- 16 Apr, 2005 40 commits
-
-
Olof Johansson authored
For prefetches of NULL (as when walking a short linked list), PPC64 will in some cases take a performance hit. The hardware needs to do the TLB walk, and said walk will always miss, which means (up to) two L2 misses as penalty. This seems to hurt overall performance, so for NULL pointers skip the prefetch alltogether. Signed-off-by:
Olof Johansson <olof@austin.ibm.com> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
Anton Blanchard authored
During some code inspection using gcc 4.0 I noticed a stack frame was being created for a number of functions that didnt require it. For example: c0000000000df944 <._spin_unlock>: c0000000000df944: fb e1 ff f0 std r31,-16(r1) c0000000000df948: f8 21 ff c1 stdu r1,-64(r1) c0000000000df94c: 7c 3f 0b 78 mr r31,r1 c0000000000df950: 7c 20 04 ac lwsync c0000000000df954: e8 21 00 00 ld r1,0(r1) c0000000000df958: 38 00 00 00 li r0,0 c0000000000df95c: 90 03 00 00 stw r0,0(r3) c0000000000df960: eb e1 ff f0 ld r31,-16(r1) c0000000000df964: 4e 80 00 20 blr It turns out we are adding -fno-omit-frame-pointer to ppc64 which is causing the above behaviour. Removing that flag results in much better code: c0000000000d5b30 <._spin_unlock>: c0000000000d5b30: 7c 20 04 ac lwsync c0000000000d5b34: 38 00 00 00 li r0,0 c0000000000d5b38: 90 03 00 00 stw r0,0(r3) c0000000000d5b3c: 4e 80 00 20 blr We dont require a frame pointer to debug on ppc64, so remove it. Signed-off-by:
Anton Blanchard <anton@samba.org> Signed-off-by:
-
Benjamin Herrenschmidt authored
The code that parses the OF device tree contains an old bogus hack which was killed a long time ago on ppc32, but survived in ppc64. It was supposed to help with a problem on the f50 which is ... a 32 bits machine :) Additionally, that hack is causing problems, so let's just get rid of it. Signed-off-by:
Benjamin Herrenschmidt <benh@kernel.crashing.org> Signed-off-by:
-
Benjamin Herrenschmidt authored
This patch adds detection of the Altivec capability of the CPU via the firmware in addition to the cpu table. This allows newer CPUs that aren't in the table to still have working altivec support in the kernel. It also fixes a problem where if a CPU isn't recognized as having altivec features, and takes an altivec unavailable exception due to userland issuing altivec instructions, the kernel would happily enable it and context switch the registers ... but not all of them (it would basically forget vrsave). With this patch, the kernel will refuse to enable altivec when the feature isn't detected for the CPU (SIGILL). Signed-off-by:
-
Benjamin Herrenschmidt authored
This patch reworks the way the ppc64 is mapped in user memory by the kernel to make it more robust against possible collisions with executable segments. Instead of just whacking a VMA at 1Mb, I now use get_unmapped_area() with a hint, and I moved the mapping of the vDSO to after the mapping of the various ELF segments and of the interpreter, so that conflicts get caught properly (it still has to be before create_elf_tables since the later will fill the AT_SYSINFO_EHDR with the proper address). While I was at it, I also changed the 32 and 64 bits vDSO's to link at their "natural" address of 1Mb instead of 0. This is the address where they are normally mapped in absence of conflict. By doing so, it should be possible to properly prelink one it's been verified to work on glibc. Signed-off-by:
-
Paul Mackerras authored
In arch/ppc64/kernel/ppc_ksyms.c, we are still exporting flush_icache_range, but that has been changed to be an inline in include/asm-ppc64/cacheflush.h which calls __flush_icache_range (defined in arch/ppc64/kernel/misc.S). This patch changes the export to __flush_icache_range, thus allowing modules to use the inline flush_icache_range. Signed-off-by:
Paul Mackerras <paulus@samba.org> Signed-off-by:
-
Benjamin Herrenschmidt authored
This patch fixes ppc64 __ioremap() so that it stops adding implicitely _PAGE_GUARDED when the cache is not writeback, and instead, let the callers provide the flag they want here. This allows things like framebuffers to explicitely request a non-cacheable and non-guarded mapping which is more efficient for that type of memory without side effects. The patch also fixes all current callers to add _PAGE_GUARDED except btext, which is fine without it. Signed-off-by:
-
Benjamin Herrenschmidt authored
This patch hacks the current PowerMac Alsa driver to add some basic support of analog sound output to some desktop G5s. It has severe limitations though: - Only 44100Khz 16 bits - Only work on G5 models using a TAS3004 analog code, that is early single CPU desktops and all dual CPU desktops at this date, but none of the more recent ones like iMac G5. - It does analog only, no digital/SPDIF support at all, no native AC3 support Better support would require a complete rewrite of the driver (which I am working on, but don't hold your breath), to properly support the diversity of apple sound HW setup, including dual codecs, several i2s busses, all the new codecs used in the new machines, proper clock switching with digital, etc etc etc... This patch applies on top of the other PowerMac sound patches I posted in the past couple of days (new powerbook support and sleep fixes). Note: This is a FAQ entry for PowerMac sound support with TI codecs: They have a feature called "DRC" which is automatically enabled for the internal speaker (at least when auto mute control is enabled) which will cause your sound to fade out to nothing after half a second of playback if you don't set a proper "DRC Range" in the mixer. So if you have a problem like that, check alsamixer and raise your DRC Range to something reasonable. Note2: This patch will also add auto-mute of the speaker when line-out jack is used on some earlier desktop G4s (and on the G5) in addition to the headphone jack. If that behaviour isn't what you want, just disable auto-muting and use the manual mute controls in alsamixer. Signed-off-by:
-
Benjamin Herrenschmidt authored
This patch improves the behaviour of the "tumbler/snapper" driver used on newer PowerMacs during sleep. It properly set the HW mutes to shut down amplifiers and does an analog shutdown of the codec. That might improve power consumption during sleep on a number of machines. Signed-off-by:
-
Benjamin Herrenschmidt authored
This patch hacks the current Alsa snd-powermac driver to add support for recent machine models with the tas3004 chip, that is basically new laptop models. The Mac Mini is _NOT_ yet supported by this patch (soon soon ...). The G5s (iMac or Desktop) will need the rewritten sound driver on which I'm working on (I _might_ get a hack for analog only on some G5s on the current driver, but no promise). Signed-off-by:
-
Benjamin Herrenschmidt authored
This patch allows Kconfig to build the MV643xx ethernet driver on Pegasos (CONFIG_PPC_MULTIPLATFORM) and adds what I think is a missing fix from Dale's batch, that is remove SA_INTERRUPT and add SA_SHIRQ in there as the interrupt is shared if I understand things correctly. Signed-off-by:
Fabio Massimo Di Nitto <fabbione@ubuntu.com> Signed-off-by:
-
Benoit Boissinot authored
make defconfig give the following error on ppc (gcc-4): arch/ppc/syslib/open_pic.c:36: error: static declaration of ‘OpenPIC’ follows non-static declaration arch/ppc/syslib/open_pic_defs.h:175: error: previous declaration of ‘OpenPIC’ was here Signed-Off-By:
Benoit Boissinot <benoit.boissinot@ens-lyon.org> Signed-off-by:
-
Benoit Boissinot authored
make defconfig give the following error on ppc (gcc-4): arch/ppc/syslib/prom_init.c:120: error: static declaration of ‘prom_display_paths’ follows non-static declaration include/asm/prom.h:17: error: previous declaration of ‘prom_display_paths’ was here arch/ppc/syslib/prom_init.c:122: error: static declaration of ‘prom_num_displays’ follows non-static declaration include/asm/prom.h:18: error: previous declaration of ‘prom_num_displays’ was here Signed-Off-By:
-
Benoit Boissinot authored
make defconfig give the following error on ppc (gcc-4): arch/ppc/kernel/time.c:92: error: static declaration of ‘time_offset’ follows non-static declaration include/linux/timex.h:236: error: previous declaration of ‘time_offset’ was here The following patch solves it (time_offset is declared in timer.c). Signed-Off-By:
-
Benoit Boissinot authored
make defconfig give the following error on ppc (gcc-4): include/asm-m68k/setup.h:365: error: array type has incomplete element type Signed-Off-By:
-
Leigh Brown authored
This patch restores the original behaviour of prep_pcibios_fixup() to only call prep_pib_init() on machines with an openpic. This allows the Powerstack II Pro4000 to boot again. Signed-off-by:
Leigh Brown <leigh@solinno.co.uk> Signed-off-by:
-
Tom Rini authored
When building a ppc32 MULTIPLATFORM kernel for a 64bit pmac, we try and build certain files or use certain functions that make no sense in that context. This catches the last of these. Signed-off-by:
Tom Rini <trini@kernel.crashing.org> Signed-off-by:
-
Giovambattista Pulcini authored
The following problem was found by Giovambattista Pulcini <gpulcini@swintel.it>, who also provided a partial patch, and this has been verified by our time guru Gabriel Paubert <paubert@iram.es>. The problem is that in do_settimeofday() we always set time_state to TIME_ERROR and except on two platforms, never re-set it. This meant that ntp_gettime() and ntp_adjtime() always returned TIME_ERROR, incorrectly. Based on Gabriel's analysis, time_state is used for leap-second processing, and ppc shouldn't be mucking with it. Signed-off-by:
-
Tom Rini authored
The CONFIG_8xx_WDT option got broken in the generic hardirq update as ppc32 had its own different request_irq that worked when other arches used setup_irq. This is the trivial fix for the problem. From: Carsten Juttner <carjay@gmx.net> Signed-off-by:
-
Kumar Gala authored
To add support for 36-bit physical addressing on e500 the following changes have been made. The changes are generalized to support any physical address size larger than 32-bits: * Allow FSL Book-E parts to use a 64-bit PTE, it is 44-bits of pfn, 20-bits of flags. * Introduced new CPU feature (CPU_FTR_BIG_PHYS) to allow runtime handling of updating hardware register (SPRN_MAS7) which holds the upper 32-bits of physical address that will be written into the TLB. This is useful since not all e500 cores support 36-bit physical addressing. * Currently have a pass through implementation of fixup_bigphys_addr * Moved _PAGE_DIRTY in the 64-bit PTE case to free room for three additional storage attributes that may exist in future FSL Book-E cores and updated fault handler to copy these bits into the hardware TLBs. Signed-off-by:
Kumar Gala <kumar.gala@freescale.com> Signed-off-by:
-
Kumar Gala authored
Allow the pfn to be offset by more than just PAGE_SHIFT in the pte. Today, PAGE_SHIFT tends to allow us to have 12-bits of flags in the pte. In the future if we have a larger pte we can allocate more bits for flags by offsetting the pfn even further. Signed-off-by:
-
Kumar Gala authored
CONFIG_PTE_64BIT & CONFIG_PHYS_64BIT are not currently consistently used in the code base. Fixed up the usage such that CONFIG_PTE_64BIT is used when we have a 64-bit PTE regardless of physical address width. CONFIG_PHYS_64BIT is used if the physical address width is larger than 32-bits, regardless of PTE size. These changes required a few sub-arch specific ifdef's to be fixed and the introduction of a physical address format string. Signed-off-by:
-
Kumar Gala authored
While the existing pte_update code handled atomically modifying a 64-bit PTE, it did not return all 64-bits of the PTE before it was modified. This causes problems in some places that expect the full PTE to be returned, like ptep_get_and_clear(). Created a new pte_update function that is conditional on CONFIG_PTE_64BIT. It atomically reads the low PTE word which all PTE flags are required to be in and returns a premodified full 64-bit PTE. Since we now have an explicit 64-bit PTE version of pte_update we can also remove the hack that existed to get the low PTE word regardless of size. Signed-off-by:
-
Benjamin Herrenschmidt authored
My previous patch that added sleep support for uninorth-agp and some AGP "off" stuff in radeonfb and aty128fb is breaking some configs. More specifically, it has problems with rage128 setups since the DRI code for these in X doesn't properly re-enable AGP on wakeup or console switch (unlike the radeon DRM). This patch fixes the problem for pmac once for all by using a different approach. The AGP driver "registers" special suspend/resume callbacks with some arch code that the fbdev's can later on call to suspend and resume AGP, making sure it's resumed back in the same state it was when suspended. This is platform specific for now. It would be too complicated to try to do a generic implementation of this at this point due to all sort of weird things going on with AGP on other architectures. We'll re-work that whole problem cleanly once we finally merge fbdev's and DRI. In the meantime, please apply this patch which brings back some r128 based laptops into working condition as far as system sleep is concerned. Signed-off-by:
-
Benjamin Herrenschmidt authored
This patch updates the PowerMac cpufreq driver. It depends on the addition of the suspend() method (my previous patch) and on the new flag I defined to silence some warnings that are normal for us. It fixes various issues related to cpufreq on pmac, including some crashes on some models when sleeping the machine while in low speed, proper voltage control on some newer machines, and adds voltage control on 750FX based G3 laptops. Signed-off-by:
-
Paul Mackerras authored
On ppc, we emulate instructions that cause alignment exceptions. If we are single-stepping an instruction and it causes an alignment exception, we will currently do the next instruction as well before taking the single-step exception. This patch fixes that, so we take the single-step exception after emulating the instruction. Signed-off-by:
-
Paul Mackerras authored
If we should happen to get an altivec assist exception while executing in the kernel, we will currently try to handle it and fail, and end up oopsing with (apparently) a segfault. (An altivec assist exception occurs for floating-point altivec instructions with denormalized inputs or outputs if the altivec unit is in java mode.) This patch checks explicitly if we are in user mode and prints a useful message if not. Signed-off-by:
-
Paul Mackerras authored
Currently the procedure in the ppc32 kernel that synchronizes the timebase registers across an SMP powermac system does so by setting both timebases to zero. That is OK at boot but causes problems if done later. So that we can do hotplug CPU on these machines, this patch changes the code so it reads the timebase from one CPU and transfers the value to the other CPU. (Hotplug CPU is needed for sleep (aka suspend to RAM) to work.) Signed-off-by:
-
Eugene Surovegin authored
This patch adds interrupt acknowledge to the PPC4xx PIC enable_irq implementation for level-sensitive IRQ sources. This helps in cases when enable/disable_irq is used in interrupt handlers for hardware, which requires IRQ acknowledge to be issued from non-interrupt context (e.g. when actual ACK in device needs an I2C transaction). For such strange hardware, interrupt handler disables IRQ and defers actual ACK to some other context. When this happens, IRQ is enabled again. For level-sensitive sources we get spurious triggering right after IRQ is enabled. This patch fixes this. Suggested by Tolunay Orkun <listmember@orkun.us>. Signed-off-by:
Eugene Surovegin <ebs@ebshome.net> Signed-off-by:
-
Paul Mackerras authored
The code that went into arch/ppc/kernel/signal.c recently to handle process freezing seems to contain a dubious assumption: that a process that calls do_signal when PF_FREEZE is set will have entered the kernel because of a system call. This patch removes that assumption. Signed-off-by:
-
James Morris authored
This patch adds SELinux support for the KOBJECT_UEVENT Netlink family, so that SELinux can apply finer grained controls to it. For example, security policy for hald can be locked down to the KOBJECT_UEVENT Netlink family only. Currently, this family simply defaults to the default Netlink socket class. Note that some new permission definitions are added to sync with changes in the core userspace policy package, which auto-generates header files. Signed-off-by:
James Morris <jmorris@redhat.com> Signed-off-by:
Stephen Smalley <sds@tycho.nsa.gov> Signed-off-by:
-
Jean Tourrilhes authored
Acked-by:
"David S. Miller" <davem@davemloft.net> Signed-off-by:
-
Herbert Xu authored
When we are not the real parent of the dst (e.g., when we're xfrm_dst and the child is an rtentry), it may already be on the GC list. In fact the current code is buggy to, we need to check dst->flags before the dec as dst may no longer be valid afterwards. Signed-off-by:
Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by:
-
Arnaldo Carvalho de Melo authored
This fixes the warning reported by Marcel Holtmann (Thanks!). Signed-off-by:
Arnaldo Carvalho de Melo <acme@conectiva.com.br> Signed-off-by:
-
David S. Miller authored
This recently got changed to include a lot of kernel internal stuff in the non-__KERNEL__ area of the header, which isn't so kosher and breaks libc builds. The fix is pretty simple. Signed-off-by:
-
Martin Hicks authored
Working on some code lately I've been getting huge values for "Cached". The cause is that get_page_cache_size() is an approximate value, and for a sufficiently small returned value of get_page_cache_size() the value underflows. Signed-off-by:
Martin Hicks <mort@sgi.com> Signed-off-by:
-
akpm@osdl.org authored
Signed-off-by:
-
akpm@osdl.org authored
) We only call pageout() for dirty pages, so this test is redundant. Signed-off-by:
-
Andrea Arcangeli authored
iscsi/lvm2/multipath needs guaranteed protection from the oom-killer, so make the magical value of -17 in /proc/<pid>/oom_adj defeat the oom-killer altogether. (akpm: we still need to document oom_adj and friends in Documentation/filesystems/proc.txt!) Signed-off-by:
Andrea Arcangeli <andrea@suse.de> Signed-off-by:
-
Jeff Moyer authored
We will return NULL from filemap_getpage when a page does not exist in the page cache and MAP_NONBLOCK is specified, here: page = find_get_page(mapping, pgoff); if (!page) { if (nonblock) return NULL; goto no_cached_page; } But we forget to do so when the page in the cache is not uptodate. The following could result in a blocking call: /* * Ok, found a page in the page cache, now we need to check * that it's up-to-date. */ if (!PageUptodate(page)) goto page_not_uptodate; Signed-off-by:
-
