Firmware/opensbi
8
0
Fork 0
forked from Mirrors/opensbi
Code Pull Requests Activity

lib: sbi: Wakeup non-coldboot HARTs early in the coldboot path

Browse Source 
Currently, all non-coldboot HARTs busy spin in wait_for_coldboot()
until the entire coldboot init sequence is completed.

This means:
1) On QEMU, all non-coldboot HARTs will eat host CPU time and
   also slow down the coldboot HART until the entire coldboot
   init sequence is completed.
2) On real HW, all non-coldboot HARTs will consume more CPU
   power until the entire coldboot init sequence is completed.

To address this, wake up all non-coldboot HARTs as early as
possible in the coldboot init sequence.

Signed-off-by: Anup Patel <apatel@ventanamicro.com>
This commit is contained in:
Anup Patel
2024-03-19 17:56:24 +05:30
committed by Anup Patel
parent beb0cd177f
commit 7862c244bc
1 changed files with 9 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
lib/sbi/sbi_init.c
View File

@@ -244,6 +244,14 @@ static void __noreturn init_coldboot(struct sbi_scratch *scratch, u32 hartid)
if (rc)
sbi_hart_hang();
/*
* All non-coldboot HARTs do HSM initialization (i.e. enter HSM state
* machine) at the start of the warmboot path so it is wasteful to
* have these HARTs busy spin in wait_for_coldboot() until coldboot
* path is completed.
*/
wake_coldboot_harts(scratch, hartid);
rc = sbi_platform_early_init(plat, true);
if (rc)
sbi_hart_hang();
@@ -355,8 +363,6 @@ static void __noreturn init_coldboot(struct sbi_scratch *scratch, u32 hartid)
sbi_hart_hang();
}
wake_coldboot_harts(scratch, hartid);
count = sbi_scratch_offset_ptr(scratch, init_count_offset);
(*count)++;
@@ -376,6 +382,7 @@ static void __noreturn init_warm_startup(struct sbi_scratch *scratch,
count = sbi_scratch_offset_ptr(scratch, entry_count_offset);
(*count)++;
/* Note: This has to be first thing in warmboot init sequence */
rc = sbi_hsm_init(scratch, hartid, false);
if (rc)
sbi_hart_hang();