December 2012 - SeaBIOS - mail.coreboot.org

[Resend][Seabios PATCH] don't boot from un-selected devices
by Amos Kong Jan. 12, 2013

Jan. 12, 2013

Current seabios will try to boot from selected devices first, if they are all failed, seabios will also try to boot from un-selected devices. For example: @ qemu-kvm -boot order=n,menu=on ... Guest will boot from network first, if it's failed, guest will try to boot from other un-selected devices (floppy, cdrom, disk) one by one. Sometimes, user don't want to boot from some devices. This patch changes seabios to boot only from selected devices. If user choose first boot device from menu, then seabios will try all the devices, even some of them are not selected. Signed-off-by: Amos Kong <akong(a)redhat.com> --- Resend for CCing seabios maillist. --- src/boot.c | 13 ++++++++----- 1 files changed, 8 insertions(+), 5 deletions(-) diff --git a/src/boot.c b/src/boot.c index 3ca7960..ee810ac 100644 --- a/src/boot.c +++ b/src/boot.c @@ -424,6 +424,10 @@ interactive_bootmenu(void) maxmenu++; printf("%d. %s\n", maxmenu , strtcpy(desc, pos->description, ARRAY_SIZE(desc))); + /* If user chooses first boot device from menu, we will treat + all the devices as selected. */ + if (pos->priority == DEFAULT_PRIO) + pos->priority = DEFAULT_PRIO - 1; pos = pos->next; } @@ -490,7 +494,10 @@ boot_prep(void) // Map drives and populate BEV list struct bootentry_s *pos = BootList; - while (pos) { + + /* The priority of un-selected device is not changed, + we only boot from user selected devices. */ + while (pos && pos->priority != DEFAULT_PRIO) { switch (pos->type) { case IPL_TYPE_BCV: call_bcv(pos->vector.seg, pos->vector.offset); @@ -513,10 +520,6 @@ boot_prep(void) } pos = pos->next; } - - // If nothing added a floppy/hd boot - add it manually. - add_bev(IPL_TYPE_FLOPPY, 0); - add_bev(IPL_TYPE_HARDDISK, 0); } -- 1.7.1

5 14

Get rid of a compile time dprintf warning
by Dave Frodin Jan. 10, 2013

Jan. 10, 2013

This patch gets rid of a compile time dprintf warning. A 'long int' was being passed to a '%d'. Signed-off-by: Dave Frodin <dave.frodin(a)se-eng.com> --- src/post.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/post.c b/src/post.c index 924b311..b84d19e 100644 --- a/src/post.c +++ b/src/post.c @@ -345,7 +345,7 @@ reloc_init(void) panic("No space for init relocation.\n"); // Copy code and update relocs (init absolute, init relative, and runtime) - dprintf(1, "Relocating low data from %p to %p (size %d)\n" + dprintf(1, "Relocating low data from %p to %p (size %ld)\n" , datalow_start, final_datalow_start, datalow_end - datalow_start); updateRelocs(code32flat_start, _reloc_datalow_start, _reloc_datalow_end , final_datalow_start - datalow_start); -- 1.7.9

3 6

[PATCH] esp: enable busmaster
by Paolo Bonzini Jan. 9, 2013

Jan. 9, 2013

The HBA does DMA, thus we must enable the busmaster bit, otherwise dma access will fail with recent qemu versions. Signed-off-by: Paolo Bonzini <pbonzini(a)redhat.com> --- src/esp-scsi.c | 2 ++ 1 file modificato, 2 inserzioni(+) diff --git a/src/esp-scsi.c b/src/esp-scsi.c index b795012..c43e55b 100644 --- a/src/esp-scsi.c +++ b/src/esp-scsi.c @@ -202,6 +202,8 @@ init_esp_scsi(struct pci_device *pci) pci_bdf_to_bus(bdf), pci_bdf_to_dev(bdf), pci_bdf_to_fn(bdf), iobase); + pci_config_maskw(bdf, PCI_COMMAND, 0, PCI_COMMAND_MASTER); + // reset outb(ESP_CMD_RESET, iobase + ESP_CMD); -- 1.7.12.1

4 10

[PATCH] enable_vga_console(): print machine UUID under seabios version message
by Laszlo Ersek Dec. 21, 2012

Dec. 21, 2012

There are users who would like to see the UUID at startup, and it probably won't bother others. Related RHBZ: 876250. Signed-off-by: Laszlo Ersek <lersek(a)redhat.com> --- src/util.h | 1 + src/bootsplash.c | 15 ++++++++++++++- src/output.c | 15 +++++++++++++++ 3 files changed, 30 insertions(+), 1 deletions(-) diff --git a/src/util.h b/src/util.h index 7723bb1..aafa9b0 100644 --- a/src/util.h +++ b/src/util.h @@ -227,6 +227,7 @@ void __set_code_invalid(struct bregs *regs, u32 linecode, const char *fname); void __set_code_unimplemented(struct bregs *regs, u32 linecode , const char *fname); void hexdump(const void *d, int len); +void format_uuid(char buf[37], const u8 uuid[16]); #define dprintf(lvl, fmt, args...) do { \ if (CONFIG_DEBUG_LEVEL && (lvl) <= CONFIG_DEBUG_LEVEL) \ diff --git a/src/bootsplash.c b/src/bootsplash.c index 78023a5..622d86e 100644 --- a/src/bootsplash.c +++ b/src/bootsplash.c @@ -12,6 +12,7 @@ #include "jpeg.h" // splash #include "vbe.h" // struct vbe_info #include "bmp.h" // bmp_alloc +#include "paravirt.h" // qemu_cfg_get_uuid /**************************************************************** @@ -33,6 +34,18 @@ call16_int10(struct bregs *br) * VGA text / graphics console ****************************************************************/ +static void +print_hello(void) +{ + u8 uuid[16] = { 0 }; + char uuid_str[37]; + + qemu_cfg_get_uuid(uuid); + format_uuid(uuid_str, uuid); + + printf("SeaBIOS (version %s)\nMachine UUID %s\n\n", VERSION, uuid_str); +} + void enable_vga_console(void) { @@ -45,7 +58,7 @@ enable_vga_console(void) call16_int10(&br); // Write to screen. - printf("SeaBIOS (version %s)\n\n", VERSION); + print_hello(); } static int diff --git a/src/output.c b/src/output.c index 83de7f4..f4cdbc0 100644 --- a/src/output.c +++ b/src/output.c @@ -474,6 +474,21 @@ hexdump(const void *d, int len) debug_serial_flush(); } +void format_uuid(char buf[37], const u8 uuid[16]) +{ + snprintf(buf, 37 + , "%02x%02x%02x%02x" + "-%02x%02x" + "-%02x%02x" + "-%02x%02x" + "-%02x%02x%02x%02x%02x%02x" + , uuid[ 0], uuid[ 1], uuid[ 2], uuid[ 3] + , uuid[ 4], uuid[ 5] + , uuid[ 6], uuid[ 7] + , uuid[ 8], uuid[ 9] + , uuid[10], uuid[11], uuid[12], uuid[13], uuid[14], uuid[15]); +} + static void dump_regs(struct bregs *regs) { -- 1.7.1

3 19

[announce] qemu firmware repo
by Gerd Hoffmann Dec. 21, 2012

Dec. 21, 2012

Hi, yum repository with fresh qemu firmware builds is available now: http://www.kraxel.org/repos/ Provides rpms with seabios, tianocore (efi), coreboot and ipxe binaries. enjoy, Gerd PS: /me will disappear into xmas season shortly, so any issues will have to wait 'til next year. jenkins should continue to provide fresh builds when new commits appear in the upstream repos.

1 0

Re: [SeaBIOS] [PATCH] display_uuid(): fix incomplete check after the loop
by Amos Kong Dec. 21, 2012

Dec. 21, 2012

Hi Laszlo, > In the v2->v3 change of what would become commit 37676f83 > <http://www.seabios.org/pipermail/seabios/2012-December/005166.html>, > the > defense against an initial "addr > end" condition ("wraparound") was > erroneously loosened. > > Signed-off-by: Laszlo Ersek <lersek at redhat.com> > --- > src/smbios.c | 4 ++-- > 1 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/src/smbios.c b/src/smbios.c > index aaa99bc..23713a2 100644 > --- a/src/smbios.c > +++ b/src/smbios.c > @@ -569,8 +569,8 @@ display_uuid(void) > addr += 2; > } > > - /* parsing finished, UUID not found */ > - if (addr == end) > + /* parsing finished or skipped entirely, UUID not found */ > + if (addr >= end) It's a stricter check, but I didn't find it will happen. > return; > > uuid = (u8 *)(addr + offsetof(struct smbios_type_1, uuid)); > -- > 1.7.1 | void | display_uuid(void) | { | u32 addr, end; | u8 *uuid; | u8 empty_uuid[16] = { 0 }; | | if (SMBiosAddr == NULL) | return; | | addr = SMBiosAddr->structure_table_address; | end = addr + SMBiosAddr->structure_table_length; structure_table_length is unsigned, end always >= addr here. | /* the following takes care of any initial wraparound too */ | while (addr < end) { | const struct smbios_structure_header *hdr; | | /* partial structure header */ | if (end - addr < sizeof(struct smbios_structure_header)) | return; | | hdr = (struct smbios_structure_header *)addr; | | /* partial structure */ | if (end - addr < hdr->length) | return; | | /* any Type 1 structure version will do that has the UUID */ | if (hdr->type == 1 && | hdr->length >= offsetof(struct smbios_type_1, uuid) + 16) | break; In this point, it just passes the while condition check, and addr doesn't change. * addr < end | | /* done with formatted area, skip string-set */ | addr += hdr->length; | | while (end - addr >= 2 && | (*(u8 *)addr != '\0' || | *(u8 *)(addr+1) != '\0')) | ++addr; | | /* structure terminator not found */ | if (end - addr < 2) | return; We have above check here, it means if it passes this check end >= addr + 2 | | addr += 2; After above sentence, end >= addr, there is only one condition that while condition check fails: * addr = end | } The above analysis means that we can only get this point by two conditions: * 'break' (addr < end) * not pass while's condition check (addr >= end) So the 'addr' will never be larger than 'end'. However, add this stricter check is harmless. | /* parsing finished, UUID not found */ | if (addr == end) | return; -- Amos.

2 2

Re: [SeaBIOS] [RFC PATCH v4 25/30] acpi_ich9: add hot-remove capability
by Blue Swirl Dec. 19, 2012

Dec. 19, 2012

On Tue, Dec 18, 2012 at 12:41 PM, Vasilis Liaskovitis <vasilis.liaskovitis(a)profitbricks.com> wrote: > --- > hw/acpi_ich9.c | 28 +++++++++++++++++++++++++++- > hw/acpi_ich9.h | 1 + > 2 files changed, 28 insertions(+), 1 deletions(-) > > diff --git a/hw/acpi_ich9.c b/hw/acpi_ich9.c > index abafbb5..f5dc1c9 100644 > --- a/hw/acpi_ich9.c > +++ b/hw/acpi_ich9.c > @@ -105,12 +105,29 @@ static uint32_t memhp_readb(void *opaque, uint32_t addr) > return val; > } > > +static void memhp_writeb(void *opaque, uint32_t addr, uint32_t val) > +{ > + switch (addr) { > + case ICH9_MEM_EJ_BASE - ICH9_MEM_BASE: > + dimm_notify(val, DIMM_REMOVE_SUCCESS); > + break; > + default: > + ICH9_DEBUG("memhp write invalid %x <== %d\n", addr, val); > + } > + ICH9_DEBUG("memhp write %x <== %d\n", addr, val); > +} > + > static const MemoryRegionOps ich9_memhp_ops = { > .old_portio = (MemoryRegionPortio[]) { > { > .offset = 0, .len = DIMM_BITMAP_BYTES, .size = 1, > .read = memhp_readb, > }, > + { > + .offset = ICH9_MEM_EJ_BASE - ICH9_MEM_BASE, > + .len = 1, .size = 1, > + .write = memhp_writeb, > + }, > PORTIO_END_OF_LIST() > }, > .endianness = DEVICE_LITTLE_ENDIAN, > @@ -234,6 +251,13 @@ static void enable_mem_device(ICH9LPCState *s, int memdevice) > g->mems_sts[memdevice/8] |= (1 << (memdevice%8)); > } > > +static void disable_mem_device(ICH9LPCState *s, int memdevice) > +{ > + struct gpe_regs *g = &s->pm.gperegs; > + s->pm.acpi_regs.gpe.sts[0] |= ICH9_MEM_HOTPLUG_STATUS; > + g->mems_sts[memdevice/8] &= ~(1 << (memdevice%8)); Spaces around '/' and '%'. > +} > + > static int ich9_dimm_hotplug(DeviceState *qdev, DimmDevice *dev, int > add) > { > @@ -243,6 +267,8 @@ static int ich9_dimm_hotplug(DeviceState *qdev, DimmDevice *dev, int > > if (add) { > enable_mem_device(s, slot->idx); > + } else { > + disable_mem_device(s, slot->idx); > } > pm_update_sci(&s->pm); > return 0; > @@ -270,7 +296,7 @@ void ich9_pm_init(void *device, qemu_irq sci_irq, qemu_irq cmos_s3) > memory_region_add_subregion(&pm->io, ICH9_PMIO_SMI_EN, &pm->io_smi); > > memory_region_init_io(&pm->io_memhp, &ich9_memhp_ops, pm, "apci-memhp0", > - DIMM_BITMAP_BYTES); > + DIMM_BITMAP_BYTES + 1); > memory_region_add_subregion(get_system_io(), ICH9_MEM_BASE, &pm->io_memhp); > > dimm_bus_hotplug(ich9_dimm_hotplug, &lpc->d.qdev); > diff --git a/hw/acpi_ich9.h b/hw/acpi_ich9.h > index 4419247..af61a2d 100644 > --- a/hw/acpi_ich9.h > +++ b/hw/acpi_ich9.h > @@ -24,6 +24,7 @@ > #include "acpi.h" > > #define ICH9_MEM_BASE 0xaf80 > +#define ICH9_MEM_EJ_BASE 0xafa0 > #define ICH9_MEM_HOTPLUG_STATUS 8 > > typedef struct ICH9LPCPMRegs { > -- > 1.7.9 >

1 0

Re: [SeaBIOS] [RFC PATCH v4 19/30] Implement "info memory-total" and "query-memory-total"
by Blue Swirl Dec. 19, 2012

Dec. 19, 2012

On Tue, Dec 18, 2012 at 12:41 PM, Vasilis Liaskovitis <vasilis.liaskovitis(a)profitbricks.com> wrote: > Returns total physical memory available to guest in bytes, including hotplugged > memory. Note that the number reported here may be different from what the guest > sees e.g. if the guest has not logically onlined hotplugged memory. > > This functionality is provided independently of a balloon device, since a > guest can be using ACPI memory hotplug without using a balloon device. > > v3->v4: Moved qmp command implementation to vl.c. This prevents a circular > header dependency problem. > > Signed-off-by: Vasilis Liaskovitis <vasilis.liaskovitis(a)profitbricks.com> > --- > hmp-commands.hx | 2 ++ > hmp.c | 7 +++++++ > hmp.h | 1 + > hw/dimm.c | 14 ++++++++++++++ > hw/dimm.h | 1 + > monitor.c | 7 +++++++ > qapi-schema.json | 11 +++++++++++ > qmp-commands.hx | 20 ++++++++++++++++++++ > vl.c | 9 +++++++++ > 9 files changed, 72 insertions(+), 0 deletions(-) > > diff --git a/hmp-commands.hx b/hmp-commands.hx > index 010b8c9..3fbd975 100644 > --- a/hmp-commands.hx > +++ b/hmp-commands.hx > @@ -1570,6 +1570,8 @@ show device tree > show qdev device model list > @item info roms > show roms > +@item info memory-total > +show memory-total > @end table > ETEXI > > diff --git a/hmp.c b/hmp.c > index 180ba2b..fb39b0d 100644 > --- a/hmp.c > +++ b/hmp.c > @@ -628,6 +628,13 @@ void hmp_info_block_jobs(Monitor *mon) > } > } > > +void hmp_info_memory_total(Monitor *mon) > +{ > + uint64_t ram_total; > + ram_total = (uint64_t)qmp_query_memory_total(NULL); > + monitor_printf(mon, "MemTotal: %lu\n", ram_total); Wrong format on 32 bit hosts, please use PRIu64. > +} > + > void hmp_quit(Monitor *mon, const QDict *qdict) > { > monitor_suspend(mon); > diff --git a/hmp.h b/hmp.h > index 0ab03be..25a3a70 100644 > --- a/hmp.h > +++ b/hmp.h > @@ -36,6 +36,7 @@ void hmp_info_spice(Monitor *mon); > void hmp_info_balloon(Monitor *mon); > void hmp_info_pci(Monitor *mon); > void hmp_info_block_jobs(Monitor *mon); > +void hmp_info_memory_total(Monitor *mon); > void hmp_quit(Monitor *mon, const QDict *qdict); > void hmp_stop(Monitor *mon, const QDict *qdict); > void hmp_system_reset(Monitor *mon, const QDict *qdict); > diff --git a/hw/dimm.c b/hw/dimm.c > index e384952..f181e54 100644 > --- a/hw/dimm.c > +++ b/hw/dimm.c > @@ -189,6 +189,20 @@ void dimm_setup_fwcfg_layout(uint64_t *fw_cfg_slots) > } > } > > +uint64_t get_hp_memory_total(void) > +{ > + DimmBus *bus; > + DimmDevice *slot; > + uint64_t info = 0; > + > + QLIST_FOREACH(bus, &memory_buses, next) { > + QTAILQ_FOREACH(slot, &bus->dimmlist, nextdimm) { > + info += slot->size; > + } > + } > + return info; > +} > + > static int dimm_init(DeviceState *s) > { > DimmBus *bus = DIMM_BUS(qdev_get_parent_bus(s)); > diff --git a/hw/dimm.h b/hw/dimm.h > index 75a6911..5130b2c 100644 > --- a/hw/dimm.h > +++ b/hw/dimm.h > @@ -85,5 +85,6 @@ DimmBus *dimm_bus_create(Object *parent, const char *name, uint32_t max_dimms, > dimm_calcoffset_fn pmc_set_offset); > void dimm_config_create(char *id, uint64_t size, const char *bus, uint64_t node, > uint32_t dimm_idx, uint32_t populated); > +uint64_t get_hp_memory_total(void); > > #endif > diff --git a/monitor.c b/monitor.c > index c0e32d6..6e87d0d 100644 > --- a/monitor.c > +++ b/monitor.c > @@ -2708,6 +2708,13 @@ static mon_cmd_t info_cmds[] = { > .mhandler.info = hmp_info_balloon, > }, > { > + .name = "memory-total", > + .args_type = "", > + .params = "", > + .help = "show total memory size", > + .mhandler.info = hmp_info_memory_total, > + }, > + { > .name = "qtree", > .args_type = "", > .params = "", > diff --git a/qapi-schema.json b/qapi-schema.json > index 5dfa052..33f88d6 100644 > --- a/qapi-schema.json > +++ b/qapi-schema.json > @@ -2903,6 +2903,17 @@ > { 'command': 'query-target', 'returns': 'TargetInfo' } > > ## > +# @query-memory-total: > +# > +# Returns total memory in bytes, including hotplugged dimms > +# > +# Returns: int > +# > +# Since: 1.4 > +## > +{ 'command': 'query-memory-total', 'returns': 'int' } > + > +## > # @QKeyCode: > # > # An enumeration of key name. > diff --git a/qmp-commands.hx b/qmp-commands.hx > index 5c692d0..a99117a 100644 > --- a/qmp-commands.hx > +++ b/qmp-commands.hx > @@ -2654,3 +2654,23 @@ EQMP > .args_type = "", > .mhandler.cmd_new = qmp_marshal_input_query_target, > }, > + > + { > + .name = "query-memory-total", > + .args_type = "", > + .mhandler.cmd_new = qmp_marshal_input_query_memory_total > + }, > +SQMP > +query-memory-total > +---------- > + > +Return total memory in bytes, including hotplugged dimms > + > +Example: > + > +-> { "execute": "query-memory-total" } > +<- { > + "return": 1073741824 > + } > + > +EQMP > diff --git a/vl.c b/vl.c > index 8406933..80803c5 100644 > --- a/vl.c > +++ b/vl.c > @@ -126,6 +126,7 @@ int main(int argc, char **argv) > #include "hw/xen.h" > #include "hw/qdev.h" > #include "hw/loader.h" > +#include "hw/dimm.h" > #include "bt-host.h" > #include "net.h" > #include "net/slirp.h" > @@ -442,6 +443,14 @@ StatusInfo *qmp_query_status(Error **errp) > return info; > } > > +int64_t qmp_query_memory_total(Error **errp) > +{ > + uint64_t info; > + info = ram_size + get_hp_memory_total(); > + > + return (int64_t)info; > +} > + > /***********************************************************/ > /* real time host monotonic timer */ > > -- > 1.7.9 >

1 0

Re: [SeaBIOS] [RFC PATCH v4 00/30] ACPI memory hotplug
by Gerd Hoffmann Dec. 19, 2012

Dec. 19, 2012

Hi, > correct, only the number of channels and ranks(dimms) per channel has been > emulated so far (2 channels of 4 dimms each). So it is still paravirtual memory > hotplug, not native. Native support still needs to be worked on. Ok. >> I think the paravirtual memory hotplug controller should be a PCI device >> (which we then can add as function to the chipset). Having some fixed >> magic addresses is bad. > > ok, so in your opinion a pci-based hotplug controller sounds better than adding > acpi ports to piix4 or ich9? > > Magic acpi_ich9 ports can be avoided if q35 native support is implemented. Yes. We should go that route for q35. > For > i440fx/piix4 it was discussed and more or less decided we would only support > a paravirtual way of memory hotplug. Sure, there is no other way to do it. It is probably a good idea to model piix4 paravirtual to work simliar to q35 native. > In the description. I meant "paravirtual memory bus" to describe a memory bus > with unlimited number of dimm devices. But the "hotplug control" has always > been acpi-based so far and not a pci device. It still can (and should) be acpi-based. It is just that: (a) Instead of using get_system_io() as parent memory region you use create a pci device and place the memory region in one of the PCI bars. (b) Instead of using OperationRegion($name, SystemIO, $magicaddress) you use OperationRegion($name, PciBarTarget, ...) to access the registers. cheers, Gerd

1 0

Re: [SeaBIOS] [RFC PATCH v4 00/30] ACPI memory hotplug
by Gerd Hoffmann Dec. 19, 2012

Dec. 19, 2012

Hi, > - multiple memory buses can be registered. Memory buses of the real hw/chipset > or a paravirtual memory bus can be added. IIRC q35 supports memory hotplug natively (picked up in some discussion). Is that correct? What does the code emulate? It doesn't look like it emulates q35 memory hotplug ... I think the paravirtual memory hotplug controller should be a PCI device (which we then can add as function to the chipset). Having some fixed magic addresses is bad. [ btw: same goes for ACPI PCI hotplug, that is hardly fixable without breaking compatibility though, for q35 we should be able to do better ]. cheers, Gerd

2 1