Hi Gustavo, On Friday 29 December 2017 03:03 PM, Gustavo Pimentel wrote: > Hi Jingoo and Kishon, > > On 28/12/2017 16:13, Jingoo Han wrote: >> On Thursday, December 28, 2017 7:44 AM, Kishon Vijay Abraham I wrote: >>> >>> Hi Gustavo, >>> >>> On Thursday 28 December 2017 05:26 PM, Gustavo Pimentel wrote: >>>> This patch adapts Keystone SoC specific driver to use the new interrupt >>> api >>>> available in pcie-designware. A new callback was added to >>> dw_pcie_host_ops >>>> to handle a specific Keystone function and msi_host_init callback is >>>> changed to simplify the access to pci data structure for keystone all >>> SoC >>>> drivers that use the structure. >>>> >>>> Signed-off-by: Gustavo Pimentel <gustavo.pimentel@xxxxxxxxxxxx> >>>> --- >>>> Change v1->v2: >>>> - Removed hardcoded num_vectors configuration (now it is done in the >>> core driver by default) >>>> Change v2->v3: >>>> - >>>> >>>> >>>> >>>> >>>> >>>> drivers/pci/dwc/pci-keystone-dw.c | 88 >> ++--------------------------- >>> ----- >>>> drivers/pci/dwc/pci-keystone.c | 1 + >>>> drivers/pci/dwc/pci-keystone.h | 4 +- >>>> drivers/pci/dwc/pci-layerscape.c | 3 +- >>>> drivers/pci/dwc/pcie-designware-host.c | 19 ++++++-- >>>> drivers/pci/dwc/pcie-designware.h | 3 +- >>> >>> The designware specific patches should be part of >>> "[1/9] pci: adding new irq api to pci-designware" > > Ok, I will move your (Kishon) fixes to the 0001 patch file, no problem. > >>> >>> Also layerscap is not related to $subject. > > Please note there are some initial changes on this patch (maybe they could be in > a isolated patch) related to pci-layerscape.c, pcie-designware-host.c and > pcie-designware.h files that changes the msi_host_init callback signature and > this affects the Keystone, Layerscape drivers and the core at the same time, > that's the reason why this modification touches on these three files and should > be together. > Was I able to pass this information clearly? Got it. $subject or the commit message didn't have that information. However as I've commented on your patch, it's better if we keep the signature of msi_host_init as it is. Thanks Kishon
