Hello.
On 08-04-2011 3:57, Greg KH wrote:
On Thu, Apr 07, 2011 at 09:56:54AM -0700, Kristen Carlson Accardi wrote:
Have you made timing/performance studies, comparing (for example) USB
disk throughput with and without threaded interrupts?
I could never do this with all the different types of hw. I suggest
that if you decide to move to a threaded irq you have a long period of
testing time where people can determine whether or not it impacts
performance on their own hw. Just from googling I can see that tglx
did some performance analysis, but you'd have to ask him what the
findings were.
I asked him about this an hour ago, and he found that he got a
performance increase in USB throughput switching to threaded irqs,
especially for usb-storage devices.
So I'm all for it now, send me the patch and let's see what breaks :)
thanks,
greg k-h
I sent him an email to see if he'll submit the patches he's been
testing with -- I figured it'd be better to use those since he's
got time on them already.
Yes, it turns out that your patch will not work properly on some
hardware in the first place, so his changes are needed. I see him
working on them right now in front of me in the conference meeting we
are currently sitting in...
Here's Thomas's first cut at doing this, totally untested and written
during a legal track at the LF Collab summit.
Kristen, can you take this and test and see if it works for you?
thanks,
greg k-h
Subject: usb.patch
From: Thomas Gleixner<tglx@xxxxxxxxxxxxx>
Date: Thu, 07 Apr 2011 23:35:16 +0200
Signed-off-by: Thomas Gleixner<tglx@xxxxxxxxxxxxx>
[...]
Index: linux-2.6/drivers/usb/core/hcd.c
===================================================================
--- linux-2.6.orig/drivers/usb/core/hcd.c
+++ linux-2.6/drivers/usb/core/hcd.c
[...]
@@ -2189,7 +2191,7 @@ EXPORT_SYMBOL_GPL (usb_hc_died);
/**
* usb_create_shared_hcd - create and initialize an HCD structure
* @driver: HC driver that will use this hcd
- * @dev: device for this HC, stored in hcd->self.controller
+s * @dev: device for this HC, stored in hcd->self.controller
I guess the patch won't compile because of this stray character...
WBR, Sergei
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
