[xen staging] x86/boot: Use fastcall for 32bit code

[patchbot@xxxxxxx]

commit 62da0b364a507d3fe48dcdb7c38592cc1848342f
Author:     Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
AuthorDate: Mon Sep 2 12:11:51 2024 +0100
Commit:     Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
CommitDate: Tue Sep 3 15:51:44 2024 +0100

    x86/boot: Use fastcall for 32bit code
    
    This is marginally more efficient, but is mostly to get rid of the use of
    stdcall in cmdline.c and reloc.c
    
    No functional change.
    
    Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
    Reviewed-by: Jan Beulich <jbeulich@xxxxxxxx>
---
 xen/arch/x86/boot/Makefile  |  2 +-
 xen/arch/x86/boot/cmdline.c |  6 +++---
 xen/arch/x86/boot/defs.h    |  1 -
 xen/arch/x86/boot/head.S    | 15 ++++++++-------
 xen/arch/x86/boot/reloc.c   | 12 ++++++------
 5 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/xen/arch/x86/boot/Makefile b/xen/arch/x86/boot/Makefile
index de47740277..8f5bbff0cc 100644
--- a/xen/arch/x86/boot/Makefile
+++ b/xen/arch/x86/boot/Makefile
@@ -13,7 +13,7 @@ $(obj)/head.o: $(head-bin-objs:.o=.bin)
 
 CFLAGS_x86_32 := $(subst -m64,-m32 -march=i686,$(XEN_TREEWIDE_CFLAGS))
 $(call cc-options-add,CFLAGS_x86_32,CC,$(EMBEDDED_EXTRA_CFLAGS))
-CFLAGS_x86_32 += -Werror -fno-builtin -g0 -msoft-float
+CFLAGS_x86_32 += -Werror -fno-builtin -g0 -msoft-float -mregparm=3
 CFLAGS_x86_32 += -nostdinc -include $(filter 
%/include/xen/config.h,$(XEN_CFLAGS))
 CFLAGS_x86_32 += $(filter -I%,$(XEN_CFLAGS))
 
diff --git a/xen/arch/x86/boot/cmdline.c b/xen/arch/x86/boot/cmdline.c
index b7375d1066..c1d21180b6 100644
--- a/xen/arch/x86/boot/cmdline.c
+++ b/xen/arch/x86/boot/cmdline.c
@@ -20,8 +20,8 @@
 
 /*
  * This entry point is entered from xen/arch/x86/boot/head.S with:
- *   - 0x4(%esp) = &cmdline,
- *   - 0x8(%esp) = &early_boot_opts.
+ *   - %eax      = &cmdline,
+ *   - %edx      = &early_boot_opts.
  */
 asm (
     "    .text                         \n"
@@ -347,7 +347,7 @@ static void vga_parse(const char *cmdline, 
early_boot_opts_t *ebo)
 #endif
 
 /* SAF-1-safe */
-void __stdcall cmdline_parse_early(const char *cmdline, early_boot_opts_t *ebo)
+void cmdline_parse_early(const char *cmdline, early_boot_opts_t *ebo)
 {
     if ( !cmdline )
         return;
diff --git a/xen/arch/x86/boot/defs.h b/xen/arch/x86/boot/defs.h
index 4d519ac4f5..d733ee9b8b 100644
--- a/xen/arch/x86/boot/defs.h
+++ b/xen/arch/x86/boot/defs.h
@@ -22,6 +22,5 @@
 
 #define __maybe_unused __attribute__((__unused__))
 #define __packed       __attribute__((__packed__))
-#define __stdcall      __attribute__((__stdcall__))
 
 #endif /* __BOOT_DEFS_H__ */
diff --git a/xen/arch/x86/boot/head.S b/xen/arch/x86/boot/head.S
index d867b015d9..12bbb97f33 100644
--- a/xen/arch/x86/boot/head.S
+++ b/xen/arch/x86/boot/head.S
@@ -613,10 +613,11 @@ trampoline_setup:
 
         /* Save Multiboot / PVH info struct (after relocation) for later use. 
*/
         push    %edx                /* Boot video info to be filled from MB2. 
*/
-        push    %ecx                /* Bottom-most low-memory stack address. */
-        push    %ebx                /* Multiboot / PVH information address. */
-        push    %eax                /* Magic number. */
+        mov     %ebx, %edx          /* Multiboot / PVH information address. */
+        /*      reloc(magic/eax, info/edx, trampoline/ecx, video/stk) using 
fastcall. */
         call    reloc
+        add     $4, %esp
+
 #ifdef CONFIG_PVH_GUEST
         cmpb    $0, sym_esi(pvh_boot)
         je      1f
@@ -848,9 +849,9 @@ trampoline_setup:
         testl   $MBI_CMDLINE,MB_flags(%ebx)
         jz      1f
 
-        lea     sym_esi(early_boot_opts),%eax
-        push    %eax
-        pushl   MB_cmdline(%ebx)
+        lea     sym_esi(early_boot_opts), %edx
+        mov     MB_cmdline(%ebx), %eax
+        /*      cmdline_parse_early(cmdline/eax, opts/edx) using fastcall. */
         call    cmdline_parse_early
 
 1:
@@ -871,7 +872,7 @@ trampoline_setup:
 
         /*
          * cmdline and reloc are written in C, and linked to be 32bit PIC with
-         * entrypoints at 0 and using the stdcall convention.
+         * entrypoints at 0 and using the fastcall convention.
          */
         ALIGN
 cmdline_parse_early:
diff --git a/xen/arch/x86/boot/reloc.c b/xen/arch/x86/boot/reloc.c
index eb9902d73f..38adca1f36 100644
--- a/xen/arch/x86/boot/reloc.c
+++ b/xen/arch/x86/boot/reloc.c
@@ -14,10 +14,10 @@
 
 /*
  * This entry point is entered from xen/arch/x86/boot/head.S with:
- *   - 0x04(%esp) = MAGIC,
- *   - 0x08(%esp) = INFORMATION_ADDRESS,
- *   - 0x0c(%esp) = TOPMOST_LOW_MEMORY_STACK_ADDRESS.
- *   - 0x10(%esp) = BOOT_VIDEO_INFO_ADDRESS.
+ *   - %eax       = MAGIC,
+ *   - %edx       = INFORMATION_ADDRESS,
+ *   - %ecx       = TOPMOST_LOW_MEMORY_STACK_ADDRESS.
+ *   - 0x04(%esp) = BOOT_VIDEO_INFO_ADDRESS.
  */
 asm (
     "    .text                         \n"
@@ -354,8 +354,8 @@ static multiboot_info_t *mbi2_reloc(uint32_t mbi_in, 
uint32_t video_out)
 }
 
 /* SAF-1-safe */
-void *__stdcall reloc(uint32_t magic, uint32_t in, uint32_t trampoline,
-                      uint32_t video_info)
+void *reloc(uint32_t magic, uint32_t in, uint32_t trampoline,
+            uint32_t video_info)
 {
     alloc = trampoline;
 
--
generated by git-patchbot for /home/xen/git/xen.git#staging