Hi, On Wed, Jan 14, 2026 at 11:10:42AM +0000, Wolf wrote: > > > On Wednesday, 14 January 2026 at 12:14, Salvatore Bonaccorso > <[email protected]> wrote: > > > > > > > Hi Wolf, > > > > On Wed, Jan 14, 2026 at 09:17:27AM +0000, Wolf wrote: > > > > > On Sunday, 11 January 2026 at 22:15, Salvatore Bonaccorso > > > [email protected] wrote: > > > > > > > Hi, > > > > > > > > So there is a similar (but not identical, and different disks) report > > > > at https://bugzilla.kernel.org/show_bug.cgi?id=220693 pointing at the > > > > same commit 9b8b84879d4a ("block: Increase BLK_DEF_MAX_SECTORS_CAP") . > > > > > > > > That particular one was fixed with 2e9832713631 ("ata: libata-core: > > > > Quirk DELLBOSS VD max_sectors"). > > > > > > > > Let's see if your report rings something to upstream. > > > > > > > > Regards, > > > > Salvatore > > > > > > Hi, > > > > > > Workaround: add the rule > > > > > > ACTION=="add|change", SUBSYSTEM=="block", KERNEL=="sda", > > > ATTR{queue/max_sectors_kb}="1280" > > > > > > in /etc/udev/rules.d/99-sda-max-sectors.rules. > > > > > > It works on linux/6.18.3+deb14-amd64. > > > > > > I think we will need a similar quirk as in the above bug for your > > device. Can you please test the following patch? > > > > diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c > > index 09d8c035fcdf..8434110a4962 100644 > > --- a/drivers/ata/libata-core.c > > +++ b/drivers/ata/libata-core.c > > @@ -4108,6 +4108,7 @@ static const struct ata_dev_quirks_entry > > __ata_dev_quirks[] = { > > / > > { "LITEON CX1-JB-HP", NULL, ATA_QUIRK_MAX_SEC_1024 }, > > { "LITEON EP1-", NULL, ATA_QUIRK_MAX_SEC_1024 }, > > + { "INTEL SSDSC2KG480G8", "XCV10120", ATA_QUIRK_MAX_SEC_1024 }, > > > > / > > * These devices time out with higher max sects. > > > > I will forward the bug upstream and keep you in the loop, but would be > > > > Regards, > > Salvatore > > Hi, > > The patch fixed the bug in linux/6.16.0-rc4+
Perfect, thanks for the confirmation. It is not the correct value to use, but I will now forward the report so we can gather input from upstream on how it should be handled. Regards, Salvatore
