qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Scott Wood <scottwood@freescale.com>
To: Alexander Graf <agraf@suse.de>
Cc: Scott Wood <scottwood@freescale.com>,
	qemu-ppc@nongnu.org, qemu-devel@nongnu.org
Subject: [Qemu-devel] [PATCH 10/15] Revert "openpic: Accelerate pending irq search"
Date: Fri, 21 Dec 2012 20:15:47 -0600	[thread overview]
Message-ID: <1356142552-13453-11-git-send-email-scottwood@freescale.com> (raw)
In-Reply-To: <1356142552-13453-1-git-send-email-scottwood@freescale.com>

This reverts commit a9bd83f4c65de0058659ede009fa1a241f379edd.

This counting approach is not robust against setting a bit that
was already set, or clearing a bit that was already clear.  Perhaps
that is considered a bug, but besides the lack of any documentation
for that restriction, it's a pretty unpleasant way for the problem
to manifest itself.

It could be made more robust by testing the current value of the
bit before changing the count, but a later patch speeds up IRQ_check
in all cases, not just when there's nothing pending.  Hopefully that
should be adequate to address performance concerns.

Signed-off-by: Scott Wood <scottwood@freescale.com>
---
 hw/openpic.c |   11 -----------
 1 file changed, 11 deletions(-)

diff --git a/hw/openpic.c b/hw/openpic.c
index e2e7079..cc3e514 100644
--- a/hw/openpic.c
+++ b/hw/openpic.c
@@ -181,7 +181,6 @@ typedef struct IRQ_queue_t {
     uint32_t queue[BF_WIDTH(MAX_IRQ)];
     int next;
     int priority;
-    int pending;    /* nr of pending bits in queue */
 } IRQ_queue_t;
 
 typedef struct IRQ_src_t {
@@ -270,13 +269,11 @@ typedef struct OpenPICState {
 
 static inline void IRQ_setbit(IRQ_queue_t *q, int n_IRQ)
 {
-    q->pending++;
     set_bit(q->queue, n_IRQ);
 }
 
 static inline void IRQ_resetbit(IRQ_queue_t *q, int n_IRQ)
 {
-    q->pending--;
     reset_bit(q->queue, n_IRQ);
 }
 
@@ -292,12 +289,6 @@ static void IRQ_check(OpenPICState *opp, IRQ_queue_t *q)
 
     next = -1;
     priority = -1;
-
-    if (!q->pending) {
-        /* IRQ bitmap is empty */
-        goto out;
-    }
-
     for (i = 0; i < opp->max_irq; i++) {
         if (IRQ_testbit(q, i)) {
             DPRINTF("IRQ_check: irq %d set ivpr_pr=%d pr=%d\n",
@@ -308,8 +299,6 @@ static void IRQ_check(OpenPICState *opp, IRQ_queue_t *q)
             }
         }
     }
-
-out:
     q->next = next;
     q->priority = priority;
 }
-- 
1.7.9.5

  parent reply	other threads:[~2012-12-22  2:16 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-12-22  2:15 [Qemu-devel] [PATCH 00/15] openpic: cleanups and fixes Scott Wood
2012-12-22  2:15 ` [Qemu-devel] [PATCH 01/15] openpic: fix debug prints Scott Wood
2013-01-03 17:31   ` Alexander Graf
2013-01-03 19:41     ` Scott Wood
2012-12-22  2:15 ` [Qemu-devel] [PATCH 02/15] openpic: lower interrupt when reading the MSI register Scott Wood
2013-01-03 17:52   ` Alexander Graf
2012-12-22  2:15 ` [Qemu-devel] [PATCH 03/15] openpic: fix sense and priority bits Scott Wood
2013-01-03 17:51   ` Alexander Graf
2013-01-03 20:12     ` Scott Wood
2012-12-22  2:15 ` [Qemu-devel] [PATCH 04/15] ppc/booke: fix crit/mcheck/debug exceptions Scott Wood
2013-01-03 17:57   ` Alexander Graf
2012-12-22  2:15 ` [Qemu-devel] [PATCH 05/15] openpic: make register names correspond better with hw docs Scott Wood
2013-01-03 18:18   ` Alexander Graf
2012-12-22  2:15 ` [Qemu-devel] [PATCH 06/15] openpic: rework critical interrupt support Scott Wood
2013-01-03 18:31   ` Alexander Graf
2013-01-03 23:07     ` Scott Wood
2013-01-04  8:04       ` Alexander Graf
2013-01-04 20:46       ` [Qemu-devel] [Qemu-ppc] " Blue Swirl
2013-01-04 20:49         ` Scott Wood
2013-01-04 21:17           ` Blue Swirl
2013-01-04 21:25             ` Scott Wood
2012-12-22  2:15 ` [Qemu-devel] [PATCH 07/15] openpic: make ctpr signed Scott Wood
2013-01-03 18:33   ` Alexander Graf
2012-12-22  2:15 ` [Qemu-devel] [PATCH 08/15] openpic/fsl: critical interrupts ignore mask before v4.1 Scott Wood
2013-01-03 18:37   ` Alexander Graf
2012-12-22  2:15 ` [Qemu-devel] [PATCH 09/15] openpic: always call IRQ_check from IRQ_get_next Scott Wood
2013-01-03 18:42   ` Alexander Graf
2013-01-03 20:09     ` Scott Wood
2012-12-22  2:15 ` Scott Wood [this message]
2012-12-22  2:15 ` [Qemu-devel] [PATCH 11/15] openpic: use standard bitmap operations Scott Wood
2013-01-03 18:49   ` Alexander Graf
2012-12-22  2:15 ` [Qemu-devel] [PATCH 12/15] openpic: IRQ_check: search the queue a word at a time Scott Wood
2013-01-03 18:53   ` Alexander Graf
2013-01-03 20:07     ` Scott Wood
2013-01-03 20:31       ` Alexander Graf
2013-01-03 20:32         ` Scott Wood
2013-01-03 20:57           ` Alexander Graf
2013-01-03 21:52             ` Scott Wood
2012-12-22  2:15 ` [Qemu-devel] [PATCH 13/15] openpic: add some bounds checking for IRQ numbers Scott Wood
2013-01-03 18:55   ` Alexander Graf
2013-01-03 19:54     ` [Qemu-devel] [Qemu-ppc] " Scott Wood
2013-01-03 21:07       ` Alexander Graf
2013-01-03 21:20         ` Scott Wood
2012-12-22  2:15 ` [Qemu-devel] [PATCH 14/15] openpic: move IACK to its own function Scott Wood
2013-01-03 18:59   ` Alexander Graf
2012-12-22  2:15 ` [Qemu-devel] [PATCH 15/15] openpic: fix CTPR and de-assertion of interrupts Scott Wood
2013-01-03 19:00   ` Alexander Graf

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1356142552-13453-11-git-send-email-scottwood@freescale.com \
    --to=scottwood@freescale.com \
    --cc=agraf@suse.de \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-ppc@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).