From: Alexander Bluhm Subject: Re: igc(4) tso To: Jan Klemkow Cc: Moritz Buhl , tech@openbsd.org, bket@openbsd.org Date: Thu, 2 May 2024 22:59:36 +0200 On Thu, May 02, 2024 at 09:35:56PM +0200, Jan Klemkow wrote: > On Thu, May 02, 2024 at 09:27:07PM GMT, jan@openbsd.org wrote: > > On Thu, May 02, 2024 at 08:51:26PM GMT, Moritz Buhl wrote: > > > I addressed your comment and also noticed that the diff broke with > > > previous convention of using the ISSET macro. > > > > > > Is this ok? > > > > After talking with bluhm. We come to the conclusion, that this diff > > is better: > > sorry, i got the mask wrong. This diff seems to have problems. panic: pool_cache_item_magic_check: mcl2k2 cpu free list modified: item addr 0xfffffd8014caca40+16 0x6e5268ce7298ae2a!=0xa33948c749a2c15c Stopped at db_enter+0x14: popq %rbp TID PID UID PRFLAGS PFLAGS CPU COMMAND 5130 8945 1000 0x18100002 0 3 tcpbench 387586 2317 0 0x14000 0x200 2 softnet2 468191 95807 0 0x14000 0x200 1 softnet0 db_enter() at db_enter+0x14 panic(ffffffff82103dab) at panic+0xc3 pool_cache_get(ffffffff82653fd8) at pool_cache_get+0x24a pool_get(ffffffff82653fd8,2) at pool_get+0x62 m_clget(0,2,802) at m_clget+0x1b4 igc_get_buf(ffff8000003045b0,371) at igc_get_buf+0x9f igc_rxfill(ffff8000003045b0) at igc_rxfill+0xce igc_intr_queue(ffff8000002fd690) at igc_intr_queue+0x55 intr_handler(ffff80005c047550,ffff80000009fe00) at intr_handler+0x72 Xintr_ioapic_edge6_untramp() at Xintr_ioapic_edge6_untramp+0x18f acpicpu_idle() at acpicpu_idle+0x11f sched_idle(ffff80005a616ff0) at sched_idle+0x272 end trace frame: 0x0, count: 3 https://www.openbsd.org/ddb.html describes the minimum info required in bug reports. Insufficient info makes it difficult to find and fix bugs. ddb{6}>