> On 17 Mar 2023, at 22:53, Moritz Buhl wrote:
>
> Any better names?
> "ffs-node-lock"
> "ffsndlock"
ffsnodelk or ffsndlck if the length shouldn’t exceed 8.
Below is an updated diff addressing the following concerns:
On Fri, Mar 17, 2023 at 03:45:59PM +0100, Alexander Bluhm wrote:
> The FreeBSD Code puts the check_nifree label before the if
> (fs->fs_cs(fs, cg).cs_nifree == 0) check. Could you move this chunk
> a little up to keep us in sync?
On Fri
till active.
> ahci0: stopping the port, softreset slot 31 was still active.
> ahci0: stopping the port, softreset slot 31 was still active.
> ahci0: stopping the port, softreset slot 31 was still active.
> ahci0: device didn't come ready after reset, TFD: 0x441
> ahci0: device didn&#
On Fri, Mar 17, 2023 at 03:23:19PM +0100, Moritz Buhl wrote:
> FreeBSD fixed this in 2013:
> https://github.com/freebsd/freebsd-src/commit/94f4ac214c403b99cbc44dd6e9cdf4db2cc9b297
>
> Below is a diff that addresses this with a exclusive lock
> using RW_SLEEPFAIL.
The FreeBSD Code puts the check_n
ive.
ahci0: device didn't come ready after reset, TFD: 0x441
ahci0: device didn't come ready after reset, TFD: 0x441
mode = 0100644, inum = 7731315, fs = /var
panic: ffs_valloc: dup alloc
Stopped at db_enter+0x10: popq%rbp
TIDPIDUID PRFLAGS PFLAGS CPU COMMAN
