Thomas Huth <th...@redhat.com> writes: > On 28/01/2024 21.22, Sven Schnelle wrote: >> When the maximum count of SCRIPTS instructions is reached, the code >> stops execution and returns, but fails to decrement the reentrancy >> counter. This effectively renders the SCSI controller unusable >> because on next entry the reentrancy counter is still above the limit. >> This bug was seen on HP-UX 10.20 which seems to trigger SCRIPTS >> loops. > > Out of curiosity: What happened there before we introduced the > reentrancy_level fix? Did it end up in an endless loop, or was it > finishing at one point? In the latter case, we might need to adjust > the "reentrancy_level > 8" to allow deeper nesting.
Without the reentrancy counter it was triggering the insn_processed limit. The HP-UX scsi driver seems to spin on some memory value during some SCSI writes (CDB with command 0x2a). So it is spinning in an endless loop until the insn_processed counter will trigger the exit. In HP-UX you will see a SCSI command timeout error in the kernel log - at least i'm assuming that's related, but can't say for sure as there's no kernel source available.
