From: Niklas Cassel <cassel@kernel.org>
To: Damien Le Moal <dlemoal@kernel.org>,
Niklas Cassel <cassel@kernel.org>,
Shaun Tancheff <shaun@tancheff.com>, Tejun Heo <tj@kernel.org>
Cc: linux-ide@vger.kernel.org
Subject: [PATCH v2 0/2] ata: DSM TRIM fix and improvements
Date: Thu, 2 Jul 2026 12:04:11 +0200 [thread overview]
Message-ID: <20260702100410.2039383-4-cassel@kernel.org> (raw)
Hello all,
This fixes DSM TRIM for 4Kn devices, which has ironically been broken
since commit ef2d7392c4ec ("libata: SCT Write Same / DSM Trim"), which
claimed to correct the DSM TRIM handling for 4Kn devices.
The first commit is a small fix that can be backported, which limits
the DSM TRIM to what we currently report in the Block Limits VPD page.
We currently report a single 512 byte page of entries, i.e. 64 Entries,
where each entry has a maximum of U16_MAX logical blocks.
The second commit:
-Improves the support by reading the MAX PAGES PER DSM COMMAND limit and
populates the Block Limits VPD page accordingly.
-Modifies ata_scsi_write_same_xlat() to properly support up to this amount
of pages.
-Removed the intermediate copying to ata_scsi_rbuf, writes to the scsi
buffer directly.
Practically, sd still caps n_block at SD_MAX_WS16_BLOCKS
(~3 pages actually used on 4Kn), so while removing the limit on
ATA_SCSI_RBUF_SIZE (2048) - i.e. 4 pages), we will still be limited
by SD_MAX_WS16_BLOCKS. The main improvement is not longer taking the
ata_scsi_rbuf_lock and doing and extra copy from ata_scsi_rbuf.
Changes since v1:
-Picked up tags from Hannes.
-Cast to u64 to avoid potential overflow when using 32-bit arithmetic.
-If no limit is reported, assume that only a single 512-byte page is
supported.
Niklas Cassel (2):
ata: libata-scsi: fix DSM TRIM for sector sizes larger than 2048 bytes
ata: libata-scsi: scale DSM TRIM payload by MAX PAGES PER DSM COMMAND
drivers/ata/libata-scsi.c | 137 ++++++++++++++++++++++++++------------
include/linux/ata.h | 13 ++++
2 files changed, 108 insertions(+), 42 deletions(-)
--
2.55.0
next reply other threads:[~2026-07-02 10:04 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-02 10:04 Niklas Cassel [this message]
2026-07-02 10:04 ` [PATCH v2 1/2] ata: libata-scsi: fix DSM TRIM for sector sizes larger than 2048 bytes Niklas Cassel
2026-07-02 10:04 ` [PATCH v2 2/2] ata: libata-scsi: scale DSM TRIM payload by MAX PAGES PER DSM COMMAND Niklas Cassel
2026-07-02 10:20 ` sashiko-bot
2026-07-02 10:52 ` Niklas Cassel
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=20260702100410.2039383-4-cassel@kernel.org \
--to=cassel@kernel.org \
--cc=dlemoal@kernel.org \
--cc=linux-ide@vger.kernel.org \
--cc=shaun@tancheff.com \
--cc=tj@kernel.org \
/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