Messages produced by cfi_probe.c and cfi_cmdset_0001.c

Messages produced by cfi_probe.c and cfi_cmdset_0001.c

[David Vrabel]

Hi,

cfi_probe.c and cfi_cmdset_0001.c both produce large quantities of
output when probing. (eg all of the CFI query data).  Is is necessary? 
Can I remove it (comment it out,#ifdef, debug only messages etc.)?

Also, in cfi_cmdset_0001.c in
cfi_intelext_write_{2_by_8,1_by_16}_buffers() we always write a full
buffer even if we have less than a buffer to write.  This affects
performace (particularly when writing short <32byte buffers).  Is there
a reason why this is done like this?

David Vrabel



To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org

Re: Messages produced by cfi_probe.c and cfi_cmdset_0001.c

[David Woodhouse]

dvrabel@arcom.co.uk said:
>  cfi_probe.c and cfi_cmdset_0001.c both produce large quantities of
> output when probing. (eg all of the CFI query data).  Is is necessary?
>  Can I remove it (comment it out,#ifdef, debug only messages etc.)?

It's only there because it came long before the actual access routines, and 
it make me happy to think that the code could do something useful :)

By all means, wrap it in #if 0


> Also, in cfi_cmdset_0001.c in cfi_intelext_write_{2_by_8,1_by_16}_buffe
> rs() we always write a full buffer even if we have less than a buffer
> to write.  This affects performace (particularly when writing short
> <32byte buffers).  Is there a reason why this is done like this?

Laziness. Please do fix it.

--
dwmw2




To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org