From: Linus Torvalds <torvalds@linux-foundation.org>
To: Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
Cc: Sparse Mailing-list <linux-sparse@vger.kernel.org>,
Bart Van Assche <bvanassche@acm.org>
Subject: Re: [PATCH 0/5] allow -1 and compares in bitwise types
Date: Mon, 27 Jun 2022 12:14:28 -0700 [thread overview]
Message-ID: <CAHk-=wiRzOgEXT95o+ZQUawKbwBaGHoy=TSzB1dN2Uh26CZfCQ@mail.gmail.com> (raw)
In-Reply-To: <20220627190540.13358-1-luc.vanoostenryck@gmail.com>
On Mon, Jun 27, 2022 at 12:05 PM Luc Van Oostenryck
<luc.vanoostenryck@gmail.com> wrote:
>
> #define is_signed_type(type) (((type)-1) <= 0)
Side note: the reason we *don't* use this form in the kernel is
because broken compilers will complain about compares with zero in
unsigned types.
Using "<=" like you do may be an acceptable way to avoid it (the most
obvious thing is to use "< 0"), but it makes me nervous.
Regardless, I think you need the cast of the zero. I think "type"
might be a pointer, and sparse should be complaining about the horrid
use of a bare 0 as NULL.
Similar issues might happen for enums, where various compilers will
complain about comparing an enum to a non-enum.
So I'm pretty sure you would want casts on both values, instead of
assuming "it's an integer type, I don't need to cast 0".
But yeah, maybe
#define is_signed_type(type) (((type)-1) <= (type)0)
works fine and avoids warnings in all the cases.
Famous last words. Warnings can happen for almost anything, and I
wonder if that use of "1" had some other reason.
Linus
next prev parent reply other threads:[~2022-06-27 19:14 UTC|newest]
Thread overview: 93+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-06-23 18:04 [PATCH 00/51] Improve static type checking for request flags Bart Van Assche
2022-06-23 18:04 ` [PATCH 01/51] treewide: Rename enum req_opf into enum req_op Bart Van Assche
2022-06-23 18:04 ` [PATCH 02/51] block: Use enum req_op where appropriate Bart Van Assche
2022-06-23 18:04 ` [PATCH 03/51] block: Change the type of the last .rw_page() argument Bart Van Assche
2022-06-23 18:04 ` [PATCH 04/51] block: Change the type of req_op() and bio_op() into enum req_op Bart Van Assche
2022-06-23 18:04 ` [PATCH 05/51] block: Introduce the type blk_opf_t Bart Van Assche
2022-06-23 18:04 ` [PATCH 06/51] block: Use the new blk_opf_t type Bart Van Assche
2022-06-23 18:04 ` [PATCH 07/51] blktrace: " Bart Van Assche
2022-06-23 18:04 ` [PATCH 08/51] blktrace: Trace remap operations correctly Bart Van Assche
2022-06-23 18:04 ` [PATCH 09/51] block/brd: Use the enum req_op type Bart Van Assche
2022-06-23 18:04 ` [PATCH 10/51] block/drbd: Use the enum req_op and blk_opf_t types Bart Van Assche
2022-06-24 5:00 ` Christoph Böhmwalder
2022-06-23 18:04 ` [PATCH 11/51] block/floppy: Fix a sparse warning Bart Van Assche
2022-06-23 18:04 ` [PATCH 12/51] block/null_blk: Fix sparse warnings in tracing code Bart Van Assche
2022-06-23 18:04 ` [PATCH 13/51] um: Use enum req_op where appropriate Bart Van Assche
2022-06-23 18:04 ` [PATCH 14/51] dm/core: Use the enum req_op and blk_opf_t types Bart Van Assche
2022-06-23 18:04 ` [PATCH 15/51] dm/bufio: Change 'int rw' into 'enum req_op op' Bart Van Assche
2022-06-23 18:04 ` [PATCH 16/51] dm/kcopyd: Rename kcopyd_job.rw into kcopyd_job.op Bart Van Assche
2022-06-23 18:04 ` [PATCH 17/51] dm/ebs: Change 'int rw' into 'enum req_op op' Bart Van Assche
2022-06-23 18:04 ` [PATCH 18/51] dm/dm-flakey: Use the new blk_opf_t type Bart Van Assche
2022-06-23 18:04 ` [PATCH 19/51] dm/dm-integrity: Use the enum req_op and blk_opf_t types Bart Van Assche
2022-06-23 18:04 ` [PATCH 20/51] dm/dm-snap: " Bart Van Assche
2022-06-23 18:04 ` [PATCH 21/51] dm/dm-zoned: Use the enum req_op type Bart Van Assche
2022-06-23 22:46 ` Damien Le Moal
2022-06-23 18:04 ` [PATCH 22/51] md/core: Use the enum req_op and blk_opf_t types Bart Van Assche
2022-06-23 18:05 ` [PATCH 23/51] md/bcache: " Bart Van Assche
2022-06-23 18:05 ` [PATCH 24/51] md/raid1: Use the new blk_opf_t type Bart Van Assche
2022-06-23 18:05 ` [PATCH 25/51] md/raid10: " Bart Van Assche
2022-06-23 18:05 ` [PATCH 26/51] md/raid5: Use the enum req_op and blk_opf_t types Bart Van Assche
2022-06-23 18:05 ` [PATCH 27/51] nvme/host: " Bart Van Assche
2022-06-26 9:25 ` Sagi Grimberg
2022-06-23 18:05 ` [PATCH 28/51] nvme/target: Use the new blk_opf_t type Bart Van Assche
2022-06-26 9:25 ` Sagi Grimberg
2022-06-23 18:05 ` [PATCH 29/51] scsi/core: Improve static type checking Bart Van Assche
2022-06-23 18:05 ` [PATCH 30/51] scsi/core: Change the return type of scsi_noretry_cmd() into bool Bart Van Assche
2022-06-23 18:05 ` [PATCH 31/51] scsi/core: Use the new blk_opf_t type Bart Van Assche
2022-06-23 18:05 ` [PATCH 32/51] scsi/device_handlers: " Bart Van Assche
2022-06-23 18:05 ` [PATCH 33/51] scsi/ufs: Rename a 'dir' argument into 'op' Bart Van Assche
2022-06-23 19:09 ` Avri Altman
2022-06-23 18:05 ` [PATCH 34/51] scsi/target: Use the new blk_opf_t type Bart Van Assche
2022-06-23 18:05 ` [PATCH 35/51] mm: " Bart Van Assche
2022-06-23 18:05 ` [PATCH 36/51] fs/buffer: " Bart Van Assche
2022-06-23 18:05 ` [PATCH 37/51] fs/direct-io: Use the enum req_op and blk_opf_t types Bart Van Assche
2022-06-23 18:05 ` [PATCH 38/51] fs/mpage: Use the new blk_opf_t type Bart Van Assche
2022-06-23 18:05 ` [PATCH 39/51] fs/btrfs: Use the enum req_op and blk_opf_t types Bart Van Assche
2022-06-23 18:05 ` [PATCH 40/51] fs/ext4: Use the new blk_opf_t type Bart Van Assche
2022-06-23 18:05 ` [PATCH 41/51] fs/f2fs: Use the enum req_op and blk_opf_t types Bart Van Assche
2022-06-23 18:05 ` [PATCH 42/51] fs/gfs2: " Bart Van Assche
2022-06-23 18:05 ` [PATCH 43/51] fs/hfsplus: " Bart Van Assche
2022-06-23 18:05 ` [PATCH 44/51] fs/iomap: Use the new blk_opf_t type Bart Van Assche
2022-06-23 18:05 ` [PATCH 45/51] fs/jbd2: Fix the documentation of the jbd2_write_superblock() callers Bart Van Assche
2022-06-23 18:05 ` [PATCH 46/51] fs/nilfs: Use the enum req_op and blk_opf_t types Bart Van Assche
2022-06-24 3:43 ` Ryusuke Konishi
2022-06-25 0:02 ` Bart Van Assche
2022-06-25 0:55 ` Ryusuke Konishi
2022-06-23 18:05 ` [PATCH 47/51] fs/ntfs3: Use enum req_op where appropriate Bart Van Assche
2022-06-23 18:05 ` [PATCH 48/51] fs/ocfs2: Use the enum req_op and blk_opf_t types Bart Van Assche
2022-06-24 2:04 ` Joseph Qi
2022-06-23 18:05 ` [PATCH 49/51] PM: " Bart Van Assche
2022-06-27 13:04 ` Rafael J. Wysocki
2022-06-23 18:05 ` [PATCH 50/51] fs/xfs: " Bart Van Assche
2022-06-25 3:23 ` Darrick J. Wong
2022-06-23 18:05 ` [PATCH 51/51] fs/zonefs: Fix sparse warnings in tracing code Bart Van Assche
2022-06-23 22:48 ` Damien Le Moal
2022-06-23 23:19 ` Bart Van Assche
2022-06-24 0:03 ` Damien Le Moal
2022-06-24 4:56 ` Christoph Hellwig
2022-06-24 19:57 ` Bart Van Assche
2022-06-25 9:23 ` Christoph Hellwig
2022-06-26 0:44 ` Bart Van Assche
2022-06-26 9:58 ` Luc Van Oostenryck
2022-06-26 15:42 ` Bart Van Assche
2022-06-26 16:24 ` Luc Van Oostenryck
2022-06-26 16:33 ` Linus Torvalds
2022-06-26 16:50 ` Linus Torvalds
2022-06-26 20:10 ` Luc Van Oostenryck
2022-06-26 19:44 ` Luc Van Oostenryck
2022-06-27 19:05 ` [PATCH 0/5] allow -1 and compares in bitwise types Luc Van Oostenryck
2022-06-27 19:05 ` [PATCH 1/5] bitwise: add testcases Luc Van Oostenryck
2022-06-27 19:05 ` [PATCH 2/5] bitwise: accept all ones as non-restricted value Luc Van Oostenryck
2022-06-27 23:32 ` Ramsay Jones
2022-06-27 19:05 ` [PATCH 3/5] bitwise: allow compares for bitwise types Luc Van Oostenryck
2022-06-27 19:20 ` Linus Torvalds
2022-06-27 23:34 ` Ramsay Jones
2022-06-27 19:05 ` [PATCH 4/5] bitwise: do not remove the signedness of " Luc Van Oostenryck
2022-06-27 19:05 ` [PATCH 5/5] bitwise: early expansion of simple constants Luc Van Oostenryck
2022-06-27 19:14 ` Linus Torvalds [this message]
2022-06-27 19:15 ` [PATCH 0/5] allow -1 and compares in bitwise types Bart Van Assche
2022-06-26 19:41 ` [PATCH 51/51] fs/zonefs: Fix sparse warnings in tracing code Steven Rostedt
2022-06-24 5:05 ` [PATCH 00/51] Improve static type checking for request flags Christoph Hellwig
2022-06-24 5:07 ` Christoph Hellwig
2022-06-28 23:10 ` Bart Van Assche
2022-06-28 22:44 ` Bart Van Assche
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='CAHk-=wiRzOgEXT95o+ZQUawKbwBaGHoy=TSzB1dN2Uh26CZfCQ@mail.gmail.com' \
--to=torvalds@linux-foundation.org \
--cc=bvanassche@acm.org \
--cc=linux-sparse@vger.kernel.org \
--cc=luc.vanoostenryck@gmail.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.