help in the MM Area of the Linux Kernel
Hello all, i'am intrested in helping and Bug Fixing in the mm area of the linux kernel. For driver development is it clear check in the staging area the TODO's. And what is the process for other areas of the kernel for example mm (Memory management X86) ? In advance many thanks Damian
Any suggestions on this topic ? Am Donnerstag, den 05.10.2017, 18:14 +0200 schrieb Damian Tometzki:
Hello all,
i'am intrested in helping and Bug Fixing in the mm area of the linux kernel.
For driver development is it clear check in the staging area the TODO's.
And what is the process for other areas of the kernel for example mm (Memory management X86) ?
In advance many thanks Damian
Any suggestions on this topic ? BR Damian Am Donnerstag, den 05.10.2017, 18:14 +0200 schrieb Damian Tometzki:
Hello all,
i'am intrested in helping and Bug Fixing in the mm area of the linux kernel.
For driver development is it clear check in the staging area the TODO's.
And what is the process for other areas of the kernel for example mm (Memory management X86) ?
In advance many thanks Damian
On Thu, 05 Oct 2017 18:14:08 +0200, Damian Tometzki said:
i'am intrested in helping and Bug Fixing in the mm area of the linux kernel.
For driver development is it clear check in the staging area the TODO's.
And what is the process for other areas of the kernel for example mm (Memory management X86) ?
Rule 1 of kernel hacking: Not every mechanic gets to work on Formula 1 engines. For mm, you'll probably need to show some expertise in other kernel areas, *plus* have a deep understanding of memory management theory. That code has already been worked over by multiple professionals, which means pretty much all the easy stuff has already been done. Oh, and you're probably going to also need knowledge of the kernel instrumentation - perf, tracepoints, and friends. If you manage to find an actual bug in that code, it is most likely going to be some weird corner case, and *much* MM clue will be required to fix it without breaking some *other* more common corner case. Remember that the same code has to Do The Right Thing on everything from an embedded system with 32M of RAM and only one major process running, to large mainframe class boxes with a terabyte of RAM, a large Oracle instance, and several hundred Apache / Tomcat / etc processes flickering in and out of existence, to multi-terabyte systems running HPC (where the use of RDMA over Infiniband by things like MPI creates challenges due to a *lot* of locked pages...) Your best bet? Find a replicable corner case (this may require access to a variety of systems), create a test-case that can cause the corner case on demand. Figure out what the mm code is doing, and how it could do it better. Write a patch, test, and then double-check on other systems that you didn't cause a regression. Submit the patch. Good luck. :)
Hello Valdis, thank you you for you reply. I'work for a big company as SAP partner and i responsible for SAP HANA and the infrastucture. My personal intrest is the memory managment. My impression was the same as you answerded. As you said i will try to figure out what the mm code is doing and try to find out what can we do better. Damian Am Freitag, den 06.10.2017, 12:34 -0400 schrieb valdis.kletnieks@vt.edu:
On Thu, 05 Oct 2017 18:14:08 +0200, Damian Tometzki said:
i'am intrested in helping and Bug Fixing in the mm area of the linux kernel.
For driver development is it clear check in the staging area the TODO's.
And what is the process for other areas of the kernel for example mm (Memory management X86) ?
Rule 1 of kernel hacking: Not every mechanic gets to work on Formula 1 engines.
For mm, you'll probably need to show some expertise in other kernel areas, *plus* have a deep understanding of memory management theory. That code has already been worked over by multiple professionals, which means pretty much all the easy stuff has already been done.
Oh, and you're probably going to also need knowledge of the kernel instrumentation - perf, tracepoints, and friends.
If you manage to find an actual bug in that code, it is most likely going to be some weird corner case, and *much* MM clue will be required to fix it without breaking some *other* more common corner case. Remember that the same code has to Do The Right Thing on everything from an embedded system with 32M of RAM and only one major process running, to large mainframe class boxes with a terabyte of RAM, a large Oracle instance, and several hundred Apache / Tomcat / etc processes flickering in and out of existence, to multi-terabyte systems running HPC (where the use of RDMA over Infiniband by things like MPI creates challenges due to a *lot* of locked pages...)
Your best bet? Find a replicable corner case (this may require access to a variety of systems), create a test-case that can cause the corner case on demand. Figure out what the mm code is doing, and how it could do it better. Write a patch, test, and then double-check on other systems that you didn't cause a regression. Submit the patch.
Good luck. :)
_______________________________________________ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
Take a look to this: https://www.youtube.com/watch?v=6S9DDTFyjrY I attended a presentation with the same name in the last Kernel Recipes and was very clarifying. Cheers! On Fri, Oct 6, 2017 at 8:29 PM, Damian Tometzki <damian.tometzki@icloud.com> wrote:
Hello Valdis,
thank you you for you reply. I'work for a big company as SAP partner and i responsible for SAP HANA and the infrastucture.
My personal intrest is the memory managment. My impression was the same as you answerded.
As you said i will try to figure out what the mm code is doing and try to find out what can we do better.
Damian
Am Freitag, den 06.10.2017, 12:34 -0400 schrieb valdis.kletnieks@vt.edu:
On Thu, 05 Oct 2017 18:14:08 +0200, Damian Tometzki said:
i'am intrested in helping and Bug Fixing in the mm area of the linux kernel.
For driver development is it clear check in the staging area the TODO's.
And what is the process for other areas of the kernel for example mm (Memory management X86) ?
Rule 1 of kernel hacking: Not every mechanic gets to work on Formula 1 engines.
For mm, you'll probably need to show some expertise in other kernel areas, *plus* have a deep understanding of memory management theory. That code has already been worked over by multiple professionals, which means pretty much all the easy stuff has already been done.
Oh, and you're probably going to also need knowledge of the kernel instrumentation - perf, tracepoints, and friends.
If you manage to find an actual bug in that code, it is most likely going to be some weird corner case, and *much* MM clue will be required to fix it without breaking some *other* more common corner case. Remember that the same code has to Do The Right Thing on everything from an embedded system with 32M of RAM and only one major process running, to large mainframe class boxes with a terabyte of RAM, a large Oracle instance, and several hundred Apache / Tomcat / etc processes flickering in and out of existence, to multi-terabyte systems running HPC (where the use of RDMA over Infiniband by things like MPI creates challenges due to a *lot* of locked pages...)
Your best bet? Find a replicable corner case (this may require access to a variety of systems), create a test-case that can cause the corner case on demand. Figure out what the mm code is doing, and how it could do it better. Write a patch, test, and then double-check on other systems that you didn't cause a regression. Submit the patch.
Good luck. :)
_______________________________________________ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
_______________________________________________ Kernelnewbies mailing list Kernelnewbies@kernelnewbies.org https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
-- Ricardo Ribalda
2017-10-07 0:34 GMT+08:00 <valdis.kletnieks@vt.edu>:
On Thu, 05 Oct 2017 18:14:08 +0200, Damian Tometzki said:
i'am intrested in helping and Bug Fixing in the mm area of the linux kernel.
For driver development is it clear check in the staging area the TODO's.
And what is the process for other areas of the kernel for example mm (Memory management X86) ?
Rule 1 of kernel hacking: Not every mechanic gets to work on Formula 1 engines.
For mm, you'll probably need to show some expertise in other kernel areas, *plus* have a deep understanding of memory management theory. That code has already been worked over by multiple professionals, which means pretty much all the easy stuff has already been done.
Oh, and you're probably going to also need knowledge of the kernel instrumentation - perf, tracepoints, and friends.
If you manage to find an actual bug in that code, it is most likely going to be some weird corner case, and *much* MM clue will be required to fix it without breaking some *other* more common corner case. Remember that the same code has to Do The Right Thing on everything from an embedded system with 32M of RAM and only one major process running, to large mainframe class boxes with a terabyte of RAM, a large Oracle instance, and several hundred Apache / Tomcat / etc
um...terabyte of RAM? Can you name one of those machine with so much RAM? Yubin
On Mon, Oct 9, 2017 at 9:50 PM, Yubin Ruan <ablacktshirt@gmail.com> wrote:
...
to Do The Right Thing on everything from an embedded system with 32M of RAM and only one major process running, to large mainframe class boxes with a terabyte of RAM, a large Oracle instance, and several hundred Apache / Tomcat / etc
um...terabyte of RAM? Can you name one of those machine with so much RAM?
4 TB: http://investors.cray.com/phoenix.zhtml?c=98390&p=irol-newsArticle&ID=185527... 64 TB: https://news.ycombinator.com/item?id=7302770 690 TB: https://en.wikipedia.org/wiki/Titan_(supercomputer) ... Jeff
On Mon, 09 Oct 2017 21:58:27 -0400, Jeffrey Walton said:
On Tue, 10 Oct 2017 09:50:21 +0800, Yubin Ruan said:
um...terabyte of RAM? Can you name one of those machine with so much RAM?
4 TB: http://investors.cray.com/phoenix.zhtml?c=98390&p=irol-newsArticle&ID=185527... 64 TB: https://news.ycombinator.com/item?id=7302770 690 TB: https://en.wikipedia.org/wiki/Titan_(supercomputer)
Heck, Dell will sell you a 4U rack-mount with 3T in it. Got probably a half-dozen of them across the hall from my office for the HPC people. And we probably got a hundred or so nodes with 512G of RAM across the various clusters. And IBM will sell you a System/Z Z14 with 32T of memory, and crazy amounts of I/O capability (Imagine the equivalent of 320 PCI slots....) https://www.theregister.co.uk/2017/07/17/ibm_latest_mainframe_z14/ https://www-01.ibm.com/common/ssi/cgi-bin/ssialias?htmlfid=ZSD03046USEN& And yes, it will run Linux, it's a s390 architecture box.
participants (5)
-
Damian Tometzki -
Jeffrey Walton -
Ricardo Ribalda Delgado -
valdis.kletnieks＠vt.edu -
Yubin Ruan