From: Douglas Gilbert <dgilbert@interlog.com>
To: SCSI development list <linux-scsi@vger.kernel.org>
Cc: "Tomáš Bžatek" <tbzatek@redhat.com>,
"Martin Pitt <mpitt@debian.org> Hannes Reinecke <hare@suse.de>
Ritesh Raj Sarraf <rrs@researchut.com> Robin H. Johnson
<robbat2@gentoo.org> Martin Wilck" <mwilck@suse.com>,
"Paul Evans" <pevans@redhat.com>,
"Alan Somers" <asomers@freebsd.org>
Subject: [Announce] sg3_utils-1.49 available
Date: Wed, 29 Jul 2026 19:33:40 -0400 [thread overview]
Message-ID: <65c9f013-eb13-43bc-a9b1-e98f05e19b26@interlog.com> (raw)
It has been a while. I no longer maintain the sg and scsi_debug drivers
in the Linux kernel but I do plan to maintain my user space packages.
First among these packages is sg3_utils which is 26 years old (if one
includes when it was part of the older sg_utils package).
For an overview of sg3_utils and downloads see either of these pages:
https://sg.danny.cz/sg/sg3_utils.html
https://doug-gilbert.github.io/sg3_utils.html
The sg_ses utility (for enclosure devices) is discussed at:
https://doug-gilbert.github.io/sg_ses.html
https://sg.danny.cz/sg/sg_ses.html
A full changelog can be found at:
https://sg.danny.cz/sg/p/sg3_utils.ChangeLog
https://doug-gilbert.github.io/p/sg3_utils.ChangeLog
My own github site:
https://github.com/doug-gilbert/sg3_utils
The last link has a 'release-1.49' tag which shows the state of the code
at the time of release [20260718]. There have been some commits since
then but none that seriously impact Linux. Some of the above pages are
awaiting updates, All the installed utilities and scripts have (or should
have) up to date manpages which are also installed. Those manpages try
to do more that list the command line options, in some cases they try to
explain "why". Some, but not all, manpages have an EXAMPLES section.
One way to report issues is with a patch via the github PR mechanism, or
raise a github Issue. Emailing me directly with a patch attached is also
good. Please note, I am no longer subscribed to this list (nor the
linux-kernel list).
Cmake build infrastructure has been added and the intention is that both
it and GNU Autotools generate the same outputs (easier said than
done). Cmake's CTest looks interesting (but has not been "wired" yet)
while cpack is functional and very useful.
The ChangeLog is long, so it has been broken into 2 parts, the first part
has utility specific changes, listed in alphabetical order. The second part
contains more general changes: to the common library or the build scripts.
The github log messages often supply more information (than a one line
summary). Note that subversion is lurking in the background with simple to
follow concepts like ascending integers for revision numbers (e.g. r1133)
ChangeLog for release sg3_utils-1.49 [20260718] [svn: r1133]
Utility specific changes:
- rescan-scsi-bus.sh: fix multipath resize without update
https://github.com/doug-gilbert/sg3_utils/pull/43.diff
and 44.diff
- fix for github.com/doug-gilbert/sg3_utils/issues/46
- github PR numbers: 68, 69, 71 and 74 by Paul Evans
- github pull req: #72 fix udevadm invocation M. Wilck
- github issue: #76 fix shellcheck warnings Xose Vazquez Perez
- scsi_logging_level: add MIT license Xose Vazquez Perez
- scsi-rescan.8 manpage added, links to rescan-scsi-bus.8
- sg_copy_results: part of github PR #74
- sg_dd: add grpnum=GN operand and --nocopy option
- sg_get_elem_status: fractional elements, sbc6r02
- correct several error messages
- sg_get_lba_status: additional LBA accessibility values
- sginfo: part of github PR #74
- sg_inq: update version descriptors to T10 table 20230814
- re-introduce the --ata option, hidden since rel 1.43
- sg_inq+sg_vpd: show depopulation time in hours and
minutes as well as its raw form: seconds
- add RSCS field to Block limits extension vpd page [sbc5r5]
- translation of NVMe FR field uses sg_last_n_non_blank()
- add NVMe Information VPD page
- clean up ATA Information VPD page
- rework designation descriptor handling
- sg_inq --export conformance for SCSI name string
by Paul Evans github PR #83
- sg_lib: add sg_last_n_non_blank() for SNT FR field
- sg_lib.[hc] add sg_strscpy() to replace strncpy()
- sg_json.c fixes from lsscsi project pull #15
- sg_json.[hc] add sgj_haj_vs_hex_bytes_nex() for
outputting IDs as ASCII hex strings
- sg_json.[hc] add sgj_haj_vs_len()
- sg_json.[hc] add sgj_haj_vs_nex()
- sg_json_sg_lib.[hc] sgj_js_designation_descriptor()
changed to sgj_haj_designation_descriptor()
- sg_logs: correct tape Device statistics lpage [0x14]
serial number parameters [0x40 and 0x41] per
https://github.com/doug-gilbert/sg3_utils/issues/45
- sg_map: handle LUNs > 255
- sg_opcodes: fix rwcdl/cdlp fields handling D. Le Moal
- sg_rem_rest_elem: add support for REMOVE ELEMENT AND
- MODIFY ZONES command, Damien Le Moal
- sg_requests: improve buffer truncation
- sg_sat_read_gplog: list names of supported GP log addresses
- sg_ses: restore some help text, github PR #79 Paul Evans
- sg_stream_ctl: add --hex, --inhex=, --json= and --raw
options; additions in sbc5r05 (perm and relative_lifetime
fields).
- sg_turs: fix missing CDB on several codepaths
- sg_vpd: inform about SPC-6 (sg_inq knew already)
https://github.com/doug-gilbert/sg3_utils/issues/40
- add --descriptors option, for version descriptors
- part of github PR #74
- some sg_vpd fixes shown under sg_inq+sg_vpd entry above
- sg_wr_mode: fix --contents and --cfile= handling
- add --mfile=MF option for decoding mask in the MF file
- fix introduced Makefile.am bug, plus only for Linux
- sg_xcopy: part of github PR #74
- utils/hxascdmp: sprint -> snprintf [github PR #78]
General changes:
- fix 'make distcheck' (again)
- add cmake build system (alongside autotools) with cpack
for deb and rpm packages; ctest not wired yet
- github PR #80 by Alan Somers which is associated
with cpack FreeBSD settings
- github PR #82 by Paul Evans to get cmake output to be
equivalent (in parity) with what Autotools produces
- wire option(DISABLE_LIBSGUTILS (def: OFF)) so when
ON sgutils2 library is static but other libraries
are shared objects (e.g. libc.so)
- github PR #84 by Paul Evans to fix cpack generated
Debian dev library path, needed by sdparm and ddpt
- retire all uses of lseek64() as _FILE_OFFSET_BITS=64
compile define should take care of older 32 bit cases
- fixes so musl-gcc builds clean
- start complying with T10 23-047r1 "Broadcom SNT Reference"
document assuming T10 SNT drafts are coming
- rearrange SNT and NVMe headers and associated source
- new error code: SG_LIB_PROGRESS_NOT_READY for long
duration 'in progress' additional sense codes
- lib/sg_pt_solaris.c : fix compilation issue per
https://github.com/doug-gilbert/sg3_utils/pull/41
- lib/sgj_haj_*() functions: tweak hex_haj argument meaning
- scripts/00-scsi-sg3_config.rules : apply
https://github.com/doug-gilbert/sg3_utils/pull/42
- rename build_debian.sh script to debian_build.sh
- rename testing directory to ttesting to avoid clash with
cmake's CTEST that creates a Testing directory
- ttesting/sg_chk_inq_vd.c: test internal table against T10
version descriptor file
- check/clean against gcc-16 -fanalyzer; clang-22 --analyze
- debian build: modernize with dh sequencer [github PR# 73]
- AIX partial support: needs getopt.h header which
surprisingly is non-standard. Keep hidden version in
lib directory and copy to include dir as needed
- still no real pass-through, using sg_dummy instead
- getopt_long directory contents -> lib directory
- JSON: make output more consistent so most command
responses have a *_parameter_data or similar sub-object
- apply https://github.com/doug-gilbert/sg3_utils/pull/39
and its revision [20230807] mainly for Android
Changelog for released sg3_utils-1.48 [20230801] [svn: r1042]
.....
Douglas Gilbert
next reply other threads:[~2026-07-29 23:40 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-29 23:33 Douglas Gilbert [this message]
2026-08-01 13:45 ` [Announce] sg3_utils-1.49 available Xose Vazquez Perez
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=65c9f013-eb13-43bc-a9b1-e98f05e19b26@interlog.com \
--to=dgilbert@interlog.com \
--cc=asomers@freebsd.org \
--cc=linux-scsi@vger.kernel.org \
--cc=mwilck@suse.com \
--cc=pevans@redhat.com \
--cc=tbzatek@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox