|
From: Dan L. M. <dma...@re...> - 2007-03-29 13:42:47
|
Hello there, I'm new to the list. I'd like to know what is needed to the lib so it can "talk" to 2.0 usb devices at full speed. I'm sorry if this has been already discussed, but I couldn't access the archives. Thanks in advance. Dan Leite Martins |
|
From: Peter S. <stu...@cd...> - 2007-03-29 15:16:05
|
Hi, On Thu, Mar 29, 2007 at 10:45:05AM -0300, Dan Leite Martins wrote: > Hello there, > > I'm new to the list. I'd like to know what is needed to the lib so > it can "talk" to 2.0 usb devices at full speed. I'm sorry if this > has been already discussed, but I couldn't access the archives. > Thanks in advance. Nothing. libusb 0.1 works with 2.0 full speed devices. Note that 0.1 is synchronous only however. Ie. only one outstanding request on the bus at any time, which wastes a lot of bus bandwidth. To get maximum throughput, you will need to use at libusb 1.0 or OpenUSB and their asynchronous features to keep many transfer queued at once. //Peter |
|
From: Peter S. <stu...@cd...> - 2007-03-29 16:51:33
|
Hi Dan, please keep the discussion on the list. Thanks! On Thu, Mar 29, 2007 at 12:39:11PM -0300, Dan Leite Martins wrote: > Thanks for the answer Peter. Synchronous communication would work > for me, but if asynchronous were available it would work much > better! However, I didn't know that libusb 1.0 was already > developed, and also don't know much about OpenUSB. Is there any > links where I can read more about them? Just to clarify, I'm using > LibUsb-win32, so maybe that's the reason I didn't see anything > about 1.0. Thank you for the attention. Aha, no I don't know about win32. libusb-win32 has been/is a separate project, but there has been talk about merging efforts for 1.0. I don't know if there's a usable version of libusb 1.0 on win32? //Peter |
|
From: Tim R. <ti...@pr...> - 2007-03-29 17:04:12
|
Dan Leite Martins wrote: > I'm new to the list. I'd like to know what is needed to the lib so it can "talk" to 2.0 usb devices at full speed. I'm sorry if this has been already discussed, but I couldn't access the archives. Thanks in advance. > Peter already provided an answer for this, but it occurred to me that your question could mean one of several things. The term "full speed" is somewhat overloaded. It could mean "at the maximum speed possible", in which case Peter was right all along. But "full speed" is also the official term for the higher 12Mbps bit rate offered by USB 1.x, which is the middle speed grade for USB 2.0. The maximum speed rate in USB 2.0 is called "high speed". So, you MIGHT have been asking "Can I talk to a USB 2.0 device as if it were hooked up to USB 1.1?" That has a very different answer (no). -- Tim Roberts, ti...@pr... Providenza & Boekelheide, Inc. |
|
From: Jake J. <ja...@ja...> - 2007-03-29 17:07:18
|
Another slant on this... If your intent is to talk to the device at full-speed to test its functionality on a USB 1.1 system, you can often setup the driver for your USB host card to operate in USB 1.1 mode. We use this for testing. We don't have any USB 1.1 hosts here, so we set our USB 2.0 host to operate as a 1.1 host and run tests on our device. Jake > -----Original Message----- > From: lib...@li... > [mailto:lib...@li...] On Behalf > Of Tim Roberts > Sent: Thursday, March 29, 2007 10:04 AM > To: lib...@li... > Subject: Re: [Libusb-devel] USB 2.0 > > Dan Leite Martins wrote: > > I'm new to the list. I'd like to know what is needed to the > lib so it can "talk" to 2.0 usb devices at full speed. I'm > sorry if this has been already discussed, but I couldn't > access the archives. Thanks in advance. > > > > Peter already provided an answer for this, but it occurred to > me that your question could mean one of several things. The > term "full speed" > is somewhat overloaded. It could mean "at the maximum speed > possible", in which case Peter was right all along. > > But "full speed" is also the official term for the higher > 12Mbps bit rate offered by USB 1.x, which is the middle speed > grade for USB 2.0. > The maximum speed rate in USB 2.0 is called "high speed". > So, you MIGHT have been asking "Can I talk to a USB 2.0 > device as if it were hooked up to USB 1.1?" That has a very > different answer (no). > > -- > Tim Roberts, ti...@pr... > Providenza & Boekelheide, Inc. > > > -------------------------------------------------------------- > ----------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the > chance to share your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge > &CID=DEVDEV > _______________________________________________ > Libusb-devel mailing list > Lib...@li... > https://lists.sourceforge.net/lists/listinfo/libusb-devel > |
|
From: Dan L. M. <dma...@re...> - 2007-03-29 17:20:21
|
Hello, Sorry about the mess, I did mean High Speed, so Peter's answer was fine, thank you. About his answer, synchronous communication would work for me, but if asynchronous were available it would work much better! However, I didn't know that libusb 1.0 was already developed, and also don't know much about OpenUSB. Is there any links where I can read more about them? Just to clarify, I'm using LibUsb-win32, so maybe that's the reason I didn't see anything about 1.0. Thank you for the attention. Dan Leite Martins On Thu, 29 Mar 2007 10:06:55 -0700, "Jake Janovetz" <ja...@ja...> wrote: > Another slant on this... > > If your intent is to talk to the device at full-speed to test its > functionality on a USB 1.1 system, you can often setup the driver for your > USB host card to operate in USB 1.1 mode. > > We use this for testing. We don't have any USB 1.1 hosts here, so we set > our USB 2.0 host to operate as a 1.1 host and run tests on our device. > > Jake > > >> -----Original Message----- >> From: lib...@li... >> [mailto:lib...@li...] On Behalf >> Of Tim Roberts >> Sent: Thursday, March 29, 2007 10:04 AM >> To: lib...@li... >> Subject: Re: [Libusb-devel] USB 2.0 >> >> Dan Leite Martins wrote: >> > I'm new to the list. I'd like to know what is needed to the >> lib so it can "talk" to 2.0 usb devices at full speed. I'm >> sorry if this has been already discussed, but I couldn't >> access the archives. Thanks in advance. >> > >> >> Peter already provided an answer for this, but it occurred to >> me that your question could mean one of several things. The >> term "full speed" >> is somewhat overloaded. It could mean "at the maximum speed >> possible", in which case Peter was right all along. >> >> But "full speed" is also the official term for the higher >> 12Mbps bit rate offered by USB 1.x, which is the middle speed >> grade for USB 2.0. >> The maximum speed rate in USB 2.0 is called "high speed". >> So, you MIGHT have been asking "Can I talk to a USB 2.0 >> device as if it were hooked up to USB 1.1?" That has a very >> different answer (no). >> >> -- >> Tim Roberts, ti...@pr... >> Providenza & Boekelheide, Inc. >> >> >> -------------------------------------------------------------- >> ----------- >> Take Surveys. Earn Cash. Influence the Future of IT >> Join SourceForge.net's Techsay panel and you'll get the >> chance to share your >> opinions on IT & business topics through brief surveys-and earn cash >> http://www.techsay.com/default.php?page=join.php&p=sourceforge >> &CID=DEVDEV >> _______________________________________________ >> Libusb-devel mailing list >> Lib...@li... >> https://lists.sourceforge.net/lists/listinfo/libusb-devel >> > > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share > your > opinions on IT & business topics through brief surveys-and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > Libusb-devel mailing list > Lib...@li... > https://lists.sourceforge.net/lists/listinfo/libusb-devel |
|
From: Peter S. <stu...@cd...> - 2007-03-29 17:25:54
|
On Thu, Mar 29, 2007 at 10:06:55AM -0700, Jake Janovetz wrote: > We use this for testing. We don't have any USB 1.1 hosts here, so > we set our USB 2.0 host to operate as a 1.1 host and run tests on > our device. The EHCI should provide a number of [UO]HCI PCI devices for USB 1.1 devices, so this should just be a matter of blocking/disabling the EHCI driver. //Peter |
|
From: Sophia Li <Sop...@Su...> - 2007-03-30 01:53:58
|
Dan Leite Martins wrote: > Hello, > > Sorry about the mess, I did mean High Speed, so Peter's answer was fine, thank you. > > About his answer, synchronous communication would work for me, but if asynchronous were available it would work much better! However, I didn't know that libusb 1.0 was already developed, and also don't know much about OpenUSB. Is there any links where I can read more about them? Just to clarify, I'm using LibUsb-win32, so maybe that's the reason I didn't see anything about 1.0. Thank you for the attention. > For openusb project, please take a look at http://openusb.sourceforge.net/. There is already some application using the openusb code and works well. If you are interested, you can give it a try. Sophia |
|
From: Tim R. <ti...@pr...> - 2007-03-29 17:36:12
|
Peter Stuge wrote: > On Thu, Mar 29, 2007 at 10:06:55AM -0700, Jake Janovetz wrote: > >> We use this for testing. We don't have any USB 1.1 hosts here, so >> we set our USB 2.0 host to operate as a 1.1 host and run tests on >> our device. >> > > The EHCI should provide a number of [UO]HCI PCI devices for USB 1.1 > devices, so this should just be a matter of blocking/disabling the > EHCI driver. > Do you have evidence that this really works? I thought the UHCI drivers came to life as children of the EHCI driver, not necessarily as independent hardware entities. However, I admit that I haven't tried it. -- Tim Roberts, ti...@pr... Providenza & Boekelheide, Inc. |
|
From: Peter S. <stu...@cd...> - 2007-03-29 21:29:24
|
On Thu, Mar 29, 2007 at 10:36:06AM -0700, Tim Roberts wrote: > Peter Stuge wrote: > > > > The EHCI should provide a number of [UO]HCI PCI devices for USB 1.1 > > devices, so this should just be a matter of blocking/disabling the > > EHCI driver. > > Do you have evidence that this really works? I thought the UHCI > drivers came to life as children of the EHCI driver, not > necessarily as independent hardware entities. However, I admit > that I haven't tried it. I think I've tried it, although it was by mistake. I didn't have the EHCI driver and could still use a USB memory stick, just very slowly. lspci reports the UHCI controllers separately. --8<-- Also from page 3 of EHCI 1.0 spec A USB 2.0 Host Controller includes one high-speed mode host controller and 0 or more USB 1.1 host controllers (see Figure 1-2). The high-speed host controller implements an EHCI interface. It is used for all high-speed communications to high-speed-mode devices connected to the root ports of the USB 2.0 host controller. This specification allows communications to Full- and Low-speed devices connected to the root ports of the USB 2.0 host controller to be provided by companion USB 1.1 host controllers. If an implementation does not include companion host controllers, the host controller must include a high-speed device permanently attached to each of the EHCI ports the implementation is planning to utilize. The EHCI controller cannot work with a Full- or Low-speed device. -->8-- //Peter |
|
From: Sophia Li <Sop...@Su...> - 2007-03-30 02:04:33
|
Tim Roberts wrote: > > Do you have evidence that this really works? I thought the UHCI drivers > came to life as children of the EHCI driver, not necessarily as > independent hardware entities. However, I admit that I haven't tried it. They are separate in most functionality except that the port ownership may be shifted from EHCI to UHCI. If EHCI driver is in place, all ports are routed to EHCI at their initial connection, but full/low-speed connection will be shifted to UHCI later after port reset. If there is no EHCI driver at all, all ports would be routed to UHCI. Sophia |