| Message ID | cover.1657149962.git.Thinh.Nguyen@synopsys.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | usb: gadget: f_tcm: Enhance UASP driver | expand |
You probably want to split this up a bit to make review easier, a natural first series would be target core improvements that can be used as-is. Also please don't just Cc people on individual patches, which makes reviewinging impossible.
On 7/6/2022, Christoph Hellwig wrote: > You probably want to split this up a bit to make review easier, a > natural first series would be target core improvements that can be > used as-is. Also please don't just Cc people on individual patches, > which makes reviewinging impossible. If you haven't noticed already, there are dependencies that the f_tcm needs in the target core to function properly. To fully test the f_tcm, we need everything here. As for the list of people Cc'ed, most are pulled using the get_maintainer.pl. The target related patches also included the USB folks for context. Likewise, the USB patches included the target/scsi list. Please take a look and see how we can split this up while it can still make sense to be able to test it. Thanks, Thinh
On Wed, Jul 06, 2022 at 04:34:20PM -0700, Thinh Nguyen wrote: > The Linux UASP gadget driver is incomplete and remained broken for a long time. > It was not implemented for performance either. This series adds some of the > required features for the UASP driver to work. It also makes some fixes to the > target core. So I can't take the USB changes without a change to the target code? Some of these seem like I can, so I do not understand the dependancy here. Can you split this into at least 2 series? One for just target, one for just USB, and maybe one for the remaining bits that require both? thanks, greg k-h
On 7/6/2022, Greg Kroah-Hartman wrote: > On Wed, Jul 06, 2022 at 04:34:20PM -0700, Thinh Nguyen wrote: >> The Linux UASP gadget driver is incomplete and remained broken for a long time. >> It was not implemented for performance either. This series adds some of the >> required features for the UASP driver to work. It also makes some fixes to the >> target core. > So I can't take the USB changes without a change to the target code? > Some of these seem like I can, so I do not understand the dependancy > here. Without the entire series, UASP Compliant Verification test will fail. The dependency is more for the CV test. > Can you split this into at least 2 series? One for just target, one for > just USB, and maybe one for the remaining bits that require both? > Ok, I can split them base on compilation dependency. Thanks, Thinh
On Thu, Jul 07, 2022 at 10:15:53AM +0000, Thinh Nguyen wrote: > On 7/6/2022, Greg Kroah-Hartman wrote: > > On Wed, Jul 06, 2022 at 04:34:20PM -0700, Thinh Nguyen wrote: > >> The Linux UASP gadget driver is incomplete and remained broken for a long time. > >> It was not implemented for performance either. This series adds some of the > >> required features for the UASP driver to work. It also makes some fixes to the > >> target core. > > So I can't take the USB changes without a change to the target code? > > Some of these seem like I can, so I do not understand the dependancy > > here. > > Without the entire series, UASP Compliant Verification test will fail. It fails today, right? So it's not an issue. > The dependency is more for the CV test. That's independant of getting patches merged, correct? > > Can you split this into at least 2 series? One for just target, one for > > just USB, and maybe one for the remaining bits that require both? > > > > Ok, I can split them base on compilation dependency. You also have to realize there are maintainer and subsystem dependencies that you are crossing. thanks, greg k-h