-
Stelian Pop authored
Signed-off-by:
-
Stelian Pop authored
Signed-off-by:
-
Stelian Pop authored
- use module_param() instead of MODULE_PARM() and __setup() - use MODULE_VERSION() Signed-off-by:
-
Michael Hunold authored
The attached patch keeps the tvaudio i2c helper module away from any saa7146 based framegrabber. All saa7146 drivers have their dedicated i2c helper modules and don't work together with tvaudio, so keep it away alltogether. This will make mixed-card configurations work. The patch was discussed and ack'ed by Gerd Knorr. - don't attach tvaudio module on saa7146 i2c busses Signed-off-by:
Michael Hunold <hunold@linuxtv.org> Acked-by:
Gerd Knorr <kraxel@bytesex.org> Signed-off-by:
-
Michael Hunold authored
The attached patch cleans up my Video4Linux "MXB" driver and the i2c helper chipset drivers used by that cards. I ran Lindent on the files and fixed some other coding style violations, which makes the patch rather big. But besides the MODULE_PARM => module_param there have been no crucial changes. - [V4L] mxb: replace MODULE_PARM with module_param - [V4L] saa7111: replace MODULE_PARM with module_param - [V4L] tda9840: replace MODULE_PARM with module_param, re-indent code with Lindent, remove unnecessary header includes, code simplification - [V4L] tea6415c: replace MODULE_PARM with module_param, re-indent code with Lindent, remove unnecessary header includes, code simplification - [V4L] tea6420: replace MODULE_PARM with module_param, re-indent code with Lindent, remove unnecessary header includes, code simplification Signed-off-by:
-
Jeff Dike authored
From: Bodo Stroesser - Set PTRACE_DTRACE and singlestep_syscall correctly. Pass out TRACESYSGOOD and make sure the parent gets notified in ptrace_syscall. Signed-off-by:
Jeff Dike <jdike@addtoit.com> Signed-off-by:
-
Jeff Dike authored
Remove an unnecessary and even dangerous check in is_syscall() The bug was inserted by my patch syscall-security-1. Signed-off-by:
Bodo Stroesser <bstroesser@fujitsu-siemens.com> Signed-off-by:
-
Jeff Dike authored
The patch syscall-security-3 is wrong (provided by me, sorry). I missed, that singlestepping_skas() used to reset PT_DTRACE. This was handled differently in tt and skas. With syscall-security-3 applied, a process in SKAS that singlestepped once continues to singlestep until the next systemcall occurs, even if it is resumed with PTRACE_CONT or PTRACE_SYSCALL. This fix unifies the usage of PT_DTRACE in TT and SKAS. PT_DTRACE now is set by ptrace(PTRACE_SINGLESTEP,...) and reset by singlestepping() and it is evaluated in kern_do_signal(). Signed-off-by:
-
Jeff Dike authored
From: Bodo Stroesser - Make singlestepping() generic. Signed-off-by:
-
Jeff Dike authored
From: Bodo Stroesser - Make singlestep_syscall generic rather than being restricted to tt mode. Signed-off-by:
-
Jeff Dike authored
From: Bodo Stroesser - Change is_syscall to check for sysenter and also that we are actually executing a system call. Signed-off-by:
-
Jeff Dike authored
This first patch lays some groundwork needed by the others. From: Bodo Stroesser - Change the interface to handle_signal so that it doesn't take the system call return value as an argument and eliminate its return value. kern_do_signal also now doesn't return immediately after determining that there is no signal to deliver. Signed-off-by:
-
Jeff Dike authored
The comment should be self-explanatory. This works around a pile of nasty-looking build output. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
Add missing newline in help string for "nosysemu" param. Run ./vmlinux --help before and after to see the difference. Trivial - must go in. Signed-off-by:
Paolo 'Blaisorblade' Giarrusso <blaisorblade_spam@yahoo.it> Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
We have re-added the "linux" target which just creates a hardlink to vmlinux. It was noted (sorry for forgetting who said this) that "make ARCH=um" will leave an old "linux" file, since the target is not listed in .PHONY. Using a symlink, instead of an hard link, was suggested because the hard link does not get updated with make vmlinux. But mv linux $destpath (the most meaningful installation command) is broken, so I instead added linux to .PHONY. Obviously, linux will not be rebuilt on "make vmlinux ARCH=um", but this makes sense (if people use the new way, don't use old tricks). Anyway, we remove "linux" in any case. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
To avoid silly messages coming out during debugging or when things break, after that we've read in CATCH_EINTR that errno == EINTR, clear it, so that it's 0 in next iteration. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
From: Jeff Dike <jdike@addtoit.com> This is a large set of small fixes and other changes: Fixed a file descriptor leak in the network driver when changing an IP address. The port channel now sets SO_REUSEADDR. Added some initcall and exitcall definitions to arch/um/include/init.h so that they can be used from userspace code. Fixed the error handling in run_helper. Added the log() facility to mem_user.c. Fixed a problem with recursive segfaults not being handled correctly. tty_log_fd and umid aren't added to the command line any more. Fixed some prints. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
From: Jeff Dike <jdike@addtoit.com> Paolo 'Blaisorblade' Giarrusso<blaisorblade_spam@yahoo.it> As Jeff Dike noted, even if SYSEMU was originally written for use in SKAS mode, there is nothing SKAS specific in SYSEMU, so use it for TT mode, too. SYSEMU simply allows one to ptrace(PTRACE_SYSEMU) a process, that will stop on the syscall entry path like with PTRACE_SYSCALL, but with the difference that it makes sure that the syscall execution is totally skipped and that the child will see the return value set by the ptracer when the child was stopped. We even hope that the SYSEMU patch can be merged in mainline sooner than SKAS. Signed-off-by: -
Paolo \'Blaisorblade\' Giarrusso authored
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade_spam@yahoo.it> Bodo Stroesser <bodo.stroesser@fujitsu-siemens.com> While testing the host ptrace(2) features, we modify the stack frame of a test child which is doing a syscall (namely getpid). Either we override the syscall number or the return value, which become the parent's pid, or we don't, and the return value is the child's pid. Actually, we only compared the return value with the child's pid, so we did not notice a bug where the return value happened to be different from both. So I added a stricter test here. Signed-off-by: -
Paolo \'Blaisorblade\' Giarrusso authored
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade_spam@yahoo.it> Bodo Stroesser <bodo.stroesser@fujitsu-siemens.com> Jeff Dike <jdike@addtoit.com> Currently, the test for the SYSEMU support on the host is completely wrong, as Bodo noticed. We should change the syscall result (inserting the host pid) and check if it is received correctly by the guest. What we actually do, without this patch, is to overwrite the syscall number. This went unnoticed because we only check that the getpid() syscall from the child does not return its pid. We don't check that it returns the correct value. Also, override the result portably, using the PT_SYSCALL_RET_OFFSET macro which abstract away the host stack frame layout (took from Jeff Dike code). Signed-off-by: -
Paolo \'Blaisorblade\' Giarrusso authored
I'm reverting the compile-time fixes to replace them with a bigger change, which also happens to fix the compilation problem. The change is in next patches. Signed-off-by:
-
Paolo \'Blaisorblade\' Giarrusso authored
This patch makes UML use print_symbol() when dumping the stack. It is not yet complete (the i386 version uses the frame pointers to walk the stack, for instance), but should be at least enough. Also, it uses bust_spinlock() when dumping errors to help them reach the console. This appears not to be enough - however I analyzed the UML console code, and it seems that UML console don't get flushed *at all* when exiting. Signed-off-by:
-
