ata fixes for 6.10-rc6

Hello Linus,

The following changes since commit fa997b0576c9df635ee363406f5e014dba0f9264:

ata: ahci: Do not enable LPM if no LPM states are supported by the HBA (2024-06-19 13:19:38 +0200)

are available in the Git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/libata/linux tags/ata-6.10-rc6

for you to fetch changes up to eeb25a09c5e0805d92e4ebd12c4b0ad0df1b0295:

ata: ahci: Clean up sysfs file on error (2024-06-30 22:23:39 +0200)

----------------------------------------------------------------
ata fixes for 6.10-rc6 (or 6.10-rc7 if this PR was sent out too late.)

- Add NOLPM quirk for for all Crucial BX SSD1 models. Considering that we now have had bug reports for 3 different BX SSD1 variants from Crucial with the same product name, make the quirk more inclusive, to catch more device models from the same generation.

- Fix a trivial null pointer dereference in the error path for ata_host_release().

- Create a ata_port_free(), so that we don't miss freeing ata_port struct members when freeing a struct ata_port.

- Fix a trivial double free in the error path for ata_host_alloc().

- Ensure that we remove the libata "remapped NVMe device count" sysfs entry on .probe() error.

----------------------------------------------------------------
Niklas Cassel (5):
ata: libata-core: Add ATA_HORKAGE_NOLPM for all Crucial BX SSD1 models
ata: libata-core: Fix null pointer dereference on error ata,scsi: libata-core: Do not leak memory for ata_port struct members
ata: libata-core: Fix double free on error
ata: ahci: Clean up sysfs file on error

 drivers/ata/ahci.c                 | 17 ++++++++++++-----
 drivers/ata/libata-core.c          | 32 +++++++++++++++++++-------------
 drivers/scsi/libsas/sas_ata.c      |  6 +++---
 drivers/scsi/libsas/sas_discover.c |  2 +-
 include/linux/libata.h             |  1 +
 5 files changed, 36 insertions(+), 22 deletions(-)

So I've pulled these, but I'm not happy with how they seem to be _very_ recent.

Yes, they all look trivial, but even trivial commits have bugs, and it looks like they have had absolutely zero testing. Four out of five commits were committed about an hour ago.

So please - don't do this. I was close to just unpulling it all again, and if this had been some final -rc, that's what I would have done. This kind of "last-minute commits in last-minute pull request" is not ok.

Linus

The four out of five patches were originally part of a larger series, sent out on Wednesday:
https://lore.kernel.org/linux-ide/[email protected]/

They have been tested using:
rmmod + modprobe + rmmod + modprobe on:
-AMD AHCI controller
-QEMU AHCI controller
-pm8001 SAS (libsas) controller

Both with a debug patch that forced an error,
and without any additional debug patch.

Ensuring that:
-The error/crash was fixed (in case of forced error) -That all devices show up as intended (in case of no forced error)

So it probably looks worse than it is.

Understood!

FWIW, even with the testing done, I would never even have considered doing the same if we were later in the release cycle.

Kind regards,
Niklas