[PATCH 18/18] dbtr: Work around specification bug in range checks

[Nicholas Piggin]

The SBI spec appears to have an off-by-one / inequality error in range
testing, reported here:

https://github.com/riscv-non-isa/riscv-sbi-doc/issues/257

This change updates range checking accordingly.

Note: may want to hold off this change until upstream spec confirms.

Signed-off-by: Nicholas Piggin <npiggin at gmail.com>
---
 lib/sbi/sbi_dbtr.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/lib/sbi/sbi_dbtr.c b/lib/sbi/sbi_dbtr.c
index f3186993..2edec1f5 100644
--- a/lib/sbi/sbi_dbtr.c
+++ b/lib/sbi/sbi_dbtr.c
@@ -689,12 +689,9 @@ int sbi_dbtr_read_trig(unsigned long smode,
 	if (!hs)
 		return SBI_ERR_FAILED;
 
-	if (trig_count >= hs->total_trigs)
-		return SBI_ERR_BAD_RANGE;
-
 	if (trig_idx_base >= hs->total_trigs ||
-	    trig_idx_base + trig_count >= hs->total_trigs)
-		return SBI_ERR_INVALID_PARAM;
+	    trig_idx_base + trig_count > hs->total_trigs)
+		return SBI_ERR_BAD_RANGE;
 
 	if (sbi_dbtr_shmem_disabled(hs))
 		return SBI_ERR_NO_SHMEM;
@@ -752,7 +749,7 @@ int sbi_dbtr_install_trig(unsigned long smode,
 	if (!hs)
 		return SBI_ERR_FAILED;
 
-	if (trig_count >= hs->total_trigs)
+	if (trig_count > hs->total_trigs)
 		return SBI_ERR_BAD_RANGE;
 
 	if (sbi_dbtr_shmem_disabled(hs))
@@ -934,7 +931,7 @@ int sbi_dbtr_update_trig(unsigned long smode,
 	if (!hs)
 		return SBI_ERR_FAILED;
 
-	if (trig_count >= hs->total_trigs)
+	if (trig_count > hs->total_trigs)
 		return SBI_ERR_BAD_RANGE;
 
 	if (sbi_dbtr_shmem_disabled(hs))
-- 
2.51.0