Skip to contentSkip to conversation

Update QEMU_UPSTREAM_REVISION

PATCH In review Anthony PERARD 4 participants 9 messages Nov 27, 2013 – Dec 4, 2013

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 Update QEMU_UPSTREAM_REVISION

---
the commit sha correspond to the last tree that pass the qemu-xen push gate (master).

Should we change this value to master once a release is done, during the developpement phase?

---
 Config.mk | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Config.mk +1 -1
index 7eeed31..4a74a33 100644
@@ -234,7 +234,7 @@ QEMU_UPSTREAM_URL ?= git://xenbits.xen.org/qemu-upstream-unstable.git
 SEABIOS_UPSTREAM_URL ?= git://xenbits.xen.org/seabios.git
 endif
 OVMF_UPSTREAM_REVISION ?= a93b0e3f6895a074b99c8817181dfa6dbc7a4807
-QEMU_UPSTREAM_REVISION ?= 1c514a7734b7f98625a0d18d5e8ee7581f26e50c
+QEMU_UPSTREAM_REVISION ?= b97307ecaad98360f41ea36cd9674ef810c4f8cf
 SEABIOS_UPSTREAM_TAG ?= rel-1.7.3.1
 # Fri Aug 2 14:12:09 2013 -0400
 # Fix bug in CBFS file walking with compressed files.
Signature · 1 line
Anthony PERARD
#1 lore raw

Anthony PERARD writes ("[PATCH] Update QEMU_UPSTREAM_REVISION"):

Quoted 7 lines

---
the commit sha correspond to the last tree that pass the qemu-xen push gate (master).

Should we change this value to master once a release is done, during the developpement phase?

Perhaps we should change it to master now. That way if further updates are made to the tree, and pass the push gate, they will be included. Which is surely the intent ?

Ian.

#2 lore raw
Quoted 6 lines · Ian Jackson

Anthony PERARD writes ("[PATCH] Update QEMU_UPSTREAM_REVISION"):

Quoted 5 lines

---
the commit sha correspond to the last tree that pass the qemu-xen push gate (master).

Quoted 6 lines
Quoted 2 lines

Should we change this value to master once a release is done, during the developpement phase?

Perhaps we should change it to master now. That way if further updates are made to the tree, and pass the push gate, they will be included. Which is surely the intent ?

It looks like it was "master" until 4de97462d34f when it was updated to "qemu-xen-4.3.0-rc1" (presumably in anticipation of branching or releasing the rc) then it was turned into a sha at 1fee00908ae9 -- which I think was likely to have been in error.

I agree that at least during the development phases it should be "master".

It should probably be a tag for a release, including rcs, so that they refer to a single consistent thing.

Given our policy of branching in the later rcs I'm not sure how we can deal with this other than as we attempted last time by committing something other than "master" to xen.git:Config.mk and remembering what to do to put it back later afterwards. (or by revisiting the branching policy, but I don't think this issue is reason enough to change).

Ian.

#3 lore raw

PATCH 1/1 Update QEMU_UPSTREAM_REVISION

Changing to master, otherwise we don't get the last updates.

---
change since previous patch:
- use master instead of a specific commit id

---
 Config.mk | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Config.mk +1 -1
index 7eeed31..ce20844 100644
@@ -234,7 +234,7 @@ QEMU_UPSTREAM_URL ?= git://xenbits.xen.org/qemu-upstream-unstable.git
 SEABIOS_UPSTREAM_URL ?= git://xenbits.xen.org/seabios.git
 endif
 OVMF_UPSTREAM_REVISION ?= a93b0e3f6895a074b99c8817181dfa6dbc7a4807
-QEMU_UPSTREAM_REVISION ?= 1c514a7734b7f98625a0d18d5e8ee7581f26e50c
+QEMU_UPSTREAM_REVISION ?= master
 SEABIOS_UPSTREAM_TAG ?= rel-1.7.3.1
 # Fri Aug 2 14:12:09 2013 -0400
 # Fix bug in CBFS file walking with compressed files.
Signature · 1 line
Anthony PERARD
Acked-by on v1 1/1
#4 lore raw
Quoted 3 lines · Anthony PERARD

Changing to master, otherwise we don't get the last updates.

WRT the release this patch is equivalent to "we should move to qemu 1.6". Did that get an exception already / is it on the list?

(please don't forget to CC George at this stage in the release for his Ack, or to reference an existing release ack he gave)

Quoted 20 lines

---
change since previous patch:
- use master instead of a specific commit id

---
 Config.mk | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Config.mk +1 -1
index 7eeed31..ce20844 100644
@@ -234,7 +234,7 @@ QEMU_UPSTREAM_URL ?= git://xenbits.xen.org/qemu-upstream-unstable.git
 SEABIOS_UPSTREAM_URL ?= git://xenbits.xen.org/seabios.git
 endif
 OVMF_UPSTREAM_REVISION ?= a93b0e3f6895a074b99c8817181dfa6dbc7a4807
-QEMU_UPSTREAM_REVISION ?= 1c514a7734b7f98625a0d18d5e8ee7581f26e50c
+QEMU_UPSTREAM_REVISION ?= master
 SEABIOS_UPSTREAM_TAG ?= rel-1.7.3.1
 # Fri Aug 2 14:12:09 2013 -0400
 # Fix bug in CBFS file walking with compressed files.
#5 lore raw
Quoted 11 lines · Ian Campbell
Quoted 3 lines · Anthony PERARD

Changing to master, otherwise we don't get the last updates.

WRT the release this patch is equivalent to "we should move to qemu 1.6". Did that get an exception already / is it on the list?

(please don't forget to CC George at this stage in the release for his Ack, or to reference an existing release ack he gave)

Actually, isn't this "Fix incorrect assignment of a specific revision instead of most-recently-passed for development branch"? It's the push into the qemu tree that would require a release ack, but that happened a month ago before the code freeze IIRC.

-George

#6 lore raw
Quoted 10 lines · George Dunlap
Quoted 9 lines · Ian Campbell
Quoted 3 lines · Anthony PERARD

Changing to master, otherwise we don't get the last updates.

WRT the release this patch is equivalent to "we should move to qemu 1.6". Did that get an exception already / is it on the list?

Quoted 7 lines
Quoted 2 lines

(please don't forget to CC George at this stage in the release for his Ack, or to reference an existing release ack he gave)

Actually, isn't this "Fix incorrect assignment of a specific revision instead of most-recently-passed for development branch"? It's the push into the qemu tree that would require a release ack, but that happened a month ago before the code freeze IIRC.

The concern was that we weren't testing the result of that push yet and that this change would be in effect merging that code into the main tree, which ought to at least be done with the release manager's awareness.

However it turns out that OSStest uses the latest qemu push gate regardless of Config.mk -- which changes the equation somewhat -- we should obviously update Config.mk to what we are testing.

Ian.

#7 lore raw
In reply to George Dunlap

George Dunlap writes ("Re: [PATCH] Update QEMU_UPSTREAM_REVISION"):

Quoted 4 lines

Actually, isn't this "Fix incorrect assignment of a specific revision instead of most-recently-passed for development branch"? It's the push into the qemu tree that would require a release ack, but that happened a month ago before the code freeze IIRC.

Yes, I think so.

Ian.

Applied
#8 lore raw
Quoted 7 lines · Ian Jackson

George Dunlap writes ("Re: [PATCH] Update QEMU_UPSTREAM_REVISION"):

Quoted 4 lines

Actually, isn't this "Fix incorrect assignment of a specific revision instead of most-recently-passed for development branch"? It's the push into the qemu tree that would require a release ack, but that happened a month ago before the code freeze IIRC.

Yes, I think so.

I've applied Anthony's patch to switch to master.