Date: Tue, 01 Jul 2003 12:11:19 -0600 From: Scott Long <scottl@freebsd.org> To: Nate Lawson <nate@root.org> Cc: cvs-all@freebsd.org Subject: Re: cvs commit: src/sys/alpha/alpha busdma_machdep.csrc/sys/alpha/include busdma_machdep.c src/sys/amd64/include bus_dma.h src/ Message-ID: <3F01CEC7.6060606@freebsd.org> In-Reply-To: <20030701103637.P87367@root.org> References: <20030701155255.12BAB37B405@hub.freebsd.org> <20030701103637.P87367@root.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Nate Lawson wrote:
> On Tue, 1 Jul 2003, Scott Long wrote:
>
>>+/*
>>+ * dflt_lock should never get called. It gets put into the dma tag when
>>+ * lockfunc == NULL, which is only valid if the maps that are associated
>>+ * with the tag are meant to never be defered.
>>+ * XXX Should have a way to identify which driver is responsible here.
>>+ */
>>+static void
>>+dflt_lock(void *arg, bus_dma_lock_op_t op)
>>+{
>>+#ifdef INVARIANTS
>>+ panic("driver error: busdma dflt_lock called");
>>+#else
>>+ printf("DRIVER_ERROR: busdma dflt_lock called\n");
>>+#endif
>>+}
>>+
>
>
> backtrace()?
>
> -Nate
>
That won't be much help because this is invoked from busdma_swi, which
of course runs in an ithread.
Scott
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3F01CEC7.6060606>