home bbs files messages ]

Forums before death by AOL, social media and spammers... "We can't have nice things"

   linux.debian.kernel      Debian kernel discussions      2,884 messages   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]

   Message 2,295 of 2,884   
   Niklas Cassel to Wolf   
   Bug#1120831: [regression] failed command   
   14 Jan 26 17:50:01   
   
   XPost: linux.debian.bugs.dist   
   From: cassel@kernel.org   
      
   Hello Wolf,   
      
   On Wed, Jan 14, 2026 at 03:11:50PM +0000, Wolf wrote:   
   > Here is find-max-sectors.sh output:   
   >   
   > Drive model:   
   > INTEL SSDSC2KG480G8   
   >   
   > Drive firmware:   
   > XCV10120   
   >   
   > SATA / AHCI controller:   
   > 00:17.0 SATA controller [0106]: Intel Corporation Cannon Lake Mobile PCH   
   SATA AHCI Controller [8086:a353] (rev 10)   
   >   
   > Drive values before running the test:   
   > /sys/block/sda/queue/max_hw_sectors_kb:32767   
   > /sys/block/sda/queue/max_sectors_kb:1280   
   > /sys/block/sda/queue/read_ahead_kb:128   
   >   
   > Running test with max_sectors 128 KiB   
   > Test: PASS   
   >   
   > Running test with max_sectors 1024 KiB   
   > Test: PASS   
   >   
   > Running test with max_sectors 2048 KiB   
   > Test: PASS   
   >   
   > Running test with max_sectors 3072 KiB   
   > Test: PASS   
   >   
   > Running test with max_sectors 4095 KiB   
   > Test: PASS   
   >   
   > Running test with max_sectors 4096 KiB   
   > Test: PASS   
      
   It is quite unexpected that 4096 KiB passes.   
      
   find-max-sectors.sh performs reads.   
      
   The commands that timed out according to dmesg shared were   
   also reads:   
      
   Dec 10 18:58:49 kernel: ata1.00: failed command: READ FPDMA QUEUED   
   Dec 10 18:58:49 kernel: ata1.00: cmd 60/00:18:50:4a:4c/20:00:0c:00:00/40 tag 3   
   ncq dma 4194304 in res 40/00:00:00:00:00/00:00:00:00:00/00 Emask 0x4 (timeout)   
      
   And also of size 4194304 == 4096 KiB.   
      
      
   The script has been able to detect different limits for other   
   "broken" drives, and was used to get the quirk value for e.g.   
   commit 2e9832713631 ("ata: libata-core: Quirk DELLBOSS VD max_sectors"),   
   so I wonder why it is not workin here (why 4 MiB reads passes).   
      
   Damien, ideas?   
      
      
   Kind regards,   
   Niklas   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)   

[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]


(c) 1994,  bbs@darkrealms.ca