Date: Mon, 07 Feb 2011 05:53:30 +0300 From: Anonymous <swell.k@gmail.com> To: Rich <rincebrain@gmail.com> Cc: Olivier Smedts <olivier@gid0.org>, FreeBSD-current <freebsd-current@freebsd.org>, Shawn Webb <lattera@gmail.com> Subject: Re: zpool v28 and -current Message-ID: <86oc6ov9lh.fsf@gmail.com> In-Reply-To: <AANLkTin0OxaHcdzE8EF6t-%2BkeLjbsXKOMnsA9WdGdrTg@mail.gmail.com> (Rich's message of "Sun, 6 Feb 2011 20:38:19 -0500") References: <AANLkTikM0S3aovnk0WQikC%2BC130DW7M4Nv6kDp%2BRCfGj@mail.gmail.com> <86k4hkiyr1.fsf@gmail.com> <AANLkTikAhuaeo3Z1O28ihBucafyJnpPiriwqs9v9Hsrv@mail.gmail.com> <86wrljuce6.fsf@gmail.com> <AANLkTimoP=AKTk-70BTzSQDbr-XqqSTLrE=y9=0ndsPR@mail.gmail.com> <AANLkTin0OxaHcdzE8EF6t-%2BkeLjbsXKOMnsA9WdGdrTg@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Rich <rincebrain@gmail.com> writes:
>> Attempted this procedure today, make fails with:
[...]
>> /usr/local/ncvs/src/sys/modules/zfs/../../cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c:188:
>> error: 'zfs_mdcomp_disable' undeclared here (not in a function)
>
> First error is just a variable undefined that was defined in the original
> file pre-patching in the function; redefined it as defined in the original.
No, it should be removed from there. See if you have arc.c.rej.
%%
@@ -186,14 +185,11 @@
TUNABLE_QUAD("vfs.zfs.arc_max", &zfs_arc_max);
TUNABLE_QUAD("vfs.zfs.arc_min", &zfs_arc_min);
TUNABLE_QUAD("vfs.zfs.arc_meta_limit", &zfs_arc_meta_limit);
-TUNABLE_INT("vfs.zfs.mdcomp_disable", &zfs_mdcomp_disable);
SYSCTL_DECL(_vfs_zfs);
SYSCTL_UQUAD(_vfs_zfs, OID_AUTO, arc_max, CTLFLAG_RDTUN, &zfs_arc_max, 0,
"Maximum ARC size");
SYSCTL_UQUAD(_vfs_zfs, OID_AUTO, arc_min, CTLFLAG_RDTUN, &zfs_arc_min, 0,
"Minimum ARC size");
-SYSCTL_INT(_vfs_zfs, OID_AUTO, mdcomp_disable, CTLFLAG_RDTUN,
- &zfs_mdcomp_disable, 0, "Disable metadata compression");
/*
* Note that buffers can be in one of 6 states:
%%
>> /usr/local/ncvs/src/sys/modules/zfs/../../cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c:3580:
>> warning: passing argument 7 of 'zio_write' discards qualifiers from pointer target type
>
> Second error is because arc_write takes a const of that property now, and
> zio_write still doesn't, resulting in the warning as shown. Dunno what the
> correct way to fix this is, as it's unclear to me why the change was done.
Did you count wrong? The 7th argument to zio_write() around 3580 line is
`&zp' vs. `zp'.
%%
@@ -3563,8 +3573,7 @@
callback->awcb_private = private;
callback->awcb_buf = buf;
- write_policy(spa, wp, &zp);
- zio = zio_write(pio, spa, txg, bp, buf->b_data, hdr->b_size, &zp,
+ zio = zio_write(pio, spa, txg, bp, buf->b_data, hdr->b_size, zp,
arc_write_ready, arc_write_done, callback, priority, zio_flags, zb);
return (zio);
%%
> This looks more broken than I might hope to fix trivially, since working
> around the second one above results in more brokenness in dbuf.c. I suppose
> I'll wait for an updated patchset to try this out again.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?86oc6ov9lh.fsf>