Re: Gnumach and I/O

Hi, Thanks people for commenting on the patches, I hope to have corrected things, so I'm resubmitting the whole updated patchset, ready for committing. Regards, Samuel ___ Bug-hurd mailing list Bug-hurd@gnu.org http://lists.gnu.org/mailman/listinfo/bu

[patch #4737] User TSS fixup

Follow-up Comment #1, patch #4737 (project hurd): Here is an update of the user tss fix, replaces file #5662 : patch-gnumach-user_tss ___ Additional Item Attachment: File name: patch-gnumach-1-user_tss Size:1 KB corrected changelog

[bug #15295] Mach lets processes write to I/O ports

Follow-up Comment #7, bug #15295 (project hurd): Here is an update of the patch that set default IO permissions to none. ___ Additional Item Attachment: File name: patch-gnumach-2-default_noio Size:1 KB corrected Changelog ; replaces fi

[bug #15295] Mach lets processes write to I/O ports

Follow-up Comment #8, bug #15295 (project hurd): Here is an update of the i386_io_port_add/remove() parameter fix. ___ Additional Item Attachment: File name: patch-gnumach-3-device_port_fix Size:1 KB Updated changelog and comment in code

[bug #15295] Mach lets processes write to I/O ports

Follow-up Comment #9, bug #15295 (project hurd): Here is an update of the patch that adds a few necessary ports (timer and speaker) ___ Additional Item Attachment: File name: patch-gnumach-4-more_ports Size:1 KB corrected Changelog ;

[bug #15295] Mach lets processes write to I/O ports

Follow-up Comment #10, bug #15295 (project hurd): Here is an update of the patch that makes io permission per task. ___ Additional Item Attachment: File name: patch-gnumach-5-io_per_taskSize:15 KB corrected changelog ; replaces file #

[bug #15295] Mach lets processes write to I/O ports

Follow-up Comment #11, bug #15295 (project hurd): Here is an update of the patch that adds an "io" device which represents all io ports. ___ Additional Item Attachment: File name: patch-gnumach-6-io_device Size:4 KB corrected changel

[bug #15295] Mach lets processes write to I/O ports

Follow-up Comment #12, bug #15295 (project hurd): Here is an update of the console-client fix for io port access. ___ Additional Item Attachment: File name: patch-hurd-ioperms Size:3 KB corrected changelog and GCS ; replaces f

Re: Gnumach and I/O

Samuel Thibault, le Sat 14 Jan 2006 01:50:35 +0100, a écrit : > Thanks people for commenting on the patches, I hope to have corrected > things, so I'm resubmitting the whole updated patchset, ready for > committing. I'm submitting them to the list again, for proof-reading. __

User TSS fixup

[gnumach]/Changelog 2006-01-14 Samuel Thibault <[EMAIL PROTECTED]> * i386/i386/iopb.c: Include "vm_param.h". (io_tss_init): Fix address and limit of user TSS. --- gnumach-20050801/i386/i386/iopb.c 2005-05-01 18:04:47.0 +0200 +++ gnumach-mine/i386/i386/iopb.c 2005-

Set io permissions to none by default

[gnumach]/Changelog 2006-01-14 Samuel Thibault <[EMAIL PROTECTED]> * i386/i386/iopb.c (iopb_create, i386_io_port_add): Set default IO permissions to none. * i386/i386/ktss.c (ktss_init): Likewise. diff -urp gnumach-mine-1-user_tss/i386/i386/iopb.c gnumach-mine-2-default_

i386_io_port_add/remove device port fix

[gnumach]/Changelog 2006-01-14 Samuel Thibault <[EMAIL PROTECTED]> * i386/i386/iopb.c (i386_io_port_add): Get the device parameter properly. (i386_io_port_remove): Likewise. diff -u -r1.1.1.1 iopb.c --- gnumach-mine-2-default_noio/i386/i386/iopb.c25 Feb 1997 21:27

more io ports

[gnumach]/Changelog 2006-01-14 Samuel Thibault <[EMAIL PROTECTED]> * i386/i386at/iopl.c (iopl_port_list): Added timer controler port. * i386/i386at/kd.c (vga_port_list): Rename to... (kd_port_list): ... this. Declare as static. Added timer and speaker por

io permissions per task

[gnumach]/Changelog 2006-01-14 Samuel Thibault <[EMAIL PROTECTED]> IO ports permissions are now task-based. * i386/i386/iopb.c (i386_io_port_add): Fix TSS access and locking accordingly. (i386_io_port_remove): Likewise. (i386_io_port_list): Likewise.

"io" device

[gnumach]/Changelog 2006-01-08 Samuel Thibault <[EMAIL PROTECTED]> Add an "io" device that represents all IO ports. * Makefile.in: Added io.c to device-files and objfiles. * device/io.c: New file. (iodevice): New variable (ioopen): New function. (

hurd console-client fix for io permissions

[hurd]/console-client/Changelog 2006-01-02 Samuel Thibault <[EMAIL PROTECTED]> * generic-speaker.c: include , and . (kd_port): New variable. (generic_speaker_start): Add IO ports access request for gnumach1. * vga-support.c: include , and .

[bug #15295] Mach lets processes write to I/O ports

Follow-up Comment #13, bug #15295 (project hurd): So, to sum it up, the patchset includes - file #5766 of patch #4737 - patch-gnumach-1-user_tss - file #3270 - patch-gnumach-2-default_noio - file #3271 - patch-gnumach-3-device_port_fix - file #3272 - patch-gnumach-4-more_ports - file #3273 - pat

Patch submission and discussion guidelines

Okie, I'm fed up with the braindead crap that the Savannah tracker is. Always send patches directly to bug-hurd first, when they have been discussed, and OKed, _then_ add it to the brain dead pile of shit that the Savannah tracker is. Not before, not later. When it has been added, to not change

Re: Patch submission and discussion guidelines

When it has been added, to not change it. Consider it as a commit to the CVS, and it is there forever. Make a new bug report for the patch, and follow the above rules. Let me clarify, instead of changing the patch, send a new bug report that fixes the filed patch, basically the same thi

Re: Patch submission and discussion guidelines

On Sat, 2006-01-14 at 02:31 +0100, Alfred M. Szmidt wrote: > Okie, I'm fed up with the braindead crap that the Savannah tracker is. What an extraordinarily differentiated statement of constructive criticism. > Always send patches directly to bug-hurd first, when they have been > discussed, and OK