Re: [U-Boot-Users] [PATCH] Part 1 of 2 MPC85XXADS-20040716.patch-1 for TLB/DDR
On Fri, 2004-07-16 at 16:51, Wolfgang Denk wrote:
In message 1090009011.23992.5.camel@blarg.somerset.sps.mot.com you wrote:
Here is a patch for the MPC8540/8560ADS and STXGP3 boards that is primarily to support larger DDR memories up to 2G.
Can you please think of a better subject line for your patches?
What would you like? I meant to convey that:
- It was a [PATCH] - It was for the MPC85XXADS boards, - A specific patch release identified by date, - That it came in multiple parts, in this case -1 and -2, - That its primary purpose was TLB/DDR related
Most of that information is geared towards potential searching of the archives, your information, my internal logging, and hopefully some form of "quick identification".
Where did I go wrong?
I find it extremely confusing when I first get a patch titled:
[PATCH] MPC85XXADS-20040716.patch for TLB/DDR
and then another one (split in 2 parts) titled
[PATCH] MPC85XXADS-20040716.patch-1 for TLB/DDR
I tend to assume that these are redundand, and dump one of them.
In fact, I was 5K over the mailing list size limit and was informed that it wasn't sent to the list. Last time I sent something to the list that was over the size limit, you specifically asked me to resend it in multiple parts.
So this time, that is exactly what I did.
So yes, it was redundant and one of the patches, the first, was expected to be /dev/nulled. In the second posts, I clearly indicated that it was a re-send of the earlier one due to size limit problems.
Looking closer, I find that at least one subject is totally wrong.
My apologies.
Wolfgang Denk
jdl
In message 1090250613.10358.7.camel@blarg.somerset.sps.mot.com you wrote:
What would you like? I meant to convey that:
Something that identifies a patch in a uniqe and correct way anc corresponds to the changelog.
... > - That its primary purpose was TLB/DDR related
Maybe for the first patch submitted under this subject.
Most of that information is geared towards potential searching of the archives, your information, my internal logging, and hopefully some form of "quick identification".
Where did I go wrong?
On 2004-07-16 you sent these messages:
[PATCH] MPC85XXADS-20040716.patch for TLB/DDR http://sourceforge.net/mailarchive/forum.php?thread_id=5136550&forum_id=...
[PATCH] Part 1 of 2 MPC85XXADS-20040716.patch-1 for TLB/DDR http://sourceforge.net/mailarchive/forum.php?thread_id=5137350&forum_id=...
[PATCH] Part 2 of 2 MPC85XXADS-20040716.patch-2 for TLB/DDR http://sourceforge.net/mailarchive/forum.php?thread_id=5137351&forum_id=...
Now is this all the same stuff, or two or maybe even three different patches?
The subject says "for TLB/DDR".
Your suggested CHANGELOG entry says:
... * Updated OR3 to only use 1M, not 16M, for CADMUS. Thanks York! * Made MPC8540/8560ADS be 33Mhz PCI by default. ... * Removed moldy CONFIG_RAM_AS_FLASH, CFG_FLASH_PORT_WIDTH_16 and CONFIG_L2_INIT_RAM options. ... * Moved board specific PCI init parts out of CPU directory. ...
The README requires:
Changesets that contain different, unrelated modifications shall be submitted as SEPARATE patches, one patch per changeset.
To make this short: which of your 3 messages shall be added?
Wolfgang Denk
On Mon, 2004-07-19 at 10:46, Wolfgang Denk wrote:
Where did I go wrong?
On 2004-07-16 you sent these messages:
[PATCH] MPC85XXADS-20040716.patch for TLB/DDR http://sourceforge.net/mailarchive/forum.php?thread_id=5136550&forum_id=...
[PATCH] Part 1 of 2 MPC85XXADS-20040716.patch-1 for TLB/DDR http://sourceforge.net/mailarchive/forum.php?thread_id=5137350&forum_id=...
[PATCH] Part 2 of 2 MPC85XXADS-20040716.patch-2 for TLB/DDR http://sourceforge.net/mailarchive/forum.php?thread_id=5137351&forum_id=...
Now is this all the same stuff, or two or maybe even three different patches?
Right. The first message was rejected as being too large. I broke it into two parts that do the same patch and resubmitted it with a -1 and a -2 suffix.
The subject says "for TLB/DDR".
Yes, I rewrote the TLB and DDR handling.
Changesets that contain different, unrelated modifications shall be submitted as SEPARATE patches, one patch per changeset.
They aren't really unreleated. For example, there were missing TLB entries for PCI address spaces that were added. The removal of the molding options was specifically to clean up the TLBs.
To make this short: which of your 3 messages shall be added?
Please add all of them as they all apply and are important.
Wolfgang Denk
Thanks, jdl
Dear Jon,
in message 1090252402.10358.21.camel@blarg.somerset.sps.mot.com you wrote:
Right. The first message was rejected as being too large.
Oops? I received it through the list, and it made it into the archives. I don't think it was rejected.
I broke it into two parts that do the same patch and resubmitted it with a -1 and a -2 suffix.
So message 2 + message 3 is the same content as message 1?
The subject says "for TLB/DDR".
Yes, I rewrote the TLB and DDR handling.
...and some other things, whichlook unrelated to me.
Changesets that contain different, unrelated modifications shall be submitted as SEPARATE patches, one patch per changeset.
They aren't really unreleated. For example, there were missing TLB entries for PCI address spaces that were added. The removal of the molding options was specifically to clean up the TLBs.
OK, but which of these has anything to do with that:
* Made MPC8540/8560ADS be 33Mhz PCI by default. ... * Removed moldy CONFIG_RAM_AS_FLASH, CFG_FLASH_PORT_WIDTH_16 and CONFIG_L2_INIT_RAM options. ... * Moved board specific PCI init parts out of CPU directory.
?
To make this short: which of your 3 messages shall be added?
Please add all of them as they all apply and are important.
This makes NO sense to me. If I apply the first patch, the other two (part 1 and 2) will not apply any more.
Best regards,
Wolfgang Denk
On Mon, 2004-07-19 at 11:15, Wolfgang Denk wrote:
Dear Jon,
in message 1090252402.10358.21.camel@blarg.somerset.sps.mot.com you wrote:
Right. The first message was rejected as being too large.
Oops? I received it through the list, and it made it into the archives. I don't think it was rejected.
I see that now too, but it didn't arrive for quite some time. I think there was weirdness, so let me look into it a bit here too...
I broke it into two parts that do the same patch and resubmitted it with a -1 and a -2 suffix.
So message 2 + message 3 is the same content as message 1?
Correct.
To make this short: which of your 3 messages shall be added?
Please add all of them as they all apply and are important.
This makes NO sense to me. If I apply the first patch, the other two (part 1 and 2) will not apply any more.
I totally misunderstood you. I thought you were asking which of the three LOG messages should be used.
Please apply either:
- The first patch without the -1 and -2 suffixes, or - The 2nd and 3rd messages with the -1 and -2 suffixes,
as they have the same content.
Sorry for the confusion, jdl
On Mon, 2004-07-19 at 11:15, Wolfgang Denk wrote:
Dear Jon,
in message 1090252402.10358.21.camel@blarg.somerset.sps.mot.com you wrote:
Right. The first message was rejected as being too large.
Oops? I received it through the list, and it made it into the archives. I don't think it was rejected.
Aha! I _did_ get confused. The patch was sent to both the public U-Boot list and the internal U-Boot list. It was rejected from the _internal_ list, but my mailer confused me and I thought it came from the public sourceforge site. I missed that it was an _internal_ sourceforge rejection.
My apologies for the confusing duplicate messages to the list.
jdl
In message 1090255620.10357.63.camel@blarg.somerset.sps.mot.com you wrote:
Aha! I _did_ get confused. The patch was sent to both
Indeed ;-)
the public U-Boot list and the internal U-Boot list. It was rejected from the _internal_ list, but my mailer confused me and I thought it came from the public sourceforge site. I missed that it was an _internal_ sourceforge rejection.
??????
There is only one relvant mailing list at SF: u-boot-users, and this is of course a public list. There is no "internal" SF list, so it is impossible that you received a reject from such a list.
I have to admit thay I am confused now. I have no idea what you might be talking about.
Best regards,
Wolfgang Denk
On Mon, 2004-07-19 at 13:31, Wolfgang Denk wrote:
In message 1090255620.10357.63.camel@blarg.somerset.sps.mot.com you wrote:
Aha! I _did_ get confused. The patch was sent to both
Indeed ;-)
Again, sorry for the confusion.
the public U-Boot list and the internal U-Boot list. It was rejected from the _internal_ list, but my mailer confused me and I thought it came from the public sourceforge site. I missed that it was an _internal_ sourceforge rejection.
??????
There is only one relvant mailing list at SF: u-boot-users, and this is of course a public list. There is no "internal" SF list, so it is impossible that you received a reject from such a list.
I have to admit thay I am confused now. I have no idea what you might be talking about.
Internal to Motorola/Freescale. There _is_ an internal U-Boot sourceforge.mot.com site where we house our CVS tree and do our U-Boot development before it gets released to the public U-Boot sources at sourceforge.net.
No, you never saw it. I did. And it looks just like any other "reject" from any sourceforge-based site. :-)
Best regards,
Wolfgang Denk
jdl
participants (2)
-
Jon Loeliger -
Wolfgang Denk