On Fri, Nov 30, 2012 at 6:38 AM, Hauke Mehrtens <hauke@xxxxxxxxxx> wrote: > On 11/30/2012 08:58 AM, Ozan Çağlayan wrote: >> On Thu, Nov 29, 2012 at 5:24 PM, Hauke Mehrtens <hauke@xxxxxxxxxx> wrote: >>> Without this patch the patches are applied in the natural sort order of >>> numbers, which caused this order for patches/network/: >>> patches/network/0001-netdev_ops.patch >>> patches/network/02-ksize.patch >>> patches/network/0002-net-misc.patch >>> patches/network/0003-netdev-needed_headroom_tailroom.patch >>> patches/network/03-rfkill.patch >>> patches/network/04-netns.patch >>> patches/network/0004-wext-namespace.patch >>> patches/network/0005-netlink-portid.patch >>> patches/network/05-usb.patch >> >> Actually it was me who changed/broke this and I really can't decide >> now which one is more intuitive. >> >> commit 46a5aef05b92a2162150326f426bebe3541d6e8c >> Author: Ozan Çağlayan <ozancag@xxxxxxxxx> >> Date: Fri Jun 22 16:30:54 2012 +0300 >> >> compat-wireless: Naturally sort patch prefixes >> >> Use ls -v to get a naturally sorted list of patches like: >> >> patches/00-vga_switcheroo_client_ops.patch >> patches/01-dma_buf_ops-addition.patch >> patches/02-revert-vm_mmap.patch >> patches/98-pr_fmt.patch >> patches/99-change-makefile.patch >> patches/9999-FIXME-dont-build-i915-for-i2c-problems.patch >> >> instead of: >> >> patches/00-vga_switcheroo_client_ops.patch >> patches/01-dma_buf_ops-addition.patch >> patches/02-revert-vm_mmap.patch >> patches/98-pr_fmt.patch >> patches/9999-FIXME-dont-build-i915-for-i2c-problems.patch >> patches/99-change-makefile.patch >> >> Signed-off-by: Ozan Çağlayan <ozancag@xxxxxxxxx> >> Signed-off-by: Luis R. Rodriguez <mcgrof@xxxxxxxxxxxxx> > > Hi Ozan, > > I missed looking which commit introduced this, I thought it was there > for ever. With ls -v the prefixes 0005 and 05 are the same and the > ordering is done by the following characters, this is the reason > 04-netns.patch is in front of 0004-wext-namespace.patch, but > 0005-netlink-portid.patch is in front of 05-usb.patch. > > For your use case adding the -v is a good option, but as we now have > some patches with four digits and some with two digits I think the > version without the -v is the better option, or we rename the patches to > one common pattern. Or we can stuff the additional 4 digit patches into a new directory, I have some patch I am working on that does something like this, let me know if this is a preferred strategy. So far I have a patch that for example moves patches under patches/ to patches/collateral-evolutions/ and then I stuff the crap/ pending-stable/ linux-next-cherry-picks/ and linux-next-pending/ into patches/ as well. If we separate the 4 digit patches from the not-so-cleaned up series of older patches for backporting its unclear to me what we'd call the newer ones if we take this approach. Thoughts? Luis -- To unsubscribe from this list: send the line "unsubscribe backports" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
