From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Helge Deller <deller@gmx.de>,
John David Anglin <dave.anglin@bell.net>
Subject: [PATCH 3.10 34/74] parisc: Do not crash 64bit SMP kernels on machines with >= 4GB RAM
Date: Fri, 8 Nov 2013 22:51:39 -0800 [thread overview]
Message-ID: <20131109065115.913429253@linuxfoundation.org> (raw)
In-Reply-To: <20131109065113.502217951@linuxfoundation.org>
3.10-stable review patch. If anyone has any objections, please let me know.
------------------
From: Helge Deller <deller@gmx.de>
commit 54e181e073fc1415e41917d725ebdbd7de956455 upstream.
Since the beginning of the parisc-linux port, sometimes 64bit SMP kernels were
not able to bring up other CPUs than the monarch CPU and instead crashed the
kernel. The reason was unclear, esp. since it involved various machines (e.g.
J5600, J6750 and SuperDome). Testing showed, that those crashes didn't happened
when less than 4GB were installed, or if a 32bit Linux kernel was booted.
In the end, the fix for those SMP problems is trivial:
During the early phase of the initialization of the CPUs, including the monarch
CPU, the PDC_PSW firmware function to enable WIDE (=64bit) mode is called.
It's documented that this firmware function may clobber various registers, and
one one of those possibly clobbered registers is %cr30 which holds the task
thread info pointer.
Now, if %cr30 would always have been clobbered, then this bug would have been
detected much earlier. But lots of testing finally showed, that - at least for
%cr30 - on some machines only the upper 32bits of the 64bit register suddenly
turned zero after the firmware call.
So, after finding the root cause, the explanation for the various crashes
became clear:
- On 32bit SMP Linux kernels all upper 32bit were zero, so we didn't faced this
problem.
- Monarch CPUs in 64bit mode always booted sucessfully, because the inital task
thread info pointer was below 4GB.
- Secondary CPUs booted sucessfully on machines with less than 4GB RAM because
the upper 32bit were zero anyay.
- Secondary CPus failed to boot if we had more than 4GB RAM and the task thread
info pointer was located above the 4GB boundary.
Finally, the patch to fix this problem is trivial by saving the %cr30 register
before the firmware call and restoring it afterwards.
Signed-off-by: Helge Deller <deller@gmx.de>
Signed-off-by: John David Anglin <dave.anglin@bell.net>
Signed-off-by: Helge Deller <deller@gmx.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
arch/parisc/kernel/head.S | 4 ++++
1 file changed, 4 insertions(+)
--- a/arch/parisc/kernel/head.S
+++ b/arch/parisc/kernel/head.S
@@ -195,6 +195,8 @@ common_stext:
ldw MEM_PDC_HI(%r0),%r6
depd %r6, 31, 32, %r3 /* move to upper word */
+ mfctl %cr30,%r6 /* PCX-W2 firmware bug */
+
ldo PDC_PSW(%r0),%arg0 /* 21 */
ldo PDC_PSW_SET_DEFAULTS(%r0),%arg1 /* 2 */
ldo PDC_PSW_WIDE_BIT(%r0),%arg2 /* 2 */
@@ -203,6 +205,8 @@ common_stext:
copy %r0,%arg3
stext_pdc_ret:
+ mtctl %r6,%cr30 /* restore task thread info */
+
/* restore rfi target address*/
ldd TI_TASK-THREAD_SZ_ALGN(%sp), %r10
tophys_r1 %r10
next prev parent reply other threads:[~2013-11-09 6:51 UTC|newest]
Thread overview: 82+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-11-09 6:51 [PATCH 3.10 00/74] 3.10.19-stable review Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 01/74] usb-storage: add quirk for mandatory READ_CAPACITY_16 Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 02/74] USB: support new huawei devices in option.c Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 03/74] USB: quirks.c: add one device that cannot deal with suspension Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 04/74] USB: quirks: add touchscreen that is dazzeled by remote wakeup Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 05/74] USB: serial: ftdi_sio: add id for Z3X Box device Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 06/74] x86: Update UV3 hub revision ID Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 07/74] cpufreq / intel_pstate: Fix max_perf_pct on resume Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 08/74] bcache: Fixed incorrect order of arguments to bio_alloc_bioset() Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 09/74] cgroup: fix to break the while loop in cgroup_attach_task() correctly Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 10/74] mac80211: correctly close cancelled scans Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 11/74] mac80211: drop spoofed packets in ad-hoc mode Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 12/74] mac80211: use sta_info_get_bss() for nl80211 tx and client probing Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 13/74] mac80211: update sta->last_rx on acked tx frames Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 14/74] mac80211: fix crash if bitrate calculation goes wrong Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 15/74] ath9k: fix tx queue scheduling after channel changes Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 16/74] cfg80211: fix warning when using WEXT for IBSS Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 17/74] mwifiex: fix SDIO interrupt lost issue Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 18/74] rtlwifi: rtl8192cu: Fix error in pointer arithmetic Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 19/74] iwlwifi: pcie: add SKUs for 6000, 6005 and 6235 series Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 20/74] jfs: fix error path in ialloc Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 21/74] can: at91-can: fix device to driver data mapping for platform devices Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 22/74] can: flexcan: fix mx28 detection by rearanging OF match table Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 23/74] can: flexcan: flexcan_chip_start: fix regression, mark one MB for TX and abort pending TX Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 24/74] SCSI: sd: call blk_pm_runtime_init before add_disk Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 25/74] ecryptfs: Fix memory leakage in keystore.c Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 26/74] raid5: set bio bi_vcnt 0 for discard request Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 27/74] raid5: avoid finding "discard" stripe Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 28/74] libata: make ata_eh_qc_retry() bump scmd->allowed on bogus failures Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 29/74] md: avoid deadlock when md_set_badblocks Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 30/74] md: Fix skipping recovery for read-only arrays Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 31/74] target/pscsi: fix return value check Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 32/74] vhost/scsi: Fix incorrect usage of get_user_pages_fast write parameter Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 33/74] clockevents: Sanitize ticks to nsec conversion Greg Kroah-Hartman
2013-11-09 6:51 ` Greg Kroah-Hartman [this message]
2013-11-09 6:51 ` [PATCH 3.10 35/74] scripts/kallsyms: filter symbols not in kernel address space Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 36/74] ARC: Incorrect mm reference used in vmalloc fault handler Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 37/74] ALSA: hda - Add missing initial vmaster hook at build_controls callback Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 38/74] ALSA: hda - Fix unbalanced runtime PM refcount after S3/S4 Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 39/74] ALSA: hda - Add a fixup for ASUS N76VZ Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 40/74] ALSA: fix oops in snd_pcm_info() caused by ASoC DPCM Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 41/74] ASoC: wm_hubs: Add missing break in hp_supply_event() Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 42/74] ASoC: dapm: Fix source list debugfs outputs Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 43/74] staging: ozwpan: prevent overflow in oz_cdev_write() Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 44/74] Staging: bcm: info leak in ioctl Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 45/74] Staging: sb105x: info leak in mp_get_count() Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 46/74] staging: wlags49_h2: buffer overflow setting station name Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 47/74] uml: check length in exitcode_proc_write() Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 48/74] xtensa: dont use alternate signal stack on threads Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 49/74] mm: make generic_access_phys available for modules Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 50/74] uio: provide vm access to UIO_MEM_PHYS maps Greg Kroah-Hartman
2013-11-09 14:22 ` Uwe Kleine-König
2013-11-09 16:10 ` Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 51/74] au1100fb: VM_IO is set by io_remap_pfn_range() Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 52/74] au1200fb: io_remap_pfn_range() sets VM_IO Greg Kroah-Hartman
2013-11-09 6:51 ` [PATCH 3.10 54/74] lib/scatterlist.c: dont flush_kernel_dcache_page on slab page Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 55/74] aacraid: missing capable() check in compat ioctl Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 56/74] clk: fixup argument order when setting VCO parameters Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 57/74] mm: numa: Do not account for a hinting fault if we raced Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 58/74] mm: Wait for THP migrations to complete during NUMA hinting faults Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 59/74] mm: Prevent parallel splits during THP migration Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 60/74] mm: numa: Sanitize task_numa_fault() callsites Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 61/74] mm: Close races between THP migration and PMD numa clearing Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 62/74] mm: Account for a THP NUMA hinting update as one PTE update Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 63/74] mm/pagewalk.c: fix walk_page_range() access of wrong PTEs Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 64/74] mm/vmalloc.c: fix an overflow bug in alloc_vmap_area() Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 65/74] drm/vmwgfx: Dont put resources with invalid ids on lru list Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 66/74] drm/vmwgfx: Dont kill clients on VT switch Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 67/74] drm: Prevent overwriting from userspace underallocating core ioctl structs Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 68/74] drm: Pad drm_mode_get_connector to 64-bit boundary Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 69/74] drm/radeon/atom: workaround vbios bug in transmitter table on rs780 Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 70/74] seq_file: always update file->f_pos in seq_lseek() Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 71/74] NTB: Add Error Handling in ntb_device_setup Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 72/74] NTB: Correct Number of Scratch Pad Registers Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 73/74] NTB: Correct USD/DSD Identification Greg Kroah-Hartman
2013-11-09 6:52 ` [PATCH 3.10 74/74] NTB: Correct debugfs to work with more than 1 NTB Device Greg Kroah-Hartman
2013-11-09 17:01 ` [PATCH 3.10 00/74] 3.10.19-stable review Guenter Roeck
2013-11-09 17:12 ` Greg Kroah-Hartman
2013-11-10 11:51 ` Satoru Takeuchi
2013-11-10 15:15 ` Greg Kroah-Hartman
2013-11-11 17:56 ` Shuah Khan
2013-11-11 22:51 ` Greg Kroah-Hartman
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20131109065115.913429253@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=dave.anglin@bell.net \
--cc=deller@gmx.de \
--cc=linux-kernel@vger.kernel.org \
--cc=stable@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).