[PATCH v2 0/2] add guest hint for memory layout, take three

[Gerd Hoffmann]

Gerd Hoffmann (2):
  add romfile_loadbool()
  update pci_pad_mem64 handling

src/romfile.h    |  1 +
 src/fw/pciinit.c | 13 ++++++++++---
 src/romfile.c    | 23 +++++++++++++++++++++++
 3 files changed, 34 insertions(+), 3 deletions(-)

[PATCH v2 1/2] add romfile_loadbool()

[Gerd Hoffmann]

Translates strings in fw_cfg files into boolean values.

Signed-off-by: Gerd Hoffmann 
---
 src/romfile.h |  1 +
 src/romfile.c | 23 +++++++++++++++++++++++
 2 files changed, 24 insertions(+)

diff --git a/src/romfile.h b/src/romfile.h
index 3e0f820047dd..ae2f4ac7e718 100644
--- a/src/romfile.h
+++ b/src/romfile.h
@@ -15,6 +15,7 @@ struct romfile_s *romfile_findprefix(const char *prefix, struct romfile_s *prev)
 struct romfile_s *romfile_find(const char *name);
 void *romfile_loadfile(const char *name, int *psize);
 u64 romfile_loadint(const char *name, u64 defval);
+u32 romfile_loadbool(const char *name, u32 defval);

void const_romfile_add_int(char *name, u32 value);

diff --git a/src/romfile.c b/src/romfile.c
index b598274edc09..8072a9150685 100644
--- a/src/romfile.c
+++ b/src/romfile.c
@@ -99,6 +99,29 @@ romfile_loadint(const char *name, u64 defval)
     return val;
 }

+u32
+romfile_loadbool(const char *name, u32 defval)
+{
+    static const char *true_strings[] = { "1", "on", "yes" };
+    static const char *false_strings[] = { "0", "off", "no" };
+    char *str = romfile_loadfile(name, NULL);
+    int i;
+
+    if (!str)
+        return defval;
+
+    for (i = 0; i < ARRAY_SIZE(true_strings); i++)
+        if (0 == strcmp(str, true_strings[i]))
+            return 1;
+
+    for (i = 0; i < ARRAY_SIZE(false_strings); i++)
+        if (0 == strcmp(str, false_strings[i]))
+            return 0;
+
+    dprintf(1, "%s: unknown bool string: %s\n", __func__, str);
+    return defval;
+}
+
 struct const_romfile_s {
     struct romfile_s file;
     void *data;

[PATCH v2 2/2] update pci_pad_mem64 handling

[Gerd Hoffmann]

Add a new possible state: '-1' means 'use default'.  In that case
seabios continue to use the current heuristic:  In case memory
above 4G is present enable 64-bit guest friendly configuration.

This allows forcing the one or the other behavior by setting the
pci_pad_mem64 variable beforehand (which is done by another patch).

Signed-off-by: Gerd Hoffmann 
---
 src/fw/pciinit.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/src/fw/pciinit.c b/src/fw/pciinit.c
index b3e359d7fe62..be76c447cf19 100644
--- a/src/fw/pciinit.c
+++ b/src/fw/pciinit.c
@@ -55,7 +55,7 @@ u64 pcimem64_end   = BUILD_PCIMEM64_END;
 // Resource allocation limits
 static u64 pci_io_low_end = 0xa000;
 static u64 pci_mem64_top  = 0;
-static u32 pci_pad_mem64  = 0;
+static u32 pci_pad_mem64  = -1;

struct pci_region_entry {
     struct pci_device *dev;
@@ -1202,8 +1202,15 @@ pci_setup(void)
         }
     }

-    if (CPUPhysBits >= 36 && CPULongMode && RamSizeOver4G)
-        pci_pad_mem64 = 1;
+    pci_pad_mem64 = romfile_loadbool("opt/org.seabios/pci64", -1);
+
+    if (pci_pad_mem64 == -1)
+        // when not set enable in case memory over 4G is present
+        pci_pad_mem64 = RamSizeOver4G ? 1 : 0;
+
+    if (!CPULongMode)
+        // force off for 32-bit CPUs
+        pci_pad_mem64 = 0;

dprintf(1, "=== PCI bus & bridge init ===\n");
     if (pci_probe_host() != 0) {

[PATCH v2 2/2] update pci_pad_mem64 handling

[Fiona Ebner]

Thank you! Works for my 32-bit Debian 12 VM :)

I wanted to briefly ask if a new SeaBIOS release is planned before/for
QEMU 10.0?

A colleague told me that commit ec0bc256 ("limit address space used for
pci devices, part two") alleviates another issue one of our users faced.
Of course we could ship an updated SeaBIOS ROM downstream, but others
might benefit from a new release too, so I wanted to ask.

Best Regards,
Fiona

Am 22.11.24 um 15:33 schrieb Gerd Hoffmann:
> Add a new possible state: '-1' means 'use default'.  In that case
> seabios continue to use the current heuristic:  In case memory
> above 4G is present enable 64-bit guest friendly configuration.
>
> This allows forcing the one or the other behavior by setting the
> pci_pad_mem64 variable beforehand (which is done by another patch).
> Signed-off-by: Gerd Hoffmann
>
> src/fw/pciinit.c | 13 ++++++++++---
>  1 file changed, 10 insertions(+), 3 deletions(-)
>
> diff --git a/src/fw/pciinit.c b/src/fw/pciinit.c
> index b3e359d7fe62..be76c447cf19 100644
> --- a/src/fw/pciinit.c
> +++ b/src/fw/pciinit.c
> @@ -55,7 +55,7 @@ u64 pcimem64_end   = BUILD_PCIMEM64_END;
>  // Resource allocation limits
>  static u64 pci_io_low_end = 0xa000;
>  static u64 pci_mem64_top  = 0;
> -static u32 pci_pad_mem64  = 0;
> +static u32 pci_pad_mem64  = -1;
>
> struct pci_region_entry {
>      struct pci_device *dev;
> @@ -1202,8 +1202,15 @@ pci_setup(void)
>          }
>      }
> if (CPUPhysBits >= 36 && CPULongMode && RamSizeOver4G)
>    pci_pad_mem64 = 1;
>
> pci_pad_mem64 = romfile_loadbool("opt/org.seabios/pci64", -1);
>
> if (pci_pad_mem64 == -1)
>    // when not set enable in case memory over 4G is present
>    pci_pad_mem64 = RamSizeOver4G ? 1 : 0;
>
> if (!CPULongMode)
>    // force off for 32-bit CPUs
>    pci_pad_mem64 = 0;
>
> dprintf(1, "=== PCI bus & bridge init ===\n");
>    if (pci_probe_host() != 0) {

[PATCH v2 2/2] update pci_pad_mem64 handling

[Gerd Hoffmann]

On Thu, Mar 06, 2025 at 11:44:01AM +0100, Fiona Ebner wrote:
> Thank you! Works for my 32-bit Debian 12 VM :)
>
> I wanted to briefly ask if a new SeaBIOS release is planned before/for
> QEMU 10.0?

Hmm.  Kevin?  1.17.0 release was briefly discussed a while back (january
IIRC).  Doing another 1.16.x feels wrong with the amount of changes we
have in master now.

I'm not aware of any pending changes right now.  So we could call code
freeze now and tag the release in ~ two weeks.  I can update seabios in
qemu to a snapshot now for some additional test coverage, and to 1.17.0
final by end of march.

comments?

take care,
  Gerd

[PATCH v2 2/2] update pci_pad_mem64 handling

[Kevin O'Connor]

On Mon, Mar 10, 2025 at 04:21:30PM +0100, Gerd Hoffmann wrote:
> On Thu, Mar 06, 2025 at 11:44:01AM +0100, Fiona Ebner wrote:
> > Thank you! Works for my 32-bit Debian 12 VM :)
> >
> > I wanted to briefly ask if a new SeaBIOS release is planned before/for
> > QEMU 10.0?
>
> Hmm.  Kevin?  1.17.0 release was briefly discussed a while back (january
> IIRC).  Doing another 1.16.x feels wrong with the amount of changes we
> have in master now.
>
> I'm not aware of any pending changes right now.  So we could call code
> freeze now and tag the release in ~ two weeks.  I can update seabios in
> qemu to a snapshot now for some additional test coverage, and to 1.17.0
> final by end of march.
>
> comments?

I don't see an issue with tagging a new release.  Alas, my schedule
has been busy recently.  How about we target mid-April for the next
release?

-Kevin

[Kevin O'Connor]

On Fri, Nov 22, 2024 at 03:33:04PM +0100, Gerd Hoffmann wrote:
> Gerd Hoffmann (2):
>   add romfile_loadbool()
>   update pci_pad_mem64 handling

Thanks.  It looks fine to me.  Feel free to commit when you are ready.

Cheers,
-Kevin
> src/romfile.h    |  1 +
>  src/fw/pciinit.c | 13 ++++++++++---
>  src/romfile.c    | 23 +++++++++++++++++++++++
>  3 files changed, 34 insertions(+), 3 deletions(-)
>
> SeaBIOS mailing list -- [e-mail]
> To unsubscribe send an email to [e-mail]

[Gerd Hoffmann]

On Tue, Dec 24, 2024 at 06:49:35PM -0500, Kevin O'Connor wrote:
> On Fri, Nov 22, 2024 at 03:33:04PM +0100, Gerd Hoffmann wrote:
> > Gerd Hoffmann (2):
> >   add romfile_loadbool()
> >   update pci_pad_mem64 handling
>
> Thanks.  It looks fine to me.  Feel free to commit when you are ready.

Pushed now.

take care,
  Gerd