Skip to contentSkip to conversation

ALSA: ctxfi: Simplify dao_clear_{left,right}_input() functions

PATCH Superseded Ethan Carter Edwards 3 participants 4 messages Dec 30, 2024 Sound

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 ALSA: ctxfi: Simplify dao_clear_{left,right}_input() functions

There was a lote of code duplication in the dao_clear_left_input() and dao_clear_right_input() functions. A new function, dao_clear_input(), was created and now the left and right functions call it instead of repeating themselves.

---
Changes in V2: Fixed formatting caused by email client. Sorry!

 sound/pci/ctxfi/ctdaio.c | 50 ++++++++++++++--------------------------
 1 file changed, 17 insertions(+), 33 deletions(-)
sound/pci/ctxfi/ctdaio.c +17 -33
index 83aaf9441ef3..d46542d0b454 100644
@@ -211,52 +211,36 @@ static int dao_set_right_input(struct dao *dao, struct rsc *input)
 	return 0;
 }
 
-static int dao_clear_left_input(struct dao *dao)
+static int dao_clear_input(struct dao *dao, unsigned int start, unsigned int end)
 {
-	struct imapper *entry;
-	struct daio *daio = &dao->daio;
-	int i;
+	struct imapper *to_free = dao->imappers[start];
+	unsigned int i;
 
-	if (!dao->imappers[0])
+	if (!to_free)
 		return 0;
-
-	entry = dao->imappers[0];
-	dao->mgr->imap_delete(dao->mgr, entry);
-	/* Program conjugate resources */
-	for (i = 1; i < daio->rscl.msr; i++) {
-		entry = dao->imappers[i];
-		dao->mgr->imap_delete(dao->mgr, entry);
+	for (i = start; i < end; i++) {
+		dao->mgr->imap_delete(dao->mgr, dao->imappers[i]);
 		dao->imappers[i] = NULL;
 	}
 
-	kfree(dao->imappers[0]);
-	dao->imappers[0] = NULL;
-
+	kfree(to_free);
 	return 0;
 }
 
-static int dao_clear_right_input(struct dao *dao)
-{
-	struct imapper *entry;
-	struct daio *daio = &dao->daio;
-	int i;
 
-	if (!dao->imappers[daio->rscl.msr])
-		return 0;
+static int dao_clear_left_input(struct dao *dao)
+{
+	u32 offset = dao->daio.rscl.msr;
 
-	entry = dao->imappers[daio->rscl.msr];
-	dao->mgr->imap_delete(dao->mgr, entry);
-	/* Program conjugate resources */
-	for (i = 1; i < daio->rscr.msr; i++) {
-		entry = dao->imappers[daio->rscl.msr + i];
-		dao->mgr->imap_delete(dao->mgr, entry);
-		dao->imappers[daio->rscl.msr + i] = NULL;
-	}
+	return dao_clear_input(dao, 0, 0 + offset);
+}
 
-	kfree(dao->imappers[daio->rscl.msr]);
-	dao->imappers[daio->rscl.msr] = NULL;
+static int dao_clear_right_input(struct dao *dao)
+{
+	u32 start = dao->daio.rscl.msr;
+	u32 offset = dao->daio.rscr.msr;
 
-	return 0;
+	return dao_clear_input(dao, start, start + offset);
 }
 
 static const struct dao_rsc_ops dao_ops = {
Signature · 1 line
2.47.1
#1 lore raw
Quoted 5 lines · Ethan Carter Edwards

There was a lote of code duplication in the dao_clear_left_input() and dao_clear_right_input() functions. A new function, dao_clear_input(), was created and now the left and right functions call it instead of repeating themselves.

Quoted 8 lines

---
Changes in V2: Fixed formatting caused by email client. Sorry!

 sound/pci/ctxfi/ctdaio.c | 50 ++++++++++++++--------------------------
 1 file changed, 17 insertions(+), 33 deletions(-)

Alright. It appears my email server is mangling my patches somehow. I have tried using both mutt and git send-email, and it appears that they are still messed up.

I'm using Protonmail as my server with its bridge locally. Does anyone have any experience with working around this? I can't immediately find anything online.

Apologies.
Ethan

Signature · 5 lines
Ethan Carter Edwards
CompTIA A+, Security+, and ISC2 (CC)
Ham Radio: AE4CE
Website: https://ethancedwards.com
GPG: 2E51 F618 39D1 FA94 7A73  00C2 34C0 4305 D581 DBFE
#2 lore raw
In reply to Ethan Carter Edwards

Hi Ethan,

Thank you for your patch. However, there are a few issues:

Quoted 7 lines · Ethan Carter Edwards

There was a lote of code duplication in the dao_clear_left_input() and dao_clear_right_input() functions. A new function, dao_clear_input(), was created and now the left and right functions call it instead of repeating themselves.

Please move the Link: tag to the SoB area without an empty line.

Quoted 2 lines

The Co-developed-by: tag should be immediately followed by the co-author's Signed-off-by: tag. Additionally, the last Signed-off-by:
tag must always belong to the patch submitter.

See: https://docs.kernel.org/process/submitting-patches.html#when-to-use-acked-by-cc-and-co-developed-by

Regards,
Kuan-Wei

#3 lore raw
In reply to Ethan Carter Edwards
Quoted 64 lines · On Mon, 30 Dec 2024 15:32:58 +0000 Ethan Carter Edwards…

There was a lote of code duplication in the dao_clear_left_input() and dao_clear_right_input() functions. A new function, dao_clear_input(), was created and now the left and right functions call it instead of repeating themselves.

---
Changes in V2: Fixed formatting caused by email client. Sorry!

 sound/pci/ctxfi/ctdaio.c | 50 ++++++++++++++--------------------------
 1 file changed, 17 insertions(+), 33 deletions(-)
sound/pci/ctxfi/ctdaio.c +10 -22
index 83aaf9441ef3..d46542d0b454 100644
@@ -211,52 +211,36 @@ static int dao_set_right_input(struct dao *dao, struct rsc *input)
 	return 0;
 }
 
-static int dao_clear_left_input(struct dao *dao)
+static int dao_clear_input(struct dao *dao, unsigned int start, unsigned int end)
 {
-	struct imapper *entry;
-	struct daio *daio = &dao->daio;
-	int i;
+	struct imapper *to_free = dao->imappers[start];
+	unsigned int i;
 
-	if (!dao->imappers[0])
+	if (!to_free)
 		return 0;
-
-	entry = dao->imappers[0];
-	dao->mgr->imap_delete(dao->mgr, entry);
-	/* Program conjugate resources */
-	for (i = 1; i < daio->rscl.msr; i++) {
-		entry = dao->imappers[i];
-		dao->mgr->imap_delete(dao->mgr, entry);
+	for (i = start; i < end; i++) {
+		dao->mgr->imap_delete(dao->mgr, dao->imappers[i]);
 		dao->imappers[i] = NULL;
 	}
 
-	kfree(dao->imappers[0]);
-	dao->imappers[0] = NULL;
-
+	kfree(to_free);
 	return 0;
 }
 
-static int dao_clear_right_input(struct dao *dao)
-{
-	struct imapper *entry;
-	struct daio *daio = &dao->daio;
-	int i;
 
-	if (!dao->imappers[daio->rscl.msr])
-		return 0;
+static int dao_clear_left_input(struct dao *dao)
+{
+	u32 offset = dao->daio.rscl.msr;

sound/pci/ctxfi/ctdaio.c static int dao_set_right_input(struct dao *dao, struct rsc *input) · line 235

Quoted 10 lines
sound/pci/ctxfi/ctdaio.c +1 -8
 
-	entry = dao->imappers[daio->rscl.msr];
-	dao->mgr->imap_delete(dao->mgr, entry);
-	/* Program conjugate resources */
-	for (i = 1; i < daio->rscr.msr; i++) {
-		entry = dao->imappers[daio->rscl.msr + i];
-		dao->mgr->imap_delete(dao->mgr, entry);
-		dao->imappers[daio->rscl.msr + i] = NULL;
-	}
+	return dao_clear_input(dao, 0, 0 + offset);

I really wouldn't bother with the temporary variable.
It is also a 'length' not an offset and you can't decide on the type. 'unsigned int' is fine.

But I suspect it really is worth checking if that *imappers[] is needed at all. (Rename and see where the build breaks!)
You might just need struct imappers *imappers_left and *imappers_right to hold base of the kmalloc()ed block.
At that point the functions collapse down even more.

David

Quoted 14 lines
sound/pci/ctxfi/ctdaio.c +6 -3
+}
 
-	kfree(dao->imappers[daio->rscl.msr]);
-	dao->imappers[daio->rscl.msr] = NULL;
+static int dao_clear_right_input(struct dao *dao)
+{
+	u32 start = dao->daio.rscl.msr;
+	u32 offset = dao->daio.rscr.msr;
 
-	return 0;
+	return dao_clear_input(dao, start, start + offset);
 }
 
 static const struct dao_rsc_ops dao_ops = {