From: "Michael S. Tsirkin" <mst@redhat.com>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: qemu-devel@nongnu.org, Anthony Liguori <anthony@codemonkey.ws>,
afaerber@suse.de
Subject: Re: [Qemu-devel] [PATCH v4 15/23] i386: add bios linker/loader
Date: Mon, 23 Sep 2013 16:47:58 +0300 [thread overview]
Message-ID: <20130923134758.GE1278@redhat.com> (raw)
In-Reply-To: <5240447E.5090601@redhat.com>
On Mon, Sep 23, 2013 at 03:39:10PM +0200, Paolo Bonzini wrote:
> Il 23/09/2013 15:36, Michael S. Tsirkin ha scritto:
> > On Mon, Sep 23, 2013 at 02:48:57PM +0200, Paolo Bonzini wrote:
> >> Il 22/09/2013 15:37, Michael S. Tsirkin ha scritto:
> >>> This adds a dynamic bios linker/loader.
> >>> This will be used by acpi table generation
> >>> code to:
> >>> - load each table in the appropriate memory segment
> >>> - link tables to each other
> >>> - fix up checksums after said linking
> >>>
> >>> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
> >>> ---
> >>> include/hw/i386/bios-linker-loader.h | 26 ++++++
> >>> hw/i386/bios-linker-loader.c | 156 +++++++++++++++++++++++++++++++++++
> >>> hw/i386/Makefile.objs | 1 +
> >>
> >> Just a naming question...
> >>
> >> If this is ACPI-specific (or if there are no other uses in sight), it
> >> should go in include/hw/acpi and hw/acpi, and renamed with s/bios/acpi/ etc.
> >>
> >> If this is i386-specific, it should go entirely in hw/i386 and the
> >> headers should be included as "bios-linker-loader.h".
> >
> > Definitely not ACPI specific.
> > It's only used on i386 so I put it all under i386.
> > You think we should replace
> > #include "hw/i386/bios-linker-loader.h" with
> > #include "bios-linker-loader.h"?
> >
> > I don't really care, even though this makes headers
> > hard to find. What's the benefit?
>
> Private headers are outside include/, public headers within it.
Ah you mean move it outside include/. OK.
> Same
> conventions as Linux. But if it's not ACPI-specific, I don't feel too
> strongly about it.
>
> > It's not easy to find a good name for it.
> > For now I'll just make all names bios_linker_loader_ consistently.
> > We can always rename it later, it's not set in stone.
>
> Ok.
>
> >>> +struct BiosLinkerLoaderEntry {
> >>> + uint32_t command;
> >>> + union {
> >>> + /*
> >>> + * COMMAND_ALLOCATE - allocate a table from @alloc_file
> >>> + * subject to @alloc_align alignment (must be power of 2)
> >>> + * and @alloc_zone (can be HIGH or FSEG) requirements.
> >>> + *
> >>> + * Must appear exactly once for each file, and before
> >>> + * this file is referenced by any other command.
> >>> + */
> >>> + struct {
> >>> + char alloc_file[BIOS_LINKER_LOADER_FILESZ];
> >>> + uint32_t alloc_align;
> >>> + uint8_t alloc_zone;
> >>> + };
> >>
> >> I would prefer to have named structs, like
> >>
> >> struct {
> >> char file[BIOS_LINKER_LOADER_FILESZ];
> >> uint32_t align;
> >> uint8_t zone;
> >> } alloc;
> >
> > Why do you prefer this? It just makes code more verbose.
>
> It doesn't, the names were already prefixed with alloc_/pointer_/cksum_.
> All it does is change underscores to periods.
>
> Paolo
You are right, that's cleaner. I'll do this.
I don't want to sprinkle QEMU_PACKED around though,
it's just there for documentation purposes, so one
place is enough.
> >>> + /*
> >>> + * COMMAND_ADD_POINTER - patch the table (originating from
> >>> + * @dest_file) at @pointer_offset, by adding a pointer to the table
> >>> + * originating from @src_file. 1,2,4 or 8 byte unsigned
> >>> + * addition is used depending on @pointer_size.
> >>> + */
> >>> + struct {
> >>> + char pointer_dest_file[BIOS_LINKER_LOADER_FILESZ];
> >>> + char pointer_src_file[BIOS_LINKER_LOADER_FILESZ];
> >>> + uint32_t pointer_offset;
> >>> + uint8_t pointer_size;
> >>> + };
> >>> +
> >>> + /*
> >>> + * COMMAND_ADD_CHECKSUM - calculate checksum of the range specified by
> >>> + * @cksum_start and @cksum_length fields,
> >>> + * and then add the value at @cksum_offset.
> >>> + * Checksum simply sums -X for each byte X in the range
> >>> + * using 8-bit math.
> >>> + */
> >>> + struct {
> >>> + char cksum_file[BIOS_LINKER_LOADER_FILESZ];
> >>> + uint32_t cksum_offset;
> >>> + uint32_t cksum_start;
> >>> + uint32_t cksum_length;
> >>> + };
> >>> +
> >>> + /* padding */
> >>> + char pad[124];
> >>> + };
> >>> +} QEMU_PACKED;
> >>> +typedef struct BiosLinkerLoaderEntry BiosLinkerLoaderEntry;
> >>> +
> >>> +enum {
> >>> + BIOS_LINKER_LOADER_COMMAND_ALLOCATE = 0x1,
> >>> + BIOS_LINKER_LOADER_COMMAND_ADD_POINTER = 0x2,
> >>> + BIOS_LINKER_LOADER_COMMAND_ADD_CHECKSUM = 0x3,
> >>> +};
> >>> +
> >>> +enum {
> >>> + BIOS_LINKER_LOADER_ALLOC_ZONE_HIGH = 0x1,
> >>> + BIOS_LINKER_LOADER_ALLOC_ZONE_FSEG = 0x2,
> >>> +};
> >>> +
> >>> +GArray *bios_linker_init(void)
> >>> +{
> >>> + return g_array_new(false, true /* clear */, sizeof(BiosLinkerLoaderEntry));
> >>> +}
> >>> +
> >>> +/* Free linker wrapper and return the linker array. */
> >>> +void *bios_linker_cleanup(GArray *linker)
> >>> +{
> >>> + return g_array_free(linker, false);
> >>> +}
> >>> +
> >>> +void bios_linker_alloc(GArray *linker,
> >>> + const char *file,
> >>> + uint32_t alloc_align,
> >>> + bool alloc_fseg)
> >>> +{
> >>> + BiosLinkerLoaderEntry entry;
> >>> +
> >>> + memset(&entry, 0, sizeof entry);
> >>> + strncpy(entry.alloc_file, file, sizeof entry.alloc_file - 1);
> >>
> >> Please use pstrcpy (also below).
> >
> > OK.
> >
> >>> + entry.command = cpu_to_le32(BIOS_LINKER_LOADER_COMMAND_ALLOCATE);
> >>> + entry.alloc_align = cpu_to_le32(alloc_align);
> >>> + entry.alloc_zone = cpu_to_le32(alloc_fseg ?
> >>> + BIOS_LINKER_LOADER_ALLOC_ZONE_FSEG :
> >>> + BIOS_LINKER_LOADER_ALLOC_ZONE_HIGH);
> >>> +
> >>> + /* Alloc entries must come first, so prepend them */
> >>> + g_array_prepend_val(linker, entry);
> >>> +}
> >>> +
> >>> +void bios_linker_add_checksum(GArray *linker, const char *file, void *table,
> >>> + void *start, unsigned size, uint8_t *checksum)
> >>> +{
> >>> + BiosLinkerLoaderEntry entry;
> >>> +
> >>> + memset(&entry, 0, sizeof entry);
> >>> + strncpy(entry.cksum_file, file, sizeof entry.cksum_file - 1);
> >>> + entry.command = cpu_to_le32(BIOS_LINKER_LOADER_COMMAND_ADD_CHECKSUM);
> >>> + entry.cksum_offset = cpu_to_le32(checksum - (uint8_t *)table);
> >>> + entry.cksum_start = cpu_to_le32((uint8_t *)start - (uint8_t *)table);
> >>> + entry.cksum_length = cpu_to_le32(size);
> >>> +
> >>> + g_array_append_val(linker, entry);
> >>> +}
> >>> +
> >>> +void bios_linker_add_pointer(GArray *linker,
> >>> + const char *dest_file,
> >>> + const char *src_file,
> >>> + GArray *table, void *pointer,
> >>> + uint8_t pointer_size)
> >>> +{
> >>> + BiosLinkerLoaderEntry entry;
> >>> +
> >>> + memset(&entry, 0, sizeof entry);
> >>> + strncpy(entry.pointer_dest_file, dest_file,
> >>> + sizeof entry.pointer_dest_file - 1);
> >>> + strncpy(entry.pointer_src_file, src_file,
> >>> + sizeof entry.pointer_src_file - 1);
> >>> + entry.command = cpu_to_le32(BIOS_LINKER_LOADER_COMMAND_ADD_POINTER);
> >>> + entry.pointer_offset = cpu_to_le32((gchar *)pointer - table->data);
> >>> + entry.pointer_size = pointer_size;
> >>> + assert(pointer_size == 1 || pointer_size == 2 ||
> >>> + pointer_size == 4 || pointer_size == 8);
> >>> +
> >>> + g_array_append_val(linker, entry);
> >>> +}
> >>> diff --git a/hw/i386/Makefile.objs b/hw/i386/Makefile.objs
> >>> index f950707..b9ca380 100644
> >>> --- a/hw/i386/Makefile.objs
> >>> +++ b/hw/i386/Makefile.objs
> >>> @@ -5,6 +5,7 @@ obj-y += pc_sysfw.o
> >>> obj-$(CONFIG_XEN) += xen_domainbuild.o xen_machine_pv.o
> >>>
> >>> obj-y += kvmvapic.o
> >>> +obj-y += bios-linker-loader.o
> >>>
> >>> iasl-option=$(shell if test -z "`$(1) $(2) 2>&1 > /dev/null`" \
> >>> ; then echo "$(2)"; else echo "$(3)"; fi ;)
> >>>
next prev parent reply other threads:[~2013-09-23 13:46 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-09-22 13:37 [Qemu-devel] [PATCH v4 00/23] qemu: generate acpi tables for the guest Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 01/23] qemu: add Error to typedefs Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 02/23] qom: pull in qemu/typedefs Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 03/23] qom: cleanup struct Error references Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 04/23] qom: add pointer to int property helpers Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 05/23] fw_cfg: interface to trigger callback on read Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 06/23] loader: support for unmapped ROM blobs Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 07/23] pcie_host: expose UNMAPPED macro Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 08/23] pcie_host: expose address format Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 09/23] q35: use macro for MCFG property name Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 10/23] q35: expose mmcfg size as a property Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 11/23] i386: add ACPI table files from seabios Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 12/23] acpi: add rules to compile ASL source Michael S. Tsirkin
2013-09-23 12:36 ` Paolo Bonzini
2013-09-23 13:39 ` Michael S. Tsirkin
2013-09-23 13:44 ` Laszlo Ersek
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 13/23] acpi: pre-compiled ASL files Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 14/23] loader: use file path size from fw_cfg.h Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 15/23] i386: add bios linker/loader Michael S. Tsirkin
2013-09-23 12:48 ` Paolo Bonzini
2013-09-23 13:36 ` Michael S. Tsirkin
2013-09-23 13:39 ` Paolo Bonzini
2013-09-23 13:47 ` Michael S. Tsirkin [this message]
2013-09-23 13:52 ` Paolo Bonzini
2013-09-23 19:08 ` Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 16/23] loader: allow adding ROMs in done callbacks Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 17/23] i386: define pc guest info Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 18/23] acpi/piix: add macros for acpi property names Michael S. Tsirkin
2013-09-22 13:37 ` [Qemu-devel] [PATCH v4 19/23] piix: APIs for pc guest info Michael S. Tsirkin
2013-09-23 9:27 ` Igor Mammedov
2013-09-23 10:10 ` Michael S. Tsirkin
2013-09-23 11:14 ` Igor Mammedov
2013-09-23 11:28 ` Michael S. Tsirkin
2013-09-22 13:38 ` [Qemu-devel] [PATCH v4 20/23] ich9: " Michael S. Tsirkin
2013-09-22 13:38 ` [Qemu-devel] [PATCH v4 21/23] pvpanic: add API to access io port Michael S. Tsirkin
2013-09-22 13:38 ` [Qemu-devel] [PATCH v4 22/23] hpet: add API to find it Michael S. Tsirkin
2013-09-22 19:22 ` Paolo Bonzini
2013-09-22 19:58 ` Michael S. Tsirkin
2013-09-22 13:38 ` [Qemu-devel] [PATCH v4 23/23] i386: ACPI table generation code from seabios Michael S. Tsirkin
2013-09-24 9:23 ` [Qemu-devel] [PATCH v4 00/23] qemu: generate acpi tables for the guest Gerd Hoffmann
2013-09-24 21:45 ` Michael S. Tsirkin
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=20130923134758.GE1278@redhat.com \
--to=mst@redhat.com \
--cc=afaerber@suse.de \
--cc=anthony@codemonkey.ws \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.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).