From b1aade87ca1f5d3788bd6ef570b438c2dc7a9428 Mon Sep 17 00:00:00 2001 From: Mattijs Korpershoek Date: Fri, 7 Jul 2023 10:13:34 +0200 Subject: lib: sparse: allocate FASTBOOT_MAX_BLK_WRITE instead of small number Commit 62649165cb02 ("lib: sparse: Make CHUNK_TYPE_RAW buffer aligned") fixed cache alignment for systems with a D-CACHE. However it introduced some performance regressions [1] on system flashing huge images, such as Android. On AM62x SK EVM, we also observe such performance penalty: Sending sparse 'super' 1/2 (768793 KB) OKAY [ 23.954s] Writing 'super' OKAY [ 75.926s] Sending sparse 'super' 2/2 (629819 KB) OKAY [ 19.641s] Writing 'super' OKAY [ 62.849s] Finished. Total time: 182.474s The reason for this is that we use an arbitrary small buffer (info->blksz * 100) for transferring. Fix it by using a bigger buffer (info->blksz * FASTBOOT_MAX_BLK_WRITE) as suggested in the original's patch review [2]. With this patch, performance impact is mitigated: Sending sparse 'super' 1/2 (768793 KB) OKAY [ 23.912s] Writing 'super' OKAY [ 15.780s] Sending sparse 'super' 2/2 (629819 KB) OKAY [ 19.581s] Writing 'super' OKAY [ 17.192s] Finished. Total time: 76.569s [1] https://lore.kernel.org/r/20221118121323.4009193-1-gary.bisson@boundarydevices.com [2] https://lore.kernel.org/r/all/43e4c17c-4483-ec8e-f843-9b4c5569bd18@seco.com/ Fixes: 62649165cb02 ("lib: sparse: Make CHUNK_TYPE_RAW buffer aligned") Signed-off-by: Mattijs Korpershoek --- include/image-sparse.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include') diff --git a/include/image-sparse.h b/include/image-sparse.h index 0572dbd0a28..282a0b25649 100644 --- a/include/image-sparse.h +++ b/include/image-sparse.h @@ -7,6 +7,8 @@ #include #include +#define FASTBOOT_MAX_BLK_WRITE 16384 + #define ROUNDUP(x, y) (((x) + ((y) - 1)) & ~((y) - 1)) struct sparse_storage { -- cgit v1.3.1