Re: [PATCH v7 1/2 hurd] libirqhelp: Add library

2024-03-03 Thread Samuel Thibault
Damien Zammit, le sam. 02 mars 2024 10:31:50 +, a ecrit: > +static error_t > +get_acpi(void) > +{ > + error_t err = 0; > + mach_port_t tryacpi, device_master; > + > + acpidev = MACH_PORT_NULL; This looks odd. If we had a previous port in acpidev, we want to deallocate it, not forget it. >

Re: [PATCH 0/1 hurd] ext2fs - new translator entries in xattrs

2024-03-03 Thread Samuel Thibault
Damien Zammit, le dim. 03 mars 2024 01:00:48 +, a ecrit: > Do we allow a soft migration where translator entries are kept as is, > but new ones are migrated only upon creation. That will be fine for now, as we test it. > Or do we write a script that will migrate all translator entries so >

Re: [PATCH 1/1 hurd] ext2fs: New default: use xattrs to store translator records

2024-03-03 Thread Samuel Thibault
Damien Zammit, le dim. 03 mars 2024 01:00:53 +, a ecrit: > - Legacy records still work with either setting. Just to confirm: did you try to make an ext2fs built before this use the xattr record? I.e. we can go back between previous hurd package and next hurd package? Also, if

Re: [PATCH hurd] libps: update ps_emit_nice_size_t to handle arbitrarily large size_t

2024-03-03 Thread Samuel Thibault
Applied, thanks! Flavio Cruz, le dim. 03 mars 2024 12:38:19 -0500, a ecrit: > Update argument types for sprint_frac_value to reflect how big they > actually are so that GCC doesn't think it needs a larger buffer than > necessary. > --- > libps/spec.c | 22 +- > 1 file

Re: [PATCH hurd] rumpdisk: do not open device if block size is 0

2024-03-03 Thread Flávio Cruz
On Sun, Mar 3, 2024 at 10:39 AM Samuel Thibault wrote: > Flávio Cruz, le jeu. 29 févr. 2024 22:01:41 -0500, a ecrit: > > On Thu, Feb 29, 2024 at 7:29 PM Samuel Thibault <[1] > samuel.thiba...@gnu.org> > > wrote: > > > > I could be wrong but if you look at this build log [3] >

[PATCH hurd] libps: update ps_emit_nice_size_t to handle arbitrarily large size_t

2024-03-03 Thread Flavio Cruz
Update argument types for sprint_frac_value to reflect how big they actually are so that GCC doesn't think it needs a larger buffer than necessary. --- libps/spec.c | 22 +- 1 file changed, 13 insertions(+), 9 deletions(-) diff --git a/libps/spec.c b/libps/spec.c index

Re: [PATCH hurd] rumpdisk: do not open device if block size is 0

2024-03-03 Thread Samuel Thibault
Flávio Cruz, le jeu. 29 févr. 2024 22:01:41 -0500, a ecrit: > On Thu, Feb 29, 2024 at 7:29 PM Samuel Thibault <[1]samuel.thiba...@gnu.org> > wrote: > > I could be wrong but if you look at this build log  > [3]https://buildd.debian.org > /status/fetch.php?pkg=perl=hurd-i386=5.38.2-3=1705087520=0 >