Re: Calling EISA experts

[Paul Gortmaker <p_gortmaker@yahoo.com>]

Eric S. Raymond wrote:

> Bingo.  I've got reliable /proc tests for ISAPNP, PCI, and MCA.  Previous
> discussion indicates I can't get one for ISA classic.  An EISA test would,
> as ever, allow me to cut the number of questions about ancient dead
> hardware that users have to see.

Minimal approach: Register motherboard EISA ID (i.e. slot zero) ports in
/proc/ioports.  Works on all kernel versions.  See $0.02 patch below.

This is probably the least intrusive way to get what you want.  It doesn't
add Yet Another Proc File, and costs zero bloat to the 99.9% of us who 
have a better chance of meeting Aunt Tillie than an EISA box. 

Possible alternative: Create something like /proc/bus/eisa/devices which
lists the EISA ID (e.g. abc0123) found in each EISA slot.   This might
have been worthwhile some 8 years ago, but now? ....

Paul.


--- arch/i386/kernel/setup.c~	Tue Nov  6 19:14:03 2001
+++ arch/i386/kernel/setup.c	Mon Jan 21 06:22:15 2002
@@ -122,6 +122,7 @@
  */
 #ifdef CONFIG_EISA
 int EISA_bus;
+struct resource eisa_id = { "EISA ID", 0xc80, 0xc83, IORESOURCE_BUSY };
 #endif
 int MCA_bus;
 
@@ -1020,6 +1021,11 @@
 	/* request I/O space for devices used on all i[345]86 PCs */
 	for (i = 0; i < STANDARD_IO_RESOURCES; i++)
 		request_resource(&ioport_resource, standard_io_resources+i);
+
+#ifdef CONFIG_EISA
+	if (EISA_bus)
+		request_resource(&ioport_resource, &eisa_id);
+#endif
 
 	/* Tell the PCI layer not to allocate too close to the RAM area.. */
 	low_mem_size = ((max_low_pfn << PAGE_SHIFT) + 0xfffff) & ~0xfffff;