Skip to contentSkip to conversation

RFC: i386: kill !4KSTACKS

DISCUSSION Adrian Bunk 7 participants 14 messages Jun 7, 2005 – Jul 5, 2005 Crypto API

lore
Share & tools
lore
#0 lore raw

4Kb kernel stacks are the future on i386, and it seems the problems it initially caused are now sorted out.

I'd like to:
- get a patch into the next -mm that unconditionally enables 4KSTACKS
- if there won't be new reports of breakages, send a patch to completely remove !4KSTACKS for 2.6.13 or 2.6.14

The only drawback is that REISER4_FS does still depend on !4KSTACKS. I told Hans back in March that this has to be changed.

Is there any ETA until that all issues with 4Kb kernel stacks in Reiser4 will be resolved?

If not people using Reiser4 might have to decide whether to switch the filesystem or the architecture...

cu
Adrian

Signature · 4 lines
       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed
#1 lore raw

tis 2005-06-07 klockan 23:27 +0200 skrev Adrian Bunk:

Quoted 8 lines

4Kb kernel stacks are the future on i386, and it seems the problems it initially caused are now sorted out.

I'd like to:
- get a patch into the next -mm that unconditionally enables 4KSTACKS
- if there won't be new reports of breakages, send a patch to completely remove !4KSTACKS for 2.6.13 or 2.6.14

Combinations of IDE/SCSI with MD/DM (maybe even stacking them ontop of eachother), NFS and a filesystem in there breaks 4KSTACKS which is a known issue so you can't just remove it leaving users with no choice.

This was not even difficult to trigger a while ago and I haven't seen any stack reduction patches in these areas.

#2 lore raw
Quoted 10 lines · Alexander Nyberg

tis 2005-06-07 klockan 23:27 +0200 skrev Adrian Bunk:

Quoted 8 lines

4Kb kernel stacks are the future on i386, and it seems the problems it initially caused are now sorted out.

I'd like to:
- get a patch into the next -mm that unconditionally enables 4KSTACKS
- if there won't be new reports of breakages, send a patch to completely remove !4KSTACKS for 2.6.13 or 2.6.14

Quoted 6 lines

Combinations of IDE/SCSI with MD/DM (maybe even stacking them ontop of eachother), NFS and a filesystem in there breaks 4KSTACKS which is a known issue so you can't just remove it leaving users with no choice.

This was not even difficult to trigger a while ago and I haven't seen any stack reduction patches in these areas.

Can you send me a backtrace for one such dump? I'd like to use that for some pessimistic checker runs.

Jörn

Signature · 2 lines
Good warriors cause others to come to them and do not go to others.
-- Sun Tzu
#3 lore raw
In reply to Alexander Nyberg
Quoted 10 lines · Alexander Nyberg

tis 2005-06-07 klockan 23:27 +0200 skrev Adrian Bunk:

Quoted 8 lines

4Kb kernel stacks are the future on i386, and it seems the problems it initially caused are now sorted out.

I'd like to:
- get a patch into the next -mm that unconditionally enables 4KSTACKS
- if there won't be new reports of breakages, send a patch to completely remove !4KSTACKS for 2.6.13 or 2.6.14

Quoted 6 lines

Combinations of IDE/SCSI with MD/DM (maybe even stacking them ontop of eachother), NFS and a filesystem in there breaks 4KSTACKS which is a known issue so you can't just remove it leaving users with no choice.

This was not even difficult to trigger a while ago and I haven't seen any stack reduction patches in these areas.

Not true. NFS patches were.

Do you have any stack overflow traces? Also "make checkstack" is your friend in looking for suspect functions.

NB: gcc 3.4.3 can use excessive stack in degenerate cases, so please include gcc version in your reports.
--
vda

#4 lore raw
Quoted 3 lines · Denis Vlasenko

...
NB: gcc 3.4.3 can use excessive stack in degenerate cases, so please include gcc version in your reports.

But this can't occur in the kernel.
We are working aroung the possible problems in SuSE gcc 3.3 and GNU gcc 3.4 and 4.0 by always disabling unit-at-a-time in the kernel on i386.

Quoted 1 line

vda

cu
Adrian

Signature · 4 lines
       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed
#5 lore raw
Quoted 6 lines · Adrian Bunk
Quoted 3 lines · Denis Vlasenko

...
NB: gcc 3.4.3 can use excessive stack in degenerate cases, so please include gcc version in your reports.

But this can't occur in the kernel.

It can. I saw the OOPS myself.
One of the functions in crypto/wp512.c was compiled with 3k+ stack usage. --
vda

crypto/wp512.c +162 -152
diff -urpN linux-2.6.12-rc2.1.be_le/crypto/wp512.c linux-2.6.12-rc2.2.wp/crypto/wp512.c
@@ -793,80 +793,90 @@ static void wp512_process_buffer(struct 
 	state[6] = block[6] ^ (K[6] = wctx->hash[6]);
 	state[7] = block[7] ^ (K[7] = wctx->hash[7]);
 
+#if 1
+ #define X(a) a ^=
+ #define XEND ;
+#else
+/* gcc -O2 (3.4.3) optimizer bug:
+** this will cause excessive spills (~3K stack used)
+** See http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21141 */
+ #define X(a) ^
+ #define XEND
+#endif
 	for (r = 1; r <= WHIRLPOOL_ROUNDS; r++) {
 
-		L[0] = C0[(int)(K[0] >> 56)       ] ^
-			   C1[(int)(K[7] >> 48) & 0xff] ^
-			   C2[(int)(K[6] >> 40) & 0xff] ^
-			   C3[(int)(K[5] >> 32) & 0xff] ^
-			   C4[(int)(K[4] >> 24) & 0xff] ^
-			   C5[(int)(K[3] >> 16) & 0xff] ^
-			   C6[(int)(K[2] >>  8) & 0xff] ^
-			   C7[(int)(K[1]      ) & 0xff] ^
-			   rc[r];
-
-		L[1] = C0[(int)(K[1] >> 56)       ] ^
-			   C1[(int)(K[0] >> 48) & 0xff] ^
-			   C2[(int)(K[7] >> 40) & 0xff] ^
-			   C3[(int)(K[6] >> 32) & 0xff] ^
-			   C4[(int)(K[5] >> 24) & 0xff] ^
-			   C5[(int)(K[4] >> 16) & 0xff] ^
-			   C6[(int)(K[3] >>  8) & 0xff] ^
-			   C7[(int)(K[2]      ) & 0xff];
-
-		L[2] = C0[(int)(K[2] >> 56)       ] ^
-			   C1[(int)(K[1] >> 48) & 0xff] ^
-			   C2[(int)(K[0] >> 40) & 0xff] ^
-			   C3[(int)(K[7] >> 32) & 0xff] ^
-			   C4[(int)(K[6] >> 24) & 0xff] ^
-			   C5[(int)(K[5] >> 16) & 0xff] ^
-			   C6[(int)(K[4] >>  8) & 0xff] ^
-			   C7[(int)(K[3]      ) & 0xff];
-
-		L[3] = C0[(int)(K[3] >> 56)       ] ^
-			   C1[(int)(K[2] >> 48) & 0xff] ^
-			   C2[(int)(K[1] >> 40) & 0xff] ^
-			   C3[(int)(K[0] >> 32) & 0xff] ^
-			   C4[(int)(K[7] >> 24) & 0xff] ^
-			   C5[(int)(K[6] >> 16) & 0xff] ^
-			   C6[(int)(K[5] >>  8) & 0xff] ^
-			   C7[(int)(K[4]      ) & 0xff];
-
-		L[4] = C0[(int)(K[4] >> 56)       ] ^
-			   C1[(int)(K[3] >> 48) & 0xff] ^
-			   C2[(int)(K[2] >> 40) & 0xff] ^
-			   C3[(int)(K[1] >> 32) & 0xff] ^
-			   C4[(int)(K[0] >> 24) & 0xff] ^
-			   C5[(int)(K[7] >> 16) & 0xff] ^
-			   C6[(int)(K[6] >>  8) & 0xff] ^
-			   C7[(int)(K[5]      ) & 0xff];
-
-		L[5] = C0[(int)(K[5] >> 56)       ] ^
-			   C1[(int)(K[4] >> 48) & 0xff] ^
-			   C2[(int)(K[3] >> 40) & 0xff] ^
-			   C3[(int)(K[2] >> 32) & 0xff] ^
-			   C4[(int)(K[1] >> 24) & 0xff] ^
-			   C5[(int)(K[0] >> 16) & 0xff] ^
-			   C6[(int)(K[7] >>  8) & 0xff] ^
-			   C7[(int)(K[6]      ) & 0xff];
-
-		L[6] = C0[(int)(K[6] >> 56)       ] ^
-			   C1[(int)(K[5] >> 48) & 0xff] ^
-			   C2[(int)(K[4] >> 40) & 0xff] ^
-			   C3[(int)(K[3] >> 32) & 0xff] ^
-			   C4[(int)(K[2] >> 24) & 0xff] ^
-			   C5[(int)(K[1] >> 16) & 0xff] ^
-			   C6[(int)(K[0] >>  8) & 0xff] ^
-			   C7[(int)(K[7]      ) & 0xff];
-
-		L[7] = C0[(int)(K[7] >> 56)       ] ^
-			   C1[(int)(K[6] >> 48) & 0xff] ^
-			   C2[(int)(K[5] >> 40) & 0xff] ^
-			   C3[(int)(K[4] >> 32) & 0xff] ^
-			   C4[(int)(K[3] >> 24) & 0xff] ^
-			   C5[(int)(K[2] >> 16) & 0xff] ^
-			   C6[(int)(K[1] >>  8) & 0xff] ^
-			   C7[(int)(K[0]      ) & 0xff];
+		L[0]  = C0[BYTE7(K[0])] XEND
+		X(L[0]) C1[BYTE6(K[7])] XEND
+		X(L[0]) C2[BYTE5(K[6])] XEND
+		X(L[0]) C3[BYTE4(K[5])] XEND
+		X(L[0]) C4[BYTE3(K[4])] XEND
+		X(L[0]) C5[BYTE2(K[3])] XEND
+		X(L[0]) C6[BYTE1(K[2])] XEND
+		X(L[0]) C7[BYTE0(K[1])] XEND
+		X(L[0]) rc[r];
+
+		L[1]  = C0[BYTE7(K[1])] XEND
+		X(L[1]) C1[BYTE6(K[0])] XEND
+		X(L[1]) C2[BYTE5(K[7])] XEND
+		X(L[1]) C3[BYTE4(K[6])] XEND
+		X(L[1]) C4[BYTE3(K[5])] XEND
+		X(L[1]) C5[BYTE2(K[4])] XEND
+		X(L[1]) C6[BYTE1(K[3])] XEND
+		X(L[1]) C7[BYTE0(K[2])];
+
+		L[2]  = C0[BYTE7(K[2])] XEND
+		X(L[2]) C1[BYTE6(K[1])] XEND
+		X(L[2]) C2[BYTE5(K[0])] XEND
+		X(L[2]) C3[BYTE4(K[7])] XEND
+		X(L[2]) C4[BYTE3(K[6])] XEND
+		X(L[2]) C5[BYTE2(K[5])] XEND
+		X(L[2]) C6[BYTE1(K[4])] XEND
+		X(L[2]) C7[BYTE0(K[3])];
+
+		L[3]  = C0[BYTE7(K[3])] XEND
+		X(L[3]) C1[BYTE6(K[2])] XEND
+		X(L[3]) C2[BYTE5(K[1])] XEND
+		X(L[3]) C3[BYTE4(K[0])] XEND
+		X(L[3]) C4[BYTE3(K[7])] XEND
+		X(L[3]) C5[BYTE2(K[6])] XEND
+		X(L[3]) C6[BYTE1(K[5])] XEND
+		X(L[3]) C7[BYTE0(K[4])];
+
+		L[4]  = C0[BYTE7(K[4])] XEND
+		X(L[4]) C1[BYTE6(K[3])] XEND
+		X(L[4]) C2[BYTE5(K[2])] XEND
+		X(L[4]) C3[BYTE4(K[1])] XEND
+		X(L[4]) C4[BYTE3(K[0])] XEND
+		X(L[4]) C5[BYTE2(K[7])] XEND
+		X(L[4]) C6[BYTE1(K[6])] XEND
+		X(L[4]) C7[BYTE0(K[5])];
+
+		L[5]  = C0[BYTE7(K[5])] XEND
+		X(L[5]) C1[BYTE6(K[4])] XEND
+		X(L[5]) C2[BYTE5(K[3])] XEND
+		X(L[5]) C3[BYTE4(K[2])] XEND
+		X(L[5]) C4[BYTE3(K[1])] XEND
+		X(L[5]) C5[BYTE2(K[0])] XEND
+		X(L[5]) C6[BYTE1(K[7])] XEND
+		X(L[5]) C7[BYTE0(K[6])];
+
+		L[6]  = C0[BYTE7(K[6])] XEND
+		X(L[6]) C1[BYTE6(K[5])] XEND
+		X(L[6]) C2[BYTE5(K[4])] XEND
+		X(L[6]) C3[BYTE4(K[3])] XEND
+		X(L[6]) C4[BYTE3(K[2])] XEND
+		X(L[6]) C5[BYTE2(K[1])] XEND
+		X(L[6]) C6[BYTE1(K[0])] XEND
+		X(L[6]) C7[BYTE0(K[7])];
+
+		L[7]  = C0[BYTE7(K[7])] XEND
+		X(L[7]) C1[BYTE6(K[6])] XEND
+		X(L[7]) C2[BYTE5(K[5])] XEND
+		X(L[7]) C3[BYTE4(K[4])] XEND
+		X(L[7]) C4[BYTE3(K[3])] XEND
+		X(L[7]) C5[BYTE2(K[2])] XEND
+		X(L[7]) C6[BYTE1(K[1])] XEND
+		X(L[7]) C7[BYTE0(K[0])];
 
 		K[0] = L[0];
 		K[1] = L[1];
@@ -877,85 +887,85 @@ static void wp512_process_buffer(struct 
 		K[6] = L[6];
 		K[7] = L[7];
 
-		L[0] = C0[(int)(state[0] >> 56)       ] ^
-			   C1[(int)(state[7] >> 48) & 0xff] ^
-			   C2[(int)(state[6] >> 40) & 0xff] ^
-			   C3[(int)(state[5] >> 32) & 0xff] ^
-			   C4[(int)(state[4] >> 24) & 0xff] ^
-			   C5[(int)(state[3] >> 16) & 0xff] ^
-			   C6[(int)(state[2] >>  8) & 0xff] ^
-			   C7[(int)(state[1]      ) & 0xff] ^
-			   K[0];
-
-		L[1] = C0[(int)(state[1] >> 56)       ] ^
-			   C1[(int)(state[0] >> 48) & 0xff] ^
-			   C2[(int)(state[7] >> 40) & 0xff] ^
-			   C3[(int)(state[6] >> 32) & 0xff] ^
-			   C4[(int)(state[5] >> 24) & 0xff] ^
-			   C5[(int)(state[4] >> 16) & 0xff] ^
-			   C6[(int)(state[3] >>  8) & 0xff] ^
-			   C7[(int)(state[2]      ) & 0xff] ^
-			   K[1];
-
-		L[2] = C0[(int)(state[2] >> 56)       ] ^
-			   C1[(int)(state[1] >> 48) & 0xff] ^
-			   C2[(int)(state[0] >> 40) & 0xff] ^
-			   C3[(int)(state[7] >> 32) & 0xff] ^
-			   C4[(int)(state[6] >> 24) & 0xff] ^
-			   C5[(int)(state[5] >> 16) & 0xff] ^
-			   C6[(int)(state[4] >>  8) & 0xff] ^
-			   C7[(int)(state[3]      ) & 0xff] ^
-			   K[2];
-
-		L[3] = C0[(int)(state[3] >> 56)       ] ^
-			   C1[(int)(state[2] >> 48) & 0xff] ^
-			   C2[(int)(state[1] >> 40) & 0xff] ^
-			   C3[(int)(state[0] >> 32) & 0xff] ^
-			   C4[(int)(state[7] >> 24) & 0xff] ^
-			   C5[(int)(state[6] >> 16) & 0xff] ^
-			   C6[(int)(state[5] >>  8) & 0xff] ^
-			   C7[(int)(state[4]      ) & 0xff] ^
-			   K[3];
-
-		L[4] = C0[(int)(state[4] >> 56)       ] ^
-			   C1[(int)(state[3] >> 48) & 0xff] ^
-			   C2[(int)(state[2] >> 40) & 0xff] ^
-			   C3[(int)(state[1] >> 32) & 0xff] ^
-			   C4[(int)(state[0] >> 24) & 0xff] ^
-			   C5[(int)(state[7] >> 16) & 0xff] ^
-			   C6[(int)(state[6] >>  8) & 0xff] ^
-			   C7[(int)(state[5]      ) & 0xff] ^
-			   K[4];
-
-		L[5] = C0[(int)(state[5] >> 56)       ] ^
-			   C1[(int)(state[4] >> 48) & 0xff] ^
-			   C2[(int)(state[3] >> 40) & 0xff] ^
-			   C3[(int)(state[2] >> 32) & 0xff] ^
-			   C4[(int)(state[1] >> 24) & 0xff] ^
-			   C5[(int)(state[0] >> 16) & 0xff] ^
-			   C6[(int)(state[7] >>  8) & 0xff] ^
-			   C7[(int)(state[6]      ) & 0xff] ^
-			   K[5];
-
-		L[6] = C0[(int)(state[6] >> 56)       ] ^
-			   C1[(int)(state[5] >> 48) & 0xff] ^
-			   C2[(int)(state[4] >> 40) & 0xff] ^
-			   C3[(int)(state[3] >> 32) & 0xff] ^
-			   C4[(int)(state[2] >> 24) & 0xff] ^
-			   C5[(int)(state[1] >> 16) & 0xff] ^
-			   C6[(int)(state[0] >>  8) & 0xff] ^
-			   C7[(int)(state[7]      ) & 0xff] ^
-			   K[6];
-
-		L[7] = C0[(int)(state[7] >> 56)       ] ^
-			   C1[(int)(state[6] >> 48) & 0xff] ^
-			   C2[(int)(state[5] >> 40) & 0xff] ^
-			   C3[(int)(state[4] >> 32) & 0xff] ^
-			   C4[(int)(state[3] >> 24) & 0xff] ^
-			   C5[(int)(state[2] >> 16) & 0xff] ^
-			   C6[(int)(state[1] >>  8) & 0xff] ^
-			   C7[(int)(state[0]      ) & 0xff] ^
-			   K[7];
+		L[0]  = C0[BYTE7(state[0])] XEND
+		X(L[0]) C1[BYTE6(state[7])] XEND
+		X(L[0]) C2[BYTE5(state[6])] XEND
+		X(L[0]) C3[BYTE4(state[5])] XEND
+		X(L[0]) C4[BYTE3(state[4])] XEND
+		X(L[0]) C5[BYTE2(state[3])] XEND
+		X(L[0]) C6[BYTE1(state[2])] XEND
+		X(L[0]) C7[BYTE0(state[1])] XEND
+		X(L[0]) K[0];
+
+		L[1]  = C0[BYTE7(state[1])] XEND
+		X(L[1]) C1[BYTE6(state[0])] XEND
+		X(L[1]) C2[BYTE5(state[7])] XEND
+		X(L[1]) C3[BYTE4(state[6])] XEND
+		X(L[1]) C4[BYTE3(state[5])] XEND
+		X(L[1]) C5[BYTE2(state[4])] XEND
+		X(L[1]) C6[BYTE1(state[3])] XEND
+		X(L[1]) C7[BYTE0(state[2])] XEND
+		X(L[1]) K[1];
+
+		L[2]  = C0[BYTE7(state[2])] XEND
+		X(L[2]) C1[BYTE6(state[1])] XEND
+		X(L[2]) C2[BYTE5(state[0])] XEND
+		X(L[2]) C3[BYTE4(state[7])] XEND
+		X(L[2]) C4[BYTE3(state[6])] XEND
+		X(L[2]) C5[BYTE2(state[5])] XEND
+		X(L[2]) C6[BYTE1(state[4])] XEND
+		X(L[2]) C7[BYTE0(state[3])] XEND
+		X(L[2]) K[2];
+
+		L[3]  = C0[BYTE7(state[3])] XEND
+		X(L[3]) C1[BYTE6(state[2])] XEND
+		X(L[3]) C2[BYTE5(state[1])] XEND
+		X(L[3]) C3[BYTE4(state[0])] XEND
+		X(L[3]) C4[BYTE3(state[7])] XEND
+		X(L[3]) C5[BYTE2(state[6])] XEND
+		X(L[3]) C6[BYTE1(state[5])] XEND
+		X(L[3]) C7[BYTE0(state[4])] XEND
+		X(L[3]) K[3];
+
+		L[4]  = C0[BYTE7(state[4])] XEND
+		X(L[4]) C1[BYTE6(state[3])] XEND
+		X(L[4]) C2[BYTE5(state[2])] XEND
+		X(L[4]) C3[BYTE4(state[1])] XEND
+		X(L[4]) C4[BYTE3(state[0])] XEND
+		X(L[4]) C5[BYTE2(state[7])] XEND
+		X(L[4]) C6[BYTE1(state[6])] XEND
+		X(L[4]) C7[BYTE0(state[5])] XEND
+		X(L[4]) K[4];
+
+		L[5]  = C0[BYTE7(state[5])] XEND
+		X(L[5]) C1[BYTE6(state[4])] XEND
+		X(L[5]) C2[BYTE5(state[3])] XEND
+		X(L[5]) C3[BYTE4(state[2])] XEND
+		X(L[5]) C4[BYTE3(state[1])] XEND
+		X(L[5]) C5[BYTE2(state[0])] XEND
+		X(L[5]) C6[BYTE1(state[7])] XEND
+		X(L[5]) C7[BYTE0(state[6])] XEND
+		X(L[5]) K[5];
+
+		L[6]  = C0[BYTE7(state[6])] XEND
+		X(L[6]) C1[BYTE6(state[5])] XEND
+		X(L[6]) C2[BYTE5(state[4])] XEND
+		X(L[6]) C3[BYTE4(state[3])] XEND
+		X(L[6]) C4[BYTE3(state[2])] XEND
+		X(L[6]) C5[BYTE2(state[1])] XEND
+		X(L[6]) C6[BYTE1(state[0])] XEND
+		X(L[6]) C7[BYTE0(state[7])] XEND
+		X(L[6]) K[6];
+
+		L[7]  = C0[BYTE7(state[7])] XEND
+		X(L[7]) C1[BYTE6(state[6])] XEND
+		X(L[7]) C2[BYTE5(state[5])] XEND
+		X(L[7]) C3[BYTE4(state[4])] XEND
+		X(L[7]) C4[BYTE3(state[3])] XEND
+		X(L[7]) C5[BYTE2(state[2])] XEND
+		X(L[7]) C6[BYTE1(state[1])] XEND
+		X(L[7]) C7[BYTE0(state[0])] XEND
+		X(L[7]) K[7];
 
 		state[0] = L[0];
 		state[1] = L[1];
@@ -979,7 +989,7 @@ static void wp512_process_buffer(struct 
 	wctx->hash[7] ^= state[7] ^ block[7];
 }
 
-static void wp512_init (void *ctx) {
+static void wp512_init(void *ctx) {
 	int i;
 	struct wp512_ctx *wctx = ctx;
 
#6 lore raw
Quoted 10 lines · Denis Vlasenko
Quoted 6 lines · Adrian Bunk
Quoted 3 lines · Denis Vlasenko

...
NB: gcc 3.4.3 can use excessive stack in degenerate cases, so please include gcc version in your reports.

But this can't occur in the kernel.

It can. I saw the OOPS myself.
One of the functions in crypto/wp512.c was compiled with 3k+ stack usage.

Strange that "make checkstack" didn't show this.

Are there any other 4KSTACKS problems you know about?

Quoted 1 line

vda

cu
Adrian

Signature · 4 lines
       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed
#7 lore raw
Quoted 9 lines
Quoted 7 lines
Quoted 4 lines
Quoted 2 lines

NB: gcc 3.4.3 can use excessive stack in degenerate cases, so please include gcc version in your reports.

But this can't occur in the kernel.

It can. I saw the OOPS myself.
One of the functions in crypto/wp512.c was compiled with 3k+ stack usage.

Strange that "make checkstack" didn't show this.

It happens with certain gcc versions only.

Quoted 1 line

Are there any other 4KSTACKS problems you know about?

Currently no.
--
vda

#8 lore raw
Quoted 11 lines · Denis Vlasenko
Quoted 9 lines
Quoted 7 lines
Quoted 4 lines
Quoted 2 lines

NB: gcc 3.4.3 can use excessive stack in degenerate cases, so please include gcc version in your reports.

But this can't occur in the kernel.

It can. I saw the OOPS myself.
One of the functions in crypto/wp512.c was compiled with 3k+ stack usage.

Strange that "make checkstack" didn't show this.

It happens with certain gcc versions only.

Which ones, exactly? If they break this way, a workaround (or forbidding them) would be the way to go...

Signature · 4 lines
Dr. Horst H. von Brand                   User #22616 counter.li.org
Departamento de Informatica                     Fono: +56 32 654431
Universidad Tecnica Federico Santa Maria              +56 32 654239
Casilla 110-V, Valparaiso, Chile                Fax:  +56 32 797513
#9 lore raw
In reply to Adrian Bunk
Quoted 10 lines · Adrian Bunk

4Kb kernel stacks are the future on i386, and it seems the problems it initially caused are now sorted out.

I'd like to:
- get a patch into the next -mm that unconditionally enables 4KSTACKS
- if there won't be new reports of breakages, send a patch to completely remove !4KSTACKS for 2.6.13 or 2.6.14

The only drawback is that REISER4_FS does still depend on !4KSTACKS. I told Hans back in March that this has to be changed.

What about ndiswrapper?
Why is it so important to make this happen unconditionally?

Will you maintain a patch for those of us that need it?

Ian

#10 lore raw
Quoted 10 lines · Ian Kent
Quoted 8 lines · Adrian Bunk

4Kb kernel stacks are the future on i386, and it seems the problems it initially caused are now sorted out.

I'd like to:
- get a patch into the next -mm that unconditionally enables 4KSTACKS
- if there won't be new reports of breakages, send a patch to completely remove !4KSTACKS for 2.6.13 or 2.6.14

Quoted 5 lines
Quoted 2 lines

The only drawback is that REISER4_FS does still depend on !4KSTACKS. I told Hans back in March that this has to be changed.

What about ndiswrapper?
Why is it so important to make this happen unconditionally?

Number of folks using ndiswrapper for acx100/acx111 while acx team needs help on native driver debugging worries me.
--
vda

#11 lore raw
Quoted 14 lines · Denis Vlasenko
Quoted 13 lines · Ian Kent
Quoted 10 lines · Adrian Bunk

4Kb kernel stacks are the future on i386, and it seems the problems it initially caused are now sorted out.

I'd like to:
- get a patch into the next -mm that unconditionally enables 4KSTACKS
- if there won't be new reports of breakages, send a patch to completely remove !4KSTACKS for 2.6.13 or 2.6.14

The only drawback is that REISER4_FS does still depend on !4KSTACKS. I told Hans back in March that this has to be changed.

Quoted 6 lines
Quoted 2 lines

What about ndiswrapper?
Why is it so important to make this happen unconditionally?

Number of folks using ndiswrapper for acx100/acx111 while acx team needs help on native driver debugging worries me.

And there's no support for any wireless usb devices that I can see as yet.

Ian

#12 lore raw
In reply to Adrian Bunk

Hello

Quoted 8 lines · Adrian Bunk

4Kb kernel stacks are the future on i386, and it seems the problems it initially caused are now sorted out.

I'd like to:
- get a patch into the next -mm that unconditionally enables 4KSTACKS
- if there won't be new reports of breakages, send a patch to completely remove !4KSTACKS for 2.6.13 or 2.6.14

Quoted 5 lines

The only drawback is that REISER4_FS does still depend on !4KSTACKS. I told Hans back in March that this has to be changed.
Is there any ETA until that all issues with 4Kb kernel stacks in Reiser4 will be resolved?

yes, it should be ready to the end of this week.

Quoted 5 lines

If not people using Reiser4 might have to decide whether to switch the filesystem or the architecture...

cu
Adrian

#13 lore raw

Hello

Quoted 11 lines · Vladimir Saveliev

Hello

Quoted 8 lines · Adrian Bunk

4Kb kernel stacks are the future on i386, and it seems the problems it initially caused are now sorted out.

I'd like to:
- get a patch into the next -mm that unconditionally enables 4KSTACKS
- if there won't be new reports of breakages, send a patch to completely remove !4KSTACKS for 2.6.13 or 2.6.14

Quoted 8 lines
Quoted 5 lines

The only drawback is that REISER4_FS does still depend on !4KSTACKS. I told Hans back in March that this has to be changed.
Is there any ETA until that all issues with 4Kb kernel stacks in Reiser4 will be resolved?

yes, it should be ready to the end of this week.

Well, this estimation was too optimistic.
I have some progress on this but this work is not yet completed. It continues. New estimation is the middle of next week.

Quoted 6 lines
Quoted 6 lines

If not people using Reiser4 might have to decide whether to switch the filesystem or the architecture...

cu
Adrian