Grumble.  After a week of uptime with my tulip patches
(the ones I posted here plus some cleanup patches) my
kernel gave me this:

------------[ cut here ]------------
WARNING: at net/sched/sch_generic.c:255 dev_watchdog+0xd9/0x137()
NETDEV WATCHDOG: cable (tulip): transmit queue 0 timed out
CPU: 0 PID: 0 Comm: swapper/0 Not tainted 3.10.2-00033-gca137af #47
Hardware name: MICRO-STAR INTERNATIONAL CO.,LTD MS-7376/MS-7376, BIOS V1.7 01/13/2009
 0000000000000000 ffffffff81028aba ffff88021fc03e68 ffff880216b38000
 ffff88021fc03eb8 ffffffff8133643a ffffffff8133643a ffffffff81028b33
 ffffffff8159fe93 ffff880200000030 ffff88021fc03ec8 ffff88021fc03e88
Call Trace:
 <IRQ>  [<ffffffff81028aba>] ? warn_slowpath_common+0x56/0x6b
 [<ffffffff8133643a>] ? qdisc_rcu_free+0x19/0x19
 [<ffffffff8133643a>] ? qdisc_rcu_free+0x19/0x19
 [<ffffffff81028b33>] ? warn_slowpath_fmt+0x47/0x49
 [<ffffffff8133634e>] ? netif_tx_lock+0x47/0x72
 [<ffffffff81336513>] ? dev_watchdog+0xd9/0x137
 [<ffffffff81031f40>] ? call_timer_fn.isra.29+0x1c/0x6f
 [<ffffffff810321a2>] ? run_timer_softirq+0x19a/0x1c2
 [<ffffffff8102de09>] ? __do_softirq+0xb9/0x171
 [<ffffffff8102df7f>] ? irq_exit+0x3a/0x7a
 [<ffffffff8101967a>] ? smp_apic_timer_interrupt+0x72/0x7e
 [<ffffffff8142a60a>] ? apic_timer_interrupt+0x6a/0x70
 <EOI>  [<ffffffff81007e34>] ? amd_e400_idle+0xbf/0xc1
 [<ffffffff81007e2c>] ? amd_e400_idle+0xb7/0xc1
 [<ffffffff8104f74d>] ? cpu_startup_entry+0x9c/0xec
 [<ffffffff8164eb91>] ? start_kernel+0x2bd/0x2c8
 [<ffffffff8164e6f7>] ? repair_env_string+0x54/0x54
---[ end trace fa3269ab5c1a15ad ]---

Since it's on the tulip device, it's probably my fault, and
since it took a week to manifest itself, it's going to be
a complete PITA to reproduce.

Anyway, I haven't forgotten.