OpenBIOS All Updates - OpenBIOS is a GPL'd Open Firmware implementation.

AgeSort asc/desc	Change
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Next
May 12, 2012
4 days 20 hours	Sparc32: avoid problems with SMP table Check CPU ID if we're running on boot CPU. If so, don't use SMP table information since it may contain uninitialized or garbage values. Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Commit 1057, by blueswirl
Apr 28, 2012
18 days 18 hours	SPARC32: Swap Forth machine memory allocation over to OFMEM memory pool. Instead of using our internal memory pool to run the Forth machine, allocate the memory using OFMEM. This enables us to dramatically increase the memory available to Forth clients whilst reducing the memory requirement for the resulting image. Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Commit 1056, by mcayland
18 days 18 hours	SPARC64: Swap Forth machine memory allocation over to OFMEM memory pool. Instead of using our internal memory pool to run the Forth machine, allocate the memory using OFMEM. This enables us to dramatically increase the memory available to Forth clients whilst substantially reducing the memory requirement for the resulting image. Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Commit 1055, by mcayland
18 days 18 hours	SPARC64: Remove majority of the video initialisation hack. Now that we have OFMEM, as long as we have a PCI bus then we can configure the address of the framebuffer automatically without having an architecture-specific hack. The only thing we need to do is override the virtual address from the OFMEM phys == virt default. Similarly we can remove all reference to the video memory variables _vmem and _evmem since the memory is allocated outside of the OpenBIOS image. This commit also fixes another couple of issues: switch video.c to use virtual instead of physical addresses for architectures that require it (the default is the existing behaviour where phys_addr == virt_addr) and also correct the framebuffer size calculation which was also adding the framebuffer start address to size of the framebuffer to be mapped. Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Commit 1054, by mcayland
18 days 18 hours	Remove architecture-specific routines from ofmem.h. This cleans up the OFMEM interface by allowing us to keep all of the architecture-specific code in separate header files; in particular ofmem_sparc32.h and ofmem_sparc64.h. PPC doesn't reference the variables from ofmem.h outside of ofmem.c, so simply redefine them as static variables. Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Commit 1053, by mcayland
18 days 18 hours	SPARC64: Move ofmem_arch_map_pages()/ofmem_arch_unmap_pages() into ofmem_sparc64.c with all the other architecture-specific code. Note that we also take some of the ITLB/DTLB helpers with us. Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Commit 1052, by mcayland
18 days 18 hours	SPARC64: Refactor tte-data code in preparation for moving architecture-specific code to ofmem_sparc64.c. Note we also change pgmap@ so it explicitly searches for a TTE entry using architecture-specific code, plus add the code to detect whether a page is marked as locked or not into the main page mapping function. Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Commit 1051, by mcayland
18 days 18 hours	SPARC32: Move ofmem_arch_map_pages() into ofmem_sparc32.c with all the other architecture-specific code. Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Commit 1050, by mcayland
18 days 18 hours	Rename ofmem_arch_early_map_pages() to ofmem_arch_map_pages(). This OFMEM call does actually map the page for the architectures that require it, so let's change its name so that we now have a symmetrical ofmem_arch_map_pages() and ofmem_arch_unmap_pages() API. Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Commit 1049, by mcayland
Mar 17, 2012
1 month 29 days	pci: fix BAR setup A change in QEMU on how PCI bridges are setup revealed a bug in OpenBIOS PCI setup. On Sparc64, the BARs just happened to get somewhat correct values by accident before the commit but not after the change. Don't use arch->io_base for PCI I/O port BARs. Fix Sparc64 PCI memory base. Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Commit 1048, by blueswirl
Aug 20, 2011
8 months 26 days	ESP: clear Unit Attention condition after reset In recent QEMU, SCSI devices are in Unit Attention condition after reset and this made Read Capacity command fail. Fix by sending a Test Unit Ready command to the devices first. Also avoid division by zero if block size is 0 when Read Capacity fails. Thanks to Paolo Bonzini for analysis. Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Commit 1047, by blueswirl
Aug 8, 2011
9 months 8 days	Make select-dev and unselect-dev compatible with OBP. From the 1275 errata for section H.8, select-dev and unselect-dev are pre Open Firmware, but not the same as open-dev/device-end. Make them Sun OBP compatible by pulling the functionality for select-dev/ unselect-dev out of (and simplifying) begin-package/end-package. For the Sun OBP definition of begin-package/end-package, along with descriptions and example usage, see the "Debugging and Testing FCode Programs" chapter of "Writing FCode 3.x Programs". Signed-off-by: Bob Breuer <breuerr@mc.net> Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@siriusit.co.uk>
Commit 1046, by mcayland
Jul 14, 2011
10 months 2 days	Fix Fcode table initialisation on SPARC32. Increase the Forth machine memory to 128K (also requiring a corresponding increase in OFMEM to 384K) to allow enough space for the Fcode tables to be initialised, and hence allow Fcode to be executed under SPARC32. Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@siriusit.co.uk> Acked-by: Bob Breuer <breuerr@mc.net>
Commit 1045, by mcayland
Jun 17, 2011
11 months 7 hours	ppc: only clear MSR_SF when clearing high MSR bits We are pretty aggressive now with how we clear the high MSR bits, clearing all of them when we find a 64-bit CPU. As Segher points out, this is not always a good idea though. There might be bits set that can be crucial for operation. Hence we should only clear MSR_SF, as that's the bit we're interested in. Reported-by: Segher Boessenkool <segher@kernel.crashing.org> Signed-off-by: Alexander Graf <agraf@suse.de>
Commit 1044, by agraf
11 months 7 hours	ppc: fix mtmsr for SF setting When running openbios-ppc32 on a ppc64 VM, we need to unset SF on every interrupt to ensure that things still work. So far we've been using mtmsr for this operation, but according to the spec mtmsr doesn't set any bit above 32bit, so we were merely exploiting a bug in qemu before. This patch adds a runtime check on MSR_SF to see if we're running on a 64-bit capable CPU. If so, we use mtmsrd, which can set the high 32bits of MSR. CC: Andreas Färber <andreas.faerber@web.de> CC: Segher Boessenkool <segher@kernel.crashing.org> Signed-off-by: Alexander Graf <agraf@suse.de> --- v1 -> v2: - use segher's awesome MSR_SF detection
Commit 1043, by agraf
11 months 7 hours	ppc: add POWER4 cpu description So far, we only supported the 970fx CPU type inside guests. Unfortunately, that CPU supports Altivec. Not all KVM capable host CPUs do have altivec though, so they're left out in the rain when running openbios in the guest. This patch adds POWER4 CPU description, allowing users to specify at least one CPU type that does not require Altivec, but is still 64-bit capable. Signed-off-by: Alexander Graf <agraf@suse.de>
Commit 1042, by agraf
May 23, 2011
11 months 24 days	ppc64: Don't set Kp bit on SLB Since QEMU 81762d6dd0d430d87024f2c83e9c4dcc4329fb7d (Clean up PowerPC SLB handling code) we never got to the ppc64 OpenBIOS banner. According to Alex' debugging this is due to the Kp bit being set. The code was supposed to be a 1:1 translation of the old mtsrin code, which did not set Kp bit. So don't set Kp bit with slbmte. Introduce a define for the shift, suggested by Alex. Signed-off-by: Andreas Färber <andreas.faerber@web.de> Acked-by: Alexander Graf <agraf@suse.de>
Commit 1041, by afaerber
May 22, 2011
11 months 24 days	switch-arch: remove non-standard "local" keyword Remove the non-standard "local" keyword from within the crosscflags() function. This was not necessary since the variables "host" and "target" were not clobbered anywhere outside of the function, but the real motivation is to support shells such as ksh93 that do not have the "local" builtin like bash. Signed-off-by: Kenneth Salerno <kennethsalerno@yahoo.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Commit 1040, by blueswirl
May 8, 2011
1 year 8 days	Fix the Forth boot word if the current package is not /chosen under SPARC32. Ensure that we deselect the current package before defining the platform-specific boot words so that they are created in the global dictionary and not the current instance dictionary. Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@siriusit.co.uk>
Commit 1039, by mcayland
May 7, 2011
1 year 10 days	Disable mixing of DMA and FIFO messages in the Solaris ESP kernel module. QEMU's ESP emulation currently does not support the mixing of DMA and FIFO messages. Setting the scsi-options property prevents the Solaris ESP kernel module from trying to use this feature by default, which otherwise would cause boot to fail. Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@siriusit.co.uk>
Commit 1038, by mcayland