| Message ID | 20190926164519.10471-21-jeffrey.t.kirsher@intel.com |
|---|---|
| State | RFC |
| Delegated to: | David Miller |
| Headers | show |
| Series | Intel RDMA/IDC Driver series | expand |
On Thu, Sep 26, 2019 at 09:45:19AM -0700, Jeff Kirsher wrote: > From: Shiraz Saleem <shiraz.saleem@intel.com> > > Mark i40iw as deprecated/obsolete. > > irdma is the replacement driver that supports X722. Can you simply delete old one and add MODULE_ALIAS() in new driver? Thanks
> Subject: Re: [RFC 20/20] RDMA/i40iw: Mark i40iw as deprecated > > On Thu, Sep 26, 2019 at 09:45:19AM -0700, Jeff Kirsher wrote: > > From: Shiraz Saleem <shiraz.saleem@intel.com> > > > > Mark i40iw as deprecated/obsolete. > > > > irdma is the replacement driver that supports X722. > > Can you simply delete old one and add MODULE_ALIAS() in new driver? > Yes, but we thought typically driver has to be deprecated for a few cycles before removing it.
On Thu, Sep 26, 2019 at 07:49:44PM +0000, Saleem, Shiraz wrote: > > Subject: Re: [RFC 20/20] RDMA/i40iw: Mark i40iw as deprecated > > > > On Thu, Sep 26, 2019 at 09:45:19AM -0700, Jeff Kirsher wrote: > > > From: Shiraz Saleem <shiraz.saleem@intel.com> > > > > > > Mark i40iw as deprecated/obsolete. > > > > > > irdma is the replacement driver that supports X722. > > > > Can you simply delete old one and add MODULE_ALIAS() in new driver? > > > > Yes, but we thought typically driver has to be deprecated for a few cycles before removing it. If you completely replace it with something that works the same, why keep the old one around at all? Unless you don't trust your new code? :) thanks, greg k-h
> Subject: Re: [RFC 20/20] RDMA/i40iw: Mark i40iw as deprecated > > On Thu, Sep 26, 2019 at 07:49:44PM +0000, Saleem, Shiraz wrote: > > > Subject: Re: [RFC 20/20] RDMA/i40iw: Mark i40iw as deprecated > > > > > > On Thu, Sep 26, 2019 at 09:45:19AM -0700, Jeff Kirsher wrote: > > > > From: Shiraz Saleem <shiraz.saleem@intel.com> > > > > > > > > Mark i40iw as deprecated/obsolete. > > > > > > > > irdma is the replacement driver that supports X722. > > > > > > Can you simply delete old one and add MODULE_ALIAS() in new driver? > > > > > > > Yes, but we thought typically driver has to be deprecated for a few cycles > before removing it. > > If you completely replace it with something that works the same, why keep the old > one around at all? Agree. Thanks! > > Unless you don't trust your new code? :) > We do :)
On Thu, 2019-09-26 at 21:55 +0200, gregkh@linuxfoundation.org wrote: > On Thu, Sep 26, 2019 at 07:49:44PM +0000, Saleem, Shiraz wrote: > > > Subject: Re: [RFC 20/20] RDMA/i40iw: Mark i40iw as deprecated > > > > > > On Thu, Sep 26, 2019 at 09:45:19AM -0700, Jeff Kirsher wrote: > > > > From: Shiraz Saleem <shiraz.saleem@intel.com> > > > > > > > > Mark i40iw as deprecated/obsolete. > > > > > > > > irdma is the replacement driver that supports X722. > > > > > > Can you simply delete old one and add MODULE_ALIAS() in new > > > driver? > > > > > > > Yes, but we thought typically driver has to be deprecated for a few > > cycles before removing it. > > If you completely replace it with something that works the same, why > keep the old one around at all? > > Unless you don't trust your new code? :) I have yet to see, in over 20 years of kernel experience, a new driver replace an old driver and not initially be more buggy and troublesome than the old driver. It takes time and real world usage for the final issues to get sorted out. During that time, the fallback is often necessary for those real world users.
On Fri, 2019-09-27 at 14:28 +0000, Saleem, Shiraz wrote: > > Subject: Re: [RFC 20/20] RDMA/i40iw: Mark i40iw as deprecated > > > > On Thu, Sep 26, 2019 at 07:49:44PM +0000, Saleem, Shiraz wrote: > > > > Subject: Re: [RFC 20/20] RDMA/i40iw: Mark i40iw as deprecated > > > > > > > > On Thu, Sep 26, 2019 at 09:45:19AM -0700, Jeff Kirsher wrote: > > > > > From: Shiraz Saleem <shiraz.saleem@intel.com> > > > > > > > > > > Mark i40iw as deprecated/obsolete. > > > > > > > > > > irdma is the replacement driver that supports X722. > > > > > > > > Can you simply delete old one and add MODULE_ALIAS() in new > > > > driver? > > > > > > > > > > Yes, but we thought typically driver has to be deprecated for a > > > few cycles > > before removing it. > > > > If you completely replace it with something that works the same, why > > keep the old > > one around at all? > > Agree. Thanks! > > > > Unless you don't trust your new code? :) > > > We do :) I don't....
On Fri, Sep 27, 2019 at 04:17:15PM -0400, Doug Ledford wrote: > On Thu, 2019-09-26 at 21:55 +0200, gregkh@linuxfoundation.org wrote: > > On Thu, Sep 26, 2019 at 07:49:44PM +0000, Saleem, Shiraz wrote: > > > > Subject: Re: [RFC 20/20] RDMA/i40iw: Mark i40iw as deprecated > > > > > > > > On Thu, Sep 26, 2019 at 09:45:19AM -0700, Jeff Kirsher wrote: > > > > > From: Shiraz Saleem <shiraz.saleem@intel.com> > > > > > > > > > > Mark i40iw as deprecated/obsolete. > > > > > > > > > > irdma is the replacement driver that supports X722. > > > > > > > > Can you simply delete old one and add MODULE_ALIAS() in new > > > > driver? > > > > > > > > > > Yes, but we thought typically driver has to be deprecated for a few > > > cycles before removing it. > > > > If you completely replace it with something that works the same, why > > keep the old one around at all? > > > > Unless you don't trust your new code? :) > > I have yet to see, in over 20 years of kernel experience, a new driver > replace an old driver and not initially be more buggy and troublesome > than the old driver. It takes time and real world usage for the final > issues to get sorted out. During that time, the fallback is often > necessary for those real world users. How many real users exist in RDMA world who run pure upstream kernel? Thanks > > -- > Doug Ledford <dledford@redhat.com> > GPG KeyID: B826A3330E572FDD > Fingerprint = AE6B 1BDA 122B 23B4 265B 1274 B826 A333 0E57 2FDD
On 9/28/2019 1:55 AM, Leon Romanovsky wrote: > On Fri, Sep 27, 2019 at 04:17:15PM -0400, Doug Ledford wrote: >> On Thu, 2019-09-26 at 21:55 +0200, gregkh@linuxfoundation.org wrote: >>> On Thu, Sep 26, 2019 at 07:49:44PM +0000, Saleem, Shiraz wrote: >>>>> Subject: Re: [RFC 20/20] RDMA/i40iw: Mark i40iw as deprecated >>>>> >>>>> On Thu, Sep 26, 2019 at 09:45:19AM -0700, Jeff Kirsher wrote: >>>>>> From: Shiraz Saleem <shiraz.saleem@intel.com> >>>>>> >>>>>> Mark i40iw as deprecated/obsolete. >>>>>> >>>>>> irdma is the replacement driver that supports X722. >>>>> >>>>> Can you simply delete old one and add MODULE_ALIAS() in new >>>>> driver? >>>>> >>>> >>>> Yes, but we thought typically driver has to be deprecated for a few >>>> cycles before removing it. >>> >>> If you completely replace it with something that works the same, why >>> keep the old one around at all? >>> >>> Unless you don't trust your new code? :) >> >> I have yet to see, in over 20 years of kernel experience, a new driver >> replace an old driver and not initially be more buggy and troublesome >> than the old driver. It takes time and real world usage for the final >> issues to get sorted out. During that time, the fallback is often >> necessary for those real world users. > > How many real users exist in RDMA world who run pure upstream kernel? I doubt too many especially the latest bleeding edge upstream kernel. That could be interesting, but I don't think it's the reality. Distro kernels could certainly still keep the old driver, and that makes a lot of sense. -Denny
On Wed, Oct 02, 2019 at 05:15:45PM -0400, Dennis Dalessandro wrote: > On 9/28/2019 1:55 AM, Leon Romanovsky wrote: > > On Fri, Sep 27, 2019 at 04:17:15PM -0400, Doug Ledford wrote: > > > On Thu, 2019-09-26 at 21:55 +0200, gregkh@linuxfoundation.org wrote: > > > > On Thu, Sep 26, 2019 at 07:49:44PM +0000, Saleem, Shiraz wrote: > > > > > > Subject: Re: [RFC 20/20] RDMA/i40iw: Mark i40iw as deprecated > > > > > > > > > > > > On Thu, Sep 26, 2019 at 09:45:19AM -0700, Jeff Kirsher wrote: > > > > > > > From: Shiraz Saleem <shiraz.saleem@intel.com> > > > > > > > > > > > > > > Mark i40iw as deprecated/obsolete. > > > > > > > > > > > > > > irdma is the replacement driver that supports X722. > > > > > > > > > > > > Can you simply delete old one and add MODULE_ALIAS() in new > > > > > > driver? > > > > > > > > > > > > > > > > Yes, but we thought typically driver has to be deprecated for a few > > > > > cycles before removing it. > > > > > > > > If you completely replace it with something that works the same, why > > > > keep the old one around at all? > > > > > > > > Unless you don't trust your new code? :) > > > > > > I have yet to see, in over 20 years of kernel experience, a new driver > > > replace an old driver and not initially be more buggy and troublesome > > > than the old driver. It takes time and real world usage for the final > > > issues to get sorted out. During that time, the fallback is often > > > necessary for those real world users. > > > > How many real users exist in RDMA world who run pure upstream kernel? > > I doubt too many especially the latest bleeding edge upstream kernel. That > could be interesting, but I don't think it's the reality. > > Distro kernels could certainly still keep the old driver, and that makes a > lot of sense. Also, they are invited to run their regression suite to verify stability and report any arising problems to upstream/vendor. Thanks > > -Denny >
diff --git a/drivers/infiniband/hw/i40iw/Kconfig b/drivers/infiniband/hw/i40iw/Kconfig index e4b45f4cd8f8..208e7525fb55 100644 --- a/drivers/infiniband/hw/i40iw/Kconfig +++ b/drivers/infiniband/hw/i40iw/Kconfig @@ -1,9 +1,11 @@ # SPDX-License-Identifier: GPL-2.0-only config INFINIBAND_I40IW - tristate "Intel(R) Ethernet X722 iWARP Driver" + tristate "Intel(R) Ethernet X722 iWARP Driver (DEPRECATED)" depends on INET && I40E depends on IPV6 || !IPV6 depends on PCI + depends on !(INFINBAND_IRDMA=y || INFINIBAND_IRDMA=m) || COMPILE_TEST select GENERIC_ALLOCATOR ---help--- Intel(R) Ethernet X722 iWARP Driver + This driver is being replaced by irdma.