Re: [PATCH] RHEL5 iBFT forward port to Fedora - part 1

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 





Martin Sivak wrote:
Note that due to some changes I made wrt iscsi in Fedora currently
startup only gets called from addTarget, so this code path will no longer
automatically execute.

startup used to be called from partitionObjectsInitialize(), but I've
removed it being called from there as that let to lots of start stop start
iscsi madness.

For iBFT we will need to call it some place sane, so that it gets done
only once (and not start stop start for each iscsi target added).

OK, what is the sane place right now?

Erm, I haven't really looked at that yet, somewhere where it gets called only once, and before we go looking for disks.



<snip>

--- a/loader/net.c
+++ b/loader/net.c
@@ -53,6 +53,7 @@
  #include "method.h"
  #include "net.h"
  #include "windows.h"
+#include "ibft.h"

  /* boot flags */
  extern uint64_t flags;

I do not see ibft.h getting added by this patch

Ehmm you are right.. the two missing files are attached to this reply


Ok, they look good, so the patch is good to go.



      if ((loaderData->netDev && (loaderData->netDev_set == 1)) &&
          !strcmp(loaderData->netDev, "link")) {
+	lookForLink = 1;
+    }
+
+    if(lookForLink){
          logMessage(INFO, "looking for first netDev with link");
          for (rc = 0; rc < 5; rc++) {
              for (i = 0; i < deviceNums; i++) {

I do not understand what the purpose of lookForLink is here, before
the

      if ((loaderData->netDev && (loaderData->netDev_set == 1)) &&
          !strcmp(loaderData->netDev, "link")) {

It is always 0, so the "if(lookForLink){" is equivalent to

      if ((loaderData->netDev && (loaderData->netDev_set == 1)) &&
          !strcmp(loaderData->netDev, "link")) {

Right now it is, there was different default value before, so I kept it there for debugging purposes and to have the possibility to change the default.

Ok.

Erm, why not just write BOOTPROTO=none (what you call static is none
AFAIK) or BOOTPROTO=dhcp depending on the iBFT info, teaching nm-system-settings about BOOTPROTO=ibft is of little use unless you plan to teach it to actually read the info from the BIOS in that scenario, so that changes in the BIOS propagate to the OS.

It is not possible, you have to read the network info from the iBFT on every startup.
Because it doesn't have to stay the same forever. And when you change the firmware setting, knowing that iBFT is supported, you expect to pick up the changes and act upon them.


Ok, so you're going to patch nm-system-settings to actually read the info from iBFT then I assume?

Regards,

Hans

_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/anaconda-devel-list

[Index of Archives]     [Kickstart]     [Fedora Users]     [Fedora Legacy List]     [Fedora Maintainers]     [Fedora Desktop]     [Fedora SELinux]     [Big List of Linux Books]     [Yosemite News]     [Yosemite Photos]     [KDE Users]     [Fedora Tools]
  Powered by Linux