[PATCH] x86_64: update ifdef to exclude the x86_64 for i386 only specific conditions

--- i386/i386/trap.c| 4 ++-- i386/intel/pmap.c | 4 ++-- i386/intel/read_fault.c | 2 +- 3 files changed, 5 insertions(+), 5 deletions(-) diff --git a/i386/i386/trap.c b/i386/i386/trap.c index a9d0d8d2..90650b31 100644 --- a/i386/i386/trap.c +++ b/i386/i386/trap.c @@ -236,7 +236,7

[PATCH] ipc_kmsg: fix print of size of mach_msg_header

msgh_size is a mach_msg_size_t which represent an unsigned int, so %u must me used there instead of %d --- ipc/ipc_kmsg.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/ipc/ipc_kmsg.c b/ipc/ipc_kmsg.c index b23cae7c..fe6582d1 100644 --- a/ipc/ipc_kmsg.c +++ b/ipc/ipc_kmsg.c @@

[PATCH] vm_page.c: fix a deadlock when running with smp enabled

the call vm_page_seg_pull_cache_page() return an vm_page (src) with his object being locked, as we don’t unlock before doing the vm_page_insert, it is still lock there, and so trying to relock it cause a deadlock. Replace this lock by an assert. This case was not seen as for non-smp locking is a no

[no subject]

Fix a deadlock when running gnumach with smp enabled (even with only 1 cpu). Tested with and without the patch, by doing a git clone (of gnumach repository).

[PATCH] configfrag: fix usage of mach_ncpus before his definition

--- configfrag-first.ac | 11 +++ configfrag.ac | 11 --- 2 files changed, 11 insertions(+), 11 deletions(-) diff --git a/configfrag-first.ac b/configfrag-first.ac index 5dc0db2c..22e34af8 100644 --- a/configfrag-first.ac +++ b/configfrag-first.ac @@ -23,6 +23,17 @@ nlpr=0

[PATCH] Add a patch to remove symbol duplication with rumpnet

The symbol was declared twice so when trying to use the libraries, the linker would fail due to multiple symbols with the same names. Theses declarations appears in: - buildrump.sh/src/sys/net/if.c - buildrump.sh/src/sys/rump/librump/rumpnet/net_stub.c --- ...umpnet_remove_struct_declaration_dupli

[PATCH] console-client: deallocate the port once finished with it

--- console-client/trans.c | 1 + 1 file changed, 1 insertion(+) diff --git a/console-client/trans.c b/console-client/trans.c index b7d56cb0..a3d6dd3c 100644 --- a/console-client/trans.c +++ b/console-client/trans.c @@ -898,6 +898,7 @@ console_setup_node (char *path) err = file_set_translator

[PATCH] console-client: Add another missing error check

--- console-client/trans.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/console-client/trans.c b/console-client/trans.c index 0c961c2f..b7d56cb0 100644 --- a/console-client/trans.c +++ b/console-client/trans.c @@ -892,6 +892,8 @@ console_setup_node (char *path) error (1, err, "Cannot

[PATCH] console-check: Add missing error check

--- console-client/trans.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/console-client/trans.c b/console-client/trans.c index 27ea5418..0c961c2f 100644 --- a/console-client/trans.c +++ b/console-client/trans.c @@ -896,6 +896,8 @@ console_setup_node (char *path) err = file_set_translato

[PATCH debian hurd] Change default hurd console configuration

To use the new way to configure the keyboard. With the switch to xkbcommon we now need to specify model, layout, variant and options to configure the keyboard and no more the keymap file. --- debian/hurd.hurd-console.default | 5 - 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/d

Re: [PATCH] console-client: use xkbcommon instead of x11 for xkb extended support

> Did you check that it builds when xkb_common is not available? > No, will do > Did you try to use compose? > No, I only tested the default > Do you have a patch for debian's /etc/default/hurd-console file? > I will create one > Etienne Brateau, le sam. 09 mars 2024 01:

Re: [PATCH 1/2] RFC enhance tracing utilities

Le sam. 9 mars 2024 à 15:03, Luca Dariz a écrit : > This extends the previous debug utility to trace system calls with > more events and the ability to control the tracing from userspace, > collecting a trace of the whole systems. > > This tool was quite useful in porting the rpc format to 64 bit

Re: [PATCH 1/5] console-client: use constant value instead of magic number

It's normal, I merged them and send them separately. Le sam. 9 mars 2024, 10:53, Samuel Thibault a écrit : > Hello, > > We however didn't receive the other 4 patches? > > Samuel >

[PATCH 1/5] console-client: use constant value instead of magic number

--- console-client/pc-kbd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/console-client/pc-kbd.c b/console-client/pc-kbd.c index afd3411e..aa5f1103 100644 --- a/console-client/pc-kbd.c +++ b/console-client/pc-kbd.c @@ -713,7 +713,7 @@ read_keycode (void) int release = 0;

Re: [PATCH hurd] libdiskfs: Support --next-task for filesystems

Isn't it possible to achieve this goal with an initramfs ? Le dim. 16 juil. 2023, 13:30, Sergey Bugaev a écrit : > Hi Damien, > > On Sun, Jul 16, 2023, 13:39 Damien Zammit wrote: > > This standardises the format of the boot parameters so that > > they all use the same parameter --next-task to s

[PATCH] vm_user: allow palign to be greater than PAGE_SIZE in allocate_contiguous

Allow to allocate contiguous if palign is greater than PAGE_SIZE and it’s still align with the begining of a page. --- vm/vm_user.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/vm/vm_user.c b/vm/vm_user.c index df597267..b3c1f09f 100644 --- a/vm/vm_user.c +++ b/vm/vm_user.c @@ -600,6 +60

Re: Some progress, Guix rumpdisk still crashes...

IDE is working, this is what I use every time I'm on hurd. It's just a bit slower because it can't use DMA (except if you patch gnumach: vm_allocate_contiguous to allow palign to be greater than page size but still multiple of it) Le jeu. 18 mai 2023 à 01:35, Damien Zammit a écrit : > Try '-M q3

[PATCH] vm_page.c: add missing object lock

The documentation of vm_page_insert says that the object must be locked. Moreover, the unlock call is here but no call was present. --- vm/vm_page.c | 1 + 1 file changed, 1 insertion(+) diff --git a/vm/vm_page.c b/vm/vm_page.c index 707be4bb..50916b74 100644 --- a/vm/vm_page.c +++ b/vm/vm_page.c

[PATCH] Fix build for smp

ad51c68171cb6a1cae15c61ca0218bbee2c05485 missed one replacement which was not discovered because it’s only when building with smp enabled. --- i386/intel/pmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/i386/intel/pmap.c b/i386/intel/pmap.c index 00a18402..63f372a9 100644

Re: [PATCH] Fix when enabling APIC without SMP

unmask_irq(3); > #endif /* NCPUS > 1 */ > +#endif /* APIC */ > > #ifdef LINUX_DEV > > I don't like put this code in this function. I don't like extremely long > functions, so I prefer put this in its own function, instead directly in > machine_init(). > >

[PATCH] Fix when enabling APIC without SMP

When we want to enable APIC, we must initialize the structure or otherwise, it will try to access to a not initialized memory addresses. --- i386/i386/smp.c | 10 ++ i386/i386at/model_dep.c | 10 -- 2 files changed, 10 insertions(+), 10 deletions(-) diff --git a/i386/i386/

[PATCH] Update patches to be able to have ahcisata and piixde compiled together

With this commit, rumpdisk needs to be build with "rumpdev_ahcisata rumpdev_piixide rumpdev_ata". --- debian/librumpdev-disk-dev.install | 3 + debian/librumpdev-disk0.install| 1 + debian/patches/ahcisata-atapi.diff | 26 --- debian/patches/ahcisata-rump.diff | 108 +-- debian/p

Re: Freeze booting from upstream gnumach

Is your hurd and glibc up to date? I remember an issue with switch to new version of glibc Le ven. 2 sept. 2022 à 18:37, Almudena Garcia a écrit : > Then, where could be the problem? > Check my previous messages > > El vie, 2 sept 2022 a las 18:28, Samuel Thibault () > escribió: > >> Almudena Ga

[PATCH] Fix types of read write and readables methods

--- boot/boot.c | 6 +++--- console-client/trans.c| 4 ++-- libdiskfs/io-read.c | 2 +- libdiskfs/io-readable.c | 2 +- libdiskfs/io-write.c | 2 +- libnetfs/io-read.c| 2 +- libnetfs/io-readable.c| 2 +- libnetfs/io-wri

[PATCH] Add missing included header

Build was broken since 6d3b4cea78b22a6562eef4256cfcc40a8d372194 --- libtrivfs/startup.c | 1 + 1 file changed, 1 insertion(+) diff --git a/libtrivfs/startup.c b/libtrivfs/startup.c index d5767df2..21555c85 100644 --- a/libtrivfs/startup.c +++ b/libtrivfs/startup.c @@ -21,6 +21,7 @@ #include #i

Re: [PATCH 13/17] ext2fs: remove a #define not_yet

Not in ext2. It was added in ext4. Le dim. 23 janv. 2022 à 10:40, Samuel Thibault a écrit : > Etienne Brateau, le dim. 23 janv. 2022 05:17:11 +0100, a ecrit: > > ext2fs will never support time precision better than second > > ? There is *_ctime_extra in newer definitions of

Re: [PATCH 17/17] ext2fs: remove 2 unused functions

Maybe yes, for debug purposes Le dim. 23 janv. 2022 à 10:48, Samuel Thibault a écrit : > They might still be useful some day? > > Etienne Brateau, le dim. 23 janv. 2022 05:17:15 +0100, a ecrit: > > --- > > ext2fs/ialloc.c | 72

[PATCH 15/17] Change a EIO by a ENOSPC in case where not enough space

Return a ENOSPC to notify that not enough space is here after the xattr entry instead of returning an EIO. --- ext2fs/xattr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/ext2fs/xattr.c b/ext2fs/xattr.c index 45f7febe..7976eaa6 100644 --- a/ext2fs/xattr.c +++ b/ext2fs/xattr.

[PATCH 09/17] ext2fs: set the f_namelen to the correct value

--- ext2fs/inode.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/ext2fs/inode.c b/ext2fs/inode.c index 47d05f47..511faf6e 100644 --- a/ext2fs/inode.c +++ b/ext2fs/inode.c @@ -541,7 +541,7 @@ diskfs_set_statfs (struct statfs *st) st->f_files = le32toh (sblock->s_inodes_coun

[PATCH 04/17] move to ext2_fs.h a #define

This makes all #define be grouped at the same place --- ext2fs/ext2_fs.h | 1 + ext2fs/ext2fs.h | 6 -- ext2fs/ialloc.c | 3 +++ 3 files changed, 4 insertions(+), 6 deletions(-) diff --git a/ext2fs/ext2_fs.h b/ext2fs/ext2_fs.h index d3701553..d838172a 100644 --- a/ext2fs/ext2_fs.h +++ b/ext

[PATCH 12/17] ext2fs: detect and warn when filesystem is an ext3 fs

--- ext2fs/hyper.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/ext2fs/hyper.c b/ext2fs/hyper.c index 9bff609f..ac18bbf1 100644 --- a/ext2fs/hyper.c +++ b/ext2fs/hyper.c @@ -134,6 +134,8 @@ get_hypermetadata (void) } if (le16toh (sblock->s_inode_size) != EXT2_GOOD_OLD_INOD

[PATCH 01/17] ext2fs Merge 2 headers

* ext2_fs_i.h: merge it into ext2_fs.h. This is also done in the linux tree --- ext2fs/ext2_fs.h | 22 ++ ext2fs/ext2_fs_i.h | 41 - ext2fs/ext2fs.h| 1 - 3 files changed, 22 insertions(+), 42 deletions(-) delete mode 100644 ext

[PATCH 07/17] ext2fs: use macro to check flags

* hyper.c: use macros to check flags --- ext2fs/hyper.c | 30 -- 1 file changed, 16 insertions(+), 14 deletions(-) diff --git a/ext2fs/hyper.c b/ext2fs/hyper.c index 93fb7dbb..6fc95eaf 100644 --- a/ext2fs/hyper.c +++ b/ext2fs/hyper.c @@ -60,6 +60,7 @@ get_hypermetada

[PATCH 02/17] ext2fs: update ext2_fs.h header

--- ext2fs/ext2_fs.h | 122 ++- ext2fs/ext2fs.h | 3 -- 2 files changed, 89 insertions(+), 36 deletions(-) diff --git a/ext2fs/ext2_fs.h b/ext2fs/ext2_fs.h index 90b1418d..d1a972b7 100644 --- a/ext2fs/ext2_fs.h +++ b/ext2fs/ext2_fs.h @@ -16,17 +16,10

[PATCH 13/17] ext2fs: remove a #define not_yet

ext2fs will never support time precision better than second --- ext2fs/inode.c | 25 +++-- 1 file changed, 3 insertions(+), 22 deletions(-) diff --git a/ext2fs/inode.c b/ext2fs/inode.c index 511faf6e..830b17a2 100644 --- a/ext2fs/inode.c +++ b/ext2fs/inode.c @@ -135,23 +135,14

[PATCH 16/17] ext2fs: remove a commented block of code

--- ext2fs/ialloc.c | 34 -- 1 file changed, 8 insertions(+), 26 deletions(-) diff --git a/ext2fs/ialloc.c b/ext2fs/ialloc.c index 73778bee..568d7eae 100644 --- a/ext2fs/ialloc.c +++ b/ext2fs/ialloc.c @@ -133,36 +133,18 @@ repeat: { avefreei = le32toh (

[PATCH 17/17] ext2fs: remove 2 unused functions

--- ext2fs/ialloc.c | 72 - 1 file changed, 72 deletions(-) diff --git a/ext2fs/ialloc.c b/ext2fs/ialloc.c index 568d7eae..e1a09a82 100644 --- a/ext2fs/ialloc.c +++ b/ext2fs/ialloc.c @@ -344,75 +344,3 @@ diskfs_alloc_node (struct node *dir, mode_t m

[PATCH 14/17] ext2fs: refactor add a method for dupplicated code

* xattr: introduce xattr_header_valid method --- ext2fs/xattr.c | 22 ++ 1 file changed, 14 insertions(+), 8 deletions(-) diff --git a/ext2fs/xattr.c b/ext2fs/xattr.c index 0c855616..45f7febe 100644 --- a/ext2fs/xattr.c +++ b/ext2fs/xattr.c @@ -413,6 +413,16 @@ xattr_entry_r

[PATCH 08/17] ext2fs: Take cares of indiannes

In ext2 data are stored in little endian to ensure portability. So enforce little endian when manipulating these bytes. --- ext2fs/balloc.c | 130 - ext2fs/dir.c | 61 ++--- ext2fs/ext2_fs.h | 36 ++--- ext2fs/ext2fs.h |

[PATCH 11/17] ext2fs: resolve a todo, check writability before writing

--- ext2fs/hyper.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/ext2fs/hyper.c b/ext2fs/hyper.c index 65a2937e..9bff609f 100644 --- a/ext2fs/hyper.c +++ b/ext2fs/hyper.c @@ -199,6 +199,9 @@ diskfs_set_hypermetadata (int wait, int clean) if (sblock_dirty) { +

[PATCH 06/17] ext2fs: reindent files

* balloc.c: reindent * bitmap.c: reindent * dir.c: reindent * ext2fs.c: reindent * getblk.c: reindent * hyper.c: reindent * ialloc.c: reindent * inode.c: reindent * msg.c: reindent * pager.c: reindent * pokel.c: reindent * truncate.c: reindent --- ext2fs/balloc.c| 264 +

[PATCH 10/17] ext2fs: set the writing time in the superblock

--- ext2fs/hyper.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/ext2fs/hyper.c b/ext2fs/hyper.c index 1bb74e53..65a2937e 100644 --- a/ext2fs/hyper.c +++ b/ext2fs/hyper.c @@ -199,6 +199,8 @@ diskfs_set_hypermetadata (int wait, int clean) if (sblock_dirty) { + /* Before writing

[PATCH 03/17] ext2fs: remove __KERNEL__

This makes the header more clear and btw it’s not anymore in the linux tree --- ext2fs/ext2_fs.h | 247 --- 1 file changed, 38 insertions(+), 209 deletions(-) diff --git a/ext2fs/ext2_fs.h b/ext2fs/ext2_fs.h index d1a972b7..d3701553 100644 --- a/ext2fs/

[PATCH 05/17] ext2fs: remove unneeded file

--- ext2fs/sblock.words | 6 -- 1 file changed, 6 deletions(-) delete mode 100644 ext2fs/sblock.words diff --git a/ext2fs/sblock.words b/ext2fs/sblock.words deleted file mode 100644 index e17b8fa3.. --- a/ext2fs/sblock.words +++ /dev/null @@ -1,6 +0,0 @@ -inodes blocks

[PATCH] Convert from K&R to ANSI

* hardclock.c: convert to ANSI * com.c: ditto * lpr.c: ditto * mem.c: ditto --- i386/i386/hardclock.c | 13 ++--- i386/i386/hardclock.h | 2 +- i386/i386at/com.c | 27 ++- i386/i386at/lpr.c | 31 +++ i386/i386at/mem.c

[PATCH 5/7] libps: fix some signed vs unsigned comparision

--- libps/proclist.c | 2 +- libps/write.c| 8 2 files changed, 5 insertions(+), 5 deletions(-) diff --git a/libps/proclist.c b/libps/proclist.c index 4e2174ac..c9ceaf0a 100644 --- a/libps/proclist.c +++ b/libps/proclist.c @@ -120,7 +120,7 @@ proc_stat_list_add_pids (struct proc_stat

[PATCH 6/7] libnetfs: fix type of variable

--- libnetfs/file-get-translator.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libnetfs/file-get-translator.c b/libnetfs/file-get-translator.c index 635653de..2dff6a13 100644 --- a/libnetfs/file-get-translator.c +++ b/libnetfs/file-get-translator.c @@ -66,7 +66,7 @@ netfs_S

[PATCH 7/7] proc: Fix variable type

--- proc/pgrp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/proc/pgrp.c b/proc/pgrp.c index 07e587a3..2995bb1e 100644 --- a/proc/pgrp.c +++ b/proc/pgrp.c @@ -182,7 +182,7 @@ S_proc_getsessionpids (struct proc *callerp, pid_t **pids,

[PATCH 3/7] libstore: fix type of a counter

--- libstore/argp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libstore/argp.c b/libstore/argp.c index a399b48f..31bae116 100644 --- a/libstore/argp.c +++ b/libstore/argp.c @@ -198,7 +198,7 @@ store_parsed_open (const struct store_parsed *parsed, int flags, return op

[PATCH 4/7] libdiskfs: improve asserts and remove a sign compare warning

--- libdiskfs/boot-start.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/libdiskfs/boot-start.c b/libdiskfs/boot-start.c index 5c4c3248..89d09d65 100644 --- a/libdiskfs/boot-start.c +++ b/libdiskfs/boot-start.c @@ -249,7 +249,8 @@ diskfs_start_bootstrap () diskfs_ker

[PATCH 1/7] libpager: fix a comparision between signed and unsigned

--- libpager/lock-object.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libpager/lock-object.c b/libpager/lock-object.c index c022d0c4..bb222ea6 100644 --- a/libpager/lock-object.c +++ b/libpager/lock-object.c @@ -30,7 +30,7 @@ _pager_lock_object (struct pager *p,

[PATCH 2/7] libpager: fix assertion unsigned against 0

Comparing an unsigned to be superior or equals to 0 is always true, instead move the assert before the substraction and compare it to the value to substract. --- libpager/pager-memcpy.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/libpager/pager-memcpy.c b/libpager/pag

[PATCH] ext2fs: fix invalid check

log2_dev_block_per_fs_block is unsigned so it won’t never be less than 0 and the check is then always false. Instead check the two values directly before doing the substraction. --- ext2fs/hyper.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/ext2fs/hyper.c b/ext2fs/hyper

Re: [PATCH 1/2] Remove name_equals and use strncmp instead

forget about this patch, it wasn’t intended to be sent Le jeu. 30 déc. 2021 à 12:23, Etienne Brateau a écrit : > * device/dev_name.c: Replace name_equals with strncmp > --- > device/dev_name.c | 24 ++-- > 1 file changed, 2 insertions(+), 22 deletions(-) >

[PATCH 2/2] Convert from K&R to ANSI

* device/dev_name.c: Convert function prototypes from K&R to ANSI. * device/subr.c: Convert function prototypes from K&R to ANSI. --- device/dev_name.c | 10 ++ device/subrs.c| 20 ++-- 2 files changed, 8 insertions(+), 22 deletions(-) diff --git a/device/dev_name.

[PATCH 1/2] Remove name_equals and use strncmp instead

* device/dev_name.c: Replace name_equals with strncmp --- device/dev_name.c | 24 ++-- 1 file changed, 2 insertions(+), 22 deletions(-) diff --git a/device/dev_name.c b/device/dev_name.c index f6133679..9f6b00f7 100644 --- a/device/dev_name.c +++ b/device/dev_name.c @@ -89,2

[bug #43320] rlimit doesn't get propagated to children

Follow-up Comment #1, bug #43320 (project hurd): GNUmach (or GNUhurd maybe ?) probably need to implement and provide the POSIX header to solve this issue. ___ Reply to this item at: __