Skip to contentSkip to conversation

catching up on the backlog

DISCUSSION [email protected] 4 participants 9 messages May 3, 2005 – May 5, 2005

lore
Share & tools
lore
#0 lore raw

I've picked through the list of patches in my backlog, and applied this bunch to my release tree ... before I ask Linus to pull, it would be useful to get some SGI testing (since the majority of the changes are SGI related).

-Tony

 arch/ia64/Kconfig                   |   19 
 arch/ia64/hp/common/sba_iommu.c     |   34 
 arch/ia64/kernel/acpi.c             |   23 
 arch/ia64/kernel/entry.S            |    2 
 arch/ia64/kernel/fsys.S             |    4 
 arch/ia64/kernel/mca_drv.c          |    4 
 arch/ia64/kernel/mca_drv_asm.S      |   18 
 arch/ia64/kernel/process.c          |   41 
 arch/ia64/kernel/signal.c           |    3 
 arch/ia64/lib/flush.S               |    6 
 arch/ia64/lib/memcpy_mck.S          |    2 
 arch/ia64/lib/memset.S              |    2 
 arch/ia64/sn/kernel/Makefile        |    7 
 arch/ia64/sn/kernel/io_init.c       |   10 
 arch/ia64/sn/kernel/mca.c           |   34 
 arch/ia64/sn/kernel/setup.c         |   31 
 arch/ia64/sn/kernel/tiocx.c         |   60 
 arch/ia64/sn/kernel/xp_main.c       |  289 +++
 arch/ia64/sn/kernel/xp_nofault.S    |   31 
 arch/ia64/sn/kernel/xpc.h           |  991 +++++++++++
 arch/ia64/sn/kernel/xpc_channel.c   | 2297 ++++++++++++++++++++++++++
 arch/ia64/sn/kernel/xpc_main.c      | 1064 ++++++++++++
 arch/ia64/sn/kernel/xpc_partition.c |  984 +++++++++++
 arch/ia64/sn/kernel/xpnet.c         |  715 ++++++++
 arch/ia64/sn/pci/pcibr/pcibr_dma.c  |    2 
 arch/ia64/sn/pci/tioca_provider.c   |    2 
 drivers/char/Kconfig                |    2 
 include/asm-ia64/sn/addrs.h         |    8 
 include/asm-ia64/sn/arch.h          |   17 
 include/asm-ia64/sn/fetchop.h       |   85 
 include/asm-ia64/sn/l1.h            |    3 
 include/asm-ia64/sn/pda.h           |    1 
 include/asm-ia64/sn/shub_mmr.h      |   24 
 include/asm-ia64/sn/shubio.h        | 3108 +++++++++++++++++-------------------
 include/asm-ia64/sn/sn_cpuid.h      |    7 
 include/asm-ia64/sn/sn_fru.h        |   44 
 include/asm-ia64/sn/sndrv.h         |   47 
 include/asm-ia64/sn/xp.h            |  436 +++++
 kernel/exit.c                       |    2 
 mm/page_alloc.c                     |    2 
 40 files changed, 8525 insertions(+), 1936 deletions(-)

commit 012914dad25bd5cacf88af4429eecda62a06020d
tree e265fa0cf7dccef220bc26a16c1f0694b4b4bbee
parent 3ea8b477b4b9d3e75b5e9b8aea41259f45031823
author Russ Anderson <([email protected])> Sat, 23 Apr 2005 00:08:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:58:17 -0700

[patch] MCA recovery module undefined symbol fix

The patch "MCA recovery improvements" added do_exit to mca_drv.c. That's fine when the mca recovery code is built in the kernel (CONFIG_IA64_MCA_RECOVERY=y) but breaks building the mca recovery code as a module (CONFIG_IA64_MCA_RECOVERY=m).

Most users are currently building this as a module, as loading and unloading the module provides a very convenient way to turn on/off error recovery.

This patch exports do_exit, so mca_drv.c can build as a module.

commit 3ea8b477b4b9d3e75b5e9b8aea41259f45031823
tree d9ab6af871931241152f0d87084a1ba0ace02480
parent 32709d8ae67356559839a9a9e4b26f79134e45a6
author Mark Maule <[email protected]> Mon, 11 Apr 2005 21:20:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:52:22 -0700

[IA64] altix: fix TIOCA dmamap list_add

Correct a bug where tioca_dma_mapped() is putting tioca dma map structs on the wrong list.

commit 32709d8ae67356559839a9a9e4b26f79134e45a6
tree 876f7e6d3f3f5d835870c8db34d141959db754b0
parent b1b901c2029aa0fd8aa4ac5f04f31648ae2358b4
author Keith Owens <[email protected]> Fri, 08 Apr 2005 14:23:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:48:55 -0700

[IA64] SAL to OS callbacks cannot call sleeping

When SAL calls back into the OS, the OS code is running with preempt disabled so it cannot call sleeping functions.

commit b1b901c2029aa0fd8aa4ac5f04f31648ae2358b4
tree 43f5dde2309d66fa2112a7c2dc46ea7c63fae459
parent 446b8831f5acf2076fa58a66286789eb84f3df2c
author Russ Anderson <[email protected]> Wed, 06 Apr 2005 00:07:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:47:42 -0700

[IA64] MCA recovery improvements

Jack Steiner uncovered some opportunities for improvement in the MCA recovery code.

1) Set bsp to save registers on the kernel stack.
2) Disable interrupts while in the MCA recovery code.
3) Change the way the user process is killed, to avoid a panic in schedule.

Testing shows that these changes make the recovery code much more reliable with the 2.6.12 kernel.

commit 446b8831f5acf2076fa58a66286789eb84f3df2c
tree 250506863a6e103444584d609cdb577e87b3a499
parent 7d5f9c0f10255000ca007fb03773c6b825c2b9ce
author David Woodhouse <[email protected]> Tue, 05 Apr 2005 17:47:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:45:39 -0700

[IA64] fix ia64 syscall auditing

Attached is a patch against David's audit.17 kernel that adds checks for the TIF_SYSCALL_AUDIT thread flag to the ia64 system call and signal handling code paths. The patch enables auditing of system calls set up via fsys_bubble_down, as well as ensuring that audit_syscall_exit() is called on return from sigreturn.

Neglecting to check for TIF_SYSCALL_AUDIT at these points results in incorrect information in audit_context, causing frequent system panics when system call auditing is enabled on an ia64 system.

I have tested this patch and have seen no problems with it.

[Original patch from Amy Griffis ported to current kernel by David Woodhouse]

From: Amy Griffis <[email protected]>
From: David Woodhouse <[email protected]>

commit 7d5f9c0f10255000ca007fb03773c6b825c2b9ce
tree 5f53e41ef4cbe5ced2c5884ee5dde8df1fcaa509
parent de7548d0e202263bb6bfd7574a7889e85a691937
author Zwane Mwaikambo <[email protected]> Wed, 30 Mar 2005 21:40:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:40:18 -0700

[IA64] reduce cacheline bouncing in cpu_idle_wait

Andi noted that during normal runtime cpu_idle_map is bounced around a lot, and occassionally at a higher frequency than the timer interrupt wakeup which we normally exit pm_idle from. So switch to a percpu variable.

I didn't move things to the slow path because it would involve adding scheduler code to wakeup the idle thread on the cpus we're waiting for.

commit de7548d0e202263bb6bfd7574a7889e85a691937
tree eee071a29e77fbfacc5dc6ff8c9c10d9dcc23f02
parent bb0fc085457cf6a865b8232b0cefab3a7819df44
author Mike Habeck <[email protected]> Fri, 25 Mar 2005 19:34:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:36:09 -0700

[IA64-SGI] Altix only: Fix for sn_dma_flush

The following patch fixes a bug in the SGI Altix sn_dma_flush code. sn_dma_flush is broken in 2.6. The code isn't waiting for the DMA data to be flushed out of the PIC ASIC. This patch is based off the linux-ia64-test-2.6.12 tree

commit bb0fc085457cf6a865b8232b0cefab3a7819df44
tree deaa9e3b58d6e40cb2bdba61898bf8373dc02688
parent c0b12422e5e1d041026dd27074de17d2d7e32c4e
author Alex Williamson <[email protected]> Thu, 24 Mar 2005 22:58:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:33:18 -0700

[IA64] use common pxm function

This patch simplifies a couple places where we search for _PXM values in ACPI namespace. Thanks,

commit c0b12422e5e1d041026dd27074de17d2d7e32c4e
tree 1818ff91a52bea353469ba787f492dd2ca303fe2
parent 3a7d555bfc4d4631d9118fb4d0ed7ab62cc2ca1c
author Colin Ngam <[email protected]> Fri, 18 Mar 2005 16:38:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:21:03 -0700

[IA64-SGI] Altix only: Register Error Interrupt

The following patch ensures that the correct error interrupt handling routine is initialized. This patch is based on the 2.6.12 ia64 release tree.

commit 3a7d555bfc4d4631d9118fb4d0ed7ab62cc2ca1c
tree 3f788937066dff5befaf3b4e70539bf3f9e4702b
parent a2d974da0afe659cff98913184a97c0ee686d02b
author Dean Nelson <[email protected]> Mon, 04 Apr 2005 13:14:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:50:22 -0700

[IA64-SGI] convert AMO address found in XPC's reserved page

This patch detects the existence of an uncached physical AMO address setup by EFI's XPBOOT (SGI) and converts it to an uncached virtual AMO address. Depends on a patch submitted on 23 March 2005 with the subject of:
[PATCH 2/3] SGI Altix cross partition functionality (2nd revision)

commit a2d974da0afe659cff98913184a97c0ee686d02b
tree 1a1479280445226104ddc062bed2e2806639352e
parent 89eb8eb927e324366c3ac0458998aaf9953fc5cd
author Dean Nelson <[email protected]> Wed, 23 Mar 2005 20:50:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:37:19 -0700

[IA64-SGI] SGI Altix cross partition functionality [3/3]

This patch contains the cross partition pseudo-ethernet driver (XPNET) functional support module.

commit 89eb8eb927e324366c3ac0458998aaf9953fc5cd
tree c5f77d88bc42821134de6ea49a5663654df38e56
parent 21223a9e78050919499d3d9039170e608eb939cc
author Dean Nelson <[email protected]> Wed, 23 Mar 2005 19:50:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:36:00 -0700

[IA64-SGI] SGI Altix cross partition functionality [2/3]

This patch contains the communication module (XPC) for cross partition communication on a partitioned SGI Altix.

commit 21223a9e78050919499d3d9039170e608eb939cc
tree 8691970d17425f6e79b0567633d106ce88db0b8e
parent b0d82bd5df874f7dadbeced1b0163473387da37c
author Tony Luck <[email protected]> Tue, 03 May 2005 12:25:50 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:25:50 -0700

[IA64] manually apply changes to arch/ia64/sn/kernel/Makefile

cg-patch couldn't apply the patch to Makefile, and my dumb script rushed on and ran cg-commit without this change.

commit b0d82bd5df874f7dadbeced1b0163473387da37c
tree c78294ddaff17034ef77253c587a7512ba813e95
parent 7fbd2a5337b2aa91266abbded97330f909904fd5
author Dean Nelson <[email protected]> Wed, 23 Mar 2005 19:46:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:16:52 -0700

[IA64-SGI] SGI Altix cross partition functionality (2nd
This patch contains the shim module (XP) which interfaces between the communication module (XPC) and the functional support modules (like XPNET).

commit 7fbd2a5337b2aa91266abbded97330f909904fd5
tree 0aca2260c63b65068e69bc82b5ab55f4da4b9421
parent 21e37283909c12e300ab87c20f5addc878cda9f9
author Dean Nelson <[email protected]> Wed, 23 Mar 2005 19:11:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:12:42 -0700

[IA64-SGI] Add some needed externs currently not defined

Add some needed externs currently not defined.

commit 21e37283909c12e300ab87c20f5addc878cda9f9
tree 0a03d3c0c90d8108eefb985272d0d49f31c2c827
parent 7223a93a5321f84337647aef62ef947afd8df41a
author Dean Nelson <[email protected]> Wed, 23 Mar 2005 19:08:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:11:38 -0700

[IA64-SGI] Define some additional SHub1 and Shub2 register symbols

Define some additional SHub1 and SHub2 register symbols.

commit 7223a93a5321f84337647aef62ef947afd8df41a
tree 10d4c8e4cb1b4ddf33ca32efbe915ca91d19fcd1
parent 2e34f07ff0c944399a6456e2d91cf0ca1d9a497c
author Dean Nelson <[email protected]> Wed, 23 Mar 2005 19:00:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:09:32 -0700

[IA64] Export node_online_map and node_possible_map

Export node_online_map and node_possible_map so that kernel modules can use the nodemask macros, like, for_each_node() and for_each_online_node().

commit 2e34f07ff0c944399a6456e2d91cf0ca1d9a497c
tree d09796e3f37d7cd809eceb5d931f1e2171a32f5b
parent ce0a3956b32650e229b68964c4400bbdc5ad3ca1
author Dean Nelson <[email protected]> Mon, 21 Mar 2005 19:41:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:07:46 -0700

[PATCH] move cnodeid_to_nasid_table out of pda
Another step in the effort to eliminate the SN pda structure. This patch moves the cnodeid_to_nasid_table field out of the pda, making it a standalone per-cpu data item, and exports it so it can be accessed by kernel modules.

commit ce0a3956b32650e229b68964c4400bbdc5ad3ca1
tree 79ea53b8321964c903f1379c48ba5c4917992c46
parent ae40aae9b9b8e336714ebb3f16410da6e69d6ac8
author Bruce Losure <[email protected]> Mon, 25 Apr 2005 19:41:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:01:07 -0700

[IA64-SGI] Altix patch to add bricktype knowledge to tiocx

Here is a patch to enable the SGI tiocx bus driver to distingush between FPGA-attached h/w and non-FPGA-attached h/w.

commit ae40aae9b9b8e336714ebb3f16410da6e69d6ac8
tree aa900ab7187e5206d30f78c1e5fad98a84a81293
parent c2d1d65ad441c8abe624bdb1c2cff2e47c8c1ee1
author Bruce Losure <[email protected]> Mon, 04 Apr 2005 13:23:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:00:01 -0700

[IA64-SGI] Altix patch to fix missing Kconfig dependency.

This is a one-liner to make the mbcs driver depend on SGI_TIOCX in the drivers/char/Kconfig file.

commit c2d1d65ad441c8abe624bdb1c2cff2e47c8c1ee1
tree 3d88b27fe6409d1ebc9849e29b0e660d437decce
parent 2074615a13a4f250e0a4e3f6ec8e3733b950a783
author Bruce Losure <[email protected]> Thu, 24 Mar 2005 12:28:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 11:58:37 -0700

[IA64-SGI] Altix only: Remove hubdev SAL call

Hi Tony,

This patch against ia64-test-2.6.12 fixes a bug where the tiocx code was inadvertently un-doing some address modifications done in earlier
fixup code. This patch just removes the offending code.

commit 2074615a13a4f250e0a4e3f6ec8e3733b950a783
tree 20af73cd640c88f5bedcb081fd5e850bf23b7b86
parent 43cc67251882f60c796c8729cefc0e05b550976c
author David Mosberger-Tang <[email protected]> Fri, 18 Feb 2005 19:09:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 11:27:33 -0700

[IA64] use fc.i for fluch_icache_range()

This is a small patch to switch fluch_icache_range() to use fc.i instead of fc. This would save time on processors which can establish i-cache coherency without flushing the cache-line out to memory (not that any current processors do). On existing processors, fc.i behaves like fc. The only caveat is that very old assemblers may not know about fc.i yet.

commit 43cc67251882f60c796c8729cefc0e05b550976c
tree 7d5a7a24eaf4f6afc76ce98b368eb9579bf59c8b
parent 6adc4cc0eead0c1897d7f1416d749c8e7c91e0bc
author Jes Sorensen <[email protected]> Thu, 17 Feb 2005 09:41:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 11:26:05 -0700

[IA64-SGI] cleanup shubio.h

This patch cleans up include/asm/sn/shubio.h by removing a ton of whitespaces and running it through Lindent, reducing it's size by almost 30KB. No actual content has been changed.

commit 6adc4cc0eead0c1897d7f1416d749c8e7c91e0bc
tree 0409af16566bde54a3d75fb0675a26a5cbc4603a
parent 9df6f705c0934a49d0f0a3468a5b5044c8aec4f1
author Jesse Barnes <[email protected]> Wed, 16 Feb 2005 16:38:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 11:16:31 -0700

[IA64-SGI] remove unused sn header files

This patch makes Jes' patch (which also contains the removal of fetchop.h) a bit smaller, and removes two other unused files at the same time, sndrv.h and sn_fru.h.

commit 9df6f705c0934a49d0f0a3468a5b5044c8aec4f1
tree 045807d5f55b61d6222c6dada311237a9badeddc
parent 512f64295f2f0049515dcc4e97c1f1ae0df9629c
author David Mosberger-Tang <[email protected]> Fri, 25 Mar 2005 00:16:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 10:56:42 -0700

[IA64] fix typos caught by new assembler

Patch below fixes 3 trivial typos which are caught by the new assembler (v2.169.90). Please apply.

[Note: fix to memcpy that was also part of this patch was separately
applied from patches by H.J. and Andreas ... so the delta here only has the other two fixes. -Tony]

commit 512f64295f2f0049515dcc4e97c1f1ae0df9629c
tree d2079e3972a7cd52f4b72cbd988dd4168dc2c0ec
parent 7a9bdd8842d0847f3b15068550a3632a2d259057
author Andreas Schwab <[email protected]> Tue, 26 Apr 2005 23:00:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 10:49:29 -0700

[IA64] Fix memcpy_mck.S for current binutils

The current ia64 assembler complains about mismatching .proc/.endp pairs. (Same patch also sent by H.J. Lu)

commit 7a9bdd8842d0847f3b15068550a3632a2d259057
tree 548d1f72a9dbdf74b67e45f5f536a71e94c26b81
parent ac09f698f1cda91e890fb75f4cb38253d60ff017
author Suresh Siddha <[email protected]> Tue, 05 Apr 2005 18:05:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 10:33:28 -0700

[IA64] Add config SCHED_SMT

Now that we have MC/MT detection patches in, appended patch allows us to configure MT scheduler optimizations. For now, we will this option off by default.

There is some discussion going on lkml about setting up sched-domains which are absolutely needed (like for example, we shouldn't setup SMT domain for non MT processors). Once that patch goes in, we can enable this option by default.

#1 lore raw

Hi Tony,

Quoted 4 lines · Tony

I've picked through the list of patches in my backlog, and applied this bunch to my release tree ... before I ask Linus to pull, it would be useful to get some SGI testing (since the majority of the changes are SGI related).

Thanks to SGI we can now help with sn testing. I cloned the testing tree today, compiled and booted on Altix350 4way successful.

- dsw

Quoted 475 lines

-Tony

 arch/ia64/Kconfig                   |   19 
 arch/ia64/hp/common/sba_iommu.c     |   34 
 arch/ia64/kernel/acpi.c             |   23 
 arch/ia64/kernel/entry.S            |    2 
 arch/ia64/kernel/fsys.S             |    4 
 arch/ia64/kernel/mca_drv.c          |    4 
 arch/ia64/kernel/mca_drv_asm.S      |   18 
 arch/ia64/kernel/process.c          |   41 
 arch/ia64/kernel/signal.c           |    3 
 arch/ia64/lib/flush.S               |    6 
 arch/ia64/lib/memcpy_mck.S          |    2 
 arch/ia64/lib/memset.S              |    2 
 arch/ia64/sn/kernel/Makefile        |    7 
 arch/ia64/sn/kernel/io_init.c       |   10 
 arch/ia64/sn/kernel/mca.c           |   34 
 arch/ia64/sn/kernel/setup.c         |   31 
 arch/ia64/sn/kernel/tiocx.c         |   60 
 arch/ia64/sn/kernel/xp_main.c       |  289 +++
 arch/ia64/sn/kernel/xp_nofault.S    |   31 
 arch/ia64/sn/kernel/xpc.h           |  991 +++++++++++
 arch/ia64/sn/kernel/xpc_channel.c   | 2297 ++++++++++++++++++++++++++
 arch/ia64/sn/kernel/xpc_main.c      | 1064 ++++++++++++
 arch/ia64/sn/kernel/xpc_partition.c |  984 +++++++++++
 arch/ia64/sn/kernel/xpnet.c         |  715 ++++++++
 arch/ia64/sn/pci/pcibr/pcibr_dma.c  |    2 
 arch/ia64/sn/pci/tioca_provider.c   |    2 
 drivers/char/Kconfig                |    2 
 include/asm-ia64/sn/addrs.h         |    8 
 include/asm-ia64/sn/arch.h          |   17 
 include/asm-ia64/sn/fetchop.h       |   85 
 include/asm-ia64/sn/l1.h            |    3 
 include/asm-ia64/sn/pda.h           |    1 
 include/asm-ia64/sn/shub_mmr.h      |   24 
 include/asm-ia64/sn/shubio.h        | 3108 +++++++++++++++++-------------------
 include/asm-ia64/sn/sn_cpuid.h      |    7 
 include/asm-ia64/sn/sn_fru.h        |   44 
 include/asm-ia64/sn/sndrv.h         |   47 
 include/asm-ia64/sn/xp.h            |  436 +++++
 kernel/exit.c                       |    2 
 mm/page_alloc.c                     |    2 
 40 files changed, 8525 insertions(+), 1936 deletions(-)

commit 012914dad25bd5cacf88af4429eecda62a06020d
tree e265fa0cf7dccef220bc26a16c1f0694b4b4bbee
parent 3ea8b477b4b9d3e75b5e9b8aea41259f45031823
author Russ Anderson <([email protected])> Sat, 23 Apr 2005 00:08:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:58:17 -0700

[patch] MCA recovery module undefined symbol fix

The patch "MCA recovery improvements" added do_exit to mca_drv.c. That's fine when the mca recovery code is built in the kernel (CONFIG_IA64_MCA_RECOVERY=y) but breaks building the mca recovery code as a module (CONFIG_IA64_MCA_RECOVERY=m).

Most users are currently building this as a module, as loading and unloading the module provides a very convenient way to turn on/off error recovery.

This patch exports do_exit, so mca_drv.c can build as a module.

commit 3ea8b477b4b9d3e75b5e9b8aea41259f45031823
tree d9ab6af871931241152f0d87084a1ba0ace02480
parent 32709d8ae67356559839a9a9e4b26f79134e45a6
author Mark Maule <[email protected]> Mon, 11 Apr 2005 21:20:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:52:22 -0700

[IA64] altix: fix TIOCA dmamap list_add

Correct a bug where tioca_dma_mapped() is putting tioca dma map structs on the wrong list.

commit 32709d8ae67356559839a9a9e4b26f79134e45a6
tree 876f7e6d3f3f5d835870c8db34d141959db754b0
parent b1b901c2029aa0fd8aa4ac5f04f31648ae2358b4
author Keith Owens <[email protected]> Fri, 08 Apr 2005 14:23:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:48:55 -0700

[IA64] SAL to OS callbacks cannot call sleeping

When SAL calls back into the OS, the OS code is running with preempt disabled so it cannot call sleeping functions.

commit b1b901c2029aa0fd8aa4ac5f04f31648ae2358b4
tree 43f5dde2309d66fa2112a7c2dc46ea7c63fae459
parent 446b8831f5acf2076fa58a66286789eb84f3df2c
author Russ Anderson <[email protected]> Wed, 06 Apr 2005 00:07:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:47:42 -0700

[IA64] MCA recovery improvements

Jack Steiner uncovered some opportunities for improvement in the MCA recovery code.

1) Set bsp to save registers on the kernel stack.
2) Disable interrupts while in the MCA recovery code.
3) Change the way the user process is killed, to avoid a panic in schedule.

Testing shows that these changes make the recovery code much more reliable with the 2.6.12 kernel.

commit 446b8831f5acf2076fa58a66286789eb84f3df2c
tree 250506863a6e103444584d609cdb577e87b3a499
parent 7d5f9c0f10255000ca007fb03773c6b825c2b9ce
author David Woodhouse <[email protected]> Tue, 05 Apr 2005 17:47:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:45:39 -0700

[IA64] fix ia64 syscall auditing

Attached is a patch against David's audit.17 kernel that adds checks for the TIF_SYSCALL_AUDIT thread flag to the ia64 system call and signal handling code paths. The patch enables auditing of system calls set up via fsys_bubble_down, as well as ensuring that audit_syscall_exit() is called on return from sigreturn.

Neglecting to check for TIF_SYSCALL_AUDIT at these points results in incorrect information in audit_context, causing frequent system panics when system call auditing is enabled on an ia64 system.

I have tested this patch and have seen no problems with it.

[Original patch from Amy Griffis ported to current kernel by David Woodhouse]

From: Amy Griffis <[email protected]>
From: David Woodhouse <[email protected]>

commit 7d5f9c0f10255000ca007fb03773c6b825c2b9ce
tree 5f53e41ef4cbe5ced2c5884ee5dde8df1fcaa509
parent de7548d0e202263bb6bfd7574a7889e85a691937
author Zwane Mwaikambo <[email protected]> Wed, 30 Mar 2005 21:40:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:40:18 -0700

[IA64] reduce cacheline bouncing in cpu_idle_wait

Andi noted that during normal runtime cpu_idle_map is bounced around a lot, and occassionally at a higher frequency than the timer interrupt wakeup which we normally exit pm_idle from. So switch to a percpu variable.

I didn't move things to the slow path because it would involve adding scheduler code to wakeup the idle thread on the cpus we're waiting for.

commit de7548d0e202263bb6bfd7574a7889e85a691937
tree eee071a29e77fbfacc5dc6ff8c9c10d9dcc23f02
parent bb0fc085457cf6a865b8232b0cefab3a7819df44
author Mike Habeck <[email protected]> Fri, 25 Mar 2005 19:34:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:36:09 -0700

[IA64-SGI] Altix only: Fix for sn_dma_flush

The following patch fixes a bug in the SGI Altix sn_dma_flush code. sn_dma_flush is broken in 2.6. The code isn't waiting for the DMA data to be flushed out of the PIC ASIC. This patch is based off the linux-ia64-test-2.6.12 tree

commit bb0fc085457cf6a865b8232b0cefab3a7819df44
tree deaa9e3b58d6e40cb2bdba61898bf8373dc02688
parent c0b12422e5e1d041026dd27074de17d2d7e32c4e
author Alex Williamson <[email protected]> Thu, 24 Mar 2005 22:58:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:33:18 -0700

[IA64] use common pxm function

This patch simplifies a couple places where we search for _PXM values in ACPI namespace. Thanks,

commit c0b12422e5e1d041026dd27074de17d2d7e32c4e
tree 1818ff91a52bea353469ba787f492dd2ca303fe2
parent 3a7d555bfc4d4631d9118fb4d0ed7ab62cc2ca1c
author Colin Ngam <[email protected]> Fri, 18 Mar 2005 16:38:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 13:21:03 -0700

[IA64-SGI] Altix only: Register Error Interrupt

The following patch ensures that the correct error interrupt handling routine is initialized. This patch is based on the 2.6.12 ia64 release tree.

commit 3a7d555bfc4d4631d9118fb4d0ed7ab62cc2ca1c
tree 3f788937066dff5befaf3b4e70539bf3f9e4702b
parent a2d974da0afe659cff98913184a97c0ee686d02b
author Dean Nelson <[email protected]> Mon, 04 Apr 2005 13:14:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:50:22 -0700

[IA64-SGI] convert AMO address found in XPC's reserved page

This patch detects the existence of an uncached physical AMO address setup by EFI's XPBOOT (SGI) and converts it to an uncached virtual AMO address. Depends on a patch submitted on 23 March 2005 with the subject of:
[PATCH 2/3] SGI Altix cross partition functionality (2nd revision)

commit a2d974da0afe659cff98913184a97c0ee686d02b
tree 1a1479280445226104ddc062bed2e2806639352e
parent 89eb8eb927e324366c3ac0458998aaf9953fc5cd
author Dean Nelson <[email protected]> Wed, 23 Mar 2005 20:50:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:37:19 -0700

[IA64-SGI] SGI Altix cross partition functionality [3/3]

This patch contains the cross partition pseudo-ethernet driver (XPNET) functional support module.

commit 89eb8eb927e324366c3ac0458998aaf9953fc5cd
tree c5f77d88bc42821134de6ea49a5663654df38e56
parent 21223a9e78050919499d3d9039170e608eb939cc
author Dean Nelson <[email protected]> Wed, 23 Mar 2005 19:50:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:36:00 -0700

[IA64-SGI] SGI Altix cross partition functionality [2/3]

This patch contains the communication module (XPC) for cross partition communication on a partitioned SGI Altix.

commit 21223a9e78050919499d3d9039170e608eb939cc
tree 8691970d17425f6e79b0567633d106ce88db0b8e
parent b0d82bd5df874f7dadbeced1b0163473387da37c
author Tony Luck <[email protected]> Tue, 03 May 2005 12:25:50 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:25:50 -0700

[IA64] manually apply changes to arch/ia64/sn/kernel/Makefile

cg-patch couldn't apply the patch to Makefile, and my dumb script rushed on and ran cg-commit without this change.

commit b0d82bd5df874f7dadbeced1b0163473387da37c
tree c78294ddaff17034ef77253c587a7512ba813e95
parent 7fbd2a5337b2aa91266abbded97330f909904fd5
author Dean Nelson <[email protected]> Wed, 23 Mar 2005 19:46:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:16:52 -0700

[IA64-SGI] SGI Altix cross partition functionality (2nd
This patch contains the shim module (XP) which interfaces between the communication module (XPC) and the functional support modules (like XPNET).

commit 7fbd2a5337b2aa91266abbded97330f909904fd5
tree 0aca2260c63b65068e69bc82b5ab55f4da4b9421
parent 21e37283909c12e300ab87c20f5addc878cda9f9
author Dean Nelson <[email protected]> Wed, 23 Mar 2005 19:11:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:12:42 -0700

[IA64-SGI] Add some needed externs currently not defined

Add some needed externs currently not defined.

commit 21e37283909c12e300ab87c20f5addc878cda9f9
tree 0a03d3c0c90d8108eefb985272d0d49f31c2c827
parent 7223a93a5321f84337647aef62ef947afd8df41a
author Dean Nelson <[email protected]> Wed, 23 Mar 2005 19:08:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:11:38 -0700

[IA64-SGI] Define some additional SHub1 and Shub2 register symbols

Define some additional SHub1 and SHub2 register symbols.

commit 7223a93a5321f84337647aef62ef947afd8df41a
tree 10d4c8e4cb1b4ddf33ca32efbe915ca91d19fcd1
parent 2e34f07ff0c944399a6456e2d91cf0ca1d9a497c
author Dean Nelson <[email protected]> Wed, 23 Mar 2005 19:00:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:09:32 -0700

[IA64] Export node_online_map and node_possible_map

Export node_online_map and node_possible_map so that kernel modules can use the nodemask macros, like, for_each_node() and for_each_online_node().

commit 2e34f07ff0c944399a6456e2d91cf0ca1d9a497c
tree d09796e3f37d7cd809eceb5d931f1e2171a32f5b
parent ce0a3956b32650e229b68964c4400bbdc5ad3ca1
author Dean Nelson <[email protected]> Mon, 21 Mar 2005 19:41:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:07:46 -0700

[PATCH] move cnodeid_to_nasid_table out of pda
Another step in the effort to eliminate the SN pda structure. This patch moves the cnodeid_to_nasid_table field out of the pda, making it a standalone per-cpu data item, and exports it so it can be accessed by kernel modules.

commit ce0a3956b32650e229b68964c4400bbdc5ad3ca1
tree 79ea53b8321964c903f1379c48ba5c4917992c46
parent ae40aae9b9b8e336714ebb3f16410da6e69d6ac8
author Bruce Losure <[email protected]> Mon, 25 Apr 2005 19:41:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:01:07 -0700

[IA64-SGI] Altix patch to add bricktype knowledge to tiocx

Here is a patch to enable the SGI tiocx bus driver to distingush between FPGA-attached h/w and non-FPGA-attached h/w.

commit ae40aae9b9b8e336714ebb3f16410da6e69d6ac8
tree aa900ab7187e5206d30f78c1e5fad98a84a81293
parent c2d1d65ad441c8abe624bdb1c2cff2e47c8c1ee1
author Bruce Losure <[email protected]> Mon, 04 Apr 2005 13:23:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 12:00:01 -0700

[IA64-SGI] Altix patch to fix missing Kconfig dependency.

This is a one-liner to make the mbcs driver depend on SGI_TIOCX in the drivers/char/Kconfig file.

commit c2d1d65ad441c8abe624bdb1c2cff2e47c8c1ee1
tree 3d88b27fe6409d1ebc9849e29b0e660d437decce
parent 2074615a13a4f250e0a4e3f6ec8e3733b950a783
author Bruce Losure <[email protected]> Thu, 24 Mar 2005 12:28:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 11:58:37 -0700

[IA64-SGI] Altix only: Remove hubdev SAL call

Hi Tony,

This patch against ia64-test-2.6.12 fixes a bug where the tiocx code was inadvertently un-doing some address modifications done in earlier
fixup code. This patch just removes the offending code.

commit 2074615a13a4f250e0a4e3f6ec8e3733b950a783
tree 20af73cd640c88f5bedcb081fd5e850bf23b7b86
parent 43cc67251882f60c796c8729cefc0e05b550976c
author David Mosberger-Tang <[email protected]> Fri, 18 Feb 2005 19:09:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 11:27:33 -0700

[IA64] use fc.i for fluch_icache_range()

This is a small patch to switch fluch_icache_range() to use fc.i instead of fc. This would save time on processors which can establish i-cache coherency without flushing the cache-line out to memory (not that any current processors do). On existing processors, fc.i behaves like fc. The only caveat is that very old assemblers may not know about fc.i yet.

commit 43cc67251882f60c796c8729cefc0e05b550976c
tree 7d5a7a24eaf4f6afc76ce98b368eb9579bf59c8b
parent 6adc4cc0eead0c1897d7f1416d749c8e7c91e0bc
author Jes Sorensen <[email protected]> Thu, 17 Feb 2005 09:41:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 11:26:05 -0700

[IA64-SGI] cleanup shubio.h

This patch cleans up include/asm/sn/shubio.h by removing a ton of whitespaces and running it through Lindent, reducing it's size by almost 30KB. No actual content has been changed.

commit 6adc4cc0eead0c1897d7f1416d749c8e7c91e0bc
tree 0409af16566bde54a3d75fb0675a26a5cbc4603a
parent 9df6f705c0934a49d0f0a3468a5b5044c8aec4f1
author Jesse Barnes <[email protected]> Wed, 16 Feb 2005 16:38:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 11:16:31 -0700

[IA64-SGI] remove unused sn header files

This patch makes Jes' patch (which also contains the removal of fetchop.h) a bit smaller, and removes two other unused files at the same time, sndrv.h and sn_fru.h.

commit 9df6f705c0934a49d0f0a3468a5b5044c8aec4f1
tree 045807d5f55b61d6222c6dada311237a9badeddc
parent 512f64295f2f0049515dcc4e97c1f1ae0df9629c
author David Mosberger-Tang <[email protected]> Fri, 25 Mar 2005 00:16:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 10:56:42 -0700

[IA64] fix typos caught by new assembler

Patch below fixes 3 trivial typos which are caught by the new assembler (v2.169.90). Please apply.

[Note: fix to memcpy that was also part of this patch was separately
applied from patches by H.J. and Andreas ... so the delta here only has the other two fixes. -Tony]

commit 512f64295f2f0049515dcc4e97c1f1ae0df9629c
tree d2079e3972a7cd52f4b72cbd988dd4168dc2c0ec
parent 7a9bdd8842d0847f3b15068550a3632a2d259057
author Andreas Schwab <[email protected]> Tue, 26 Apr 2005 23:00:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 10:49:29 -0700

[IA64] Fix memcpy_mck.S for current binutils

The current ia64 assembler complains about mismatching .proc/.endp pairs. (Same patch also sent by H.J. Lu)

commit 7a9bdd8842d0847f3b15068550a3632a2d259057
tree 548d1f72a9dbdf74b67e45f5f536a71e94c26b81
parent ac09f698f1cda91e890fb75f4cb38253d60ff017
author Suresh Siddha <[email protected]> Tue, 05 Apr 2005 18:05:00 -0700 committer Tony Luck <[email protected]> Tue, 03 May 2005 10:33:28 -0700

[IA64] Add config SCHED_SMT

Now that we have MC/MT detection patches in, appended patch allows us to configure MT scheduler optimizations. For now, we will this option off by default.

There is some discussion going on lkml about setting up sched-domains which are absolutely needed (like for example, we shouldn't setup SMT domain for non MT processors). Once that patch goes in, we can enable this option by default.

Quoted 7 lines

-
To unsubscribe from this list: send the line "unsubscribe linux-ia64" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

--------------------------------------------------
Darren Williams <dsw AT gelato.unsw.edu.au>
Gelato@UNSW <www.gelato.unsw.edu.au>
--------------------------------------------------

author
#2 lore raw
In reply to [email protected]
Quoted 7 lines
Quoted 4 lines

I've picked through the list of patches in my backlog, and applied this bunch to my release tree ... before I ask Linus to pull, it would be useful to get some SGI testing (since the majority of the changes are SGI related).

Thanks to SGI we can now help with sn testing. I cloned the testing tree today, compiled and booted on Altix350 4way successful.

Darren,

Thanks for running the test. I notice that you said "I cloned the testing tree" ... which tree did you pull from? Despite my earlier assertion that I was going to filter things through:
rsync://rsync.kernel.org/pub/scm/linux/kernel/git/aegl/test-2.6.git
before putting them into the release tree:
rsync://rsync.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6.git
that huge bunch of backlog all went into the release tree.

-Tony

#3 lore raw
In reply to [email protected]

Tony,

It appears that you've missed the following two patches:

   date    subject
   03/22   [PATCH] move nodepda pointer out of pda
   03/23   [PATCH] convert some sn SAL_CALLs to ia64_sal_oemcall calls

When I apply them to your 'release' tree then everything builds cleanly. By everything, I'm including the 'SGI Altix cross partition functionality' patches you just applied. (They're the xp* files listed below.)

Note that to build these I had to set 'CONFIG_IA64_SGI_SN_XP=m' during the 'make oldconfig'. (I had copied 'arch/ia64/configs/sn2_defconfig' to .config.)

The system booted and ran just fine on an SGI Altix.

Thanks,
Dean

Quoted 48 lines · On Tue, May 03, 2005 at 02:08:58PM -0700, [email protected] wrote:

I've picked through the list of patches in my backlog, and applied this bunch to my release tree ... before I ask Linus to pull, it would be useful to get some SGI testing (since the majority of the changes are SGI related).

-Tony

 arch/ia64/Kconfig                   |   19 
 arch/ia64/hp/common/sba_iommu.c     |   34 
 arch/ia64/kernel/acpi.c             |   23 
 arch/ia64/kernel/entry.S            |    2 
 arch/ia64/kernel/fsys.S             |    4 
 arch/ia64/kernel/mca_drv.c          |    4 
 arch/ia64/kernel/mca_drv_asm.S      |   18 
 arch/ia64/kernel/process.c          |   41 
 arch/ia64/kernel/signal.c           |    3 
 arch/ia64/lib/flush.S               |    6 
 arch/ia64/lib/memcpy_mck.S          |    2 
 arch/ia64/lib/memset.S              |    2 
 arch/ia64/sn/kernel/Makefile        |    7 
 arch/ia64/sn/kernel/io_init.c       |   10 
 arch/ia64/sn/kernel/mca.c           |   34 
 arch/ia64/sn/kernel/setup.c         |   31 
 arch/ia64/sn/kernel/tiocx.c         |   60 
 arch/ia64/sn/kernel/xp_main.c       |  289 +++
 arch/ia64/sn/kernel/xp_nofault.S    |   31 
 arch/ia64/sn/kernel/xpc.h           |  991 +++++++++++
 arch/ia64/sn/kernel/xpc_channel.c   | 2297 ++++++++++++++++++++++++++
 arch/ia64/sn/kernel/xpc_main.c      | 1064 ++++++++++++
 arch/ia64/sn/kernel/xpc_partition.c |  984 +++++++++++
 arch/ia64/sn/kernel/xpnet.c         |  715 ++++++++
 arch/ia64/sn/pci/pcibr/pcibr_dma.c  |    2 
 arch/ia64/sn/pci/tioca_provider.c   |    2 
 drivers/char/Kconfig                |    2 
 include/asm-ia64/sn/addrs.h         |    8 
 include/asm-ia64/sn/arch.h          |   17 
 include/asm-ia64/sn/fetchop.h       |   85 
 include/asm-ia64/sn/l1.h            |    3 
 include/asm-ia64/sn/pda.h           |    1 
 include/asm-ia64/sn/shub_mmr.h      |   24 
 include/asm-ia64/sn/shubio.h        | 3108 +++++++++++++++++-------------------
 include/asm-ia64/sn/sn_cpuid.h      |    7 
 include/asm-ia64/sn/sn_fru.h        |   44 
 include/asm-ia64/sn/sndrv.h         |   47 
 include/asm-ia64/sn/xp.h            |  436 +++++
 kernel/exit.c                       |    2 
 mm/page_alloc.c                     |    2 
 40 files changed, 8525 insertions(+), 1936 deletions(-)
author
#4 lore raw
In reply to [email protected]
Quoted 5 lines

It appears that you've missed the following two patches:

  date    subject
  03/22   [PATCH] move nodepda pointer out of pda
  03/23   [PATCH] convert some sn SAL_CALLs to ia64_sal_oemcall calls

Ok, I've applied these locally. I can't connect to kernel.org at the moment (I think the machine I need is being upgraded at the moment). I'll push them up as soon as I can.

Quoted 7 lines

When I apply them to your 'release' tree then everything builds cleanly. By everything, I'm including the 'SGI Altix cross partition functionality' patches you just applied. (They're the xp* files listed below.)

Note that to build these I had to set 'CONFIG_IA64_SGI_SN_XP=m' during the 'make oldconfig'. (I had copied 'arch/ia64/configs/sn2_defconfig' to .config.)

Do you want to send me an updated sn2_defconfig?

Quoted 1 line

The system booted and ran just fine on an SGI Altix.

Great. Thanks for testing.

-Tony

author
#5 lore raw
In reply to [email protected]
Quoted 3 lines

Ok, I've applied these locally. I can't connect to kernel.org at the moment (I think the machine I need is being upgraded at the moment). I'll push them up as soon as I can.

Ok. These are now up on kernel.org.

-Tony

#6 lore raw
In reply to [email protected]
Quoted 5 lines · Tony Luck

I've picked through the list of patches in my backlog, and applied this bunch to my release tree ... before I ask Linus to pull, it would be useful to get some SGI testing (since the majority of the changes are SGI related).

The MCA recovery code runs on Altix with no noticeable regressions. It was tested both integrated in the kernel and as a module.

Signature · 2 lines
Russ Anderson, OS RAS/Partitioning Project Lead  
SGI - Silicon Graphics Inc          [email protected]
#7 lore raw
In reply to [email protected]

Hi Luck,

Quoted 10 lines · Tony
Quoted 7 lines
Quoted 4 lines

I've picked through the list of patches in my backlog, and applied this bunch to my release tree ... before I ask Linus to pull, it would be useful to get some SGI testing (since the majority of the changes are SGI related).

Thanks to SGI we can now help with sn testing. I cloned the testing tree today, compiled and booted on Altix350 4way successful.

Darren,

Quoted 7 lines

Thanks for running the test. I notice that you said "I cloned the testing tree" ... which tree did you pull from? Despite my earlier assertion that I was going to filter things through:
rsync://rsync.kernel.org/pub/scm/linux/kernel/git/aegl/test-2.6.git
before putting them into the release tree:
rsync://rsync.kernel.org/pub/scm/linux/kernel/git/aegl/linux-2.6.git
that huge bunch of backlog all went into the release tree.

I cloned from rsync://rsync.kernel.org/pub/scm/linux/kernel/git/aegl/test-2.6.git

Just tested the relaese tree now:
Build OK, boot OK

Quoted 2 lines

-Tony

- dsw

--------------------------------------------------
Darren Williams <dsw AT gelato.unsw.edu.au>
Gelato@UNSW <www.gelato.unsw.edu.au>
--------------------------------------------------

#8 lore raw
In reply to [email protected]
Quoted 5 lines · Tony
Quoted 3 lines

Ok, I've applied these locally. I can't connect to kernel.org at the moment (I think the machine I need is being upgraded at the moment). I'll push them up as soon as I can.

Ok. These are now up on kernel.org.

I again cg-clone'd your release tree and now that the two previously missing patches are in it, everything (with 'CONFIG_IA64_SGI_SN_XP=m' configured) builds cleanly, boots and runs on a SGI Altix.

Thanks.